VEGETATION RECOVERY ON ABANDONED ROAD SEGMENTS OF HIGHWAY 16
IN NORTHWESTERN BRITISH COLUMBIA
by
Kimberley Lutz
B.Sc., University of Alberta, 2010

THESIS SUBMITTED IN PARTIAL FULLFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
NATURAL RESOURCE AND ENVIRONMENTAL STUDIES

UNIVERSITY OF NORTHERN BRITISH COLUMBIA
December 2022
© Kimberley Lutz, 2022

Abstract
Vegetation recovery on abandoned road segments of Highway 16 in northwestern British
Columbia were examined across a climate gradient. The time since road abandonment ranged
from 16 to 57 years on sites sampled. Plant cover on asphalt roads was compared with that
found on gravel road shoulders using paired t-tests. Plant cover by growth form was further
evaluated in response to climate and other environmental predictor variables using multiple
regression ‘best fit’ models. Plant community ordination analysis using nonmetric
multidimensional scaling was conducted to describe patterns across study sites in species
space along environmental gradients. Key drivers of current plant community composition
include time since road abandonment, road substrate type, and several different annual
climate variables. The best predictor of vascular plant cover and total plant cover was time
since road abandonment, but plant community composition was strongly driven by the coastto-interior climate gradient. Non-vascular cover was more abundant on asphalt roads
compared to gravel substrates. Woody plant cover was greatest on gravel shoulders compared
to gravel or asphalt road centers. Exotic plant cover was negatively correlated with mean
annual relative humidity. Plant diversity and species richness were not driven by the climate
gradient but instead reflected more site-specific environmental variables. Primary succession
on abandoned roads in this study area may be constrained by continued anthropogenic
disturbance post-abandonment.

ii

Table of Contents
List of Tables ............................................................................................................................iv
List of Figures ............................................................................................................................ v
List of Acronyms and Abbreviations ...................................................................................... vii
Acknowledgements................................................................................................................ viii
1.0 Introduction.......................................................................................................................... 1
1.1 Road Ecology....................................................................................................................... 1
1.2 Research Objectives............................................................................................................. 6
2.0 Methods ............................................................................................................................... 8
2.1 Study Area ........................................................................................................................... 8
2.2 Biogeoclimatic Zones in Study Area ................................................................................... 9
2.3 Site Selection ..................................................................................................................... 12
2.4 Time Since Abandonment ................................................................................................. 14
2.5 Site Sampling Procedures .................................................................................................. 17
2.5.1 Environmental Data ........................................................................................................ 19
2.5.2 Soil sampling .................................................................................................................. 20
2.5.3 Vegetation Sampling ...................................................................................................... 22
2.6 Vegetation Analysis ........................................................................................................... 23
3.0 Results ............................................................................................................................... 26
3.1 General Site Descriptions .................................................................................................. 26
3.2 Plant Species Frequency and Diversity ............................................................................. 39
3.3 Substrate Differences ......................................................................................................... 43
3.4 Environmental Drivers....................................................................................................... 44
3.5 Plant Community Ordination ............................................................................................. 49
4.0 Discussion .......................................................................................................................... 53
4.1 Time Since Road Abandonment ........................................................................................ 55
4.2 Substrate Differences as an Environmental Driver ........................................................... 56
4.3 Trends Along the Climate Gradient ................................................................................... 58
4.3.1 Correlation Trends Along the Climate Gradient ............................................................ 59
4.3.2 Model Trends Along the Climate Gradient .................................................................... 60
4.4 Adjacency Effects .............................................................................................................. 61
4.5 Plant Community Patterns and Correlations...................................................................... 62
4.6 Research Applications and Limitations ............................................................................. 64
5.0 Conclusions and Recommendations .................................................................................. 66
Literature Cited ........................................................................................................................ 69
Appendix 1. Species and vegetation summary. ....................................................................... 74
Appendix 2. Road and shoulder vegetation comparisons. ....................................................... 94
Appendix 3. Vegetation correlations with hypothesized environmental drivers. .................... 95
Appendix 4. Multi-factorial model selection. .......................................................................... 98

iii

List of Tables
Table 2-1. Disturbance types observed during fieldwork in 2018 and 2019 at abandoned road
segments of Highway 16 in northwestern BC. ........................................................................ 13
Table 3-1. List of the most frequent and highest cover species found in road and shoulder
quadrats, organized by growth form in each of the sampled habitats. .................................... 40
Table 3-2. Mean (standard deviation) and results of paired t- tests or Wilcoxon* tests
comparing plant cover organized by growth form on asphalt road surfaces and gravel
shoulders within abandoned road segments (n=11). . ............................................................. 43
Table 3-3. Summary of Pearson’s and Spearman’s correlation coefficients of road surface
vegetation by plant group and different interpolated climate variables, light, and time since
road abandonment. ................................................................................................................... 45
Table 3-4. Summary of Pearson’s and Spearman’s correlation coefficients of gravel shoulder
surface vegetation with environmental variables. ................................................................... 47
Table 3-5. Best fit linear models of environmental predictors for the abundance of plant
growth forms at road centers on abandoned road segments (n=30). ....................................... 48
Table 3-6. Best fit linear models of environmental predictors for the abundance of plant
growth forms on the gravel shoulder surface (n=30). . ........................................................... 48
Table 3-7. Pearson's correlation coefficients of climate and environmental variables for Axis
1 and Axis 2. ............................................................................................................................ 51
Table 3-8. Pearson's correlation coefficients of plant species abundance from NMS
ordination of Axis 1. ................................................................................................................ 52
Table 3-9. Pearson's correlation coefficients of plant species abundance from NMS
ordination of Axis 2. ................................................................................................................ 53

iv

List of Figures
Figure 1- 1. A model of ecosystem dynamics in a road ecosystem. .......................................... 4
Figure 2-1. The study area in northwestern British Columbia with locations of sampled
abandoned sections of Highway 16 and biogeoclimatic regions. .............................................. 9
Figure 2-2. Archival photo of the Highway 16 opening ceremony in 1944 from the George
Little House in Terrace, BC. .................................................................................................... 16
Figure 2-3. Photo of an interpretive sign about the old highway 16 road segments at
Exchamsiks Provincial Park, British Columbia....................................................................... 17
Figure 2-4. Abandoned road segment sampling layout. .......................................................... 19
Figure 3-1. A photo of the Cut Block abandoned Highway 16 segment west of Terrace, BC,
with bryophyte cover and exotic plant cover on the asphalt surface. ...................................... 27
Figure 3-2. A photo of the Zymacord abandoned Highway 16 segment (asphalt road
substrate) west of Terrace, BC. The vegetation appeared to have been mowed in the past and
walked on periodically. This research site included several grass species in addition to the
mosses and forbs. ..................................................................................................................... 28
Figure 3-3. A photo of the Exstew abandoned Highway 16 segment (asphalt) west of Terrace,
BC. This site shares a ditch with the new highway. ................................................................ 29
Figure 3-4. A photo of the SE of Exchamsiks Park abandoned Highway 16 segment west of
Terrace, BC, with an asphalt surface next to CN rail line which shares a ditch...................... 30
Figure 3-5. A photo of the Exstew abandoned Highway 16 segment showing shallow soil
development with plant litter on the asphalt surface. .............................................................. 31
Figure 3-6. A photo of the 30 Mile abandoned Highway 16 segment with asphalt road
exposed inside a quadrat along the transect line. ..................................................................... 32
Figure 3-7. A photo of the Legate abandoned Highway 16 segment east of Terrace, BC, with
exposed asphalt road surface. The abandoned road segment is lower than the new Highway
16 and they share a ditch. ........................................................................................................ 32
Figure 3-8. A photo of the Delta Creek abandoned Highway 16 segment west of Terrace, BC
with asphalt road exposed. This is an example of another site with intermittent vehicle use
but it met sampling criteria. ..................................................................................................... 33
Figure 3-9. A photo of the St Croix abandoned Highway 16 segment east of Terrace, BC with
an asphalt road surface and secondary forest immediately adjacent to both sides. ................. 34
v

