BIOPHYSICAL DIVERSITY ON LANDSLIDES
IN THE PEACE RIVER REGION OF BRITISH COLUMBIA
by
Victoria Kress
B.Sc. NRM, Forestry and Wildlife, University of Northern British Columbia, 2000
MF, University of British Columbia, 2014

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
IN
NATURAL RESOURCES AND ENVIRONMENTAL STUDIES

UNIVERSITY OF NORTHERN BRITISH COLUMBIA
December 2023
©Victoria Kress, 2023

ABSTRACT
The aim of this dissertation is to investigate and quantify biophysical diversity on
landslides in the boreal forest of the Peace River Region of northern British Columbia (BC),
sampling three landslides that occurred in the last 50 years. Landslides are increasing in the
boreal forest, likely driven by climate changes such as increased precipitation and permafrost
thaw, and as a derivative of large wildfires. An understanding of ecosystem recovery on
landslides is important for conservation and management purposes. Several studies have been
done in southern parts of the world to elucidate processes of landslide recovery. However, few
studies have addressed landslide recovery in northern climates, and little is known about the
biophysical diversity of landslides in this region. This research investigates whether landslides
are more biophysically diverse than the surrounding relatively undisturbed terrain, and whether
microsite variables or geomorphic diversity are predictors of plant community diversity. Using a
series of field sampling campaigns and GIS (geographic information system) mapping exercises,
I show that landslides are more biophysically diverse than the surrounding terrain in some
respects, while the surrounding undisturbed terrain is more diverse in other aspects. The age and
size of landslides also appear to influence diversity. Microsite diversity does not necessarily
predict plant diversity. The research highlights the role of invasive plant species in slope
stabilisation and plant community makeup. I also show that landslide ponds are
disproportionately concentrated on rotational landslides, and that most ponds on landslides occur
in the body and toe. I note post-slide modifications such as drainage of landslide ponds and
lowering of landslide ridges, but many geomorphic features are expected to endure for decades to
millennia. Overall, the research shows that vegetation recovery is complex and may take
decades to fully manifest. This study contributes knowledge about plant community and site
diversity on landslides by providing quantitative data and comparing those traits with those found
ii

on surrounding terrain. These findings can be used as guidance when identifying conservation
and management practices for ecological restoration of disturbed slopes.

iii

TABLE OF CONTENTS
ABSTRACT ......................................................................................................................... ii
TABLE OF CONTENTS ..................................................................................................... iv
LIST OF TABLES ............................................................................................................... vi
LIST OF FIGURES ........................................................................................................... viii
ACKNOWLEDGEMENTS ............................................................................................... xiii
DEDICATION................................................................................................................... xiv
Chapter 1. Introduction to landslide biophysical diversity research ....................................... 1
1.1 Introduction................................................................................................................. 1
1.2 Background to the study .............................................................................................. 2
1.3 The research problem .................................................................................................. 6
1.4 Research aims, objectives, and questions ..................................................................... 8
1.5 Significance .............................................................................................................. 13
1.6 Structural outline of the dissertation .......................................................................... 13
1.7 References ................................................................................................................ 15
Chapter 2. Biophysical alpha diversity of landslides and comparison with surrounding
undisturbed terrain.............................................................................................................. 18
2.1 Introduction and Background .................................................................................... 18
2.2 Study areas ................................................................................................................ 27
2.3 Methods and Analysis ............................................................................................... 31
2.4 Results ...................................................................................................................... 46
2.5 Discussion ................................................................................................................. 81
2.6 Conclusions and recommendations .......................................................................... 106
2.7 References .............................................................................................................. 113
Chapter 3. Beta diversity as a measure of biophysical turnover on landslides .................... 122
3.1 Introduction and Background .................................................................................. 122
3.2 Methods .................................................................................................................. 127
3.3 Results .................................................................................................................... 131
3.4 Discussion ............................................................................................................... 141
3.5 Conclusions and Recommendations ........................................................................ 159
3.6 References .............................................................................................................. 162

iv

Chapter 4. Landslide-generated ponds in the Fort St. John area, British Columbia:
Characteristics, distribution, and ecological implications .................................................. 165
4.1. Introduction............................................................................................................ 165
4.2. Study area .............................................................................................................. 168
4.3. Methods ................................................................................................................. 170
4.4. Results ................................................................................................................... 177
4.5. Discussion .............................................................................................................. 193
4.5. Conclusions and recommendations ......................................................................... 203
4.6 References .............................................................................................................. 205
Chapter 5. Conclusions and Recommendations for Landslide Recovery and Management 208
5.1 Conclusions............................................................................................................. 208
5.2 Recommendations ................................................................................................... 212
5.3 Final thoughts ......................................................................................................... 215
APPENDICES.................................................................................................................. 217
Appendix 1 Material Origin classes .................................................................................. 218
Appendix 2 Vegetation summary tables - Mean species cover (% of total relevé area) ...... 219
Appendix 3 Vegetation cover by growth form ................................................................... 236
Appendix 4 Landslide geotypes glossary .......................................................................... 256
Appendix 5 Geotype polygons .......................................................................................... 259
Appendix 6 Microtopography elevation summary tables - BEC plots................................ 274
Appendix 7 Landslide pond descriptions - Mapsheet 94A ................................................. 278
Appendix 8 Pond size statistics by landslide type.............................................................. 302
Appendix 9 Pond size by geomorphic location per slide type ............................................ 303

v

LIST OF TABLES
Table 2-1. Ten most abundant plant species for relevés at all study sites,
comparing landslide and undisturbed results. Values are mean percent cover of
the entire relevé area +/- standard error of the mean.

48

Table 2-2. Relevé mean growth form cover as a percentage of mean total
vegetation cover, along with proportions, for all study sites. The table provides a
comparison between landslide and undisturbed vegetation growth form cover.

51

Table 2-3. Combined* and mean (+/- standard error) vegetation diversity indices
for relevés

52

Table 2-4. Hill numbers for vegetation diversity on relevés.

53

Table 2-5. Pearson correlations of BEC plot plant species abundance with NMS
Axis 1. N = 116.

61

Table 2-6. Pearson correlations of BEC plot plant species abundance with NMS
Axis 2. N = 116.

61

Table 2-7. Pearson correlations of environment variables and Axis 1 and Axis 2
of the ordination space. N=116.

62

Table 2-8. Site types/series found on all landslides and undisturbed terrain in the
study.

63

Table 2-9. Diversity indices for site series polygons - all study sites. BE =
Beatton River, CE = Cecil Lake, HA = Hasler Flats.

69

Table 2-10. Landslide geotype summary by area (m2) and
percentage.

73

Table 2-11. Geotype diversity indices for landslide geotype polygons.

74

Table 2-12. Undisturbed geotype summary by area (m2) and percentage.

75

Table 2-13. Geotype diversity indices for undisturbed geotype polygons.

76

Table 2-14. BEC 50m2 plot LiDAR points: Elevation coefficient of variation
(CV) summary statistics.

78

Table 3-1. Landslide transect vegetation beta diversity (Bw) values.

132

Table 3-2. Landslide transect vegetation alpha diversity statistics.

133

Table 3-3. Undisturbed transect vegetation beta diversity (Bw) values.

134

vi

Table 3-4. Undisturbed transect vegetation alpha diversity statistics.

135

Table 3-5. Landslide transect environment beta diversity (Bw) values.

136

Table 3-6. Landslide transect environment alpha diversity statistics.

137

Table 3-7. Undisturbed transect environment beta diversity (Bw) values.

138

Table 3-8. Undisturbed transect environment alpha diversity statistics.

139

Table 4-1. Summary information on ponds and landslides in the 768.25 km2
(76,825 ha) study area.

179

vii

LIST OF FIGURES
Figure 1-1. Beatton River landslide. Top image is an oblique 3D presentation of
an ortho mosaic draped over drone imagery (flown 2016), at vertical
exaggeration 1X. Bottom image is an aerial view of the landslide, illustrating
temporary partial blockage of the Beatton River. (Aerial photo on bottom was
taken October 20, 2015, by M. Geertsema. Used with permission.)

10

Figure 1-2. Cecil Lake landslide. Top image is an oblique 3D presentation of
World Imagery layer drape over LiDAR imagery, vertical exaggeration 1X. Inset
in top image is a photograph of the headscarp and part of the landslide, facing
west. Inset photo was taken May 12, 2016. Bottom image is an ortho mosaic
presentation of the landslide, showing variable topography and ponds.

11

Figure 1-3. Hasler Flats landslide. Top image is an ortho mosaic draped over
drone imagery, vertical exaggeration 1X. Bottom image is an aerial view of the
landslide, facing south. (Aerial photo on bottom was taken September 30, 2013,
by M. Geertsema. Used with permission.)

12

Figure 2-1. Study areas of Beatton River, Cecil Lake, and Hasler Flats
landslides, shown in Google Earth Pro. The landslides are in the Peace River
Region of northeastern British Columbia, Canada. Landslide locations are
outlined in yellow and identified with red locator balloons. Inset map shows
landslide locations at a provincial scale and in relation to Mapsheet 94A, which is
outlined in red. BE = Beatton River landslide, CE = Cecil Lake landslide, and
HA = Hasler Flats landslide.

29

Figure 2-2. Vegetation and site diversity on landslides in the study. From top
left clockwise: horst and graben complex and mature rafts (Hasler Flats. Photo
July 12, 2018); unvegetated weathered pillar next to heavy brush (Cecil Lake.
Photo September 17, 2017); dewatered/revegetated pond site (Cecil Lake. Photo
August 23, 2018); steep unvegetated scarps interspersed with relatively level
swaths of abundant invasive vegetation and the occasional pond (Beatton River.
Photo August 15, 2017).

47

Figure 2-3. Comparison of rank abundance distributions for relevé plot plant
species for all study sites. Abundance rank by species is on the x-axis.

55

Figure 2-4. Kolmogorov-Smirnov two sample test for all relevés, comparing
landslide and undisturbed plant abundance for Beatton River (BE), Cecil Lake
(CE), and Hasler Flats (HA) study areas. Results where the maximum D statistic
is greater than the critical value indicate a significant difference in plant
community between two samples.

57

viii

Figure 2-5. NMS (nonmetric multidimensional scaling) ordination biplot of 116
subjectively located averaged BEC (biogeoclimatic ecosystem classification)
plots and 163 plant species sampled on three landslide study areas. The scale of
the biplot has been increased to 700% due to the crowded distribution of the
plots. Axis 1 explains 50% of the variation and represents an apparent gradient
from forest community to open/grassland community, from left to right across
the figure. Axis 2 explains 24% of the variation and represents an apparent soil
moisture gradient from drier to wetter, moving from the upper to lower extent of
the figure. Both axes are significant using the Monte Carlo test (p = 0.0040 and
final stress = 18.20). Study areas are colour-coded, as shown in the legend. BE =
Beatton River, CE = Cecil Lake, and HA = Hasler Flats. The environmental
variables correlated with the ordination axes are: MesoslopePos (mesoslope
position), Moisture (soil moisture regime), SlopeGrad (slope gradient in %), HLI
(heat load index), and MatOrigin (material origin). Strength and nature of
environmental variable relationships are represented by the length and direction
of the red vectors. Undisturbed benchmark plots BE16-29, BE16-30, CE-35a,
and HA16-01 are highlighted in yellow in the biplot.

59

Figure 2-6. Site series/types mapping example for Beatton River landslide study
site. Site series/types codes (shown in blue) are 101 = Sw – Trailing raspberry –
Step moss, 101$ = At – Rose – Creamy peavine, 103 = SwPl – Soopolallie –
Fuzzy-spiked wildrye, 110$ = At – Highbush-cranberry – Oak fern, 111 = Sw –
Currant - Horsetail, Gg = Grassland, Ws = Wetland/pond (swamp). Black lines
represent site series polygon boundaries. The pink line is the perimeter of the
undisturbed terrain study site surrounding the landslide.

64

Figure 2-7. Site series percent area paired graphs for Beatton River, Cecil Lake,
and Hasler Flats study areas, with categories ordered from low (left) to high
(right) moisture regime.

66

Figure 2-8. Site series Kolmogorov-Smirnov two sample tests for the paired
study sites of each study area of Beatton River, Cecil Lake, Hasler Flats. For all
three tests, the critical value is greater than the D statistic, indicating no
significant difference.

70

Figure 2-9. Example of digitised geotyping on the Cecil Lake landslide.
Geotype codes (shown in blue) are PO = Pond/wet area, RI = Ridge, RA = Raft,
PI = Pillar, Fl = Flood deposits -seasonal, SW = Swale, GU = Gully. The black
lines are polygon boundaries.

71

Figure 2-10. Rank abundance curves comparing geotype diversity on landslides
and undisturbed terrain.

77

Figure 2-11. Relationships between relevé vegetation diversity and geotype
diversity in terms of richness, Shannon diversity, Pielou’s evenness, and
Simpson’s diversity for all three landslide study areas. Blue dots represent
landslide sites and orange dots represent undisturbed sites for each study area.
ix

80

Regression lines are not shown, as no significant regression relationship was
found for any of the comparisons.
Figure 3-1. Sampling transects with a series of 1 x 1 m quadrats along a
measuring tape. Image on the left is Beatton River undisturbed transect BEtu1a.
(Photo August 26, 2018) Image on the right is Beatton River landslide transect
BEtl1a. (Photo July 18, 2017, by P. Burton. Used with permission.)

129

Figure 3-2. Landslide and undisturbed transect beta diversity (Bw) regression:
vegetation vs environment. Blue symbols represent Cecil Lake transects, green
symbols represent Beatton River transects, and orange symbols represent Hasler
Flats transects. The blue dashed line is the regression line for landslide transect
beta diversity (Bw). There was no significant regression for the undisturbed
transects.

140

Figure 3-3. Cecil Lake landslide transect CEtl3c, showing high diversity of
habitats as illustrated by Alnus spp. next to a pond, with sparsely vegetated southfacing slopes on the other side of the pond. (Photo August 25, 2018)

144

Figure 3-4. Cecil Lake landslide transect CEtl1, showing lower diversity of
adjacent habitats as illustrated by expanse of level ground and relatively
homogeneous vegetation. (Photo August 25, 2017)

144

Figure 3-5. Sweetclover (Melilotus spp.) carpeting the toe of the Beatton River
landslide. Transect is BEtl4. (Photo August 16, 2017)

146

Figure 3-6. Vegetation diversity on the Hasler Flats landslide, illustrated by
dense layers of varied forbs, short shrubs, and tall shrubs. Transect is HAtl1a.
(Photo July 28, 2018)

147

Figure 3-7. Invasive/exotic plants Lactuca serriola (prickly lettuce), Sonchus
spp., Melilotus spp., and Elymus repens (quackgrass) on the Beatton River
landslide. Transect shown is BEtl1. (Photo August 17, 2017)

148

Figure 3-8. Example of moist cottonwood (Populus balsamifera ssp.
balsamifera) habitat in undisturbed terrain at Beatton River. Transect is BEtu1a.
(Photo August 26, 2018)

149

Figure 3-9. Relatively homogeneous vegetation (young aspen – Populus
tremuloides) in undisturbed terrain at Hasler Flats. Transect is HAtu1a. (Photo
August 10, 2018)

150

Figure 3-10. Beatton River landslide transect BEtl6, showing microsite diversity
of hummocks, depressions, ponds, and fissures. (Photo taken September 5, 2018)

153

Figure 3-11. Hasler Flats undisturbed transect HAtu3 showing broken terrain.
Transect traverses uneven terrain and small fissures for the first 15 m
(foreground), then ascends a steep slope for 8 m (visible in the distance), and
x

155

finally levels out (evident in the photo where the sky shows through). (Photo
August 3, 2018)
Figure 3-12. Cecil Lake undisturbed transect CEtu2 showing uniform, level
terrain. (Photo August 17, 2018)

156

Figure 4-1. Landslide ponds study area – Mapsheet 94A (1:250,000) Peace River
Region of northeastern British Columbia, Canada. The yellow areas indicate
history of landslides along the Peace River and its tributaries, based on shape file
linework by Severin (2004). Shape file linework imagery reproduced with
permission.

169

Figure 4-2. Sample of landslide pond mapping. Location is Cecil Lake. Ponds
are outlined in blue. Yellow lines indicate landslide features (previously mapped
by Severin 2004). Severin landslide shape file lines used with permission.

172

Figure 4-3. Simplified anatomy of a landslide showing head, body, and toe.
Adapted from Cruden and Varnes 1996. Credit: Transportation Research Board.
1996. Landslides: Investigations and Mitigation. Special Report 247.
https://doi.org/10.17226/11057. Reproduced with permission from the National
Academy of Sciences, Courtesy of the National Academies Press, Washington,
D.C.

174

Figure 4-4. Frequency distribution graphs showing landslide pond size
distribution in hectares (ha). The upper graph shows all landslide-generated
ponds in Mapsheet 94A, while the lower graph shows the size distribution for
landslide ponds ≤ 1.00 ha.

180

Figure 4-5. Total number of landslide ponds per geomorphic location.

181

Figure 4-6. Total landslide pond area (ha) by geomorphic location.

181

Figure 4-7. Mean pond size (ha) by geomorphic location on landslide.

182

Figure 4-8. Paired graph showing overall number of landslides per landslide type
compared to number of pond-bearing landslides per landslide type. Landslide
type names have been abbreviated to accommodate graph. Type abbreviations:
CO = Compound; EF = Earth flow; MF = Mobile flow; MR = Multi-level
rotational; RA = Ravelling; RE = Retrogressive rotational; RO = Rotational; SR
= Shallow retrogressive. Note: No ponds were found on Ravelling (RA) or Earth
flow (EF) landslide types.

183

Figure 4-9. Number of ponds per landslide type.

184

Figure 4-10. Pond area (ha) in each pond-bearing landslide type.

184

Figure 4-11. Mean pond size (ha) per landslide type.

185

xi

Figure 4-12. Number of ponds on head per landslide type.

186

Figure 4-13. Pond area (ha) on head per landslide type.

187

Figure 4-14. Number of ponds on body per landslide type.

188

Figure 4-15. Total pond area (ha) on body per landslide type.

189

Figure 4-16. Number of ponds on toe per landslide type.

190

Figure 4-17. Pond area (ha) on toe per landslide type.

191

Figure 4-18. Very new pond on landslide (Beatton River Landslide). The most
recent movement on the landslide occurred in 2015, so pond was approximately
three years old at time of photograph. (Photo June 19, 2018)

192

Figure 4-19. Newer pond on landslide (Beatton River Landslide). Pond is on an
older part of the landslide near the toe. (Photo August 9, 2017)

192

Figure 4-20. Persistent pond on landslide (Cecil Lake Landslide). Landslide
occurred in 1998, thus the pond was approximately 20 years old at time of
photograph. (Photo June 23, 2017)

193

Figure 4-21. Recent beaver gnawing activity on young sapling near pond (Cecil
Lake Landslide). (Photo September 7, 2017)

199

Figure 4-22. Beaver pond on the Cecil Lake Landslide. (Photo June 22, 2017)

199

Figure 4-23. Older beaver gnawing activity on the Cecil Lake Landslide. (Photo
September 10, 2017)

200

Figure 4-24. Recent beaver activity at the Hasler Flats landslide. Top image
shows well-used beaver path (bottom centre of photo) leading from the sidescarp
to a landslide pond. Lower image shows very recent felling of large aspen trees
(Populus tremuloides) by beavers just above the sidescarp. (Photos August 14,
2018)

201

xii

ACKNOWLEDGEMENTS

Financial support for this study was provided by the Province of British Columbia Ministry of
Forests, the BC Conservation Federation, Dr. Phil Burton’s NSERC (Natural Sciences and
Engineering Research Council of Canada) funds, and NSTP (Northern Scientific Training
Program) award funding. Further funding was made available through various UNBC student
awards: UNBC research project award, UNBC travel grant, the Susan Stevenson Memorial
Award, and the Al Nevison Award.
I would first like to thank my supervisors Dr. Phil Burton and Dr. Marten Geertsema for agreeing
to take me on as a student without firm funding in place and with only a kernel of an idea for a
PhD topic. Thank you to Phil for all his advice and enthusiasm, for sharing his vast knowledge
about disturbance ecology, and for always pushing me to reach further. Thank you to Marten for
his steady support and encouragement, for somehow always finding funds to keep me employed,
for opening the world of landslides up to me, and for providing me with interesting opportunities
to get out into the field and learn firsthand. I also thank my other Committee members Dr. Paul
Sanborn and Dr. Roger Wheate, and Richard Kabzems, MSc, PAg, RPF (Ret). I thank Roger for
his direct approach and patience with my GIS questions. Thank you to Paul for his good nature,
willingness to provide advice, and for sharing his wealth of knowledge on soils. And thank you
to Richard for sharing his expansive expertise regarding deciduous and mixedwood ecosystems
of the Peace River Region.
I would not have been able to complete this work without the assistance of those willing to share
their knowledge or provide a word of encouragement. Thank you to Ken Simonar for field
training and mentoring on site-level ecological sampling and classification, and to Benita Kaytor
for assistance with field data collection on Peace River landslides. Thank you to Kim Lutz for
assistance with PCORD and for her all-around support. I also thank Dr. Darwyn Coxson for
advice on PCORD. I appreciate the late Bruce Rogers for his willingness to answer all my
questions about ecological classification over the years. His presence and contributions will be
greatly missed. Thank you to Alexandre Bevington for help with my remote sensing and
technological questions. Thank you to Phil Burton, Richard Kabzems, Benita Kaytor, Ken
Simonar, Curtis Bjork, Nick Hamilton, and Marika Cameron for assistance with plant
identification. Finally, thank you to my dear friends who believed in me and encouraged me all
along, even when they didn’t understand the work I was doing. You know who you are.

xiii

DEDICATION

I dedicate this PhD Dissertation to my Mom and Dad, who instilled in me a love of the natural
world and the North, appreciation of simple things, a sense of humour regardless of life’s
situations, and perseverance in the face of adversity. God Bless.

~In loving memory of my father, Richard Kress, who left this world June 2, 2019~

xiv

Chapter 1. Introduction to landslide biophysical diversity research
1.1 Introduction
Throughout the world, landslides are known for their destructive impacts on ecosystems,
infrastructure, and human life (Geertsema et al. 2009). However, relatively little has been
studied regarding the contribution of landslides to ecosystem health and diversity over
various scales. Following a disturbance such as a landslide, succession is initiated, and new
plant communities begin to form. Many factors influence the pathways of succession,
including dispersal processes, substrate availability, nutrient and moisture levels, and
competition (Clements 1916; Gleason 1917; Gleason 1926; Connell and Slatyer 1977; Pickett
and White 1985). Landslides are unusual disturbances in that primary succession and
secondary succession occur simultaneously, due to the presence of biological legacies
(Walker and Shiels 2013). Successional processes can also be reset if the landslide
reactivates. Landslides are thus challenging to study.
This research aims to describe and compare biophysical diversity on landslides and in the
surrounding relatively undisturbed terrain in the landslide-prone Peace River Region of
northeastern British Columbia, Canada. The research is multidisciplinary, drawing on
theories, applications, and literature of landslide ecology, geomorphology, plant ecology,
remote sensing and spatial applications, ecosystem restoration and ecological engineering,
and hydrology.
This chapter will present an introduction to the research by first discussing the background
and context, followed by the research problem, the research aims, objectives and questions,
and the significance of the research work. The chapter ends with an outline of the structure
of the dissertation.
1

1.2 Background to the study
Globally, catastrophic landslides draw the attention of communities, governments, land
managers, and researchers. Landslides can be large and destructive and often occur suddenly
and rapidly (Cruden and Varnes 1996a). They can kill people and animals, destroy forests,
and seriously damage buildings and transportation infrastructure. Landslides can also alter or
obliterate habitat for fish and wildlife. The most common triggers of naturally occurring
landslides are intense rainfall, rapid snowmelt, alterations in water level, volcanic eruptions,
and earthquakes (Wieczorek 1996). Because of the impacts of landslides on nature and
society, study of these events is ever-increasing.
Broadly, landslides can be divided into prehistoric and historic types. Prehistoric landslides
are those that occurred in the period before recorded history and they can be thousands of
years old. Study of prehistoric landslides can help explain how and why a landscape formed
as it did and can help predict the nature and frequency of future landslides over space and
time. Prehistoric landslide evolution is analysed using signatures or clues left behind in the
landslide, such as organic material that can be radiocarbon dated. Historic landslides are
those that have occurred in the time since history was written down. It is usually easier to
determine the cause and triggers of these landslides, as they have not been subjected to
significant weathering or reworking. Historic landslides can also help predict future events,
although perhaps on shorter time scales.
The Peace River area features a unique combination of landforms due to various processes
that created and changed the landscape over different scales of time. The valley and
tributaries consist of thick layers of glaciolacustrine material from the former Glacial Lake
Peace (Hickin et al. 2015). Over thousands of years, deep postglacial incision carved out the
2

current valley, creating steep slopes along the major river systems. When present on steep
slopes, the glaciolacustrine material is unstable. The area has a long history of natural
disturbance caused by landslides, interspersed with periods of stability. Some of the
landslides were one-time events, while others appear to have reactivated, burying previous
slides to some extent. Some landslides are large, such as the Attachie landslide that blocked
the Peace River in 1976. There is evidence of many other landslides that are quite small
“slumps”. Above the river valleys there are rolling hills and plateaus.
Landslide ecology
Landslide events are both erosional and depositional processes, removing existing natural
features while at the same time creating new features on the landscape. These processes can
result in a unique diversity of soil, vegetation, and microsite types across the landslide
environment (Geertsema and Pojar 2007). Landslide ecology is a branch of landscape
ecology that investigates how plant and animal communities respond and interact as a result
of the changes caused by landslides. Some key components of study are nutrient cycling, soil
development, plant adaptations, dispersal and colonisation, new mixes of native and nonnative species, and successional trajectories. Studies address these relationships on multiple
scales of space and time. Landslide ecology also involves the integration of these biological
learnings into management practices for slope hydrology, soil erosion, and slope stability
(Walker and Shiels 2013).
Geomorphology
Geomorphology is the study of the physical or morphological features or properties
(landforms) of the Earth’s outer crust in relation to geological features (Schaetzl and
3

Thompson 2015). Geomorphic or landform processes involve mechanical transport of
organic and inorganic material (Swanson et al. 1988) and are strongly influenced by slope,
pore water pressure, and soil cohesion. Landforms influence air and ground temperature,
moisture availability, and nutrient availability at a site. They affect the flow of vegetative
propagules, energy, organisms, and material through a landscape. Landforms also can affect
the frequency and spatial distribution of other natural disturbances (e.g. fire) and control
geomorphic processes that alter biotic features. Concepts of geomorphology are central to
the present study, with its focus on geomorphic processes and recovery of landslides.
Plant ecology
Plant ecology is a subdiscipline of ecology that studies plant abundance and distribution,
effects of environmental factors on plant abundance, and interactions among and between
plants and other organisms (Keddy 2007). This discipline can be applied from the microsite
to landscape scales, and from seasonal to millennial temporal scales. Plant ecology plays a
central role in the present research, as the diversity of plant communities across the landslide
is documented and compared to the surrounding area. The theories and principles of the
discipline are incorporated as they relate to plant colonisation and abundance, successional
trajectories, deterministic microsite features, and the influence of other organisms on plants.
Remote sensing and spatial applications
Remote sensing refers to any type of technology that captures an image from a distance. The
most commonly used types of remote sensing are aerial photographs, satellite images, and
more recently LiDAR (light detection and ranging). In this study a combination of LiDARconstructed DEMs (digital elevation models), drone imagery, LANDSAT (land-use satellite)
4

imagery, GIS (geographical information system) applications, and aerial photographs were
used to identify, delineate, and analyse various components of biophysical diversity.
Ecosystem restoration and ecological engineering
Ecological or ecosystem restoration involves intentional activities by humans to start or
accelerate recovery of an ecosystem to maintain or enhance ecosystem health. In many
cases, the ecosystem has been damaged or destroyed by direct or indirect human activities. In
other circumstances, the ecosystem has been degraded by a natural agent such as wildfire,
flood, or landslide to the point where it is unable to recover to its historic successional
trajectory. Restoration efforts attempt to regain this trajectory. The restored ecosystem will
not necessarily recover its former state, developing instead along an altered trajectory
(Burton 1991; SER Working Group 2004).
In many industries, native plants are being used to help restore ecosystems damaged by both
human-caused and natural activities (Polster 1997; Stokes et al. 2009; Walker et al. 2009).
The selected plants help stabilise soils and slopes, control water flow, and increase wildlife
habitat, among many other services. This practice of applying engineering principles to
natural earth materials is referred to as geological engineering. The present study draws
heavily on the concepts, theories, and findings related to ecological restoration and
geological engineering when providing conclusions and recommendations.
Hydrology
Hydrology studies the pathways, distribution and storage, and quality of water, and includes
the hydrologic cycle (Schaetzl and Thompson 2015). Water is very important in landslide
processes. It can both trigger a landslide and transport debris and sediments downslope and
5

downstream. Soil strength properties are a function of soil water content (Meusburger and
Alewell 2008), and landslide occurrence is closely linked to pore water pressure thresholds.
Prolonged periods of rainfall often trigger landslides. The probability of slope failure is
determined by the balance between precipitation and evapotranspiration by plants (Van Beek
and Van Asch 2004), therefore vegetation plays a large role in this process. Mechanisms of
landslide movement can significantly alter the hydrology of a slope, making it more unstable
and changing soil development processes, which can also result in the formation of ponds
and wetlands. Hydrological features such as gullies can limit the spatial extent of a slide
(Geertsema et al. 2010). The present study incorporates hydrological principles and concepts
when assessing landslide initiation and evolution. It similarly employs these theories to
investigate persistence of water bodies formed by landslides.
1.3 The research problem
a) Current state of research
While the geological and geomorphological processes of landslides have been extensively
studied throughout the world, much less research has focused on the ecological processes. In
northern British Columbia, even the geological processes are not as well studied, and
ecological processes have not been researched to any extent. This presents a challenge to
landslide management, as it is important to understand the ecological processes of landslides
to fully appreciate landslide evolution and recovery over multiple temporal and spatial scales.
Some work has been done on plant succession and ecosystem recovery on landslides
(Francescato et al. 2001; Dale and Adams 2003), with several studies occurring in tropical
climates (Shiels and Walker 2003; Shiels 2006; Shiels et al. 2006; Restrepo et al. 2009).

6

These studies have helped build knowledge on topics such as the role of organic materials
and birds in landslide recovery. In recent years, landslide ecology research has focused on
the application of knowledge about landslide recovery processes to restoration practices
(Turner et al. 1998; Pickett et al. 2009; Walker and del Moral 2009; Walker et al. 2009).
These findings have contributed greatly to the advancement of a more holistic approach to
landslide management.
b) Literature gaps
Although there are increasingly more studies on landslide ecology and recovery, peerreviewed literature quantifying biophysical diversity on landslides is lacking. Even less
available is any research on landslide biophysical diversity in northern climates. In northern
British Columbia (BC), there visually appears to be a wider diversity of plant communities
and microtopography types on landslide surfaces as compared to the surrounding landscape
(Geertsema and Pojar 2006). However, other than a coastal study (Smith 1986) there have
not been efforts in the northern part of the province to quantify vegetative and environmental
differences between landslides and surrounding terrain and analyse these differences to
determine their significance. On a broader scale, studies incorporating this type of
biophysical diversity analysis in other parts of the world do not appear to exist. The present
study is the first known large-scale and comprehensive comparison of biophysical diversity
on landslides and surrounding undisturbed terrain.
c) Problem
With changing climate, landslides in BC are projected to become more frequent and have
greater magnitude, due to increased precipitation and degradation of permafrost (Geertsema
7

2006). It is expected there will be increased interest in prevention and mitigation of
landslides, as well as restoration of ecosystems altered by a disturbance such as this. A lack
of knowledge about succession, the plant assemblages that form, and the influence of
biophysical factors may hinder the ability to manage landslides and carry out conservation
measures. This research attempts to quantify these facets of landslide ecology to obtain a
better understanding and provide lessons for restoration.
1.4 Research aims, objectives, and questions
This research aims to quantify, describe, and analyse components of biophysical diversity on
landslides in the Peace River Region of northeastern British Columbia. The research further
aims to compare this vegetation and site diversity with surrounding, relatively undisturbed
terrain where possible, and provide some recommendations for restoration and land
management on landslides.
Research objectives
There are five main objectives of the study:
(1) To quantify and analyse biophysical diversity on landslides and compare diversity with
that found on the surrounding undisturbed terrain.
(2) To quantify, analyse and compare the distribution of site-level ecological classification on
landslides and on the surrounding undisturbed terrain.
(3) To quantify, analyse and compare spatial turnover (beta diversity) of plant species and
microsites on landslides and on the surrounding undisturbed terrain.

8

(4) To use the findings on biophysical diversity to assess whether geomorphic diversity is a
predictor of vegetation diversity.
(5) To quantify and analyse the presence and distribution of landslide ponds on an area of
the Peace River Region.
Research questions
There are five key research questions to be answered in this study:
(1) Are landslides demonstrably more biophysically diverse than undisturbed ecosystems?
(2) To what extent do landslides rearrange the relative abundance of site-level ecological
classifications on a slope compared to adjacent undisturbed terrain?
(3) What is the extent of turnover of microsite and plant species diversity on landslides, and
how does this compare to adjacent undisturbed terrain?
(4) Is vegetation diversity on landslides significantly related to geomorphological diversity?
(5) What is the distribution and abundance of landslide ponds at regional and local scales,
and what are the ecological and management implications?
A large portion of the research centres around three landslides in the Peace River Region,
presented in Figures 1-1 (Beatton River), 1-2 (Cecil Lake), and 1-3 (Hasler Flats). These
landslides are described in detail in Chapter 2.

9

Figure 1-1. Beatton River landslide. Top image is an oblique 3D presentation of an ortho mosaic draped over
drone imagery (flown in 2016), at vertical exaggeration 1X. Bottom image is an aerial view of the landslide,
illustrating temporary partial blockage of the Beatton River. (Aerial photo was taken October 20, 2015, by M.
Geertsema. Used with permission.)

10

Figure 1-2. Cecil Lake landslide. Top image is an oblique 3D presentation of World Imagery layer draped
over LiDAR imagery, vertical exaggeration 1X. Inset in top image is a photograph of the headscarp and part of
the landslide, facing west. Inset photo was taken May 12, 2016. Bottom image is an ortho mosaic presentation
of the landslide, showing variable topography and ponds.

11

Figure 1-3. Hasler Flats landslide. Top image is an ortho mosaic draped over drone imagery, vertical
exaggeration 1X. Bottom image is an aerial view of the landslide, facing south. (Aerial photo on bottom was
taken September 30, 2013, by M. Geertsema. Used with permission.)

12

1.5 Significance
At a time when researchers are more frequently collaborating across disciplines or
incorporating other disciplines into their research, the current study is truly a
multidisciplinary endeavour. Overall, this research will contribute to the body of knowledge
on disturbance ecology and restoration by describing and quantifying biophysical diversity of
landslides in northeastern BC and providing recommendations for land management. Key
contributions include measurement of alpha diversity and site-level ecological classification
on landslides and comparison with nearby undisturbed areas, and quantification of
geomorphological diversity on landslides with comparison to undisturbed terrain. An
additional advance in knowledge is the quantification and analysis of spatial turnover (beta
diversity) on landslides, which indicates the degree of differentiation among biological
communities. Further contributions include quantification and description of surface water on
landslides at a landscape level. All these learnings can enhance understanding of succession,
recovery, and restoration of landslides and similar disturbances. The research findings may
be especially useful in northeastern BC, where a new hydroelectric dam is under construction
on the banks of the landslide-prone Peace River.
1.6 Structural outline of the dissertation
In Chapter 1, the context of the research has been introduced and framed within the wider
realms of scientific endeavours. The research questions and objectives have been outlined,
and an argument for the value of this research has been presented. Finally, a structural
outline of the dissertation has been provided.

13

In Chapter 2, a detailed background on alpha diversity will be presented and methods will be
laid out regarding data collection and diversity analysis for vegetation BEC (biogeoclimatic
ecosystem classification) plots, relevés, site-level ecological classification mapping,
geomorphic type mapping, microtopography variation, and the multivariate ordination of
vegetation data. A detailed description of results from several lines of analysis will be
presented. The results will be discussed in the context of existing research literature.
Conclusions and recommendations on the findings will be provided, followed by a list of
references.
In Chapter 3, background on beta diversity (i.e. turnover) will be provided, followed by a
detailed presentation of the methods used for data collection and analysis for assessment of
beta diversity on a series of field transects. Results will be presented, which will then be
discussed in detail in the context of previous research and implications for management.
Conclusions and recommendations on the findings will be provided, followed by a list of
references.
In Chapter 4, background on landslide ponds will be presented and then detailed methods
will be laid out regarding data collection and analysis of landslide ponds over the area
covered by geographical mapsheet 94A (Charlie Lake). The results will be presented,
followed by a discussion of key findings. Conclusions and recommendations will finish the
chapter, closing out with references cited.
In Chapter 5, a summary of key findings and insights will be provided, followed by
implications and recommendations for future applications.
Finally, appendices will provide additional information to supplement the various chapters.
14

1.7 References
Burton, P. 1991. Ecosystem restoration versus reclamation: the value of managing for
biodiversity. Proceedings of the 15th Annual British Columbia Mine Reclamation
Symposium in Kamloops, BC, 1991. The Technical and Research Committee on
Reclamation. pp. 17-26.
Clements, F. 1916. Plant succession: an analysis of the development of vegetation.
Carnegie Institution of Washington publication 242.
Connell, J., and R. Slatyer. 1977. Mechanisms of succession in natural communities and
their role in community stability and organization. The American Naturalist, 111(982):
1119-1144.
Cruden, D., and D. Varnes. 1996. Landslide types and processes. In: Turner, A., Schuster, R.
(eds.), Special Report 247: Landslides Investigation and Mitigation. National Research
Council, Transportation Research Board, Washington, DC, pp. 36-75.
Dale, V., and W. Adams. 2003. Plant reestablishment 15 years after the debris avalanche at
Mount St. Helens, Washington. The Science of the Total Environment, 313:101-113.
Francescato, V., M. Scotton, D. Zarin, J. Innes, and D. Bryant. 2001. Fifty years of natural
revegetation on a landslide in Franconia Notch, New Hampshire, U.S.A. Canadian
Journal of Botany, 79: 1477-1485.
Geertsema, M. 2006. Hydrogeomorphic hazards in northern British Columbia. National
Geographical Studies 341. Utrecht. 184 p.
Geertsema, M., L. Highland, and L. Vaugeouis. 2009. Environmental impacts of landslides.
In Landslides – Disaster Risk Reduction. K. Sassa and P. Canuti (eds), Springer-Verlag.
Berlin, Heidelberg. 650 pp
Geertsema, M., and J. Pojar. 2007. Influence of landslides on biophysical diversity – A
perspective from British Columbia. Geomorphology, 89: 55-69.
Geertsema, M., J. Schwab, P. Jordan, T. Millard, and T. Rollerson. 2010. Ch. 8, Hillslope
processes In Compendium of Forest Hydrology and Geomorphology in British Columbia,
Volume2 of 2. Pike, R., T. Redding, R. Moore, R. Winkler and K. Bladon (eds). B.C.
Ministry of Forests and Range, Forest Science Program, Victoria, B.C. and FORREX Forum
for Research and Extension in Natural Resources, Kamloops, B.C. Land Management
Handbook 66.
Gleason, H. 1917. The structure and development of the plant association. Bulletin of the
Torrey Botanical Club, 44(10): 463-481.
Gleason, H. 1926. The individualistic concept of the plant association. Bulletin of the
Torrey Botanical Club, 53(1): 7-26.
15

Hickin, A., O. Lian, V. Levson, and Y. Cui. 2015. Pattern and chronology of glacial Lake
Peace shorelines and implications for isostacy and ice-sheet configuration in northeastern
British Columbia, Canada. Boreas, 44(2): 288-304. DOI 10.1111/bor.12110
Keddy, P. 2007. Plants and Vegetation: Origins, Processes, Consequences. Cambridge:
Cambridge University Press, First Edition. 706 p
Meusburger, K., and C. Alewell. 2008. Impacts of anthropogenic and environmental factors
on the occurrence of shallow landslides in an alpine catchment (Urseren Valley, Switzerland).
Natural Hazards and Earth System Sciences, 8: 509-520.
Pickett, S., M. Cadenasso, and S. Meiners. 2009. Ever since Clements: from succession to
vegetation dynamics and understanding to intervention. Applied Vegetation Science, 12(1):
9-21.
Pickett, S., and P. White. 1985. The Ecology of Natural Disturbance and Patch Dynamics.
Academic Press. San Diego, CA. 472 pp.
Polster, D. 1997. Restoration of landslides and unstable slopes: considerations for
bioengineering in Interior locations. Proceedings of the 21st Annual British Columbia Mine
Reclamation Symposium in Cranbrook, BC. The Technical and Research Committee on
Reclamation.
Restrepo, C., L. Walker, A. Shiels, R. Bussmann, L. Claessens, S. Fisch, P. Lozano, G. Negi,
L. Paolini, G. Poveda, C. Ramos-Scharrón, M. Richter, and E. Velázquez. 2009. Landsliding
and its multiscale influence on mountainscapes. BioScience, 59(8): 685-698.
Schaetzl, R., and M. Thompson. 2015. Soils: genesis and geomorphology. Cambridge
University Press, New York, NY.
SER Working Group. 2004. The SER International Primer on Ecological
Restoration.www.ser.org & Tucson: Society for Ecological Restoration International.
Shiels, A. 2006. Leaf litter decomposition and substrate chemistry of early successional
species on landslides in Puerto Rico. Biotropica, 38(3): 348-353.
Shiels, A., and L. Walker. 2003. Bird perches increase forest seeds on Puerto Rican
landslides. Restoration Ecology, 11(4): 457-465.
Shiels, A., L. Walker, and D. Thompson. 2006. Organic matter inputs create variable
resource patches on Puerto Rican landslides. Plant Ecology, 184: 223-236.
Smith, R. 1986. Soils, vegetation, and forest growth on landslides and surrounding
logged and old-growth areas on the Queen Charlotte Islands. BC Ministry of
Forests Land Management report 41.

16

Stokes, A., C. Atger, A. Bengough, T. Fourcaud, and R. Sidle. 2009. Desirable plant root
traits for protecting natural and engineered slopes against landslides. Plant Soil 324:1–30.
Swanson, F., T. Kratz, N. Caine, and R. Woodmansee. 1988. Landform effects on ecosystem
patterns and processes. BioScience 38(2): 92-98.
Turner, M, W. Baker, C. Peterson, and R. Peet. 1998. Factors influencing succession:
lessons from large, infrequent natural disturbances. Ecosystems, 1(6): 511-523.
Van Beek, L., and T. Van Asch. 2004. Regional assessment of the effects of land-use change
on landslide hazard by means of physically based modelling. Natural Hazards, 31: 289-304.
Walker, L., and A. Shiels. 2013. Landslide Ecology. Cambridge University Press. New
York, NY. 300 pp.
Walker, L., and R. del Moral. 2009. Lessons from primary succession for restoration of
severely damaged habitats. Applied Vegetation Science, 12(1): 55-67.
Walker, L., E. Velázquez, and A. Shiels. 2009. Applying lessons from ecological succession
to the restoration of landslides. Plant and Soil, 324(1-2): 157-168.
Wieczorek, G.F. 1996. Landslides: investigation and mitigation. Chapter 4 Landslide
triggering mechanisms. Transportation Research Board Special Report, (247).

17

Chapter 2. Biophysical alpha diversity of landslides and comparison with surrounding
undisturbed terrain
2.1 Introduction and Background
Landslides display a wide variety of sites, soils, and vegetation patterns compared to the
surrounding undisturbed landscape (Geertsema et al. 2006) and they stand out visually from
the adjacent terrain in terms of vegetation types and coverage, surface soils, and relief.
Extremes in surface roughness, nutrients, and moisture are evident over very short distances
on landslides.
Landslides are both erosional and depositional (Cruden and Varnes 1996). The erosional
process can remove deep layers of soil that have developed over centuries, effectively setting
the clock back for pedogenesis (Geertsema and Pojar 2007). Landslides mainly change soil
properties by exposing parent material, creating a variety of stages of pedogenic development
(Geertsema et al. 2009). Where once a Brunisol existed, there now may be an Orthic
Regosol. As material moves downslope, it becomes jumbled and turbated, resulting in
unique soil layers and buried organic material (Phillips and Lorz 2008). The depositional
process may transport large amounts of organic material from the slopes above, depositing it
in one place near the toe of the slide. These organic components contribute an influx of
nutrients in the depositional toe zone (Walker and Shiels 2013), and result in a loss of
nutrients from the erosional scarp zone. Tree uprooting on the landslide can affect soil
morphology, distribution of rock fragments, and evolution of regolith, at the same time both
creating and enhancing layering and vertical contrasts (Phillips and Lorz 2008). Ultimately,
erosion and deposition cause increases in extremes of nutrient regimes, facilitating a variety
of plant community development trajectories.

18

The processes of erosion and deposition may also create many distinct geomorphic
formations on the landslide. At the headscarp, erosion can result in steep, dry cliffs.
Material moving at a high speed can gain momentum and form ridges and pillars on the
landslide body as it comes to a stop (Geertsema et al. 2006). Subsequent erosion may create
deep gullies, and additional material can be deposited in flows. Weathering transforms
ridges and pillars over time, reducing surface roughness. In a rotational slide, the backward
tilting at times creates a depression where a sag pond can form (Takaoka 2015). Spreads
produce horsts and grabens (raised and lowered blocks, respectively) as the material pulls
apart between faults, and water may accumulate in the grabens. Small ponds may also form
anywhere there is a depression in the landslide surface. Overall, the topography formed by
landsliding can result in extremes of moisture. Very dry sites often abut very wet sites, and
very rich sites often lie close to very poor sites.
In regions where much of the terrain is inherently unstable, landslide processes and the
resulting variety of geomorphic landforms and site characteristics can create a very disturbed
landscape in terms of vegetation processes, soil processes, hydrology, and habitat. Habitat
diversity is strongly related to patterns of disturbance and recovery (Sousa 1984; Geertsema
et al. 2009; Walker and Shiels 2013). If both site and soil change, the soil changes may
persist much longer and have more profound ecological effects, such as the formation of
wetlands or the infilling of valleys.
Observations across northern British Columbia (BC), Canada, indicate landslides in this part
of the province are much more biophysically diverse than nearby undisturbed terrain
(Geertsema and Pojar 2007). However, biophysical diversity on landslides in northeastern
BC has not been measured or quantified in any substantial way. This chapter investigates
19

whether landslides are demonstrably more biologically and physically diverse than adjacent
undisturbed ecosystems. Four key questions were asked: 1) How does plant species
abundance and distribution differ on landslides compared to adjacent undisturbed terrain? 2)
To what extent do landslides rearrange the relative abundance of site-level ecological
classifications on a slope compared to adjacent undisturbed terrain? 3) Is landslide
geomorphology significantly more diverse than adjacent terrain, and 4) Is vegetation
diversity on landslides significantly related to geomorphological diversity?
This chapter assesses alpha diversity of vegetation and environmental sites (i.e. biophysical
diversity) within landslides in the study area and compares it with the surrounding
undisturbed terrain where possible. The chapter also investigates the possibility of
correlations between vegetation diversity and geomorphic diversity. Biophysical diversity in
this study refers to the variety of vegetation and physical sites; for vegetation this variety can
include both compositional and structural elements (Pitkanen 2000). It is hypothesised that
there is higher vegetation diversity, site-level ecological classification diversity, and
geomorphic diversity on landslides compared to the surrounding relatively undisturbed
terrain, and that geomorphic diversity positively influences vegetation diversity.
Because of the complexities and many theoretical layers involved in the study of diversity, it
is necessary to first examine the concept of biophysical diversity and its measurement to lay
the groundwork for this research chapter.
Alpha diversity
In general, diversity in nature can be described in a series of scales forming a hierarchy. The
three basic levels of diversity are alpha diversity, beta diversity, and gamma diversity
20

(Whittaker 1960). These three levels are related, and often, quantification of one level is
required before calculation of another. Alpha diversity refers to within-site diversity, while
gamma diversity is described at a regional level (resultant of alpha and beta diversities) and
beta diversity is generally understood as change in diversity between sites. Alpha diversity
addresses diversity at the stand or plot level and represents the range of species that may
interact with each other (Noss 1990; Magurran 2004).
Biological diversity or biodiversity represents the variability among organisms from all
sources and is essentially a comparative science (Magurran 2004). At different scales and
contexts, the biodiversity of plants describes the range of alleles or genotypes in a population,
the diversity of species or growth forms in a community, or the diversity of vegetation types
across a landscape (Noss 1990; Burton et al. 1992). Diversity can be partitioned into two
components: element richness and evenness (Simpson 1949). Species richness refers to the
number of species in a unit of study (McIntosh 1967). Evenness is the variability in species
abundances: the more equal the proportion of abundances between species, the higher the
evenness, and the more diverse the community (Magurran 2004). Additionally, abundance is
a surrogate measure of niche size, and statistical models assume that abundance is in some
way related to a species’ ecological importance (Magurran 2004). Alpha diversity describes
richness and evenness of species of a particular stand or community or group of organisms.
These concepts and measurements of diversity can be applied to various physical elements of
ecosystems as well, including vegetation structure and substrates.

21

Measurement of diversity
The measurement of species diversity is based on three assumptions: 1) all species are equal,
2) all individuals are equal, and 3) species abundance has been recorded using appropriate
and comparable units (Peet 1974). Diversity statistics are classified as either species richness
measures (McIntosh 1967), or heterogeneity measures which combine richness and evenness
components (Good 1953). Evenness measures assess the departure of the observed species
abundance pattern from the expected pattern in a hypothetical assemblage (Lloyd and
Ghelardi 1964).
Measurement of species richness
In its simplest form, species richness is the total number of species in a sampling area.
Species richness can be estimated from samples through species-area curves, parametric
methods, and nonparametric estimators (Magurran 2004). Species-area curves are a wellknown measure and plot the cumulative number of species recorded as a function of
sampling effort (Colwell and Coddington 1994). These curves can be extrapolated to give an
estimate of total richness of the assemblage, and they illustrate the rate at which new species
are found. Species richness can also be estimated using jack-knife statistics or bootstrapping,
resampling methods that only require incidence data.
Differences in species richness between samples can be assessed in various ways. Richness
estimates can be compared by using richness estimators such as Chao1 (non-parametric
estimator that takes into account rare species) or ACE (i.e. abundance-based coverage
estimator) to deduce overall minimum estimates (Chao 1984). A more accurate method of
assessment of differences is rarefaction, which uses the information provided by all species
22

collected to estimate the richness of a smaller sample. The two samples can then be
compared directly. However, rarefaction is computationally taxing. It also assumes that
individuals are randomly dispersed, which means that richness in clumped communities will
be overestimated. Rarefaction curves converge at small sample sizes, so sampling size must
be sufficient to characterise the community (Gotelli and Colwell 2001). Rarefaction did not
appear suitable for the purposes of the present study due to the prevalence of clumped
communities and relatively small sample size; therefore, it was not used.
Measurement of species abundance/evenness
Many different species abundance models have been devised to describe the relationship
between the number of species and the number of individuals of each species. Models
include the geometric series (Motomura 1932) and Fisher’s logarithmic series (Fisher et al.
1943). Some models are better than others for showing species abundance distributions, but
none are equally applicable to all assemblages due to local variations and the dependence on
local influencing factors (Magurran 2004). However, distributions generated by models can
still provide insight into processes determining biodiversity, because of the linkage of
abundance of species with successful competition for limited resources.
One of the most common and useful methods of displaying abundance data is the
rank/abundance plot. Species are plotted sequentially from most to least abundant on the
horizontal (x) axis. Abundances are usually shown in log10 format on the y-axis to capture
the full range of abundances of different species. Data sets can also be displayed as
percentages or proportions to allow for comparison between samples (Whittaker 1965). The
rank/abundance plot distinctly highlights contrasting patterns of species richness and

23

differences in evenness among assemblages and can be very useful for showing changes
following a disturbance (e.g. Bazzaz 1975). The shape of the rank/abundance plot can be
used to infer which species abundance model best describes the data.
Diversity indices
In addition to the various measures of species richness and species abundance, there are also
several diversity indices that have been developed. A diversity index is a single statistic that
incorporates information on both richness and evenness components and is essentially a
measure of heterogeneity (Good 1953; Hurlburt 1971). The weighting that is assigned to one
component relative to the other can markedly affect the level of diversity calculated and the
way sites or assemblages are ranked (Magurran 2004). Because each diversity index
emphasises either the richness or evenness component of diversity, there is no single perfect
index.
Species diversity indices can be used to compare communities, but different measures may
produce different rankings of sites. There are both parametric and nonparametric measures
of diversity. Parametric measures or indices depend on distribution of species abundances
while nonparametric measures do not. One of the most well-known nonparametric diversity
measures is the Shannon index or Shannon’s diversity index, H’ (Shannon 1948; Magurran
2004; Ortiz-Burgos 2016), which assumes that individuals are randomly sampled from an
infinitely large community, and that all species are represented in the sample. The Shannon
index estimates species diversity in a community by considering the number of species and
their relative abundances (evenness). It represents the degree of uncertainty in predicting the
species of a given individual selected at random from a community. The Shannon index H’

24

usually ranges from 1.5 to 3.5 in temperate zones (Magurran 2004). The higher the H’ value,
the higher the diversity in a particular community. Some disadvantages of the Shannon
index are that it is constrained and usually yields low numbers (Magurran 2004), and it can
be difficult to interpret since it confounds richness and evenness.
Simpson’s index, D (Simpson 1949) provides an alternative method of describing diversity.
It calculates the probability that any two individuals drawn at random from an infinitely large
community will belong to the same species (or some other category). As Simpson’s index D
decreases, diversity increases. Simpson’s index is heavily weighted towards the most
abundant species in the sample, so is less sensitive to species richness (Magurran 2004). It is
one of the most meaningful and robust diversity measures, as it captures the variance of the
species abundance distribution. Simpson’s index is also much less sensitive to sample size
than the Shannon index. A variation on Simpson’s index D is Simpson’s index of diversity,
1-D (used in this chapter), also known as the Gini-Simpson index, which represents the
probability that two individuals randomly selected from a sample will belong to different
species or categories. In this second formula, the greater the value of 1-D, the more diverse
the sample. Both Simpson measures are on a scale of 0 to 1.
Pielou’s evenness, J (Pielou 1966) measures the relative abundance of the different species
making up the richness in an area (Magurran 2004). As evenness increases, so does
diversity. A community where just a few species are dominant is less diverse than one where
the abundances are more evenly distributed among several species. Evenness is reported on a
scale of 0 to 1.

25

Hill (1973) devised a way of describing the relationship between indices by defining a
diversity index as the reciprocal mean proportional abundances and classifying according to
the weighting the indices give to rare species. Hill related this classification to the fact that
diversity measures emphasise either species richness or dominance. The conclusion about
whether one site is more diverse than another can thus depend on the choice of diversity
measure.
Hill numbers (qD) are a parametric class of true diversity measures that integrate species
richness and species abundances and represent a hierarchy of diversity values. Essentially,
Hill numbers are the ‘effective number of species’ or ‘species equivalents’ (MacArthur 1965,
1972), representing the number of equally abundant species that would be needed to give the
same value of a diversity measure such as the Shannon index or Simpson’s index. Hill
numbers present a simplified interpretation of results, since the units always denote the
effective number of species, regardless of position in the hierarchy (Morris et al. 2014). The
parameter q determines the sensitivity of the measure to the relative abundances and
quantifies how much the measure discounts rare species (Chiu and Chao 2014). The main
Hill numbers (q) are q = 0, q = 1, q = 2, and q = ∞. Hill number q = 0 is simply richness and.
Hill q = 1 is the exponential of the Shannon index and corresponds to the weighted harmonic
mean of the species’ proportional abundances. Hill q = 2 is the inverse of Simpson’s
concentration index and is associated with the weighted arithmetic mean of abundances. Hill
q = ∞ represents infinity. As q nears infinity, the weighted generalised mean with exponent
q-1 approaches the maximum proportional abundance of the most abundant species in the set
of data. The formula for Hill number calculations is provided in Section 2.3.2.3 of this
chapter.

26

In ecology, diversity measures are normally calculated on data for living organisms, but they
can also be applied to any set of measurable categories of items, such as mapped units of
landscape features (Nagendra 2002; Ricotta and Avena 2003). These calculations are
generally less computationally intensive than for species, but the results still provide a robust
measurement of diversity.
2.2 Study areas
The regional area of interest is the Peace River Region of northeastern BC, on landslides
<100 years old in glaciolacustrine sediments. The area was subjected to advances of the
Laurentide Ice Sheet on at least three separate occasions, with the most recent retreat
occurring more than 27,400 years ago (Mathews 1978; Mathews 1980). Evidence of glacial
events is presented in interglacial fluviatile gravel units. The study area is mostly within the
Alberta Plateau of the Interior Plains Region subdivision of the Canadian physiographic
classification system, and it is drained by the Peace River (Holland 1976). The Interior Plains
are east of the Rocky Mountain Foothills and consist of plateaus, plains, prairies, and
lowlands. The Plains are underlain by sedimentary rocks chiefly of Cretaceous age,
primarily of the Fort St. John Group with thick series of shales and sandstones near the top.
This area is also comprised of the Dunvegan Formation, which is hard cliff-forming
sandstone, and the Smoky Group, which is interbedded shales and sandstones. A small
portion of the study area is within the Rocky Mountain Foothills subdivision of the Eastern
System of the Canadian Cordillera (Holland 1976). The Foothills are underlain completely by
sedimentary rock, mainly from the Mesozoic age and consisting of a variety of limestones,
siltstones, sandstones, and shales. The Foothills were covered by continental ice during the
Pleistocene.
27

The regional study area is completely within the Boreal Forest region of Canada and the
Boreal Plains ecoregion of BC (Demarchi 2011) and is also entirely within the Boreal White
and Black Spruce moist warm (BWBSmw) biogeoclimatic unit, as defined using the
Provincial BEC (Biogeoclimatic Ecosystem Classification) guidelines (DeLong et al. 2011).
The climate of the area is continental, with low annual precipitation (Chilton 1981). Winters
are cold and long, with frequent inputs of continental arctic air. Summers are warm and short
but have long daylight hours that benefit agriculture. The most common soils in the area are
Grey Luvisols, but Luvic Gleysols, Eutric Brunisols, Chernozems, Solods, Organic Soils, and
Regosols are also present (Valentine 1978; Lord and Green 1986).
Following a series of field reconnaissance inspections, three landslides and their surrounding
undisturbed terrain were selected as smaller study areas within the regional area of interest:
these are known as the Beatton River, Cecil Lake, and Hasler Flats Landslides (Figure 2-1).
Each of these study areas consisted of two paired study sites: the landslide itself and a
delineated equivalent portion of the immediate surrounding undisturbed terrain. All three
landslides occurred near tributaries of the Peace River. The Peace River and its tributaries
have a long history of extensive and recurring slope instability. In addition to the above
criteria, the landslide study areas were chosen based on local knowledge, safe and efficient
access, and the presence of surrounding relatively undisturbed terrain at least equal in area to
the landslide.