Figure 3-10. A photo of the Rainbow Pass abandoned Highway 16 segment (gravel) site
which was the furthest west in the CWH biogeoclimatic zone. There was evidence of
recreational vehicle traffic as shown by the parallel gravel tracks. ......................................... 35
Figure 3-11. A photo of the Yellow Cedar Lodge abandoned Highway 16 segment (asphalt)
west of Terrace, BC in the CWH biogeoclimatic zone where there was significant litter and
plant cover on road surface. ..................................................................................................... 35
Figure 3-12. A photo of the Flint Creek abandoned Highway 16 segment (gravel) east of
Terrace, BC in the ICH biogeoclimatic zone........................................................................... 36
Figure 3-13. A photo of the Witset (east of Witset Village) abandoned Highway 16 segment
(gravel) in the ICH biogeoclimatic zone. ................................................................................ 36
Figure 3-14. A photo of the Sheraton Mill abandoned Highway 16 segment (gravel) site (east
of the mill site and closer to the river) in the SBS biogeoclimatic zone. The left side of the
road was a wet riparian area as shown by the row of cottonwood trees. ................................. 37
Figure 3-15. A photo of the Priestly Hill abandoned Highway 16 segment (gravel) east of
Burns Lake, BC in the drier SBS biogeoclimatic zone. This site could be used periodically by
recreational traffic if accessed by adjacent private land. ......................................................... 37
Figure 3-16. A photo of the Chimdemash East Transect 1 abandoned Highway 16 segment
(gravel) east of Terrace, BC and running perpendicular to the New Highway 16. ................. 38
Figure 3-17. A photo of the Chimdemash East Transect 2 abandoned Highway 16 segment
(gravel) east of Terrace, BC, running perpendicular to the new Highway 16 but higher in
elevation along the creek than Transect 1, and just before an old bridge crossing. ................ 39
Figure 3-18. Road center plant richness versus time since road abandonment (n=30). .......... 41
Figure 3-19. Shoulder plant richness versus time since road abandonment (n=30). ............... 42
Figure 3-20. Shoulder Shannon-Wiener diversity index compared with time since road
abandonment (n=30). ............................................................................................................... 42
Figure 3-21. Scatterplot and linear regression of vascular plant cover versus time since road
abandonment (TSA) for the road center (n=30). ..................................................................... 46
Figure 3-22. NMS (nonmetric multidimensional scaling) ordination of 60 averaged plots (R =
road plot, S = shoulder plot) and 254 plant species sampled on 30 abandoned road segments.
................................................................................................................................................. 50
Figure 4-1. Road vascular plant cover model (in Table 3-5 with 0.329 R2 and a p value of
0.0046) predicted versus observed values of percent cover. ................................................... 56
vi

List of Acronyms and Abbreviations
AHM (annual heat-moisture index, ((MAT+10)/(MAP/1000)))
AICc (Akaike’s information criterion scores)
bFFP (day of the year the frost-free period begins)
CWH (Coastal Western Hemlock biogeoclimatic zone)
DD1040 (degree-days above 10°C and below 40°C)
eFFP (day of the year on which frost-free period ends)
Eref (Hargreaves reference evaporation in millimeters)
EXT (extreme maximum temperature over 30 years in oC)
ICH (Interior Cedar-Hemlock biogeoclimatic zone)
MAP (mean annual precipitation in mm)
MAR (mean annual solar radiation in MJ m‐2 d‐1)
MAT (mean annual temperature in ℃)
MCMT (mean coldest month temperature in °C)
MOTI (BC Ministry of Transportation and Infrastructure)
MWMT (mean warmest month temperature in °C)
NMS (nonmetric multidimensional scaling)
PAS (precipitation as snow in mm/year)
RH (mean annual relative humidity in %)
SBS (Sub-Boreal Spruce biogeoclimatic zone)
TD (temperature difference between MWMT and MCMT, or continentality in °C)
TSA (time since road abandonment in years)
vii

Acknowledgements
This thesis would not have been possible without the financial support of the University of
Northern British Columbia, Research Project Award which helped me execute my fieldwork
and complete my soil and vegetation lab analysis. Additional scholarship or stipend support
was provided by the UNBC Al Nevison Award, Seabridge Gold KSM Project Bursary, and the
Natural Resources Canada Emergency Management Strategy.
I appreciate the opportunity to be a regional student out of the Terrace campus where I reside
and supervised by a local professor, Dr. Phil Burton. Thank you, Phil, for taking me on as a
part-time graduate student to complete my M.Sc. while I had full-time employment.
My supervisory committee consisted of UNBC professors: Dr. Carla Burton, Dr. Darwyn
Coxson, and Dr. Mike Rutherford. I appreciate their willingness to help and kindness when
they made time for my urgent questions during their busy schedules. I thank Dr. Nancy
Shackelford (University of Victoria) for serving as my external examiner, and for her
suggestions to improve the final version of this thesis. Thank you for the opportunity to learn
from all your expertise.
I owe my research assistants Charlie Bourque, Rose Coffey, and Marie Blouin a great deal of
gratitude for helping me conduct my field data collection during the summers of 2018 and
2019. It was an honour to work with such capable, interesting, and strong women.
I really appreciate the information I received from staff in the BC Ministry of Transportation
and Infrastructure offices in Terrace and Smithers. Thank you, Grant Watson, Daena
Bilodeau Cooper, and Marlene Keehn for your assistance in helping me locate abandoned
sections of Highway 16 in northwestern BC. I was fortunate to have a colleague, Mark
Shumki (GIS Analyst) with the BC Ministry of Forests, create my research sites map outside
office hours. Thank you, Mark!
I would like to thank Don Hill, the laboratory services supervisor at the Coast Mountain
College, Terrace campus, for assisting me while I conducted my soil pH and electrical
conductivity analysis at the college’s physics laboratory.
Thank you to the following people for helping me successfully complete my statistical
analysis using R Studio and PC-ORD: Sunny Tseng and Lisa Koetke (UNBC Applied
Analysis Hub), Andrew Boxwell, Dr. Cedar Welsh, Dr. Heather Bryan, Dr. Jerilyn Peck and
Victoria Kress.
And thank you, David Duddy, for your Legal Land Survey experience that helped me
interpret the abandoned road locations and timelines. Also, I appreciate your patience and
support as my partner throughout my entire graduate studies. You are my rock and part of the
foundation which allowed me to pursue this milestone goal.

viii

1.0 Introduction
1.1 Road Ecology
There are two main types of roads in Canada: paved primary roads surfaced with
asphalt and secondary gravel roads. Asphalt is “a heated bituminous (petroleum) mixture of
92% aggregates and 8% glue pressed flat and cooled” to form an impermeable road surface
(Forman et al. 2003). In Canada, the province of British Columbia’s large land base requires
an extensive network of roads to connect communities and gain access to resources such as
timber, energy, and minerals (FLNRO n.d.). Paved primary roads were built to create access
for vehicle travel along corridors often already established by rail or water to access towns
and cities. Rural areas contain a network of gravel roads for residential, agricultural, and
natural resource extraction (Forman et al. 2003). The Canadian Federal Government manages
about 2% of roads in Canada which includes cross-country primary highways, federal Indian
Reserve land roads, National Parks, and historical military roads. Provincial or territorial
jurisdiction includes 25% of the Canadian road network, and local governments manage the
remaining 73% (Forman et al. 2003). Road networks of all kinds are linear, anthropogenic
disturbances on the land base that impact ecosystems by removing natural vegetation,
changing adjacent vegetation patterns and native plant community composition, redirecting
water flow, creating soil erosion from water runoff on road surfaces, fragmenting wildlife
habitat, and interfering with animal movement (Forman et al. 2003). Landscape connectivity
is altered by roads, which can negatively affect smaller organisms such as salamanders
(unable to cross for food or breeding) or positively affect the hunting success of predators
such as wolves and birds of prey (Lindenmayer and Fischer 2006). In landscape ecology,

1

linear corridors such as roads can function as either conduits or barriers to the movement of
water and animals (Forman and Godron 1986).
Road construction is a severe disturbance that removes vegetation, naturally occurring
soil horizons, and organic matter important to ecosystem productivity (Forman et al. 2003).
Vegetative seed banks may exist in road substrates but cannot be activated or effective to
enable germination because roads are constructed by compressing or grading a soil surface
and removing all biota. Primary roads are then surfaced with crushed rock or gravel that is
compacted and covered with asphalt or concrete (Walker 2012). The compacted road surface
prevents plant seed establishment, restricting root growth and road maintenance herbicide
application destroys seedlings. Abandoned roads leave no biological legacies of the original
ecosystem prior to road construction. Road building and maintenance methods such as
bulldozing, mounding, tarring, oiling, or spraying inhibit vegetation recovery during road use
and vegetation recovery can occur only through primary succession (Walker and del Morel
2008, Walker et al. 2003). This means that all plants must establish from propagules (seeds,
spores) that disperse to the former road surface from elsewhere and must somehow establish
and grow in a largely open, severe, and nutrient-poor environment.
Road building material is often a mixture of rock and subsoil created by regional
climate and material brought in from another location with different geological parent
material. Roadside ditches and shoulders are considered novel microhabitats with plant
species adapted to repeated disturbance (mowing, herbicide application, vehicle exhaust)
during road operation. Overall plant species richness can be higher in road ecosystems than
nearby native plant communities because the roadside plant community is influenced by
adjacent land uses, creating a blend of native and non-native vegetation from adjacent forests,
wetlands, farms, horticultural species from urban areas, and industry (Forman et al. 2003).
2

Roads act as invasion corridors for exotic plants to spread through seed dispersal by wind,
wildlife, and people or vehicles passing through. Some organisms avoid the “repulsion zone”
(i.e., an active or abandoned road) because of its harsh attributes such as increased heat or
cold, road salt, herbicides, heavy metals, and increased predator and herbivore activity
searching for food in edge habitats (Forman and Godron 1986). Linear corridors within a
forested landscape matrix are also open spaces with no or little forest canopy, resulting in
warmer conditions than found in adjacent ecosystems on sunny days, and the opposite on
cold windy days. Linear corridors (e.g., right of ways, rail line, highways) adjacent to roads
and ditches can be dominated by edge species, whereas road centers suffer from repeated
disturbance by humans and wildlife travel (Forman and Godron 1986).
Roads, their shoulders, ditches, and associated right of ways are linear disturbance
corridors within a land base. These habitats host organisms adapted to their specific
environment, share exchanges with adjacent natural ecosystems and change (erode or
degrade) their structure over time. This means roads can be considered a distinctive
ecosystem (Lugo and Gucinski 2000), as portrayed in Figure 1-1, which shows inputs and
outputs of a road ecosystem from construction materials to abiotic and biotic factors on or
next to roads, and their economic costs and benefits. Road ecology is the relationship
between roads and local ecosystem processes and organisms (Forman et al. 2003). It is
considered a relatively new emerging ecological science with contributions from different
disciplines: ecology, geography, engineering, and planning (Coffin 2007). Roadsides and
abandoned roads are also novel or emerging ecosystems, which are new combinations of
species and biodiversity because of human action, environmental change, and intended or
unintended introduction of new species (Hobbs et al. 2006).

3

Figure 1- 1. A model of ecosystem dynamics in a road ecosystem (from Lugo and Gucinski 2000).
The dashed line represents the boundary of the road ecosystem. Abbreviations are as follows: et is
evapotranspiration, and C is carbon.

Roads are impermeable barriers that can alter drainage patterns, sometimes resulting
in nearby flooding and soil erosion (Walker 2012). Consequently, if roads are abandoned and
no longer maintained for vehicle travel, they may be subject to erosion by flooding or
landslides. The movement of water accelerates along impermeable road surfaces, often
picking up sediment and depositing it downslope. During heavy rainfall events, flooding
from increased road runoff in riparian areas can negatively affect stream channels through the
deposition of sediment in the aquatic habitat of creeks and rivers. Logs, branches, and

4

boulders can be carried in high flows to other locations, sometimes blocking natural drainage
patterns. Pools form in otherwise dry areas and affect terrestrial plant growth more suited to
drier sites (Trammell and Carreiro 2011). Road construction cuts hillslope banks and
removes native vegetation which would otherwise absorb water and retain soil and
unconsolidated material with its roots. Road networks can negatively affect ecosystem
function; therefore, there is incentive to reduce their impacts through abandonment. This
could involve leaving materials to degrade and allowing soil development and native
vegetation to recover. Linear corridor ecosystems such as roads can change with repeated
disturbance from wind, water, wildlife, human use, mineral nutrient cycling, but will recoup
on their own if given time (Forman and Godron 1986). Other alternatives include reclamation
or revegetation efforts that could decrease erosion of road sediment into fish-bearing streams,
restrict motorized access, and recreate habitat for wildlife (Forman and Alexander 1998).
Ecological succession is a change in species composition, abundance, or dominance
that occurs naturally over time. Plant succession studies are often distinguished by the
severity of the disturbance that triggers them, where severely disturbed sites undergo primary
succession and less severely disturbed sites recover by secondary succession where seeds,
root stalks or rhizomes persist (Clements 1916, Prach & Walker 2018). The constraints of
primary succession are absent or infertile soils, nonexistent plant material or seed banks in
surface substrates to aid in natural regeneration, and a lack of nitrogen required for plant
establishment and growth. Primary succession follows severe natural or anthropogenic
disturbances such as volcanic eruptions, sand dune movement, glacial retreat exposing
unvegetated moraines, landslides or erosion removing vegetation and soil layers, the
introduction of toxic substances that sterilize the soil, and road building (Walker and del
Moral 2008, Tilman 1988). Primary succession after road abandonment starts from a barren
5

substrate without seed propagules, much like glacial till, sand dunes, and lava beds. The
seeds or spores of all colonizers must be carried onto abandoned roads by wind, water, or
animals for succession and ecosystem recovery to proceed (Walker et al. 2003).
Studies of ecological dynamics on road networks and abandoned road segments are
rarer than roadside (ditch and right of way) vegetation research focused on exotic vegetation
management (Forman et al. 2003). Research on plant recovery on anthropogenic disturbances
in northwestern British Columbia is also sparse. Therefore, this study aims to increase the
knowledge base of ecosystem recovery after severe disturbance in the Coastal Western
Hemlock (CWH), Interior Cedar-Hemlock (ICH), and Sub-Boreal Spruce (SBS)
biogeoclimatic zones (Meidinger and Pojar 1991). It has been generalized that the rapidity of
primary succession increases along dry to wet climate gradients and with the level of seed
encroachment from nearby ecosystems (Walker 2012). These general trends are expected to
be reflected in the vegetation found today on abandoned road segments in northwestern
British Columbia.
1.2 Research Objectives
This research aims to provide an understanding of vegetation recovery across a
climatic gradient that spans the Coastal Western Hemlock zone (wetter coastal climate), and
the Interior Cedar-Hemlock and Sub-boreal Spruce zones (drier interior climate) in
northwestern British Columbia, where road construction and abandonment was the
anthropogenic disturbance. Research was undertaken to answer the following questions,
among others:
•

Do longer road abandonment times lead to increased cover by all plant growth
forms?

6

•

Are plant cover by growth form, species richness and diversity higher on
gravel road substrates compared to asphalt substrates?

•

Are plant cover, species richness and diversity higher at the wetter coastal
locations compared to the drier, more continental interior locations?

The rate, composition, and limitations of ecosystem recovery on abandoned roads
were explored to learn more about primary succession and natural ecosystem recovery under
current conditions, but also reflects an underlying scientific curiosity about abandoned
infrastructure degradation moving forward and what will remain or how it will impact
ecosystems into the future.
This research could enhance various reclamation or restoration projects in the
northwest by providing a list of plant species found in post-disturbance habitats, by
identifying exotic plant species that may require control along Highway 16 or the adjacent
CN railway, and by providing evidence where natural regeneration is more likely or would
benefit from assisted recovery. Results of this research may also help design native seed
mixes, provide some knowledge of soil and climatic factors that facilitate or constrain the
natural recovery of abandoned road ecosystems, and thereby inform restoration professionals
working in northern British Columbia. To date, most road research in BC is related to
deactivation and rehabilitation of resource forest roads used for timber harvesting (Moore
1994, Anonymous 1999), and not asphalt or gravel highways.
An ecological chronosequence study is defined by Walker et al. (2010) as “a set of
sites formed from the same parent material or substrate that differs in the time since they
were formed.” This research is, in part, a vegetation recovery chronosequence study
comparing the same substrate, where each road segment was abandoned 16 to 57 years prior

7

to sampling. The Highway 16 road segments in this study were discontinued at different
times, depending on when the provincial government budget was available to improve new
highway infrastructure, typically associated with road straightening, reducing grades, and
improving creek and gully crossings for upgraded sections of Highway 16. Some old or
unused highway segments were blocked off by rock piles, sections were paved over as part of
the new highway segment or left abandoned but accessible by foot or vehicle as traffic was
shifted to the newly upgraded highway. Unlike the research reported here, temporal scales in
other road abandonment research largely consist of groups of sites abandoned at two or more
abandonment times, or all roads sharing the same abandonment year (Howard 1985, Bolling
1996, Heyne 2000, Bochet and Garcia-Fayos 2004, Bolling 2000).
2.0 Methods
2.1 Study Area
This study was conducted in northwestern British Columbia, Canada, spanning three
biogeoclimatic zones: Coastal Western Hemlock zone (CWH), Interior Cedar-Hemlock zone
(ICH), and Sub-Boreal Spruce zone (SBS) (Meidinger and Pojar 1991). Research sites were
located along Highway 16 between Prince Rupert and Fraser Lake, British Columbia (Figure
2-1). The thirty sites sampled represent a gradient from a temperate coastal, strongly
maritime climate to a sub-boreal interior somewhat continental climate.

8

Figure 2-1. The study area in northwestern British Columbia with locations of sampled abandoned
sections of Highway 16 (site names in the legend in the top right corner) and biogeoclimatic regions.

2.2 Biogeoclimatic Zones in Study Area
The abandoned roads in this study act as anthropogenically disturbed sites compared
to mature plant communities with climax vegetation. This plant recovery research was never
predicted to become a mature plant community like each of the biogeoclimatic zones because
abandoned roads are a severe disturbance with different substrate. The disturbed roads are
undergoing primary succession with younger plant communities but are presumably in the
process of recovering the composition characteristic of climax communities described for the
biogeoclimatic zones referenced in this study, as summarized in Banner et al. (1993).