28

Figure 2-1. Study areas of Beatton River, Cecil Lake, and Hasler Flats landslides, shown in Google Earth Pro.
The landslides are in the Peace River Region of northeastern British Columbia, Canada. Landslide locations are
outlined in yellow and identified with red locator balloons. Inset map shows landslide locations at a provincial
scale and in relation to Mapsheet 94A, which is outlined in red. BE = Beatton River landslide, CE = Cecil Lake
landslide, and HA = Hasler Flats landslide.

The Beatton River landslide was the youngest disturbance, to a large degree a reactivation of
older landslide deposits, with the most recent movement occurring in 2015. It is a rotational
earth slide / earth flow and has somewhat more complex stratigraphy than the Cecil Lake or
Hasler Flats slides, including exposed bedrock. The Beatton River landslide (56°21’57” N
120°42’26” W) is approximately 30 ha in size, with an elevation range of 655 m to 447 m (δ
29

208 m) and an average slope of ~17.2°. It is northeast facing, situated below (east of) a
cultivated hay field and between mature forest stands to the north and a grassland/mature
forest mix to the south. When the landslide occurred it temporarily blocked a portion of the
channel of the Beatton River to the east.
The Hasler Flats landslide is the smallest slide at approximately 1.5 ha. It is a southwestfacing spread which occurred in 2013 and is situated in post-glacial lake sand (4 m in
thickness) over silty clay. It is below a young, regenerated deciduous cutblock to the east
and between mature deciduous stands to the north and south. The Hasler Flats landslide
(55°36’39” N 122°0’45” W) has an elevation range of 615 m to 598 m (δ 17 m) and an
average slope of ~6.3°. It is adjacent to the Pine River to the west, and the slide deposited
debris into the river channel when it first occurred. This small landslide was chosen due to
its distinctive horst-graben formation and small size, two features which could be studied and
contrasted against the other two landslides. It is also conveniently located along the highway
west of Chetwynd and had easy access.
The Cecil Lake landslide was both the oldest and the largest of the three selected landslides.
It is a spread with a rupture surface in glacial lake sediments underlying glacial till and is
largely comprised of silty clay. The Cecil Lake landslide (56°23’48” N 120°38’11” W)
occurred in 1998 and is approximately 56 ha in size. It has an elevation range of 665 m to
556 m (δ 109 m) and an average slope of ~9.6°. It is a north-facing landslide less than a
kilometre east of the Beatton River, and at its north boundary (i.e. toe) it contains a narrow
but deeply incised unnamed creek that empties into the river. A small strip north of the creek
was included in the landslide study site, as there was disturbance that appeared to be the same
age as the rest of the landslide. The Cecil Lake landslide lies below a cultivated field and
30

mature deciduous and mixedwood forests at its south end where the headscarp is. To the east
is an older landslide, while to the west the Cecil Lake landslide abuts mature mixedwood
forests. To the north above the creek is a steep south-facing mosaic of grasslands and
mixedwood forests.
2.3 Methods and Analysis
2.3.1 Subjectively placed BEC plots
2.3.1.1 Field sampling of subjectively located BEC plots
In the first sampling season (2016), 91 subjectively placed 50 m2 plots were established on
the three landslide study sites (Beatton River, Cecil Lake, Hasler Flats) using the Provincial
BEC (Biogeoclimatic Ecosystems Classification) field sampling methodology (BC Gov
2010, 2015), modified to accommodate smaller plots. In addition, one 400 m2 (20 m x 20 m)
benchmark mature forest plot each was established in the surrounding undisturbed terrain
study site at Beatton River and Hasler Flats, along with one 50 m 2 benchmark grassland plot
in the Beatton River undisturbed terrain study site. In 2017, an additional 21 50 m2 plots on
the Cecil Lake landslide and one 400 m2 mature forest plot on the surrounding undisturbed
terrain were sampled to complete coverage of the Cecil Lake study area, as seasonal
limitations in 2016 ended the sampling prematurely. The objective of this sampling program
was to assess and describe the variation in plant communities and site characteristics present
on each landslide study site. The intent was to describe as much of the variability as possible
through field reconnaissance across the landslide. In total 116 plots were sampled across the
three study areas: 30 plots at Beatton River, 30 plots at Hasler Flats, and 56 plots at Cecil
Lake.

31

Data were recorded on FS882 field cards normally used in British Columbia BEC field
sampling (BC Gov 2010). Procedures followed the Describing Ecosystems field guide (BC
Gov 2010, 2015) and provincial terrain classification guidelines (Howes and Kenk 1997).
The Biogeoclimatic Ecosystem Classification (BEC) system is a hierarchical site-level
ecological classification method developed in BC that combines climatic, vegetation, and site
classifications (Pojar et al. 1987; Meidinger and Pojar 1991). The BEC system provides a
framework for organising ecological information and ecosystem management learnings,
creates common terminology among forest resource managers, and is used to prescribe and
monitor treatments at the site level (DeLong et al. 2011). The biogeoclimatic subzone is the
basic unit used in climatic classification and is a group of ecosystems that have developed
under the influence of the same regional climate. Each subzone has a distinctive sequence of
related ecosystems ranging from dry to wet sites, influenced not only by the regional climate,
but also by local soil and topographic features. The site series is the basic unit of site
classification and is defined by using late seral or climax vegetation. Site series represent site
units with similar environmental properties and potential vegetation. An eight-class scale has
been developed, based on relative soil moisture regime, relative soil nutrient regime, and
various other environmental factors. The standard size of plot used in BEC sampling is 400
m2, or 20 m x 20 m.
Sampling intensity and location of the subjective BEC plots was based on vegetative and
geomorphic differences observed in the field, with plot locations determined while in the
field traversing the study sites. Plots on the landslide were either circular or rectangular,
depending on the extent and configuration of the geomorphic/vegetation type. Most plots
were circular (i.e. 50 m2 – 3.99 m radius). If plots needed to be rectangular to capture a

32

microsite, they were configured to cover an area of 50 m2. The smaller plot size of the
subjective plots was chosen due to the broken-up and small-scale nature of many microsites
on the landslide. Plots in the surrounding undisturbed terrain were generally larger, as more
extensive areas of distinct plant communities were present. At each plot, vegetation, soil,
and environment information was recorded and representative photographs were taken. Plant
species were identified and recorded using BC’s seven letter coding system (BC Forest
Service 2016), and abundance was described in units of percent cover. Plant species
identification was aided by consulting with specialists as well as several published and online resources (i.e. MacKinnon et al. 1992; Johnson et al. 1995; Kershaw et al. 1998; Douglas
et al. 1998-2002; Klinkenberg 2021). For soils, information on geology, terrain, and organic
and mineral horizons/layers was recorded. Environment information included microsite
slope, microsite aspect, plant community structural stage, successional status, mesoslope
position, surface substrate composition, topography, moisture regime, nutrient regime, and
elevation. Soils were classified according to national standards (Soil Classification Working
Group 1998).
An initial attempt was made to classify each sample to BEC site series using the appropriate
guidebooks, recognising that some sites are a complex of multiple site series, as is also often
noted when mapping post-logging site types. Forested ecosystems or precursors to these
types of ecosystems were classified using the BWBS zone field guidebook (DeLong et al.
2011). Wetland and water features were classified using the Wetlands of British Columbia
guide (MacKenzie and Moran 2004) and Technical Report 68, Biogeoclimatic Ecosystem
Classification of Non-Forested Ecosystems in British Columbia (MacKenzie 2012).
Grasslands were also classified using the MacKenzie (2012) guide. Rationale for
33

classifications was supported by field photos and field notes, as well as analysis of and
comparison with the digital imagery. For all classifications, it was recognised that landslide
sites are seral or in a state of successional development as a precursor to supporting the
mature plant community likely to develop on each site.
The 2016 data were entered into the VPro [VENUS (Vegetation and Environment NexUS)
PROfessional] database program (MacKenzie and Klassen 2004), which is used primarily by
BC research ecologists and contains data fields that mirror the FS882 cards. The program
allows transfer of entered data to a spreadsheet and generation of various reports for analysis.
Data can also be directly exported for statistical analysis in ecological multivariate analysis
programs such as PC-ORD (PC-ORD 2015). The 2017 Cecil Lake plot data were entered in
an Excel spreadsheet, due to some technical problems with VPro.
The subjectively located BEC 50 m 2 and 400 m2 plots were originally intended to gather
general baseline information about plant and site characteristics of the study sites but were
ultimately central in some analysis components of the research. Although most of these plots
were smaller than the random 20 m x 20 m relevé main plots and were established using
different criteria, and thus could not be used for direct comparison in some parts of the study,
the plot information was instrumental in analysing plant and site diversity and mapping out
site series/types. The subjectively located plot information was used to assist in mapping out
polygons to describe site series/types and terrain diversity. Other key analyses using these
plots were a nonmetric multidimensional scaling (NMS) ordination of the vegetation and
selected environmental data and calculation of microtopographic variability. The plots could
eventually be registered in the Provincial BEC database (BECWeb 2023) if the smaller size
and variable configuration are acceptable.
34

2.3.1.2 Nonmetric multidimensional scaling (NMS) ordination of BEC plot plant species and
environmental variables
Nonmetric multidimensional scaling (NMS) ordination (Kruskal 1964; Mather 1976) was
used to analyse vegetation and selected environmental data collected on the subjectively
located BEC plots, to identify and describe any patterns in species composition in relation to
environment. NMS is especially suited to heterogenous ecological datasets such as this one,
because it does not require any specific design or model form, and so avoids parametric
assumptions (Peck 2016). The multivariate analysis program PCORD v. 7 (McCune and
Mefford 2016) was used to run the NMS ordination. Plant species and abundance data were
loaded from a spreadsheet into the main matrix. Values for five select environmental
variables were loaded into the secondary matrix. A random starting configuration was
employed. The Gower distance measure was used, as some plots had no vegetation (as
indicated by empty rows of data). The PC-ORD program user guide, citing previous works
(Gower 1971; Legendre and Legendre 1998), notes that the Gower distance measure is a
flexible, universal measure and is suitable for data sets containing empty rows, but not empty
columns. Three runs with real data were carried out on the dataset. A Mantel test was
performed to assess redundancy between each pair of runs. The dimensionality (i.e.
appropriate number of axes) of the data was assessed by performing a stress test on autopilot
mode at “thorough” setting and six axes, using 250 runs each of random and real data to
determine the best solution. The stability of the final solution was assessed by checking to
see if the stress leveled out and plateaued over time. The proportion of variance of each axis
was calculated based on the r 2 (i.e. coefficient of determination) between distance in the
ordination space and distance in the original species space.

35

The plant species most strongly associated with the environmental variables of interest were
identified using Pearson correlations with the axes of the ordination. The five environmental
variables assessed in the NMS ordination in the secondary matrix were soil moisture regime,
mesoslope position, slope gradient, heat load index (HLI), and material origin. Heat load
index and material origin variables were not specifically measured in the field but are derived
variables, incorporating information from the field data.
Heat load index is a measure of heat on a site based on slope, aspect, and latitude (McCune
and Keon 2002). The index is derived from direct incident radiation, which is the radiant
solar energy that hits the earth’s surface (Belessiotis and Delyannis 2011). To get a true
representation of the heat potential for each plot, the aspect was “folded” about the northeastsouthwest line using the formula ABS(180-ABS(Aspect -225)) provided in McCune and
Keon (2002). The following formula (Equation 3 in the 2002 paper) was then used to
calculate heat load index for each plot:
HLI = 0.339+0.808*COS(RADIANS(latitude))*COS(RADIANS(slope))0.196*SIN(RADIANS(latitude))*SIN(RADIANS(slope))-0.482*COS(RADIANS(folded
aspect))*SIN(RADIANS(slope))
where latitude is the site’s location in degrees north of the equator, slope is in degrees, and
folded aspect is calculated from aspect (in degrees azimuth from North) as shown above.
The material origin variable was created to categorise the level of soil development on each
plot. Five classes were identified, based on information obtained from the field data and
photos. Class 1 was the most stable level of soil material development, representing mature
in situ material in the surrounding terrain. Class 2 represented intact mature material on the
36

landslide that had been rafted from the surrounding terrain. Class 3 was material on the
landslide body that had an intermediate level of soil development, with no A horizon and a
weakly developed B horizon. Class 4 represented Orthic Regosols, having only a C horizon.
Class 5 was ponds, with arrested development of soil, which in most cases was likely only a
C horizon. Appendix 1 provides a more detailed description of each class.
2.3.2 Randomly located 400 m2 relevés
2.3.2.1 Field sampling of randomly located 400 m2 relevés
In the 2017 and 2018 sampling seasons, three randomly placed 400 m2 (20 m x 20 m) square
relevé plots were established on each landslide study site (Beatton River, Cecil Lake, Hasler
Flats) to sample vegetation and environment components. Plot locations were determined
beforehand using a grid system and random number generator. The same plot establishment
and sampling procedures were followed for an equivalent area of the surrounding
undisturbed landscape, for a total of six random relevés per study area, with three relevés on
each of six paired study sites. In contrast to the subjectively located plots, the objective of
this particular sampling program was to randomly sample vegetation and site conditions to
compare and contrast mean diversity on the landslides and in the surrounding undisturbed
terrain.
Data collection methods for these plots were very similar to those used on the subjectively
located 50 m2 and 400 m2 plots, following the same BEC field manuals cited in Section
2.3.1.1. For the sake of clarity and simplicity, however, the random plots will be henceforth
referred to as relevés and the subjectively located plots will be referred to as BEC plots for
the remainder of this paper. At each 20 m x 20 m relevé, vegetation, soil, and environment
37

details were recorded on FS882 field cards and representative photographs were taken.
Environmental information was collected on slope, aspect, plant community structural stage,
successional status, surface substrate composition, topography, mesoslope position, soil
moisture regime, soil nutrient regime, and elevation. Vegetation cover was tallied by percent
abundance and plant species were recorded using BC’s seven letter coding system (BC Forest
Service 2016, 2020). Plant species were identified using the same resources as in Section
2.3.1.1. Individual tree species, diameters and heights were also recorded. Information was
collected on geology, terrain, and organic and mineral horizons, but with a reduced focus on
soils. Field data were entered into Excel spreadsheets.
2.3.2.2 Rank abundance curves of plant species in relevés
Analysis of the relevé plot data was carried out in Excel. The mean relative abundance per
species was calculated for the three relevés on each of the six study sites, and then the species
were ordered from greatest to least percent cover. Mean rank abundance curves for the
relevés were graphed for each study site, plotting the log base 10 of ranked plant cover
abundance for each species. Individual curves were also plotted for each relevé. These rank
abundance plots were examined and compared among study sites and between landslides and
undisturbed terrain using the slope of a linear regression. The rank abundance method does
not require a goodness of fit test, but rather equates diversity of the assemblage with the slope
of the relationship, which reflects evenness of abundances. The steeper the slope, the lower
the diversity of the sample, since the higher ranked species have much greater abundances
than the lower ranked species. The richness of the sample is represented by the number of
species on the horizontal axis.

38

2.3.2.3 Plant diversity indices and measures for relevés
To assess alpha diversity on the landslide and compare within-slide alpha diversity with
undisturbed terrain, plant species richness and various abundance measures were calculated
for the relevé plot data of each landslide (n=3) and compared to diversity values obtained
from the three relevés on the surrounding terrain. Species richness (S = total number of
species), the Shannon index [H’ = -∑pi*ln(pi), where p is the proportional abundance of each
species], Simpson’s index of diversity [1-D, or 1-∑(n/N)2, where n is the total abundance of a
particular species and N is the total abundance of all species], and Pielou’s evenness value [J
= H’/ln(S)], were calculated.
To compare plant diversity between landslide and undisturbed sites for the relevés, two types
of values were calculated. First, the overall diversity values were calculated by obtaining the
mean cover value of each plant species over all three relevés for each study site. These
means were then used to calculate richness, Shannon index of diversity, Pielou’s evenness,
and Simpson’s index of diversity (1-D) for the study site. The second set of values was
obtained by first calculating diversity values for each individual relevé in a study site, and
then using these results to calculate the mean diversity values and the standard deviations for
the study site.
In addition, true diversity Hill numbers q = 0, 1, 2, and ∞ were calculated for the plant
species composition of the relevés on each study site, using a pre-formulated Excel
spreadsheet obtained online (Goepel 2012). The formula for the different Hill q values is:

39

Where:
R = Richness
q = Hill numbers order or effective number of species
pi = proportional abundance of the ith species

2.3.2.4 Plant growth form and species abundance in relevés
The composition and distribution of plant growth forms on a site can provide indicators of
the environmental forces shaping an ecological community (Landau 2004). Abundances of
individual plant species can also provide information about the influence of landslide
disturbance on plant community development. To investigate a possible relationship
between disturbance and growth form, each plant species in the three combined relevés on
the landslide and the three relevés on the surrounding terrain for each landslide was classified
and grouped by growth form in a table, along with each species’ average percent cover
(abundance). The standard provincial growth form categories of tree, shrub, forb, graminoid,
fern and fern allies, bryophyte, and lichen were used (BC Gov. 2010, 2015). The mean plant
species abundances by study site were subtotalled for each growth form category and their
proportions of the total cover were then calculated and compared among study sites. Each
individual plant species was also ranked in descending order by mean percent cover for each
study site and a list of the ten most abundant species by cover was compiled for each study
site. Plant species autecology guidance resources were consulted for interpretation of results
(Haeussler et al. 1990; FEIS 2023).

40

2.3.2.5 Two-sample Kolmogorov-Smirnov test to compare plant assemblages in relevés
The two-way sample K-S test was used to test for significant differences between species
abundance distributions of the two assemblages of relevés, for each of the three pairs of
landslide-undisturbed terrain study sites. The two-sample Kolmogorov-Smirnov (K-S) test
(Tokeshi 1993; Sokal and Rohlf 1995) is a nonparametric test of the equality of continuous or
discontinuous one-dimensional probability distributions. It quantifies the distance between
the empirical distribution functions of two samples and determines the likelihood of the two
sets of samples occurring if they were drawn from the same (though unknown) probability
distribution. The K-S test is considered one of the most useful nonparametric methods for
comparing two samples, due to its sensitivity to differences in both location and shape of
their empirical cumulative distribution functions. The maximum cumulative proportional
difference (K-S test D statistic) of abundances between sites was calculated in Excel
(Microsoft Corporation 2010) entering formulas by hand and compared with the critical
value (1.36/√n), obtained from tables (Zaiontz 2017) based on a sample size >40 with a pvalue of 0.05. If the maximum D-statistic was greater than the critical value, the null
hypothesis of no difference between samples was rejected.
2.3.3 Mapping, classifying, and analysing site series/types using multiple types of plot
data
2.3.3.1 Site types/series
The site type/series study incorporated the Beatton River, Cecil Lake, and Hasler Flats field
data and photos from the BEC and relevé plots and a series of randomly located 30 m
transects (described and analysed in Chapter 3) to type out vegetation and characterise patch
41

diversity on the landslides and the surrounding undisturbed terrain. In British Columbia, the
“site series” is the key unit of site-level ecosystem classification categories, based on
distinctive plant associations and typical soil, parent material, or slope position characteristics
(described in Section 2.3.1.1). In this paper, categories will be referred to interchangeably as
either site types or site series. Prior to mapping the site series, the perimeter of the landslide
was first digitised on mosaic transparent drone imagery, and then an equivalent area of
adjacent surrounding relatively undisturbed terrain was also digitised. This analysis was a
combination of site series classification using field manuals and GIS mapping exercises in
Global Mapper (Global Mapper 2020) using high-resolution drone (UAV – unmanned aerial
vehicle) imagery (i.e. mosaic transparent group TIF (tag image format) file) and an underlay
of LiDAR (light detection and ranging) DSM (digital surface model). Pre-processed,
government-purchased LiDAR data was used, which was collected at an average of 1.1 to 1.3
points per m2. The data collection project was flown in 2006 at heights of 1200 to 1550 m
using an Optech 3100 LiDAR system. In some areas the coverage was sparser, while in
other places there was overlap and almost twice as many points per m2. The configuration of
the outer perimeter of the undisturbed terrain was constrained by the limited availability of
the same high-resolution imagery that was available for the landslide site series mapping.
Prior to digitising site types/series polygons, an attempt was made to classify the site series of
each relevé using the same manuals and guides as for the BEC plots (Section 2.3.1.1). When
mapping the site types, plot data were employed to verify vegetation cover and serve as a
means of ground-truthing. The digital drone imagery and field photos were used as crossreferences to confirm plot classifications. Although some preliminary site series
classification was done on the BEC plots during the first field season (2016), these plot

42

classifications were reassessed based on a greater understanding of the local conditions
following two more field seasons, as well as an assessment of the high-resolution digital
imagery that subsequently became available. Classifications were revised where necessary.
Site types/series were initially delineated and digitised on the digital imagery in Global
Mapper and assigned classifications based on vegetation and identifiable environmental
indicators such as aspect and steepness of slope. Aerial interpretation methods were applied
to assist in distinguishing different types of vegetation (Sayn-Wittgenstein 1960; Avery 1969;
Sayn-Wittgenstein 1978). The GPS (global positioning system) plot locations for all
subjectively and randomly located plots for each landslide were then transferred into Global
Mapper, complete with their site series classifications.
Site series information determined from the plot data was used to verify the mapped types.
Conversely, some plot site series classifications were modified after assessing the digital
imagery and classification of other nearby plots. A plant indicator guide was consulted for
site series that appeared transitional (Beaudry et al. 1999). The same steps for classification
were followed for an equivalent area of the surrounding undisturbed terrain for each of the
three study areas. Due to the transient and jumbled nature of vegetation, moisture, and soil
material that is often characteristic of disturbed ecosystems such as landslides, not all plot
sites fit into neat categories. As a result, some adjustments were made to the classifications.
Additionally, some new site series/type categories were created to reflect site types
influenced by human activities or types not described in the field guides, such as cultivated
fields.

43

The areas of individual polygons of each site type were subtotalled, and proportions of each
site series were calculated for the total area of each landslide and associated undisturbed area
once the mapping was complete. The proportions of site series on and off the landslide were
graphically compared in a two-way bar graph in order of moisture regime for each of the
three study areas. Rank abundance curves were plotted for each study site, and KolmogorovSmirnov two-sample tests were done to compare landslide and undisturbed results and assess
whether the two paired sites came from the same distribution or community. Finally,
Shannon, Simpson’s (1-D), and Pielou’s diversity measures were calculated for the site types
found at each study site and compared between landslide and undisturbed sites.
2.3.4 Mapping, classifying, and analysing biophysical features using multiple types of
plot data
2.3.4.1 Geotyping
To describe and assess geophysical diversity, distinct individual geomorphic features were
mapped out on each landslide study site (Beatton River, Cecil Lake, and Hasler Flats). Each
feature was classified as a specific geomorphic type using Global Mapper to view the
imagery and digitise polygon boundaries. These features were referred to as geotypes for the
purposes of this study. The same digital imagery and digitised perimeter boundaries used in
the site series work described in Section 2.3.3.1 were utilised.
The ultimate delineation for geotypes was based on terrain, although the presence of the
“pond” geotype was at times first identified by the presence of cattail (Typha spp. – most
likely Typha latifolia). The provincial terrain classification guidebook (Howes and Kenk
1998) was used as a baseline reference, and features were classified with one of the
44

categories from the guidebook where possible. However, classification into pre-existing
categories was not always possible due to the nature of some landslide features on slides
adjacent to fields or cutblocks. Geotype classification was assisted by reference to the field
card data, notes, and photographs from the BEC plots and relevés, as well as from random
transects sampled for the beta diversity chapter (see Chapter 3).
The mapping was done on high-resolution drone photogrammetric imagery (mosaic
transparent group TIF file), using Global Mapper. Digital elevation model (DEM)
topographical imagery (DSM TIF file) was also used to assist in distinguishing geomorphic
features. Once all features on each landslide were mapped and classified, the total areas of
all the polygons for each geotype were added up. Summary statistics were then calculated.
Further analysis of the geotyping results was carried out by calculating the Shannon and
Simpson’s diversity indices, as well as Pielou’s evenness, for each landslide study site. Rank
abundance curves were plotted for each of the six study sites to compare landslides and
undisturbed terrain. Finally, a regression analysis was done to check for any relationship
between relevé vegetation diversity and geotype diversity.
2.3.4.1 Microtopography/surface roughness of BEC plots
Microtopography refers to the amount of soil surface roughness at the local level, at a scale
that can fundamentally influence nutrient and groundwater regimes, and thus plant
establishment and reproduction and wetland processes. Data preparation for assessment of
microtopography on the study areas involved clipping out the buffered BEC plots from
LiDAR or drone cloud points in Global Mapper and running an analysis on the standardised
elevation of the points for each plot. The surface roughness/microtopography of each plot on

45

each study site was represented by the coefficient of variation (in %), calculated as the
standard deviation of standardised point elevation values divided by the mean standardised
elevation for each plot. Generally, the higher the coefficient of variation, the higher the
spread of data relative to the mean standardised elevation, and thus the higher the microvariability of the terrain within the plot. Although the standard deviation of elevation method
of assessing surface roughness used in this study is not as computationally intensive as some
other methods, it has been shown to perform equally as well as other more complex measures
(Rozycka et al. 2016).
2.4 Results
2.4.1 Diversity of relevés
The landslide study areas exhibited a visually diverse array of plant communities and sites
over short distances. Very dry, sparsely vegetated sites could be found juxtaposed with
ponds and rafts, while level, heavily vegetated sites occurred next to steep, unvegetated
scarps. An illustration of some of the diversity found in the study areas is presented in Figure
2-2.

46

Figure 2-2. Vegetation and site diversity on landslides in the study. From top left clockwise: horst and graben
complex and mature rafts (Hasler Flats. Photo July 12, 2018); unvegetated weathered pillar next to heavy brush
(Cecil Lake. Photo September 17, 2017); dewatered/revegetated pond site (Cecil Lake. Photo August 23,
2018); steep unvegetated scarps interspersed with relatively level swaths of abundant invasive vegetation and
the occasional pond (Beatton River. Photo August 15, 2017).

47

Summaries of the ten overall most abundant plant species on landslide and undisturbed
relevés at each study site are presented in Table 2-1. The species abundances are presented
as mean percent cover of entire relevé area. A complete list of plant species and mean
abundances for these plots is provided in Appendix 2.
Table 2-1. Ten most abundant plant species for relevés at all study sites, comparing landslide and undisturbed
results. Values are mean percent cover of the entire relevé area +/- standard error of the mean.

48

For the Beatton River landslide study site relevés, the forb yellow sweetclover (Melilotus
officinalis) and the fern ally common horsetail (Equisetum arvense) were the leading species
for percentage of total plant cover, at 28.8% and 12.2%, respectively. On the Beatton River
undisturbed study site, the shrub saskatoon (Amelanchier alnifolia) and the tree paper birch
(Betula papyrifera) were the leading species overall (10.8% and 10.7%, respectively), and
wild sarsaparilla (Aralia nudicaulis) and twinflower (Linnaea borealis) were the leading
forbs (7.8% and 6.8%, respectively). On the Cecil Lake landslide study site, the leading
species overall was common horsetail at 33.5%. The leading shrubs on the Cecil Lake
landslide study site were green alder (Alnus viridus ssp. sinuata) at 24%, followed by a
willow (Salix sp.) at 5.6%. On the Cecil Lake undisturbed study site, the top two leading
species were trees: white spruce (Picea glauca) at 22.9% and trembling aspen (Populus
tremuloides) at 9.1%. The next most abundant species were the shrubs prickly rose (Rosa
acicularis) at 6.9% and highbush cranberry (Viburnum edule) at 6.8%. For both Hasler Flats
study sites, the fern ally common horsetail was the overall leading species (16.4% landslide,
15.6% undisturbed), followed by trembling aspen (10.8% landslide, 13.6% undisturbed). On
the Hasler Flats landslide study site, seven out of ten leadings species were shrubs, while on
the surrounding undisturbed study site, five out of ten leading species were shrubs.
There was also a difference between landslides in the presence of undisturbed species that
were also found on the landslide. For the Beatton River study area, none of the top ten
undisturbed species were also found on the landslide. On the Cecil Lake landslide, four
shrub species were shared between the undisturbed terrain and the landslide: Populus
tremuloides, Rosa acicularis, Shepherdia canadensis, and Alnus viridis ssp. sinuata. Hasler
Flats study area had six top species in common between the paired study sites: Equisetum

49

arvense, Populus tremuloides, Rosa acicularis, Alnus viridis ssp. sinuata, Cornus stolonifera,
and Symphoricarpos albus.
2.4.1.1 Plant growth form composition and species abundances for relevés
Plant growth form abundances varied significantly between relevé study sites. Table 2-2
provides a summary of mean vegetation covers by growth form as a percentage of mean total
vegetation cover on the plots, as well as the subsequent proportions, for all six study sites. A
complete list of abundances and proportions of all species by growth form in relation to mean
total vegetation cover for each study site is provided in Appendix 3.

50

Table 2-2. Relevé mean growth form cover as a percentage of mean total vegetation cover, along with
proportions, for all study sites. The table provides a comparison between landslide and undisturbed vegetation
growth form cover.
Study site
Vegetation cover
Mean total vegetation cover (% )

Beatton
Landslide

Beatton
Cecil
Undisturbed Landslide

Cecil
Hasler
Undisturbed Landslide

Hasler
Undisturbed

59.68

97.35

104.16

87.02

81.69

100.81

Cover (%)

0.20

35.62

11.22

18.46

0.34

17.09
17.55

10.44

Proportion of mean total (%)

10.02

40.94

13.74

18.31

Cover (%)

1.74

37.27

43.83

27.57

41.53

46.04

Proportion of mean total (%)

2.92

38.29

42.08

31.68

50.84

45.67

Cover (%)

45.27

26.38

13.82

13.08

6.86

20.08

Proportion of mean total (%)

75.86

27.09

13.26

15.03

8.40

19.91

Cover (%)

0.13

15.97

1.67

2.75

2.97

0.39

Proportion of mean total (%)

0.21

16.40

1.60

3.16

3.63

0.39

Cover (%)

12.17

0.13

33.50

0.03

16.42

15.63

Proportion of mean total (%)

20.39

0.13

32.16

0.03

20.10

15.51

Cover (%)

0.17

0.43

0.90

7.93

2.69

0.21

Proportion of mean total (%)

0.28

0.44

0.86

9.11

3.30

0.21

Cover (%)

N/A

0.08

0.02

0.05

N/A

N/A

Proportion of mean total (%)

N/A

0.08

0.02

0.06

N/A

N/A

Trees

Shrubs (% )

Forbs (% )

Graminoids

Ferns & Fern allies

Bryophytes

Lichens (% )

For Beatton River, forbs dominated on the landslide (>75% of the total cover) while shrubs
were leading on the undisturbed terrain (38.3% of total cover), followed by forbs (27.1%).
For the Cecil Lake landslide study site, shrubs were dominant (43.8% of cover) followed by
ferns and fern allies (33.5% of cover). On the Cecil Lake undisturbed terrain study site, trees
dominated (41% of cover) followed closely by shrubs (>31% of cover). For the Hasler Flats
landslide site, shrubs prevailed (>50% of cover) followed by ferns and fern allies (>20% of
cover) and then trees (>13%). On the Hasler Flats undisturbed terrain, shrubs comprised
51

>45% of the total cover, followed by forbs (20% of cover) and then trees (>18% of cover).
Fern and fern allies were close behind, at 15.5%.
2.4.1.2 Plant diversity indices and measures for relevés
The results for relevé plant diversity assessment using both the mean cover values and the
individual relevé cover values show that in all comparisons, the Shannon index and Pielou’s
evenness were higher on the undisturbed terrain compared to the landslide (Table 2-3).
However, all landslide study sites had more variation around the mean than their paired
undisturbed study sites for the Shannon index.
Table 2-3. Combined* and mean (+/- standard error) vegetation diversity indices for relevés

Study site
Beatton Landslide
Beatton Undisturbed
Cecil Landslide
Cecil Undisturbed
Hasler Landslide
Hasler Undisturbed

Combined
Mean
Shannon
Shannon
index
index (H')
(H')
(n=3)

Combined
Pielou's
evenness
(J)

Mean
Pielou's
evenness (J)
(n=3)

Combined
Simpson's
index
(1-D)

Mean
Combined
Simpson's Richness
index
(S)
(1-D)
(n=3)

Me an
Richness
(S)
(n=3)

1.75

1.78 ± 0.26

0.47

0.55 ± 0.10

0.72

0.59 ± 0.22

41

25.33 ± 4.04

3.22

2.58 ± 0.17

0.71

0.66 ± 0.05

0.95

0.89 ± 0.03

91

52.33 ± 13.01

2.46

2.25 ± 0.36

0.53

0.55 ± 0.08

0.84

0.82 ± 0.06

106

61.00 ± 18.25

2.65

2.43 ± 0.30

0.64

0.65 ± 0.08

0.90

0.88 ± 0.05

65

43.33 ± 4.04

2.82

2.64 ± 0.33

0.63

0.63 ± 0.07

0.92

0.90 ± 0.03

90

65.33 ± 4.93

2.94

2.63 ± 0.12

0.66

0.65 ± 0.01

0.93

0.90 ± 0.01

84

57.00 ± 10.15

*where combined values are based on first averaging plant abundance data from the three relevés and then
calculating the diversity indices.

Simpson’s index of diversity was higher on the Beatton River and Cecil Lake undisturbed
study sites compared to the landslide study sites, with an especially marked difference
between the Beatton River landslide and undisturbed sites. However, the Beatton River
landslide had a much higher standard deviation than the undisturbed study site. Simpson’s

52

index of diversity was slightly higher on the Hasler Flats undisturbed site compared to the
landslide site using the mean cover metric, while the two values were the same for the
individual cover metric. Richness was lower on the Beatton River landslide compared to the
undisturbed terrain, while it was higher on the landslide for both Cecil Lake and Hasler Flats.
Hill numbers were calculated using the mean percent cover data of plant species of relevés
for each study site, as shown in Table 2-4.
Table 2-4. Hill numbers for vegetation diversity on relevés.
Mean Hill Numbe rs -True Diversity qD:
Generalised Beatton Beatton Cecil Cecil Hasler - Hasler Order q
Mean
Landslide
Undisturbed Landslide Undisturbed Landslide Undisturbed
0
1
2
∞

harm
geom
avg
inf

25.00
6.04
4.20
2.63

52.33
13.36
8.85
4.38

61.00
9.84
5.61
3.23

43.33
11.74
8.01
4.39

65.33
14.54
9.68
5.15

57.00
13.94
8.86
4.37

For the Cecil Lake and Hasler Flats study areas, mean Hill number 0 (species richness) was
higher on the landslide study site than on the undisturbed terrain study site, while at the
Beatton River study area, Hill number 0 was much higher on the undisturbed terrain. For the
rest of the Hill numbers 1, 2, and infinity (∞), the values were higher on the undisturbed
terrain than on the landslide for the Beatton River and Cecil Lake study areas, but higher on
the landslide than the undisturbed terrain study site for Hasler Flats. The Beatton River study
area exhibited the biggest difference between landslide and undisturbed terrain study sites,
for all four Hill numbers. The Beatton River landslide study site had the lowest Hill numbers
of all three landslides while the Hasler Flats landslide study site had the highest Hill
numbers. Overall, the Hasler Flats study area had the smallest differences in Hill number
values between landslide and undisturbed terrain study sites, while the Beatton River study
area generally had the largest differences.
53

2.4.1.3 Rank abundance curves for plant species on relevés
Rank abundance curves provide a visual representation of the richness and evenness of a
study site. The rank abundance curves for all relevé study sites are shown in Figure 2-3.
Results indicate a higher evenness of the Cecil Lake and Hasler Flats undisturbed study sites
compared to the landslide sites, as shown by the steeper gradient of the curves for the
undisturbed sites. The opposite is apparent for the Beatton River study area, with the
landslide study site having higher evenness than the undisturbed site.
The rank abundance curve for the Cecil Lake study site highlights the greater richness on the
landslide compared to the surrounding undisturbed terrain. The Cecil Lake landslide site also
had the highest species richness overall. The Beatton River landslide study site had a much
lower richness than the surrounding undisturbed site, with less than half the number of
species (41 compared to 91).

54

Figure 2-3. Comparison of rank abundance distributions for relevé plot plant species for all study sites.
Abundance rank by species is on the x-axis.

2.4.1.4 Kolmogorov-Smirnov two sample comparison for relevés
The Kolmogorov-Smirnov two-sample test to compare the plant composition between the
relevés on the paired study sites revealed that for the Beatton River and Cecil Lake study
areas, the maximum proportional difference (D statistic) was greater than the critical value
(Fig. 2-4). These results rejected the null hypothesis of no difference between the two
samples, and indicated there was indeed a difference in plant community diversity structure
55

between the landslide and undisturbed study sites. The opposite was true for the two Hasler
Flats study sites, where the maximum D statistic was less than the critical value by half. The
null hypothesis of no difference could not be rejected. Overall, the Beatton River study area
K-S two sample test had the biggest difference between maximum D statistic and critical
value.

56

Figure 2-4. Kolmogorov-Smirnov two sample test for all relevés, comparing landslide and undisturbed plant
abundance for Beatton River (BE), Cecil Lake (CE), and Hasler Flats (HA) study areas. Results where the
maximum D statistic is greater than the critical value indicate a significant difference in plant community
between two samples.

57

2.4.2 NMS ordination of BEC plot data: Assessing patterns and correlations of plant
communities and environmental factors
2.4.2.1 Plant species and environment biplot of BEC plots
Nonmetric multidimensional scaling (NMS) ordination using PC-ORD was carried out on the
vegetation and environment data for all BEC plots (landslide and undisturbed benchmark) to
seek pattern within a matrix of multiple responses, in this case plant species. The Mantel
test for redundancy for the six pairs of runs of real data yielded values of 93.84%, 88.37%,
88.92%, 90.62%, 90.46%, and 86.00%, indicating high redundancy. The final (best) solution
from the stress test had three dimensions or axes, with stress (i.e. residual sum of squares) of
18.20. Stress directly measures the quality of an ordination. The Monte Carlo test result
used 250 randomised runs, with the probability that a similar final stress could have been
obtained by chance being 0.0040, or significantly low. The final solution had 138 iterations,
with the stress levelling out and plateauing, indicating stability of the solution.
The resulting biplot (Figure 2-5) depicts all the plots in species space, paired with the
environmental factors of interest. Axis 1 accounted for 50.7% of the variation in plant
species composition, while Axis 2 accounted for 24.0% of the variation, totaling 74.7% of
the plant community composition explained. Axis 3 is not shown, but accounted for 12.5%
which, combined with Axis 1 and 2, explained a total of 87.2% of the variation in species
composition among sites. The point distribution reflects the makeup of the plant community,
with similar plots close together in the biplot.

58

Figure 2-5. NMS (nonmetric multidimensional scaling) ordination biplot of 116 subjectively located averaged
BEC (biogeoclimatic ecosystem classification) plots and 163 plant species sampled on three landslide study
areas. The scale of the biplot has been increased to 700% due to the crowded distribution of the plots. Axis 1
explains 50% of the variation and represents an apparent gradient from forest community to open/grassland
community, from left to right across the figure. Axis 2 explains 24% of the variation and represents an apparent
soil moisture gradient from drier to wetter, moving from the upper to lower extent of the figure. Both axes are
significant using the Monte Carlo test (p = 0.0040 and final stress = 18.20). Study areas are colour-coded, as
shown in the legend. BE = Beatton River, CE = Cecil Lake, and HA = Hasler Flats. The environmental
variables correlated with the ordination axes are: MesoslopePos (mesoslope position), Moisture (soil moisture
regime), SlopeGrad (slope gradient in %), HLI (heat load index), and MatOrigin (material origin). Strength and
nature of environmental variable relationships are represented by the length and direction of the red vectors.
Undisturbed benchmark plots BE16-29, BE16-30, CE-35a, and HA16-01 are highlighted in yellow in the biplot.

The resulting statistically significant (randomisation p test = 0.0040) three-dimensional
ordination solution shows the distribution of plots for all study areas (Beatton, Cecil, and
59

Hasler). Axis 1 and Axis 2 represent condensed gradients of differences in the composition
of the data matrix. The environmental variables soil moisture regime, mesoslope position,
heat load index, slope gradient, and material origin are represented as red vectors in the
biplot, radiating out from the centroid.
Axis 1 appeared to represent a gradient from forest community to open/grassland community,
moving from left to right. The benchmark plots on the undisturbed terrain (highlighted in
Figure 2-5 as BE16-29, BE16-30, CE-35a, and HA16-01) appeared quite different in
composition from the associated landslides, as they were much further away from most plots
in ordination space. These plots were also different from each other, based on their location
on the biplot. Interpretation of Axis 2 is less clear, but it appears to represent a moisture
gradient, transitioning from drier to wetter when moving from the lower to upper extent of
the axis in ordination space.
2.4.2.2 Plant species correlations with axes using Pearson’s correlation coefficient r, for
BEC plots
Using the NMS ordination results, Pearson’s correlation r was calculated in PC-ORD for
each plant species for Axis 1 and Axis 2. Plant species that were strongly correlated with
Axis 1 either positively or negatively are shown in Table 2-5. The exotic forb yellow salsify
(Tragopogon dubius) was most positively correlated with Axis 1, and the next four most
positively correlated species were also grassland species. The forb creamy peavine (Lathyrus
ochroleucus) was most negatively associated with Axis 1, followed by the shrub red
honeysuckle (Lonicera dioica) and the tree trembling aspen (Populus tremuloides).

60

Table 2-5. Pearson correlations of BEC plot plant species abundance with NMS Axis 1. N = 116.

Plant species that were most strongly correlated with Axis 2 are shown in Table 2-6. The
strongest positive correlation was the shrub species red swamp currant (Ribes triste),
followed by the forb red clover (Trifolium pratense) and the fern ally field horsetail
(Equisetum arvense), and then the tree balsam poplar (Populus balsamifera ssp.
balsamifera). The plant species with the strongest negative correlation was the forb western
meadowrue (Thalictrum occidentale), followed by the shrub Sitka alder (Alnus viridis ssp.
sinuata) and then an unknown fern species. Also showing strong negative correlations with
Axis 2 were the forb large-leaved avens (Geum macrophyllum) and the pixie cup lichen
(Cladonia pyxidata).
Table 2-6. Pearson correlations of BEC plot plant species abundance with NMS Axis 2. N = 116.
Positive correlation
Plant species
Common name
Ribes triste
Red swamp currant
Trifolium pratense
Red clover
Equisetum arvense
Field horsetail
Populus balsamifera ssp. balsamifera Balsam poplar
Fragaria virginiana
Wild strawberry

Negative correlation
r
0.366
0.351
0.281
0.270
0.264

Plant species
Common name
Thalictrum occidentale Western meadowrue
Alnus viridis ssp. sinuata Sitka alder
Unknown fern sp.
Fern sp.
Geum macrophyllum
Large-leaved avens
Cladonia pyxidata
Pixie cup lichen

r
-0.483
-0.400
-0.400
-0.392
-0.356

None of the environmental variables tested exhibited overly strong associations with either of
the axes (Table 2-7). However, material origin (MatOrigin) showed a somewhat strong
correlation with Axis 2, the apparent moisture gradient, while slope gradient (SlopeGrad) had
a weak correlation with Axis 1, the forest-grassland gradient. Mesoslope position
61

(MesoslopePos), heat load index (HLI), and soil moisture regime (Moisture) all had weak
correlations with Axis 1. Of the three variables associated with the forest-grassland gradient,
mesoslope position had the strongest association, followed by moisture regime.
Table 2-7. Pearson correlations of environment variables and Axis 1 and Axis 2 of the ordination space. N=116

Variable
MatOrigin
Moisture
SlopeGrad
MesoslopePos
HLI

Axis 1 (R2= 50.7) Axis 2 (R2= 24.0)
0.253
0.153
-0.023
-0.020
0.185
0.024
-0.086
-0.101
-0.103
0.010

PCORD does not report p-values for correlations between variables and ordination axes,
because sample sizes are typically large enough that even a very small correlation is
“statistically significant”. Therefore, usually the lowest r-value is more conservative than the
one determined by the p-value (McCune and Grace 2002).
2.4.3 Mapping, classifying, and analysing site series/types using multiple types of plot
data
2.4.3.1 Mapping, classifying, and analysing site types/series
The site series digitising exercise yielded a variety of configurations on each of the study
sites, with some very small site type polygons next to very large polygons. In total, 20
different site types/site series were identified (Table 2-8).

62

Table 2-8. Site types/series found on all landslides and undisturbed terrain in the study.
Beatton
Beatton
Cecil
Cecil
Hasler
Hasler
landslide undisturbed landslide undisturbed landslide undisturbed
"Site series"
Description*
101
Sw – Trailing raspberry – Step moss
102
Pl – Kinnikinnick – Lingonberry
103
SwPl – Soopolallie – Fuzzy-spiked wildrye
110
Sw – Oak fern – Sarsaparilla
111
Sw – Currant - Horsetail
101$
At – Rose – Creamy peavine
103$
At – Rose – Fuzzy-spiked wildrye
110$
At – Highbush-cranberry – Oak fern
111$6B.1
Acb – Dogwood – Highbush-cranberry
111$6B.2
At – Cow-parsnip – Meadowrue
112 (Fm02) AcbSw – Mountain alder – Dogwood
C
Cultivated field
Fl
Flood deposits - seasonal
FL
Fluvial (creek/river)
Gb
Brushland
Gb51
Saskatoon – Blue wildrye
Gg
Grassland
Gg51
Sle nder whe atgrass – Pasture sage
Ro
Rock outcrop
Rt
Talus
W
Wetlands/ponds
Totals

m2
31190
977
79743
42487
30142
82840
0.0
8258
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2009
0.0
242
23510
818
302216

m2
38200
0.0
35440
17430
57600
48945
13090
0.0
0.0
0.0
14460
0.0
0.0
0.0
0.0
46340
11740
17927
0.0
0.0
1044
302216

m2
161417
0.0
31387
30929
63888
134646
2257
38737
58348
0.0
0.0
0.0
6571
5989
0.0
0.0
0.0
0.0
0.0
0.0
27563
561732

m2
123924
0.0
0.0
6990
18784
250756
626
0.0
0.0
0.0
0.0
26670
0.0
4600
12550
0.0
62477
0.0
0.0
53537
1878
562792

m2
0.0
0.0
134
0.0
0.0
6097
1558
628
1257
1056
2770
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1482
14982

m2
0.0
0.0
0.0
0.0
0.0
10130
0.0
807
886
0.0
3142
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
24
14989

*Sw=White spruce (Picea glauca ); Pl=Lodgepole pine (Pinus contorta ); At=Trembling aspen (Populus tremuloides );
Acb=Black cottonwood (Populus balsamifera ssp. balsamifera )

The majority of the site series in Table 2-8 are defined and described in the BWBS field
guide (DeLong et al. 2011). As BWBS forests in the Peace River Region contain a notable
component of mature deciduous trees, the field guide describes a number of deciduous site
series, denoted by the dollar sign symbol $. There are four newly created or modified “site
series” categories: cultivated field [C], flood deposits (seasonal) [Fl], fluvial (creek/river)
[FL], and wetlands/ponds [W].

63

Figure 2-6 illustrates an example of site series mapping on the Beatton River landslide.

Figure 2-6. Site series/types mapping example for Beatton River landslide study site. Site series/types codes
(shown in blue) are 101 = Sw – Trailing raspberry – Step moss, 101$ = At – Rose – Creamy peavine, 103 =
SwPl – Soopolallie – Fuzzy-spiked wildrye, 110$ = At – Highbush-cranberry – Oak fern, 111 = Sw – Currant Horsetail, Gg = Grassland, Ws = Wetland/pond (swamp). Black lines represent site series polygon boundaries.
The pink line is the perimeter of the undisturbed terrain study site surrounding the landslide.

The results of the site series mapping and classification exercise showed a change in site
series composition, area, and polygon count when comparing the landslide study site to the
surrounding undisturbed terrain. The site series/types were graphed in order of moisture
regime on paired bar charts, from driest to wettest, for each study area (Figure 2-7). For the
64

Beatton River landslide study site, the 101$ (i.e. mesic deciduous) site series occupied the
highest proportion of the total study site area (27.4%), followed closely by 103, a somewhat
drier site series, at 26.4 %. The next most abundant type on the Beatton River landslide was
the moist site series 110 (14.1%), followed by the mesic coniferous 101 (10.3%) and then the
very dry Rt site series (7.8%). The two mesic site series occupied 37.7% of the total
landslide area. For the Beatton River undisturbed terrain study site, the leading site series
was 111, a coniferous wet type (19.1%), followed closely by the deciduous mesic 101$
(16.2%) and the dry Gb51 (brushland) at 15.3%. The next highest abundance on the Beatton
River undisturbed site was the coniferous mesic site series 101 (12.6%) and then the drier
103 (11.7%). Combined, the two mesic site series occupied 28.8% of the total undisturbed
area.

65

Figure 2-7. Site series percent area paired graphs for Beatton River, Cecil Lake, and Hasler Flats study areas,
with categories ordered from low (left) to high (right) moisture regime.

66

Overall, there were more site series polygons on the Beatton River landslide study site (43)
than the undisturbed study site (29). The landslide study site had a higher percentage of
mesic site series area, as well as a higher ratio of deciduous to coniferous mesic site series
area. There was a similar number of mesic polygons on the Beatton River landslide and on
the undisturbed terrain, but the mean sizes and ranges were markedly different. The
landslide had a higher proportion of dry site series area compared to the undisturbed terrain.
Although the landslide contained five times more pond polygons than the undisturbed study
site, the undisturbed ponds were on average six times larger.
For the Cecil Lake landslide study site, the mesic 101 site series per cent area was leading
(28.7%) followed closely by its mesic deciduous equivalent 101$ (24.0%). The next two
most abundant site series were 111 (11.4%) and 111$6B.1 (10.4%), which were wet types.
For the Cecil Lake surrounding undisturbed terrain study site, the mesic deciduous 101$ site
series area dominated (44.6%), followed by its coniferous equivalent 101 (22.0%). The next
most abundant site series by per cent of total undisturbed area were grassland (Gg 11.1%)
and talus (Rt 9.5%), which are dry types. Overall, mesic site series dominated on both the
landslide and undisturbed terrain of the Cecil Lake study area, with coniferous 101 per cent
area greater than deciduous 101$ on the landslide, and the opposite trend on the undisturbed
terrain.
There were more than four times as many site series/types polygons on the Cecil Lake
landslide compared to the undisturbed terrain (381 polygons vs. 86), but almost 75% of the
landslide polygons were ponds. There were 71 times more pond polygons on the landslide
study site than on the surrounding undisturbed terrain. However, the ponds on the Cecil Lake
undisturbed terrain were an average of almost five times larger than the landslide ponds.
67

On the Hasler Flats landslide study site, the 101$ deciduous site series was leading (40.7% of
the total area), followed by a much wetter site series, 112 (Fm02), which was 18.5 %. The
pattern was similar for the undisturbed terrain study site, with 101$ at 67.6 % and 112
(Fm06) at 21.0 % of the total study site area. The third most abundant site series on the
Hasler Flats landslide was a very dry type, 103$ (10.4 %), followed by ponds (W) at 9.9 %,
and then two very wet site series, 111$6B.1 (8.4 %) and 111$6B.2 (7.0 %). On the Hasler
Flats undisturbed study site, the third most abundant site type was 111$6B.1 (5.9 %)
followed by the moist site 110$ (5.4 %). Ponds occupied only 0.2 % of the total undisturbed
area.
Overall, the Hasler Flats landslide had more site series/types than the undisturbed terrain and
there were significantly more individual polygons. The Hasler Flats landslide study site had
eight site series, while the undisturbed terrain study site had five site series. The landslide
had 47 polygons, while the undisturbed terrain had just seven polygons. The mean size of
most polygons was smaller on the landslide. However, the pond polygons (W) on the
landslide had a much larger mean size than those on the undisturbed study site, and there
were also thirteen times more individual pond polygons on the landslide study site.
The diversity indices of the site series for the study sites for all three study areas (Table 2-9)
indicate that the Beatton River undisturbed terrain was more diverse in site series/type
composition than the landslide for Shannon diversity H’, Pielou’s evenness J, and Simpson’s
index of diversity 1-D. In contrast, the Cecil Lake and Hasler Flats landslide study sites were
more diverse in site series /type composition than the surrounding undisturbed terrain
regarding Shannon diversity, Pielou’s evenness, and Simpson’s index of diversity. In terms
of richness, the value was the same both on and off the slide for Beatton River and Cecil
68

Lake study areas. The Hasler Flats study area, however, had a higher richness of site
series/types on the landslide study site compared to the surrounding undisturbed terrain (8
site series vs. 5). Of the three study areas, Hasler Flats also had the biggest difference
between each pair of index values comparing landslide and undisturbed study sites.
Table 2-9. Diversity indices for site series polygons - all study sites. BE = Beatton River, CE = Cecil Lake,
HA = Hasler Flats.

Richness
(S)
Site
BE -Landslide
BE - Undisturbed
CE - Landslide
CE - Undisturbed
HA - Landslide
HA - Undisturbed

11
11
11
11
8
5

Shannon
diversity index
(H')
1.817
2.170
1.959
1.624
1.712
0.927

Pielou's
Simpson's
evenness index of diversity
(J)
(1-D)
0.758
0.808
0.905
0.872
0.817
0.823
0.677
0.728
0.823
0.766
0.576
0.493

The Kolmogorov-Smirnov two sample test for all three study areas (Fig. 2-8) showed that
there was no significant difference between landslide and undisturbed site series diversity.
This finding is evidenced by the fact that the critical value is higher than the maximum D
statistic in all three cases.

69

Figure 2-8. Site series Kolmogorov-Smirnov two sample tests for the paired study sites of each study area of
Beatton River, Cecil Lake, Hasler Flats. For all three tests, the critical value is greater than the D statistic,
indicating no significant difference.

70

2.4.4 Mapping, classifying, and analysing biophysical features using multiple types of
plot data
2.4.4.1 Geotyping: Mapping, classifying, and analysing geomorphic features
Landslide Geotypes
In the geotyping exercise, 20 different geomorphic types were identified and digitised over
the three landslide study sites. A glossary describing each classification is found in
Appendix 4. Figure 2-9 shows an example of the digitised geotyping on the Cecil Lake
landslide study site.

Figure 2-9. Example of digitised geotyping on the Cecil Lake landslide. Geotype codes (shown in blue) are PO
= Pond/wet area, RI = Ridge, RA = Raft, PI = Pillar, Fl = Flood deposits -seasonal, SW = Swale, GU = Gully.
The black lines are polygon boundaries.

71

The summaries of the resultant geotyped areas for each of the three landslides are presented
in Table 2-10. Comprehensive tables listing each geotype polygon and its associated area for
each landslide are provided in Appendix 5. For Beatton River and Cecil Lake, the slide
matrix geotype occupied the largest percentage of the total area of the landslide study site.
The slide matrix component was highest (almost 69%) on the Cecil Lake landslide and
lowest (36.83%) on the Hasler Flats landslide. The leading geotype by percent area for
Hasler Flats was rafts, at 45.72%. The second highest geotype percentage for both Beatton
River and Cecil Lake was hummocks, at 14.25% and 10%, respectively, while for Hasler
Flats the second highest geotype area was the landslide matrix (36.83%). The third and
fourth highest geotype percentages were scarp (9.16%) and raft (8.54%) for Beatton River,
raft (6.65%) and pond (4.91%) for Cecil Lake, and pond (9.89%) and scarp (6.03%) for
Hasler Flats.

72

Table 2-10. Landslide geotype summary by area (m 2) and percentage.

The number and composition of geotype polygons differed markedly among landslide study
sites. Beatton River landslide study site had 56 separate geotype polygons, Cecil Lake had
327 polygons, and Hasler Flats had 39 polygons. For all three landslides, ponds had the
highest geotype polygon count. On the Cecil Lake landslide study site, ponds comprised
almost 87% of the total number of polygons, but they only added up to 4.9% of the total area.
Ridges had the second highest polygon count on the Cecil Lake landslide, followed closely
by rafts. On the Beatton River and Hasler Flats landslides, rafts had the second highest
polygon count. The third highest polygon count for the Beatton River landslide study site

73

was represented by scarps, while for Hasler Flats scarps and ridges were tied for third place.
The Beatton River study site had 15 different geotypes, Cecil Lake had 12 different geotypes,
and Hasler Flats had five different geotypes.
The diversity indices for geotypes (Table 2-11) showed that the Beatton River landslide had
the highest geotype richness, Shannon index, and Simpson’s diversity of the three landslide
study sites. The Beatton River landslide study site also had a higher Shannon diversity,
Pielou’s evenness, and Simpson’s diversity than the oldest landslide, Cecil Lake. Hasler
Flats had the lowest richness but the highest Pielou’s evenness.
Table 2-11. Geotype diversity indices for landslide geotype polygons.

Richness
Shannon
(S)
diversity inde x
(H')
Site
Beatton River
Cecil Lake
Hasler Flats

15
12
5

1.759
1.176
1.188

Pielou's
evenness
(J)

Simpson's
index of
diversity
(1-D)

0.650
0.473
0.738

0.739
0.506
0.642

Undisturbed Geotypes
The results of the digitising of geotypes in the undisturbed terrain are presented in Table 212. Observations of the terrain in the field and on digital imagery showed a general trend of
larger, more contiguous geotype polygons in the undisturbed areas for all three landslides.
Digitising and summarising the polygons confirmed this, as all three undisturbed study sites
had fewer types but larger, more extensive polygons. However, there appeared to be a
greater diversity of geotypes on the landslides compared to the surrounding undisturbed
terrain.

74

Table 2-12. Undisturbed geotype summary by area (m 2) and percentage.

Diversity indices were calculated for the undisturbed geotypes (Table 2-13). Results show
that the Cecil Lake study site had the highest richness, while Hasler Flats had the lowest
richness. The Cecil Lake undisturbed terrain also had the highest Shannon index and
Simpson’s index. The Hasler Flats undisturbed study site had the lowest Shannon and
Simpson’s indices, but the highest Pielou’s evenness value.

75

Table 2-13. Geotype diversity indices for undisturbed geotype polygons.

Richness
Shannon
Pielou's Simpson's
(S)
diversity index evenness index of
(H')
(J)
diversity
(1-D)
Site
Beatton River
8
1.547
0.744
0.743
Cecil Lake
9
1.599
0.728
0.761
Hasler Flats
6
1.450
0.810
0.731

Comparing landslide geotype diversity vs undisturbed geotype diversity
Rank abundance curves for geotype data were plotted for each of the six study sites (Fig. 210). The results show that the undisturbed sites had a lower evenness overall, illustrated by
the steeper curve. However, the landslide sites had higher richness for Beatton River and
Cecil Lake.

76

Figure 2-10. Rank abundance curves comparing geotype diversity on landslides and undisturbed terrain.