9

The CWH zone is dominated by temperate rainforest in the coastal mountains and
boasts the highest rainfall and productivity in BC. In the northern part of the province, it is
associated with cool summers and high snowfall accumulation with relatively mild
temperatures during winter. Mean annual precipitation is approximately 1000 to 4400
millimetres and mean annual temperature is 5.2 to 10.5 ℃ (Meidinger and Pojar 1991) for
the entire CWH zone. Study sites located in the CWH zone ranged in elevation from 15
meters above sea level along lower reaches of the Skeena River to 145 m in a higher coastal
pass (Rainbow Summit). Descriptions of the CWH subzone variants are as follows:
CWHws1is a wet area with submaritime climate, the CWHvm1 variant is very wet with a
maritime climate, and the CWHvh2 is very wet with a hypermaritime climate that has the
highest precipitation (Meidinger and Pojar 1991).
There were thirteen study sites in the wet submaritime (CWHws1) typically with the
following subzone climax vegetation: western hemlock (Tsuga heterophylla), amabilis fir
(Abies amabilis), some western redcedar (Thuja plicata), with bunchberry (Cornus
unalaschkensis), five-leaved bramble (Rubus pedatus), and Queen’s cup (Clintonia uniflora)
in the understory. Four study sites were in the wet maritime (CWHvm1) subzone where the
typical dominant vegetation is western hemlock, amabilis fir, western redcedar and in the
understory Alaskan blueberry (Vaccinium alaskaense) dominates, then sparse traces of
Alaska bunchberry (Cornus unalaschkensis), deer fern (Struthiopteris spicant) and spiny
wood fern (Dryopteris expansa). There was only one study site (Rainbow Pass) in the very
wet hypermaritime (CWHvh2) subzone, which typically has the following dominant or
climax vegetation: western redcedar, yellow cedar (Callitropsis nootkatensis), western
hemlock, sometimes lodgepole pine (Pinus contorta) and mountain hemlock (Tsuga
mertensiana), and in the understory salal (Gaultheria shallon), Alaskan blueberry, false
10

azalea (Menziesia ferruginea), deer fern, and Alaska bunchberry with step moss
(Hylocomium splendens) and lanky moss (Rhytidiadelphus loreus) on the forest floor (Banner
et al. 1993).
The ICH zone is dominated by diverse temperate forest further inland and east of the
coastal mountains where summers are drier and warmer, and winters are cool and wet. Mean
annual precipitation is 500 to 1200 millimeters and mean annual temperature is from 2 to 8.7
℃ for the entire ICH zone (Meidinger and Pojar 1991). The ICH study sites ranged in
elevation from 155 meters near Little Oliver Creek to 353 meters near the Suskwa River. In
this study, the ICHmc2 was the only subzone variant located in the ICH. It is a moist region
with a cold climate. There were seven study sites in the ICHmc2 subzone, in which western
hemlock is the most dominant tree species. Intermittent trees include western redcedar,
lodgepole pine, hybrid spruce (Picea glauca x sitchensis), subalpine fir (Abies lasiocarpa),
paper birch (Betula papyrifera) and trembling aspen (Populus tremuloides). The understory
species include Alaska blueberry, false azalea, Canada bunchberry (Cornus canadensis) and
twinflower (Linnaea borealis) (Banner et al. 1993).
The SBS zone further inland is hilly with variable weather from moderate snowfall in
winter and summers that are damp and warm. Mean annual precipitation ranges from 440 to
900 millimeters and mean annual temperature is 1.7 to 5℃ for the entire SBS zone
(Meidinger and Pojar 1991). The SBS research sites range in elevation from 505 meters at
Telkwa to 763 meters on Priestly Hill. The SBS units included in this study were the SBSdk
and the SBSmc2. The SBSdk, mostly following the valleys, is a dry area with a cool climate
and the SBSmc2 variant, mostly found on the uplands, is a moist area with a cold climate
(Meidinger and Pojar 1991). There were three study sites in the SBSdk subzone, where
typical climax vegetation is dominated by hybrid white spruce (Picea engelmannii x glauca),
11

with some lodgepole pine and trembling aspen, and in the understory prickly rose (Rosa
acicularis), birch-leaved spirea (Spiraea betulifolia), soapberry (Shepherdia canadensis),
purple peavine (Lathyrus nevadensis), showy aster (Eurybia conspicua), Canada bunchberry
and fireweed (Chamaenerion angustifolium). Two study sites were in the SBSmc2 subzone
with dominant vegetation being lodgepole pine, subalpine fir, hybrid white spruce, and in the
understory black huckleberry (Vaccinium membranaceum), Canada bunchberry, five-leaved
bramble, twinflower, and heart leaved arnica (Arnica cordifolia), and on the forest floor there
is red-stemmed feathermoss (Pleurozium schreberi), step moss, and knight’s plume (Ptilium
crista-castrensis) (Banner et al. 1993).
2.3 Site Selection
Preparatory research took place during 2018 and 2019 and consisted of identifying
suitable research plot locations of Highway 16 abandoned road segments using information
from the following sources:
•

Province of British Columbia - Ministry of Transportation and Infrastructure (MOTI)
legal road survey maps.

•

In-person interviews and emails with MOTI staff from the Terrace and Smithers
offices.

•

Use of the BC Land Title and Survey Authority (LTSA 2022) to locate roads.

•

Google Earth desktop used in conjunction with the legal road surveys.

•

Field visits to 57 abandoned road segments to determine if they met sampling criteria.
Each site’s location along Highway 16 was measured using Google Earth and

Ministry of Transportation and Infrastructure highway markers, starting at Prince Rupert by
kilometer. The main sampling criteria for research sites were freedom from severe recent

12

disturbance, and a minimum of continuous vegetation/site conditions on at least 30 meters of
the abandoned road segment. Table 2-1 shows a generalized list of anthropogenic and
wildlife disturbances patterns observed on the road segments of Highway 16 (n=57) in
northwestern BC. Some of the higher disturbance impacts on road vegetation prohibited
further research and quantitative sampling from being conducted on those road segments.
Table 2-1. Disturbance types observed during fieldwork in 2018 and 2019 at abandoned road
segments of Highway 16 in northwestern BC.

Recreation
Fishing foot access
trail
Dog walking trail
Off highway vehicle
travel
Camping
Access to private land
Parking area for
hiking trail access

Industry
CN rail access road

Wildlife Sign
Moose scat

Waste
Full garbage bags

Timber harvesting
access
Trapline

Moose tracks

Powerline access

Bear tracks

Rangeland for cattle
Old run of the river
project (small
operation)

Wolf scat

Discarded
suitcases
Abandoned
vehicles
Household
appliances
Yard clippings
Sawdust piles

Bear scat

Old vegetable garden

Old log cabin
pieces

A numbered scale was created to characterize the severity of anthropogenic disturbance on
road vegetation along the transect for each abandoned road sampled (Table A1-2):
0. No sign of post-abandonment disturbance.
1. Signs of recurrent foot, minor disturbance, no vehicular traffic.
2. History of mowing, seeding, or other vegetation modifying activity; no ongoing
vehicular traffic.
3. Signs of light or old vehicular traffic, vegetation persisting in wheel tracks.
4. Signs of heavy, recent, or recurrent vehicular traffic, with no vegetation in wheel
tracks.
13

A number of potential sites, typically those given a disturbance score of 4, were omitted for
various reasons. Field sampling was not conducted where there was evidence of CN Rail
herbicide application on vegetation, difficulty in locating a road segment from maps and
Google Earth images in the field, wildlife safety concerns for the author and research
assistants (negative grizzly bear behaviour), erosional sediment covering road segment
vegetation, and evidence of high human or vehicular disturbance from recreational (hiking,
off-highway vehicles, camping, etc.) or industrial use (CN Rail or timber harvesting access
roads) of abandoned roads.
2.4 Time Since Abandonment
The time since road abandonment was largely based on legal road survey dates for new
Highway 16 sections, which prompted the desertion of old highway segments. The
approximate time since road abandonment (TSA) was determined by using the following
sources:
•

Gazettes (publications from newspapers stating when the provincial government takes
ownership of an existing road);

•

Legal road surveys from MOTI (2018) and Duddy (2020);

•

Assistance from a Legal Land Survey Technician who used the BC Land Title and
Survey Authority (Duddy 2022); and

•

Discussions with landowners, contractors, and the public about when some road
segments were decommissioned based on memory.

The road surveys ranged from 1962 to 1983 (MOTI 2018, Duddy 2020). The old highway
would have been abandoned right before or after the new highway road survey date, or
before or after the new highway was in use (Duddy 2020). The time since road abandonment

14

year is an approximation within three years before or after the legal land surveys were
completed for Highway 16 upgrades, which prompted the abandoning of old sections. The
old highway segments would have been abandoned before or after the TSA because legal
road surveys are conducted immediately before or after road construction. In discussions with
MOTI employees at the Terrace and Smithers offices, there is no official record of when old
highway sections were abandoned. This means the road surveys were the most accessible
method to approximate TSA. Records were compiled for road deactivation locations,
approximate abandonment timelines, and whether the road segments were asphalt or gravel
surfaces at the time of abandonment which was determined by field visits to the 57
abandoned roads. Many old, surrendered highway sections wound up and around creeks
parallel or tangential (or perpendicular) to the new highway before larger two-lane bridges
were constructed, which is why several of the study sites in this research are named after
those creeks; others are named after nearby towns or landmarks.
One example of Highway 16 improvements in the northwest was between the
Exchamsiks River and Andesite Creek during the early 1990s. The nine kilometers of
highway realignment was intended to improve passenger safety from overhanging icefall and
adjust speed limits to reduce vehicle collisions, reduce road maintenance expenses, decrease
wait time for vehicles during avalanche control in winter, and (where possible) improve fish
and wildlife habitat affected by highway and railway corridors (Bonwick et al. 1992).
The study site named “30 Mile” (based on the adjacent CN railway distance marker
near the new Highway 16) does not have a legal road survey for the new Highway 16 to help
estimate the old segment’s time since road abandonment, but the project maps imply it was
likely part of those highway improvements around 1995. A few legal road surveys from 1946

15

were located for the original highway from Prince Rupert to Little Oliver Creek, in addition
to the La Verge segment which was surveyed in 1930 (MOTI 2018, Duddy 2020).
In northwestern British Columbia, Highway 16 was officially opened in 1944 from
Prince Rupert to Hazelton, as seen in the photo below from the archives at George Little
House in Terrace (Figure 2-2). East of Hazelton, the Telkwa Curve study site was surveyed
in 1930. The “Witset curve” site wasn’t surveyed until 1971, as noted from a gazette, because
it was likely on an existing private road before the province took ownership of it as primary
highway (Duddy 2020). That road segment was subsequently abandoned in about 1993 when
the highway was straightened.

Figure 2-2. Archival photo of the Highway 16 opening ceremony in 1944 from the George Little
House in Terrace, BC.

16

The “Exchamsiks Provincial Park” old road study site (Figure 2-3) underwent a
restoration project around 2002 where asphalt was removed, and native plants were
transplanted to aid plant recovery. Due to the 2002 restoration, Exchamsiks has the shortest
abandonment time, with sampling conducted sixteen years post-abandonment. Today, there is
a walking trail on the old highway inside the park with interpretive signs that share the
restoration project's story.

Figure 2-3. Photo of an interpretive sign about the old highway 16 road segments at Exchamsiks
Provincial Park, British Columbia.

2.5 Site Sampling Procedures
Approximately half of the field sites were sampled in July and August 2018 (n=14)
and the other half were sampled in July and August 2019 (n=16). Road, shoulder, ditch, and
17

forest attributes were sampled at each site because there were noticeable differences in plant
species across each road ecosystem and between substrates (e.g., forest versus asphalt road).
By sampling each habitat, multiple environmental variables could be explored to determine
which contributed most to plant recovery on the abandoned roads e.g., propagules from
adjacent habitats, etc. Sites were characterized for vegetation and environmental conditions
using linear transects and square quadrats (plots) distributed at regular intervals along each
road segment. A 30-m tape measure was laid out as a transect along the center line of a
suitably homogenous segment of abandoned roadway. Using a clinometer, road slope was
recorded at 0 meters of each transect, facing towards the end of the transect at 30 meters.
Road slopes ranged from 0 to 5 degrees. Five quadrats, measuring 2 m x 2 m each, were
spaced 5 meters apart and on alternating sides of the transect. A die was rolled, with an odd
or even number designating whether the first quadrat was on the left or right side, then
alternating each quadrat after the first. Five plots measuring 1 m x 1 m were sampled on the
shoulders adjacent to asphalt or running surface. These plots matched the 30-meter road
transect but were staggered (located on the opposite side of the road) from road quadrats to
avoid overlap (Figure 2-4). Ditch vegetation was assessed by walking the transect length and
recording plant cover by species over an approximately 3 m x 30 m area. The composition of
adjacent forest cover was assessed in a 7.98 m radius circular plot (200 m2), located 10 to 60
m from the forest edge as determined by the roll of a die. Dominant tree species (in adjacent
forests) was also measured by plotless prism sampling (Terry and Chilingar 1955), recording
the basal area (m2/ha) of tree species using a basal area factor (BAF) of 5 or less depending
on tree diameter at breast heights (1.3 m).
Often there were two shoulders, ditches, and forest stands on either side of each road
segment. There was variation among selected study sites in that forests and ditches adjacent
18

were sometimes replaced by another road (often new upgraded highway), waterbodies, steep
rocky slopes, rail line, etc. For example, two sites did not have forest on either side of the
road (Exstew and SE of Exchamsiks Park). Instead, there were CN railway tracks and ditches
that were shared with new Highway 16, and a waterbody existed instead of adjacent forests.

Figure 2-4. Abandoned road segment sampling layout (asphalt/paved or gravel roads).

2.5.1 Environmental Data
Abiotic data were recorded in each transect on the road and shoulder. Cover of dead
woody debris was estimated inside quadrats (Tables A1-4 and A1-5) in the following
categories (adapted from different Province of British Columbia land management
handbooks):
•

Coarse woody debris > 7.5cm in diameter

•

Medium woody debris 1 to 7.5cm in diameter

•

Fine woody debris <1 cm in diameter
19

Light or canopy openness was measured at waist height standing inside the middle of
each road and shoulder quadrat and then averaged for each site for road and shoulder. A
concave spherical densiometer (Forestry Suppliers Inc.) was used to estimate percent canopy
openness or light available for plant growth, measured for each road and shoulder quadrat
and separately averaged for road quadrats and shoulder quadrats by recording the N, S, W, E
facing measurement inside each quadrat. Intercepts of the open sky or overhead foliage
(shade) were counted on a grid pattern of 96 points and converted to percent light.
Climate attributes for each study site were interpolated using the ClimateNA model
version 7.10 (Wang et al. 2016) based on annual averages from 1961 to 1990 normals.
Elevations for each site were recorded in the field with a GPS unit and verified in Google
Earth, instead of using default elevations in ClimateNA.
2.5.2 Soil sampling
Samples of the substrate (soil) now found on abandoned roads were collected in the
field and samples were analyzed in a lab to characterize selected attributes that might affect
vegetation establishment although this data was not used in the analysis because it was not
central to any specific hypothesis. Methods for soil sampling and analysis followed those
employed in similar forest and abandoned road ecosystem studies (Kalra and Maynard 1991,
Heyne 2000, Lloyd et al. 2013). Soil samples were collected from the center of the 10
quadrats per study site. Samples were separately bulked for the road and for the shoulder of
each road segment and air dried for analysis later (n=60). On roads with an asphalt surface
(n=11), soil depth varied from 0 to 9 centimeters depending on the abundance of canopy
cover that had generated litter available for soil development. Soil and litter depths were
recorded for each quadrat on road and on shoulder surfaces. Litter depth was measured inside
quadrats (and later averaged for road and for shoulder per study site) with an aluminum ruler
20

in centimeters by taking an average of five depth measurements throughout each quadrat.
Soil and litter compaction were measured together inside each quadrat using a DICKEY-john
soil compaction tester or penetrometer (Dickey-John Corporation, Auburn, Illinois) to
measure pounds per square inch of pressure (PSI) exerted until further penetration was halted
in surface substrates on the road and on the shoulder. Soil rooting depth was measured inside
quadrats (and later averaged for road and for shoulder per study site) by probing the soil
penetrometer into five different parts of each quadrat and taking an average depth in
centimeters from the markings on the device.
Soil lab analysis included measuring pH, total carbon and total nitrogen on a dry
weight basis, and electrical conductivity in milliSiemens per meter (Kalra and Maynard
1991). Soil samples were air dried, shaken in a brass soil sieve with 1 millimeter diameter
openings to separate out the smaller particle size needed for total carbon and nitrogen lab
analysis (and then used for pH and electrical conductivity lab analysis later), and weighed
with a top-loading balance (Metter Toledo model TLE3002E) (Mettler Toledo, Mississauga,
Ontario). Soil pH and electrical conductivity were measured at the Coast Mountain College
lab in Terrace using a Hanna 211 Processor (HANNA instruments, Laval, Quebec). The
electrode was calibrated with pH 4 and 7 buffer solutions. Organic soil samples were mixed
with a deionized water ratio of 1:5 and inorganic soil samples were mixed into a 1:2 water
ratio. Samples were stirred with a glass rod five times over 30 minutes and then left to settle
for another 30 minutes. Soil slurries were poured onto a size 5 filter paper inside a Buchner
funnel over a flask creating a vacuum to separate sediment from liquid. The liquid was then
used to test electrical conductivity and pH (to prevent ion release into samples from pH
measurements which would void electrical conductivity measurements). Total carbon and
total nitrogen measurements on a dry weight basis were completed at the Northern Analytical
21

Laboratory of the University of Northern British Columbia in Prince George. An elemental
analyzer (ECS 4010 – CHNS-O elemental combustion system) was used to combust the soil
samples with the standard sequential combustion/reduction setup recommended by the ECS
4010 manual (Costech Analytical Technologies Ltd. 2022) with a flowrate of 100 mL/min of
helium 5.0 as the carrier gas. The combustion oven was set to 1000°C, the reduction oven to
800°C, and the gas conversion oven to 70°C. Gases were then separated on a standard 3m
column and quantified with the built-in thermal conductivity detector to determine the
percent nitrogen and percent carbon of each bulked road and bulked shoulder sample per site.
2.5.3 Vegetation Sampling
Each road and shoulder quadrat (plot) was visually assessed for percent plant cover
by species, where the plant was not required to be rooted within a plot to be counted
(Causton 1988). Plant percent cover estimates were informed and calibrated by comparison
with published diagrams (Terry & Chilingar 1955). Percent ground cover of plant litter, dead
wood, rock, and exposed mineral soil was estimated similarly. Overall percent plant cover by
vegetation layer (trees, shrubs, herbs, and ground cover of moss and lichens) was recorded,
followed by percent cover estimates for each plant species. This ecological sampling method
allowed for the data to be summarized and analyzed by plant growth forms (Appendix 1)
(Causton 1988).
Plant species were identified by the author and research assistants in the field in
consultation with the following sources: Klinkenberg (2021), Mackinnon et al. (1992),
McCune and Geiser (2009), Meidinger and Pojar (1991), Pojar and Mackinnon (1994), Royer
and Dickson (2006), Schofield (1992), and Whitson (2009). Plant species that could not be
identified in the field were collected and taken to the lab for keying. Problematic specimens
were mailed to Curtis Bjork (Enlichened Consulting Ltd.; primarily lichens and bryophytes)
22