77

2.4.4.2 Surface roughness/microtopography of BEC plots
The results of the microtopography (surface roughness) exercise showed a range of
coefficients of variation between the BEC plots for each landslide site, as shown in Table 210. The number of points for each buffered plot varied based on the quality of the LiDAR
imagery and ranged from 48 to 120 points. A table summarising elevation point statistics for
each buffered BEC plot is provided in Appendix 6.
Table 2-14. BEC 50m2 plot LiDAR points: Elevation coefficient of variation (CV) summary statistics.

Beatton River, which was the most recent landslide, had the highest mean elevation CV,
whereas the Hasler Flats landslide had the lowest mean CV. The same pattern existed for the
range of CV, where Beatton River landslide had the highest range and Hasler Flats landslide
had the lowest range. Cecil Lake and Hasler Flats landslides had a much smaller minimum
CV overall (22.46% and 29.93%, respectively) than Beatton River (41.82%). Beatton River
also had the highest maximum CV, at 139.42%. All three landslides contained at least one
plot with an elevation CV greater than 100%.
Only a few plots were available for an assessment of the surface roughness on the
surrounding undisturbed terrain for each study area, as the original plot sampling focused
mainly on the landslide, with one plot in the undisturbed terrain for Cecil Lake and Hasler
Flats and two plots in the undisturbed terrain for Beatton River. There were insufficient plots
for a true comparison with the landslide plots. Additionally, the LiDAR point data used for
analysing microtopography on the landslides study sites did not extend enough beyond the
78

landslide area to provide adequate coverage of the adjacent undisturbed sites to calculate
elevation CV for these areas. However, observations obtained from traversing the
undisturbed terrain during subsequent sampling for this chapter showed that this terrain was
generally smoother and less varied than the landslide body. This was also evidenced by the
greatly reduced number of ponds on the undisturbed study sites for all study areas.
2.4.4.3 Assessing correlation between vegetation diversity and geotype diversity
Simple linear regression analyses were run to assess whether there was a significant
association of vegetation diversity with geotype diversity in terms of richness, Shannon
diversity, Pielou’s evenness, and Simpson’s diversity (Fig. 2-11).

79

Figure 2-11. Relationships between relevé vegetation diversity and geotype diversity in terms of richness,
Shannon diversity, Pielou’s evenness, and Simpson’s diversity for all three landslide study areas. Blue dots
represent landslide sites and orange dots represent undisturbed sites for each study area. Regression lines are
not shown, as no significant regression relationship was found for any of the comparisons.

80

Regression results were obtained for landslide richness (R2 = 0.632, F (1,1) =1.718, p =
0.415), landslide Shannon diversity (R2 = 0.778, F (1,1) =3.495, p = 0.313), landslide
Pielou’s evenness (R2 = 0.607, F (1,1) =1.546, p = 0.431), and landslide Simpson’s diversity
(R2 =0.404, F (1,1) = 0.677, p = 0.562). In addition, regression results were obtained for
undisturbed richness (R2 = 0.868, F (1,1) = 6.601, p = 0.236), undisturbed Shannon diversity
(R2 = 0.798, F (1,1) = 3.944, p = 0.297), undisturbed Pielou’s evenness (R2 = 0.0376, F (1,1)
= 0.0391, p = 0.876), and undisturbed Simpson’s diversity (R2 = 0.926, F (1,1) = 12.485, p =
0.176). For all comparisons, and contrary to expectations, geomorphological diversity did
not significantly predict vegetation diversity.
2.5 Discussion
This research chapter set out to answer some key questions related to the biophysical
diversity of landslides in northeastern BC. The main hypothesis underpinning the research
was that landslides are biophysically more diverse than nearby relatively undisturbed terrain,
with a greater diversity of plant communities, species abundances, geomorphic types, and
microsites. The methods employed to answer the questions consisted of a series of subjective
and random vegetation sampling, as well as NMS ordination, species composition and
abundance analyses, diversity index calculations, and classification and GIS mapping of
biophysical features. The findings of this research point to a partial confirmation of the
hypothesis that landslides are more diverse than the surrounding landscape, but this is not
always the case. Results also indicate that landslide ecology is more complex than
conventional succession theories would suggest.

81

1) How does plant species composition, abundance, and distribution differ on landslides
compared to adjacent undisturbed terrain?
Plant diversity of relevés
Contrary to what was expected, the rank abundance curves and calculated evenness,
Shannon, and Simpson diversity indices for the Beatton River and Cecil Lake study areas
show the vegetation of the relevés on the undisturbed study sites is generally more diverse
than on the landslides, while for the Hasler Flats study area there is no significant difference
between disturbed and undisturbed relevés. The lower evenness on the landslides indicates a
few species, especially exotics such as Melilotus spp. and Sonchus spp., are very abundant
and there are many species with low relative abundances. These findings could be due to the
surrounding terrain being much older than the landslide. As the landslides ranged between
two and twenty years old at the time of sampling, not enough time had passed to allow the
full spectrum of available plant species to establish. In addition, the headscarp was adjacent
to an anthropogenically modified site for all three landslides: Beatton River and Cecil Lake
landslides were both next to agricultural fields of forage crops, while Hasler Flats landslide
was adjacent to a young deciduous cutblock. Anthropogenically developed or modified areas
are usually relatively low in plant species diversity compared to the surrounding landscape.
Further, the plants in cultivated agricultural fields are commonly grasses and other fast
growing pioneer species which spread easily via wind and runoff.
The greater richness on the landslide compared to the undisturbed terrain for Cecil Lake and
Hasler Flats landslides in contrast to Beatton River may be because the Beatton River
landslide study site was the most recently disturbed slide and had a higher proportion of bare

82

or sparsely vegetated areas. The Beatton landslide was largely a reactivation of a previous
landslide. Thus, the randomly placed relevés were more likely to land on areas of low
vegetation cover consisting of mainly exotic species. The Cecil Lake landslide study site had
the highest species richness overall, likely because it was the oldest of the three slides and
successional processes had allowed more species to establish. However, on the Cecil Lake
landslide the leading species was still a fern ally (Equisetum arvense), while the second
leading species was a shrub (Alnus viridis ssp. sinuata).
Evenness measures how evenly species abundances are distributed in a plant community, and
as its value increases, so does diversity. Mean Pielou’s evenness, J, was higher on the
undisturbed terrain study sites compared to the landslide study sites for all three study areas,
and all three undisturbed study sites had almost identical evenness values. In addition, the
Beatton River and Cecil Lake study areas had identical evenness values on the landslide
study sites, although Beatton River had slightly more variability. The paired Hasler Flats
study sites had a much smaller difference between each other compared to Beatton River and
Cecil Lake, and the Hasler Flats landslide had the highest evenness of the three landslide
study sites. Overall, these findings suggest that the landslides are very similar in distribution
of species abundances, despite the differences in sizes and ages. However, analysis of the
top ten leading species on each study area revealed that although the three landslides all had
high values of Equisetum arvense, Rubus idaeus, and Populus tremuloides, they each had a
different array of leading species. The differences in species composition and abundances
are most likely due to the propagule sources and amounts, as well as the available substrates
on each landslide.

83

The Shannon index, H’, and Simpson’s index of diversity, 1-D, both indicated that vegetation
was more diverse on the undisturbed terrain for Beatton River and Cecil Lake study areas but
was very similar for the paired Hasler Flats study sites. Mean Shannon indices and mean
Simpson’s indices of diversity were higher on the undisturbed terrain than on the landslide
for the Beatton River and Cecil Lake study areas, but slightly lower on the undisturbed
terrain compared to the landslide for the Hasler Flats study area. The Beatton River study
area had the biggest difference in both Shannon and Simpson’s diversity between its paired
study sites. The lower diversity on the Beatton River and Cecil Lake landslide study sites
compared to the surrounding undisturbed terrain may be due to vegetation dynamics
influenced by disturbance. Vegetation dynamics are driven by site availability, species
availability, and species performance, and changes in any of these conditions can alter plant
communities (Pickett et al. 2009). On the Beatton River landslide, Melilotus spp. and other
exotics dominated on large portions of recently disturbed substrate, while on the Cecil Lake
landslide, Alnus spp. and Equisetum arvense were widespread. Vegetation dynamics are
intrinsically connected to landscape ecology, disturbance ecology, competition, invasion
ecology, and community assembly. The process of recovery of landslide surfaces is complex
due to the high spatial and temporal variability of soil stability and fertility (Walker et al.
2009). Surface soil erosion and patchiness of soil fertility can significantly hinder plant
community development.
The minimal difference in Shannon and Simpson’s diversity between the landslide and
undisturbed terrain at Hasler Flats compared to the Beatton River and Cecil Lake study areas
may be because the Hasler Flats landslide was much smaller, more gently sloped, and had a
much high proportion of rafted material originating from the surrounding terrain, providing a

84

mosaic of stable and fertile substrates and propagules for new vegetation. Further, the
adjacent cutblock above the headscarp at Hasler was approximately 15 years old, with
established shrubs and forbs in the understory. A greater diversity of plant species was
available from this source compared to above the headscarps at the Beatton River or Cecil
Lake study areas.
The biological legacies or residuals left after disturbance seem to be interacting with
landslide size, age, and disturbance intensity in the study areas to influence vegetation
dynamics, in agreement with other studies (Turner et al. 1998). Residual vegetation from
vegetated rafts and chunks of intact soil can spread via seed banks, propagules, suckers,
rhizomes, or serotinous cones. The life history traits of plants present at the time of
disturbance interact with the disturbance intensity to influence the species composition of
residuals. Succession occurs on a continuum of the role of residuals compared to new
invaders, as well as a separate continuum of soil development (Franklin et al. 2000; Franklin
and MacMahon 2000; Dale et al. 2005). Residuals are affected by the spatial variability and
intensity of the disturbance, and thus larger disturbances such as the Cecil Lake landslide
may present a greater degree of uncertainty and variability in successional pathways (Foster
et al. 1998; Turner et al. 1998).
Hill numbers represent true diversity, or the effective number of equally abundant species
required to achieve a specific diversity measure value. For the Cecil Lake and Hasler Flats
study areas, the Hill number q = 0 (richness) was higher on the landslide study sites, whereas
on the Beatton River study area, q = 0 was higher on the undisturbed study site. However,
for all other Hill numbers (q = 1, 2, and ∞), the value was higher on the undisturbed terrain
for Cecil Lake and Beatton River but higher on the landslide for Hasler Flats. These findings
85

are consistent with the trends from the calculations of Shannon and Simpson’s diversities.
The Hasler Flats study area has the smallest Hill number differences between paired study
sites, while the Beatton River study area shows the biggest differences between paired study
sites. These findings indicate once again that the Beatton River study area has the biggest
difference in diversity between landslide and undisturbed terrain, and the Hasler Flats study
area has the smallest difference. The lower diversity on the Beatton River landslide study
site compared to the undisturbed terrain is most likely attributable to the young age of the
landslide and the fact it is still quite active. In addition, the proximity of the hay field just
above the scarp, along with the steep slope and prevailing winds, could have contributed to
abundant reseeding with exotic species such as sweetclover (Melilotus spp.) as well as
noxious weeds such as thistles (Sonchus spp., Cirsium spp.).
As Hill numbers reached q = ∞, the differences on the undisturbed study sites became less
different from each other, with values for all three sites being quite similar but all still higher
than the landslide study sites. However, the landslide study sites maintained a distinct
difference in value between each other, with the Hasler landslide study site having the
highest Hill number value and the Beatton landslide study site having the lowest value. The
maintenance of this difference indicates the sites are robustly different from each other in
their diversities, and this difference appears to be influenced by spatial and temporal factors.
Hasler Flats is the smallest landslide while Beatton River is the youngest landslide.
The distribution of growth forms on a disturbed site can change over time and is an indicator
of successional status (Dale and Adams 2003). The plant growth form abundances for the
relevés differed significantly between paired study sites (landslide vs. undisturbed) and also
between landslide study areas. On the Beatton River landslide study site, most of the cover
86

was forbs (>75% of total cover), followed by fern and fern allies (20% of total cover) and
then distantly by shrubs (<3% of total cover). On the Beatton River undisturbed study site,
shrubs were leading, followed by forbs and then trees and graminoids. The high dominance
of forbs on the Beatton River landslide study site may be due to adjacency to a field of mixed
forage species. In addition, the Beatton River landslide appears to be more active and is
steeper than the other two slides, which could prevent more persistent woody species from
establishing on the headscarp and large secondary scarps. The Beatton River landslide also
had less pond cover and fewer wet areas in general, indicating a limited hydrological system.
Although a large proportion of the forb species on the Beatton River landslide was exotic
invasives, the landslide surface could still eventually become populated with persistent native
species, as introduced species may switch from competition to facilitation in their
relationship with native species. In a study of revegetation on a landslide caused by the
eruption of Mount St. Helens, vegetation plots populated with nonnative species had higher
vegetation cover and more native species richness than those sites not invaded, indicating any
type of invasion may facilitate primary succession (Dale and Adams 2003). Annual plant
species were most common in the first two years after the eruption, likely the result of wind
dispersal and the ability to grow in poor soil. In that study, the introduced species were more
successful due to their ability to establish, spread, and fix nitrogen. On the Beatton River
study area, the prevalence of shrubs and forbs over trees on the surrounding undisturbed
terrain is reflected in the dominance of the first two growth forms on the landslide. Although
the landslide was surrounded mostly by mature mixedwood and deciduous forests, a
significant portion of the landscape was a grassland/shrubland complex.

87

For both Hasler Flats and Cecil Lake landslide relevés, shrubs dominated, followed by ferns
and fern allies. The dominance of shrubs on both landslide sites may be due to propagule
contributions from rafts, as well as the advanced age of the Cecil Lake landslide, which could
have allowed time for shrubs to seed in via wind and animal dispersal. On the undisturbed
terrain, the leading growth form for Hasler Flats was again shrubs, while on the Cecil Lake
undisturbed site trees dominated. The presence of more shrubs on the Hasler Flats
undisturbed site may be the result of the deciduous, somewhat open overstory, while the
Cecil Lake undisturbed terrain overstory was mainly coniferous or mixedwood with a denser
canopy. As deciduous stands shed their leaves in the autumn, more sunlight and nutrients
reach the understory, stimulating growth of shrubs in the spring.
A comparison can be made of the autecology of the dominant species on the three different
landslide surfaces. On the Beatton River landslide, the youngest slide, six out of the top ten
most abundant species were exotics. Most of these exotics are perennials and can spread by
both seeds and rhizomes. Wind and water are the primary modes of seed dispersal. The
seeds were most likely transported downslope from the field above via prevailing winds and
runoff. Most of the species are also seed-banking and have a longer range of seed viabilities
or “shelf life” (FEIS 2023). The most abundant species on the Beatton River landslide, the
exotic forb Melilotus officinalis (yellow sweetclover), dominated the toe of the landslide. It
is an aggressive invader but is also a nitrogen-fixer, thus serving to improve the soil for
future successional species. Melilotus officinalis is usually a biennial but is sometimes an
annual or short-lived perennial and can have viable seed for up to 30 years (FEIS 2023).
Thus, seed could be stored in the soil and reemerge following a future movement of the
landslide. The four native species in the top ten most abundant list for the Beatton River

88

landslide are the fern ally common horsetail (Equisetum arvense), the forb mugwort
(Artemisia sp.), the shrub red raspberry (Rubus idaeus), and the forb Lindley’s aster (Aster
ciliolatus). All four species are perennials and can spread by both spores/seeds and
rhizomes.
At Hasler Flats, the second oldest and the smallest landslide, seven of the ten most abundant
plant species were shrubs, and there was one tree species, one fern ally species, and one moss
species. The five most abundant species were the fern ally common horsetail (Equisetum
arvense), followed by the tree species trembling aspen (Populus tremuloides), and the shrubs
prickly rose (Rosa acicularis), green alder (Alnus viridis ssp. sinuata), and red raspberry
(Rubus idaeus). None of the top ten species were exotic. This difference in species
composition and abundances compared to Beatton River can be explained by the presence of
ample native seed sources in the form of multiple rafts down the slope of the slide, in a series
of horsts and grabens. These rafts often spanned much of the width of the slide, providing
many potential seed sources to populate exposed soil. In addition, because the slide was so
small (<2 ha) and was abutted by mature forests on both flanks, seeds and propagules could
be readily dispersed from the adjacent forest.
On the Cecil Lake landslide, which was the oldest of the three slides and was 15 years older
than the Hasler slide, five of the top ten most abundant species were shrubs, three species
were trees, one species was a fern ally, and only one species was a forb. There were no
exotic species in the list of the ten most abundant species. The five most abundant species
were the fern ally Equisetum arvense, the shrub Alnus viridis ssp. sinuata, the forb palmate
coltsfoot (Petasites frigidus), an unknown willow species (Salix sp.), and the tree balsam
poplar (Populus balsamifera ssp. balsamifera). The common horsetail Equisetum arvense is
89

a moisture indicator. It absorbs silicon from the soil, and then in turn can absorb excess
moisture. Equisetum arvense has a perennial rhizomatous stem system which can extend into
the soil up to 1.8 m (FEIS 2023), and it also reproduces via spores. Alnus viridis ssp. sinuata
also prefers moist sites. It disperses seed over long distances by wind or water and can fix
nitrogen and stabilise slopes (Haeussler 1990). Petasites frigidus grows on moist to wet sites.
It is a perennial forb that expands via a creeping root. Salix spp. in general prefer moister
sites and produce large amounts of seeds which are dispersed by wind and water and
germinate best on exposed mineral soil. Salix spp. also reproduce by sprouts, which can grow
very rapidly. Populus balsamifera ssp. balsamifera is an indicator of high moisture and
nutrients. Seeds are easily dispersed long distances by wind. Populus balsamifera ssp.
balsamifera can regenerate from root suckers, stump sprouts, and buried branch pieces.
Suckering is most abundant where mineral soil is exposed.
The composition of the ten most abundant plant species on the Cecil Lake landslide is
indicative of progression toward a more mature forested community. This makes sense, as
the Cecil Lake landslide is the oldest of the three landslides. However, observations in the
field showed large areas of the landslide at the north-facing headscarp that were still bare or
sparsely vegetated. This delayed succession condition contrasts with a study of landslide
recovery on a volcanic site (Saito et al. 2021), which found that grass vegetation recovery is
quicker on shady slopes due to moisture and can recover within 12 years. It is likely the
slower recovery of the north-facing headscarp at Cecil Lake is due to the colder climate of
northeastern BC compared to Japan. In addition, there was evidence of reactivated mass
movements along the Cecil Lake headscarp.

90

The two sample Kolmogorov-Smirnov (K-S) test showed the relevé communities of landslide
vegetation were significantly different from the undisturbed plant communities for Beatton
River and Cecil Lake study areas. These findings were consistent with the diversity values
for the same relevés, which also indicated significant differences between the paired study
sites for these two study areas. The K-S two-sample test also showed the two Beatton River
study sites to be significantly more different from each other than the two Cecil Lake study
sites were from each other. This pronounced difference on the Beatton River study area may
be due to the younger age and active nature of the Beatton River landslide, in concert with
the diversity of ecosystems (grassland, shrubland, mixedwood, mature aspen, mature conifer)
in the surrounding undisturbed terrain. The K-S test did not show a significant difference in
plant communities between the landslide and undisturbed study sites for Hasler Flats. This
result also agreed with the diversity indices findings which showed that the plant
communities were very similar on the landslide and on the undisturbed terrain.
NMS ordination of plant and environment data from BEC plots
In general, the NMS ordination of the plant and environment data from the BEC plots
showed a combination clustered-scattered pattern, with few clear, strong associations. The
concentration of more than half of each set of landslide plots in a central cluster surrounded
by a somewhat uniformly scattered distribution of the remaining plots seems to indicate a
significant degree of similarity among all three landslides. This redundancy may be due to
the early successional stage of most of the plant communities, with different species still
trying to establish and occupy a niche.

91

Although the ordination biplot did not reveal overly strong patterns, it did indicate an
association of certain plant communities with landslide age. For the Beatton River landslide,
the youngest slide, the plots that weren’t clustered at the centre were largely associated to the
open/grassland side of the gradient. In contrast, the scattered Hasler Flats plots were mainly
distributed on the left side of the biplot, indicating forested communities. The plant
communities of the oldest landslide, Cecil Lake, appear to be much more diverse than on the
other landslides, as the plot locations on the biplot that weren’t clustered at the centroid were
scattered throughout much of the species space.
For all three study areas, the plant communities of the benchmark plots in the undisturbed
terrain appear markedly different from the landslide plots, as they all occur as outliers on the
biplot. This increased distance in species space confirms field observations that the plant
communities of the surrounding undisturbed terrain were distinctly different from the
landslide communities. The benchmark plots are also quite different from each other, as they
are all far apart on the ordination biplot. The Beatton River and Hasler Flats benchmark tree
plots in undisturbed terrain are closer to each other on the biplot and thus more like each
other in composition than either of the plots is to the Cecil Lake tree plot. This similarity
between the two study sites is consistent with the field data, as the Beatton River and Hasler
Flats plots were in deciduous stands, while the Cecil Lake plot was in a mixedwood stand.
Of the environmental variables tested in the biplot, material origin and slope gradient appear
to be bigger drivers of plant communities than mesoslope position, soil moisture regime, or
heat load index. However, mesoslope position seems to have a greater influence on plant
composition than soil moisture regime or heat load index. These findings are consistent with
research on coastal BC comparing landslide revegetation with that of adjacent logged areas
92

of the same age (Smith 1986), where the main factors influencing the rate and pattern of plant
recovery on the bare surfaces were associated rock types and slope position. On the
landslides, the development of vegetation and forest productivity were influenced by the
post-disturbance soil materials, which were modified by mixing, surface erosion, and
sloughing. Smith (1986) found the revegetation rate was slower on the steep upper and
middle thirds of a landslide compared to the lower slopes, where revegetation was quite
rapid. He identified a gradient of plant communities along the slope gradient of the
landslides. Conifers dominated on the middle and upper slide zones, especially on coarse
and acidic materials. Red alder (Alnus rubra) was predominant on the lower parts of the
slides, particularly in fine-textured materials. Shrub and bryophyte covers were higher on the
logged areas, while the forb cover was the same on the landslide and the logged area.
Landslides had a much lower productivity, producing only a third as much wood volume as
the harvested areas after 60 years.
In the Peace River Region study areas, field observations did not fully support the findings of
Smith (1986) regarding the plant community gradient – slope gradient relationship. On the
Hasler Flats landslide, shrubs and trees were interspersed down the length of the slope. On
the Cecil Lake landslide, there were very few trees on the upper slope, scattered mixedwood
rafts, young aspen (Populus tremuloides) patches and alder (Alnus spp.) swales on the middle
slopes, and young (~10 years old) spruce (Picea glauca) and aspen scattered throughout the
hummocky toe of the slide. The Beatton River landslide had very little tree cover overall,
apart from some small rafts and a patch of Picea glauca seedlings in a rocky, rubbly portion
of the slide. The main vegetation was the expanse of sweet clover (Melilotus spp.), thistle
(Sonchus spp.), and other invaders occupying the hummocky toe of the slope. These

93

differences may be due to the younger ages of the landslides in this study, as well as the very
different climates between the two studies.
Positive Pearson’s correlations associated with the apparent successional plant community
gradient (Axis 1) are strongest for exotic and grassland species and negative correlations are
strongest with trees and shrubs. For Axis 2, the apparent moisture gradient, the species most
positively associated were the shrub Ribes triste, followed by the forb Trifolium pratense,
and then the moisture indicators Equisetum arvense (fern ally) and Populus balsamifera ssp.
balsamifera (tree).
2) To what extent do landslides rearrange relative abundance of site series/types on a slope
compared to adjacent undisturbed terrain?
Site types/series
Landslides appear to rearrange relative abundances of site types to varying degrees compared
to the adjacent terrain, and this variation seems to be driven by the age of the landslide at
least to some extent. Overall, the general trend for all three study areas was more site
series/types on the landslide than in the surrounding undisturbed terrain, or in situations
where there was the same number of site series, they differed in classification between the
two paired study sites. Another general trend was a greater number of individual polygons
on the landslide for each site series, even when the proportion of the site series was similar
between slide and undisturbed sites. In the Beatton River study area, the landslide study site
had a higher percentage of mesic site series and a higher ratio of deciduous to coniferous
mesic sites compared to the surrounding undisturbed terrain. On the Beatton River
undisturbed terrain study site mesic sites also comprised the largest proportion of the total
94

area. The Beatton River landslide and undisturbed study sites each contained a similar
number of mesic polygons but the mean sizes and ranges were markedly different, and most
mesic polygons were larger on the undisturbed terrain. The landslide also had a higher
proportion of dry sites than the undisturbed terrain. Overall, there were more site series
polygons on the Beatton River landslide compared to the undisturbed terrain. The slide had
five times as many ponds, but the average pond on the undisturbed terrain was six times
larger.
Although there were some similarities with the other two study areas, the Cecil Lake study
area had the most notable differences in site type distribution between landslide and
undisturbed terrain among the three landslides. On the Cecil Lake study area, for both study
sites mesic types dominated with deciduous 101$ greater than coniferous 101. In contrast,
the next two most abundant site series/types on the landslide were wet types (111$6B.1 and
111), while on the undisturbed terrain the next two most abundant site series/types were very
dry types (Gg -grassland and Rt -talus). The main reason for the difference of moisture
gradient on the undisturbed terrain appears to be the influence of the south-facing slopes
across the creek, whereas the Cecil Lake landslide itself is north-facing and seems to be
heavily influenced by groundwater. The Cecil Lake landslide had more than four times as
many site type/series polygons as the undisturbed terrain, but almost 75% of these were
ponds. The average size of the ponds on the undisturbed terrain was almost five times
greater than those on the landslide. On the slide, the ponds appeared in clusters just below
the scarp and on the toe of the slope, where the terrain was more uneven and marked with
depressions. The ponds in the undisturbed terrain were more stable and vegetated. Overall,
the age, size, geology, and aspect may all have contributed to the distinctness of site type

95

distribution on the Cecil Lake landslide compared to Beatton River and Hasler Flats
landslides.
The Hasler Flats study area yielded site type distributions most like what might be expected
when comparing landslides and undisturbed terrain. On the Hasler Flats study area, for both
study sites the mesic 101$ was leading, followed by the much wetter 112(Fm02). However,
on the undisturbed terrain 101$ was a much higher proportion than on the landslide. Further,
on the landslide the third most abundant site series was the very dry 103$, while on the
undisturbed terrain the third most abundant site series was the very wet 111$6B.1. These
results indicate that on this smaller, younger landslide, mesic sites are reduced overall and
there is more diversity of dry and wet site types compared to the undisturbed terrain. The
landslide also has many more polygons and a few more site series than the undisturbed
terrain. The mean size of the pond (W) polygons was much greater on the landslide, and
there were 13 times more pond polygons than on the undisturbed area. The findings
regarding larger average pond size on the landslide are opposite of the Beatton River and
Cecil Lake study areas, but the findings about the greater number of ponds on the landslide
compared to the undisturbed terrain were the same as for Beatton River and Cecil Lake. The
larger pond size on the Hasler Flats landslide is likely due to the series of horsts (ridges) and
grabens (trenches or depressions), with water accumulating in the grabens. The surrounding
terrain is much more uniform, with fewer depressions containing persistent water. Overall,
the contrasts in the Hasler Flats study area site type distributions are due to the movement
type and subsequent features (horsts/grabens), and these differences are more pronounced
because of the small size of the landslide.

96

The pond distribution patterns on the three landslides are consistent with a study in Japan
(Takaoka 2015), which found that ponds mainly occur in displaced masses downslope from
ridges and in linear depressions along main ridges. This pattern was most pronounced on the
Cecil Lake landslide, which also had numerous small ponds at the base of the headscarp.
The Japanese study also found that 90% of the ponds on the landslide were <1000 m2. This
finding is consistent with the three landslides of the present study, where average pond size
was 54.5 m2 (Beatton River), 97.1 m2 (Cecil Lake), and 56.9 m2 (Hasler Flats). The
characteristic of small landslide pond size is also confirmed at a regional scale in the pond
mapping exercise in Chapter 4 of this study. Although ponds on landslides occupy small
areas, they are a very important component of biodiversity. They provide connectivity,
shelter, and a water source for larger animals such as ungulates and birds. Pond size
influences species richness, as a series of smaller ponds have more species and higher
conservation value than a single large pond of the same total area (Oertli et al. 2002).
Alpha diversity of site types on landslides appears to be influenced by the passage of time.
Site series/type diversity was greater on the landslide than the undisturbed terrain for the
Cecil Lake and Hasler Flats study areas in terms of Shannon diversity, Pielou’s evenness, and
Simpson’s index of diversity. Richness was equal on the Cecil Lake paired study sites but
higher on the Hasler Flats undisturbed terrain. Overall, the alpha diversity findings for Cecil
Lake and Hasler Flats confirm observations in the field and indicate that site series are more
diverse on older landslides compared to the surrounding undisturbed area. Conversely, site
series diversity was greater on the undisturbed terrain for the Beatton River study area for all
measures except richness, which was equal on the paired study sites. The findings for the
Beatton River study area are consistent with observations in the field and indicate that site

97

series diversity is reduced on more recent, steep landslides which are still in a state of
heightened activity. The lower site series diversity indices on the Beatton River landslide
study site compared to the surrounding terrain and the other landslides are likely due to the
landslide having large areas of uniform terrain and fewer rafts. The Hasler Flats study area
shows the largest positive difference in site series diversity indices between the landslide and
the undisturbed terrain. This difference may be because not only does the Hasler Flats
landslide contain rafts and remnants of the surrounding more developed plant communities,
but it also has ridges and depressions where new species have established, and new
combinations of soil moisture and soil nutrient regimes are created. The small size of the
Hasler Flats landslide also increases the proportions of those site series found on rafts.
Overall, the differences in site series diversity and distributions between the Peace River
Region study areas may be driven by both initial pre-slide vegetation and slope conditions
and post-slide species interactions and geomorphic alterations over time. All three landslides
in this study had evidence of past mass movements in the surrounding terrain, but the degree
of diversity created by these past events differed between study areas. Vegetation may
modify slope stability, thus influencing landslide size and severity, and at the landscape level
vegetation may influence the spatial distribution of landslides through its interaction with the
substrate (Restrepo et al. 2009). There may be both downslope and horizontal environmental
gradients on a landslide, with conditions usually the mildest at landslide edges and harshest at
the landslide centre. Over time, ecosystems on landslides can reorganise and embark on
different successional trajectories, due to time lags caused by changing interactions, in
addition to feedback between biota and substrate attributes. For example, differences in leaf

98

litter chemical makeup and rate of decomposition between some early successional species
may influence organic matter dynamics and nutrient cycling rates (Shiels 2006).
3) Are landslides significantly more geomorphically diverse than the surrounding
undisturbed terrain?
Geotyping
Geodiversity indices have been increasingly used to try to quantify geophysical diversity on
larger areas as a proxy for species diversity (Wallis et al. 2021). The geodiversity index is
the sum of the spatial diversity of multiple environmental variables measured within each site
and its surroundings, and comprises climate, habitat, and soil. The index quantifies the
degree of variation across space. However, there is mixed evidence as to how effective
geodiversity is at predicting biodiversity and ecosystem functions at the regional scale, and
Wallis et al. (2021) found that climate variables are more important predictors than habitat
and soil variables at the regional scale. Overall, the results in that study showed the
geodiversity index only explained a small amount of the variation in plants and ecosystems,
while environmental conditions and resources explained most of the variation.
While newer geodiversity indices may have limited utility, the geotype diversity work done
in the present study with traditional diversity indices of Shannon and Simpson can be used
for benchmark comparisons with plant and site series diversity. In general, the findings of
the study highlight a relationship between geomorphic diversity and time, in view of the ages
of the landslides. The geotyping analysis revealed that in terms of geomorphic diversity
Cecil Lake, the oldest landslide, had the lowest Shannon diversity index, Pielou’s evenness,
and Simpson’s index of diversity, while the Beatton River landslide yielded the highest
99

richness, Shannon diversity index and Simpson’s index of diversity. The Hasler Flats
landslide registered the lowest richness but had higher Shannon diversity and Simpson’s
index of diversity than the Cecil Lake landslide. The overall higher geomorphic diversity
indices on the Beatton River landslide compared to the other landslides was likely because
the slide was very recent, and there was still large-scale movement occurring. In addition,
the Beatton River landslide was situated in more complex stratigraphy than either Cecil Lake
or Hasler Flats, with exposed bedrock and boulder fields resulting from rockfall. The
Beatton River landslide had some much steeper slopes, which were susceptible to slumping.
Many of the geotypes on the Beatton River landslide—such as aprons, scarps, rubble, and
blocky areas—were sparsely vegetated due to either their steepness or the lack of organic
material. Sparse vegetation may also result from greater distances from a dispersal source.
Studies have found revegetation occurs inward toward the slide body from the landslide
edges and outward from islands or rafts of vegetation (Francescato et al. 2001). Some of the
most sparsely vegetated geotypes at Beatton River were in the centre of the slide.
The Cecil Lake landslide had the lowest geotype diversity indices of the three slides, except
for richness. This seems contradictory at first, considering the variety of features observed in
the field. However, features weather and become less distinct over time. The Cecil Lake
landslide was very large, and some expansive areas were occupied by just one or two
geotypes. Some examples of these geotypes include the long headscarp, as well as the large
area of hummocks at the toe of the slide. The headscarp was quite sparsely vegetated, even
though the landslide was twenty years old at time of sampling. The headscarp also still had
mainly exotics and native grasses. This condition is likely due to the width and steepness of
the slope, which can slow the rate of successful revegetation (Francescato et al. 2001).

100

Further, although most revegetation is initiated from the landslide edge, a significant length
of the headscarp was bordered by a cultivated hay field. Therefore, the main source of plant
propagules was likely from the field. Francescato et al. (2001) found that revegetation rate
on landslides decreases over time, likely due to early saturation by easily established species.
This situation may be occurring on the Cecil Lake landslide headscarp, due to the
colonisation by agronomic species from the adjacent field.
Hasler Flats and Cecil Lake landslides both contained ridges, with those on Cecil Lake much
more prominent and extensive. The larger ridges on the Cecil Lake landslide were likely
because the disturbance area was much bigger and steeper than the Hasler Flats landslide, so
the material gained more momentum as it moved downslope. In addition, the material on the
Cecil Lake landslide was silty clay, while on Hasler Flats the material was fine sand for the
first four metres of thickness. The clay ridges would likely persist longer, while the sand
ridges would not be as pronounced from the start, and they would become subdued with
weathering.
Although ponds did not occupy a notable amount of area on any of the slides, they were
numerous and distributed throughout, especially on the Cecil Lake landslide. The trend of
high numbers of small ponds was evident over a large area of the region, as described in
Chapter 4. This abundance and widespread distribution of ponds indicated the presence of
depressions throughout the slides, as well as the level of the groundwater. The ponds were in
different stages of development depending on the slide and specific locations within the slide.
Ponds provide habitat for many different types of wildlife. Evidence of wildlife use of the
ponds was present, especially signs of beaver (Castor canadensis) activity such as felled
trees, gnawed bark, and well-worn trails.
101

In all three study areas, vegetated rafts were a significant component of the landslide area.
Rafts and other fragments of vegetative legacies are important for initiation of succession
following disturbances. In one of only a few detailed quantitative studies of northern British
Columbia landslides, Smith (1986) compared revegetation on landslides with surrounding
logged areas as well as old growth areas on various islands of the northern coastal
archipelago of Haida Gwaii. Initial revegetation on those landslides was shown to depend on
the availability of stable microsites and islands of debris and other remnant organic material
(Smith 1986).
Microtopography/surface roughness
It is assumed that over time, landslides will undergo smoothing of the surface due to slope
degradation and local deposition of materials. The microtopography exercise in this study
showed that all three landslides had a high degree of variability around the mean coefficient
of variation (CV) of elevation on microsites and also a high mean CV overall, indicating
variability of elevation is high both within sites and among sites. The most recent landslide,
Beatton River, had the highest variability on average. This higher surface roughness could
be due to the greater steepness of the slope in many places, as well as the presence of more
rubbly areas and exposed bedrock and rockfall. It should be noted that only drone data were
available for the Beatton River landslide for this exercise, so all elevation points were
unclassified. Therefore, some of the points could be vegetation rather than ground points.
However, there were only a few heavily vegetated plots, so the results are largely
representative of ground surface elevations. The Beatton River landslide also had the highest
range of elevational variability, which could be explained by somewhat equal proportions of
steep, smooth slopes, rubbly moderate slopes, and hummocky gentle slopes at the toe.
102

The Cecil Lake landslide had the second highest elevational mean, maximum, and range of
CV, even though it was older than the Hasler Flats landslide. This higher surface roughness
could be mainly due to the steeper slope of the Cecil Lake landslide compared to the Hasler
Flats landslide. Steeper slopes may increase the diversity of elevation variation as material
falls and slides, creating rubbly piles at the same time as smooth scarps form.
When compared with field photographs taken at the microsite level, individual plot
elevational CV results were not always representative of on-the-ground plot characteristics.
For example, on the Beatton landslide one of the highest CVs (Plot 1 at 68.58%) was a flat
grassy raft. The high CV calculation may be due to the vegetation present, as bare ground
points were not filtered out. Overall, however, the elevation CVs appeared to accurately
reflect the plot characteristics. The results were also consistent with other microtopography
research (Rozycka et al. 2016).
4) Is vegetation diversity on landslides significantly related to geomorphological diversity?
One of the key premises of this study was that the diverse array of geomorphic features on
landslides would positively influence the diversity of vegetation communities. Regression
analyses were performed for each landslide to assess whether there were correlations
between the relevé alpha diversity values of vegetation richness, Shannon index, evenness,
and Simpson’s index and the same measures for geomorphic diversity, using the geotyping
results obtained for both on the landslide and in the undisturbed terrain. All four regressions
showed that there was not a significant relationship for any of the diversity indices tested in
any of the three study areas. This was somewhat surprising, considering how field
observations suggested that vegetation and site diversity were closely related. There was not

103

even a relatively stronger, but not significant, relationship on the landslides compared to the
undisturbed terrain, as the R2 and p-values were similar for both study site categories. A
possible explanation could be that more successional time is required before true
relationships have become established and are statistically discernible. Alternatively, it could
be that the scale of the exercise masked any relationships, as the geotyping was done at the
landscape level. It is possible the sample size of three relevés per study site was simply too
small to obtain an accurate result. It is also possible a different analysis would be more
appropriate. More study is required on this aspect of the research.
Limitations
One of the limitations of the study design was the fact that study areas were partly chosen
based on access. This naturally resulted in selecting landslides with road access and
increased the likelihood that the slides would be next to a field or other managed landscape.
The presence of development in turn likely influenced the establishment of exotics from the
adjacent field. A more random selection of study areas could result in some more remote
landslides where revegetation is primarily influenced by native forest vegetation from the
surrounding terrain. Alternatively, a deliberate selection of some more remote landslides
would have allowed for a focused or comparative study of natural succession in contrast with
succession influenced by exotic species. In addition, it would have been beneficial to study
more landslides, to better assess correlations between vegetation and geomorphic diversity.
A further limitation was the small sample size of three landslides. Each landslide was unique
in age, size, and slide type, which presented challenges when attempting to generalise
findings. The differences among the landslides did provide a basis for identifying spatial and

104

temporal trends, which is very useful as a baseline dataset. However, a series of landslides
from different age classes, size classes, and landslide types would help to complete the
picture. The sample size for this study was selected based on time and resource constraints,
but future studies could expand on the work presented here.
An additional limitation of the study was the lack of sufficient LiDAR bare ground point data
for assessing microtopography on undisturbed terrain. This deficit of imagery precluded a
robust analysis of the undisturbed microtopography and did not allow a full comparison with
the landslide study sites. The microtopography assessment component was added several
years into the study and thus was not planned for, but for future landslide studies researchers
should ensure adequate LiDAR is available before setting objectives and delineating study
parameters where mapwork is planned. A related limitation was the limited availability of
the LiDAR mosaic high-resolution imagery for the site series and geotype mapping on the
undisturbed terrain for each study area. The outer perimeters were thus determined in part by
the boundaries of the available high-resolution imagery.
The timing of the field data collection posed another possible limitation. Due to funding
availability and other scheduling factors, vegetation plot sampling was done from early
summer into the fall. During the late summer and fall sampling, there was less foliage and it
was thus more difficult to identify species and estimate cover. However, extra effort was
made to identify species and estimate cover based on stems, fallen leaves, and plant crown
expanses. Therefore, it is believed that species identification and cover estimates were
accurate within an acceptable margin of error. Regardless, it is still preferable to conduct
sampling when plants are at their most robust and easily identifiable during the growing
season.
105

A further limitation of the study design was the restriction of landslides to the Peace River
Region. The Peace River Region is somewhat unusual in the province in that most slides
occur in glaciolacustrine material. Therefore, the results are more specific to this area and
less generalisable. However, basic elements of succession and site variability can still be
applied to other parts of the province, country, or world to see if the results are comparable.
In addition, the findings could apply to any part of the world that has been affected by
glaciation.
Finally, a limitation of the data collection was the learning curve involved during some field
sampling. This was especially the case when estimating plant species cover, particularly on
the larger 400 m2 plots. The implications of this are that some of the initial plots may have
slightly inaccurate cover estimates. An alternative to this approach could be to have more
than one person estimating the cover on the first few plots, and then compare notes and
calibrate measurements before proceeding, or to take drone imagery of each plot and assess
plant species cover by image analysis.
2.6 Conclusions and recommendations
Some general trends appear in the analysis of biophysical alpha diversity on landslides in the
Peace River region of northeastern BC. Overall, plant communities vary depending on the
age and size of the landslide and the slope and soil development of the various geotypes
present. Exotic forb species are more dominant in the early stages of a landslide’s
successional development. Shrubs and trees become more established as the landslide
stabilises. However, landslides can once again reactivate, setting plant succession back on
some areas of the landslide.

106

The landslides in the study tend to have more diverse site series units compared to an equal
area of the surrounding undisturbed terrain. In general, the landslides contain a larger
proportion of mesic sites compared to surrounding undisturbed terrain, but they also have
more pronounced extremes of site series at either end of the soil moisture regime gradient.
The site series polygons are generally smaller on the landslide, compared to larger swaths of
intact site series in the undisturbed areas.
The landslides in the study area are more geomorphically diverse than the surrounding
undisturbed terrain. Geomorphic types tend to be more diverse on the landslides, due to mass
movement, rearrangement of stratigraphy, and erosion of unstable substrates. Geomorphic
diversity also varies inversely with the age of the landslide. Surface roughness also appears
more diverse on landslides and tends to decline with landslide age. The most recent landslide
had the highest surface roughness.
Ultimately, landslides in the study are generally less diverse than surrounding undisturbed
terrain regarding plant diversity, but more diverse in abundance and distribution of site series
and geomorphology. Although the landslides are lower in plant diversity due to rapid and
wide-reaching establishment by invasive species and persistence of early successional
species, the site series and geomorphological diversity present provide conditions for a
greater variety of plant communities and wildlife habitats over time.
The findings of this study of alpha diversity on landslides in northeastern BC provide the
groundwork for several recommendations regarding landslide restoration, management, and
research. Landslides targeted for restoration should be prioritised and selected based on the
risk of future instability to human safety and health, infrastructure, and ecosystems.

107

Remotely located landslides may be left to revegetate and restabilise unaided if future
movements do not threaten to block waterways or cause heavy sedimentation. Restoration
should first focus on landslides near communities and infrastructure, such as Old Fort and the
Site C hydroelectric dam along the Peace River. Restoration should also focus on landslides
near cultivated areas, as well as next to rivers, creeks, or other water bodies used for human
drinking water or designated as sensitive wildlife habitat.
Restoration of landslides is essentially managed succession (Walker and del Moral 2008).
Long-term research on primary succession on landslides provides information on temporal
vegetation dynamics that serve as lessons for restoration practices. Applicable learnings
include information on site amendments, development of community structure, nutrient
dynamics, species life history traits, species interactions (competition), and modeling of
transitions and trajectories. When considering restoration of landslides, it is important to
realise that no plant community is ever static (Pickett et al. 2009). Successful restoration
depends on accurate identification of where on the temporal gradient of succession a site is
located, and appropriate intervention in disturbance, colonisation, and competition processes.
Landslides typically cause a loss of vertical vegetative structure, soil nutrients, and soil seed
bank (Shiels and Walker 2003), as was observed on the three landslides studied. Restoration
on landslides is particularly challenging due to the high spatial and temporal heterogeneity of
soil stability and fertility (Walker et al. 2009). Surface soil erosion is a serious problem, as is
patchiness of soil fertility. These conditions were especially evident on the Beatton River
landslide but were also widespread along the headscarp of the Cecil Lake landslide.
Persistent soil erosion is the first barrier to succession and restoration on landslides, and

108

measures to reshape slopes are often unsuccessful because of subsequent erosion,
compaction, and other problems (Walker et al. 2009).
Recovery of stable plant communities can be achieved by first stabilising the substrates with
bioengineering techniques such as live stakes, brush layers, and wattle fences (Polster 2003;
Singh 2010; Punetha et al. 2019), supplemented with establishment of early successional and
mid-successional native plant cover and amendment of the soil with nutrients and organic
matter. Successful growth of established plants on landslides may require different
environmental conditions than colonising plants. The emphasis should be on assuring a
diversity of functionally redundant species, so that if some plants do not survive, others with
a similar life history can fill in the gaps (Walker et al. 2009). These species can stabilise and
fertilise the surface and should have rapid growth and reproduction traits. On Peace River
region landslides, native species to use in restoration plantings could include grasses such as
blue wildrye (Elymus glaucus) and Canada bluejoint (Calamagrostis canadensis), forbs such
as wild sarsaparilla (Aralia nudicaulis), palmate coltsfoot (Petasites frigidus), showy aster
(Aster conspicuus) and false Solomon’s seal (Smilacina racemosa) and shrubs such as
prickly rose (Rosa acicularis), saskatoon (Amelanchier alnifolia), alder species (Alnus spp.),
red-osier dogwood (Cornus stolonifera), and highbush cranberry (Viburnum edule). Because
the deposition zone is generally more stable, restoration efforts in this part of the slide should
focus on management of species interactions, such as reducing competition between native
species and exotics such as Melilotus officinalis. Management of interactions should retain
some exotics for nitrogen fixing and nutrient input, but exotic densities should be reduced
around native species. Restoration in the headscarp and chute zones should prioritise
proactive erosion control and enhanced seed dispersal, perhaps using hydroseeding. In

109

addition, planting of native shrubs and tree seedlings could be done throughout the landslide
wherever slopes are already stable, to diversify the species composition and increase the rate
of succession.
Two strategies to accelerate succession on the large bare or sparse zones and the grassdominated areas on the Beatton River and Cecil Lake landslides could be to increase natural
dispersal by birds and amend the substrate surface with nutrients (Shiels and Walker 2001,
2003). The dispersal strategy with birds involves establishing perches in the open areas to
encourage birds to fly closer to the centre of the landslide, where they may drop, regurgitate,
or defecate seeds. The goal is to facilitate the establishment of native woody plants to shade
out the graminoids and aggressive forbs. Success would depend on the bird species and the
proximity of the perches from the forest edge. For seed deposition from birds to occur, there
must be some vegetation on the slide (Shiels and Walker 2001). However, this poses a
problem for restoration, because forest seed germination may be inhibited by competition
with grasses, exotics, and other fast-growing pioneers. Forest seed germination may also be
hindered on bare sites due to low organic matter and extreme microclimates. Commercial
fertiliser and amendments with mature forest soil may help speed up plant colonisation.
However, research has shown that organic matter input on landslides does not influence plant
growth unless it is forest soil (Shiels et al. 2006). The type of organic substrate, frequency of
organic deposition, and presence or absence of biota all directly influence soil nutrient patch
dynamics, which ultimately affects plant development.
Overall, restoration on landslides in the Peace River Region should focus on ecosystem
recovery and biodiversity, rather than plant species composition. It is important to consider
long-term recovery trends and effects of nonnative species succession because restoration
110

practices often incorporate nonnative species (Dale and Adams 2003). Further, the influence
of invasive species and other disturbances needs to be fully integrated into successional
theory.
Biogeomorphic systems such as landslides are viewed as open and path-dependent systems
by many researchers (Stallins 2006). Plant communities change as the substrate changes, and
the communities also actively change the substrate. The material impacts of organisms
serving as ecosystem engineers may be just as important as the trophic links, since they
stabilise substrates, enhance weathering, provide habitat, and promote facilitation. There are
many developmentally connected feedbacks between geomorphic and ecological
components, and multiple causes and recurrences should be considered when studying
succession and planning restoration projects.
Further research on landslides in northern BC could enhance understanding of plant
community development, geomorphic evolution, relationships between geomorphic diversity
and vegetation diversity on landslides, and broader ecological processes on these
disturbances. Resampling of vegetation plots on each of the three study areas after some
years have passed could provide insights on whether species composition and site conditions
are changing or becoming more diverse over time. Sampling on remote landslides away
from the influence of cultivated fields or roadways would create a dataset for comparison of
vegetation dynamics with solely native vegetation. In addition, a variety of landslides of
different movement types, sizes, and ages could be sampled to provide a broader
understanding of geomorphic, spatial, and temporal influences on alpha diversity.
Experiments with bird perches, organic matter amendments, bioengineering, and
hydroseeding on the bare slopes of the slides would present valuable lessons on the
111

feasibility and effectiveness of management options. Finally, detailed studies on wildlife
habitat, wildlife use of landslide features, and connectivity to the surrounding undisturbed
terrain would provide a perspective of the ecological role of landslides at the landscape level.
All these findings could be applied more generally to our understanding of landslide ecology
(Walker and Shiels 2013), and the options for management, and restoration here and
elsewhere in the world.

112

2.7 References
Avery, T.E. 1969. Forester’s guide to aerial photo interpretation. U.S. Department of
Agriculture, Forest Service. Agriculture Handbook No. 308. Washington, D.C. 49 pp.
Bazzaz, F. A. 1975. Plant species diversity in old‐field successional ecosystems in southern
Illinois. Ecology, 56(2), 485-488.
BC Gov (British Columbia Ministry of Forests and Range and British Columbia Ministry of
Environment). 2010. Field manual for describing terrestrial ecosystems, 2nd edn.
BCMFR Research Branch and BCMOE Resource Inventory Branch, Victoria, B.C.
(Reprint with updates 2015.)
BC Forest Service Research Branch. 2016. British Columbia plant species codes.
BC Forest Service Research Branch. 2020. British Columbia plant species codes Version 14.
Beaudry, L., R. Coupe, C. DeLong, and J. Pojar. 1999. Plant indicator guide for northern
British Columbia: Boreal, Sub-Boreal, and Subalpine biogeoclimatic zones (BWBS, SBS,
SBPS, and northern ESSF). Land Management Handbook 46, BC Ministry of Forests,
Research Branch, Victoria, BC
BECWeb. 2023. Biogeoclimatic Ecosystem Classification Program. Information website.
https://www.for.gov.bc.ca/hre/becweb/resources/information-requests/index.html
Belessiotis, V, and E. Delyannis. 2011. Solar drying. Solar Energy, 85: 1665-1691.
Burton, P. 1991. Ecosystem restoration versus reclamation: the value of managing for
biodiversity. Proceedings of the 15th Annual British Columbia Mine Reclamation
Symposium in Kamloops, BC, 1991. The Technical and Research Committee on
Reclamation. pp. 17-26.
Burton, P.J., A.C. Balisky, L.P. Coward, S.G. Cumming, and D.D. Kneeshaw. 1992. The
value of managing for biodiversity. The Forestry Chronicle 68(2): 225-237. DOI:
10.5558/tfc68225-2
Chao, A. 1984. Nonparametric estimation of the number of classes in a population.
Scandinavian Journal of Statistics,11(4): 265-270.
Chilton, R.R.H. 1981. A summary of climatic regimes of British Columbia. Ministry of
Environment, Air Studies Branch, Assessment and Planning Division. 44 pp.
113

Chiu, C-H., and A. Chao. 2014. Distance-based functional diversity measures and their
decomposition: a framework based on Hill numbers. PLoS ONE 9(7): e100014 1-17.
Colwell, R.K., and J.A. Coddington. 1994. Estimating terrestrial biodiversity through
extrapolation. Philosophical Transactions of the Royal Society of London B Biological
Sciences, 345: 101-118.
Cruden, D., and D. Varnes. 1996. Landslide types and processes. In: Turner, A., Schuster, R.
(eds.), Special Report 247: Landslides Investigation and Mitigation. National Research
Council, Transportation Research Board, Washington, DC, pp. 36-75.
Dale, V., and W. Adams. 2003. Plant reestablishment 15 years after the debris avalanche at
Mount St. Helens, Washington. The Science of the Total Environment, 313: 101-113.
Dale, V. H., F.J. Swanson, and C.M. Crisafulli. 2005. Disturbance, survival, and succession:
understanding ecological responses to the 1980 eruption of Mount St. Helens. In: Dale, V.H.,
F.J. Swanson, and C.M. Crisafulli (eds). Ecological Responses to the 1980 Eruption of
Mount St Helens. Springer, New York, NY. https://doi.org/10.1007/0-387-28150-9_1. pp. 311
DeLong, C., A. Banner, W. H. Mackenzie, B. J. Rogers, and B. Kaytor. 2011. A field guide
to ecosystem identification for the Boreal White and Black Spruce Zone of British Columbia.
B.C. Ministry of Forests and Range, Forest Sciences Program, Victoria, B.C. Land
Management Handbook No. 65.
Demarchi, D.A. 2011. An introduction to the ecoregions of British Columbia. Third
Edition. Ecosystem Information Section, Ministry of Environment, Victoria, B. C. 163 pp.
Douglas, G.W., D.V. Meidinger, and J. Pojar (eds). 1998-2002. Illustrated Flora of British
Columbia, volumes 1 to 8. B.C. Ministry of Environment, Lands and Parks, and B.C.
Ministry of Forests, Victoria, B.C.
FEIS. 2023. Fire Effects Information System. Syntheses about fire ecology and fire regimes
in the United States. United States Forest Service, Department of Agriculture. Online source.
https://feis-crs.org/feis/
Fisher, R.A., A.S. Corbet, and C.B. Williams. 1943. The relationship between the number
of species and the number of individuals in a random sample of an animal population.
Journal of Animal Ecology,12:42–58.

114

Foster, D. R., D.H. Knight, and J.F. Franklin. 1998. Landscape patterns and legacies resulting
from large, infrequent forest disturbances. Ecosystems, 1: 497-510.
Francescato, V., M. Scotton, D. Zarin, J. Innes, and D. Bryant. 2001. Fifty years of natural
revegetation on a landslide in Franconia Notch, New Hampshire, U.S.A. Canadian
Journal of Botany, 79: 1477-1485.
Franklin, J.F., D. Lindenmayer, J.A. MacMahon, A. McKee, J. Magnuson, D.A. Perry, R.
Waide, and D. Foster. 2000. Threads of continuity. Conservation Biology in Practice, 1(1): 816.
Franklin, J. F., and J.A. MacMahon. 2000. Messages from a mountain. Science, 288(5469):
1183-1184.
Geertsema, M., J.J. Clague, J.W. Schwab, and S.G. Evans. 2006. An overview of recent
large catastrophic landslides in northern British Columbia, Canada. Engineering Geology,
83: 120-143.
Geertsema, M., L. Highland, and L. Vaugeouis. 2009. Environmental impact of landslides.
In Landslides – Disaster Risk Reduction. K. Sassa and P. Canuti (eds), Springer-Verlag.
Berlin, Heidelberg. pp. 589-607.
Geertsema, M., and J. Pojar. 2007. Influence of landslides on biophysical diversity – A
perspective from British Columbia. Geomorphology, 89: 55-69.
Global Mapper. 2020. Global Mapper v23 64 bit. Blue Marble Geographics.
Goepel, K.D. 2012. Diversity Excel Template – Diversity Indices and True Diversity.
Business Performance Management Singapore. https://bpmsg.com/diversity-18-12-12/
Good, I.J. 1953. The population frequencies of species and the estimation of population
parameters. Biometrika, 40: 237-264.
Gotelli, N., and R.K. Colwell. 2001. Quantifying biodiversity: procedures and pitfalls in the
measurement and comparison of species richness. Ecology Letters, 4: 379-391.
Gower, J.C. 1971. Statistical methods of comparing different multivariate analyses of the
same data. In Hodson, F.R., D.G. Kendal, and P. Tautu, (eds) Mathematics in the
archaeological and historical sciences. Edinburgh University Press, Edinburgh. pp. 138-149.

115

Haeussler, S., D. Coates, and J. Mather. 1990. Autecology of common plants in British
Columbia: a literature review. Forest Resource Development Agreement (FRDA) report 158,
275 pp. Joint publication of BC Ministry of Forests, Research Branch and Forestry Canada,
Pacific Forestry Centre, Victoria, B.C.
Hill, O. 1973. Diversity and evenness: a unifying notation and its consequence. Ecology,
54(2):427-432.
Holland, S.S. 1976. Landforms of British Columbia: A physiographic outline. British
Columbia Geological Survey, British Columbia Department of Mines and Petroleum
Resources, Mineralogical Branch, Province of British Columbia, Bulletin 48. 137 pp.
Howes, D., and E. Kenk. 1997. Terrain classification system for British Columbia (version
2). BC Ministry of Environment, Recreational Fisheries Branch, and BC Ministry of Crown
Lands, Surveys and Resource Mapping Branch, Victoria.
Hurlburt, S.H. 1971. The non-concept of species diversity: a critique and alternative
parameters. Ecology, 52: 577-586.
Johnson, D., L. Kershaw, A. MacKinnon, J. Pojar, T. Goward, and D. Vitt. 1995. Plants of
the western boreal forest and aspen parkland. Lone Pine Publishing and Canadian Forest
Service, Edmonton, Alberta. 392 pp.
Kershaw, L., A. MacKinnon, and J. Pojar. 1998. Plants of the Rocky Mountains. Lone Pine
Publishing, Edmonton, Alberta. 384 pp.
Klinkenberg, B. (editor). 2021. E-Flora BC: Electronic atlas of the plants of British
Columbia. Department of Geography, University of British Columbia, Vancouver, BC. First
Edition 2007.
Kruskal, J.B. 1964. Multidimensional scaling by optimizing goodness of fit to a nonmetric
hypothesis. Psychometrika 29(1): 1-27.
Landau, F. 2004. Plant life forms (growth forms). Article retrieved online from
landau.faculty.unlv.edu/lifeforms.htm
Legendre, P., and L. Legendre. 1998. Numerical Ecology. Second English Edition. Elsevier,
Amsterdam. 853 pp.
Lloyd, M., and R.J. Ghelardi. 1964. A table for calculating the “equitability” component of
species diversity. Journal of Animal Ecology, 33: 217-255.
116

Lord, T.M, and A.J. Green. 1986. Soils of the Fort St. John – Dawson Creek area, British
Columbia. British Columbia Soil Survey, No. 42. Ottawa: Agriculture Canada Research
Branch. 140 pp.
MacArthur, R. H. 1965. Patterns of species diversity. Biological Reviews, 40: 510-533.
MacArthur, R.H. 1972. Geographical Ecology. New York: Harper & Row. 269 pp.
MacKenzie, W. 2012. Biogeoclimatic ecosystem classification of non-forested ecosystems
in British Columbia. Province of British Columbia, Victoria, B.C. Technical Report 068
MacKenzie, W., and R. Klassen. 2004. VPro User Guide. British Columbia Ministry of
Forests, Forest Science Program. Victoria, B.C. 106 pp.
MacKenzie, W., and J.R. Moran. 2004. Wetlands of British Columbia: a guide to
identification. Land Management Handbook No. 52. Research Branch, B.C. Ministry of
Forests, Victoria, B.C.
MacKinnon, A., J. Pojar, and R. Coupe. 1992. Plants of northern British Columbia. BC
Ministry of Forests and Canadian Forest Resource Development Agreement. 352 pp.
Magurran, A. 2004. Measuring biological diversity. Blackwell Publishing. Oxford, UK.
256 pp.
Mather, P.M. 1976. Computational methods of multivariate analysis in physical geography.
J. Wiley and Sons, London. 544 pp.
Mathews, W.H. 1978. Quaternary stratigraphy and geomorphology of Charlie Lake (94A)
map-area, British Columbia. Geological Survey of Canada, Ottawa. Paper 76-20.
Mathews, W.H. 1980. Retreat of the last ice sheets in northeastern British Columbia and
adjacent Alberta. Energy, Mines, and Resources Canada, Geological Survey of Canada,
Bulletin 31, Ottawa. 28 pp.
McCune, B., J.B. Grace, and D.L. Urban. 2002. Analysis of Ecological Communities. MjM
Software Design, Gleneden Beach, OR. 300 pp.
McCune, B., and D. Keon. 2002. Equations for potential annual direct incident radiation and
heat load. Journal of Vegetation Science, 13(4): 603-606.
McCune, B., and M.J. Mefford. 2016. PC-ORD. Multivariate analysis of Ecological Data,
Version 7.0 for Windows. Wild Blueberry Media, Corvallis, Oregon, U.S.A.
McIntosh, R.P. 1967. An index of diversity and the relation of certain concepts to diversity.
Ecology, 48: 392-404.