or to Nick Hamilton (Research Range Agrologist with the BC Ministry of Forests; primarily
for grasses).
2.6 Vegetation Analysis
Road quadrat data and shoulder quadrat data were averaged separately for summary
metrics of cover by growth form (trees, shrubs, woody, broadleaf forbs, graminoids, ferns
and fern allies, herbs, vascular, bryophytes, lichen, non-vascular, exotic, and total plant
cover) and used as response variables in the statistical analysis (below and Appendix 1).
Species richness (total species count) was tabulated as the number of species encountered
separately in all the road, shoulder, ditch, or forest plots at a site. Richness values are
comparable for a given habitat among study site locations, but not among habitats because of
different reference areas inventoried: 20 m2 for road centers, 5 m2 for road shoulders, 90 m2
for road ditches, and 200 m2 for adjacent forest.
The averaged plot data on the road, shoulder, ditch, and the forest at each site were
then used to calculate a Shannon-Wiener diversity index for each habitat at each site. The
Shannon-Wiener diversity index (Smith and Smith 1998) measures species evenness (relative
dominance of any individual species in a plant community) and in this case measures species
across a large area at each site. The Shannon-Wiener diversity index formula is provided
below, where H’ is the species diversity index, s is the number of species, and pi is the
proportion of cover of each species belonging to the ith species (Nolan and Callahan 2006).
s
H’ = - Σ pi log10 pi
i =1
Designation as an exotic species was derived from the E-Flora BC website
(Klinkenberg 2021). The influence of site and climate variables (n=30) was tested

23

statistically using the averaged quadrat data, and microsite and soil attributes were averaged
separately for road and for the shoulder plots (n=60).
Plant cover on gravel versus asphalt road surfaces was compared in R Studio (R
Studio Team, 2022) using the dplyr package (Wickham et al. 2022) for paired t-tests for each
growth form group, species richness, and the Shannon diversity index. Some comparisons
were made between shoulder plots and all road center plots, regardless of substrate
differences; other tests were made only between shoulder plots and unpaved road center plots
(n=19) to test for positional or edge effects. T-tests comparing asphalt road plots (n=11) with
their associated shoulder plots were also conducted but may conflate substrate effects with
positional effects.
Pearson’s and Spearman’s rank correlation tests were run in R Studio using the
tidyverse package (Wickham et al. 2019) to determine the strongest environmental predictors
to the vegetation response variables (Appendix 2), including plant groups, species richness
and the Shannon-Wiener diversity index.
An analysis of selected plant groups by growth form (Appendix 1) was conducted
using multiple regression analysis in R Studio using the MuMIn package known as multimodel inference (Burnham and Anderson 2002) to determine the best fit linear models for
each plant response group. Broader plant groups were not used in models (non-vascular,
herb, woody, and fern and fern allies cover) because more specific plant groups were tested
in the models which also represented the broader groups overall, e.g., tested bryophytes and
lichens separately instead of non-vascular cover. Road and shoulder plots were considered
separately to evaluate positional or substrate effects on vegetation. The strongest correlated
climate predictor and other environmental predictors that could have impacted plant growth
(road substrate, time since road abandonment, and light/canopy openness) were inserted into
24

models to determine the best fit model (Tables A4-1 to A4-15) for each vegetation response
by comparing Akaike’s information criterion (AICc) scores (where “c” denotes smaller
sample sizes <40), and Akaike weight and delta scores under two (Appendix 4). This
information-theoretic approach makes it possible to compare the weight of evidence for any
number of plant group responses (Logan 2010). The AICc scores determined how well the
models fit the data and help rank the importance of predictor influence on vegetation
responses by listing the best fit models using a data dredge (Symonds and Moussalli, 2011),
which runs all the possible combinations of models to alleviate the potential of missing the
best fit model (Logan, 2010). The dredge was used after carefully selecting ecologically
significant environmental predictors for each model and eliminating any colinear variables
with a variance inflation factor over five that could affect model validity. In most cases road
center and shoulder habitats were highly correlated with the same environmental predictors
and were both measured in percent plant cover. Light was not tested in models where
vegetation groups had the potential to create shade or affect light availability to road and
shoulder surface vegetation, compared to reacting to light, i.e., when testing tree, shrub,
woody, vascular, or total plant cover (as all those categories could include trees that cast
shade).
Plant community analysis was completed using nonmetric multidimensional scaling
(NMS) using PC-ORD v. 7.0 (McCune and Mefford 2016). NMS was used on vegetation
data collected in road and shoulder plots to describe patterns across study sites in species
space along environmental gradients. The response matrix consisted of percent cover for 254
taxa in 60 plots. Vectors portraying association with environmental gradients had r>0.5 in the
biplot because they were the most strongly correlated to the vegetation data. A Sorensen
(Bray and Curtis, 1957) random starting distance measure was applied to the vegetation NMS
25

ordination and the environmental matrix was summarized using Euclidean distances. A twodimensional NMS interpretation was conducted (randomized test p = 0.004) with final stress
of 19.675 (real stress was 15.605) after several NMS solutions and comparisons. Pearson’s
correlations were used to determine which plant species were strongly associated with Axis
1, Axis 2, and environmental gradients.
3.0 Results
3.1 General Site Descriptions
The sites studied varied widely in their land use context, nearby plant communities,
and the structure and composition of current vegetation. They also had different climates,
slopes (0 to 3 degrees), road surface widths (6 to 14 meters), recent and past disturbances,
canopy cover (shaded or open), litter cover, and jurisdictions (right of ways, private land,
etc.). The study area spans a climate gradient from maritime coastal to subcontinental interior
conditions. The old road segments were mostly gravel surfaces (n=19), others with asphalt
paving (n=11). Environmental data are tabulated in separate tables in Appendix 1: site
characteristics (Table A1-2), interpolated climate data (Table A1-3); road quadrat data (Table
A1-4), shoulder quadrat data (Table A1-5), road soil characteristics (Table A1-6), and
shoulder soil characteristics (Table A1-7). Photographs were taken at each site to help
represent visual differences between sites. Two asphalt sites, the 30 Mile and Cut Block sites
(e.g., Figure 3-1), had almost no vascular plant cover on the road, except some exotic species.
There was evidence of mowed vegetation on the road at Suskwa (adjacent to private
land and near residential buildings) and Zymacord (Kitksumkalum First Nation reserve land
with a residential building nearby) sites (e.g., Figure 3-2).

26

Figure 3-1. A photo of the Cut Block abandoned Highway 16 segment west of Terrace, BC, with
bryophyte cover and exotic plant cover on the asphalt surface, which is exposed on the left side.

27

Figure 3-2. A photo of the Zymacord abandoned Highway 16 segment (asphalt road substrate) west of
Terrace, BC. The vegetation appeared to have been mowed in the past and walked on periodically.
This research site included several grass species in addition to the mosses and forbs.

Several sites were without adjacent forest on one or two sides of the abandoned road
because they shared a ditch with new Highway 16, the CN rail line, or there was a waterbody
instead. These sites had incomplete forest data, so were not included when analyzing the
influence of forest vegetation on road vegetation recovery. Forest tree cover was not
significantly correlated with any of the plant group cover sums (Tables A3-3 and A3-6) and
therefore not used in the multiple regression models. In cases where the old highway and the
new Highway 16 were in close proximity, it was observed that the shared ditch had several
exotic plant species that likely impacted the plant community composition on the old road
segment, as shown in Figure 3-3 at Exstew. Old road segments were often running parallel to
the new Highway 16 or perpendicular to it where sections connected to the new highway.
28

This close proximity of the old and new highway by orientation or distance could have aided
exotic species movement by seed dispersal. The Cut Block research site was parallel to and
only 19 meters from the new Highway 16 and shared a ditch.

Figure 3-3. A photo of the Exstew abandoned Highway 16 segment (asphalt) west of Terrace, BC.
This site shares a ditch with the new highway. Rose Coffey (Research Assistant) is in the ditch
recording vegetation cover.

The old road segments tangential to the new highway often followed up along a creek
to where old bridge crossings would have existed. Some different examples of old road
placement in relation to the new Highway 16 were Witset (perpendicular at 12 meters), Delta
Creek (parallel at 40 meters), Exstew at 16 meters, SE of Exchamsiks Park (perpendicular at
11 meters), 30 Mile (parallel at 16 meters), Telkwa (parallel at 17 meters), Legate (parallel at
18 meters), Smithers Research Farm (perpendicular at 24 meters), and Big Oliver East
(parallel at 9 meters). The sites with CN rail line next to them where Exstew (11 meters
distant), SE of Exchamsiks Park (28 meters away the current highway) (Figure 3-4), and
Sheraton Mill (distance not measured but visible from the transect across a wetland). La

29

Verge and Exstew sites were next to old back channels of the Skeena River on one side of the
abandoned roads, so forest data were not recorded here.