117

Meidinger, D., and J. Pojar (eds). 1991. Ecosystems of British Columbia. Special Report
Series No. 6. BC Ministry of Forests, Victoria, BC.
Microsoft Corporation. 2010. Microsoft 365: Microsoft Excel. Retrieved from:
https://office.microsoft.com/excel
Morris, E.K., T. Caruso, F. Buscot, M. Fischer, C. Hancock et al. 2014. Choosing and using
diversity indices: insights for ecological applications from the German Biodiversity
Exploratories. Ecology and Evolution, 4(18): 3514-3524.
Motomura, I. 1932. On the statistical treatment of communities. Zoological Magazine
Tokyo, 44: 379-383.
Nagendra, H. 2002. Opposite trends in response for the Shannon and Simpson indices of
landscape diversity. Applied Geography, 22: 175-186.
Noss, R.F. 1990. Indicators for monitoring biodiversity: a hierarchical approach.
Conservation Biology, 4(4): 355-364.
Oertli, B., D.A. Joye, E. Castella, R. Juge, D. Cambin, and J.-B. Lachavanne. 2002. Does
size matter? The relationship between pond area and biodiversity. Biological Conservation,
104: 59-70.
Ortiz-Burgos, S. 2016. Shannon-Weaver Diversity Index. In: Kennish, M.J. (eds)
Encyclopedia of Estuaries. Encyclopedia of Earth Sciences Series. Springer, Dordrecht.
https://doi.org/10.1007/978-94-017-8801-4_233
PC-ORD. 2015. PC-ORD website. https://www.pcord.com/
Peck, J.E. 2016. Multivariate Analysis for Ecologists: Step-by-Step, Second edition. MjM
Software Design, Gleneden Beach, OR. 192 pp.
Peet, R.K. 1974. The measurement of species diversity. Annual Review of Ecology and
Systematics, 5: 285-307.
Phillips, J., and C. Lorz. 2008. Origins and implications of soil layering. Earth-Science
Reviews, 89: 144-155.
Pickett, S., M. Cadenasso, and S. Meiners. 2009. Ever since Clements: from succession to
vegetation dynamics and understanding to intervention. Applied Vegetation Science, 12(1):
9-21.

118

Pielou, E.C. 1966. The measurement of diversity in different types of biological collections.
Journal of Theoretical Biology, 13: 131-144.
Pitkanen, S. 2000. Classification of vegetational diversity in managed boreal forests in
eastern Finland. Plant Ecology, 146(1): 11-28.
Pojar, J., K. Klinka, and D.V. Meidinger. 1987. Biogeoclimatic ecosystem classification in
British Columbia. Forest Ecology Management, 22: 119-154.
Polster, D. F. 2003. Soil bioengineering for slope stabilization and site restoration. Paper
presented Sudbury Mining and the Environment III, May 25-28. Available online at
https://pdf.library.laurentian.ca/medb/conf/Sudbury03/Amendments/122.pdf
Punetha, P., M. Samanta, and S. Sarkar. 2019. Bioengineering as an effective and
ecofriendly soil slope stabilization method: A review. In: Prahdan,S., V. Vishal, T. Singh
(eds) Landslides: Theory, practice and modelling. Advances in Natural and Technological
Hazards Research, vol 50. Springer, Cham. pp. 201-224.
Restrepo, C., L. Walker, A. Shiels, R. Bussmann, L. Claessens, S. Fisch, P. Lozano, G. Negi,
L. Paolini, G. Poveda, C. Ramos-Scharrón, M. Richter, and E. Velázquez. 2009. Landsliding
and its multiscale influence on mountainscapes. BioScience, 59(8): 685-698.
Ricotta, C., and G. Avena. 2003. On the relationship between Pielou’s evenness and
landscape dominance within the context of Hill’s diversity profiles. Ecological Indicators, 2:
361-365.
Rozycka, M., P. Migon, and A. Michniewicz. 2016. Topographic Wetness Index and Terrain
Ruggedness Index in geomorphic characterisation of landslide terrains, on examples from
the Sudetes, SW Poland. Zeitschrift fur Geomorphologie, Supplementary Issue, 1-20.
Saito, H., S. Uchiyama, and K. Teshirogi. 2021. Rapid vegetation recovery at landslide scars
detected by multitemporal high-resolution satellite imagery at Aso volcano, Japan.
Geomorphology, https://doi.org/10.1016/j.geomorph.2021.107989
Sayn-Wittgenstein, L. 1960. Recognition of tree species on air photographs by crown
characteristics. Forest Research Division Technical Note 95. Department of Forestry, Forest
Research Laboratory, Government of Canada, Ottawa, Ontario. 56 pp.
Sayn-Wittgenstein, L. 1978. Recognition of tree species on aerial photographs. Information
Report FMR-X-118. Forest Management Institute, Canadian Forest Service, Department of
the Environment, Ottawa. 106 pp.

119

Shannon, C.E. 1948. A mathematical theory of communication. Bell System Technical
Journal. 27: 379-423. Doi: 10.1002/j.1538-7305.1948.tb01338.x.
Shiels, A. 2006. Leaf litter decomposition and substrate chemistry of early successional
species on landslides in Puerto Rico. Biotropica, 38(3): 348-353.
Shiels, A., and L. Walker. 2001. Accelerating plant colonisation on landslides in Puerto Rico
by additions of bird perches and organic matter. Tropical Ecosystems: Structure, Diversity
and Human Welfare. Proceedings of the International Conference on Tropical
Ecosystems. K.N. Ganeshaiah, R. Uma Shaanker and K.S. Bawa (eds). Published by
Oxford-IBH, New Delhi. pp. 661-664.
Shiels, A., and L. Walker. 2003. Bird perches increase forest seeds on Puerto Rican
landslides. Restoration Ecology, 11(4): 457-465.
Shiels, A., L. Walker, and D. Thompson. 2006. Organic matter inputs create variable
resource patches on Puerto Rican landslides. Plant Ecology, 184: 223-236.
Simpson, E.H. 1949. The measurement of diversity. Nature 163:688.
Singh, A. K. 2010. Bioengineering techniques of slope stabilization and landslide
mitigation. Disaster Prevention and Management: An International Journal, 19(3): 384-397.
DOI:10.1108/09653561011052547
Smith, R. 1986. Soils, vegetation, and forest growth on landslides and surrounding
logged and old-growth areas on the Queen Charlotte Islands. Land Management Report 41.
BC Ministry of Forests, Victoria, BC.
Soil Classification Working Group. 1998. The Canadian system of soil classification. Third
edition. Agriculture and Agri-Food Canada, Research Branch, Publication 1646, 187 pp.
Sokal, R.R., and F.J. Rohlf. 1995. Biometry: The Principles and Practice of Statistics in
Biological Research. Third Edition, W.H. Freeman and Col., New York. 887 pp.
Sousa, W. P. 1984. The role of disturbance in natural communities. Annual Review of
Ecology and Systematics, 15(1), 353-391.
Stallins, J.A. 2006. Geomorphology and ecology: Unifying themes for complex systems in
biogeomorphology. Geomorphology, 77(3-4):207-216.

120

Takaoka, S. 2015. Origin and geographical characteristics of ponds in a high mountain
region of central Japan. Limnology, 16:103-112.
Tokeshi, M. 1993. Species abundance patterns and community structure. Advances in
Ecological Research, 24:112-186.
Turner, M., W. Baker, C. Peterson, and R. Peet. 1998. Factors influencing succession:
lessons from large, infrequent natural disturbances. Ecosystems, 1(6): 511-523.
Valentine, K.W.G. 1978. The soil landscapes of British Columbia. Ministry of
Environment, Resource Analysis Branch, Victoria, BC. 197 pp.
Walker, L., and R. del Moral. 2008. Lessons from primary succession for restoration of
severely damaged habitats. Applied Vegetation Science, 12(1): 55-67.
Walker, L., and A. Shiels. 2013. Landslide Ecology. Cambridge University Press. New
York, NY. 300 pp.
Walker, L., E. Velazquez, and A. Shiels. 2009. Applying lessons from ecological succession
to the restoration of landslides. Plant and Soil, 324(1-2): 157-168.
Wallis, C.I.B., Y.C. Tiede, E. Beck, K. Bohning-Gaese, R. Brandl et al. 2021. Biodiversity
and ecosystem functions depend on environmental conditions and resources rather than the
geodiversity of a tropical biodiversity hotspot. Scientific Reports 11: 24530.
Whittaker, R. 1960. Vegetation of the Siskiyou Mountains, Oregon and California.
Ecological Monographs, 30:279-338.
Whittaker, R. H. 1965. Dominance and diversity in land plant communities. Science, 147:
250-260.
Zaiontz, C. 2017. “Two Sample Kolmogorov-Smirnov Table.” Real Statistics Using Excel.
Online source: real-statistics.com/non-parametric-tests/goodness-of-fit-tests/two-samplekolmogorov-smirnov-test/

121

Chapter 3. Beta diversity as a measure of biophysical turnover on landslides
3.1 Introduction and Background
Through marked alteration of vegetation and environment, landslides create diversity on a
landscape. Diversity in nature can be described in a series of scales forming a hierarchy. As
outlined in Chapter 2, the three basic levels of diversity are alpha diversity, beta diversity,
and gamma diversity (Whittaker 1960). These three scales of diversity are related, and often
quantification of one level is required before quantifying another. Alpha and gamma
diversity refer to defined geographic units and are classified as inventory diversity (Whittaker
1972). Alpha diversity is within-site diversity, while gamma diversity is described at a
regional level. These two diversities may be measured in similar ways, although they are not
transferable (Tuomisto 2010a). Whittaker coined the concept of a third category of diversity,
namely beta diversity. Beta diversity is generally understood as diversity between sites or
communities. It refers to variation between samples and is classified as differentiation
diversity. It requires a different set of techniques for measurement than inventory diversity.
Essentially, beta diversity is assessed in relation to complementarity or relative similarity or
dissimilarity between communities (Magurran 2004).
Beta diversity represents a key aspect of the spatial pattern of biodiversity. Whittaker (1960)
described beta diversity in its simplest form as the change in species composition over an
area, or the extent of change of community composition in relation to a gradient or pattern of
environments. For this reason, it has also been referred to as spatial turnover. Generally,
beta diversity increases as similarity in species composition decreases. A landscape with a
certain number of species and little overlap in species composition over an area will be more
diverse than an equivalent landscape with equally speciose assemblages but many shared
122

species from site to site. Beta diversity has been found to be more effective than alpha
diversity at detecting change in ecological communities (Seguin et al. 2013), and therefore is
valuable in assessing the effects of disturbances over time as well as space.
3.1.1 Measurement of beta diversity
Unfortunately, many different definitions and concepts of beta diversity have arisen over the
years, accompanied by an equally varying number of methods to describe and quantify it.
This lack of standardisation has been the source of debate and discussion in recent years
(Koleff et al. 2003; Moreno and Rodriguez 2010; Tuomisto 2010a; Tuomisto 2010b;
Tuomisto 2010c; Anderson et al. 2011). Koleff et al. (2003) noted at least 24 measures of
beta diversity for presence/absence data. All the variants of beta diversity measure some sort
of differentiation or heterogeneity, but they each represent very different phenomena
(Tuomisto 2010a). The many different measurements of these variants cannot be compared
numerically. The debate over a lack of a unified concept and measure has led some to call
beta diversity a key concept in ecology, while others deem it of no use (Jurasinski et al.
2009). In this current study, beta diversity refers to the change in species composition over
space, or species turnover, and will focus on turnover as it relates to distance or area.
There are two main categories of beta diversity measures: species richness (coefficients that
examine variation in richness across scales) and species composition (coefficients that
examine variation in species composition between samples) (Jurasinski et al. 2009). Species
richness measures assess the extent of difference between two or more areas of alpha
diversity relative to gamma diversity, where gamma diversity is usually measured as total
species richness. Species composition measures focus on the differences in species

123

composition amongst areas of alpha diversity and were designed to measure complementarity
or similarity/dissimilarity (Barwell et al. 2015; Chao and Chiu 2016). They employ
similarity/dissimilarity coefficients, slope of distance-decay relationships, sum of squares
species matrices, or gradient length in ordination space techniques. Species composition
measures essentially assess the biotic distinctness of assemblages and include the Jaccard and
Bray-Curtis coefficients. A third category of measures use the species-area relationship to
measure turnover related to species accumulation with area.
Beta diversity indices
As with alpha diversity, various indices have been devised for beta diversity. Most indices
use presence/absence data and so focus on the species richness component of diversity. One
of the simplest and most effective measures is Whittaker’s (1960) measure Bw, which is
obtained by dividing total species recorded in the system by average sample diversity
(species richness). Of six beta diversity measures tested by Wilson and Shmida (1984) for
effectiveness in measuring community turnover, Whittaker’s measure was found to fulfill
most of the authors’ criteria of number of community (assemblage) changes, additivity,
independence from alpha diversity, and independence from excessive sampling, with fewest
restrictions.
The more complementary two sites are, the higher their beta diversity. Complementarity
describes the difference between sites in terms of the species they support. The beta diversity
of pairs of sites can be described by using a similarity/dissimilarity coefficient. A matrix can
be constructed using the Jaccard (1908) similarity index, which tallies species gained and
species lost.

124

The Sorensen (1948) measure is another similarity measure that is popular and viewed as
very effective. It uses presence/absence and is identical to the Bray-Curtis presence/absence
coefficient. One disadvantage of this measure is that if samples differ markedly in species
richness, the Sorensen measure will always be large (Lennon et al 2001). However, it can be
modified to accommodate quantitative data (Bray and Curtis 1957).
A significant advantage of the beta diversity measures discussed is their ease of calculation
and application. However, the coefficients do not take into consideration relative abundance
of species. Dominant species have no more weight in a presence/absence beta diversity
measure than a species represented by one individual. This has resulted in the development
of similarity/dissimilarity measures based on quantitative data, such as the Bray-Curtis index
(Bray and Curtis 1957).
3.1.2 Landslide beta diversity
Landslides create harsh environments that can act as ecological filters for plant community
development (Furusawa et al. 2023). Because they are large-scale disturbances caused by
stochastic events, landslides are difficult to study and generalise. The environment that
results from a landslide often consists of a patchwork of rafted materials amid a variety of
exposed and deposited substrates, which can lead to increased patchiness across the
landscape. This greater patchiness presumably translates to increased beta diversity.
Community assembly occurs by both stochastic/random (i.e. ecological drift) and
deterministic (i.e. niche selection) processes. General patterns of community responses to
landslides and environmental influences on community assemblages after a landslide are
poorly understood. The extreme environment following a landslide is expected to favour

125

niche-assembled communities comprised of specialist species, but little is known about this
process.
Disturbances such as landslides can alter beta diversity by changing the relative importance
of community assembly mechanisms that influence clustering of species across landscapes
(Myers et al. 2015). Disturbance can increase clumping either through divergent selection of
niches across environmental gradients or through reduced dispersal if species become rare
after disturbance, or by a combination of the two processes. Disturbed landscapes may thus
have a lower species richness and a lower overall alpha diversity than undisturbed
landscapes, yet greater beta diversity.
In this chapter, I attempt to answer the question: To what extent does turnover or beta
diversity in microsites and plant species occur on landslides? I also investigate whether
vegetation beta diversity is correlated in any way with site beta diversity. For both lines of
investigation, I also assess whether there is a difference between different sizes or ages of
landslides and between landslide and undisturbed terrain study sites.
3.1.3 Study areas
The sampling transects for this study were established on the same three study areas
described in Chapter 2, in the Peace River Region of northeastern BC on landslides < years
old in glaciolacustrine material. The three landslides were Beatton River, Cecil Lake, and
Hasler Flats. The Beatton River landslide most recently activated in 2015 and was
approximately 30 ha. The Cecil Lake landslide occurred in 1998 and was about 56 ha. The
much smaller Hasler Flats landslide occurred in 2013 and was only 1.5 ha in size. All three
landslides initiated on plateaus and travelled downslope into rivers or streams. Detailed
126

information on each landslide and its associated surrounding terrain can be found in Chapter
2 on alpha diversity, in Section 2.2.
3.2 Methods
3.2.1 Data collection
Sampling of vegetation cover and environment variables was carried out on paired landslide
and undisturbed study sites in the Peace River Region of British Columbia on the three study
areas: Beatton River, Cecil Lake, and Hasler Flats. On the Beatton River and Cecil Lake
landslide study sites, seven randomly located transects were sampled, comprised of up to 30
m of a series of 1 x 1 m quadrats (plots). On the Hasler Flats landslide study site, just three
transects were established, due to the much smaller size of the landslide (1.5 ha). In total, 17
landslide transects were sampled. On the associated undisturbed study sites surrounding each
landslide, three transects were sampled for each of the three study areas, for a total of nine
undisturbed transects. Transect commencement points for each transect were randomly
located on a grid in the office prior to beginning field work, with point locations and transect
directions determined using an online random number generator.
To further ensure randomness of data, field sampling was done to the right of the transect
defined by a 30 m tape measure on even numbered transects and to the left of the transect on
odd numbered transects. Individual quadrats were delineated using a 1 x 1 m square plot
made of PVC (polyvinyl chloride) tubing (Figure 3-1). At each plot, plant species and
associated percent covers were recorded, along with ground substrate percentages as per
Land Management Handbook 25 (BC Gov 2010), as well as predominant slope and
predominant aspect. Slope and aspect were measured at the microsite level (i.e. the 1 x 1 m

127

quadrat plot). Aspect was eventually converted to folded aspect values along the north-south
access to facilitate its analysis as a continuous variable (McCune and Keon 2002). Any
vegetation or site type changes along the transect were recorded. Adjacent plots were added
successively along the transect until two plots were sampled consecutively with no new
species or the 30 m transect ended, whichever came first. If plots fell on a transition between
two visibly different vegetation or substrate types, the plot was moved to the next metre
marker where it was completely within the new type. Additionally, if there were no new
species between two consecutive plots but there was a type change further along the 30 m of
the transect, the next plot was established at the next metre marker fully within the new type.
The number of quadrats per landslide transect ranged from 15-30 for Beatton River and 1430 for Cecil Lake, while all transects for Hasler Flats had 30 quadrats. All undisturbed
transects for Beatton River and Hasler Flats had 30 quadrats, while for Cecil Lake the
number of quadrats per undisturbed transect ranged from 28-30. Sampling intensity for the
landslide study sites was 0.23 transects/ha and 5.63 quadrats/ha for Beatton River, 0.12
transects/ha and 3.00 quadrats/ha for Cecil Lake, and 2.00 transects/ha and 60 quadrats/ha for
Hasler Flats. Sampling intensity for the undisturbed study sites worked out to 0.1
transects/ha and 3.00 quadrats/ha for Beatton River, 0.05 transects/ha and 1.57 quadrats/ha
for Cecil Lake, and 2.00 transects/ha and 60 quadrats/ha for Hasler Flats. For all transects,
representative photographs were taken at various locations along the length.

128

Figure 3-1. Sampling transects with a series of 1 x 1 m quadrats along a measuring tape. Image on the left is
Beatton River undisturbed transect BEtu1a. (Photo August 26, 2018) Image on the right is Beatton River
landslide transect BEtl1a. (Photo July 18, 2017, by P. Burton. Used with permission.)

3.2.2 Data analysis
Vegetation data for each transect was entered into Excel spreadsheets, formatted, and then
exported to the multivariate analysis program PC-ORD v. 7.10 (McCune and Mefford 2016;
McCune and Mefford 2018) for analysis. For plots with no vegetation or 100 percent water,
separate columns were created titled “NO_PLANTS” and “WATER”. Two transects on the
Beatton River landslide had no vegetation the entire length of the transect, and thus were
eliminated from the PCORD analysis because it was discovered after field sampling that a
beta diversity value could not be calculated in PCORD when empty rows were present. For

129

each remaining transect on each landslide and in the associated surrounding undisturbed
terrain, an initial value of Whittaker’s measure of beta diversity (Bw) was calculated in PCORD, using the formula Bw = Gamma (overall) diversity/Alpha (local) diversity.
Environment data for each transect was entered into Excel spreadsheets and formatted for
analysis in PC-ORD. Slope and folded aspect were inputted in separate columns. Folded
aspect was calculated in Excel using the following formula:
ABS(180 - ABS(aspect -225°)
where ABS refers to the absolute value, and aspect was measured in degrees azimuth from
true north. Folding aspect around the southwest slope-facing direction is considered
representative of heat intensity on a site in the northern hemisphere (McCune and Keon
2002). Substrate values for organic matter, decaying wood, bedrock, rock (cobbles and
stones), mineral soil, and water were inputted in separate columns. Once the environment
data was formatted, each transect dataset was imported into PC-ORD and Whittaker’s beta
diversity Bw was calculated. Descriptive statistics were then calculated in Excel for the
vegetation and environment beta diversity data for each study site.
To provide background information and a fuller picture of diversity on the sites, alpha
diversity values for vegetation and environment variables were calculated for each transect
by averaging the values of all quadrats/plots along the transect. Richness, evenness, Shannon
index (H’), and Simpson’s index (1-D) were all calculated for each transect in PC-ORD. The
higher the evenness of a sample, the more similar the abundances are to each other in an area,
and the higher the alpha diversity. The Shannon index incorporates richness and evenness to
quantify the uncertainty in predicting the identity of an individual taken at random from a
130

dataset. It is related to the weighted geometric mean of the proportional abundances
(evenness). The higher the Shannon index, the greater the uncertainty. Simpson’s index of
diversity ranges between 0 and 1 and represents the probability that two individuals randomly
selected from a sample will belong to different groupings. The higher the Simpson value, the
greater the sample alpha diversity. Descriptive statistics were calculated in Excel for the
vegetation and environment alpha diversity data for each study site, using the individual
transect data generated in PC-ORD.
A scatter graph was created to display the relationship between the resultant vegetation beta
diversity values and environment beta diversity values for all landslide transects, and for all
undisturbed transects. Regression lines of vegetation beta diversity as a function of
environment beta diversity and associated regression statistics were then calculated,
displayed, and compared for the landslide and undisturbed transects.
3.3 Results
3.3.1 Transect vegetation beta diversity
3.3.1.1 Landslide transect vegetation beta diversity
The results for the landslide transect vegetation beta diversity calculations in PC-ORD
revealed a range of values, both within and among study sites (Table 3-1).

131

Table 3-1. Landslide transect vegetation beta diversity (Bw) values.

Transect Be ta dive rsity (Bw)
Be atton River
Mean Bw
BEtl1a
2.7
BEtl1
1.8
BEtl4
0.8
2.42
BEtl5
3.0
BEtl6
3.8
Cecil Lake
CEtl1
1.9
CEtl2
2.3
CEtl3c
5.5
CEtl4
2.8
3.10
CEtl5
2.3
CEtl6
4.2
CEtl7
2.7
Hasler Flats
HAtl1a
2.9
HAtl2
3.0
3.17
HAtl3
3.6
Ove rall (all transects)

2.89

SD

CV %

1.15

47.69

1.29

41.52

0.38

11.96

1.11

38.39

Cecil Lake transects showed the greatest range of landslide vegetation beta diversity (Bw)
values, while Hasler Flats transects had the smallest range. Cecil Lake also yielded the single
transect with the highest beta diversity value (CEtl3c at 5.5), whereas Beatton River had the
transect with the lowest beta diversity value (BEtl4 at 0.8) of all the landslide transects.
Hasler Flats presented the highest mean landslide transect vegetation Bw (3.17) whereas
Beatton River had the lowest mean Bw (2.42) of the three study sites. Cecil Lake showed the
highest standard deviation (SD), Beatton River had the highest coefficient of variation (CV),
and Hasler Flats exhibited the lowest SD and CV. Both Hasler Flats and Cecil Lake had a
higher mean than the overall mean Bw value for all landslide vegetation transects in the
study.

132

For background information, alpha diversity values for vegetation were calculated for each
landslide transect study site (Table 3-2). Mean richness, evenness, Shannon index (H’), and
Simpson’s index (D) were all calculated in Excel from individual transect values generated in
PC-ORD. Although these values were generally lower than those calculated for the landslide
relevés in Chapter 2, the overall trend was the same for ranking the landslides.
Table 3-2. Landslide transect vegetation alpha diversity statistics.

Diversity Measure
Beta diversity (Bw)
Richness (S)
Eve nness (E)
Shannon (H')
Simpson (D)

Mean
2.42
6.64
0.58
0.96
0.46

Beatton River
SD
CV %
1.15
47.69
2.61
39.32
0.09
15.41
0.37
38.54
0.15
32.54

Mean
3.1
11.4
0.67
1.55
0.66

Cecil Lake
SD
CV %
1.29
41.52
2.99
26.26
0.05
6.73
0.39
25.05
0.15
22.55

Mean
3.17
12.3
0.63
1.47
0.64

Hasler Flats
SD
CV%
0.38
11.96
1.57
12.78
0.11
18.28
0.22
15.05
0.07
11.22

Overall - All landslide transects
Mean
SD
CV %
2.89
1.11
38.39
9.99
3.51
35.08
0.63
0.08
13.06
1.34
0.43
32.41
0.59
0.16
27.07

Of the three landslides, Hasler Flats had the highest vegetation richness (12.3) while Beatton
River had the lowest richness (6.64) but the highest CV (39.32%). Hasler Flats and Cecil
Lake mean richness values were higher than the overall mean. Cecil Lake showed the
highest evenness while Hasler Flats yielded the lowest evenness value. At the same time,
Hasler Flats had the highest CV. The Shannon index was highest on the Cecil Lake landslide
and lowest on the Beatton River landslide. Beatton River showed the highest CV for
Shannon index. Cecil Lake and Hasler Flats both had a mean Shannon index that was higher
than that of the mean overall value for all landslide transects. Cecil Lake also yielded the
highest Simpson index while Beatton River was the lowest. Cecil Lake and Hasler Flats both
had a higher mean Simpson index than the overall mean.
3.3.1.2 Undisturbed transect vegetation beta diversity
Table 3-3 shows the vegetation beta diversity values calculated in PC-ORD for each
undisturbed transect.

133

Table 3-3. Undisturbed transect vegetation beta diversity (Bw) values.

Transect Be ta dive rsity (Bw)
Be atton River
Mean Bw
BEtu1a
3.3
BEtu2b
1.8
2.50
BEtu3a
2.4
Ce cil Lake
CEtu1
2.4
CEtu2
3.2
2.87
CEtu3a
3.0
Hasler Flats
HAtu1a
2.3
HAtu2
1.4
2.03
HAtu3
2.4
Ove rall (all transects)

2.47

SD

CV %

0.75

30.20

0.42

14.52

0.55

27.09

0.63

25.40

Of the three undisturbed study sites, Beatton River yielded the greatest range of transect
vegetation Bw values while Cecil Lake showed the smallest range. Beatton River contained
the undisturbed transect with the overall highest Bw value (BEtu1a at 3.3), while Hasler Flats
had the transect with the lowest Bw value (HAtu2 at 1.4).
For the undisturbed transect vegetation, Hasler Flats yielded the lowest mean Bw, at 2.03.
Cecil Lake presented the highest mean Bw, at 2.87. Cecil Lake also had the lowest SD and
CV. Both Beatton River and Cecil Lake showed a mean Bw that was higher than the overall
mean for all the undisturbed transects. Beatton River also had a higher SD and CV than
these overall values for all undisturbed transects.
For background information, alpha diversity values for vegetation were calculated for each
undisturbed transect study site (Table 3-4). Mean richness, evenness, Shannon index (H’),
and Simpson’s index (D) were all calculated in Excel from the individual transect values
generated in PC-ORD. Although these alpha diversity values were for the most part lower
134

than those calculated for the undisturbed relevés in Chapter 2, the results ranked the
undisturbed study sites in the same order.
Table 3-4. Undisturbed transect vegetation alpha diversity statistics.

Diversity Me asure
Beta diversity (Bw)
Richness (S)
Evenness (E)
Shannon (H')
Simpson (D)

Mean
2.50
10.50
0.73
1.69
0.73

Beatton River
SD
CV %
0.75
30.20
1.15
10.98
0.05
7.43
0.19
11.44
0.06
8.36

Mean
2.87
10.17
0.70
1.56
0.69

Cecil Lake
SD
CV %
0.42
14.52
1.89
18.57
0.04
6.33
0.27
17.60
0.09
12.76

Mean
2.03
16.53
0.74
2.04
0.80

Hasler Flats
Overall - All undisturbed transects
SD
CV %
Mean
SD
CV %
0.55
27.09
2.47
0.63
25.40
1.50
9.06
12.40
3.38
27.25
0.04
5.10
0.72
0.04
6.10
0.15
7.14
1.77
0.28
16.05
0.04
4.94
0.74
0.08
10.25

Hasler Flats showed the highest mean undisturbed transect vegetation richness value (16.53),
while Cecil Lake had the lowest richness value (10.17). Only Hasler Flats exhibited a mean
richness higher than the overall mean richness for all undisturbed study sites. Regarding
evenness, all three undisturbed study sites were similar in mean value, but Beatton River
yielded the highest CV while Hasler Flats had the lowest CV. The Shannon index was
highest at Hasler Flats and lowest at Cecil Lake, but Cecil Lake had the highest SD and CV.
Only Hasler Flats showed a higher Shannon index than the overall mean. Hasler Flats also
had the highest Simpson index, while Cecil Lake yielded the lowest Simpson index.
Additionally, Cecil Lake had the highest SD and CV for Simpson’s index.
3.3.2 Transect environment beta diversity
The results for the environment beta diversity calculations in PC-ORD generally showed a
smaller range of values as well as lower beta diversity values overall, compared to the
vegetation beta diversity.
3.3.2.1 Landslide transect environment beta diversity
Table 3-5 presents the beta diversity values calculated in PC-ORD for landslide transect
environment variables.
135

Table 3-5. Landslide transect environment beta diversity (Bw) values.

Transect Be ta dive rsity (Bw)
Be atton River
Mean Bw
BEtl1a
0.2
BEtl1
0.2
BEtl4
0.0
0.22
BEtl5
0.1
BEtl6
0.6
Cecil Lake
CEtl1
0.2
CEtl2
0.2
CEtl3c
0.9
CEtl4
0.3
0.43
CEtl5
0.2
CEtl6
0.9
CEtl7
0.3
Hasler Flats
HAtl1a
0.6
HAtl2
0.3
0.43
HAtl3
0.4
Ove rall (all transects)

0.36

SD

CV %

0.23

103.65

0.33

75.87

0.15

35.25

0.27

75.56

The Cecil Lake study site showed the greatest range in landslide transect environment Bw
values while Hasler Flats had the smallest range. Beatton River contained the transect with
the overall lowest Bw (BEtl4, at 0.0), while Cecil Lake contained two transects with the
highest Bw (CEtl3c and CEtl6, at 0.9).
Hasler Flats yielded the highest mean landslide transect environment Bw, at 0.43, while
Beatton River had the lowest mean Bw, at 0.22. Hasler Flats showed the lowest SD and CV.
Both Hasler Flats and Cecil Lake had mean Bw values that were higher than the overall mean
environment Bw value for all landslide transects.
For background information, alpha diversity values for environment variables were
calculated for each landslide transect study site (Table 3-6). Mean richness, evenness,
136

Shannon index (H’), and Simpson’s index (D) were all calculated in Excel using the
individual transect values generated in PC-ORD. Alpha diversity values were generally
lower than those calculated for landslide geotypes in Chapter 2, but the results overall ranked
the landslide study sites in the same order.
Table 3-6. Landslide transect environment alpha diversity statistics.

Diversity Measure
Beta diversity (Bw)
Richness (S)
Evenness (E)
Shannon (H')
Simpson (D)

Mean
0.22
4.42
0.67
1.00
0.55

Beatton River
SD
CV %
0.23
103.65
0.30
6.86
0.05
6.79
0.11
11.28
0.05
8.60

Mean
0.43
3.87
0.69
0.92
0.52

Ce cil Lake
SD
CV %
0.33
75.87
0.17
4.40
0.08
11.03
0.12
13.23
0.07
14.05

Mean
0.43
4.20
0.63
0.90
0.48

Hasler Flats
Overall -All landslide transects
SD
CV %
Mean
SD
CV %
0.15
35.25
0.36
0.27
75.56
0.50
11.90
4.12
0.37
9.04
0.04
7.13
0.67
0.06
9.36
0.14
15.13
0.94
0.12
12.79
0.06
11.90
0.52
0.06
12.11

Beatton River showed the highest landslide transect environment richness while Cecil Lake
had the lowest richness. Hasler Flats yielded the highest CV, while Cecil Lake had the
lowest CV. Both Beatton River and Hasler Flats showed a higher mean environment
richness than the overall average for all landslide transects. Of the three study sites, Cecil
Lake had the highest environment evenness, while Hasler Flats exhibited the lowest
evenness. Cecil Lake also showed the highest CV. For landslide environment Shannon
diversity, Beatton River had the highest value, while Hasler Flats yielded the lowest Shannon
index. However, Hasler Flats also had the highest CV. Only Beatton River showed a higher
mean Shannon than the overall mean Shannon value. For Simpson diversity, Beatton River
had the highest value while Hasler Flats showed the lowest value. Beatton River was higher
than the mean overall Simpson value, while Cecil Lake was tied with the overall mean.
3.3.2.2 Undisturbed transect environment beta diversity
Table 3-7 shows the environment beta diversity values for each undisturbed transect.

137

Table 3-7. Undisturbed transect environment beta diversity (Bw) values.

Transect Be ta dive rsity (Bw)
Be atton River
Mean Bw
BEtu1a
0.4
BEtu2b
0.4
0.30
BEtu3a
0.1
Ce cil Lake
CEtu1
0.1
CEtu2
0.1
0.13
CEtu3a
0.2
Hasler Flats
HAtu1a
0.2
HAtu2
0.4
0.37
HAtu3
0.5
Ove rall (all transects)

0.27

SD

CV %

0.17

57.74

0.06

43.30

0.15

41.66

0.16

59.29

Beatton River and Hasler Flats were tied for the highest range of undisturbed transect
environment Bw values, while Cecil Lake had the lowest range of Bw values. Hasler Flats
yielded the highest mean Bw value of the three undisturbed transect environment data sets,
whereas Cecil Lake showed the lowest Bw. Beatton River had the highest CV, while Hasler
Flats had the lowest CV. Of the three study sites, only Hasler Flats exhibited a higher mean
environment Bw than the overall mean value for all undisturbed transects.
For background information, alpha diversity values for environment variables were
calculated for each undisturbed transect study site (Table 3-8). Mean richness, evenness,
Shannon index (H’), and Simpson’s index (D) were all calculated in Excel using diversity
values that were generated in PC-ORD for each transect. These alpha diversity values were
overall lower than those calculated for undisturbed geotypes in Chapter 2, and for some
indices the undisturbed study sites were ranked differently than in Chapter 2.

138

Table 3-8. Undisturbed transect environment alpha diversity statistics.

Diversity Measure
Beta diversity (Bw)
Richness (S)
Eve nne ss (S)
Shannon (H')
Simpson (D)

Mean
0.30
3.60
0.81
1.03
0.60

Beatton River
SD
CV %
0.17
57.74
0.00
0.00
0.07
8.64
0.09
8.74
0.04
6.24

Me an
0.13
3.53
0.65
0.81
0.49

Cecil Lake
SD
CV %
0.06
43.30
0.21
5.89
0.04
5.80
0.02
2.46
0.03
6.26

Mean
0.37
3.47
0.71
0.87
0.53

Hasler Flats
Overall - All undisturbed transe cts
SD
CV %
Mean
SD
CV %
0.15
41.66
0.27
0.16
59.29
0.21
6.00
3.53
0.16
4.47
0.05
6.98
0.72
0.08
11.64
0.10
11.66
0.90
0.12
13.14
0.04
7.31
0.54
0.06
10.47

For the undisturbed transect environment alpha diversity data, Beatton River showed the
highest environment richness whereas Hasler Flats had the lowest richness. Hasler Flats
yielded the highest richness CV, while Beatton River had the lowest CV. The Beatton River
richness value was also higher than the mean overall richness, while richness for Cecil Lake
was tied with the overall mean. Beatton River showed the highest evenness, while Cecil
Lake had the lowest evenness. Only Beatton River had a higher evenness than the overall
average for all undisturbed transects. Beatton River also yielded the highest Shannon and
Simpson diversities, while Cecil Lake had the lowest values. Beatton River was the only
study site with environment Shannon and Simpson diversities higher than the overall
average. Of the three undisturbed study sites, Hasler Flats had the highest CV for Simpson
index.
3.3.3 Regression analysis: assessing the relationship between vegetation and environment
beta diversity
Simple linear regression was used to test if environment Bw significantly predicted
vegetation Bw both on the landslide and in the surrounding undisturbed terrain (Figure 3-2).

139

Figure 3-2. Landslide and undisturbed transect beta diversity (Bw) regression: vegetation vs environment. Blue
symbols represent Cecil Lake transects, green symbols represent Beatton River transects, and orange symbols
represent Hasler Flats transects. The blue dashed line is the regression line for landslide transect beta diversity
(Bw). There was no significant regression for the undisturbed transects.

A regression line with the R2 (coefficient of determination) value was drawn for the landslide
transects, and a separate analysis was conducted for the undisturbed transects, to determine
whether there was any relationship or correlation between vegetation beta diversity and
environmental beta diversity. The fitted regression model was: Vegetation Bw =
1.617631918 + 3.525096525*(environment Bw). The regression for the landslide data was
statistically significant (R2 = 0.75, F (1, 13) = 38.71, p = 0.000031). The standard error of
the slope coefficient was 0.5767 indicating the observed values for the landslide data fell an
140

average of 0.58 units from the regression line. The low standard error suggested the model
was precise enough to be used for predictions. Using the model, environment Bw
significantly predicted vegetation Bw on landslides. The landslide transect vegetation beta
diversity had a strong relationship with the beta diversity of the environmental variables. The
R2 value of 0.7486 indicated that almost 75% of the variation in vegetation Bw was
explained by environment Bw. The undisturbed transects did not yield a significant
relationship between vegetation beta diversity and environment beta diversity. When the
landslide and undisturbed plots were combined in a single analysis (not reported), there was a
significant, though somewhat weaker relationship than for just the landslide plots.
3.4 Discussion
The purpose of this research was to assess turnover of vegetation and site (environment)
variables on landslides and to determine if there was any correlation between the two
components. The research also investigated whether any correlations or relationships were
influenced by size or age of the landslide disturbance, and whether there was a difference
between landslide and undisturbed terrain areas.
In this study, vegetation and environment beta diversity were generally higher on landslides
than on the surrounding undisturbed terrain. In addition, there were notable differences in
diversity between the three study areas of Beatton River, Cecil Lake, and Hasler Flats.
Average transect vegetation and environment beta diversity were both higher on the landslide
than on the surrounding undisturbed terrain for the Cecil Lake and Hasler Flats study areas,
but for Beatton River, vegetation and environment beta diversity were higher on the
undisturbed terrain. The lower beta diversity on the Beatton River landslide was likely
because the slide was newer and more active than the other two study sites, resulting in large
141

areas of continuous bare ground. The Beatton River landslide also had the largest change in
elevation from the top of the landslide to the bottom. In addition, this landslide was steeper
and had more extreme exposures than the other two landslides, hindering the establishment
of a variety of species across the disturbed area. The size of the three different study areas
also likely affected the beta diversity values, as area influences both environmental gradients
and the ecological mechanisms (such as species sorting) that drive spatial variation in species
composition within a region unit (Heino et al. 2014).
The occurrences of lower alpha diversity but higher beta diversity on the landslide compared
to the surrounding terrain for some of the study areas may be because while mature forest
species were lost, the patchiness of the landslide habitats enabled more novel assemblages of
species or life history strategies (Boinot and Alignier 2023). The structural complexity and
heterogeneity of pre-existing and remaining habitats contributes to local and regional biotic
successional processes and ultimately influences beta diversity (Abbasi et al. 2023). This
impact is mainly seen through changes in both taxonomic and functional beta diversity. At
the local level, biotic processes are determined by species-energy relationships and
availability of resources.
Overall, vegetation beta diversity had a strong positive relationship with site or environment
beta diversity on landslides in the study. As environment beta diversity increased, so did
vegetation beta diversity. On the surrounding undisturbed terrain, however, there was no
such relationship, either positive or negative.

142

3.4.1 Transect vegetation beta diversity
3.4.1.1 Vegetation beta diversity on the landslide
There were notable differences in vegetation beta diversity between the three landslide study
sites of Beatton River, Cecil Lake, and Hasler Flats. Average transect vegetation beta
diversity (Bw) was highest on the Hasler Flats landslide, but Cecil Lake had the greatest
range or variability. The higher variation of the landslide vegetation beta diversity for the
Cecil Lake transects was likely directly due to the larger size of the landslide, its age, and the
adjacent landscape. The passage of time allowed more types of vegetation to encroach from
the surrounding landscape and to propagate from rafts and seed banks. Overall, the
surrounding landscape makeup may have strongly influenced the recovery rate of species
diversity (Furusawa et al. 2023).
The higher among-site variability in plant species composition (beta diversity) on the Cecil
Lake landslide suggests both a random, stochastic process of ecological drift and niche
selection were critical for community assembly following the landslide disturbance. Field
observations indicated the Cecil Lake landslide had a greater variety of site types in close
proximity compared to Beatton River and Hasler Flats, allowing for the potential for higher
turnover of plant species along any given transect distance. For example, the transect CEtl3c
had the highest landslide vegetation beta diversity value (Bw = 5.5) and intersected a pond
for ten metres of the transect length (Figure 3-3). This local site diversity likely contributed
to the higher species turnover, as the pond and surrounding shorelines presented different
habitats where new plants and unique communities could establish. Conversely, the Cecil
Lake landslide transect with the lowest beta diversity, CEtl1 (for which Bw =1.9), had a

143

uniform site type consisting of dense shrubs throughout the transect (Figure 3-4). This
uniformity is expressed in the lower turnover denoted by a lower Bw value.

Figure 3-3. Cecil Lake landslide transect CEtl3c, showing high diversity of adjacent habitats as illustrated by
Alnus spp. next to a pond, with sparsely vegetated south-facing slopes on the other side of the pond. (Photo
August 25, 2018)

Figure 3-4. Cecil Lake landslide transect CEtl1, showing lower diversity of adjacent habitats as illustrated by
expanse of level ground and relatively homogeneous vegetation. (Photo August 25, 2017)

144

The higher average beta diversity of the Hasler Flats landslide reflects the abundance of very
different site types situated very close to each other, coupled with the much smaller area of
the Hasler Flats landslide (Heino et al. 2014), compared to the Cecil Lake landslide. The
Hasler Flats landslide consisted of a series of interlocking young aspen cutblock rafts from a
15 year old clearcut stand, exposed horsts (ridges of sand or clay) and grabens, and pond site
types. Following the landslide disturbance, vegetation dispersal distances would have been
shorter both from the surrounding landscape and within the landslide area. In addition, small
samples in areas with high gamma diversity tend to have inflated beta diversity values, due to
the dependence on sample size that interacts with gamma diversity (Cao et al. 2021). Despite
its high site type diversity, Hasler Flats likely showed the smallest range of transect
vegetation beta diversity values because of its smaller sample size. Hasler Flats had less than
half the number of landslide transects of Cecil Lake, so this could have reduced the range just
because there were fewer samples to begin with.
Although Hasler Flats had the highest mean vegetation beta diversity, it had the lowest CV.
This lower CV indicates much less variation around the mean compared to Beatton River and
Cecil Lake. This finding was in accord with field observations, as the Hasler Flats landslide
overall appeared more consistent in diversity than Beatton River or Cecil Lake.
The lower average vegetation beta diversity of the Beatton River landslide compared to the
other two landslide study sites was likely due to the large proportion of the slide that was
sparsely vegetated because of recent movement and steep, unstable slopes. The Beatton
River landslide also contained the transect with the lowest overall vegetation beta diversity,
BEtl4 (Figure 3-5), for which Bw = 0.8. The transect BEtl4 was on a sparsely vegetated

145

rubbly colluvial side slope and most of the vegetation was the invader exotic pioneers
Melilotus officinalis and Melilotus albus (sweetclovers).

Figure 3-5. Sweetclover (Melilotus spp.) carpeting the toe of the Beatton River landslide. Transect is BEtl4.
(Photo August 16, 2017)

Research has shown that sites with lower beta diversity may have higher alpha diversity, and
the opposite can be true as well (Boinot and Alignier 2023). The background alpha diversity
results of the present study showed variable outcomes. The Hasler Flats landslide had the
highest richness (Figure 3-6), likely due to the many species rafted or otherwise dispersed
from the adjacent aspen cutblock and the surrounding aspen (Populus tremuloides) and
cottonwood (P. balsamifera ssp. balsamifera) stands with shrubby understories. The greater
evenness, Shannon, and Simpson values at Cecil Lake may have resulted from the higher
diversity of site types and the greater age and size of the landslide. Of the three landslides,
146

Beatton River exhibited the lowest vegetation richness, evenness, Shannon, and Simpson
measures. The lower alpha diversity values for Beatton River were likely due to the younger
age of the landslide, as well as the fact that it was still quite active, so even when plants did
establish, they could quickly die off or slide downslope and be buried or otherwise damaged.
Depending on the severity of the disturbance, landslides can cause a shift from stochastic or
random to more deterministic or predictable plant community assembly, with distinct
responses in terms of beta diversity (Seguin et al. 2013). More frequent or severe landslides
can lead to a homogenised species distribution or a high proportion of invasive species, as
found on the Beatton River transect BEtl1 (Figure 3-7), while a lower disturbance frequency
or severity can cause greater stochasticity.

Figure 3-6. Vegetation diversity on the Hasler Flats landslide, illustrated by dense layers of varied forbs, short
shrubs, and tall shrubs. Transect is HAtl1a. (Photo July 28, 2018)

147

Figure 3-7. Invasive/exotic plants Lactuca serriola (prickly lettuce), Sonchus spp., Melilotus spp., and Elymus
repens (quackgrass) on the Beatton River landslide. Transect shown is BEtl1. (Photo August 17, 2017)

3.4.1.2 Vegetation beta diversity in the surrounding undisturbed terrain
The variable levels of beta diversity in the undisturbed areas surrounding each landslide
highlights the complexities of assessing beta diversity. The greater range of transect
vegetation beta diversity values and the higher variability in the adjacent undisturbed terrain
of the Beatton River site compared to Cecil Lake and Hasler Flats may have been due to the
higher diversity of ecological habitats surrounding the Beatton River landslide. Habitats
included mature conifer forest, mature aspen forest, mature cottonwood forest (Figure 3-8),
shrublands, and grasslands.

148

Figure 3-8. Example of moist cottonwood (Populus balsamifera ssp. balsamifera) habitat in undisturbed terrain
at Beatton River. Transect is BEtu1a. (Photo August 26, 2018)

Cecil Lake exhibited the lowest range of vegetation beta diversity values of the three
undisturbed study sites but yielded the highest average undisturbed vegetation transect beta
diversity. The Cecil Lake undisturbed study site also had the lowest CV, indicating less
variation around the mean compared to Beatton River and Hasler Flats. The comparatively
149

higher undisturbed transect vegetation beta diversity for Cecil Lake could have been due to
the larger area, increasing the likelihood of transects falling on more heterogeneous terrain.
Compared to the associated landslides, the mean vegetation beta diversity was lower in the
surrounding undisturbed terrain for Cecil Lake and Hasler Flats, but higher in the undisturbed
terrain than on the landslide at the Beatton River study area. Although there may be no
significant differences in species diversity between an undisturbed and landslide community,
much higher among-site variability may be found on landslides (Furusawa et al. 2023). The
generally lower vegetation beta diversity on the undisturbed terrain for the three study areas
was likely due to the presence of greater areas of homogeneous habitats, such as the adjacent
cutblock next to Hasler Flats (Figure 3-9), as well as the passage of time that allowed certain
successional species to outcompete pioneers and take over.

Figure 3-9. Relatively homogeneous vegetation (young aspen – Populus tremuloides) in undisturbed terrain at
Hasler Flats. Transect is HAtu1a. (Photo August 10, 2018)

150

Overall, the assessment of vegetation alpha diversity in the undisturbed terrain among the
three study areas revealed variability between study sites, with higher alpha diversity values
for Beatton River and Hasler Flats and lower values for Cecil Lake in the undisturbed terrain
compared to the landslide areas. The higher alpha diversity richness value of Hasler Flats
compared to Cecil Lake may have been due to the greater array of shrub species in the terrain
surrounding the Hasler Flats landslide. Hasler Flats also yielded the highest Shannon and
Simpson values, further suggesting the surrounding undisturbed terrain vegetation had higher
alpha diversity than the Beatton River and Cecil Lake study sites. Cecil Lake had the lowest
Shannon and Simpson values, indicating the vegetation on the undisturbed terrain was less
diverse. The similar evenness for all three undisturbed sites suggests the sites were similar in
the proportional distribution of various species. It should be noted that “undisturbed” was a
relative term, and for all three landslides there were signs of past disturbance in the
surrounding terrain. There was evidence of older landslides, historical logging, cultivation of
fields, and possibly other natural disturbances. All these factors undoubtedly contributed to a
diversity of plant community growth forms and successional stages.
3.4.2 Transect environment beta diversity
3.4.2.1 Environment beta diversity on the landslide
Overall, the environment beta diversity values on the landslides were much lower than
vegetation beta diversity and had smaller ranges, suggesting a lower turnover of site diversity
compared to vegetation. Disturbance such as a landslide can either increase or decrease
environmental heterogeneity, and likewise either increase or decrease beta diversity (Maab et
al. 2014). The environment of an area can locally filter dispersing plant species, which may

151

establish either through niche partitioning or dispersal (Maab et al. 2014). The variable
environment beta diversity values among landslides in this study indicate different drivers
influencing diversity for each landslide. Just as for vegetation beta diversity, the Cecil Lake
landslide study site had the greatest range of environment beta diversity values, while Hasler
Flats showed the smallest range. The higher range of beta diversity values at Cecil Lake
could have been due to the greater diversity of site types observed on the landslide compared
to Hasler Flats. The findings could also simply be explained by the fact that Cecil Lake
landslide was almost 40 times larger and had more transects than Hasler Flats, increasing the
potential for a random transect to fall on a diverse site. However, Hasler Flats had a much
higher sampling intensity given its size, so the influence may not have been that great. Either
way, the higher range indicates an increased variability in turnover of environment features
on the Cecil Lake landslide. The Cecil Lake and Hasler Flats landslide study sites were tied
for the highest environment beta diversity of the three landslide study sites, even though
Cecil Lake landslide was more variable. Hasler Flats was a smaller and younger slide, and it
appears these factors contributed to a higher beta diversity. The landslide was still in the
process of weathering and becoming more subdued. The lower average environment beta
diversity of Beatton River is consistent with field observations, as large parts of the landslide
were uniform in slope, aspect, and substrate. However, Beatton River also had the highest
CV (over 100%), indicating a large amount of variation among transects.
The variability in environment alpha diversity results for the three landslide study sites
highlights the influence of space and time on environment heterogeneity (Maab et al. 2014).
The higher landslide environment alpha diversity results for Beatton River contrasted with
the lower beta diversity results. Higher alpha diversity could be explained by the site

152

diversity present along some of the Beatton River transects. One of the transects, BEtl6,
passed through rubbly boulders, across a pond, and then up a steep slope (Figure 3-10).
Other transects on the landslide had patches with no vegetation. The lower environment
richness of Cecil Lake compared to Beatton River and Hasler Flats could be due to the age of
the landslide, as geomorphological features weather over time, becoming more uniform. The
lower alpha diversity values for Hasler Flats may have been a result of the small size of the
landslide and the more consistent configuration of the environment features.

Figure 3-10. Beatton River landslide transect BEtl6, showing microsite diversity of hummocks, depressions,
ponds, and fissures. (Photo taken September 5, 2018)

153

3.4.2.2 Environment beta diversity in the surrounding undisturbed terrain
Overall, mean environment beta diversity values for the undisturbed transects were lower
than for the landslide transects, but on an individual basis the Cecil Lake and Hasler Flats
undisturbed study sites had lower average beta diversity while Beatton River had higher beta
diversity compared to the associated landslides. Lower environment beta diversity in the
undisturbed terrain could be due to the surfaces exhibiting a lower diversity of microsites
over short distances compared to the landslides, as a result of minimal disturbance over time.
The higher average beta diversity value for Hasler Flats and the lowest average beta diversity
for Cecil Lake undisturbed samples could be explained by the fact that Hasler Flats samples
were in more broken terrain (Figure 3-11) in the surrounding area, while Cecil Lake
undisturbed samples fell in more uniform terrain (Figure 3-12). The similarity in beta
diversity range for Beatton River and Hasler Flats is likely due to the similar undisturbed
terrain of these two study sites.

154

Figure 3-11. Hasler Flats undisturbed transect HAtu3 showing broken terrain. Transect traverses uneven
terrain and small fissures for the first 15 m (foreground), then ascends a steep slope for 8 m (visible in the
distance), and finally levels out (evident in the photo where the sky shows through). (Photo August 3, 2018)

155

Figure 3-12. Cecil Lake undisturbed transect CEtu2 showing uniform, level terrain. (Photo August 17, 2018)

The variable environment alpha diversity of transects on undisturbed sites was reflective of
the variation among landslide study areas. The higher environment alpha diversity values of
richness, evenness, Shannon, and Simpson values for transects in the Beatton River
undisturbed terrain could be explained by the fact that Beatton River had a variety of site
types and ecosystems. Conversely, Hasler Flats exhibited the lowest richness, while Cecil
Lake had the lowest evenness, Shannon, and Simpson values. The lower environment

156

richness of the undisturbed Hasler Flats transects reflects more uniform terrain and may also
be due to the smaller area over which transect locations were selected. The lower alpha
diversity values for Cecil Lake may have been the result of a more subdued landscape
overall, given the larger scale of the site as well as its advanced age.
3.4.3 Relationship between transect vegetation and environment beta diversity
Environmental factors shape both local and regional biotic processes, mainly through
changes in functional and taxonomic beta diversity patterns (Abbasi et al. 2023) but also
alpha diversity patterns (Boinot and Alignier 2023). The current study in the Peace River
Region revealed a significant relationship between vegetation beta diversity and environment
beta diversity on the landslide study sites, while there was no such relationship on the
undisturbed sites. Generally, as environment beta diversity increased on the landslides, so
did vegetation beta diversity, even though the environment beta diversity values were notably
lower than those for vegetation beta diversity. The stronger association between vegetation
and environment beta diversity on the Cecil Lake landslide study site compared to Beatton
River and Hasler Flats could be explained by the greater age of this landslide, where plant
communities have evolved with the changing microsites over time. It is also possible the
plant communities altered the environment as time passed. The lack of a strong relationship
between environment and vegetation beta diversity on the undisturbed areas is most likely
due to the overall lower beta diversity for both environment and vegetation on these sites.
The sites have stabilised, and thus more resilient plant species have tended to dominate.
Research has shown beta diversity is positively associated with local topographical
heterogeneity as well as with community-level niche specialisation and niche marginality
(Cao et al. 2021). Higher niche marginality, or ability to occupy the peripheries, indicates
157

larger niche space, which may allow more species to use more variable resources, while
higher niche specialisation enables species to specialise on narrower subsets of the resources
present. Therefore, on a highly variable topography where niche specialisation or
marginalisation are at work, a greater turnover of species can be found from one point to
another across a given distance. Local processes of topographical heterogeneity and the
resultant niche differentiation drive beta diversity at scales of 15-52 ha. Greater
environmental variability on a site may ultimately lead to reduced evenness in communities
(Furusawa et al. 2023).
Connectivity between habitats is important for driving the recovery of disturbed sites such as
landslides. Connectivity is a product of the synergies between natural history, dispersal
mechanisms, and the quantity, quality, and pattern of habitat patches, at the landscape level
(Chiantore et al. 2018). Connectedness of habitats and the successional state of ecological
communities together play an important role in understanding consequences of change in
different ecosystems. Of the three landslides studied, Hasler Flats appeared to have the
greatest connectivity, mainly due to its small size and the presence of many rafts. From the
perspective of beta diversity as an indicator of ecological connectivity, recovery does not
necessarily depend on the available species at the regional level. Instead, local ecological
attributes affiliated with beta diversity and the creation of habitats by living organisms drive
community assembly by way of species replacement (turnover) and habitat filtering in
disturbed areas. Habitats and communities can become homogenised with an increased
disturbance frequency or severity, exhibiting an accelerated homogenisation with increasing
scale of disturbance. This condition appeared to be present on portions of the Beatton River
landslide, where disturbance was still active and exotics dominated large portions of the

158

landslide. Boinot and Alignier (2023) found alpha and beta diversity can be maintained with
a variety of weed species and life strategies in highly disturbed sites, even if rare and more
sensitive species are removed from the community. Homogeneous sites also recover faster,
which results in diminished complexity and biodiversity over time (Chiantore et al. 2018).
3.4.4 Limitations
One of the limitations to the study was the fact that the number of samples and sampling
intensity per hectare at each study site were not the same. The only study area with the same
amount of transects on the landslide and in the surrounding undisturbed terrain was Hasler
Flats. Another limitation was that not all transects were of equal length or had the same
number of plots in a sample. For example, two transects could have both been 30 m long, but
one might have only 20 plots because a portion of the transect was skipped once there were
no new species. Plots were sampled again along the transect if there was a new type, but it is
possible there were microsites in between that contained new species. Therefore, beta
diversity could have been underestimated. Further, the landslides varied in age, type, and
size, rendering the results more challenging to generalise. Overall, however, the methods
used did provide a useful synopsis of beta diversity on landslides and in the surrounding
terrain, and the study provided many insights into turnover on diverse disturbed sites.
3.5 Conclusions and Recommendations
Although there was some variability between transects on some landslides, overall, average
vegetation and environment beta diversity (Bw) appears to be higher on landslides than on
surrounding undisturbed terrain for sites in northeastern British Columbia. The trend is more
pronounced for vegetation beta diversity. There is a strong relationship between vegetation

159

beta diversity and environment beta diversity on the landslides, even though the environment
beta diversity values are much smaller. This finding suggests that environment beta diversity
does not need to be very high to produce increased vegetation beta diversity.
The differences in vegetation and environment beta diversity values between landslides of
differing sizes and ages indicates that spatial and temporal factors may influence the level of
turnover on a given landslide. The smallest landslide (Hasler Flats) had the highest beta
diversity, and it was the second oldest of three landslides. The newest and most active
landslide (Beatton River) had the lowest beta diversity.
The findings of this study of beta diversity on landslides can help inform restoration planning
for other landslides in the Peace River Region, as well as for low-gradient, deep-seated
landslides throughout the circumboreal zone which occur on glaciolacustrine unconsolidated
material. Beta diversity is a primary signal of the health of a community and the proper
functioning of ecosystems. It can provide important information about ecosystem
mechanisms for the purposes of restoration. However, patterns of beta diversity on their own
should not be used as a benchmark for restoration, since observed change in beta diversity in
disturbed areas such as landslides is chiefly due to random sampling effects influenced by
changes in local community size (Myers et al. 2015). Sites with very similar habitat might
have quite different makeup of species because of the order that species arrive on the
landslide. This phenomenon can produce multiple stable equilibria.
Knowledge about the connections between ecosystem disturbance and taxonomic, functional,
and beta diversity of landslides may provide crucial information for natural resources
management (De et al. 2023) as well as restoration. Beta diversity presents important

160

potential for management of environmental monitoring and conservation since diverse
processes can result in the same beta diversity pattern (Maab et al. 2014). The present study
of landslides has shown that when only a few plant species prevail because of accelerated
habitat disruption, the species composition becomes more uniform and the diversity of the
community decreases, as evident on the Beatton River landslide. High alpha diversity may
obscure deleterious effects of habitat disturbances on plant species diversity (Dehling and
Dehling 2023). Just as for beta diversity, alpha diversity alone should not be viewed as a
representative indicator of disturbance impacts since it does not consider possible local or
regional changes in species composition or turnover. Habitat alterations resulting from
disturbances such as landslides can eliminate vital functional roles, reducing an ecologically
complex system to one that is simpler. In this situation, unique native species disappear, and
more common and often invasive species take over, creating homogenised communities that
diminish regional diversity. Alpha and beta diversity can be maintained with a variety of
weed species and life history strategies in simple, severely disturbed areas, even when
uncommon and more disturbance-sensitive species are eliminated (Boinot and Alignier
2023). However, this type of diversity is not a good indicator for conservation or healthy
functioning of ecosystems, because the original natural community system has been
modified. As there are different drivers for alpha and beta diversity, both types of diversity
can be advanced by implementing different arrays of management practices (Boinot and
Alignier 2023). Any attempts at restoration of landslides should retain patchiness and
maintain beta diversity. As climates change in an unpredictable way in coming years, beta
diversity may provide important wildlife habitat at various scales and enhance resilience of
populations in uncertain times.

161

3.6 References
Abbasi, U.A., E. Mattsson, S.P. Nissanka, and A. Ali. 2023. Species α-diversity promotes
but β-diversity restricts aboveground biomass in tropical forests, depending on stand
structure and environmental factors. Journal of Forest Research, 34: 889-901.
Anderson, M. J., T.O. Crist, J.M. Chase, M. Vellend, B.D. Inouye, A.L. Freestone, N.J.
Sanders, H.V. Cornell, L.S. Comita, K.F. Davies, S.P. Harrison, N.J.B. Kraft, J.C. Stegen,
and N.G. Swenson. 2011. Navigating the multiple meanings of β diversity: a roadmap for the
practicing ecologist. Ecology Letters 14, 19–28. doi:10.1111/j.1461-0248.2010.01552.x
Barwell, L.J., N.J.B. Isaac, and W.E. Kunin. 2015. Measuring β-diversity with species
abundance data. Journal of Animal Ecology, 84: 1112-1122.
BC Gov (British Columbia Ministry of Forests and Range [BCMFR] and British Columbia
Ministry of Environment [BCMOE]). 2010. Field manual for describing terrestrial
ecosystems. 2nd edn. BCMFR Research Branch and BCMOE Resource Inventory Branch,
Victoria, B.C. (Reprint with updates 2015.)
Boinot, S., and A. Alignier. 2023. Discrepancies between the drivers of alpha and beta plant
diversity in arable field margins. Proceedings Royal Society B, 290: 20222179 (11 pages).
Bray, J.R., and J.T. Curtis. 1957. An ordination of the upland forest communities in
Southern Wisconsin. Ecological Monographs, 27:325-349.
Cao, K. et al. 2021. Species packing and the latitudinal gradient in beta diversity.
Proceedings Royal Society B, 288. 20203045 (8 pages).
Chao, A., and C.-H. Chiu. 2016. Bridging the variance and diversity decomposition
approaches to beta diversity via similarity and differentiation measures. Methods in Ecology
and Evolution, 7: 919-928.
Chiantore, M., S.F. Thrush, V. Asnaghi, and J.E. Hewitt. 2018. The multiple roles of βdiversity help untangle community assembly processes affecting recovery of temperate rocky
shores. Royal Society Open Science, 5: 171700 (9 pages).
De, K., A.P. Singh, A.S. Sarkar, K. Singh, M. Siliwal, V.P. Uniyal, and S.A. Hussain. 2023.
Relationship between species richness, taxonomic distinctness, functional diversity, and local
contribution to β diversity and effects of habitat disturbance in the riparian spider community
of the Ganga River, India. Ecological Processes, 12: 13 (13 pages).
Dehling, D.M., and J.M. Dehling. 2023. Elevated alpha diversity in disturbed sites obscures
regional decline and homogenization of amphibian taxonomic, functional, and phylogenetic
diversity. Scientific Reports, 13: 1710 (9 pages).

162

Furusawa, J., K. Makoto, and S. Utsumi. 2023. A large-scale field experiment of artificially
caused landslides with replications revealed the response of the ground-dwelling beetle
community to landslides. Ecology and Evolution, 13: e9939.
https://doi.org/10.1002/ece3.9939
Heino, J., A.S. Melo, and L.M. Bini. 2015. Reconceptualising the beta diversityenvironmental heterogeneity relationship in running water systems. Freshwater Biology, 60:
223-235.
Jaccard, P. 1908. Nouvelle recherches sur la distribution florale. Bulletin de la Societe
Vaudoise des Sciences Naturelles, 44: 223-270.
Jurasinski, G., V. Retzer, and C. Beierkuhnlein. 2009. Inventory, differentiation, and
proportional diversity: a consistent terminology for quantifying species diversity. Oecologia,
159:15- 26 DOI 10.1007/s00442-008-1190-z
Koleff, P., K. Gaston, and J. Lennon. 2003. Measuring beta diversity for presence–absence
data. Journal of Animal Ecology, 72: 367-382.
Lennon, J., P. Koleff, J.J.D. Greenwood, and K.J. Gaston. The geographical structure of
British bird distributions: diversity, spatial turnover, and scale. Journal of Animal Ecology,
70: 966-979.
Maab, S., M. Migliorini, M. Rillig, and T. Caruso. 2014. Disturbance, neutral theory, and
patterns of beta diversity in soil communities. Ecology and Evolution, 4(24): 4766-4774.
Magurran, A. 2004. Measuring biological diversity. Blackwell Publishing. Oxford, UK.
256 p.
McCune, B., and D. Keon. 2002. Equations for potential annual direct incident radiation and
heat load. Journal of Vegetation Science, 13: 603-606.
McCune, B., and M. J. Mefford. 2016. PC-ORD. Multivariate Analysis of Ecological Data.
Version 7. MjM Software Design, Gleneden Beach, Oregon, U.S.A.
McCune, B., and M. J. Mefford. 2018. PC-ORD. Multivariate Analysis of Ecological Data.
Version 7.10. MjM Software Design, Gleneden Beach, Oregon, U.S.A.
Moreno, C., and P. Rodriguez. 2010. A consistent terminology for quantifying species
diversity? Oecologia, 163:279-282.
Myers, J.A., J.M. Chase, R.M. Crandall, and I. Jimenez. 2015. Disturbance alters betadiversity but not the relative importance of community assembly mechanisms. Journal of
Ecology, 103: 1291-1299.

163

Seguin, A., D. Gravel, and P. Archambault. 2014. Effect of disturbance regime on alpha and
beta diversity of rock pools. Diversity, 6: 1-17.
Sørensen, T. 1948. A method of establishing groups of equal amplitude in plant sociology
based on similarity of species content. Biologiske Skrifter, 5:1-34.
Tuomisto, H. 2010a. A diversity of beta diversities: straightening up a concept gone awry.
Part1. Defining beta diversity as a function of alpha and gamma diversity. Ecography, 33:222.
Tuomisto, H. 2010b. A diversity of beta diversities: straightening up a concept gone awry.
Part 2. Quantifying beta diversity and related phenomena. Ecography, 33: 23-45.
Tuomisto, H. 2010c. A consistent terminology for quantifying species diversity? Yes, it does
exist. Oecologia, 164:853-860.
Whittaker, R. 1960. Vegetation of the Siskiyou Mountains, Oregon and California.
Ecological Monographs, 30: 279-338.
Whittaker, R. 1972. Evolution and measurement of species diversity. Taxon, 21(2/3): 213251.
Wilson, M.V., and A. Shmida. 1984. Measuring beta diversity with presence-absence data.
Journal of Ecology, 72(3) (Nov. 1984): 1055-1064.

164

Chapter 4. Landslide-generated ponds in the Fort St. John area, British Columbia:
Characteristics, distribution, and ecological implications
4.1. Introduction
Traditionally, landslides have been viewed as destructive to the habitats and ecology of a
landscape, in that they remove vegetation, bury other vegetation, disintegrate soil profiles, fill
in water bodies, and alter hydrology. However, the same processes that are seen as
destructive can serve to enhance ecological diversity by creating new geomorphic and
hydrological features. In addition to depositing sediment nutrients and habitat debris into
river systems and changing stream morphology, landslides often produce a diverse array of
topographies on the landslide surface, with many pond depressions that contain water
ephemerally, seasonally, or year-round. The size, number, and distribution of such water
bodies is variable and may depend on the material of the landslide as well as the landslide
type. The focus of the research reported here is the ecological contribution of these landslide
ponds to diversity and implications for management.
Water bodies on landslides form important habitats that contribute to biodiversity in many
different environments (Geertsema and Pojar 2007; Shapley et al. 2019). At the landscape
level, these aquatic habitats can create a network of reproductive and dispersal routes for
pond-dependent fauna such as invertebrates, amphibians, and reptiles; this network of various
sizes of ponds provides more value than one large lake (Pop and Chitu 2013). While
landslides themselves create ponds, they also can provide ideal habitat for beavers (Castor
canadensis) (Krueger and Johnson 2016), and beavers can then further modify the
geomorphology and hydrology of the landslide (Butler and Malanson, 1995; Westbrook et al.
2017) and increase the ecological diversity of ponds (Nummi et al. 2019).

165

Although few detailed studies of landslide ponds have been reported, existing research has
revealed interesting characteristics. Sasaki and Sugai (2015) found that landslide ponds in the
mountains of Japan chiefly occur on bigger landslides, in large or deep depressions along
scarps, or in small depressions at pressure ridges, and are mainly fed by groundwater. The
authors of the study also found that pond size is constrained by the topography of the
landslide, and ponds at different stages of development can be present concurrently due to
new activity on an existing landslide. Ponds can persist and expand over time (Cruden et al.
1997). Surface water infiltration through boulder debris and fractures on an existing slide can
saturate the substrate and facilitate further slide activity (Coe et al. 2003), which then forms
new depressions that develop into ponds. On larger slides, different temperature and water
regimes may exist among head, body, and toe positions of the landslide due to differences in
elevation, which can influence infiltration and pond persistence (Coe et al. 2003). Coe et al.
(2009) found that pond location and persistence may be controlled by basal topography, in
that the landslide ponds may persist in one location for over 100 years even while the
landslide material moves around, below, and through the pond.
The studies discussed above have helped to explain pond distribution and development on
landslides in other parts of the world. The key purposes of the research reported here are to
compile an inventory of landslide ponds in a designated study area of the Peace River Region
of northeastern British Columbia, Canada using GIS applications, to describe characteristics
of size and distribution of these ponds within landslides and between landslide types, to
identify ecological importance, and to provide recommendations for management.
The study of landslides in the Charlie Lake 1:250,000 mapsheet area and in the Peace River
Region in general is of interest to various government and industrial entities because of the
166

potential and existing impacts of mass movement events on industrial development and
public safety. Some of the earliest geomorphology studies were carried out in anticipation of
a proposed hydroelectric dam, Site C (now under construction), near the town of Taylor
(Mathews 1978; Catto 1991). These investigations focused on the geological setting of
historic landslides and attempted to determine slide mechanics. Since then, major work has
been done to describe the geological history and stratigraphic makeup of the mapsheet area
(Mathews 1978, 1980; Bobrowsky et al. 1990; Bobrowsky and Rutter 1992; Catto et al.
1996; Hartman 2005). In recent years, more detailed assessments of movement mechanisms
have been performed on specific landslides, such as the Cache Creek slide (Van Esch 2012),
the Attachie slide (Fletcher et al. 2002; Van Esch 2012), and the Beatton River slide
(Dandurand 2018).
An inventory of all landslides in the Charlie Lake mapsheet area was compiled by Severin
(2004), with each mapped landslide categorised according to factors such as movement type
and activity level. Severin (2004) found that the majority of identified landslides occurred in
the Peace River and Beatton River Valleys within the pre-glacial valley limits, where valley
fill and rebound features in the shale bedrock are more prominent. In the current research,
the aim is to investigate landslides in the Peace River Region with a focus on landslide ponds
and their potential ecological value, using the same study area as Severin (2004).
The objectives of the landslide pond research were to: 1) Obtain a snapshot in time of pond
locations on landslides in the study area; 2) Enumerate and graphically illustrate pond area
and distribution in relation to landslide type and geomorphic location; and 3) Investigate
potential ecological and management implications of the findings.

167

What follows is a synthesis of the research and findings of the landslide pond study. First,
the study area is described, followed by a detailed description of the methods that were
employed to address the research question regarding the nature of landslide pond presence,
abundance, distribution, and possible ecological roles in a specified area of the Peace River
Region. The results are presented, including some representative pond photographs from a
few pond-bearing landslides in the area. Presentation of the results is followed by a
discussion of key findings regarding ponds on landslides and possible ecological
implications. The discussion includes a section on the influence of beavers, which often play
a role of keystone species at the landscape level. Finally, the chapter concludes with some
recommendations for further research and land management considerations. The work
reported here is the first known large-scale detailed regional inventory and description of
ponds on landslides.
4.2. Study area
The study area covers the entire NTS (National Topographic System) mapsheet 94A (Charlie
Lake mapsheet, 1:250,000), an area of approximately 16,000 km2 (1,600,000 ha) that
includes the communities of Fort St. John, Hudson’s Hope, and Taylor in the Peace River
Region of British Columbia (Figure 4-1). The Beatton River and Cecil Lake Landslide study
sites described in Chapters 2 and 3 are located within this mapsheet. The area is represented
by the Boreal White and Black Spruce (BWBS) biogeoclimatic zone (Meidinger and Pojar
1991) and is completely within the boreal forest biome. The climate is temperate, with
warm, wet summers and cold winters (Meidinger and Pojar 1991). Much of the precipitation
falls in the summer months. Vegetation consists of pure and mixed conifer and aspen forests
on the uplands, transitional aspen parkland, and grasslands along the river slopes.
168

Historically, the main forms of land use in the area were farming, forestry, and fur trapping.
However, in recent decades, oil and gas activity has expanded exponentially, greatly
increasing the industrial footprint on the land through exploration, extraction, and
infrastructure.

Figure 4-1. Landslide ponds study area – Mapsheet 94A (1:250,000) Peace River Region of northeastern
British Columbia, Canada. The yellow areas indicate history of landslides along the Peace River and its
tributaries, based on shape file linework by Severin (2004). Shape file linework imagery reproduced with
permission.

169

The geology of the area consists of Quaternary stratigraphy from the historical episodes of
Glacial Lake Mathews and Glacial Lake Peace (Mathews 1978, 1980) formed by successive
advances and retreats of Cordilleran and Laurentide ice sheets. Many of the sediments are
fine clay and silt overlaying shale bedrock (Catto 1991). The Shaftesbury (marine shale) and
Dunvegan (marine and deltaic sandstone with shale interbeds) formations prevail in this area
(Mathews 1978), and most existing landslides occurred within the Lower Cretaceous
Shaftesbury Formation (Severin 2004). Many tributaries of the Peace River were formed by
the rapid carving of valleys following isostatic rebound as the glaciers retreated and the land
masses rose (Mathews 1978; Catto 1991; Hartman and Clague 2008). Prehistoric and
contemporary landslides are abundant along the Peace River and its tributaries.
4.3. Methods
This study used computerised geographic information system (GIS) techniques and large
Excel spreadsheets to compile and analyse the distribution and spatial characteristics of
pond-bearing landslides and their associated water bodies within the area of interest. It also
incorporated some field observations obtained during data collection for Chapters 2 and 3 to
enhance the discussion on pond description and development.
To limit scope for the time-intensive digitising and data entry work, the study was restricted
to a specific 1:250,000 mapsheet of the Peace River Region, mapsheet 94A, and only ponds
on landslides were inventoried and described on this 16,000 km2 area. Ponds were digitised
by hand rather than using digital image processing (i.e. supervised image analysis), as many
of the ponds were smaller than what could be captured in processing, and the aim was to map
all ponds visually discernible on the imagery.
170

To begin the mapping of landslide ponds, Mapsheet 94A (NTS 2021) was loaded from the
World Imagery layer (World Imagery 2021) into the Global Mapper GIS application (Global
Mapper 2021). Google Earth Pro imagery (Google Earth Pro 2021) was loaded on a separate
computer monitor. World Imagery and Google Earth Pro are platforms that synthesise aerial
photographs, and it was these photographs that were analysed. The resolution of both sets of
imagery was intermediate, with World Imagery having a slightly better resolution. The two
sets of imagery were used as cross-references to each other, to verify whether the features
identified were indeed landslides and ponds, or something else. While the World Imagery
had clearer features, Google Earth Pro had the advantage of providing a 3D view as well as
rapid zooming capabilities. The vintage for both types of imagery varied and covered a
range of years, typically 2006 to 2018.
Using the World Imagery map layer and a map tile grid in Global Mapper, and Google Earth
Pro as a cross-reference, the entire mapsheet was systematically assessed visually for the
presence of ponds, and all ponds located on landslides were digitised working at a scale of
approximately 1:500 (Figure 4-2). The ponds were then labelled with numbers and classified
as either ponds, dried out ponds, or wetlands. Each pond was assigned a unique number.
Although some ponds were partially or completely dried out, all were counted, since the
drying appeared to be seasonal. Early spring imagery was not always available to confirm
the extent of water pooling, but other indicators such as variations in vegetation were used to
verify. The occurrence and extent of dried ponds was evidenced by aquatic or wetland
vegetation still present, mainly recognisable as cattail (Typha spp. – most likely Typha
latifolia), which contrasted in texture and colour against the surrounding terrain. Because

171

wetlands in general are essentially a type of pond, they were combined with the ponds when
compiling and analysing the data.

Figure 4-2. Sample of landslide pond mapping. Location is Cecil Lake. Ponds are outlined in blue. Yellow
lines indicate landslide features that had been previously mapped by Severin (2004). Severin landslide shape
file lines used with permission.

Once the ponds were mapped, Severin’s (2004) landslide shape file was loaded into Global
Mapper as an overlay on the pond polygons. Severin’s digital GIS rendering of landslides
(2004) was in the format of linework rather than enclosed area polygons (see Figure 4-2), and
thus individual areas of these landslides were not obtainable in the GIS dataset. Therefore,
the table from the inventory created by Severin and the landslide areas provided in Appendix
IV of his thesis were used to obtain the initial total area of landslides. In the present study, a
small number of additional landslides that were not included in the 2004 study were
identified and digitised, as well as portions of other landslides where new movement had

172

occurred since the 2004 work done by Severin. The new area was added to the total
landslide area prior to analysis of landslide ponds. No minimum or maximum size limit was
assigned for either ponds or landslides; anything that could be discerned by zooming in on
the imagery was mapped.
A large Excel spreadsheet was created to record various data related to the ponds and
associated landslides (Appendix 7). In the spreadsheet, a unique number was assigned to
each pond-bearing landslide, in numerical order starting from the first landslide with ponds
encountered in the pond digitising exercise. The landslides were not numerically labelled in
Global Mapper, as several landslides were either eliminated or renumbered in the final
spreadsheet list. However, the geographical location (UTMs – Universal Transverse
Mercator) of the centre of each pond-bearing landslide was recorded in the spreadsheet.
Once pond and landslide numbers were assigned, the general location of each pond on the
landslide was recorded in the spreadsheet, for example if it was above or below a scarp or on
a debris apron. The geomorphic location of each pond was also recorded and classified as to
whether it was on the head, body, or toe of the landslide, using a simplified version of
Cruden and Varne’s (1996) rendering of landslide anatomy (Figure 4-3). In some situations,
ponds appeared to be on the borderline between geomorphic locations. If a pond seemed
borderline between the head and the body, the pond was classified as being on the head if it
was at the base of the main scarp before the slope changed by more than approximately 5
degrees. If the pond was located after a change in slope, it was classified as being on the
body. A similar method was applied to ponds that appeared borderline between the body and
the toe, in that ponds located at the base of the body before the slope changed significantly
were classified as being on the body, and those occurring after a slope change were classified
173

as being on the toe. The Global Mapper Profile tool applied on an Advanced Spaceborne
Thermal Emission and Reflection Radiometer (ASTER) Global Digital Elevation Model
(DEM) image of 1 arc-second resolution (ASTER GDEM v2 2021) was used to assist in
determining slope. Geomorphic location classifications were assigned initially by
incorporating information on landslides found in Cruden and Varnes (1996) and Highland
and Bobrowsky (2008) but decisions were ultimately subjective based on the characteristics
of each individual location.

Figure 4-3. Simplified anatomy of a landslide showing head, body, and toe. Adapted from Cruden and Varnes
1996. Credit: Transportation Research Board. 1996. Landslides: Investigations and Mitigation. Special
Report 247. https://doi.org/10.17226/11057. Reproduced with permission from the National Academy of
Sciences, Courtesy of the National Academies Press, Washington, D.C.

For each pond, the size in km2 was recorded as calculated in Global Mapper, and values were
then converted to hectares (ha) for the purposes of the study. The movement type of each
pond-bearing landslide was classified according to the relevant descriptions and diagrams
174

laid out in Severin’s (2004) thesis. In the study area, six pond-bearing landslide types were
identified: compound failure, mobile flow, multi-level rotational failure, retrogressive
rotational failure, rotational failure, and shallow retrogressive rotational failure. The
definitions of these types are outlined in Severin (2004) and are based on key landslide
classification works (Varnes 1978; Hutchinson 1988; Cruden and Varnes 1996). However,
Severin modified some classifications to reflect the unique characteristics of landslides in the
Peace River Region. Compound failures occur on pre-sheared surfaces in bentonitic-rich
layers of Shaftesbury shale and generally have a low angle. As the slope range parameter of
compound failures was not specified in Severin (2004), a subjective judgment was made in
the present study to classify as a compound failure any slide with a relatively long, straight
profile and an average slope of approximately 20% or less. This classification method relied
heavily on the Global Mapper Profile feature to assess slope profile characteristics. Mobile
flows usually occur in gullies along the river and fan out at the bottom of the gully. They are
shallow and are triggered during heavy rainfall and snowmelt seasons. Rotational failures
develop along pre-sheared surfaces mainly in pre-glacial sediments, mostly on south-facing
slopes. The debris is often broken up across the slope, rather than appearing as a classic
slumped block. Multi-level rotational failures take place along several weak shear planes at
different levels. The failures can occur either separately or simultaneously. Retrogressive
rotational failures develop when rupture surfaces lengthen along a weak layer in the slope
and there is progression of a single rotational slip upslope, ending at a curved back scarp.
The initial small failure destabilises the toe, causing further slumping, and there are often
back-tilted blocks of debris present. Shallow retrogressive rotational failures are similar to
retrogressive rotational failures, but only occur in Glacial Lake Peace clay. The till below
acts as a base that the failure plane cannot penetrate. The basal shear plane normally
175

coincides with the surface of the Wisconsinan till. The sediment remobilises and flows over
a stable bench and quickly disintegrates at a slope angle of 3.5 degrees.
Although the original landslide inventory (Severin 2004) categorised each landslide by
movement type and assigned it a reference number, no digital or hard copy geographical
reference database or information in the GIS shape files was available to determine the
reference number or landslide type designation of any specific landslide within the mapsheet
study area. Therefore, classification of landslide types in the current study mostly relied on
Severin’s geomorphological symbols in the map file and descriptions in the thesis text,
combined with the Profile feature in Global Mapper applied on a Digital Elevation Model
(DEM) image and my understanding of geomorphological features.
The activity level of each slide was classified using Severin’s (2004) activity classifications,
as that information was present in the shape files themselves. Severin’s classifications for
activity level included Very Active, Active, Low Activity – Dormant, Low Activity –
Abandoned, and Modified. Low Activity – Dormant landslides refer to those landslides that
are probably older than 50 years, but still have active erosion near the toe, while Low
Activity – Abandoned landslides are older than 50 years and do not currently have active
erosion at the toe. Landslides were classified as Active that were originally Dormant if there
was any new activity on them, as Active slides are defined as those with activity in the last 50
years (Severin 2004). Modified landslides were those that had been stabilised with artificial
earth works following failure.
The last column in the spreadsheet was a comments section, where pertinent information
about the ponds or landslides was recorded. After the initial inventory was completed, each
map tile of the mapsheet was methodically re-checked to verify classifications and record any
176

missed features. Some pond polygons were deleted from the final list due to not falling on an
actual landslide, while other pond polygons were deleted because they were not actually
ponds, but instead were shadows. The ponds that remained in the data set retained their
original number designations. Therefore, there were some gaps in the final numbering. For
the final spreadsheet, comments were modified or deleted as issues were addressed.
This spreadsheet was then used as a baseline for various calculations and graphs. The overall
total number and area of ponds and the total number and area of landslides for the mapsheet
were summarised. A frequency distribution of landslide ponds by size class was calculated
and graphed. The total number of landslides for each landslide type and the total number of
pond-bearing landslides for each landslide type were then calculated and compared in a
paired graph. Mean pond size overall, mean pond size per landslide type, and mean pond
size per geomorphic location within landslides were then calculated and graphed. Minimum,
maximum, and median pond sizes were also determined for each of the preceding categories.
Due to time constraints, no attempt was made to map and analyse ponds on undisturbed
terrain.
4.4. Results
4.4.1 Pond characteristics in overall study area
Of the total 1,638 landslides identified in the study area, 223 landslides with ponds were
identified, and 755 ponds were recorded on these pond-bearing landslides (Table 4-1). There
appeared to be a clustering pattern, with some large landslides containing many small ponds.
Other smaller landslides only had a few ponds. Many of the ponds appeared to be persistent,

177

with signs of cattails (Typha spp. – most likely Typha latifolia) and development into wetland
ecosystems.
Ponds occupied only a small fraction of the total landslide area. The total area of landslides
in the study area was 768.25 km2 (76,825 ha), including some added area not digitised in
Severin’s work. The total area of ponds on these landslides was 111.647 ha. Therefore,
landslide ponds only occupied 0.14% of the total landslide area. The minimum landslide
pond size was 0.0009 ha while the maximum pond size was 5.890 ha, showing a wide range
of sizes. However, most ponds were in the <1.00 ha size class (Figure 4-4). The mean pond
size was 0.15 ha, and the median pond size was 0.05 ha. Because the landslides in Severin’s
(2004) work were not digitised as enclosed polygons with a fixed area, it was not possible to
calculate the area of pond-bearing landslides separately for the present study. Therefore,
calculation of the proportion of total area occupied by ponds on just pond-bearing landslides
was not possible. Re-digitising 1,600 landslides would have added significantly to the
workload for this study and was not feasible with the time constraints and scope of the
project.

178

Table 4-1. Summary information on ponds and landslides in the 768.25 km2 (76,825 ha) study area.
Total number of landslides mapped

1,638

Total number of landslides with ponds

223

Percent (%) of total landslides that contain ponds

13.61

Total number of ponds on landslides

755

Mean number of ponds/pond-bearing landslide

3.386

Minimum number of ponds/pond-bearing

1

landslide
Maximum number of ponds/pond-bearing

29

landslide
Total pond area (ha) on landslides

111.647

Minimum landslide pond size (ha)

0.0009

Maximum landslide pond size (ha)

5.890

Mean landslide pond size (ha)

0.148

Total landslide area (ha)

76,918.527

Percentage of total landslide area covered by

0.14

ponds

179

Figure 4-4. Frequency distribution graphs showing landslide pond size distribution in hectares (ha). The upper
graph shows all landslide-generated ponds in Mapsheet 94A, while the lower graph shows the size distribution
for landslide ponds ≤ 1.00 ha.

4.4.1.1 Pond number and total area by geomorphic location
Over the study area, the majority of the 755 ponds were located on the body of landslides
(370 ponds, 49%), followed by the toe (236 ponds, 31%) and then the head (149 ponds,
20%). (Figure 4-5).

180

Figure 4-5. Total number of landslide ponds per geomorphic location.

Ponds on the body of the slides also comprised the highest total area (53.252 ha), which
represented almost 48% of the total pond area (Figure 4-6). This was followed by pond area
on the toe (27%) and pond area on the head (25%).

Figure 4-6. Total landslide pond area (ha) by geomorphic location.

181

4.4.1.2 Pond size by geomorphic location
Pond size was quite variable both within and among different geomorphic location types.
Ponds on the head had the largest average size at 0.186 ha, while ponds on the toe were the
smallest average size at 0.130 ha (Figure 4-7). The average size of ponds on the body was
0.144 ha. Overall, a location on the body contained a pond with the smallest size (0.0009 ha)
and a location on the body also contained the largest pond size (5.890 ha). The head had the
largest range of pond sizes (1.569 ha), while the body had the smallest range (0.059 ha). The
range of pond sizes on the toe was (0.356 ha).

Figure 4-7. Mean pond size (ha) by geomorphic location on landslide.

4.4.1.3 Landslide types for slides containing ponds
Landslide ponds occurred more frequently on certain landslide types (Figure 4-8). The
highest number of pond-bearing slides were retrogressive rotational failures, followed by
rotational failures and then compound failures. Proportionally, retrogressive rotational

182

failures comprised approximately 39.5% of the total landslides with ponds, followed by
rotational failures at 32.7%, and then compound failures (14.8%).

Figure 4-8. Paired graph showing overall number of landslides per landslide type compared to number of pondbearing landslides per landslide type. Landslide type names have been abbreviated to accommodate graph.
Type abbreviations: CO = Compound; EF = Earth flow; MF = Mobile flow; MR = Multi-level rotational; RA =
Ravelling; RE = Retrogressive rotational; RO = Rotational; SR = Shallow retrogressive. Note: No ponds were
found on Ravelling (RA) or Earth flow (EF) landslide types.

4.4.2 Ponds and landslide type
4.4.2.1 Number of ponds per landslide type
Pond numbers ranged widely across landslide types (Figure 4-9). Retrogressive rotational
failures contained by far the greatest overall number and proportion of the total 755 ponds
(394 ponds, approximately 52%). The next highest proportion of ponds was on rotational
failures (20%, 151 ponds), followed by compound failures (14%, 107 ponds). Shallow
retrogressive failures had the lowest proportion of total ponds (approximately 2%, 13 ponds).
183

Figure 4-9. Number of ponds per landslide type.

4.4.2.2 Pond area per landslide type
Within the study area, the planimetric surface area occupied by landslide ponds primarily
occurred in one landslide type (Figure 4-10). By far, retrogressive rotational failures
contained the highest total area (77.696 ha, 69.59 %), followed distantly by rotational failures
(18.877 ha, 16.91 %) and then compound failures (5.737 ha, 5.14 %).

Figure 4-10. Pond area (ha) in each pond-bearing landslide type.

184

4.4.2.3 Mean pond size per landslide type
Pond size varied widely between landslide types in the study area (Figure 4-11 and Appendix
8). Retrogressive rotational failures had the largest average pond size (0.197 ha), followed
by rotational failures (0.125 ha), and then multi-level rotational failures (0.102 ha). Shallow
retrogressive slides had the smallest average pond size (0.051 ha). There was quite a range
of pond sizes both among and within slide types, with large standard deviations.
Retrogressive rotational failures had the smallest minimum pond size (0.0009 ha), as well as
the largest maximum pond size (5.890 ha). Mobile flows had the largest minimum pond size
(0.005 ha). Shallow retrogressive failures had the smallest maximum pond size (0.287 ha).
Within slide types, retrogressive rotational failures had the greatest pond size range (5.889
ha), followed by rotational failures (1.651 ha). Shallow retrogressive rotational failures had
the smallest pond size range (0.285 ha). However, it should be noted that there were only a
few shallow retrogressive rotational failures in the data set. The next smallest pond size range
was within compound failures, at 0.290 ha.

Figure 4-11. Mean pond size (ha) per landslide type.

185

4.4.3 Geomorphic location of ponds and landslide type
This section presents the results of compilation and analysis of pond area and size for various
combinations of pond geomorphic locations and landslide types. In all cases for the results
regarding pond size, the standard deviations were quite large, often much larger than the
average size (see Appendix 9).
4.4.3.1 Head and landslide type
4.4.3.1a Number and proportion of ponds on head per landslide type
Ponds on the head of landslides occurred much more frequently on some landslide types than
on others (Figure 4-12). The highest number and proportion of ponds on the head were on
retrogressive rotational failures (80 ponds, 54%). Rotational failures had the next highest
number and proportion of ponds (29 ponds, 19%), followed by multi-level rotational failures.
The smallest number and proportion of ponds on the head occurred on the single mobile flow
on which landslide ponds were detected.

Figure 4-12. Number of ponds on head per landslide type.

186

4.4.3.1b Pond area on head per landslide type
The area occupied by ponds on the head of slides varied strongly by landslide type (Figure 413). Most of the total area of ponds on the head of slides was represented by retrogressive
rotational failures, comprising 22.493 ha (81.21 %). The second highest area was on
rotational failures, with a much smaller value of 2.377 ha (8.58 %), followed by compound
failures (1.302 ha, 4.70 %).

Figure 4-13. Pond area (ha) on head per landslide type.

4.4.3.2 Body and landslide type
4.4.3.2a Number and proportion of ponds on body per landslide type
Occurrence of ponds on the body of landslides varied but showed a tendency towards certain
landslide types (Figure 4-14). With a similar trend as for ponds on the head, the highest
number and proportion of ponds on the body occurred on retrogressive rotational failures
(192 ponds, 50.26%). Rotational failures had the second highest number and proportion of
187

ponds (81 ponds, 21.2 %). Shallow retrogressive failures represented the smallest number
and proportion of ponds on the body (5 ponds, 1.31%).

Figure 4-14. Number of ponds on body per landslide type.

4.4.3.2b Pond area on body per slide type
The area occupied by ponds on the body of slides showed a strong tendency toward just one
or two landslide types in the study area (Figure 4-15). Retrogressive rotational failures
contained the highest overall area of ponds on the body, at 36.946 ha (69.38 %). Rotational
failures were a distant second at 9.983 ha (18.75 %), followed by compound failures (2.849
ha, 5.35 %).

188

Figure 4-15. Total pond area (ha) on body per landslide type.

4.4.3.3 Toe and landslide type
4.4.3.3a Number and proportion of ponds on toe per landslide type
The occurrence of ponds on the toe of a landslide was variable, but more evenly distributed
among landslide types than that of ponds on the body (Figure 4-16). More than half of ponds
on the slide toe occurred on retrogressive rotational failures (122 ponds, 51.69%). The next
highest number of ponds on the toe was represented by rotational failures (41 ponds,17.37%),
similar to the trend for ponds on the head and body. No ponds on the toe were located on
mobile flows, and the next lowest pond number on the toe was found on shallow
retrogressive failures (3 ponds, 1.27%).

189

Figure 4-16. Number of ponds on toe per landslide type.

4.4.3.3b Pond area on toe per landslide type
Total area of ponds on the toe of landslides was somewhat variable between landslide types
(Figure 4-17). The pond area on the toe was highest on retrogressive rotational failures, at
18.257 ha (59.47 %), and this slide type also had the greatest overall pond area in the study
(see Figures 4-13 and 4-14). The second highest area of ponds on the toe was on rotational
failures (6.517 ha, 21.23 %), followed by multi-level rotational failures (4.124 ha, 13.43 %).

190

Figure 4-17. Pond area (ha) on toe per landslide type.

4.4.4 Landslide pond development
The landslide-generated ponds in the study area exhibited varying stages of development,
both between and within individual landslides. Because of the seasonal nature of pond water
levels and vegetation, it would not have been accurate to attempt to categorise the ponds in
the study according to development stage. However, some representative photographs
illustrate the diversity of vegetation development present on landslide-bearing ponds in the
mapsheet study area. These photographs were taken during fieldwork for Chapters 2 and 3 in
the same area. Figure 4-18 shows a pond recently formed in fresh unvegetated soil.
Figure 4-19 shows a pond with a more stable bank, with vegetation growing around the
perimeter of the pond. Figure 4-20 shows a pond with an obviously stable bank, surrounded
by vegetation, and containing aquatic vegetation including cattail (Typha spp. – most likely
Typha latifolia).

191

Figure 4-18. Very new pond on landslide (Beatton River Landslide). The most recent major movement on the
landslide occurred in 2015, so the pond was approximately three years old at time of photograph. (Photo June
19, 2018)

Figure 4-19. Newer pond on landslide (Beatton River Landslide). Pond is on an older part of the landslide near
the toe. (Photo August 9, 2017)

192

Figure 4-20. Persistent pond on landslide (Cecil Lake Landslide). Landslide occurred in 1998, thus the pond
was approximately 20 years old at time of photograph. (Photo June 23, 2017)

4.5. Discussion
This study set out to compile an inventory of landslide ponds on NTS Mapsheet 94A and
describe area, size, and distribution of the ponds in relation to overall number and area of
landslides, geomorphic locations on the landslide, and occurrence on landslide types. The
overarching purpose was to employ this information to identify potential ecological
implications of pond occurrence and distribution and provide recommendations for
management.

193

4.5.1 Landslide pond characteristics and distribution
The results showed a wide range of pond sizes, with the average size of ponds for each
landslide type falling within an intermediate range. Research has shown that intermediate
sized ponds (ranging in size from 200 to 4000 m2, or 0.02 to 0.4 ha) contain the highest
density, richness, and diversity of pond-breeding amphibians (Semlitsch et al. 2015). In the
present study, 497 ponds (over 65%) fell within the intermediate size range. This could be
significant for maintenance of amphibian diversity in the study area. Although no amphibians
or aquatic insects were observed in the ponds encountered on the Cecil Lake, Beatton River,
and Hasler Flats landslides (Chapters 2 and 3), waterfowl were seen on some of the ponds.
The ponds in the study area also exhibited a range of stages of evolution. This diversity of
developmental stages could allow for a variety of different macroinvertebrate communities to
develop, each taking advantage of the particular hydrologic, sedimentary, and vegetative
conditions present (Jeffries 2011).
The characteristics of each geomorphic location on the landslide may influence the
distribution of ponds. Generally, each part of a landslide has different kinds and orientations
of geomorphic structures (Parise 2003). In the present study, ponds were most prevalent on
the body of landslides, followed by the toe. Ponds on the body represented the highest total
number, as well as the highest total area. Greater pond presence on the landslide body may be
due to the diversity in topography that results as the slide material is moved, creating
depressions and cutting off drainage. As the landslide stabilises, vegetation starts to
encroach, further stabilising ponds. The conditions are generally better for plants to persist on
the body compared to the steeper nature of the head, which prevents soil from building up
and hinders plant root establishment and persistence (Walker and Shiels 2013).
194

As discussed in the Introduction, a network of varying sizes of ponds can provide value for
dispersal of water-dependent organisms (Pop and Chitu 2013). Ponds on the toe represented
the second highest number and total area of all the mapped ponds but had the smallest
average size. These toe ponds had a size range that was less than the head and greater than
the body, and thus had greater size diversity than ponds on the body. This size variation
could be due to the broken up and uneven nature of the terrain often present on the toe, where
large amounts of material are deposited, often rapidly.
Although ponds were more prevalent on the landslide body, the head had the largest
maximum pond size and the greatest size range. Most ponds on the head of the landslide
were at or near the transition zone with the body. Deep, long depressions were more likely to
be present at this sharp transition in slope from positive to negative, forming the larger ponds.
The smaller ponds on the head tended to be in small depressions. These findings suggest that
the head of a landslide may provide more diverse habitats for some aquatic organisms, as
well as for larger animals seeking water or shade.
The apparent relationship between landslide type and number of ponds in the study area may
be a function of the underlying material and the topographical characteristics of the landslide.
Most of the ponds in the data set were located on either retrogressive rotational slides or
rotational slides, followed by compound slides. By far, most ponds were found on
retrogressive rotational slides. The persistence of ponds on this landslide type may be due to
pre-existing fault planes that form depressions and restrict drainage following movement of
slide material. Retrogressive rotational slides have multiple weak layers and fractures
(Severin 2004), creating many potential sites for ponds. Ponds on retrogressive rotational
slides also had the greatest mean size and the greatest range of sizes, followed by ponds on
195

rotational slides. Retrogressive rotational slides tend to produce a series of geomorphic
features as the land mass slumps and shifts. This process may result in a great variation of
depressions and cracks where ponds can form. Rotational failures have a similar
configuration, but with fewer fracture planes (Cruden and Varnes 1996; Severin 2004). The
compound failures possibly have a relatively high proportion of ponds in the study because
the overall slope on this type of landslide is gentler, enhancing the ability of the material to
retain water. The fewest number of ponds were on shallow retrogressive failures. It is
possible the fracture planes of this landslide type were not severe enough to create persistent
depressions for water retention.
Geomorphic location of ponds on the landslide varied somewhat with landslide type. For
ponds on the head, by far the highest number and total area were on retrogressive rotational
failures, followed by rotational failures and then compound failures. This prevalence on
retrogressive rotational failures could be due to the relatively steep headscarp of these types
of landslides and the associated deep and wide fissures at the transition zone with the body.
Regarding ponds on the body, the highest number and area of ponds was again on
retrogressive rotational failures, followed by rotational failures and then shallow
retrogressive failures. Total pond area on the body was higher than for ponds on the head for
most landslide types noted. For ponds on the toe, the trend was the same, with the highest
number and area of toe ponds occurring on retrogressive rotational failures and rotational
failures. Retrogressive rotational failures have multiple shear zones (Cruden and Varnes
1996), which can produce depressions and fissures at all geomorphic locations of the
landslide. Overall, the results suggest that retrogressive rotational failures are more

196

ecologically diverse than the other landslides sampled regarding pond size, area, distribution,
and persistence.
4.5.2 Landslide pond dynamics
Although ponds on landslides can persist for many decades, they evolve and change over this
time. The ponds in the study area showed a diversity of stages of evolution. Initially, when
most ponds form after a landslide event, they are in fresh unvegetated soil (see Figure 4-18).
Over time, the vegetation starts to grow in around the pond from propagules either within the
soil or from surrounding rafts of vegetation or from seeds dispersed from adjacent forest
vegetation (see Figure 4-19). This vegetation can serve to stabilise the pond, and it also can
influence the ecology of the site. Eventually, the area surrounding the pond may become
fully vegetated and the pond may persist for years (see Figure 4-20).
As habitat and water persistence change, so do the populations and compositions of plants
and animals. For example, macroinvertebrate communities develop and change in response
to pond persistence and hydrological cycles, and at times there is a fine threshold between
different community compositions (Jeffries 2011). The gradual revegetation of the pond site
can also influence the size and persistence of the pond. Surrounding vegetation as well as
aquatic vegetation can shrink the pond either seasonally or over years, as established
vegetation takes up the water through transpiration, eventually lowering the overall reserves.
4.5.3 Beaver influence on landslide ponds
Evidence on the study sites visited for field sampling in Chapters 2 and 3 suggested that
beavers can play a part in forming or altering ponds on landslides. The North American
beaver (Castor canadensis) is common in the study area and throughout northeastern British
197

Columbia and can significantly impact landscapes that have water present. Beavers can alter
hydrogeomorphic and ecological processes through dam building and the associated felling
of trees, and excavation and transport of large amounts of sediments, resulting in the flooding
of various terrains (Butler and Malanson 1995; Westbrook et al. 2017). Beaver-modified
landscape patches produce distinct habitats that can increase richness and abundance of
terrestrial and semi-aquatic mammals (Nummi et al. 2019). Although it was difficult to
identify beaver ponds in the imagery used, in the field there was evidence of beaver activity
on the Cecil Lake landslide study site (Figures 4-21, 4-22, and 4-23), as well as the Hasler
Flats landslide study site (Figure 4-24) sampled in work for Chapters 2 and 3. The beavers
appear to have significantly influenced the configuration and size of the bigger ponds on site
at the Cecil Lake landslide (Figure 4-22). Field evidence of old weathered, advanced-decay
gnawed logs and stumps (Figure 4-23) indicate beavers have maintained a presence for many
years and operated in cycles on different parts of the landslide. In many areas on the Cecil
Lake landslide there were small beaver-browsed sapling stumps and beaver trails throughout
the woods leading to ponds. On the Hasler Flats landslide, there was evidence of very recent
felling of large aspen trees in addition to well-worn trails leading from the sidescarp to the
landslide ponds (Figure 4-24).

198

Figure 4-21. Recent beaver gnawing activity on young sapling near pond (Cecil Lake Landslide). (Photo
September 7, 2017)

Figure 4-22. Beaver pond on the Cecil Lake Landslide. (Photo June 22, 2017)

199

Figure 4-23. Older beaver gnawing activity on the Cecil Lake Landslide. (Photo September 10, 2017)

200

Figure 4-24. Recent beaver activity at the Hasler Flats landslide. Top image shows well-used beaver path
(bottom centre of photo) leading from the sidescarp to a landslide pond. Lower image shows very recent felling
of large aspen trees (Populus tremuloides) by beavers just above the sidescarp. (Photos August 14, 2018)

201

4.5.4 Limitations of the study
A limitation to the landslide ponds study was the intermediate and variable resolution of the
imagery used when determining and mapping landslide ponds. Some ponds may have been
missed if they were too small, or if the available imagery in the area was of a poorer
resolution. However, the vast majority of ponds present were most likely identified and
digitised, and any missed ponds probably did not affect the overall results.
Another limitation of the study was the fact that the pre-existing digitised landslides did not
have individual areas assigned to them in a format that could be georeferenced. Thus,
calculation of proportions of total area of pond-bearing landslides occupied by ponds was not
possible. This information would have been a useful metric to have for comparison and
consideration. However, the landslide area summary information that was available provided
an initial baseline for calculating the proportion of total landslide area occupied by ponds.
The georeferenced and numbered landslide type designations that Severin (2004) assigned to
each landslide were also unavailable, so in this study I assigned my own classifications for
each individual pond-bearing landslide based on Severin’s definitions. Landslide types were
classified mainly using Severin’s geomorphic symbols and the DEM imagery. My
designations may not have been completely accurate, as I did not have access to high
resolution aerial photographs as stereo pairs and am not a trained geomorphologist.
Nevertheless, the classifications were likely accurate enough for distinguishing between
obviously different landslide types and describing pond distribution.

202

An additional limitation to the study was the lack of previous research on landslide ponds, for
comparison. There have not been many studies to inventory landslide ponds, let alone to
describe ecological elements. This study is thus a baseline for future research.
4.5. Conclusions and recommendations
In this study I provided an initial inventory of ponds on landslides in an area of northeastern
British Columbia that is particularly susceptible to landslides. I also presented details on the
number, size, and distribution of ponds on different geomorphic locations within landslides
and among different landslide types. Although they only occupy a small proportion of the
total landslide area in the mapsheet study area, landslide ponds may provide important
ecological roles for amphibians, ungulates, birds, and other wildlife both at the local and
landscape scales in the Peace River Region. These ponds serve many purposes, including
nesting, feeding, shelter, water, protection, connectivity, and biodiversity.
Pond size and distribution impact wildlife species richness and overall diversity, as an array
of smaller ponds yields a greater number of species and higher conservation value than a
large pond of the same total area (Oertli et al. 2002).
This study provides the first detailed baseline information on landslide pond distribution in
the Peace River Region and adds a valuable component to the knowledge base on water
bodies in the area. The findings present a benchmark for conservation considerations in the
Region. As landslide ponds do not occur in great numbers on the landscape here and yet are
potentially high in ecological value, efforts should be made to conserve them. Conservation
would allow for preservation of breeding, nesting, feeding, and shelter habitats for various
species, as well as connectivity for migrating species. Knowledge about the specific

203

ecological value of these landslide ponds is significantly lacking and requires further
investigation. To better understand how these ponds are used by wildlife, a subset of the
pond-bearing landslides should be selected to sample in the field. Ecological information
such as pond plant species composition, aquatic invertebrates, and signs of wildlife use
should be recorded in detail for each pond in the subset. Ideally, such a study would be
carried out over multiple seasons and years to capture the full breadth of use of the pond, as
well as any changes to the pond. It would also be informative to carry out similar research in
landscapes dominated by other landslide types such as rock falls.
Although the ecological value of landslide ponds is recognised, geohazard assessment
considers the presence of pooled water on unstable slopes a dangerous situation to be
avoided. In fact, some management measures recommend draining ponds on landslides
(Kansas Geological Survey 1999). Therefore, a balance must be sought between conserving
important habitat and preventing catastrophic reactivation of landslides. Landslide ponds
may serve an important role as an indicator for land management decisions concerning
infrastructure, resource development, home building, and other activities on the land base.
Persistent ponds signify a high water table and soil saturation, conditions which can indicate
the potential for slope instability and increased possibility of flooding. Care should be taken
to develop away from areas where landslide ponds are present. If landslides with ponds
already exist near developed areas or sensitive fisheries habitat, they should be monitored on
a regular basis and assessed for reactivation. It is possible some ponds should be drained if
there are signs of imminent danger of slope movement. For landslides in remote areas,
however, the ponds should be left intact to provide connectivity, habitat, and other ecological
roles.

204

4.6 References
ASTER GDEM v2. 2021. Advanced Spaceborne Thermal Emission and Reflection
Radiometer (ASTER) Global Digital Elevation Map. https://asterweb.jpl.nasa.gov/gdem.asp
Bobrowsky, P.T., N. Catto, and V. Levson. 1990. Reconnaissance Quaternary geological
investigations in Peace River District, British Columbia (93P, 94A). Geological Fieldwork,
Paper 1991-1.
Bobrowsky, P.T., and N.W. Rutter. 1992. The Quaternary geologic history of the Canadian
Rocky Mountains. Geographie physique et Quaternaire, 46(1): 5-50.
Butler, D.R., and G.P. Malanson. 1995. Sedimentation rates and patterns in beaver ponds in
a mountain environment. Geomorphology, 13: 255-269.
Catto, N.R. 1991. Quaternary geology and landforms of the eastern Peace River region,
British Columbia NTS 94A/1, 2, 7, 8. Province of British Columbia, Ministry of Energy,
Mines and Petroleum Resources, Mineral Resources Division, Geological Survey Branch,
Open File 1991-11, 23 pages.
Catto, N., D.G.E. Liverman, P.T. Bobrowsky, and N.W. Rutter. 1996. Laurentide,
Cordilleran, and Montane glaciation in the Western Peace River – Grande Prairie Region,
Alberta and British Columbia, Canada. Quaternary International, 32: 21-32.
Coe, J.A., W.L. Ellis, J.W. Godt, W.Z. Savage, J.E. Savage, J.A. Michael, J.D. Kibler, P.S.
Powers, D.J. Lidke, and S. Debray. 2003. Seasonal movement of the Slumgullion landslide
determined from Global Positioning System surveys and field instrumentation, July 1998 –
March 2002. Engineering Geology, 68: 67-101.
Coe, J.A., J.P. McKenna, J.W. Godt, and R.L. Baum. 2009. Basal-topographic control of
stationary ponds on a continuously moving landslide. Earth Surface Processes and
Landforms, 34: 264-279.
Cruden, D.M., Z.-Y. Lu, and S. Thomson. 1997. The 1939 Montagneuse River landslide,
Alberta. Canadian Geotechnical Journal, 34: 799-810.
Cruden, D., and D. Varnes. 1996. Landslide types and processes. In: Turner, A., Schuster, R.
(eds.), Special Report 247: Landslides Investigation and Mitigation. National Research
Council, Transportation Research Board, Washington, DC, pp. 36-75.
Dandurand, R. 2018. Investigating regional groundwater flow influences on slope stability
in unlithified materials. Unpublished Master of Science Thesis, Simon Fraser University,
139 pages.
Fletcher, L., O. Hungr, and S.G. Evans. 2002. Contrasting failure behaviour of two large
landslides in clay and silt. Canadian Geotechnical Journal, 39: 46-62.
Geertsema, M., and J. Pojar. 2007. Influence of landslides on biophysical diversity - A
perspective from British Columbia, Geomorphology, 89: 55-69.
Global Mapper. 2021. https://www.bluemarblegeo.com/global-mapper/
205

Google Earth Pro. 2021. https://www.google.com/earth/versions/
Hartman, G.M.D. 2005. Quaternary stratigraphy and geologic history of the Charlie Lake
(NTS 94A) map-area, British Columbia. M.Sc. thesis, Simon Fraser University, Burnaby,
B.C., 165 pp.
Hartman, G.M.D., and J.J. Clague. 2008. Quaternary stratigraphy and glacial history of the
Peace River valley, northeast British Columbia. Canadian Journal of Earth Science, 45: 549564.
Highland, L.M., and P. Bobrowsky. 2008. The landslide handbook – A guide to
understanding landslides. Reston, Virginia, U.S. Geological Survey Circular 1325. 129 pp.
Hutchinson, J.N. 1988. General Report: Morphological and geotechnical parameters of
landslides in relation to geology and hydrogeology. In Proceedings, 5th International
Symposium on Landslides, Lausanne, Vol. 1, pp. 3-35.
Jeffries, M.J. 2011. The temporal dynamics of temporary pond macroinvertebrate
communities over a 10-year period. Hydrobiologia, 661:391-405.
Kansas Geological Survey. 1999. Public Information Circular (PIC) 13. Kansas Geological
Survey, Public Outreach. http://www.kgs.ku.edu/Publications/pic13/pic13_5.html
Krueger, K.A., and B.G. Johnson. 2016. The effect of sub-alpine landslides on headwater
stream gradient and beaver habitat. Physical Geography, 37(5): 344-360. DOI:
10.1080/02723646.2016.1218723
Mathews, W.H. 1978. Quaternary stratigraphy and geomorphology of Charlie Lake (94A)
map-area, British Columbia. Geological Survey of Canada, Paper 76-20, 25 pages.
Mathews, W.H. 1980. Retreat of the last ice sheets in northeastern British Columbia and
adjacent Alberta. Geological Survey of Canada Bulletin 331, 22 pages.
Meidinger, D., and J. Pojar. (Eds.) 1991. Ecosystems of British Columbia (342 p). Special
Report Series 6, Victoria, BC: B.C. Ministry of Forests and Range Research Branch.
National Topographic System (NTS). 2021. Website. https://naturalresources.canada.ca/earth-sciences/geography/topographic-information/maps/nationaltopographic-system-maps/9767
Nummi, P., W. Liao, O. Huet, and E. Scarpulla. 2019. The beaver facilitates species
richness and abundance of terrestrial and semi-aquatic mammals. Global Ecology and
Conservation, 20: 1-10.
Parise, M. 2003. Observation of surface features on an active landslide, and implications for
understanding its history of movement. Natural Hazards and Earth System Sciences, 3: 569580.
Pop, D.E., and Z. Chitu. 2013. Landslides and biodiversity conservation: the importance of
an integrated approach. A case study: the Subcarpathian part of the Doftana watershed
(Prahova County, Romania). Revista de geomorfologie, 2: 57-68.