Figure 3-4. A photo of the SE of Exchamsiks Park abandoned Highway 16 segment west of Terrace,
BC, with an asphalt surface next to CN rail line which shares a ditch. This site appeared to have
intermittent vehicle traffic likely to access the rail line, but it was included in the samples because
some of the more disturbed sites were sampled to reach the minimum 30 abandoned roads sample size
for this research project. From the more disturbed sites, there still was continuous and relatively
uniform vegetation along the road to meet sampling criteria.

In most situations, the asphalt road surface was covered with vascular and nonvascular vegetation, but in some cases, asphalt was visible (intact or in remnant pieces) and
recorded as a substrate in quadrats (Tables A1-4 and A1-5). The sites with exposed asphalt
that met sampling criteria included 30 Mile, Legate, Cut Block, Delta Creek, Exstew, SE of
Exchamsiks Park, Shames West, Delta Creek, Exstew, Zymacord, and St Croix. Witset, La
Verge, and Priestly Hill sites showed evidence of possibly being paved in the past, but the
asphalt had presumably been ripped up and removed, as the old road was dominantly gravel
when field work was conducted. In those cases, very small pieces of remnant asphalt were

30

recorded in plots or just observed along the abandoned road or in shoulders and ditches. At
the Exstew abandoned road segment soil and litter cover on the asphalt road surface was very
thin and could be pushed aside to see the old road, but asphalt was not visible during
vegetation sampling (Figure 3-5).

Figure 3-5. A photo of the Exstew abandoned Highway 16 segment showing shallow soil
development with plant litter on the asphalt surface. Asphalt is exposed in grey on the left side above
the trowel.

The photo from the 30 Mile site (Figure 3-6) shows a large patch of exposed asphalt
inside a quadrat on the road surface and a minimal amount of leaf litter from adjacent trees
on the road. Legate was sampled on a portion of the road where there was adjacent forest
cover but in Figure 3-7 you can see exposed asphalt leading up to the start of the transect.
This photo also shows Highway 16 running parallel to the old road segment with a shared
ditch. At Delta Creek, there are visible vehicle tracks on the asphalt as shown in Figure 3-8.
This site is easily accessible from the new Highway 16 and therefore periodically driven on.

31

Figure 3-6. A photo of the 30 Mile abandoned Highway 16 segment with asphalt road exposed (grey
colour) inside a quadrat along the transect line.

Figure 3-7. A photo of the Legate abandoned Highway 16 segment east of Terrace, BC, with exposed
asphalt road surface. The abandoned road segment is lower than the new Highway 16 and they share a
ditch. The transect starts past the blue bin (inside yellow circle) in the shade where there was
continuous plant cover for 30 meters.

32

Figure 3-8. A photo of the Delta Creek abandoned Highway 16 segment west of Terrace, BC with
asphalt road exposed. The true center of the road with the yellow paint line is underneath the shovel
and transect line. Percent asphalt was recorded inside quadrats were visible. This is an example of
another site with intermittent vehicle use but it met sampling criteria.

The St. Croix site (Figure 3-9) is an asphalt site with a gravel shoulder where the adjacent
trees have grown out over the road to take advantage of the higher light available. There was
an obvious lack of recent impact on the vegetation, although some was flattened or leaning
from wildlife passing through. This lack of disturbance allowed for litter build up and soil
development on the asphalt surface resulting in vigorous vascular plant growth (grasses,
forbs, exotics).

33

Figure 3-9. A photo of the St Croix abandoned Highway 16 segment east of Terrace, BC with an
asphalt road surface and secondary forest immediately adjacent to both sides. Charlie Bourque
(Research Assistant) is standing next to the transect line.

Examples of wetter coastal abandoned roads in the CWH biogeoclimatic sites (e.g.,
Figures 3-10 and 3-11) and the drier interior sites in the ICH and SBS biogeoclimatic sites
are presented below. Increased site productivity wasn’t limited to coastal sites as shown at
the Flint Creek ICH site (Figure 3-12), where the road shows moderate grass cover, but it
was not as dense as the CWH sites overall. An example of a dry site with less plant cover on
the road was the Witset Curve ICH site (Figure 3-13). Interior sites near wet areas also had
higher plant cover as seen at the Sheraton Mill site, which was next to a low-lying riparian
area with aquatic vegetation near the road (Figure 3-14). In contrast, the Priestly Hill SBS
site was sparsely covered on the road center (Figure 3-15).
34

Figure 3-10. A photo of the Rainbow Pass abandoned Highway 16 segment (gravel) site which was
the furthest west in the CWH biogeoclimatic zone. There was evidence of recreational vehicle traffic
as shown by the parallel gravel tracks.
Figure 3-11. A photo of the Yellow Cedar
Lodge abandoned Highway 16 segment
(asphalt) west of Terrace, BC in the CWH
biogeoclimatic zone where there was
significant litter and plant cover on road
surface.

35

Figure 3-12. A photo of the Flint Creek abandoned Highway 16 segment (gravel) east of Terrace, BC
in the ICH biogeoclimatic zone.

Figure 3-13. A photo of the Witset (east of Witset Village) abandoned Highway 16 segment (gravel)
in the ICH biogeoclimatic zone.

36

Figure 3-14. A photo of the Sheraton Mill abandoned Highway 16 segment (gravel) site (east of the
mill site and closer to the river) in the SBS biogeoclimatic zone. The black umbrella is in the middle
of the abandoned road with a quadrat to the right. The left side of the road was a wet riparian area as
shown by the row of cottonwood trees.

Figure 3-15. A photo of the Priestly Hill abandoned Highway 16 segment (gravel) east of Burns Lake,
BC in the drier SBS biogeoclimatic zone. This site could be used periodically by recreational traffic if
accessed by adjacent private land.

37

In one case, three transects were sampled within the same drainage; Chimdemash
West, Chimdemash East Transect 1 and Chimdemash East Transect 2. The two eastern
transects showed a marked difference in plant cover moving away from the adjacent new
Highway 16 (Transect 1 in Figure 3-16) and higher up the creek towards an existing bridge
crossing where the site was noticeably drier at Transect 2 (Figure 3-17). The western site had
a small deconstructed hydroelectric installation next to the creek and modern dwellings
nearby that showed evidence of horticultural (garden) and agricultural (hay) vegetation from
the former homestead.

Figure 3-16. A photo of the Chimdemash East Transect 1 abandoned Highway 16 segment (gravel)
east of Terrace, BC and running perpendicular to the New Highway 16.

38

Figure 3-17. A photo of the Chimdemash East Transect 2 abandoned Highway 16 segment (gravel)
east of Terrace, BC, running perpendicular to the new Highway 16 but higher in elevation along the
creek than Transect 1, and just before an old bridge crossing.

3.2 Plant Species Frequency and Diversity
A total of 312 plant taxa were identified during this research (Table A1-1). The most
frequent and highest cover species on the road and shoulder surfaces are listed in Table 3-1
which includes four exotic species, eleven native species and an unknown lichen.

39

Table 3-1. List of the most frequent and highest cover species found in road and shoulder quadrats,
organized by growth form in each of the sampled habitats. Values denote frequency as a count of the
locations at which a species occurred and mean percent cover across all locations (including those
with zero cover of the species).

Road, n=30
Mean
Frequency Cover %

Shoulder, n=30
Mean
Frequency Cover %

Alnus rubra
Populus balsamifera spp.
trichocarpa
Tsuga heterophylla

16

19.5

17

16.9

17
16

6.5
5.2

19
21

6.6
12.2

Alnus viridis ssp. sinuata
Rubus parviflorus

6
14

2.7
5.8

6
15

2.2
7.4

Galium triflorum
Tanacetum vulgare*
Trifolium hybridum*

13
3
5

0.41
0.32
2.4

10
2
4

0.35
0.77
0.15

Bromus inermis*
Poa pratensis*
Ferns & Fern Allies:
Athyrium filix-femina
Bryophytes:
Hylocomium splendens
Niphotrichum ericoides
Lichens:
Peltigera britannica
Stereocaulon tomentosum
(unknown lichen)

3
8

0.55
4.4

3
6

0.62
0.11

9

2.6

8

1.6

13
8

1.9
6.9

9
5

1.4
2.6

1
3
0

0.01
0.31
0

2
2
2

0.09
0.06
0.11

Plant Species
Trees:

Shrubs:

Forbs:

Graminoids:

* exotic to British Columbia, as per Klinkenberg (2021)

Species richness (number of plant species) and the Shannon-Wiener diversity index
were calculated for plant communities on the road, shoulder, ditch, and forest at each of the
30 locations (Table A1-8). Road, shoulder, ditch, and forest plots were different sizes (Figure
2-4) which did not allow for comparison of richness and diversity values between habitats at
each site. Species richness was greatest on the road at Suskwa (43 taxa), Exchamsiks Park
(39), Legate (36), and Rainbow Pass (35) (Table A1-8). The sites with the most diverse
40

vegetation on the road were SE of Exchamsiks Park (Shannon-Wiener score of 1.29) and
Suskwa (Shannon-Wiener 1.27) sites (Table A1-8). On the road centers, plant species
richness was significantly correlated with plant recovery over time (Figure 3-18), but plant
diversity was not significantly (p value = 0.3579) correlated with time since abandonment.