206

Sasaki, N., and T. Sugai. 2015. Distribution and development processes of wetlands on
landslides in the Hachimantai volcanic group, NE Japan. Geographical Review of Japan
Series 8, 87(2):103-114.
Semlitsch, R.D., W.E. Peterman, T.L. Anderson, D.L. Drake, and B.H. Ousterhout. 2015.
Intermediate pond sizes contain the highest density, richness, and diversity of pond-breeding
amphibians. PLoS ONE 10(4): e0123055. doi:10.1371/journal.pone.0123055
Severin, J.M. 2004. Landslides in the Charlie Lake map sheet, Fort St. John. Unpublished
Master’s thesis, University of British Columbia, 395 pp.
Shapley, M.D., B.P. Finney, and C.R. Krueger. 2019. Characteristics of landslide-formed
lakes of central Idaho: high-resolution archives of watershed productivity and clastic
sediment delivery. Geological Society of America, Special Paper 536, pp 241-258.
Varnes, D.J. 1978. Slope movement types and processes. In Landslides – Analysis and
Control. Special Report of the National Research council and Transportation Research
Board, 176. Edited by R.L. Schuster and R.J. Krizek. pp. 11-33.
Van Esch, K.J.B. 2012. Failure behaviour of bedrock and overburden landslides of the Peace
River Valley near Fort St. John, British Columbia (T). University of British Columbia.
Retrieved from https://open.library.ubc.ca/collections/ubctheses/24/items/1.0072987
Walker, L.R., and A.B. Shiels. 2013. Landslide Ecology. Cambridge University Press, New
York. 300 pages.
Westbrook, C.J., D.J. Cooper, and C.B. Anderson. 2017. Alteration of hydrogeomorphic
processes by invasive beavers in southern South America. Science of the Total Environment,
574: 183-190.
World Imagery. 2021.
https://www.arcgis.com/home/item.html?id=10df2279f9684e4a9f6a7f08febac2a

207

Chapter 5. Conclusions and Recommendations for Landslide Recovery and
Management
5.1 Conclusions
This research set out to describe and quantify various aspects of biophysical diversity on
landslides in the landslide-prone Peace River Region of northeastern British Columbia and
present ecological implications and management recommendations based on the findings.
The central research questions for this research were:
(1) Are landslides demonstrably more biophysically diverse than undisturbed ecosystems?
(2) To what extent do landslides rearrange the relative abundance of site series/types on a
slope compared to adjacent undisturbed terrain?
(3) What is the extent turnover in microsite and plant species diversity on landslides, and
how does this compare to adjacent undisturbed terrain?
(4) Is vegetation diversity on landslides significantly related to geomorphological diversity?
(5) What is the distribution and abundance of landslide ponds at a regional and local scale,
and what are the ecological and management implications?
The overarching purpose of the work was to compare biophysical diversity within and among
landslides, and between landslides and surrounding undisturbed terrain. An additional
objective was to investigate possible correlations between vegetation diversity and site
diversity. Vegetation and site characteristics on three different landslides were measured and
analysed for alpha diversity and beta diversity values (Chapter 2 and Chapter 3, respectively),
and characteristics of ponds on landslides within a larger area in the region were also
assessed, described, and analysed (Chapter 4). The findings of this study revealed that

208

quantification and prediction of biophysical diversity on a severe disturbance such as a
landslide is complex and highly dependent on individual study site characteristics. Despite
these challenges, it is still possible to analyse the results and draw learnings for application in
management, ecology, and rehabilitation on landslides in the Peace River Region and for
landslides at the broader scale.
This work showed that following a mass movement in the Peace River Region, plant
community composition on landslides varies depending on the age and size of the landslide
and the slope and soil development of the various geomorphological features present. Exotic
forb species tend to dominate in early stages of landslide revegetation and can persist if the
disturbance periodically reactivates, effectively preventing the establishment of shrubs and
trees.
The landslides in the study were overall less diverse in alpha diversity of plant communities
than the surrounding undisturbed terrain, a finding which diverged from initial expectations.
However, the landslides were more diverse in abundance and distribution of site types/series
and geomorphic features than the undisturbed terrain. Although there was a greater
proportion of mesic sites on landslides, there were also more extreme site series on a scale of
moisture regime. Geomorphic types were overall more diverse on the landslides due to mass
movement and substrate rearrangement, and type diversity and surface roughness both tended
to decrease with age of the landslide. Therefore, although the landslides were lower in plant
alpha diversity, the site series and geomorphic diversity present provide conditions for a
greater variety of plant communities and wildlife habitats over time.

209

Beta diversity often reflects the health and proper functioning of an ecosystem. Beta
diversity or turnover of both vegetation and environment variables was generally higher on
landslides than the surrounding undisturbed terrain, with a strong relationship between the
vegetation and environment beta diversities. Beta diversity was also affected by spatial and
temporal characteristics of the study area. The largest and oldest landslide, Cecil Lake,
showed the highest mean vegetation beta diversity and also the most variable vegetation beta
diversity. This suggests that processes affecting turnover of plant communities may be at
work to varying degrees on different parts of a landslide within a given time period. The
vegetation beta diversity on the surrounding terrain was both lower and much less variable,
suggesting a state of relative stability. The smallest landslide, Hasler Flats, had the highest
beta diversity but also had much less variability. This suggests size can inversely influence
beta diversity, with smaller areas facilitating greater relative patchiness. Overall,
environment beta diversity was much lower than vegetation beta diversity for all study sites,
and it was also lower on undisturbed areas compared to the landslides. Cecil Lake and
Hasler Flats landslides both had environment beta diversity values that were nearly twice as
much as the youngest landslide, Beatton River, but Cecil Lake had more than twice the
variability as Hasler Flats. Beatton River had a markedly higher variability. Interestingly,
Beatton River had some higher vegetation and environment alpha diversity values than the
oldest landslide, Cecil Lake. The above findings highlight the variability of the effects of
patchiness on biophysical diversity in general.
This study of landslides in the Peace River Region uncovered variable relationships between
vegetation and site or environment diversity, depending on the spatial and temporal scale of
the samples. The NMS ordination analysis on the BEC vegetation plots showed some weak

210

relationships between plant community development and slope gradient, and to a lesser
degree mesoslope position, heat load index, and moisture. Contrary to expectations, a
significant relationship was not found between vegetation alpha diversity and environment
(geomorphic) alpha diversity on any of the three landslides studied (Chapter 2). However,
there was a significant positive relationship between vegetation beta diversity and
environment beta diversity on the transects sampled in Chapter 3. These findings suggest
that while within-plot vegetation diversity seems to be independent of within-plot
environmental diversity, vegetation turnover over space is distinctly linked to microsite
condition turnover. The complex nature of these relationships between vegetation and
environment indicates a scale-dependency that is not yet clear and requires further
investigation. In general, it appears that during the first 20 or more years following landslide
occurrence, plant community succession is still sorting out, at the same time as the terrain is
weathering and evolving.
Although both prehistoric and historic landslides are abundant along the Peace River and its
tributaries, the research in Chapter 4 found only a small proportion of these landslides
contained persistent ponds. The results showed trends in landslide pond size, geomorphic
location on the landslide, and association with certain landslide types. Most ponds were
under one hectare in surface area, with most being less than one-tenth that size. Further,
these ponds tended to occur more frequently and in greater numbers on certain types of
landslides, namely retrogressive rotational, rotational, and multi-level rotational landslides.
In addition, ponds occurred in greater concentrations on the body of the landslide, followed
by the toe and then least frequently on the head. However, ponds on average tended to be
larger on the head of landslides. This is likely due to the presence of the rupture zone below

211

the headscarp, where movement of material away from the zone followed by stabilisation can
cause large depressions, restricting drainage. Landslide ponds encountered during fieldwork
for Chapter 2 and Chapter 3 had evidence of beaver activity and alteration, as well as use by
waterfowl and other animals.
5.2 Recommendations
The findings of this research provide a foundation to begin managing landslides in the Peace
River Region from an ecological perspective, considering succession and the influence of
spatial and temporal scales. Landslides close to communities, infrastructure, or other
important developments or ecosystems should be the priority for management and
restoration. Landslides abutting large rivers used for drinking water or important fishbearing streams should be given special attention, as sediment input can be substantial. The
type of landslide and its geomorphological characteristics may also provide guidance for
prioritising management. Rotational landslides tend to be more susceptible to reactivation,
especially those with steeper slopes of perhaps 20 degrees or more. The presence of moving
water within the landslide also tends to reactivate portions of the slide. The Beatton River
landslide had some areas of seepage and debris flow, while the Cecil Lake landslide was
influenced by a creek running through it from the south, creating an active gully. When
landslides first occur, they should be assessed in the field to determine potential risks for
reactivation.
Once a landslide has been assessed and is determined safe to work on, initial measures
should focus on revegetating and stabilising the terrain to enhance the rate of ecological
succession. Invasive, exotic plant species such as Melilotus spp. and Sonchus spp. are
obviously very effective at colonising, stabilising, and enriching freshly disturbed terrain.
212

However, these species pose the problem of creating reduced plant community diversity and
structure over time. To alleviate this, land managers should also plant a variety of
competitive, rapidly growing native pioneer grass seed mixes such as blue wildrye (Elymus
glaucus) and Canada bluejoint (Calamagrostis canadensis), forbs including wild sarsaparilla
(Aralia nudicaulis), palmate coltsfoot (Petasites frigidus), showy aster (Aster conspicuus)
and false Solomon’s seal (Smilacina racemosa), and common shrubs prickly rose (Rosa
acicularis), saskatoon (Amelanchier alnifolia), alder species (Alnus spp.), red-osier dogwood
(Cornus stolonifera), and highbush cranberry (Viburnum edule), as found most abundantly on
the landslides in this study. In addition, ecologically suitable deep-rooted or high
evapotranspiration trees such as trembling aspen (Populus tremuloides), white spruce (Picea
glauca) and cottonwood (Populus balsamifera) should be planted where possible and safe to
do so to help stabilise the soil and take up extra moisture. Decisions on species selection
should incorporate local and regional goals for landscape ecosystem health and measures
around climate change adaptation. Organic amendments and bird perches in open areas may
also be necessary to ensure adequate substrate conditions, plant dispersal and establishment.
Further, these sites should be monitored over time, and managers should be prepared to replant or stabilise the slope if steep portions of the landslide are reactivated. It is also
important to note that there are often other disturbances interacting with landslides, such as
floods and fires. These disturbances should also be taken into consideration when planning
and monitoring restoration.
Because of the complexities of landslide succession and the potential for reactivation,
restoration on landslides should focus on ecosystem recovery and biodiversity, rather than
species composition. Ecosystem recovery is not intended to return the disturbed system to its

213

historic condition, but rather to its historic trajectory. Work should involve stabilising the
soil through planting and bioengineering measures, rehabilitating water courses, and creating
a diverse multi-level vegetation cover of mostly native species. The measures described
above could be applicable to deep-seated, moderate to steep landslides occurring on glaciated
sites in unconsolidated glaciolacustrine material elsewhere in British Columbia and in other
parts of the world.
Much remains to be learned about landslide recovery and ecology in the Peace River Region,
and the Beatton River landslide could provide an outdoor laboratory for restoration
experimentation and trials. Because the landslide is so young and much of the surface is
steep and unstable, there are many opportunities to try different slope stabilisation and
revegetation measures. The trials could be easily monitored, as access to the landslide is
good.
Further research should focus on a variety of landslide types, sizes, and ages in the region,
and these methods could also be extended to other regions. Similar vegetation and microsite
sampling for both the landslide and the surrounding terrain should be employed, preferably
with an increased sample size for the relevés and transects and greater utilisation of remotesensed imagery. To better understand the dynamics of colonisation by native plant species,
an attempt should be made to locate remote landslides free from the influence of invasive
exotic species.
Although relatively sparse at the regional scale and generally small in size, landslide ponds
may have important implications for landslide stability and ecosystem health. The presence
of ponds indicates saturation, therefore pond-bearing landslides near priority management

214

areas should be monitored so that they can be stabilised quickly if they reactivate. These
ponds should also be conserved for their ecological value for amphibians and invertebrates,
as well as for habitat and a feeding and drinking source for waterfowl and fur-bearing
mammals. Further research is needed to investigate the ecological significance of these
features.
5.3 Final thoughts
The research that formed the basis of this dissertation was initiated to address the gap in
knowledge about biophysical diversity and recovery on landslides in northeastern British
Columbia’s Peace River Region, a glaciated area highly susceptible to landslide activity. The
work is the first of its kind in the region, as there were no previous published diversity
studies that collected and analysed detailed field data on landslide vegetation species or
ponds, let alone at such a large scale of study.
This research provided quantitative confirmation of the increased diversity of some
biophysical aspects of landslides compared to undisturbed terrain in northeastern British
Columbia and demonstrated that both vegetation and environmental diversity vary with age
and size of landslides. The results suggest landslide diversity and recovery evolve over time
and may take decades to settle out, and both invasion by exotic vegetation and reactivation of
the slide can hinder ecological succession. The findings obtained from the vegetation surveys
in this study were used to present recommendations on native plant species to use for
restoration of landslides in the Peace River Region and other similar areas. This study also
provided valuable baseline information about ponds on landslides, a topic that warrants
further investigation given the ecological value of networks of small ponds. Overall, this
research has significantly enhanced understanding of landslide diversity and recovery in the
215

Peace River Region and provided timely recommendations for restoration and management
of landslides both locally and for similar glaciolacustrine sites around the world.

216

APPENDICES

217

Appendix 1 Material Origin classes
The Material Origin variable for the BEC 50m2 plots is a new variable created from
components of the soil and site description data collected during BEC sampling. It is
intended to describe the primary level of soil development present on each plot site.
The coding separates out those plots which have intermediate soil development, as well as
those plots which are ponds. Coding is as follows:
1. Mature in situ (benchmark plot in surrounding terrain)
Category 1 consists of intact material with a well-developed soil profile, situated in the
surrounding terrain. Soils have at least a B horizon, and an A horizon may also be present.
In Dystric Brunisols, the A horizon is commonly absent.
2. Mature raft
Category 2 consists of mature intact material transported from the surrounding terrain onto
the landslide through slope movement processes. The soil is well-developed and has at least
a B horizon. An A horizon may also be present.
3. Intermediate development
Category 3 consists of partially-developed landslide body soils with an immature B horizon
and no A horizon. The B horizon is thin, usually less than 10 cm in thickness.
4. Orthic Regosol
Category 4 consists of material on the landslide body which has only a C horizon exposed.
The A/B horizons have either been buried, stripped away by movement of material
downslope, or eroded away by weathering.
5. Pond
Category 5 consists of soils that are inundated under various ponds on the landslide body.
Soil development at present is mostly arrested, with soils likely comprised of just a C horizon
at the time of flooding.

218

Appendix 2 Vegetation summary tables - Mean species cover (% of total relevé area)
Beatton River Landslide Relevés BErl1, BErl2, BErl3

Mean cover SD
(%)
(+/-)
Melilotus officinalis
Yellow sweet-clover
28.833
44.312
Equisetum arvense
Common horsetail
12.167
19.776
Melilotus alba
White sweet-clover
6.000
3.500
Sonchus arvensis
Perennial sow-thistle
4.100
2.456
Artemisia sp. 2
Sage sp.
1.767
0.874
Lactuca serriola
Prickly lettuce
1.410
1.417
Rubus idaeus
Red raspberry
1.087
1.874
Aster ciliolatus
Fringed aster
0.700
1.127
Solidago canadensis
Canada goldenrod
0.650
0.589
Taraxacum officinale
Common dandelion
0.443
0.501
Fragaria virginiana
Wild strawberry
0.400
0.529
Cornus stolonifera
Red-osier dogwood
0.283
0.407
Unknown forb (white)
Forb sp.
0.167
0.289
Brachythecium sp.
Ragged moss sp.
0.167
0.289
Populus balsamifera ssp. balsamifera
Black cottonwood
0.167
0.208
Chenopodium album
Lamb's-quarters
0.133
0.115
Rosa acicularis
Prickly rose
0.133
0.153
Achillea millefolium
Yarrow
0.123
0.155
Tragopogon dubius
Goat's-beard, Yellow salsify
0.103
0.095
Hieracium triste
Wooly hawkweed
0.100
0.100
Elymus repens (aka Agropyron repens ) Quackgrass, couch grass
0.100
0.173
Aster sp.
Aster sp.
0.083
0.144
Vicia americana
American vetch
0.073
0.110
Ranunculus sp.
Buttercup sp.
0.067
0.115
Salix sp. 2
Willow sp. 2
0.067
0.115
Salix sp.
Willow sp.
0.060
0.053
Shepherdia canadensis
Soopolallie
0.043
0.051
Epilobium angustifolium
Fireweed
0.037
0.055
Lathyrus ochroleucus
Creamy peavine
0.033
0.058
Medicago sativa
Alfalfa
0.033
0.058
Elaeagnus commutata
Wolf-willow
0.033
0.058
Salix sp. 1
Willow sp. 1
0.033
0.058
Picea glauca
White spruce
0.033
0.058
Cirsium arvense
Canada thistle
0.013
0.006
Species

Common name(s)

219

Beatton River Landslide Relevés BErl1, BErl2, BErl3 -cont'd
Species

Common name(s)

Elymus glaucus
Elymus lanceolatus ssp. lanceolatus
Lily sp.
Trifolium hybridum
Bromus inermis ssp. inermis
Elymus trachycaulus ssp. trachycaulus
Ribes lacustre

Blue wildrye
Thickspike wheatgrass
Lily sp.
Alsike clover
Smooth brome
Slender wheatgrass
Black gooseberry

220

Mean cover
(%)
0.010
0.010
0.003
0.003
0.003
0.003
0.003

SD
(+/-)
0.017
0.017
0.006
0.006
0.006
0.006
0.006

Beatton River Undisturbed Relevés BEru1, BEru2d, BEru3

Species

Common name(s)

Amelanchier alnifolia
Betula papyrifera
Aralia nudicaulis
Linnaea borealis
Carex sp. 1
Viburnum edule
Rosa acicularis
Picea glauca
Carex sp. 3
Aster conspicuus
Alnus viridis ssp. sinuata
Cornus stolonifera
Leymus innovatus
Calamagrostis canadensis
Shepherdia canadensis
Symphoricarpos albus
Apocynum androsaemifolium
Salix sp. 2
Artemisia sp. 1
Populus tremuloides
Rubus pubescens
Elaeagnus commutata
Tragopogon dubius
Hesperostipa curtiseta
Pyrola asarifolia
Prunus virginiana
Lonicera dioica
Populus balsamifera ssp. balsamifera
Spiraea betulifolia
Lathyrus ochroleucus
Cornus canadensis
Platydictya jungermannioides
Vicia americana

Saskatoon
Paper birch
Wild sarsaparilla
Twinflower
Sedge sp. 1
Highbush-cranberry
Prickly rose
White spruce
Sedge sp. 3
Showy aster
Sitka alder
Red-osier dogwood
Fuzzy-spiked wildrye
Bluejoint
Soopolallie
Common snowberry
Spreading dogbane
Willow sp. 2
Sage sp. 1
Trembling aspen
Trailing raspberry
Wolf-willow
Goat's-beard, yellow salsify
Needle-and-thread grass
Pink wintergreen
Choke cherry
Red honeysuckle
Black cottonwood
Birch-leaved spirea
Creamy peavine
Bunchberry
False willow moss, Spruce's leskea
American vetch

221

Mean cover SD
(%)
(+/-)
10.846
16.598
10.667
18.475
7.824
2.299
6.832
6.135
5.667
9.815
5.550
4.877
4.925
0.393
4.727
6.724
4.000
6.928
3.852
3.372
3.500
6.062
2.683
2.675
2.625
2.770
2.614
3.105
2.407
3.049
2.003
3.038
1.833
3.175
1.510
2.615
1.417
2.454
1.297
1.101
1.268
1.874
1.083
1.876
1.000
1.732
0.667
1.155
0.500
0.621
0.422
0.298
0.396
0.487
0.396
0.351
0.333
0.577
0.329
0.333
0.277
0.392
0.233
0.400
0.233
0.354

Beatton River Undisturbed Relevés BEru1, BEru2d, BEru3 - cont'd
Mean cover SD
(%)
(+/-)
Galium boreale
Northern bedstraw
0.213
0.191
Mitella nuda
Common mitrewort
0.169
0.291
Maianthemum canadense
Wild lily-of-the-valley
0.168
0.148
"Silver grass"
Grass sp.
0.167
0.289
Fragaria vesca
Wood strawberry
0.165
0.188
Salix sp. 3
Willow sp. 3
0.162
0.193
Disporum trachycarpum
Rough-fruited fairybells
0.149
0.157
Ribes lacustre
Black gooseberry
0.148
0.129
Mertensia paniculata
Tall bluebell
0.143
0.128
Equisetum arvense
Common horsetail
0.127
0.217
Ribes triste
Red swamp currant
0.125
0.110
Actaea rubra
Baneberry
0.100
0.163
Orthilia secunda
One-sided wintergreen
0.095
0.095
Fragaria virginiana
Wild strawberry
0.094
0.126
Epilobium angustifolium
Fireweed
0.093
0.089
Comandra umbellata
Bastard toadflax
0.093
0.162
Aster ciliolatus
Fringed aster
0.083
0.018
Achnatherum nelsonii
Columbia needlegrass
0.083
0.144
Hesperostipa spartea ( aka Stipa spartea) Porcupinegrass
0.083
0.144
Peltigera sp.
Pelt lichen sp.
0.080
0.122
Rubus idaeus
Red raspberry
0.079
0.097
Lonicera involucrata
Black twinberry
0.073
0.082
Osmorhiza chilensis
Mountain sweet-cicely
0.068
0.117
Hylocomium splendens
Step moss
0.066
0.095
Plagiomnium cuspidatum
Baby tooth moss, woodsy thyme-moss
0.056
0.048
Maianthemum canadense
Violet sp.
0.055
0.048
Ribes oxyacanthoides
Northern gooseberry
0.048
0.048
Eurynchiastrum pulchellum
Elegant feather-moss
0.040
0.037
Galium triflorum
Sweet-scented bedstraw
0.040
0.038
Achillea millefolium
Yarrow
0.039
0.053
Geocaulon lividum
False toadflax, northern comandra
0.033
0.058
Sonchus sp.
Sow-thistle sp.
0.033
0.058
Androsace septentrionalis
Pygmyflower rockjasmine
0.033
0.058
Species

Common name(s)

222

Beatton River Undisturbed Relevés BEru1, BEru2d, BEru3 - cont'd
Species

Common name(s)

Ribes sp.
Smilacina stellata
Taraxacum officinale
Erigeron glabellus
Koelaria macrantha
Pleurozium schreberi
Pohlia nutans
Pyrola sp.
Erigeron acris
Agropyron sp.
Syntrichia ruralis
Solidago canadensis
Salix sp. 1
Trifolium pratense
Petasites frigidus
Dicranum polysetum
Gymnocarpium dryopteris
Delphinium sp.
Dryas drummondii
Maianthemum stellatum
Rhinanthus minor
Glyceria sp.
Platanthera dilitata
Cladonia sp.
Juniperus communis

Gooseberry/currant sp.
Star-flowered false Solomon's seal
Common dandelion
Smooth fleabane
Junegrass
Red-stemmed feathermoss
Nodding thread-moss
Wintergreen sp.
Bitter fleabane
Slender wheatgrass
Twisted moss
Canada goldenrod
Willow sp. 1
Clover sp.
Sweet coltsfoot
Broom moss, wavyleaf moss
Oak fern
Larkspur sp.
Yellow mountain avens
Starry false lily of the valley
Yellow rattle
Manna grass sp.
White northern bog-orchid
Pixie cup sp.
Common juniper

223

Mean cover SD
(%)
(+/-)
0.033
0.058
0.023
0.030
0.018
0.018
0.017
0.029
0.017
0.029
0.014
0.017
0.013
0.022
0.012
0.014
0.010
0.017
0.010
0.017
0.008
0.013
0.008
0.014
0.008
0.013
0.007
0.012
0.004
0.007
0.003
0.004
0.003
0.004
0.003
0.006
0.003
0.006
0.003
0.006
0.003
0.006
0.003
0.006
0.002
0.003
0.002
0.003
0.002
0.003

Cecil Lake Landslide Relevés CErl1, CErl2, CErl3

Species

Common name(s)

Equisetum arvense
Common horsetail
Alnus viridis ssp. sinuata
Sitka alder
Petasites frigidus
Sweet coltsfoot
Salix sp. 1
Willow sp. 1
Populus balsamifera ssp. balsamifera Black cottonwood
Shepherdia canadensis
Soopolallie
Picea glauca
White spruce
Populus tremuloides
Trembling aspen
Rosa acicularis
Prickly rose
Salix sp. 2
Willow sp. 2
Trifolium hybridum
Alsike clover
Salix sp. 3
Willow sp. 3
Viburnum edule
Highbush-cranberry
Cornus stolonifera
Red-osier dogwood
Calamagrostis canadensis
Bluejoint
Taraxacum officinale
Common dandelion
Typha sp.
Cattail sp.
Salix sp. 6
Willow sp. 6
Eurynchiastrum pulchellum
Elegant feather-moss
Rubus idaeus
Red raspberry
Aster ciliolatus
Fringed aster
Epilobium angustifolium
Fireweed
Betula papyrifera
Paper birch
Cirsium arvense
Canada thistle
Linnaea borealis
Twinflower
Fragaria virginiana
Wild strawberry
Sonchus arvensis
Perennial sow-thistle
Rubus pubescens
Trailing raspberry
Vicia americana
American vetch
Salix sp. 4
Willow sp. 4
Lathyrus ochroleucus
Creamy peavine
Aralia nudicaulis
Wild sarsaparilla
Pyrola asarifolia
Pink wintergreen

224

Mean cover SD
(%)
(+/-)
33.500
12.173
24.092
20.687
6.733
11.490
5.630
4.352
4.767
6.280
3.128
0.327
2.679
2.065
2.642
2.455
2.412
0.390
1.958
2.895
1.664
2.516
1.518
1.885
1.492
1.784
1.325
0.282
1.258
1.906
0.833
0.441
0.675
0.622
0.667
1.155
0.567
0.787
0.523
0.285
0.518
0.729
0.415
0.388
0.350
0.563
0.348
0.410
0.332
0.282
0.329
0.170
0.298
0.495
0.265
0.190
0.253
0.137
0.225
0.390
0.203
0.086
0.197
0.341
0.194
0.319

Cecil Lake Landslide Relevés CErl1, CErl2, CErl3 - cont'd

Species
Aster conspicuus
Brachythecium sp. 2
Salix sp. 5
Solidago canadensis
Amelanchier alnifolia
Mertensia paniculata
Ribes oxyacanthoides
Cornus canadensis
Phleum pratense
Elymus trachycaulus
Brachythecium sp. 1
Salix exigua
Hordeum jubatum
Symphoricarpos albus
Delphinium glaucum
Lonicera involucrata
Goodyera repens
"Hairy grass"
Brachythecium sp. 3
Pohlia nutans
Cinna latifolia
Galium triflorum
Fragaria vesca
Glyceria sp.
Achillea sibirica
Melilotus officinalis
Bromus inermis ssp. inermis
"Tallwheat grass" sp.
Achillea millefolium
Ribes sp.
Geum macrophyllum
Galium boreale
Matricaria perforata

Mean cover
(%)
Showy aster
0.170
Ragged moss sp.
0.167
Willow sp. 5
0.167
Canada goldenrod
0.161
Saskatoon
0.117
Tall bluebell
0.108
Northern gooseberry
0.092
Bunchberry
0.091
Timothy
0.090
Slender wheat grass
0.084
Moss sp.
0.067
Sandbar willow
0.067
Foxtail barley
0.064
Common snowberry
0.060
Tall larkspur
0.057
Black twinberry
0.057
Dwarf rattlesnake-plantain
0.050
"Hairy" grass sp.
0.049
Moss sp.
0.039
Nodding thread-moss
0.037
Nodding wood-reed
0.033
Sweet-scented bedstraw
0.026
Wood strawberry
0.024
Manna grass sp. 1
0.021
Siberian yarrow
0.020
Yellow sweet-clover
0.019
Smooth brome
0.019
"Tallwheat" grass sp.
0.018
Common yarrow
0.016
Gooseberry/currant sp.
0.016
Large-leaved avens
0.015
Northern bedstraw
0.008
Scentless chamomile
0.008
Common name(s)

225

SD
(+/-)
0.255
0.289
0.289
0.177
0.126
0.091
0.103
0.077
0.087
0.146
0.115
0.115
0.051
0.100
0.098
0.068
0.087
0.085
0.068
0.064
0.058
0.017
0.042
0.034
0.026
0.033
0.033
0.032
0.018
0.027
0.022
0.010
0.014

Cecil Lake Landslide Relevés CErl1, CErl2, CErl3 - cont'd

Species
Ribes triste
Dicranum sp.
Funaria sp.
Epilobium ciliatum
Mitella nuda
Spiraea sp.
Agrostis stolonifera
Cladonia sp.
Nephroma sp.
Lonicera dioica
Bromus ciliatus
Poa nemoralis ssp. interior
Elymus lanceolatus ssp. lanceolatus
Peltigera sp.
Actaea rubra
Hylocomium splendens
Achillea sp.
Chenopodium sp.
Crepis tectorum
Leucantheum sp.
Sanicula marilandica
Koelaria macrantha
Rush sp.
Ribes sp. 2
Ribes lacustre
Rumex crispus
Platanthera dilitata
Viola sp.
Poa pratensis
Ribes laxiflorum
Marchantia sp.
"Bottlebrush moss" sp.
Pleurozium schreberi

Mean cover
(%)
Red swamp currant
0.008
Broom moss sp.
0.007
Rope moss sp.
0.007
Purple-leaved willowherb
0.007
Common mitrewort
0.007
Hardhack sp.
0.007
Redtop
0.007
Pixie cup sp.
0.007
Kidney lichen sp.
0.007
Red honeysuckle
0.006
Fringed brome
0.005
Inland blue grass
0.005
Thickspike wheatgrass
0.005
Pelt lichen sp.
0.005
Baneberry
0.004
Step moss
0.003
Yarrow sp.
0.003
Goosefoot sp.
0.003
Annual hawksbeard
0.003
Daisy sp.
0.003
Maryland black snakeroot
0.003
Junegrass
0.003
Rush sp.
0.003
Gooseberry/currant sp. 2
0.003
Black gooseberry
0.003
Curly dock
0.002
White northern bog-orchid
0.002
Violet sp.
0.002
Kentucky bluegrass
0.002
Trailing black currant
0.002
Liverwort sp.
0.001
"Bottlebrush" moss sp.
0.001
Red-stemmed feathermoss
0.001
Common name(s)

226

SD
(+/-)
0.014
0.012
0.012
0.012
0.012
0.012
0.012
0.012
0.012
0.010
0.009
0.009
0.009
0.009
0.007
0.006
0.006
0.006
0.006
0.006
0.004
0.006
0.004
0.006
0.004
0.003
0.003
0.001
0.003
0.003
0.001
0.001
0.001

Cecil Lake Landslide Relevés CErl1, CErl2, CErl3 - cont'd
Species

Common name(s)

Hieracium triste
Senecio eremophilus
Sonchus arvensis
Gentianella amarella
"Bigwheat grass" sp.
Glyceria sp. 2
Beckmannia syzigachne

Wooly hawkweed
Cutleaf groundsel
Perennial sow thistle
Autumn gentian
"Bigwheat" grass sp.
Manna grass sp. 2
American sloughgrass

227

Mean cover
(%)
0.001
0.001
0.001
0.001
0.001
0.001
0.001

SD
(+/-)
0.001
0.001
0.001
0.001
0.001
0.001
0.001

Cecil Lake Undisturbed Relevés CEru1, CEru2a, CEru3

Species

Common name(s)

Picea glauca
Populus tremuloides
Rosa acicularis
Viburnum edule
Aralia nudicaulis
Hylocomium splendens
Shepherdia canadensis
Alnus viridis ssp. sinuata
Betula papyrifera
Linnaea borealis
"Fuzzy grass" sp.
Cornus stolonifera
Pleurozium schreberi
Aster conspicuus
Cornus canadensis
Amelanchier alnifolia
Rubus pubescens
Ribes triste
Symphoricarpos albus
Mertensia paniculata
Ptilium crista-castrensis
Ribes lacustre
Lonicera dioica
"Thin grass" sp.
Calamagrostis canadensis
Salix sp.
Lathyrus ochroleucus
Clematis sp.
Lonicera involucrata
Rhytidiadelphus triquetrus
Mitella nuda
Eurynchiastrum pulcellum
Disporum trachycarpum

White spruce
Trembling aspen
Prickly rose
Highbush-cranberry
Wild sarsaparilla
Step moss
Soopolallie
Sitka alder
Paper birch
Twinflower
"Fuzzy" grass sp.
Red-osier dogwood
Red-stemmed feathermoss
Showy aster
Bunchberry
Saskatoon
Trailing raspberry
Red swamp currant
Common snowberry
Tall bluebell
Knight's plume
Black gooseberry
Red honeysuckle
"Thin" grass sp.
Bluejoint
Willow sp.
Creamy peavine
Clematis sp.
Black twinberry
Electrified cat's-tail moss
Common mitrewort
Elegant beaked moss
Rough-fruited fairybells

228

Mean cover SD
(%)
(+/-)
22.893
14.687
9.107
12.483
6.908
3.425
6.762
3.772
6.460
1.935
6.210
5.743
4.388
4.020
4.015
2.953
3.624
5.017
3.137
3.102
2.253
3.894
1.922
1.738
1.118
1.779
1.118
0.986
1.058
0.820
0.971
0.806
0.782
0.500
0.445
0.404
0.426
0.224
0.349
0.327
0.323
0.543
0.297
0.285
0.279
0.106
0.253
0.439
0.240
0.235
0.190
0.329
0.146
0.156
0.133
0.231
0.115
0.133
0.110
0.191
0.108
0.166
0.087
0.150
0.086
0.012

Cecil Lake Undisturbed Relevés CEru1, CEru2a, CEru3 - cont'd

Mean cover
(%)
Orthilia secunda
One-sided wintergreen
0.081
Geocaulon lividum
Bastard toad-flax
0.060
Aster ciliolatus
Fringed aster
0.053
Galium boreale
Northern bedstraw
0.053
Pyrola chlorantha
Green wintergreen
0.047
Peltigera polydactylon
Pioneer pelt
0.047
Pyrola asarifolia
Pink wintergreen
0.038
Ribes oxyacanthoides
Northern gooseberry
0.035
Fragaria virginiana
Wild strawberry
0.033
Rubus idaeus
Red raspberry
0.029
Brachythecium sp.
Ragged moss sp.
0.027
Fragaria vesca
Wood strawberry
0.027
Moss sp.
Moss sp.
0.020
Fern sp.
Fern sp.
0.020
Epilobium angustifolium
Fireweed
0.020
Actaea rubra
Baneberry
0.018
Osmorhiza chilensis
Mountain sweet-cicely
0.018
Dicranum polysetum
Wavy leaf moss, broom moss
0.014
Galium triflorum
Sweet-scented bedstraw
0.012
Dicranum scoparium
Broom forkmoss
0.010
Plagiomnium cuspidatum
Baby tooth moss, woodsy thyme moss
0.007
Viola sp.
Violet sp.
0.007
Vicia americana
American vetch
0.005
Delphinium glaucum
Tall larkspur
0.004
Equisetum arvense
Common horsetail
0.003
Goodyera oblongifolia
Rattlesnake-plantain
0.003
Listera sp. (aka Neottia sp.) Twayblade sp.
0.003
Maianthemum canadense
Wild lily-of-the-valley
0.003
Taraxacum officinale
Common dandelion
0.003
Cladonia sp.
Pixie cup sp.
0.003
Prunus virginiana
Choke cherry
0.003
Gymnocarpium dryopteris
Oak fern
0.002
Species

Common name(s)

229

SD
(+/-)
0.059
0.060
0.084
0.051
0.072
0.045
0.013
0.061
0.058
0.026
0.046
0.025
0.035
0.035
0.035
0.011
0.028
0.025
0.013
0.017
0.012
0.012
0.005
0.007
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.003

Hasler Flats Landslide Relevés HArl1b, HArl2b, HArl3a
Species

Common name(s)

Equisetum arvense
Populus tremuloides
Rosa acicularis
Alnus viridis ssp. sinuata
Rubus idaeus
Cornus stolonifera
Symphoricarpos albus
Pohlia nutans
Viburnum edule
Lonicera dioica
Smilacina racemosa
Amelanchier alnifolia
Aralia nudicaulis
Prunus virginiana
Calamagrostis canadensis
Bromus inermis
Aster conspicuus
Lonicera involucrata
Elymus glaucus ssp. glaucus
Ribes oxyacanthoides
Salix sp. 2
Rubus pubescens
Ribes triste
Populus balsamifera ssp. balsamifera
Mertensia paniculata
Salix sp. 1
Petasites frigidus
Vicia americana
Streptopus amplexifolius
Viola sp.
Marchantia polymorpha
Aster ciliolatus
Lathyrus ochroleucus

Common horsetail
Trembling aspen
Prickly rose
Sitka alder
Red raspberry
Red-osier dogwood
Common snowberry
Nodding thread-moss
Highbush-cranberry
Red honeysuckle
False Solomon's-seal
Saskatoon
Wild sarsaparilla
Choke cherry
Bluejoint
Smooth brome
Showy aster
Black twinberry
Blue wildrye
Northern gooseberry
Willow sp. 2
Trailing raspberry
Red swamp currant
Black cottonwood
Tall bluebell
Willow sp. 1
Sweet coltsfoot
American vetch
Clasping twistedstalk
Violet sp.
Common liverwort
Fringed aster
Creamy peavine

230

Mean cover
(%)
16.417
10.758
10.375
6.308
5.595
4.900
4.145
2.423
2.423
2.348
1.623
1.528
1.403
1.320
1.088
1.051
0.925
0.716
0.696
0.461
0.434
0.348
0.335
0.300
0.281
0.266
0.248
0.247
0.232
0.199
0.198
0.176
0.176

SD
(+/-)
1.041
5.329
2.211
4.242
3.055
3.776
2.427
0.505
1.273
2.706
2.537
1.671
2.147
0.437
0.934
1.818
0.839
0.584
0.654
0.491
0.475
0.378
0.273
0.173
0.110
0.142
0.161
0.178
0.401
0.313
0.208
0.172
0.204

Hasler Flats Landslide Relevés HArl1b, HArl2b, HArl3a - cont'd
Species

Common name(s)

Galium boreale
Northern bedstraw
Betula papyrifera
Paper birch
Heracleum lanatum
Cow-parsnip
Actaea rubra
Baneberry
Smilacina stellata
Star-flowered false Solomon's-seal
Taraxacum officinale
Common dandelion
Fragaria vesca
Wood strawberry
Epilobium angustifolium
Fireweed
Cornus canadensis
Bunchberry
Mitella nuda
Common mitrewort
Galium triflorum
Sweet-scented bedstraw
Picea glauca
White spruce
Dicranum polysetum
Wavy leaf moss, broom moss
Pyrola asarifolia
Pink wintergreen
Thalictrum occidentale
Western meadowrue
Fragaria virginiana
Wild strawberry
Cirsium arvense
Canada thistle
Spiraea betulifolia
Birch-leaved spirea
Maianthemum canadense
Wild lily-of-the-valley
Poa palustris
Fowl blue grass
Sonchus arvensis
Perennial sow-thistle
Epilobium ciliatum
Purple-leaved willowherb
Linnaea borealis
Twinflower
Veratrum viride
Indian hellebore
"Oat grass" sp.
"Oat grass" grass sp.
Rubus parviflorus
Thimbleberry
Bromus ciliatus
Fringed brome
Typha sp.
Cattail sp.
Leymus innovatus
Fuzzy-spiked wildrye
Veronica beccabunga
European speedwell, brooklime
Elymus trachycaulus ssp. trachycaulus Slender wheatgrass
Carex sp.
Sedge sp.
Carex rostrata
Beaked sedge

231

Mean cover
(%)
0.163
0.103
0.098
0.090
0.088
0.088
0.084
0.082
0.074
0.068
0.065
0.063
0.062
0.061
0.054
0.047
0.046
0.040
0.039
0.038
0.030
0.029
0.027
0.023
0.019
0.019
0.016
0.015
0.012
0.010
0.010
0.010
0.008

SD
(+/-)
0.142
0.003
0.100
0.089
0.153
0.052
0.135
0.085
0.004
0.089
0.061
0.097
0.107
0.093
0.019
0.081
0.058
0.034
0.046
0.026
0.014
0.020
0.031
0.040
0.033
0.019
0.027
0.020
0.020
0.017
0.011
0.017
0.009

Hasler Flats Landslide Relevés HArl1b, HArl2b, HArl3a - cont'd
Species

Common name(s)

Cirsium vulgare
Festuca sp.
Achillea millefolium
Trisetum cernuum
"Limp grass" sp.
Syntrichia ruralis
Plagiomnium cuspidatum
Cyclamen hederifolium
Hieracium sabaudum
Orthilia secunda
"Frothy grass" sp.
Shepherdia canadensis
Leptobryum pyriforme
Fern sp.
Equisetum variegatum
Geum macrophyllum
Chenopodium capitatum
Boechera stricta
Plantago major
"Pondweed" sp.
Trifolium hybridum
Phleum pratense
Ribes hudsonianum
Ribes laxiflorum

Bull thistle
Fescue sp.
Yarrow
Nodding trisetum
"Limp grass" grass sp.
Twisted moss
Baby tooth moss, woodsy thyme-moss
Hardy cyclamen, ivy leaved cyclamen
European hawkweed, Savoy hawkweed
One-sided wintergreen
"Frothy grass" grass sp.
Soopolallie
Golden thread-moss
Fern sp.
Variegated scouring-rush
Large-leaved avens
Strawberry-blite
Canada rockcress
Broadleaf plantain
Pondweed sp.
Alsike clover
Timothy
Northern black currant
Trailing black currant

232

Mean cover
(%)
0.007
0.007
0.005
0.005
0.005
0.004
0.004
0.003
0.003
0.003
0.003
0.003
0.002
0.002
0.002
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001

SD
(+/-)
0.012
0.012
0.005
0.009
0.009
0.007
0.007
0.006
0.006
0.004
0.004
0.004
0.003
0.003
0.003
0.003
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001

Hasler Flats Undisturbed Relevés HAru1, HAru2a, HAru3a
Species

Common name(s)

Equisetum arvense
Populus tremuloides
Alnus viridis ssp. sinuata
Cornus stolonifera
Viburnum edule
Aralia nudicaulis
Symphoricarpos albus
Rosa acicularis
Populus balsamifera ssp. balsamifera
Heracleum lanatum
Aster conspicuus
Amelanchier alnifolia
Lonicera dioica
Prunus virginiana
Mitella nuda
Rubus idaeus
Salix sp. 2
Ribes oxyacanthoides
Smilacina racemosa
Actaea rubra
Rubus pubescens
Thalictrum occidentale
Picea glauca
Smilacina stellata
Ribes triste
Lonicera involucrata
Viola sp.
Lathyrus ochroleucus
Maianthemum canadense
Streptopus amplexifolius
Vicia americana
Petasites frigidus
Angelica lucida

Common horsetail
Trembling aspen
Sitka alder
Red-osier dogwood
Highbush-cranberry
Wild sarsaparilla
Common snowberry
Prickly rose
Black cottonwood
Cow-parsnip
Showy aster
Saskatoon
Red honeysuckle
Choke cherry
Common mitrewort
Red raspberry
Willow sp. 2
Northern gooseberry
False Solomon's-seal
Baneberry
Trailing raspberry
Western meadowrue
White spruce
Star-flowered false Solomon's-seal
Red swamp currant
Black twinberry
Violet sp.
Creamy peavine
Wild lily-of-the-valley
Clasping twistedstalk
American vetch
Sweet coltsfoot
Seacoast angelica, sea-watch

233

Mean cover SD
(%)
(+/-)
15.625
6.945
13.580
5.074
9.863
17.084
7.358
4.376
6.983
5.925
6.856
4.865
5.378
5.090
5.283
1.617
4.335
6.321
3.844
5.059
2.556
2.311
2.503
1.616
1.976
2.900
1.609
1.262
1.397
1.984
1.236
1.720
1.026
0.876
0.904
1.521
0.653
0.481
0.580
0.544
0.551
0.240
0.516
0.532
0.510
0.515
0.488
0.420
0.487
0.318
0.438
0.193
0.415
0.595
0.328
0.285
0.327
0.375
0.276
0.386
0.235
0.147
0.231
0.138
0.199
0.179

Hasler Flats Undisturbed Relevés HAru1, HAru2a, HAru3a - cont'd

Species
Salix sp. 1
Calamagrostis canadensis
Pyrola asarifolia
Aster foliaceus
Ribes hudsonianum
Mertensia paniculata
Cirsium arvense
Aster ciliolatus
Dicranum polysetum
Galium boreale
Shepherdia canadensis
Galium triflorum
"Mystery grass 2" sp.
Disporum trachycarpum
Plagiomnium cuspidatum
Fragaria vesca
Plagiomnium ellipticum
Osmorhiza chilensis
Cornus canadensis
Salix sp. 3
Acer glabrum
Orthilia secunda
Bromus inermis ssp. inermis
Fragaria virginiana
Taraxacum officinale
"Hairy grass"
Epilobium ciliatum
Veronica beccabunga
Carex disperma
"Mystery grass" sp. 1
Veratrum viride
Linnaea borealis
Bromus ciliatus

Mean cover
(%)
Willow sp. 1
0.195
Bluejoint
0.185
Pink wintergreen
0.163
Leafy-bracted aster
0.131
Northern black currant
0.130
Tall bluebell
0.124
Creeping thistle
0.123
Fringed aster
0.110
Wavy leaf moss, broom moss
0.105
Northern bedstraw
0.085
Soopolallie
0.083
Sweet-scented bedstraw
0.077
Mystery grass sp. 2
0.077
Rough-fruited fairybells
0.074
Baby tooth moss, woodsy thyme-moss
0.068
Wood strawberry
0.056
Marsh thyme-moss
0.041
Mountain sweet-cicely
0.035
Bunchberry
0.034
Willow sp. 3
0.033
Douglas maple
0.033
One-sided wintergreen
0.032
Smooth brome
0.028
Wild strawberry
0.027
Common dandelion
0.026
"Hairy grass" grass sp.
0.024
Purple-leaved willowherb
0.019
European speedwell, brooklime
0.018
Soft-leaf sedge, two-seed sedge
0.018
"Mystery grass" grass sp. 1
0.017
Indian hellebore
0.014
Twinflower
0.011
Fringed brome
0.008
Common name(s)

234

SD
(+/-)
0.287
0.215
0.249
0.218
0.225
0.027
0.214
0.132
0.173
0.071
0.144
0.028
0.133
0.128
0.118
0.097
0.056
0.044
0.049
0.058
0.058
0.051
0.018
0.031
0.030
0.042
0.033
0.030
0.016
0.027
0.025
0.019
0.014

Hasler Flats Undisturbed Relevés HAru1, HAru2a, HAru3a - cont'd

Species

Common name(s)

Poa pratensis
Equisetum pratense
Spiraea betulifolia
Leymus innovatus
Elymus lanceolatus ssp. lanceolatus
Epilobium angustifolium
Geum rivale
Phleum pratense
"Soft grass" sp
Grass sp.
Carex sp.
Rubus parviflorus
Botrychium virginianum
Cyclamen hederifolium
Glyceria sp.
Achillea millefolium
Monotropa uniflora
Ribes lacustre

Kentucky bluegrass
Meadow horsetail, shade horsetail
Birch-leaved spirea
Fuzzy-spiked wildrye
Thickspike wheatgrass
Fireweed
Water avens, purple avens
Timothy
"Soft grass" grass sp.
Unknown grass sp.
Sedge sp.
Thimbleberry
Rattlesnake fern
Hardy cyclamen, ivy leaved cyclamen
Manna grass sp.
Yarrow
Ghost pipe, Indian-pipe
Black gooseberry

235

Mean cover
(%)
0.008
0.007
0.005
0.005
0.005
0.003
0.003
0.003
0.003
0.003
0.003
0.003
0.002
0.002
0.002
0.001
0.001
0.001

SD
(+/-)
0.014
0.006
0.009
0.009
0.004
0.003
0.006
0.004
0.004
0.006
0.006
0.004
0.003
0.003
0.003
0.001
0.001
0.001

Appendix 3 Vegetation cover by growth form
Beatton River Landslide Relevés - Total mean vegetation cover = 59.68%

Species
Trees
Populus balsamifera ssp. balsamifera
Picea glauca

Mean cover (%)

Total
Proportion of all cover

0.167
0.033
0.200
0.335

Total
Proportion of all cover

1.087
0.283
0.133
0.067
0.060
0.043
0.033
0.033
0.003
1.742
2.919

Shrubs
Rubus idaeus
Cornus stolonifera
Rosa acicularis
Salix sp. 2
Salix sp.
Shepherdia canadensis
Elaeagnus commutata
Salix sp. 1
Ribes lacustre

Forbs
Melilotus officinalis
Melilotus alba
Sonchus arvensis
Artemisia sp. 2
Lactuca serriola
Aster ciliolatus
Solidago canadensis
Taraxacum officinale
Fragaria virginiana
Unknown forb (white)
Chenopodium album
Achillea millefolium
Tragopogon dubius
Hieracium triste
Aster sp.
Vicia americana
Ranunculus sp.

28.833
6.000
4.100
1.767
1.410
0.700
0.650
0.443
0.400
0.167
0.133
0.123
0.103
0.100
0.083
0.073
0.067

236

Beatton River Landslide Relevés - cont'd

Forbs -cont'd
Epilobium angustifolium
Lathyrus ochroleucus
Medicago sativa
Cirsium arvense
Lily sp.
Trifolium hybridum
Total
Proportion of all cover
Graminoids
Elymus repens (aka Agropyron repens )
Elymus glaucus
Elymus lanceolatus ssp. lanceolatus
Bromus inermis ssp. inermis
Elymus trachycaulus ssp. trachycaulus

0.037
0.033
0.033
0.013
0.003
0.003
45.274
75.861

Total
Proportion of all cover

0.100
0.010
0.010
0.003
0.003
0.126
0.211

Total
Proportion of all cover

12.167
12.167
20.387

Total
Proportion of all cover

0.167
0.167
0.280

Ferns & Fern allies
Equisetum arvense

Bryophytes
Brachythecium sp.

237

Beatton River Undisturbed Relevés - Total mean vegetation cover = 97.348%

Species
Trees
Betula papyrifera
Picea glauca
Populus tremuloides
Populus balsamifera ssp. balsamifera

Mean cover (%)

Total
Proportion of all cover

10.667
4.727
1.297
0.396
17.087
17.552

Total
Proportion of all cover

10.846
5.550
4.925
3.500
2.683
2.407
2.003
1.510
1.268
1.083
0.422
0.396
0.162
0.148
0.125
0.079
0.073
0.048
0.033
0.008
0.002
37.271
38.286

Shrubs
Amelanchier alnifolia
Viburnum edule
Rosa acicularis
Alnus viridis ssp. sinuata
Cornus stolonifera
Shepherdia canadensis
Symphoricarpos albus
Salix sp. 2
Rubus pubescens
Elaeagnus commutata
Prunus virginiana
Lonicera dioica
Salix sp.
Ribes lacustre
Ribes triste
Rubus idaeus
Lonicera involucrata
Ribes oxyacanthoides
Ribes sp.
Salix sp. 1
Juniperus communis

Forbs
Aralia nudicaulis
Linnaea borealis
Aster conspicuus

7.824
6.832
3.852

238

Beatton River Undisturbed Relevés - cont'd

Species
Forbs - cont'd
Apocynum androsaemifolium
Artemisia sp. 1
Tragopogon dubius
Pyrola asarifolia
Spiraea betulifolia
Lathyrus ochroleucus
Cornus canadensis
Vicia americana
Galium boreale
Mitella nuda
Maianthemum canadense
Fragaria vesca
Disporum trachycarpum (aka Prosartes trachycarpa )
Mertensia paniculata
Actaea rubra
Orthilia secunda
Fragaria virginiana
Epilobium angustifolium
Comandra umbellata
Aster ciliolatus
Osmorhiza chilensis
Maianthemum canadense
Galium triflorum
Achillea millefolium
Geocaulon lividum
Sonchus sp.
Smilacina stellata
Taraxacum officinale
Erigeron glabellus
Pyrola sp.
Erigeron acris
Solidago canadensis
Trifolium pratense
Petasites frigidus
Delphinium sp.

239

Mean cover (%)
1.833
1.417
1.000
0.500
0.333
0.329
0.277
0.233
0.213
0.169
0.168
0.165
0.149
0.143
0.100
0.095
0.094
0.093
0.093
0.083
0.068
0.055
0.040
0.039
0.033
0.033
0.023
0.018
0.017
0.012
0.010
0.008
0.007
0.004
0.003

Beatton River Undisturbed Relevés - cont'd

Species
Forbs - cont'd
Dryas drummondii
Maianthemum stellatum
Rhinanthus minor
Platanthera dilitata

Mean cover (%)

Total
Proportion of all cover
Graminoids
Carex sp. 1
Carex sp. 3
Leymus innovatus
Calamagrostis canadensis
Hesperostipa curtiseta
"Silver grass"
Achnatherum nelsonii
Hesperostipa spartea (aka Stipa spartea )
Androsace septentrionalis
Koelaria macrantha
Elymus trachycaulus ssp. trachycaulus
Glyceria sp.

0.003
0.003
0.003
0.002
26.376
27.095

Total
Proportion of all cover

5.667
4.000
2.625
2.614
0.667
0.167
0.083
0.083
0.033
0.017
0.010
0.003
15.969
16.404

Total
Proportion of all cover

0.127
0.003
0.130
0.134

Ferns & Fern allies
Equisetum arvense
Gymnocarpium dryopteris

Bryophytes
Platydictya jungermannioides
Hylocomium splendens
Plagiomnium cuspidatum
Eurynchiastrum pulchellum
Pleurozium schreberi

0.233
0.066
0.056
0.040
0.014

240

Beatton River Undisturbed Relevés - cont'd
Species
Bryophytes - cont'd
Pohlia nutans
Syntrichia ruralis
Dicranum polysetum

Mean cover (%)

Total
Proportion of all cover

0.013
0.008
0.003
0.433
0.445

Total
Proportion of all cover

0.080
0.002
0.082
0.084

Lichens
Peltigera sp.
Cladonia sp.

241

Cecil Lake Landslide Relevés - Total mean vegetation cover = 104.160%

Species
Trees
Populus balsamifera ssp. balsamifera
Picea glauca
Populus tremuloides
Betula papyrifera

Mean cover (%)

Total
Proportion of all cover

4.767
2.679
2.642
0.350
10.438
10.021

Total
Proportion of all cover

24.092
5.630
3.128
2.412
1.958
1.518
1.492
1.325
0.667
0.523
0.265
0.225
0.167
0.117
0.092
0.067
0.060
0.057
0.016
0.008
0.006
0.003
0.003
0.002
43.833
42.082

Shrubs
Alnus viridis ssp. sinuata
Salix sp. 1
Shepherdia canadensis
Rosa acicularis
Salix sp. 2
Salix sp. 3
Viburnum edule
Cornus stolonifera
Salix sp. 6
Rubus idaeus
Rubus pubescens
Salix sp. 4
Salix sp. 5
Amelanchier alnifolia
Ribes oxyacanthoides
Salix exigua
Symphoricarpos albus
Lonicera involucrata
Ribes sp.
Ribes triste
Lonicera dioica
Ribes sp. 2
Ribes lacustre
Ribes laxiflorum

242

Cecil Lake Landslide Relevés - cont'd

Species
Forbs
Petasites frigidus
Trifolium hybridum
Taraxacum officinale
Typha sp.
Aster ciliolatus
Epilobium angustifolium
Cirsium arvense
Linnaea borealis
Fragaria virginiana
Sonchus arvensis
Vicia americana
Lathyrus ochroleucus
Aralia nudicaulis
Pyrola asarifolia
Aster conspicuus
Solidago canadensis
Mertensia paniculata
Cornus canadensis
Delphinium glaucum
Goodyera repens
Galium triflorum
Fragaria vesca
Achillea sibirica
Melilotus officinalis
Achillea millefolium
Geum macrophyllum
Galium boreale
Matricaria perforata (aka Tripleurospermum inodorum )
Epilobium ciliatum
Mitella nuda
Spiraea sp.
Actaea rubra
Achillea sp.
Chenopodium sp.

243

Mean cover (%)
6.733
1.664
0.833
0.675
0.518
0.415
0.348
0.332
0.329
0.298
0.253
0.203
0.197
0.194
0.170
0.161
0.108
0.091
0.057
0.050
0.026
0.024
0.020
0.019
0.016
0.015
0.008
0.008
0.007
0.007
0.007
0.004
0.003
0.003

Cecil Lake Landslide Relevés - cont'd

Species
Forbs - cont'd
Crepis tectorum
Leucantheum sp.
Sanicula marilandica
Rumex crispus
Platanthera dilitata
Viola sp.
Hieracium triste
Senecio eremophilus
Sonchus arvensis
Gentianella amarella

Mean cover (%)

Total
Proportion of all cover
Graminoids
Calamagrostis canadensis
Phleum pratense
Elymus trachycaulus
Hordeum jubatum
"Hairy grass"
Cinna latifolia
Glyceria sp.
Bromus inermis ssp. inermis
"Tallwheat grass"
Agrostis stolonifera
Bromus ciliatus
Poa nemoralis ssp. interior
Elymus lanceolatus ssp. lanceolatus
Koelaria macrantha
Rush sp.
Poa pratensis
"Bigwheat grass"
Glyceria sp. 2
Beckmannia syzigachne
Total
Proportion of all cover

244

0.003
0.003
0.003
0.002
0.002
0.002
0.001
0.001
0.001
0.001
13.815
13.263
1.258
0.090
0.084
0.064
0.049
0.033
0.021
0.019
0.018
0.007
0.005
0.005
0.005
0.003
0.003
0.002
0.001
0.001
0.001
1.669
1.602

Cecil Lake Landslide Relevés - cont'd

Species
Ferns & Fern allies
Equisetum arvense

Mean cover (%)

Total
Proportion of all cover

33.500
33.500
32.162

Total
Proportion of all cover

0.567
0.167
0.067
0.039
0.037
0.007
0.007
0.003
0.001
0.001
0.001
0.897
0.861

Total
Proportion of all cover

0.007
0.007
0.005
0.019
0.018

Bryophytes
Eurynchiastrum pulchellum
Brachythecium sp.
Brachythecium sp. 1
Brachythecium sp.
Pohlia nutans
Dicranum sp.
Funaria sp.
Hylocomium splendens
Marchantia sp.
"Bottlebrush moss"
Pleurozium schreberi

Lichens
Cladonia sp.
Nephroma sp.
Peltigera sp.

245

Cecil Lake Undisturbed Relevés - Total mean vegetation cover = 87.022%
Species
Trees
Picea glauca
Populus tremuloides
Betula papyrifera

Me an cover (%)

Total
Proportion of all cover

22.893
9.107
3.624
35.624
40.937

Total
Proportion of all cover

6.908
6.762
4.388
4.015
1.922
0.971
0.782
0.445
0.426
0.297
0.279
0.190
0.115
0.035
0.029
0.003
27.567
31.678

Shrubs
Rosa acicularis
Viburnum edule
Shepherdia canadensis
Alnus viridis ssp. sinuata
Cornus stolonifera
Amelanchier alnifolia
Rubus pubescens
Ribes triste
Symphoricarpos albus
Ribes lacustre
Lonicera dioica
Salix sp.
Lonicera involucrata
Ribes oxyacanthoides
Rubus idaeus
Prunus virginiana

Forbs
Aralia nudicaulis
Linnaea borealis
Aster conspicuus
Cornus canadensis
Mertensia paniculata
Lathyrus ochroleucus
Clematis sp.
Mitella nuda
Disporum trachycarpum (aka Prosartes trachycarpa )

246

6.460
3.137
1.118
1.058
0.349
0.146
0.133
0.108
0.086

Cecil Lake Undisturbed Relevés - cont'd

Spe cie s
Forbs - cont'd
Orthilia secunda
Geocaulon lividum
Aster ciliolatus
Galium boreale
Pyrola chlorantha
Pyrola asarifolia
Fragaria virginiana
Fragaria vesca
Epilobium angustifolium
Actaea rubra
Osmorhiza chilensis
Galium triflorum
Viola sp.
Vicia americana
Delphinium glaucum
Goodyera oblongifolia
Listera s p. (aka Neottia sp.)
Maianthemum canadense
Taraxacum officinale

Mean cover (%)

Total
Proportion of all cover

0.081
0.060
0.053
0.053
0.047
0.038
0.033
0.027
0.020
0.018
0.018
0.012
0.007
0.005
0.004
0.003
0.003
0.003
0.003
13.083
15.034

Total
Proportion of all cover

2.253
0.253
0.240
2.746
3.156

Total
Proportion of all cover

0.020
0.003
0.002
0.025
0.029

Graminoids
"Fuzzy grass"
"Thin grass"
Calamagrostis canadensis

Ferns & Fern Allies
Fern sp.
Equisetum arvense
Gymnocarpium dryopteris

247

Cecil Lake Undisturbed Relevés - cont'd
Species
Bryophyte s
Hylocomium splendens
Pleurozium schreberi
Ptilium crista-castrensis
Rhytidiadelphus triquetrus
Eurynchiastrum pulcellum
Brachythecium sp.
Moss sp.
Dicranum polysetum
Dicranum scoparium
Plagiomnium cuspidatum

Me an cover (%)

Total
Proportion of all cover

6.210
1.118
0.323
0.110
0.087
0.027
0.020
0.014
0.010
0.007
7.926
9.108

Total
Proportion of all cover

0.047
0.003
0.050
0.057

Lichens
Peltigera polydactylon
Cladonia sp.

248

Hasler Flats Landslide Relevés - Total mean vegetation cover = 81.687%

Species
Mean cover (%)
Trees
Populus tremuloides
10.758
Populus balsamifera ssp. balsamifera
0.300
Betula papyrifera
0.103
Picea glauca
0.063
Total
11.224
Proportion of all cover
13.740
Shrubs
Rosa acicularis
10.375
Alnus viridis ssp. sinuata
6.308
Rubus idaeus
5.595
Cornus stolonifera
4.900
Symphoricarpos albus
4.145
Viburnum edule
2.423
Lonicera dioica
2.348
Amelanchier alnifolia
1.528
Prunus virginiana
1.320
Lonicera involucrata
0.716
Ribes oxyacanthoides
0.461
Salix sp.
0.434
Rubus pubescens
0.348
Ribes triste
0.335
Salix sp.
0.266
Rubus parviflorus
0.019
Shepherdia canadensis
0.003
Ribes hudsonianum
0.001
Ribes laxiflorum
0.001
Total
41.526
Proportion of all cover
50.836
Forbs
Smilacina racemosa
1.623
Aralia nudicaulis
1.403
Aster conspicuus
0.925
Mertensia paniculata
0.281
Petasites frigidus
0.248

249

Hasler Flats Landslide Relevés - cont'd

Species
Forbs - cont'd
Vicia americana
Streptopus amplexifolius
Viola sp.
Aster ciliolatus
Lathyrus ochroleucus
Galium boreale
Heracleum lanatum
Actaea rubra
Smilacina stellata
Taraxacum officinale
Fragaria vesca
Epilobium angustifolium
Cornus canadensis
Mitella nuda
Galium triflorum
Pyrola asarifolia
Thalictrum occidentale
Fragaria virginiana
Cirsium arvense
Spiraea betulifolia
Maianthemum canadense
Sonchus arvensis
Epilobium ciliatum
Linnaea borealis
Veratrum viride
Typha sp.
Veronica beccabunga
Cirsium vulgare
Achillea millefolium
Cyclamen hederifolium
Hieracium sabaudum
Orthilia secunda
Geum macrophyllum
Chenopodium capitatum

Mean cover (%)
0.247
0.232
0.199
0.176
0.176
0.163
0.098
0.090
0.088
0.088
0.084
0.082
0.074
0.068
0.065
0.061
0.054
0.047
0.046
0.040
0.039
0.030
0.029
0.027
0.023
0.015
0.010
0.007
0.005
0.003
0.003
0.003
0.002
0.001

250

Hasler Flats Landslide Relevés - cont'd

Species
Forbs - cont'd
Boechera stricta
Plantago major
"Pondweed"
Trifolium hybridum
Total
Proportion of all cover
Graminoids
Calamagrostis canadensis
Bromus inermis
Elymus glaucus ssp. glaucus
Poa palustris
"Oat grass"
Bromus ciliatus
Leymus innovatus
Elymus trachycaulus ssp. trachycaulus
Carex sp.
Carex rostrata
Festuca sp.
Trisetum cernuum
"Limp grass"
"Frothy grass"
Phleum pratense
Total
Proportion of all cover
Ferns & Fern allies
Equisetum arvense
Equisetum variegatum
Fern sp.
Total
Proportion of all cover

Mean cover (%)
0.001
0.001
0.001
0.001
6.859
8.397
1.088
1.051
0.696
0.038
0.019
0.016
0.012
0.010
0.010
0.008
0.007
0.005
0.005
0.003
0.001
2.969
3.635
16.417
0.002
0.002
16.421
20.102

251

Hasler Flats Landslide Relevés - cont'd

Species
Bryophytes
Pohlia nutans
Marchantia polymorpha
Dicranum polysetum
Syntrichia ruralis
Plagiomnium cuspidatum
Leptobryum pyriforme
Total
Proportion of all cover

Mean cover (%)
2.423
0.198
0.062
0.004
0.004
0.002
2.693
3.297

252

Hasler Flats Undisturbed Relevés - Total mean vegetation cover = 100.807%

Species
Trees
Populus tremuloides
Populus balsamifera ssp. balsamifera
Picea glauca
Acer glabrum

Mean cover (%)

Total
Proportion of all cover

13.580
4.335
0.510
0.033
18.458
18.310

Total
Proportion of all cover

9.863
7.358
6.983
5.378
5.283
2.503
1.976
1.609
1.236
1.026
0.904
0.551
0.487
0.438
0.195
0.130
0.083
0.033
0.003
0.001
46.040
45.671

Shrubs
Alnus viridis ssp. sinuata
Cornus stolonifera
Viburnum edule
Symphoricarpos albus
Rosa acicularis
Amelanchier alnifolia
Lonicera dioica
Prunus virginiana
Rubus idaeus
Salix sp. 2
Ribes oxyacanthoides
Rubus pubescens
Ribes triste
Lonicera involucrata
Salix sp. 1
Ribes hudsonianum
Shepherdia canadensis
Salix sp. 3
Rubus parviflorus
Ribes lacustre

Forbs
Aralia nudicaulis
Heracleum lanatum
Aster conspicuus
Mitella nuda

6.856
3.844
2.556
1.397

253

Hasler Flats Undisturbed Relevés - cont'd

Species
Forbs - cont'd
Smilacina racemosa
Actaea rubra
Thalictrum occidentale
Smilacina stellata
Viola sp.
Lathyrus ochroleucus
Maianthemum canadense
Streptopus amplexifolius
Vicia americana
Petasites frigidus
Angelica lucida
Pyrola asarifolia
Aster foliaceus
Mertensia paniculata
Cirsium arvense
Aster ciliolatus
Galium boreale
Galium triflorum
Disporum trachycarpum (aka Prosartes trachycarpa )
Fragaria vesca
Osmorhiza chilensis
Cornus canadensis
Orthilia secunda
Fragaria virginiana
Taraxacum officinale
Epilobium ciliatum
Veronica beccabunga
Veratrum viride
Linnaea borealis
Spiraea betulifolia
Epilobium angustifolium
Geum macrophyllum

254

Mean cover (%)
0.653
0.580
0.516
0.488
0.415
0.328
0.327
0.276
0.235
0.231
0.199
0.163
0.131
0.124
0.123
0.110
0.085
0.077
0.074
0.056
0.035
0.034
0.032
0.027
0.026
0.019
0.018
0.014
0.011
0.005
0.003
0.003

Hasler Flats Undisturbed Relevés - cont'd

Species
Forbs - cont'd
Cyclamen hederifolium
Achillea millefolium
Monotropa uniflora

Mean cover (%)

Total
Proportion of all cover
Graminoids
Calamagrostis canadensis
"Mystery grass 2"
Bromus inermis ssp. inermis
"Hairy grass"
Carex disperma
"Mystery grass"
Bromus ciliatus
Poa pratensis
Leymus innovatus
Elymus lanceolatus ssp. lanceolatus
Phleum pratense
"Soft grass"
Unknown grass sp.
Carex sp.
Glyceria sp.

0.002
0.001
0.001
20.075
19.914

Total
Proportion of all cover

0.185
0.077
0.028
0.024
0.018
0.017
0.008
0.008
0.005
0.005
0.003
0.003
0.003
0.003
0.002
0.389
0.386

Total
Proportion of all cover

15.625
0.007
0.002
15.634
15.509

Total
Proportion of all cover

0.105
0.068
0.041
0.214
0.212

Ferns & Fern allies
Equisetum arvense
Equisetum pratense
Botrychium virginianum

Bryophytes
Dicranum polysetum
Plagiomnium cuspidatum
Plagiomnium ellipticum

255

Appendix 4 Landslide geotypes glossary

The geotype concept was developed in this study to delineate and classify geomorphic
diversity as it relates to potential microsite and ecosystem diversity. No explicit guidelines
existed for this work; therefore, a novel classification system was designed, using various
sources for reference. The geotype definitions draw from several disciplines, including
geography, geology, geomorphology, and hydrology. The main feature of the geotype
classifications is that they describe identification of features on digital LiDAR imagery
(rather than in the field), with the assistance of a digital elevation model.

AP – Apron
A relatively gentle, fanned out slope at the foot of a steeper slope, which is formed by
materials from the steeper, upper slope.
BA – Bank
The sides of the channel of a river or creek, as well as the level land adjacent.
BE – Bench
A long, relatively narrow strip of level or gently sloped land with distinctly steeper slopes
above and below it.
BL – Blocky site
An accumulation of large (>25 cm) chunks of consolidated, mostly unvegetated substrate.
BS – Sandstone bedrock
Exposed sedimentary rock, identified as sandstone based on background knowledge of local
stratigraphy.
CU – Cultivated field
Raft of material broken off and transported from adjacent agricultural field.
DF – Debris flow
A mass of loose mud, sand, soil, rocks, vegetation, and air progressing down a slope.
Seepage is present, and a water source is evident.
DP – Depression
A low spot, identified by a darker circle compared to surrounding terrain.

256

EF – Earthflow
A moving mass of saturated fine-grained materials (eg. clay, fine sand, silt) progressing
downslope. Visually distinguished from debris flows by the lack of large materials in the
matrix.
FL – Fluvial
A river or creek channel with flowing water evident.
Fl – Flood deposits – seasonal
Fine blanket of material deposited by high water events. Identified by paler colour and flat
appearance compared to surrounding terrain.
FP – Floodplain
A relatively flat, extensive area next to a river or creek, where water seasonally rises and
covers the terrain. Identified by pools of standing water and shorter, saturation-tolerant
vegetation.
GU – Gully
Sharp erosions of soil due to running water, usually on a hill side. Can be tens of metres
wide and deep.
HU – Hummocks
Clusters of small knolls or mounds, usually <15 m tall.
PI – Pillar
A tall vertical column of slide material that remains after surrounding material has wasted
away.
PL – Plateau
A flat, elevated landform rising distinctly above the surrounding terrain on at least one side.
PO – Pond/wet area
Any depression that contains standing water for any period of time, as visible on drone
imagery. There is no lower size or depth limit. Identified in imagery by much darker colour
than surrounding terrain and/or presence of cattail (Typha spp.).
RA – Raft
Intact pieces of vegetated terrain that have been transported down the landslide. Can consist
of surrounding mature forest, grassland, or cultivated field.
RB – Rotational blocks
Blocks of slide material that are tilted due to rotational movement.
257

RF – Rockfall/topple
A concentrated accumulation of rock that has fallen freely from a cliff face and moved
downslope.
RI – Ridge
A chain of eroded hills that form a continuous elevated crest for some distance.
RU – Rubble
A blanket of irregular broken stone fragments on gentle or flat terrain.
SB – Sandbar
A ridge or island of sand in a river or creek channel.
SC – Scarp
A steep, often sparsely vegetated surface at the upper edge or within the landslide, caused by
movement of displaced material away from stable material.
SG – Steep grassy slope
A stable slope generally > 35% gradient, covered in grasses and some short shrubs.
SM – Slide matrix
The main portion of the slide material within which other geomorphic features are found.
Usually sparsely vegetated regosol.
ST – Steep treed slope
A stable slope generally > 35% gradient, covered in mature trees.
SW – Swale/wrinkle
Gently rolling lateral extensions of material populated by thick shrub growth.
TA – Talus/scree
A sheet of loose rock fragments over a slope.

258

Appendix 5 Geotype polygons

Beatton River Landslide Geotype polygons
Geotype code Description
AP
Apron - colluvial
BL
Blocky
BL
Blocky
BL
Blocky
BL
Blocky
CU
Cultivated field
CU
Cultivated field
DF
Debris flow
FL
Fluvial deposit
GU
Gully
HU
Hummocks
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
PO
Pond/wet area
RA
Raft
RA
Raft
RA
Raft
RA
Raft
RA
Raft
RA
Raft
RA
Raft
RA
Raft

Area (m2 ) Index in Layer
9240
39
1447
26
4889
0
5550
22
7630
27
27.9
15
197.3
12
2996
11
165.6
18
370.7
36
43060
34
8
54
8.8
46
9.8
52
9.9
47
13.4
48
21.9
49
32.4
53
35.7
43
60.4
51
72.4
42
77.2
50
91.6
41
114.4
45
129.9
44
132.3
40
67.7
6
197.9
30
518
19
610
32
670
16
718
14
938
29
1492
2

259

Beatton River Landslide Geotype polygons - continued
RA
RA
RA
RA
RA
RB
RB
RF
RF
RF
RF
RU
RU
RU
SB
SC
SC
SC
SC
SC
SC

Raft
Raft
Raft
Raft
Raft
Rotational block
Rotational block
Rockfall/topple
Rockfall/topple
Rockfall/topple
Rockfall/topple
Rubble
Rubble
Rubble
Sandstone bedrock
Scarp
Scarp
Scarp
Scarp
Scarp
Scarp

2110
2899
3014
3560
9010
211.6
844
859
2335
3102
4519
364.8
7440
11890
802
439.1
591
2312
3011
5360
15960

13
17
28
23
7
9
1
4
3
31
5
35
37
21
10
24
33
8
20
38
25

Beatton River Undisturbed Geotype polygons
Ge otype code
CU
DP
FP
PO
PO
PO
RU
SB
SG
ST
ST

Area (m2) Index in Layer
Description
Cultivated field
11750
0
Depression
9250
3
Floodplain
65000
10
Pond/wet area
606
6
Pond/wet area
340.5
7
Pond/wet area
97.5
8
Rubble
11900
2
Sandbar
9310
1
Steep grassy slope
99100
4
Steep treed slope
30250
5
Steep treed slope
64600
9

260

Cecil Lake Landslide Geotype polygons
Geotype code
EF
FL
FL
FL
Fl
Fl
Fl
Fl
GU
HU
HU
PI
PI
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO

Area (m2 ) Index in Layer
Description
Earth flow
240
13
Fluvial
500
312
Fluvial
1135
319
Fluvial
4354
311
Flood deposits -seasonal
543
315
Flood deposits -seasonal
868
314
Flood deposits -seasonal
1108
313
Flood deposits -seasonal
4052
19
Gully
222.8
12
Hummocks
2520
0
Hummocks
7790
1
Pillar
135.2
17
Pillar
143.9
18
Pond/wet area
0.7
55
Pond/wet area
0.7
222
Pond/wet area
0.8
58
Pond/wet area
0.8
200
Pond/wet area
0.8
219
Pond/wet area
0.9
50
Pond/wet area
0.9
57
Pond/wet area
0.9
210
Pond/wet area
1
94
Pond/wet area
1
180
Pond/wet area
1
192
Pond/wet area
1
198
Pond/wet area
1
216
Pond/wet area
1
224
Pond/wet area
1.1
166
Pond/wet area
1.1
217
Pond/wet area
1.3
201
Pond/wet area
1.4
177
Pond/wet area
1.6
56
Pond/wet area
1.6
197
Pond/wet area
1.8
129

261

Cecil Lake Landslide Geotype polygons - continued
Geotype code
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO

Description
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area

Area (m2 ) Index in Layer
1.8
227
1.9
43
2
178
2.1
148
2.1
175
2.2
54
2.2
236
2.3
52
2.4
157
2.5
37
2.5
220
2.5
225
2.7
51
2.7
114
2.7
174
2.7
179
2.7
223
2.8
61
2.8
221
2.9
205
3.1
199
3.2
228
3.3
38
3.4
49
3.4
102
3.5
30
3.5
122
3.6
209
3.7
176
3.9
86
3.9
111
3.9
133
3.9
230
4
35

262

Cecil Lake Landslide Geotype polygons - continued
Geotype code
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO

Description
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area

Area (m2 ) Index in Layer
4
208
4.1
118
4.3
202
4.4
110
4.4
207
4.5
146
4.5
161
4.6
135
4.6
215
4.6
218
4.7
112
4.7
116
4.7
181
4.8
66
4.8
137
4.9
143
5
41
5.1
65
5.1
206
5.2
36
5.2
132
5.3
79
5.5
282
5.6
144
5.6
167
5.6
170
5.7
93
5.9
60
6
46
6
117
6
125
6.1
140
6.2
194
6.3
78

263

Cecil Lake Landslide Geotype polygons - continued
Geotype code
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO

Description
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area

Area (m2 ) Index in Layer
6.3
81
6.3
171
6.5
168
6.5
226
6.6
203
6.6
212
6.7
238
6.8
136
6.9
47
7.1
34
7.1
204
7.2
237
7.7
113
7.9
53
8
97
8.1
59
8.2
45
8.3
300
8.5
40
8.5
100
8.6
274
8.7
147
8.8
101
8.9
134
9
44
9.1
138
9.7
156
10.1
301
10.4
164
10.6
191
10.9
235
11
119
11.2
42
11.4
92

264

Cecil Lake Landslide Geotype polygons - continued
Geotype code
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO

Description
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area

Area (m2 ) Index in Layer
11.4
95
11.5
303
11.6
173
11.8
32
12
169
12.2
77
12.3
149
12.4
145
12.5
29
13
162
13.5
232
13.8
39
13.9
172
14
70
14.6
121
14.7
229
15.1
165
15.2
245
15.5
275
16
128
16.3
67
16.8
76
16.8
98
16.8
182
17.2
297
17.2
48
17.2
163
17.4
160
17.5
63
17.8
159
17.9
142
17.9
270
18.3
83
18.5
73

265

Cecil Lake Landslide Geotype polygons - continued
Geotype code
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO

Description
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area

Area (m2 ) Index in Layer
18.9
273
19.3
33
20.2
247
20.4
62
20.9
272
21.3
103
21.5
75
22
213
22.1
261
22.7
260
22.8
120
23.6
151
24.5
107
24.6
195
24.8
126
25.4
87
25.6
211
26.4
187
27.4
233
28
105
28.7
280
29.1
302
29.1
234
29.3
306
29.4
82
30.4
256
30.7
186
31
27
31.1
296
31.2
71
31.2
269
32.1
64
32.1
255
32.4
271

266

Cecil Lake Landslide Geotype polygons - continued
Geotype code
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO

Description
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area

Area (m2 ) Index in Layer
32.7
72
32.8
295
33.4
239
34.5
246
34.9
74
34.9
281
35.5
307
35.7
141
35.7
183
36.1
189
36.7
289
38.8
299
40.4
304
40.5
254
41.5
80
41.9
264
42.2
158
44.4
104
45
109
45.6
90
46.4
190
46.7
196
46.8
139
47.3
88
47.3
294
47.6
130
48.2
106
49.2
279
51.8
193
52.1
89
57.5
131
58.4
309
61.3
276
64.2
286

267

Cecil Lake Landslide Geotype polygons - continued
Geotype code
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO

Description
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area

Area (m2 ) Index in Layer
66.5
278
67
115
67.2
85
67.4
127
68.4
99
68.6
108
69.8
310
71.8
288
75.9
259
77.9
184
79.2
305
82.6
152
88.2
284
91.3
185
92
231
97.1
285
98.8
84
100.8
96
102
287
109.8
124
110.1
253
114.3
68
114.3
123
115
28
117
91
118.8
188
126.3
250
128.9
290
129
248
129.2
263
138.2
298
138.7
214
142
244
147.9
150

268

Cecil Lake Landslide Geotype polygons - continued
Geotype code
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
RA
RA
RA
RA
RA
RA
RA
RA
RA

Description
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Raft
Raft
Raft
Raft
Raft
Raft
Raft
Raft
Raft

Area (m2 ) Index in Layer
179.5
31
184.5
243
185.8
262
189.6
277
216.9
293
218.4
308
219.8
292
262.9
155
283.4
153
287.7
69
306.3
283
317.7
268
318.8
251
362.6
154
499.9
258
539
257
636
267
695
291
797
241
1115
252
1141
266
1424
249
2231
240
2281
242
5790
265
221.6
21
390.5
3
429
22
1009
2
1211
14
1277
5
1387
7
2827
322
4466
8

269

Cecil Lake Landslide Geotype polygons - continued
Geotype code
RA
RA
RI
RI
RI
RI
RI
RI
RI
RI
RI
RI
RI
RI
SC
SC
SW
SW
SW

Description
Raft
Raft
Ridge
Ridge
Ridge
Ridge
Ridge
Ridge
Ridge
Ridge
Ridge
Ridge
Ridge
Ridge
Scarp
Scarp
Swale/wrinkle
Swale/wrinkle
Swale/wrinkle

Area (m2 ) Index in Layer
6840
6
17320
4
408.5
317
586
15
607
323
645
25
965
24
1667
324
2117
16
2244
23
2719
316
2786
20
5330
26
5790
318
2218
321
9490
320
617
10
902
11
1404
9

270

Cecil Lake Undisturbed Geotype polygons
Geotype code
BA
CU
CU
FL
FL
FL
FP
PL
PL
PL
PO
PO
PO
PO
SG
ST
ST
TA
TA
TA

2

Area (m ) Index in Layer
Description
Bank
19210
17
Cultivated field
10290
0
Cultivated field
16380
1
Fluvial
177.8
7
Fluvial
314.1
4
Fluvial
4108
2
Floodplain
1014
19
Plateau
29450
12
Plateau
32980
14
Plateau
50700
16
Pond/wet area
28.1
11
Pond/wet area
437.9
10
Pond/wet area
570
8
Pond/wet area
842
9
Steep grassy slope 184300
15
Steep treed slope
13660
13
Steep treed slope
144200
18
Talus/scree
957
6
Talus/scree
17440
3
Talus/scree
35140
5

271

Hasler Flats Landslide Geotype polygons
Ge otype code
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
PO
RA
RA
RA
RA
RA
RA
RA
RA
RI
RI
SC
SC

Description
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Pond/wet area
Raft
Raft
Raft
Raft
Raft
Raft
Raft
Raft
Ridge
Ridge
Scarp
Scarp

Area (m2) Index in Layer
4.6
5.2
6.3
7.2
7.6
9.4
10.3
10.4
10.8
12.4
15.1
17.5
22.6
26
38
38.1
39.9
53.3
55.5
58.2
87.6
104.3
105
111.6
134.5
490.3
121
132.9
287.6
366
383.4
486.1
616
4456
88.5
142
173.5
730

36
17
24
33
12
34
23
16
30
28
11
13
22
15
21
26
27
14
25
35
29
31
32
19
18
20
0
2
1
5
4
6
3
38
10
8
37
9

272

Hasler Flats Undisturbed Geotype polygons
Ge otype code
BE
BE
BE
BE
DP
DP
FP
FP
PL
PO
PO
SC
SC
SC
SC
SC

Area (m2) Index in Layer
De scription
Bench
440.2
4
Bench
385.4
9
Bench
335.5
11
Bench
750
14
Depression
428.8
7
Depression
497.9
13
Floodplain
1411
2
Floodplain
1760
3
Plateau
6190
5
Pond/wet area
16
0
Pond/wet area
7.5
1
Scarp
340.3
6
Scarp
661
8
Scarp
363.2
10
Scarp
352.1
12
Scarp
1046
15

273

Appendix 6 Microtopography elevation summary tables - BEC plots

Beatton River BEC plot elevation summary (n = 28 plots)

Plot
BE1
BE2
BE3
BE4
BE5
BE6
BE7
BE8
BE9
BE10
BE11
BE12
BE13
BE14
BE15
BE16
BE17
BE18
BE19
BE20
BE21
BE22
BE23
BE24
BE25
BE26
BE27
BE28

No. of Mean
points elev
(m)
49
644.81
50
648.03
49
640.05
49
645.42
49
643.06
50
644.54
50
642.77
51
629.00
49
594.54
47
591.38
48
594.38
45
591.42
48
575.60
47
558.80
48
577.12
47
513.62
49
511.33
50
489.28
49
491.24
48
480.94
46
475.02
50
473.63
48
484.04
47
546.41
49
644.69
51
649.54
50
645.65
49
640.08

Mean
Min
Max
Min
Standardised Standardised
microsite
microsite
microsite
elev
SD
CV
elev (m) elev (m) elev (m)
(m)
(+/-m)
(%)
643.56
1.25
0
4.88
0.87
69.41
645.57
2.46
0
4.34
1.21
49.38
639.29
0.76
0
1.39
0.38
50.09
642.10
3.33
0
7.97
1.86
55.90
640.92
2.14
0
6.17
1.14
53.38
638.61
5.93
0
11.30
3.33
56.20
641.81
0.95
0
4.47
0.84
88.29
627.26
1.75
0
3.72
1.04
59.34
593.78
0.77
0
2.33
0.54
70.87
591.05
0.33
0
1.88
0.46
139.42
593.30
1.08
0
2.21
0.60
55.69
590.68
0.74
0
1.37
0.36
49.12
575.01
0.58
0
1.81
0.47
80.36
556.37
2.43
0
5.02
1.34
55.20
574.13
2.99
0
6.14
1.66
55.73
512.43
1.19
0
2.09
0.55
46.17
510.04
1.28
0
4.28
0.92
71.57
488.87
0.42
0
1.09
0.28
67.53
490.39
0.85
0
2.33
0.51
60.20
480.24
0.70
0
1.33
0.32
45.71
474.59
0.43
0
0.91
0.21
48.15
473.31
0.32
0
1.02
0.21
64.33
482.30
1.74
0
3.03
0.73
41.82
545.57
0.84
0
1.56
0.39
45.74
643.36
1.34
0
2.90
0.92
69.12
645.96
3.58
0
7.25
2.04
56.98
645.17
0.48
0
0.99
0.24
49.85
636.22
3.86
0
7.58
2.02
52.19

274

Cecil Lake BEC plot elevation summary (n = 55 plots)

Plot
CE1
CE2
CE3
CE4
CE5
CE6
CE7
CE8
CE9
CE10
CE11
CE12
CE13
CE14
CE15
CE16
CE17
CE18
CE19
CE20
CE21
CE22
CE23
CE24
CE25
CE26
CE27
CE28
CE29
CE30

No. of Mean
points elev
(m)
49
670.75
46
661.89
48
646.31
48
647.64
46
647.60
49
633.92
49
629.59
50
663.58
48
658.92
48
663.48
46
658.25
48
647.26
49
655.66
48
646.35
51
641.23
50
642.02
46
632.64
49
629.66
50
626.75
51
614.60
47
621.29
49
652.20
50
644.48
46
638.38
49
636.20
50
634.52
51
627.78
48
616.41
48
600.86
46
579.89

Mean
Min
Max
Min
Standardised Standardised
elev microsite microsite microsite
SD
CV
elev (m) elev (m) elev (m)
(m)
(+/-m)
(%)
669.25
1.50
0
3.54
0.93
62.48
660.42
1.47
0
2.44
0.68
46.45
646.08
0.23
0
0.84
0.19
79.53
646.23
1.41
0
2.91
0.89
63.39
645.41
2.19
0
3.63
1.11
50.66
631.44
2.48
0
4.77
1.38
55.63
629.17
0.42
0
0.78
0.22
52.74
661.83
1.75
0
3.04
0.83
47.07
657.26
1.66
0
2.66
0.73
44.05
662.53
0.95
0
1.89
0.50
53.23
656.34
1.91
0
3.76
1.17
61.18
645.07
2.19
0
4.71
1.27
58.05
653.88
1.78
0
2.76
0.78
43.95
645.84
0.51
0
0.73
0.15
29.17
640.83
0.40
0
0.72
0.18
46.01
641.23
0.79
0
1.40
0.29
37.31
631.10
1.54
0
3.39
1.01
65.54
628.45
1.21
0
1.80
0.47
38.93
626.00
0.75
0
1.25
0.28
37.60
614.45
0.15
0
0.54
0.12
74.66
620.42
0.87
0
1.80
0.45
51.40
650.59
1.61
0
2.59
0.74
45.71
643.92
0.56
0
0.88
0.23
41.73
637.38
1.00
0
1.88
0.54
54.41
635.57
0.63
0
1.42
0.37
59.57
634.35
0.17
0
0.87
0.19
112.80
627.42
0.36
0
1.03
0.25
69.62
615.01
1.40
0
2.60
0.76
54.15
600.26
0.60
0
1.64
0.39
65.89
578.68
1.21
0
2.23
0.57
46.81

275

Cecil Lake BEC plot elevation summary - cont'd

Plot
CE31
CE32
CE33
CE34
CE36
CE37
CE38
CE39
CE40
CE41
CE42
CE43
CE44
CE45
CE46
CE47
CE48
CE49
CE50
CE51
CE52
CE53
CE54
CE55
CE56

No. of Mean
points elev
(m)
50
564.17
50
568.46
50
578.46
48
574.87
50
640.36
48
645.77
49
646.10
50
641.34
48
639.52
52
626.70
49
611.79
50
610.97
50
626.78
47
588.08
52
633.83
48
632.16
48
616.63
47
588.08
51
580.27
49
578.59
49
572.10
48
572.15
49
570.99
49
582.11
49
581.21

Mean
Min
Max
Min
Standardised Standardised
elev microsite microsite microsite
SD
CV
elev (m) elev (m) elev (m)
(m)
(+/-m)
(%)
564.08
0.09
0
0.21
0.04
43.51
567.72
0.74
0
1.36
0.38
52.25
577.21
1.25
0
2.55
0.79
63.03
574.77
0.10
0
0.14
0.02
22.46
640.28
0.08
0
0.15
0.03
40.87
644.88
0.89
0
1.88
0.56
62.80
644.99
1.11
0
1.93
0.54
49.08
640.08
1.26
0
2.19
0.57
45.74
638.94
0.58
0
0.88
0.18
30.82
626.57
0.13
0
0.47
0.09
72.30
611.35
0.44
0
1.10
0.32
72.94
610.76
0.21
0
1.23
0.25
118.97
625.46
1.32
0
2.60
0.84
63.42
586.01
2.07
0
4.05
1.08
52.10
633.03
0.80
0
1.45
0.37
45.97
631.60
0.56
0
1.47
0.43
77.63
616.02
0.61
0
1.53
0.36
59.16
586.01
2.07
0
4.05
1.08
52.10
579.31
0.96
0
1.96
0.62
65.05
575.91
2.68
0
4.32
1.24
46.13
571.97
0.13
0
0.54
0.10
72.33
571.18
0.97
0
1.48
0.34
34.63
570.47
0.52
0
1.10
0.29
55.65
581.57
0.54
0
1.14
0.30
55.37
580.97
0.24
0
0.91
0.21
88.92

276

Hasler Flats BEC plot elevation summary (n = 29)
No. of Mean
points elev
Plot
(m)
HA2
41
598.72
HA3
51
599.56
HA4
52
601.95
HA5
62
599.19
HA6
65
601.10
HA7
99
609.13
HA8
74
608.47
HA9
66
606.40
HA10 43
606.87
HA11 48
604.89
HA12 36
603.79
HA13 58
604.59
HA14 74
603.80
HA15 90
604.29
HA16 102 601.01
HA17 79
601.15
HA18 92
601.81
HA19 70
599.28
HA20 69
600.72
HA21 23
600.69
HA22 36
598.98
HA23 29
597.25
HA24 60
599.87
HA25 52
600.39
HA26 42
602.12
HA27 81
603.00
HA28 58
607.83
HA29 112 602.58
HA30 78
601.15

Mean
Min
Max Standardised Standardised
Min
SD
CV
elev microsite microsite microsite
elev (m) elev (m) elev (m)
(+/-m)
(%)
(m)
598.24
0.48
0
1.24
0.26
54.05
598.74
0.82
0
1.74
0.47
57.18
601.33
0.62
0
1.51
0.40
64.81
598.17
1.01
0
2.55
0.73
71.48
600.41
0.69
0
1.38
0.29
41.77
607.97
1.16
0
2.82
0.78
67.24
606.90
1.57
0
2.48
0.68
43.32
605.33
1.07
0
1.93
0.56
52.16
606.50
0.37
0
0.82
0.18
48.51
604.59
0.30
0
0.55
0.13
42.29
603.38
0.41
0
1.81
0.38
91.17
603.77
0.82
0
1.86
0.50
60.75
603.17
0.63
0
1.03
0.21
34.09
602.92
1.37
0
3.18
0.84
61.53
600.00
1.01
0
1.64
0.46
45.51
599.94
1.21
0
1.81
0.53
44.20
600.75
1.06
0
1.57
0.34
31.77
598.76
0.52
0
1.09
0.25
48.00
600.20
0.52
0
1.09
0.28
53.71
599.93
0.76
0
1.04
0.23
29.93
598.74
0.24
0
0.67
0.16
64.27
597.13
0.12
0
0.22
0.06
51.50
599.09
0.78
0
1.43
0.33
41.90
599.83
0.56
0
1.05
0.33
60.08
600.58
1.54
0
2.34
0.54
35.14
602.00
1.00
0
1.96
0.54
53.84
606.65
1.18
0
4.23
1.16
98.30
601.63
0.95
0
1.90
0.59
62.31
600.87
0.28
0
0.66
0.17
61.94

277

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0009 515 141 10U 635411 6231647 Retrogressive rotational failure
0.002 154 61 10U 648617 6262829 Retrogressive rotational failure
0.0023 517 141 10U 635411 6231647 Retrogressive rotational failure
0.0026 680 196 10U 658130 6232188 Retrogressive rotational failure
0.0026 52
71 10U 660889 6229983 Shallow retrogressive failure
0.0028 96
45 10U 645103 6237317 Retrogressive rotational failure
0.003
51
71 10U 660889 6229983 Shallow retrogressive failure
0.0032 616 173 10U 647595 6263055 Compound failure
0.0032 153 61 10U 648617 6262829 Retrogressive rotational failure
0.0034 497 136 10U 627698 6232647 Rotational failure
0.0034 587 47 10U 640425 6241038 Retrogressive rotational failure
0.0036 88
40 10U 648407 6238441 Multi-level rotational failure
0.0036 344 94 10U 586742 6231586 Rotational failure
0.004 280 83 10U 611833 6238026 Retrogressive rotational failure
0.0043 424 113 10U 607910 6245770 Rotational failure
0.0043 378 100 10U 598414 6248139 Compound failure
0.0045 48
71 10U 660889 6229983 Shallow retrogressive failure
0.0046 679 196 10U 658130 6232188 Retrogressive rotational failure
0.0047 283 83 10U 611833 6238026 Retrogressive rotational failure
0.0049 82
40 10U 648407 6238441 Multi-level rotational failure
0.005
37
27 10U 661452 6225535 Compound failure
0.005 367 100 10U 598414 6248139 Compound failure
0.0051 476 126 10U 631665 6307193 Rotational failure
0.0054 456 122 10U 619929 6209531 Mobile flow
0.0054 345 94 10U 586742 6231586 Rotational failure
0.0054 519 141 10U 635411 6231647 Retrogressive rotational failure
0.0055 520 141 10U 635411 6231647 Retrogressive rotational failure
0.0056 87
41 10U 647664 6238258 Retrogressive rotational failure
0.0056 522 140 10U 634447 6231204 Retrogressive rotational failure
0.0059 514 141 10U 635411 6231647 Retrogressive rotational failure
0.006 342 94 10U 586742 6231586 Rotational failure
0.006 521 141 10U 635411 6231647 Retrogressive rotational failure
Pond location on slide
Flat area midway down slide
In fresh material
Flat area midway down slide
Right next to creek
Just below headscarp (small, intact)
Below small scarp (intact)
Below headscarp (small, intact)
Just below new part of headscarp (small, intact)
In fresh material
Above small scarp (intact), near river
Above small scarp (in debris)
Debris apron
Debris apron
Toe
Below small scarp (in debris)
Debris apron near river
Just below headscarp (small, intact)
Next to creek
At bottom of slide
Debris apron below large scarp (intact)
Toe
Debris apron near river
Base of headscarp (small, intact)
Near top of slide
Debris apron
Above small scarp (intact)
Base of slide, next to creek
Below small scarp (intact)
Above small scarp (intact)
Flat area midway down slide
Debris apron
Below small scarp (intact)

278

Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Active
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A
Geomorphic
location
of pond
Body
Toe
Toe
Toe
Head
Toe
Body
Head
Toe
Body
Toe
Toe
Body
Toe
Head
Toe
Head
Toe
Toe
Toe
Toe
Toe
Head
Body
Body
Toe
Toe
Body
Body
Body
Body
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.006 678 196 10U 658130 6232188 Retrogressive rotational failure
0.0061 457 122 10U 619929 6209531 Mobile flow
0.0063 36
27 10U 661452 6225535 Compound failure
0.0064 90
43 10U 646755 6238196 Retrogressive rotational failure
0.0064 317 89 10U 583121 6249823 Multi-level rotational failure
0.0064 381 100 10U 598414 6248139 Compound failure
0.0066 100 46 10U 640572 6238860 Retrogressive rotational failure
0.0066 637 182 10U 653926 6268718 Retrogressive rotational failure
0.0067 459 122 10U 619929 6209531 Mobile flow
0.0068 506 139 10U 633271 6230890 Compound failure
0.0068 25
60 10U 647207 6252566 Rotational failure
0.0068
4
57 10U 645882 6252696 Retrogressive rotational failure
0.0068 50
71 10U 660889 6229983 Shallow retrogressive failure
0.0069 335 90 10U 584584 6239849 Retrogressive rotational failure
0.0069 767 226 10U 673115 6239339 Multi-level rotational failure
0.0069 393 104 10U 607196 6251797 Compound failure
0.007
49
71 10U 660889 6229983 Shallow retrogressive failure
0.0071 484 130 10U 633609 6300494 Compound failure
0.0072 379 100 10U 598414 6248139 Compound failure
0.0072 615 173 10U 647595 6263055 Compound failure
0.0073 592 162 10U 639763 6241926 Retrogressive rotational failure
0.0074 542
3 10U 635265 6222327 Retrogressive rotational failure
0.0075 77
44 10U 646758 6237400 Retrogressive rotational failure
0.0075 369 100 10U 598414 6248139 Compound failure
0.0077 38
27 10U 661452 6225535 Compound failure
0.0077 377 100 10U 598414 6248139 Compound failure
0.0079 639 184 10U 657390 6240045 Rotational failure
0.0079 584 161 10U 639518 6240605 Retrogressive rotational failure
0.0081 423 113 10U 607910 6245770 Rotational failure
0.0081 530 144 10U 626024 6221319 Rotational failure
0.0081 635 182 10U 653926 6268718 Retrogressive rotational failure
0.0081 410 108 10U 607734 6248896 Rotational failure
Pond location on slide
Next to creek
Near top of slide
Toe
Just above small scarp (intact)
Debris apron, right at base of large scarp (intact)
Debris apron near river
Below small scarp (in debris)
Below/within small scarp (intact), near river
Near top of slide (midway across)
On hillside in small depression
Debris apron
Below small scarp (intact)
Just below headscarp (small, intact)
Just above small retrogressive scarp (intact)
Above small scarp (intact)
Debris apron
Just below headscarp (small, intact)
Base of headscarp (small, intact)
Debris apron
Just below new part of headscarp (small, intact)
Below headscarp (large, intact)
Below small scarp (in debris)
Below small scarp (in debris)
Debris apron near river
Toe
Debris apron near river
Debris apron near creek
Below small scarp (intact), near river
Base of headscarp (large, intact)
Below small scarp (intact)
Above small scarp (intact), near river
Above small scarp (intact)

279

Dormant
Active
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Toe
Body
Toe
Toe
Toe
Toe
Body
Toe
Body
Body
Toe
Toe
Body
Body
Body
Body
Body
Head
Body
Head
Head
Body
Body
Toe
Toe
Toe
Toe
Toe
Head
Body
Toe
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0083 417 111 10U 608499 6247479 Compound failure
0.0083 660 189 10U 654871 6235759 Retrogressive rotational failure
0.0084 373 100 10U 598414 6248139 Compound failure
0.0086 737 15 10U 655647 6215120 Retrogressive rotational failure
0.0087 402 105 10U 607741 6250210 Compound failure
0.0087 16
57 10U 645882 6252696 Retrogressive rotational failure
0.0088 472 126 10U 631665 6307193 Rotational failure
0.0089 659 189 10U 654871 6235759 Retrogressive rotational failure
0.0089 341 94 10U 586742 6231586 Rotational failure
0.0089 430 114 10U 607383 6243249 Retrogressive rotational failure
0.009 421 112 10U 608153 6247351 Rotational failure
0.009 607 53 10U 641413 6249461 Retrogressive rotational failure
0.009 617 173 10U 647595 6263055 Compound failure
0.0091 498 136 10U 627698 6232647 Rotational failure
0.0093 316 89 10U 583121 6249823 Multi-level rotational failure
0.0093 586 47 10U 640425 6241038 Retrogressive rotational failure
0.0094 548 145 10U 634428 6225821 Multi-level rotational failure
0.0095 47
71 10U 660889 6229983 Shallow retrogressive failure
0.0097 231
4 10U 637998 6220826 Multi-level rotational failure
0.0097 730 14 10U 655144 6217752 Compound failure
0.0098 346 94 10U 586742 6231586 Rotational failure
0.01
57
33 10U 657471 6230119 Compound failure
0.01
408 107 10U 607530 6250024 Compound failure
0.01
609 169 10U 642693 6252730 Retrogressive rotational failure
0.01
766 225 10U 672724 6237484 Retrogressive rotational failure
0.0101 20
57 10U 645882 6252696 Retrogressive rotational failure
0.0101 471 126 10U 631665 6307193 Rotational failure
0.0102 588 47 10U 640425 6241038 Retrogressive rotational failure
0.0102 254
1 10U 631931 6221078 Retrogressive rotational failure
0.0103 418 111 10U 608499 6247479 Compound failure
0.0104 163 22 10U 668246 6222087 Compound failure
0.0106 286 83 10U 611833 6238026 Retrogressive rotational failure
Pond location on slide
Above small scarp (in debris)
Above small scarp (intact)
Debris apron near river
Next to South side scarp
Within small scarp (in debris) below headscarp (small, intact)
Just below headscarp (large, intact)
Below headscarp (small, intact)
Above small scarp (intact)
Debris apron
Near creek/river
Below small scarp (intact)
Below headscarp (large, intact)
Below new part of headscarp (small, intact)
Above small scarp (intact), near river
Debris apron below large scarp (intact)
Next to small scarp (in debris)
Debris apron
Just below headscarp (small, intact)
Debris apron - near creek
Within small scarp (in debris)
Debris apron
Below small scarp (intact)
Within headscarp (small, intact) by south sidescarp
Below headscarp (small, intact)
Next to small scarp (intact)
Near small scarp (in debris)
Below headscarp (small, intact)
Below small scarp (in debris), near river
Above backtilt
Below headscarp (large, intact)
Below small scarp (intact)
At bottom of slide

280

Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Body
Body
Toe
Body
Body
Head
Head
Body
Body
Toe
Toe
Body
Head
Body
Toe
Toe
Toe
Head
Toe
Head
Body
Body
Head
Head
Body
Body
Body
Toe
Body
Head
Body
Toe

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0106 602 165 10U 641712 6246986 Rotational failure
0.0106 558 148 10U 635615 6217619 Retrogressive rotational failure
0.0106 372 100 10U 598414 6248139 Compound failure
0.0106 31
24 10U 662143 6223550 Rotational failure
0.011 404 105 10U 607741 6250210 Compound failure
0.0111 668 38 10U 652244 6236095 Retrogressive rotational failure
0.0112 502 137 10U 628079 6232225 Multi-level rotational failure
0.0112 550 145 10U 634428 6225821 Multi-level rotational failure
0.0113 309 87 10U 580867 6250762 Compound failure
0.0114 399 105 10U 607741 6250210 Compound failure
0.0118 368 100 10U 598414 6248139 Compound failure
0.0118 677 196 10U 658130 6232188 Retrogressive rotational failure
0.0118 749 214 10U 664400 6210645 Multi-level rotational failure
0.0122 376 100 10U 598414 6248139 Compound failure
0.0122 308 87 10U 580867 6250762 Compound failure
0.0123 219
4 10U 637998 6220826 Multi-level rotational failure
0.0124 478 127 10U 632593 6304450 Rotational failure
0.0125 518 141 10U 635411 6231647 Retrogressive rotational failure
0.0125 422 112 10U 608153 6247351 Rotational failure
0.0127 429 114 10U 607383 6243249 Retrogressive rotational failure
0.0127 638 183 10U 652608 6268058 Retrogressive rotational failure
0.0128 400 105 10U 607741 6250210 Compound failure
0.0128 489 132 10U 633416 6298822 Rotational failure
0.0129 724 207 10U 655366 6218822 Retrogressive rotational failure
0.0129 289 83 10U 611833 6238026 Retrogressive rotational failure
0.0129 425 113 10U 607910 6245770 Rotational failure
0.0129 535
1 10U 631931 6221078 Retrogressive rotational failure
0.013 371 100 10U 598414 6248139 Compound failure
0.013 509 140 10U 634447 6231204 Retrogressive rotational failure
0.0134 403 105 10U 607741 6250210 Compound failure
0.0134 512 141 10U 635411 6231647 Retrogressive rotational failure
0.0134 582 161 10U 639518 6240605 Retrogressive rotational failure
Pond location on slide
Base of headscarp (large, intact)
Above small scarp (intact)
Debris apron right by river
Above small scarp (in debris)
Above small scarp (in debris)
Next to river at bottom of slide
Just above small scarp (intact), near river
Debris apron
Debris apron - below headscarp (small, intact), above large scarp (intact)
Within small scarp (in debris) below headscarp (small, intact)
Debris apron near river
Below small scarp (in debris), near creek
Below headscarp (small, intact)
Debris apron
Just below headscarp (small, intact), above large scarp (intact)
Next to small scarp (intact)
Base of headscarp (small, intact)
Above small scarp (intact)
Above small scarp (intact)
Near creek/river
Below/within small scarp (intact)
Below small scarp (in debris)
Below small scarp (in debris), near river
Within large scarp (intact)
At bottom of slide
Below small scarp (in debris)
Below small scarp (in debris)
Debris apron near river
Below small scarp (intact)
Below small scarp (in debris)
Flat area midway down slide
Below small scarp (intact), near river

281

Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Body
Toe
Toe
Body
Body
Toe
Body
Toe
Body
Body
Toe
Toe
Head
Body
Body
Body
Head
Toe
Body
Toe
Toe
Body
Toe
Toe
Toe
Head
Body
Toe
Body
Body
Body
Toe

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0135 99
46 10U 640572 6238860 Retrogressive rotational failure
0.0136 566 153 10U 637499 6223113 Rotational failure
0.0138 781 233 10U 682783 6224622 Rotational failure
0.014 723 207 10U 655366 6218822 Retrogressive rotational failure
0.0141 80
40 10U 648407 6238441 Multi-level rotational failure
0.0141 18
57 10U 645882 6252696 Retrogressive rotational failure
0.0142 470 125 10U 622324 6288717 Rotational failure
0.0144 44
71 10U 660889 6229983 Shallow retrogressive failure
0.0144 445 121 10U 608754 6224408 Rotational failure
0.0144 191 10 10U 652580 6220750 Retrogressive rotational failure
0.0144 419 111 10U 608499 6247479 Compound failure
0.0145 415 111 10U 608499 6247479 Compound failure
0.0145 681 196 10U 658130 6232188 Retrogressive rotational failure
0.0145 613 172 10U 641260 6254208 Rotational failure
0.0147 365 100 10U 598414 6248139 Compound failure
0.0147 416 111 10U 608499 6247479 Compound failure
0.0147 655 189 10U 654871 6235759 Retrogressive rotational failure
0.0148 511 141 10U 635411 6231647 Retrogressive rotational failure
0.0149 374 100 10U 598414 6248139 Compound failure
0.015 774 229 10U 674428 6241207 Retrogressive rotational failure
0.0151 527 143 10U 624881 6220596 Retrogressive rotational failure
0.0152 86
41 10U 647664 6238258 Retrogressive rotational failure
0.0154 294 84 10U 612699 6237705 Retrogressive rotational failure
0.0154 763 224 10U 673168 6235959 Retrogressive rotational failure
0.0155 630 180 10U 634582 6300315 Retrogressive rotational failure
0.0157 329 90 10U 584584 6239849 Retrogressive rotational failure
0.0157 732 14 10U 655144 6217752 Compound failure
0.0158 98
45 10U 645103 6237317 Retrogressive rotational failure
0.0159 631 180 10U 634582 6300315 Retrogressive rotational failure
0.0162 526 143 10U 624881 6220596 Retrogressive rotational failure
0.0163 204 14 10U 655144 6217752 Compound failure
0.0163 485 130 10U 633609 6300494 Compound failure
Pond location on slide
Just below small scarp (in debris)
Within small scarp (in debris)
Debris apron below headscarp (large, intact)
Below small scarp (intact)
Debris apron below large scarp (intact)
Next to small scarp (in debris)
Below headscarp (small, intact)
Just below headscarp (small, intact)
Debris apron near river
Below small scarp (intact)
Below headscarp (large, intact)
Below small scarp (in debris)
Below headscarp (large, intact)
Debris apron below headscarp (large, intact)
Debris apron near river
Above small scarp (in debris)
Beside creek
Flat area below small scarp (in debris)
Debris apron
Within headscarp (small, intact)
Below small scarp (intact)
Below small scarp (intact)
Below small scarp (in debris)
Below small scarp (intact)
Base of small scarp (intact)
Just below small retrogressive scarp (intact)
Below small scarp (in debris)
Below small scarp (in debris)
Below small scarp (intact)
Below small scarp (intact)
Below small scarp (in debris)
Below small scarp (intact)

282

Dormant
Active
Dormant
Dormant
Active
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Body
Body
Toe
Body
Toe
Body
Body
Head
Toe
Toe
Head
Toe
Body
Body
Toe
Body
Toe
Toe
Body
Head
Toe
Body
Toe
Body
Body
Body
Body
Body
Body
Toe
Body
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0166 557 148 10U 635615 6217619 Retrogressive rotational failure
0.0168 420 111 10U 608499 6247479 Compound failure
0.0169 683 197 10U 659267 6231807 Rotational failure
0.0169 682 196 10U 658130 6232188 Retrogressive rotational failure
0.017 460 122 10U 619929 6209531 Mobile flow
0.017 363 99 10U 598103 6246815 Rotational failure
0.0173 414 110 10U 608193 6247991 Compound failure
0.0173 411 109 10U 608185 6248414 Compound failure
0.0173 667 38 10U 652244 6236095 Retrogressive rotational failure
0.0173 505 138 10U 632037 6230791 Rotational failure
0.0174 282 83 10U 611833 6238026 Retrogressive rotational failure
0.0174 664 191 10U 649843 6238652 Rotational failure
0.0174 326 90 10U 584584 6239849 Retrogressive rotational failure
0.0174 375 100 10U 598414 6248139 Compound failure
0.0175 562 151 10U 637222 6220172 Compound failure
0.0176 362 99 10U 598103 6246815 Rotational failure
0.0177 718 204 10U 654675 6219635 Retrogressive rotational failure
0.0181 742 212 10U 656103 6216700 Retrogressive rotational failure
0.0182 30
24 10U 662143 6223550 Rotational failure
0.0184 328 90 10U 584584 6239849 Retrogressive rotational failure
0.0187 382 100 10U 598414 6248139 Compound failure
0.0187 641 185 10U 656213 6239712 Rotational failure
0.0188 176 20 10U 679534 6219946 Rotational failure
0.0189 392 103 10U 589643 6282999 Rotational failure
0.019 474 126 10U 631665 6307193 Rotational failure
0.0191 531 144 10U 626024 6221319 Rotational failure
0.0191 263 83 10U 611833 6238026 Retrogressive rotational failure
0.0194 748 214 10U 664400 6210645 Multi-level rotational failure
0.0194 440 116 10U 606417 6242052 Rotational failure
0.0194 302 85 10U 612971 6237140 Retrogressive rotational failure
0.0195 315 89 10U 583121 6249823 Multi-level rotational failure
0.0196 469 125 10U 622324 6288717 Rotational failure
Above small scarp (intact)
Below small scarp (in debris)
Debris apron at base of headscarp (large, intact)
Just above small scarp (in debris)
Near top of slide (midway across)
Within small scarp (in debris)
Below headscarp (small, intact)
Debris apron near river/creek
Next to river at bottom of slide
Debris apron
At bottom of slide
Within headscarp (large, intact)
Above small retrogressive scarp (intact)
Debris apron right by river
Middle of slide
Within small scarp (in debris)
Below headscarp (small, intact)
Bottom of slide, near creek
Above small scarp (intact)
Within small retrogressive scarp (intact)
Debris apron near river
Base of headscarp (large, intact)
Below headscarp (large, intact)
Debris apron
Base of headscarp (small, intact)
Above small scarp (intact)
Below headscarp (large, intact), near west sidescarp
Above small scarp (in debris)
Debris apron
Below small scarp (in debris)
Debris apron below large scarp (intact)
Halfway down slide

Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

283

Pond location on slide

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Toe
Toe
Body
Body
Body
Body
Body
Toe
Toe
Toe
Toe
Head
Body
Toe
Body
Body
Head
Toe
Body
Body
Toe
Body
Body
Body
Head
Body
Body
Body
Toe
Toe
Toe
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0198 213
9 10U 639115 6218246 Rotational failure
0.02
634 182 10U 653926 6268718 Retrogressive rotational failure
0.0201 725 207 10U 655366 6218822 Retrogressive rotational failure
0.0201 481 127 10U 632593 6304450 Rotational failure
0.0204 686 198 10U 660187 6227739 Retrogressive rotational failure
0.0204 398 105 10U 607741 6250210 Compound failure
0.0204 327 90 10U 584584 6239849 Retrogressive rotational failure
0.0206 447 122 10U 619929 6209531 Mobile flow
0.0208 473 126 10U 631665 6307193 Rotational failure
0.0208 370 100 10U 598414 6248139 Compound failure
0.021 644 186 10U 655856 6238250 Retrogressive rotational failure
0.021 483 129 10U 633999 6301639 Rotational failure
0.0212 722 207 10U 655366 6218822 Retrogressive rotational failure
0.0212 250
1 10U 631931 6221078 Retrogressive rotational failure
0.0215 360 97 10U 597665 6245826 Rotational failure
0.0216 409 108 10U 607734 6248896 Rotational failure
0.0221 490 132 10U 633416 6298822 Rotational failure
0.0222 126 57 10U 645882 6252696 Retrogressive rotational failure
0.0223 23
57 10U 645882 6252696 Retrogressive rotational failure
0.0224 606 53 10U 641413 6249461 Retrogressive rotational failure
0.0224 338 93 10U 585645 6232574 Rotational failure
0.0225 593 74 10U 636436 6241818 Compound failure
0.0226 649 187 10U 655524 6236858 Multi-level rotational failure
0.0226 661 190 10U 653976 6235516 Retrogressive rotational failure
0.0226 278 83 10U 611833 6238026 Retrogressive rotational failure
0.0233 21
57 10U 645882 6252696 Retrogressive rotational failure
0.0235 500 136 10U 627698 6232647 Rotational failure
0.0235 687 199 10U 659140 6227440 Multi-level rotational failure
0.0237 380 100 10U 598414 6248139 Compound failure
0.024 395 105 10U 607741 6250210 Compound failure
0.024
84
41 10U 647664 6238258 Retrogressive rotational failure
0.0241 160 63 10U 656420 6272207 Rotational failure
Pond location on slide
Below small scarp (intact)
Above small scarp (intact), near river
Within large scarp (intact)
Below headscarp (small, intact)
Below small scarp (intact), right by river
At base of headscarp (large, intact)
Above small retrogressive scarp (intact)
Near base of slide
Below headscarp (small, intact)
Debris apron near river
Below large headscarp (intact)
Just below headscarp (small) and above small scarp (both intact)
Base of headscarp (small, intact)
Above backtilt
On level portion of headscarp (large, intact)
Next to small scarp (intact)
Below small scarp (in debris), near river
Below small scarp (in debris)
Below small scarp (in debris)
Below headscarp (large, intact)
Bottom of slide, by river
Below headscarp (large, intact)
Halfway down slide, by south sidescarp
Apron at top of slide below headscarp (large, intact)
Toe
Below headscarp (large, intact)
Above small scarp (intact), near river
Below small scarp (in debris)
Debris apron
Base of headscarp (small, intact)
Below small scarp (intact)
Just below small scarp (in debris)

284

Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Active
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Toe
Toe
Toe
Body
Toe
Head
Body
Body
Head
Toe
Head
Body
Head
Body
Head
Body
Toe
Body
Toe
Body
Toe
Body
Body
Head
Toe
Body
Body
Toe
Body
Head
Body
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0241 696 10 10U 652580 6220750 Retrogressive rotational failure
0.0242 701 10 10U 652580 6220750 Retrogressive rotational failure
0.0243 764 224 10U 673168 6235959 Retrogressive rotational failure
0.0246 394 105 10U 607741 6250210 Compound failure
0.0246 464 122 10U 619929 6209531 Mobile flow
0.0251 325 90 10U 584584 6239849 Retrogressive rotational failure
0.0253 790 20 10U 679534 6219946 Rotational failure
0.0254 614 172 10U 641260 6254208 Rotational failure
0.0254 340 94 10U 586742 6231586 Rotational failure
0.0255 300 85 10U 612971 6237140 Retrogressive rotational failure
0.0255 32
25 10U 661743 6223747 Compound failure
0.0257 386 103 10U 589643 6282999 Rotational failure
0.0258 264 83 10U 611833 6238026 Retrogressive rotational failure
0.0259 758 219 10U 672065 6223288 Retrogressive rotational failure
0.0264 773 228 10U 676317 6244415 Retrogressive rotational failure
0.0265 529 144 10U 626024 6221319 Rotational failure
0.0266 260
3 10U 635265 6222327 Retrogressive rotational failure
0.0267 523 140 10U 634447 6231204 Retrogressive rotational failure
0.0268 40
29 10U 659264 6228605 Rotational failure
0.0271 656 189 10U 654871 6235759 Retrogressive rotational failure
0.0273 675 195 10U 655685 6231888 Rotational failure
0.0274 303 85 10U 612971 6237140 Retrogressive rotational failure
0.0276 706 10 10U 652580 6220750 Retrogressive rotational failure
0.0276 102 41 10U 647664 6238258 Retrogressive rotational failure
0.0277 274 83 10U 611833 6238026 Retrogressive rotational failure
0.0279 56
33 10U 657471 6230119 Compound failure
0.0281 689 200 10U 658361 6228323 Multi-level rotational failure
0.0283 544
3 10U 635265 6222327 Retrogressive rotational failure
0.0285 662 190 10U 653976 6235516 Retrogressive rotational failure
0.0285 780 232 10U 681946 6227934 Retrogressive rotational failure
0.0286 756 217 10U 662941 6224282 Retrogressive rotational failure
0.0287 413 110 10U 608193 6247991 Compound failure
Pond location on slide
Above small scarp (in debris)
Above small scarp (in debris)
Within/below headscarp (small, intact)
Small scarp (in debris)
Top of slide along west sidescarp
Above small retrogressive scarp (intact)
Base of headscarp (large, intact)
Debris apron at base of new headscarp (large, intact)
Debris apron
Below small scarp (in debris)
Below small scarp (intact)
Debris apron
Below small scarp (in debris)
Below small scarp (intact)
Within headscarp (small, intact)
Just above small scarp (intact)
Above small scarp, below headscarp (large, intact)
Right above small scarp (intact)
Debris apron
Beside creek
Above small scarp (intact)
Below small scarp (in debris)
Next to small scarp (in debris)
Just below large scarp (intact)
At base of large scarp (intact)
Below large scarp (intact)
Below small scarp (in debris)
Below small scarp (in debris)
Apron at top of slide below headscarp (large, intact)
Base of slide next to creek
Above small scarp (intact)
Below small scarp (in debris)

285

Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Abandoned
Dormant
Abandoned
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Head
Body
Head
Body
Head
Body
Body
Head
Body
Toe
Body
Body
Body
Body
Head
Toe
Head
Body
Toe
Toe
Body
Toe
Body
Toe
Body
Toe
Toe
Body
Head
Toe
Body
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0288 589 47 10U 640425 6241038 Retrogressive rotational failure
0.0288 784 235 10U 685591 6226178 Multi-level rotational failure
0.0288 389 103 10U 589643 6282999 Rotational failure
0.0289 396 105 10U 607741 6250210 Compound failure
0.0289 125 57 10U 645882 6252696 Retrogressive rotational failure
0.0291 230
4 10U 637998 6220826 Multi-level rotational failure
0.0296 426 114 10U 607383 6243249 Retrogressive rotational failure
0.0296 757 218 10U 663556 6223926 Retrogressive rotational failure
0.0299 597 163 10U 640344 6243435 Multi-level rotational failure
0.0302 571 152 10U 637789 6222601 Retrogressive rotational failure
0.0304 524 142 10U 630879 6227122 Retrogressive rotational failure
0.0307 229
4 10U 637998 6220826 Multi-level rotational failure
0.0307 716 204 10U 654675 6219635 Retrogressive rotational failure
0.0307 58
33 10U 657471 6230119 Compound failure
0.0309 705 10 10U 652580 6220750 Retrogressive rotational failure
0.0309 772 227 10U 673362 6239998 Retrogressive rotational failure
0.031 499 136 10U 627698 6232647 Rotational failure
0.0316 646 186 10U 655856 6238250 Retrogressive rotational failure
0.0318 698 10 10U 652580 6220750 Retrogressive rotational failure
0.0319 359 97 10U 597665 6245826 Rotational failure
0.0321 625 178 10U 637541 6284374 Retrogressive rotational failure
0.0321 727 208 10U 655089 6218426 Retrogressive rotational failure
0.0321 322 90 10U 584584 6239849 Retrogressive rotational failure
0.0322 603 166 10U 641866 6247606 Retrogressive rotational failure
0.0327 491 133 10U 633983 6299243 Rotational failure
0.033 777 230 10U 673848 6240822 Rotational failure
0.0331
2
57 10U 645882 6252696 Retrogressive rotational failure
0.0334 406 107 10U 607530 6250024 Compound failure
0.0334 401 105 10U 607741 6250210 Compound failure
0.0335 465 122 10U 619929 6209531 Mobile flow
0.0338 124 57 10U 645882 6252696 Retrogressive rotational failure
0.0338 719 205 10U 653493 6218800 Multi-level rotational failure
Pond location on slide
Below small scarp (in debris)
Base of large scarp (intact), in debris apron
Debris apron near river
Within small scarp (in debris)
Above small scarp (in debris)
Debris apron - near creek
Below headscarp (large, intact)
Below headscarp (large, intact), above small scarp (intact)
Below headscarp (large, intact)
Above small scarp (intact)
Base of headscarp (small, intact)
Debris apron
Above small scarp (intact)
Above small scarp (intact)
Below small scarp (in debris)
Within headscarp (small, intact)
Above small scarp (intact), near river
Base of small scarp (in debris), by creek
Below small scarp (in debris)
Within small scarp (in debris)
Above small scarp (in debris)
Above small scarp (intact)
Just above small retrogressive scarp (intact)
Below small scarp (intact)
Debris apron
Below headscarp (small, intact)
Below small scarp (intact)
Below headscarp (small, intact)
Within small scarp (in debris)
Top of slide along west sidescarp
Below headscarp (large, intact)
Base of headscarp (small, intact)

286

Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Active
Active
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Body
Toe
Toe
Body
Body
Toe
Head
Body
Head
Body
Head
Toe
Body
Body
Body
Head
Body
Toe
Body
Body
Body
Body
Body
Body
Toe
Body
Toe
Body
Body
Head
Body
Head

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.034 190 10 10U 652580 6220750 Retrogressive rotational failure
0.0343 292 84 10U 612699 6237705 Retrogressive rotational failure
0.0346 731 14 10U 655144 6217752 Compound failure
0.0347 436 115 10U 606706 6242217 Rotational failure
0.035 275 83 10U 611833 6238026 Retrogressive rotational failure
0.0351 525 142 10U 630879 6227122 Retrogressive rotational failure
0.0354 741 212 10U 656103 6216700 Retrogressive rotational failure
0.0356 227
4 10U 637998 6220826 Multi-level rotational failure
0.0356 455 122 10U 619929 6209531 Mobile flow
0.0357 754 215 10U 664950 6210554 Multi-level rotational failure
0.0357 17
57 10U 645882 6252696 Retrogressive rotational failure
0.0361 738 15 10U 655647 6215120 Retrogressive rotational failure
0.0363 692 201 10U 655452 6222977 Retrogressive rotational failure
0.0365 279 83 10U 611833 6238026 Retrogressive rotational failure
0.0367 513 141 10U 635411 6231647 Retrogressive rotational failure
0.0375 510 140 10U 634447 6231204 Retrogressive rotational failure
0.0375 676 195 10U 655685 6231888 Rotational failure
0.0379 621 176 10U 638057 6276259 Retrogressive rotational failure
0.0382 707 10 10U 652580 6220750 Retrogressive rotational failure
0.0384 488 132 10U 633416 6298822 Rotational failure
0.0384 541
3 10U 635265 6222327 Retrogressive rotational failure
0.0385 313 89 10U 583121 6249823 Multi-level rotational failure
0.0385 46
71 10U 660889 6229983 Shallow retrogressive failure
0.0387 181 17 10U 672017 6212944 Rotational failure
0.0389 549 145 10U 634428 6225821 Multi-level rotational failure
0.0389 717 204 10U 654675 6219635 Retrogressive rotational failure
0.0397 650 188 10U 655498 6235114 Retrogressive rotational failure
0.0398 585 47 10U 640425 6241038 Retrogressive rotational failure
0.04
583 161 10U 639518 6240605 Retrogressive rotational failure
0.0402 605 53 10U 641413 6249461 Retrogressive rotational failure
0.0405
3
57 10U 645882 6252696 Retrogressive rotational failure
0.0405 441 117 10U 605801 6243804 Rotational failure
Pond location on slide
Below small scarp (intact)
Just below small scarp (intact)
Below small scarp (in debris)
Near creek/river
At base of large scarp (intact)
Base of headscarp (small, intact)
Bottom of slide, by creek
Debris apron
Near top of slide
Base of headscarp (small, intact)
Just above small scarp (in debris)
Above/within small scarp (in debris)
Bottom of slide
Toe
Flat area midway down slide
Below small scarp (in debris)
Below headscarp (small, intact)
Below headscarp (large, intact)
Below small scarp (in debris)
Below large scarp (intact), near river
Below small scarp (in debris)
Debris apron next to river
Below headscarp (small, intact)
Below small scarp (intact)
Debris apron
Above small scarp (intact), below headscarp (small, intact)
Below headscarp (large, intact)
Halfway down slide
Below small scarp (intact), near river
Below headscarp (large, intact)
Below small scarp (intact)
Below headscarp (large, intact), above small scarp (in debris)

287

Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Toe
Body
Body
Toe
Body
Head
Toe
Toe
Body
Head
Body
Body
Toe
Toe
Toe
Body
Body
Head
Body
Toe
Body
Toe
Body
Body
Toe
Head
Head
Body
Toe
Body
Toe
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0406 307 87 10U 580867 6250762 Compound failure
0.0406 201 13 10U 656024 6217434 Compound failure
0.0409 736 210 10U 655750 6214657 Rotational failure
0.0414 60
36 10U 658291 6231793 Retrogressive rotational failure
0.0414 22
57 10U 645882 6252696 Retrogressive rotational failure
0.0415 129 64 10U 641133 6264748 Retrogressive rotational failure
0.0417 688 200 10U 658361 6228323 Multi-level rotational failure
0.0418 598 164 10U 640997 6242676 Rotational failure
0.0419 284 83 10U 611833 6238026 Retrogressive rotational failure
0.0419 570 152 10U 637789 6222601 Retrogressive rotational failure
0.0421 336 92 10U 584580 6235119 Rotational failure
0.0422 788 236 10U 675638 6221766 Multi-level rotational failure
0.0425 101 45 10U 645103 6237317 Retrogressive rotational failure
0.0427 504 137 10U 628079 6232225 Multi-level rotational failure
0.0428 199 12 10U 656272 6218290 Rotational failure
0.0429 306 87 10U 580867 6250762 Compound failure
0.0435 81
40 10U 648407 6238441 Multi-level rotational failure
0.0435 568 154 10U 636856 6222710 Compound failure
0.0436 387 103 10U 589643 6282999 Rotational failure
0.0437 141 74 10U 636436 6241818 Compound failure
0.0438 626 178 10U 637541 6284374 Retrogressive rotational failure
0.0439 39
28 10U 660309 6228226 Compound failure
0.044 543
3 10U 635265 6222327 Retrogressive rotational failure
0.0442 35
27 10U 661452 6225535 Compound failure
0.0443 85
41 10U 647664 6238258 Retrogressive rotational failure
0.0446 785 235 10U 685591 6226178 Multi-level rotational failure
0.0447 791 238 10U 679053 6219617 Rotational failure
0.0451 492 133 10U 633983 6299243 Rotational failure
0.0452 397 105 10U 607741 6250210 Compound failure
0.0452 712 10 10U 652580 6220750 Retrogressive rotational failure
0.0456 461 122 10U 619929 6209531 Mobile flow
0.0456 182 17 10U 672017 6212944 Rotational failure
Pond location on slide
Just below headscarp (small, intact)
Below small scarp (in debris)
Below headscarp (small, intact), next to north sidescarp
Toe
Below small scarp (in debris)
Below headscarp (large, intact)
Above small scarp (intact)
Above small scarp (in debris)
At bottom of slide
Base of small scarp (intact)
Within small scarp (in debris)
Debris apron at base of slide
Just above small scarp (intact)
Above small scarp (intact), near east sidescarp
Above small scarp (in debris)
Just below headscarp (small, intact)
Debris apron below large scarp (intact)
Above small scarp (intact), near east sidescarp
Debris apron right by river
Below headscarp (large, intact)
Below small scarp (intact), near east sidescarp
Toe
Below small scarp (in debris)
Body
Below small scarp (intact)
Base of large scarp (intact), in debris apron
Debris apron
Debris apron
Debris apron near river
Next to small scarp (in debris)
Top of slide along west sidescarp
Below small scarp (intact)

288

Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Head
Body
Head
Toe
Toe
Head
Body
Body
Toe
Body
Body
Toe
Body
Body
Body
Head
Toe
Body
Toe
Body
Toe
Toe
Body
Body
Body
Toe
Body
Toe
Toe
Head
Head
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0457 721 207 10U 655366 6218822 Retrogressive rotational failure
0.0458 159 63 10U 656420 6272207 Rotational failure
0.0462 740 211 10U 655836 6216390 Retrogressive rotational failure
0.0462 708 10 10U 652580 6220750 Retrogressive rotational failure
0.0463 192 10 10U 652580 6220750 Retrogressive rotational failure
0.0465 453 122 10U 619929 6209531 Mobile flow
0.0466 312 88 10U 582374 6251325 Rotational failure
0.047 281 83 10U 611833 6238026 Retrogressive rotational failure
0.0472 223
5 10U 637848 6220421 Retrogressive rotational failure
0.0473 162 22 10U 668246 6222087 Compound failure
0.0473 768 227 10U 673362 6239998 Retrogressive rotational failure
0.0473 390 103 10U 589643 6282999 Rotational failure
0.0474 700 10 10U 652580 6220750 Retrogressive rotational failure
0.0474 152 61 10U 648617 6262829 Retrogressive rotational failure
0.0474 553 146 10U 635074 6225748 Multi-level rotational failure
0.0476 792 89 10U 583121 6249823 Multi-level rotational failure
0.0478 252
1 10U 631931 6221078 Retrogressive rotational failure
0.048 622 176 10U 638057 6276259 Retrogressive rotational failure
0.0485 188 10 10U 652580 6220750 Retrogressive rotational failure
0.0488 442 118 10U 605649 6243226 Rotational failure
0.0491 388 103 10U 589643 6282999 Rotational failure
0.0494 268 83 10U 611833 6238026 Retrogressive rotational failure
0.0498 486 130 10U 633609 6300494 Compound failure
0.0498 225
5 10U 637848 6220421 Retrogressive rotational failure
0.0498 534
1 10U 631931 6221078 Retrogressive rotational failure
0.0499 711 10 10U 652580 6220750 Retrogressive rotational failure
0.0499 454 122 10U 619929 6209531 Mobile flow
0.05
26
60 10U 647207 6252566 Rotational failure
0.0501 120 54 10U 642211 6250812 Rotational failure
0.0506 651 188 10U 655498 6235114 Retrogressive rotational failure
0.0507 195 11 10U 656217 6219302 Retrogressive rotational failure
0.0508 361 98 10U 598431 6246464 Compound failure
Pond location on slide
Base of headscarp (small, intact)
Below small scarp (in debris)
Bottom of slide, near west sidescarp
Below small scarp (in debris)
Below small scarp (intact)
Near base of slide
Within debris of small scarp (in debris)
Toe
Just above small scarp (in debris)
Above small scarp (intact)
Base of headscarp (small, intact)
Debris apron near river
Below small scarp (in debris)
In fresh material below small scarp (in debris)
Above small scarp (in debris)
Debris apron next to river
Above backtilt
Within small scarp (intact)
Below small scarp (intact)
Just above small scarp (intact)
Debris apron
Above large scarp (intact), below small scarp (in debris)
Within small scarp (intact)
Below headscarp (small, intact)
Below small scarp (in debris)
Below headscarp (small, intact)
Near centre of slide
Debris apron
In debris apron at base of headscarp (large, intact)
Near creek
Just below small scarp (in debris)
Just below headscarp (small, intact)

289

Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Active
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Head
Toe
Toe
Body
Toe
Body
Toe
Toe
Body
Body
Head
Toe
Body
Toe
Head
Toe
Body
Body
Toe
Body
Body
Body
Body
Head
Body
Head
Body
Toe
Body
Toe
Body
Head

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0508 673 194 10U 656093 6232341 Retrogressive rotational failure
0.0512 431 114 10U 607383 6243249 Retrogressive rotational failure
0.0513 265 83 10U 611833 6238026 Retrogressive rotational failure
0.0516 769 227 10U 673362 6239998 Retrogressive rotational failure
0.0522 205 14 10U 655144 6217752 Compound failure
0.0525 726 208 10U 655089 6218426 Retrogressive rotational failure
0.0528 149 76 10U 633480 6244644 Retrogressive rotational failure
0.0528 775 230 10U 673848 6240822 Rotational failure
0.0529 729 14 10U 655144 6217752 Compound failure
0.053 564
4 10U 637998 6220826 Multi-level rotational failure
0.0531 135 70 10U 631913 6311160 Retrogressive rotational failure
0.0533 269 83 10U 611833 6238026 Retrogressive rotational failure
0.0535 89
43 10U 646755 6238196 Retrogressive rotational failure
0.0539 463 122 10U 619929 6209531 Mobile flow
0.0539 507 139 10U 633271 6230890 Compound failure
0.054 528 143 10U 624881 6220596 Retrogressive rotational failure
0.0547 118 53 10U 641413 6249461 Retrogressive rotational failure
0.0547 285 83 10U 611833 6238026 Retrogressive rotational failure
0.0548 83
40 10U 648407 6238441 Multi-level rotational failure
0.0551 540
3 10U 635265 6222327 Retrogressive rotational failure
0.0551 443 119 10U 599356 6238157 Multi-level rotational failure
0.0552 439 116 10U 606417 6242052 Rotational failure
0.056 508 140 10U 634447 6231204 Retrogressive rotational failure
0.0563 311 88 10U 582374 6251325 Rotational failure
0.0565 642 186 10U 655856 6238250 Retrogressive rotational failure
0.0565 734 210 10U 655750 6214657 Rotational failure
0.0567 697 10 10U 652580 6220750 Retrogressive rotational failure
0.0569 458 122 10U 619929 6209531 Mobile flow
0.0571 314 89 10U 583121 6249823 Multi-level rotational failure
0.0576 364 99 10U 598103 6246815 Rotational failure
0.0578 293 84 10U 612699 6237705 Retrogressive rotational failure
0.0579 95
45 10U 645103 6237317 Retrogressive rotational failure
Pond location on slide
Below headscarp (large, intact)
Below large scarp (intact)
Above large scarp (intact)
Base of headscarp (small, intact)
Above small scarp (in debris)
Base of headscarp (small, intact)
Just above small scarp (in debris)
Within headscarp (small, intact)
Above small scarp (in debris)
Base of small scarp (intact)
Below headscarp (small, intact)
At base of headscarp (large, intact)
Just below small scarp (intact)
Top of slide near west sidescarp
Above small scarp (in debris)
Below small scarp (intact), near north sidescarp
Just below headscarp (large, intact)
At bottom of slide
Just below large scarp (intact)
Above small scarp (in debris)
Base of large scarp (intact)
Debris apron
Above small scarp (intact)
Just below small scarp (in debris)
Below headscarp (large, intact)
Just above small scarp (in debris)
Above small scarp (in debris)
Near top of slide
Debris apron next to river
At base of headscarp (large, intact), by sidescarp
Within small scarp (in debris)
Below small scarp (intact)

290

Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Active
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Body
Body
Body
Head
Body
Head
Body
Head
Body
Body
Body
Body
Body
Head
Body
Toe
Body
Toe
Body
Body
Body
Toe
Body
Toe
Head
Body
Head
Body
Toe
Head
Body
Toe

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0587 674 194 10U 656093 6232341 Retrogressive rotational failure
0.0588 451 122 10U 619929 6209531 Mobile flow
0.0589 645 186 10U 655856 6238250 Retrogressive rotational failure
0.0593 337 92 10U 584580 6235119 Rotational failure
0.0594 53
71 10U 660889 6229983 Shallow retrogressive failure
0.0598 733 209 10U 655227 6215278 Retrogressive rotational failure
0.06
220
4 10U 637998 6220826 Multi-level rotational failure
0.0603 632 181 10U 654030 6269428 Rotational failure
0.0608 298 85 10U 612971 6237140 Retrogressive rotational failure
0.061
41
31 10U 657777 6229034 Compound failure
0.0611 143 75 10U 633471 6243680 Retrogressive rotational failure
0.0621 466 85 10U 612971 6237140 Retrogressive rotational failure
0.0622 79
40 10U 648407 6238441 Multi-level rotational failure
0.0628 702 10 10U 652580 6220750 Retrogressive rotational failure
0.0629 782 234 10U 684625 6226289 Retrogressive rotational failure
0.0633 448 122 10U 619929 6209531 Mobile flow
0.0644 55
72 10U 659938 6229825 Retrogressive rotational failure
0.065 339 94 10U 586742 6231586 Rotational failure
0.0651 728 14 10U 655144 6217752 Compound failure
0.0651 759 220 10U 673724 6221673 Multi-level rotational failure
0.0656 434 114 10U 607383 6243249 Retrogressive rotational failure
0.0658 383 101 10U 598608 6247865 Compound failure
0.066 624 177 10U 638157 6284497 Rotational failure
0.0662 575 157 10U 643272 6221186 Retrogressive rotational failure
0.0665 310 87 10U 580867 6250762 Compound failure
0.0667 119 53 10U 641413 6249461 Retrogressive rotational failure
0.0669 765 222 10U 672726 6234762 Retrogressive rotational failure
0.0673 34
27 10U 661452 6225535 Compound failure
0.0679 45
71 10U 660889 6229983 Shallow retrogressive failure
0.068 217
4 10U 637998 6220826 Multi-level rotational failure
0.0686 384 102 10U 596707 6242884 Rotational failure
0.069 222
5 10U 637848 6220421 Retrogressive rotational failure
Pond location on slide
Just below small scarp (in debris)
Centre of slide
Below large headscarp (intact)
Just below headscarp (large, intact)
Just below small scarp (intact)
Base of small scarp (in debris)
Below small scarp (intact)
Bottom of slide
Below small scarp (in debris)
Below small scarp (intact)
Just below small scarp (intact)
Below small scarp (in debris)
Debris apron
Above small scarp (in debris)
Below/within small scarp (intact)
Near base of slide by west sidescarp
Near base of headscarp (large), above small scarp (in debris)
At base of headscarp (large, intact)
Bottom of slide, near north sidescarp
Above small scarp (intact)
Next to south sidescarp
Debris apron below headscarp (small, intact)
Base of small scarp (intact)
Below large scarp (intact)
Debris apron, just above large scarp (small, intact)
Below headscarp (large, intact)
Next to creek
Body
Below headscarp (small, intact)
Below headscarp (large, intact)
Above small scarp (in debris)
Below headscarp (small, intact)

291

Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Active
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Body
Body
Head
Body
Toe
Body
Body
Toe
Toe
Toe
Body
Body
Toe
Body
Body
Body
Body
Head
Toe
Body
Body
Body
Toe
Head
Body
Body
Toe
Body
Body
Head
Body
Head

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0691 24
59 10U 646337 6252998 Retrogressive rotational failure
0.0702 196 11 10U 656217 6219302 Retrogressive rotational failure
0.0705 743 212 10U 656103 6216700 Retrogressive rotational failure
0.0707 301 85 10U 612971 6237140 Retrogressive rotational failure
0.0709 559 149 10U 635558 6216354 Rotational failure
0.0711 180 17 10U 672017 6212944 Rotational failure
0.0721 324 90 10U 584584 6239849 Retrogressive rotational failure
0.0723 569 152 10U 637789 6222601 Retrogressive rotational failure
0.0726 435 115 10U 606706 6242217 Rotational failure
0.0732 27
60 10U 647207 6252566 Rotational failure
0.0733 714 202 10U 654150 6220760 Retrogressive rotational failure
0.0739 108 48 10U 641990 6244350 Rotational failure
0.0747 33
26 10U 661704 6225287 Shallow retrogressive failure
0.075 693 201 10U 655452 6222977 Retrogressive rotational failure
0.0751 793 90 10U 584584 6239849 Retrogressive rotational failure
0.0754 405 106 10U 607981 6249961 Rotational failure
0.0755 685 33 10U 657471 6230119 Compound failure
0.0767 670 194 10U 656093 6232341 Retrogressive rotational failure
0.0771 93
45 10U 645103 6237317 Retrogressive rotational failure
0.0777 643 186 10U 655856 6238250 Retrogressive rotational failure
0.0779 710 10 10U 652580 6220750 Retrogressive rotational failure
0.0782 433 114 10U 607383 6243249 Retrogressive rotational failure
0.079 140 74 10U 636436 6241818 Compound failure
0.0793 91
43 10U 646755 6238196 Retrogressive rotational failure
0.0798 636 182 10U 653926 6268718 Retrogressive rotational failure
0.0809 54
71 10U 660889 6229983 Shallow retrogressive failure
0.0816 13
58 10U 646995 6252365 Compound failure
0.0816 438 115 10U 606706 6242217 Rotational failure
0.0817 663 39 10U 651511 6236995 Rotational failure
0.0821 437 115 10U 606706 6242217 Rotational failure
0.0823 699 10 10U 652580 6220750 Retrogressive rotational failure
0.0836 355 96 10U 597364 6244753 Compound failure
Pond location on slide
Toe
Just below small scarp (in debris)
Above small scarp (intact)
Below small scarp (in debris)
Within small scarp (in debris)
Within headscarp (small, intact)
Just above small retrogressive scarp (intact)
Base of headscarp (small, intact)
Above small scarp (intact)
Debris apron
Within small scarp (intact)
At base of headscarp (large, intact)
In debris apron
Within small scarp (in debris)
Just above small retrogressive scarp (intact)
Debris apron near river
Above large scarp (intact), near river
Below headscarp (large, intact)
Below small scarp (intact)
Base of headscarp (large, intact)
Below headscarp (small, intact)
Above small scarp (intact)
At base of headscarp (large, intact)
Just below small scarp (intact)
Just above small scarp (intact), near river
Just above small scarp (intact)
Below small scarp (in debris)
Near west sidescarp
Debris apron
Below headscarp (large, intact)
Below small scarp (in debris)
Just below small scarp (in debris)

292

Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Toe
Body
Body
Toe
Body
Head
Body
Head
Body
Toe
Toe
Head
Toe
Toe
Body
Body
Body
Head
Toe
Head
Head
Body
Body
Toe
Toe
Toe
Toe
Body
Toe
Body
Body
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.0837 288 83 10U 611833 6238026 Retrogressive rotational failure
0.086 444 120 10U 603024 6239487 Rotational failure
0.0863 601 50 10U 641564 6246268 Compound failure
0.0868 151 78 10U 637889 6241550 Rotational failure
0.0874 75
44 10U 646758 6237400 Retrogressive rotational failure
0.0874 760 221 10U 670755 6236278 Rotational failure
0.0876 783 234 10U 684625 6226289 Retrogressive rotational failure
0.0877 385 103 10U 589643 6282999 Rotational failure
0.0881 407 107 10U 607530 6250024 Compound failure
0.089 343 94 10U 586742 6231586 Rotational failure
0.0915 123 55 10U 641685 6251522 Multi-level rotational failure
0.0923 779 231 10U 680412 6226004 Multi-level rotational failure
0.0927 391 103 10U 589643 6282999 Rotational failure
0.0928 287 83 10U 611833 6238026 Retrogressive rotational failure
0.093 161 22 10U 668246 6222087 Compound failure
0.0931 11
58 10U 646995 6252365 Compound failure
0.0942 753 215 10U 664950 6210554 Multi-level rotational failure
0.0944 709 10 10U 652580 6220750 Retrogressive rotational failure
0.0947 539
3 10U 635265 6222327 Retrogressive rotational failure
0.095 321 90 10U 584584 6239849 Retrogressive rotational failure
0.0958 273 83 10U 611833 6238026 Retrogressive rotational failure
0.0976 187 10 10U 652580 6220750 Retrogressive rotational failure
0.0995 200 13 10U 656024 6217434 Compound failure
0.0996 29
23 10U 662280 6223095 Rotational failure
0.1017 144 75 10U 633471 6243680 Retrogressive rotational failure
0.1019 563
4 10U 637998 6220826 Multi-level rotational failure
0.1028 745 212 10U 656103 6216700 Retrogressive rotational railure
0.1031 690 30 10U 658107 6228756 Compound failure
0.1033 197 12 10U 656272 6218290 Rotational failure
0.1035 672 194 10U 656093 6232341 Retrogressive rotational failure
0.1045 122 55 10U 641685 6251522 Multi-level rotational failure
0.1045 595 162 10U 639763 6241926 Retrogressive rotational failure
Pond location on slide
At bottom of slide
Just below headscarp (large, intact)
Within small scarp (in debris)
Just below headscarp (small)
Below small scarp (in debris)
Base of headscarp (large, intact)
Below small scarp (intact)
At base of headscarp (small, intact)
Below headscarp (small, intact) near south sidescarp
Debris apron
Debris apron
Debris apron at base of large scarp (intact)
Debris apron near river
At bottom of slide
Below small scarp (intact)
Below headscarp (small, intact)
Base of headscarp (small, intact)
Below small scarp (in debris)
Just below small scarp (intact)
Just above small retrogressive scarp (intact)
Below large scarp (intact)
Below small scarp (intact)
Above small scarp (in debris), just below headscarp (small, intact)
Above debris apron
Below small scarp (intact), next to small scarp (in debris)
Below small scarp (intact)
Below headscarp (large, intact)
Below small scarp (in debris)
Above small scarp (in debris)
Just below small scarp (in debris)
Debris apron
Below small scarp (in debris)

293

Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Abandoned
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Abandoned
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Toe
Head
Body
Head
Body
Head
Body
Head
Body
Body
Toe
Toe
Toe
Toe
Body
Head
Head
Body
Body
Body
Body
Toe
Head
Body
Body
Body
Body
Toe
Body
Body
Toe
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.1049 794 97 10U 597665 6245826 Rotational failure
0.1055 648 187 10U 655524 6236858 Multi-level rotational failure
0.1057 215 86 10U 639774 6220198 Retrogressive rotational failure
0.1074 581 161 10U 639518 6240605 Retrogressive rotational failure
0.1086 612 172 10U 641260 6254208 Rotational failure
0.1091 462 122 10U 619929 6209531 Mobile flow
0.1094 786 235 10U 685591 6226178 Multi-level rotational failure
0.1096 482 128 10U 632934 6302871 Rotational failure
0.1101 216
4 10U 637998 6220826 Multi-level rotational failure
0.1103 226
4 10U 637998 6220826 Multi-level rotational failure
0.1106 654 189 10U 654871 6235759 Retrogressive rotational failure
0.1112 752 215 10U 664950 6210554 Multi-level rotational failure
0.1116 778 227 10U 673362 6239998 Retrogressive rotational failure
0.1121 750 214 10U 664400 6210645 Multi-level rotational failure
0.1121 776 230 10U 673848 6240822 Rotational failure
0.1129 565 152 10U 637789 6222601 Retrogressive rotational failure
0.1136
6
59 10U 646337 6252998 Retrogressive rotational failure
0.1139 577 158 10U 643429 6220210 Rotational failure
0.1152 720 206 10U 655413 6219392 Retrogressive rotational failure
0.1161 323 90 10U 584584 6239849 Retrogressive rotational failure
0.117 762 81 10U 673096 6235267 Retrogressive rotational failure
0.1173 590 47 10U 640425 6241038 Retrogressive rotational failure
0.1194 15
57 10U 645882 6252696 Retrogressive rotational failure
0.1195 449 122 10U 619929 6209531 Mobile flow
0.1205 92
45 10U 645103 6237317 Retrogressive rotational failure
0.1218 145 75 10U 633471 6243680 Retrogressive rotational failure
0.1223 19
57 10U 645882 6252696 Retrogressive rotational failure
0.1229 256
2 10U 633364 6222821 Compound failure
0.1233 142 74 10U 636436 6241818 Compound failure
0.1233 432 114 10U 607383 6243249 Retrogressive rotational failure
0.1235 331 91 10U 585686 6240356 Compound failure
0.1238 12
57 10U 645882 6252696 Retrogressive rotational failure
Pond location on slide
At base of headscarp (large, intact)
Below headscarp (large, intact)
Below small scarp (intact)
Below small scarp (in debris), near river
Debris apron beside creek
Top of slide near west sidescarp
Base of large scarp (intact), in debris apron
Base of headscarp (large, intact)
Below headscarp (large, intact), above small scarp (intact)
Debris apron
Below headscarp (large, intact)
Base of headscarp (small, intact)
Base of slide, near creek
Below headscarp (small, intact)
Within headscarp (small, intact)
Above small scarp (intact), next to east sidescarp
Below small scarp (in debris)
Below small scarp (in debris), by north sidescarp
Below headscarp (small, intact)
Below small scarp (in debris)
Above small scarp (intact), near creek
Halfway down slide
Just above small scarp (in debris)
Near base of slide by west sidescarp
Within large scarp (intact), below headscarp (small, intact)
Just below small scarp (intact)
Below small scarp (in debris)
Below small scarp (in debris)
Below small scarp (intact)
Below large scarp (intact)
Just below small scarp (intact)
Just above small scarp (in debris)

294

Dormant
Dormant
Dormant
Dormant
Active
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Active
Dormant
Active
Active
Abandoned
Dormant
Dormant
Dormant
Active

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Head
Head
Head
Toe
Toe
Head
Toe
Body
Head
Toe
Body
Head
Toe
Head
Head
Body
Body
Toe
Head
Toe
Body
Body
Body
Body
Head
Body
Body
Toe
Body
Body
Body
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.1244 467 123 10U 615696 6235258 Multi-level rotational failure
0.1269 573 156 10U 643924 6221112 Rotational failure
0.1275 479 127 10U 632593 6304450 Rotational failure
0.1281 354 96 10U 597364 6244753 Compound failure
0.1284 412 109 10U 608185 6248414 Compound failure
0.131 107 48 10U 641990 6244350 Rotational failure
0.131 127 56 10U 644848 6253033 Retrogressive rotational failure
0.1311 671 194 10U 656093 6232341 Retrogressive rotational failure
0.1317 658 189 10U 654871 6235759 Retrogressive rotational failure
0.1324 212
9 10U 639115 6218246 Rotational failure
0.1338 501 137 10U 628079 6232225 Multi-level rotational failure
0.137 139 73 10U 634577 6242061 Compound failure
0.1383 537
3 10U 635265 6222327 Retrogressive rotational failure
0.1396 358 97 10U 597665 6245826 Rotational failure
0.1409 666 38 10U 652244 6236095 Retrogressive rotational failure
0.1415 771 227 10U 673362 6239998 Retrogressive rotational failure
0.1417 761 222 10U 672726 6234762 Retrogressive rotational failure
0.1418 255
1 10U 631931 6221078 Retrogressive rotational failure
0.1419 228
4 10U 637998 6220826 Multi-level rotational failure
0.1423 665 192 10U 649524 6238522 Rotational failure
0.1425 277 83 10U 611833 6238026 Retrogressive rotational failure
0.1431 224
5 10U 637848 6220421 Retrogressive rotational failure
0.1437 604 167 10U 642365 6247706 Rotational failure
0.1439 578 159 10U 649510 6220866 Rotational failure
0.1441 14
57 10U 645882 6252696 Retrogressive rotational failure
0.1441 580 160 10U 648168 6224229 Compound failure
0.1442 116 53 10U 641413 6249461 Retrogressive rotational failure
0.1448 179 19 10U 675077 6219231 Rotational failure
0.1452 450 122 10U 619929 6209531 Mobile flow
0.1457 591 47 10U 640425 6241038 Retrogressive rotational failure
0.1467
7
59 10U 646337 6252998 Retrogressive rotational failure
0.1476 715 203 10U 654136 6220177 Retrogressive rotational failure
0.1481 703 10 10U 652580 6220750 Retrogressive rotational failure
Debris apron near river
Within small scarp (in debris)
Within small scarp (intact)
Within small scarp (in debris)
Debris apron near river/creek
At base of headscarp (large, intact)
Below small scarp
Below small scarp (in debris)
Above small scarp (intact)
Below small scarp (intact)
Below small scarp (intact), by southwest sidescarp
Below small scarp (in debris)
Above small scarp (intact)
At base of headscarp (large, intact)
Next to river at bottom of slide
Base of headscarp (small, intact)
Above small scarp (in debris), near creek
Above small scarp (intact)
Debris apron
Below headscarp (small, intact)
Below small scarp (in debris)
Between 2 small scarps (in debris)
Bottom of slide, next to creek
Next to west side scarp, near base of slide
At base of headscarp (large, intact)
Above small scarp (in debris)
At base of headscarp (large, intact)
Debris apron
Centre of slide
Below, to the right of small scarp (in debris)
Below headscarp (large, intact)
Above small scarp (in debris), near south sidescarp
Below small scarp (in debris)

Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant

295

Pond location on slide

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Toe
Body
Body
Body
Toe
Head
Toe
Body
Body
Toe
Body
Body
Body
Head
Toe
Head
Toe
Body
Toe
Body
Body
Body
Toe
Toe
Head
Body
Body
Body
Body
Body
Body
Toe
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.1507 620 176 10U 638057 6276259 Retrogressive rotational failure
0.1518 94
45 10U 645103 6237317 Retrogressive rotational failure
0.1518 97
45 10U 645103 6237317 Retrogressive rotational failure
0.1534 189 10 10U 652580 6220750 Retrogressive rotational failure
0.1542 547 145 10U 634428 6225821 Multi-level rotational failure
0.1546 221
4 10U 637998 6220826 Multi-level rotational failure
0.1546 198 12 10U 656272 6218290 Rotational failure
0.1557 290 83 10U 611833 6238026 Retrogressive rotational failure
0.1561 138 73 10U 634577 6242061 Compound failure
0.1578 356 96 10U 597364 6244753 Compound failure
0.1591 291 83 10U 611833 6238026 Retrogressive rotational failure
0.1597 247
1 10U 631931 6221078 Retrogressive rotational failure
0.1607 551 145 10U 634428 6225821 Multi-level rotational failure
0.1614 735 15 10U 655647 6215120 Retrogressive rotational failure
0.1616 579 159 10U 649510 6220866 Rotational failure
0.1622 744 212 10U 656103 6216700 Retrogressive rotational failure
0.1634 684 33 10U 657471 6230119 Compound failure
0.1639 572 155 10U 644195 6222091 Multi-level rotational failure
0.1664 496 75 10U 633471 6243680 Retrogressive rotational failure
0.1668 746 213 10U 659241 6219068 Retrogressive rotational failure
0.1673 567 154 10U 636856 6222710 Compound failure
0.1674 755 216 10U 663258 6218654 Multi-level rotational failure
0.1689
1
15 10U 655647 6215120 Retrogressive rotational failure
0.1718 42
30 10U 658107 6228756 Compound failure
0.1726 242
1 10U 631931 6221078 Retrogressive rotational failure
0.1729 259
3 10U 635265 6222327 Retrogressive rotational failure
0.1733 249
1 10U 631931 6221078 Retrogressive rotational failure
0.1735 669 193 10U 657047 6232296 Retrogressive rotational failure
0.1761 206 16 10U 655340 6214299 Rotational failure
0.1761 276 83 10U 611833 6238026 Retrogressive rotational failure
0.1765 546 145 10U 634428 6225821 Multi-level rotational failure
0.1811 751 214 10U 664400 6210645 Multi-level rotational failure
Pond location on slide
Base of headscarp (large, intact)
Below small scarp (intact)
Below small scarp (intact)
Below small scarp (intact)
Debris apron
Below small scarp (intact)
Just below headscarp (small, intact)
At bottom of slide
Next to small scarp (intact)
Next to north sidescarp
At bottom of slide
Backtilt below small scarp (intact)
Debris apron
Below headscarp (small, intact), along south sidescarp
Above small scarp (intact)
Below headscarp (large, intact)
Below large scarp (intact), near river
Below small scarp (intact)
Below small scarp (intact)
Below small scarp (intact)
Below small scarp (intact), near east sidescarp
Above small scarp (intact), near river
Just below headscarp (small, intact)
Next to small scarp (in debris)
Below small scarp (intact)
Above small scarp, below headscarp (large, intact)
Next to small scarp (in debris)
Below headscarp (large, intact)
On headscarp (small, intact)
Below small scarp (in debris)
Debris apron
Within headscarp (small, intact)

296

Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Abandoned
Dormant
Abandoned
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Head
Toe
Toe
Toe
Toe
Body
Head
Toe
Toe
Body
Toe
Body
Toe
Head
Body
Head
Toe
Body
Body
Toe
Body
Toe
Head
Toe
Body
Head
Body
Head
Head
Body
Toe
Head

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.1825 657 189 10U 654871 6235759 Retrogressive rotational failure
0.1828 121 55 10U 641685 6251522 Multi-level rotational failure
0.1841 76
44 10U 646758 6237400 Retrogressive rotational failure
0.1863 117 53 10U 641413 6249461 Retrogressive rotational failure
0.1874 596 163 10U 640344 6243435 Multi-level rotational failure
0.189 243
1 10U 631931 6221078 Retrogressive rotational failure
0.1897 704 10 10U 652580 6220750 Retrogressive rotational failure
0.1905 261
3 10U 635265 6222327 Retrogressive rotational failure
0.191 175 20 10U 679534 6219946 Rotational failure
0.1911 713 10 10U 652580 6220750 Retrogressive rotational failure
0.1937 113 50 10U 641564 6246268 Compound failure
0.197 158 63 10U 656420 6272207 Rotational failure
0.1975 115 52 10U 641810 6248528 Rotational failure
0.1978 251
1 10U 631931 6221078 Retrogressive rotational failure
0.2006 747 213 10U 659241 6219068 Retrogressive rotational failure
0.2009 640 185 10U 656213 6239712 Rotational failure
0.2012 137 73 10U 634577 6242061 Compound failure
0.2022 241
1 10U 631931 6221078 Retrogressive rotational failure
0.2033 789 237 10U 680337 6220929 Retrogressive rotational failure
0.209 299 85 10U 612971 6237140 Retrogressive rotational failure
0.2105 599 48 10U 641990 6244350 Rotational failure
0.2115 258
3 10U 635265 6222327 Retrogressive rotational failure
0.2182 594 162 10U 639763 6241926 Retrogressive otational failure
0.222 574 157 10U 643272 6221186 Retrogressive rotational failure
0.2223 428 114 10U 607383 6243249 Retrogressive rotational failure
0.2225 516 141 10U 635411 6231647 Retrogressive rotational failure
0.2228 533
1 10U 631931 6221078 Retrogressive rotational failure
0.2237 576 157 10U 643272 6221186 Retrogressive rotational failure
0.2251 618 174 10U 642740 6258340 Retrogressive rotational failure
0.2332 240
1 10U 631931 6221078 Retrogressive rotational failure
0.234 136 70 10U 631913 6311160 Retrogressive rotational failure
0.2342 610 170 10U 642859 6253490 Retrogressive rotational failure
0.236 545
3 10U 635265 6222327 Retrogressive rotational failure
Pond location on slide
Below headscarp (large, intact)
Debris apron
Next to small scarp (in debris)
At base of headscarp (large, intact)
Below large scarp (intact)
Below small scarp (intact)
Below small scarp (in debris)
Just below small scarp near headscarp (large, intact)
Below headscarp (large, intact)
Below small scarp (intact)
Below small scarp (intact)
Just above small scarp (in debris)
In debris apron below headscarp (large, intact)
Next to small scarp (intact) and backtilt
Below small scarp (intact)
Base of headscarp (large, intact)
Bottom of slide
Below small scarp (intact)
Just above large scarp (intact), near river
Below small scarp (in debris)
Base of headscarp (large, intact), next to small scarp (in debris)
Above small scarp, below headscarp (large, intact)
Below headscarp (large, intact)
Above small scarp (in debris)
Below small scarp (intact)
Flat area midway down slide
Between 2 small scarps (intact), by river
Above small scarp (in debris)
Base of headscarp (large, intact), above small scarp (intact)
Below small scarp (intact)
Body
Above small scarp (intact)
Below, to east of small scarp (in debris)

297

Dormant
Abandoned
Dormant
Dormant
Dormant
Dormant
Dormant
Abandoned
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Abandoned
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Head
Toe
Body
Body
Body
Body
Body
Head
Body
Body
Body
Body
Body
Body
Toe
Body
Toe
Body
Toe
Toe
Body
Head
Body
Toe
Body
Toe
Toe
Body
Body
Body
Body
Body
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.2365 64
38 10U 652244 6236095 Retrogressive rotational failure
0.2384 538
3 10U 635265 6222327 Retrogressive rotational failure
0.2384 691 201 10U 655452 6222977 Retrogressive rotational failure
0.2407 608 168 10U 642871 6251602 Rotational failure
0.2466
5
59 10U 646337 6252998 Retrogressive rotational failure
0.249
63
38 10U 652244 6236095 Retrogressive rotational failure
0.2519 106 48 10U 641990 6244350 Rotational failure
0.2536 319 90 10U 584584 6239849 Retrogressive rotational failure
0.2545 109 48 10U 641990 6244350 Rotational failure
0.2559 619 175 10U 641263 6274322 Retrogressive rotational failure
0.2574 246
1 10U 631931 6221078 Retrogressive rotational failure
0.2578 475 126 10U 631665 6307193 Rotational failure
0.2581 236
1 10U 631931 6221078 Retrogressive rotational failure
0.2585 146 76 10U 633480 6244644 Retrogressive rotational failure
0.2613 770 227 10U 673362 6239998 Retrogressive rotational failure
0.2664 28
57 10U 645882 6252696 Retrogressive rotational failure
0.2759 155 62 10U 650179 6264258 Compound failure
0.2824 552 146 10U 635074 6225748 Multi-level rotational failure
0.2827 218
4 10U 637998 6220826 Multi-level rotational failure
0.2869 266 83 10U 611833 6238026 Retrogressive rotational failure
0.2874 43
71 10U 660889 6229983 Shallow retrogressive failure
0.2893 556 148 10U 635615 6217619 Retrogressive rotational failure
0.2906 304 87 10U 580867 6250762 Compound failure
0.2919 174 81 10U 673096 6235267 Retrogressive rotational failure
0.2934 203 14 10U 655144 6217752 Compound failure
0.3078 66
38 10U 652244 6236095 Retrogressive rotational failure
0.3181 452 122 10U 619929 6209531 Mobile flow
0.3193 248
1 10U 631931 6221078 Retrogressive rotational failure
0.3212 446 122 10U 619929 6209531 Mobile flow
0.3226 480 127 10U 632593 6304450 Rotational failure
0.3227 334 90 10U 584584 6239849 Retrogressive rotational failure
0.3239 427 114 10U 607383 6243249 Retrogressive rotational failure
0.3251 72
38 10U 652244 6236095 Retrogressive rotational failure
Pond location on slide
Toe
Just below small scarp (intact)
Bottom of slide
Debris apron next to small scarp (intact)
Below small scarp (in debris)
Below small scarp (in debris)
At base of headscarp (large, intact)
Below headscarp (large, intact)
Just above debris apron
Below small scarp (in debris)
Backtilt below small scarp (intact)
Base of headscarp (small, intact)
Below small scarp (intact) - near river
Below small scarp (in debris)
Base of headscarp (small, intact)
Below headscarp (large, intact)
Just below headscarp (small, intact)
Above small scarp (intact)
Below headscarp (large, intact), next to small scarp
Above large scarp (intact)
Just at base of headscarp (small, intact)
Base of headscarp (small, intact)
Within headscarp (small, intact)
Below headscarp (large, intact)
Within small scarp (in debris), just below headscarp (small, intact)
Just below headscarp (large, intact)
Centre of slide
Above backtilt
Near base of slide
Below headscarp (small, intact)
Just above small retrogressive scarp (intact)
Base of headscarp (large, intact)
Just below headscarp (large, intact)

298

Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Active
Dormant
Active
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Toe
Body
Toe
Body
Body
Toe
Body
Head
Body
Toe
Body
Head
Toe
Body
Head
Body
Head
Body
Body
Body
Head
Head
Head
Body
Head
Head
Body
Body
Body
Body
Body
Head
Head

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.328 112 49 10U 641388 6245365 Multi-level rotational failure
0.3531 320 90 10U 584584 6239849 Retrogressive rotational failure
0.3538 234
1 10U 631931 6221078 Retrogressive rotational failure
0.3548 194 11 10U 656217 6219302 Retrogressive rotational failure
0.3553 695 10 10U 652580 6220750 Retrogressive rotational failure
0.3565 114 51 10U 642403 6248773 Rotational failure
0.3582 318 90 10U 584584 6239849 Retrogressive rotational failure
0.3589 239
1 10U 631931 6221078 Retrogressive rotational failure
0.3648 647 186 10U 655856 6238250 Retrogressive rotational failure
0.3731 633 182 10U 653926 6268718 Retrogressive rotational failure
0.3802 68
38 10U 652244 6236095 Retrogressive rotational failure
0.3815 245
1 10U 631931 6221078 Retrogressive rotational failure
0.3828 357 97 10U 597665 6245826 Rotational failure
0.3829 104 47 10U 640425 6241038 Retrogressive rotational failure
0.3898 532
1 10U 631931 6221078 Retrogressive rotational failure
0.397 739 211 10U 655836 6216390 Retrogressive rotational failure
0.4329 493 134 10U 634160 6291151 Rotational failure
0.436 694 159 10U 649510 6220866 Rotational failure
0.4523 270 83 10U 611833 6238026 Retrogressive rotational failure
0.4636 296 84 10U 612699 6237705 Retrogressive rotational failure
0.4648 130 66 10U 640009 6272311 Retrogressive rotational failure
0.4707 70
38 10U 652244 6236095 Retrogressive rotational failure
0.471
10
57 10U 645882 6252696 Retrogressive rotational failure
0.4942
9
59 10U 646337 6252998 Retrogressive rotational failure
0.506 652 189 10U 654871 6235759 Retrogressive rotational failure
0.51
134 70 10U 631913 6311160 Retrogressive rotational failure
0.515 232
6 10U 638745 6221151 Retrogressive rotational failure
0.519 178 18 10U 672960 6219166 Retrogressive rotational failure
0.525 186 10 10U 652580 6220750 Retrogressive rotational failure
0.527 600 48 10U 641990 6244350 Rotational failure
0.528 267 83 10U 611833 6238026 Retrogressive rotational failure
0.532 653 189 10U 654871 6235759 Retrogressive rotational failure
0.547 208
7 10U 637199 6218231 Retrogressive rotational failure
Pond location on slide
Below small scarp (in debris)
Just below headscarp (large, intact)
Above small scarp (intact) next to river
Just above small scarp (in debris)
Below headscarp (small, intact), above small scarp (in debris)
In debris apron below headscarp (large, intact), next to south sidescarp
Below headscarp (large, intact)
Backtilt below headscarp (large, intact)
Below small scarp (in debris), right by creek
Above small scarp (intact)
Above small scarp (intact)
Backtilt below headscarp (large, intact)
Just below headscarp (large, intact)
Below headscarp (large, intact)
Below small scarp (intact), by river
Base of headscarp (large, intact)
Bottom of slide
Below small scarp (in debris)
Below large scarp (intact)
At base of small scarp (intact)
Below headscarp, above small scarp (in debris)
Just below headscarp (large, intact)
Between 2 small scarps (in debris)
Below small scarp (in debris)
Below headscarp (large, intact)
Below headscarp (small, intact)
Just above small scarp (in debris)
Above small scarp (in debris)
Above small scarp (in debris), below headscarp (small, intact)
Base of headscarp (large, intact)
Above large scarp (intact)
Below headscarp (large, intact)
Below headscarp (east) - (small, intact)

299

Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Very active
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Body
Head
Toe
Body
Head
Body
Toe
Head
Toe
Body
Body
Head
Body
Head
Toe
Head
Toe
Head
Body
Toe
Body
Head
Body
Body
Body
Head
Body
Body
Head
Body
Body
Body
Body

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
0.551 262 82 10U 564797 6216748 Rotational failure
0.552
71
38 10U 652244 6236095 Retrogressive rotational failure
0.565 214
9 10U 639115 6218246 Rotational failure
0.578 468 124 10U 621307 6287220 Rotational failure
0.582 185 10 10U 652580 6220750 Retrogressive rotational failure
0.582
69
38 10U 652244 6236095 Retrogressive rotational failure
0.6
297 85 10U 612971 6237140 Retrogressive rotational failure
0.601 148 76 10U 633480 6244644 Retrogressive rotational failure
0.606 494 135 10U 632986 6245453 Retrogressive rotational failure
0.608 353 95 10U 589923 6234772 Multi-level rotational failure
0.609 244
1 10U 631931 6221078 Retrogressive rotational failure
0.614 787 235 10U 685591 6226178 Multi-level rotational failure
0.617 253
1 10U 631931 6221078 Retrogressive rotational failure
0.617
8
59 10U 646337 6252998 Retrogressive rotational failure
0.627 237
1 10U 631931 6221078 Retrogressive rotational failure
0.637 627 179 10U 636025 6286564 Retrogressive rotational failure
0.64
477 127 10U 632593 6304450 Rotational failure
0.648 555 147 10U 636595 6219522 Multi-level rotational failure
0.661 235
1 10U 631931 6221078 Retrogressive rotational failure
0.668 271 83 10U 611833 6238026 Retrogressive rotational failure
0.672 295 84 10U 612699 6237705 Retrogressive rotational failure
0.675 333 90 10U 584584 6239849 Retrogressive rotational failure
0.758 184 10 10U 652580 6220750 Retrogressive rotational failure
0.815 183 10 10U 652580 6220750 Retrogressive rotational failure
0.858
0
15 10U 655647 6215120 Retrogressive rotational failure
0.874 623 177 10U 638157 6284497 Rotational failure
0.917 495 135 10U 632986 6245453 Retrogressive rotational failure
0.965 103 47 10U 640425 6241038 Retrogressive rotational failure
0.969 628 179 10U 636025 6286564 Retrogressive rotational failure
0.976
65
39 10U 651511 6236995 Rotational failure
0.989 536
3 10U 635265 6222327 Retrogressive rotational failure
1.025
78
38 10U 652244 6236095 Retrogressive rotational failure
1.062 561 150 10U 635070 6214869 Retrogressive rotational failure
Pond location on slide
At base of debris apron
Below headscarp (large, intact) , above small scarp (in debris)
Below small scarp (intact)
Base of slide, near river
Above small scarp (in debris), just below headscarp (backtilt - small, intact)
Just below headscarp (large, intact)
Below small scarp (intact)
Below and within small scarp (in debris)
Below small scarp (in debris)
Below small scarp (intact) near river
Backtilt below small scarp (intact)
Base of large scarp (intact), in debris apron
Above backtilt
Just below headscarp (large, intact)
Backtilt below headscarp (large, intact)
Bottom of slide
Below small scarp (intact)
Debris apron
Backtilt below small scarp (in debris)
Below large scarp (intact)
At base of small scarp (in debris)
Just below headscarp (large, intact)
Above small scarp (in debris), in backtilt below headscarp
Above small scarp (intact)
Just above small scarp (in debris), below headscarp (small, intact)
Above/next to small scarp (intact)
Below small scarp (in debris)
Below small scarp (in debris)
Bottom of slide
Debris apron
Below small scarp (intact)
Toe
Base of headscarp (small, intact)

300

Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Toe
Head
Toe
Toe
Head
Head
Toe
Body
Body
Toe
Body
Toe
Body
Head
Head
Toe
Body
Toe
Body
Body
Toe
Head
Head
Body
Head
Body
Body
Toe
Toe
Toe
Body
Toe
Head

Pond
Pond Slide
Size
Slide UTMs
Slide Type
#
#
(ha)
1.094 272 83 10U 611833 6238026 Retrogressive rotational failure
1.14
487 131 10U 632364 6300428 Rotational failure
1.22
105 48 10U 641990 6244350 Rotational failure
1.255
67
38 10U 652244 6236095 Retrogressive rotational failure
1.279 193 11 10U 656217 6219302 Retrogressive rotational failure
1.314
74
38 10U 652244 6236095 Retrogressive rotational failure
1.355 177 18 10U 672960 6219166 Retrogressive rotational failure
1.371 238
1 10U 631931 6221078 Retrogressive rotational failure
1.48
332 90 10U 584584 6239849 Retrogressive rotational failure
1.572 147 76 10U 633480 6244644 Retrogressive rotational failure
1.654 611 171 10U 641105 6253591 Rotational failure
2.376
73
44 10U 646758 6237400 Retrogressive rotational failure
3.227 629 179 10U 636025 6286564 Retrogressive rotational failure
5.89
207
7 10U 637199 6218231 Retrogressive rotational failure
Pond location on slide
At base of large scarp (intact)
Middle of slide
At base of headscarp (large, intact)
Just below headscarp (large, intact)
Below small scarp (in debris)
Just below headscarp (large, intact)
Below small scarp (in debris)
Backtilt below headscarp (large, intact)
Just below headscarp (large, intact)
Just below headscarp (small, intact), just above small scarp (in debris)
Bottom of slide, partly blocking creek
Just below headscarp (large, intact), above small scarp (in debris)
Bottom of slide
Below headscarp (east) - (small, intact)

301

Dormant
Dormant
Dormant
Dormant
Active
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant
Dormant

Slide
Activity

Appendix 7 Landslide pond descriptions - Mapsheet 94A - cont'd
Geomorphic
location
of pond
Body
Body
Body
Head
Body
Head
Toe
Head
Head
Head
Toe
Body
Toe
Body

Appendix 8 Pond size statistics by landslide type

Slide type
Compound failure
Mobile flow
Multi-level rotational
Retrogressive rotational
Rotational
Shallow retrogressive

Mean
pond size
(ha)
0.054
0.077
0.102
0.197
0.125
0.051

Minimum Maximum Median
pond size pond size pond size
(ha)
(ha)
(ha)
0.003
0.293
0.026
0.005
0.321
0.048
0.004
0.648
0.055
0.001
5.890
0.059
0.003
1.654
0.046
0.003
0.287
0.014

302

Standard Deviation
of pond size
(ha)
0.062
0.089
0.131
0.430
0.234
0.074

Appendix 9 Pond size by geomorphic location per slide type

Ponds on head
Landslide type
Compound failure
Mobile flow
Multi-level rotational failure
Retrogressive rotational failure
Rotational failure
Shallow retrogressive failure
Ponds on body
Landslide type
Compound failure
Mobile flow
Multi-level rotational failure
Retrogressive rotational failure
Rotational failure
Shallow retrogressive failure
Ponds on toe
Landslide type
Compound failure
Mobile flow
Multi-level rotational failure
Retrogressive rotational failure
Rotational failure
Shallow retrogressive failure

Mean
pond size
(ha)

Minimum
pond size
(ha)

Maximum
pond size
(ha)

Median
pond size
(ha)

SD of
pond size
(ha)

0.072
0.053
0.078
0.281
0.082
0.064

0.003
0.025
0.012
0.007
0.004
0.003

0.293
0.109
0.181
1.572
0.436
0.287

0.022
0.033
0.081
0.131
0.065
0.009

0.100
0.030
0.047
0.366
0.090
0.112

0.050
0.085
0.104
0.192
0.123
0.025

0.007
0.005
0.007
0.001
0.003
0.003

0.293
0.321
0.060
5.890
1.654
0.068

0.033
0.046
0.058
0.062
0.042
0.007

0.046
0.100
0.099
0.489
0.222
0.025

0.050
0.000
0.109
0.150
0.159
0.072

0.004
0.000
0.004
0.002
0.007
0.059

0.201
0.000
0.648
0.358
1.654
0.081

0.017
0.000
0.044
0.041
0.047
0.075

0.058
0.000
0.160
0.355
0.310
0.009

303