Plant Species Richness
(number of species)

50
45

40
35
30
25
20
15
10
5
0
0

10

20

30

40

50

60

Time Since Road Abandonment (years)
CWH

ICH

SBS

regression

Figure 3-18. Road center plant richness versus time since road abandonment (n=30). The regression
line is y = 0.52079 x – 1.721; p = 0.0489, R2 = 0.1315.

On the shoulders, plant species richness was also significantly correlated with time
since abandonment (Figure 3-19), but also was plant diversity (Shannon-Wiener index)
(Figure 3-20). No significant correlation or linear relationship was found between road and
shoulder plant species richness or Shannon-Wiener diversity index and the climate indicators.

41

Plant Species Richness
(number of species)

40
35
30
25
20
15
10
5
0
0

10

20

30

40

50

60

Time Since Road Abandonment (years)
CWH

ICH

SBS

regression

Figure 3-19. Shoulder plant richness versus time since road abandonment (n=30). The regression line
is y = 0.44705 x – 2.227746; p = 0.0249, R2 = 0.1671.

Shannon-Wiener Diversity Index

1.4
1.2
1

0.8
0.6
0.4
0.2
0
0

10

20

30

40

50

60

Time Since Road Abandonment (years)
CWH

ICH

SBS

regression

Figure 3-20. Shoulder Shannon-Wiener diversity index compared with time since road abandonment
(n=30). The regression line is y = 0.01143 x + 0.34236; p = 0.0157, R2 = 0.1911.

42

3.3 Substrate Differences
Paired t-tests and Wilcoxon tests were run to determine which plant growth groups
were more frequent on gravel or asphalt surfaces. All shoulders consisted of gravel substrate,
while road substrates were either gravel or asphalt. Table 3-2 presents mean plant cover by
growth form and diversity metrics for asphalt road and gravel shoulder surfaces, and the
associated t-test or Wilcoxson test (where data were not normally distributed) results.
Table 3-2. Mean (standard deviation) and results of paired t- tests or Wilcoxon* tests comparing plant
cover organized by growth form on asphalt road surfaces and gravel shoulders within abandoned road
segments (n=11). Significant statistical relationships with bolded p values (<0.05).

Variables
Tree cover
Shrub cover
Woody cover
Forb cover
Fern and Fern Allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-vascular cover
Vascular cover
Exotic cover
Total plant cover
Shannon-Wiener index

Asphalt
Road Center
Cover Mean, %
47.8 (42.6)
21.8 (20.4)
69.6 (47.0)
14.7 (12.9)
1.2 (2.1)
12.8 (15.1)
28.7 (20.6)
30.3 (31.0)
0 (0)
30.3 (31.0)
98.3 (43.2)
12.5 (15.3)
128.6 (32.1)
0.9 (0.3)

Gravel Shoulder
Cover Mean, %
72.7 (27.6)
39.4 (23.8)
112.2 (32.7)
12.8 (16.2)
2.9 (3.5)
4.2 (6.0)
19.9 (20.1)
7.6 (15.4)
0.02 (0.06)
7.6 (15.5)
132.1 (30.8)
6.2 (8.7)
139.7 (33.7)
0.9 (0.2)

p value
0.0425
0.0132
0.0052
*0.5771
*0.0591
*0.0666
*0.2783
*0.0080
*0.3711
*0.0080
0.0280
*0.2662
0.4164
0.9683

* denotes Wilcoxon tests used where data were non-normal.

Bryophyte cover and non-vascular plant cover were significantly greater on asphalt than
gravel substrates (Table 3-2 and Appendix 2). During field sampling, bryophytes were
observed more often on moist asphalt roads where surface water pooled on the impenetrable
surface, creating wetter site conditions for bryophytes. They were not observed on welldrained gravel sites. Lichen cover and fern and fern allies cover did not favor asphalt over

43

gravel specifically, as represented by their non-significant p values (Appendix 2). As
expected, woody cover (tree and shrub) was significantly different between substrates (Table
3-2 and Table A2-1), with substantially more woody cover on gravel shoulders where rooting
depths (Table A1-4) were greater than on compact or impenetrable road substrates (Table
A1-5). Consequently, vascular plant cover was significantly greater on gravel substrates than
on asphalt road centers at the same sites (Table 3-2).
In some cases, road versus shoulder positional effects on vegetation existed regardless
of the substrate (Appendix 2). The Shannon-Wiener diversity index showed a significant
difference between gravel road and gravel shoulder surfaces (Table A2-2), with higher
species diversity at road center on the same substrate, likely due to more light available for
plant growth. Forb, graminoid, herb, exotic and total plant cover were also significantly
greater at gravel road centers than on the adjacent gravel shoulders.
3.4 Environmental Drivers
Vegetation responses were correlated with several environmental predictor variables,
including time since road abandonment (TSA), light (canopy openness), and multiple climate
variables estimated for each site. The predictor variables listed in Table 3-3 had the strongest
correlation coefficients with response variables, some of which have significant p values
(p<0.05). Table 3-3 and Tables A3-1 to A3-3 summarize variable correlations with
vegetation attributes on the road centers, with gravel or asphalt surfaces not distinguished.
Correlations of shoulder vegetation attributes with selected environmental predictors are
summarized in Table 3-4 and Tables A3-4 to A3-6.

44

Table 3-3. Summary of Pearson’s and Spearman’s correlation coefficients of road surface vegetation
by plant group and different interpolated climate variables, light, and time since road abandonment.
Significant statistical relationships are denoted with bolded p values (p<0.05).
Vegetation variables
Tree cover
Tree cover
Shrub cover
Woody cover
Woody cover
Forb cover
Forb cover
Forb cover
Fern and fern allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-vascular cover
Vascular cover
Vascular cover
Exotic cover
Total plant cover
Total plant cover
Species richness
Shannon-Wiener diversity index

Environmental
variables
Pearson's
Light
-0.592
TSA
0.232
MAR
0.288
Light
-0.635
TSA
0.370
TSA
0.365
MAT
-0.525
DD1040
-0.501
Eref
-0.403
AHM
0.410
TSA
0.386
Eref
0.520
bFFP
0.409
Eref
0.514
TSA
0.497
Light
-0.539
RH
-0.686
TSA
0.546
Light
-0.577
TSA
0.261
EXT
-0.327

p value
0.000564
0.216900
0.122600
0.000166
0.044260
0.047290
0.002886
0.004826
0.027300
0.024570
0.035000
0.003206
0.024800
0.003685
0.005233
0.002109
0.000029
0.001790
0.000839
0.163100
0.077390

Spearman's
-0.643
0.396
0.119
-0.592
0.360
0.406
-0.398
-0.438
0.113
0.196
0.350
0.628
0.329
0.598
0.513
-0.461
-0.531
0.615
-0.382
0.327
-0.183

p value
0.000127
0.030140
0.532000
0.000563
0.050460
0.026040
0.029580
0.015560
0.112500
0.299700
0.057680
0.000205
0.075660
0.000479
0.003725
0.010390
0.002536
0.000299
0.037470
0.077590
0.334000

Light (% canopy openness), TSA (time since road abandonment in years), MAR (mean annual solar radiation in
MJ m‐2 d‐1), MAT (annual temperature in ℃), DD1040 (degree-days above 10°C and below 40°C), Eref
(Hargreaves reference evaporation in millimeters), AHM (annual heat-moisture index,
(MAT+10)/(MAP/1000)), bFFP (day of the year the frost free period begins) and EXT (extreme maximum
temperature over 30 years, oC).

Time since road abandonment (TSA) was significantly correlated with woody cover
(trees and shrubs), forb cover, herb cover, vascular plant cover and total plant cover on the
road surface of abandoned highway segments, shown by the bolded p values in Table 3-3. As
shown in Figure 3-21, TSA had a significant impact (R2=0.25, p=0.005) on vascular plant
cover on the road centers. High levels of light or canopy openness were negatively associated
with high levels of tree cover, woody cover, vascular plant cover and total plant cover on the
45

road surface, shown by the negative coefficients and bolded p values in Table 3-3.
Surprisingly, tree cover in the adjacent vegetation did not influence road or shoulder plant
cover, species richness or vegetation diversity (Tables A3-3 and A3-6). Vegetation responses
were also significantly (p<0.05) correlated with eight climate variables estimated for the
different locations sampled (Table 3-3).

Figure 3-21. Scatterplot and linear regression of vascular plant cover versus time since road
abandonment (TSA) for the road center (n=30). The 95% confidence interval for the predictive
equation (denoted by the solid line) is highlighted in the shaded dark grey area.

Time since road abandonment (TSA) was not as strong a predictor on the shoulder
vegetation, as shown in Table 3-4 and Table A3-4, where there was only one marginally
significant p value (p<0.10) for species richness. TSA did not have a significant impact
(p>0.05) on vascular plant cover on the gravel shoulders adjacent to the road center. Light or
canopy openness was negatively associated with tree cover, woody cover (trees and shrubs),
vascular plant cover and total plant cover on the shoulders, as found in the road plots (Table
46

3-4 and Table A3-2). Six climate variables were significantly correlated with shoulder
vegetation but with less significant responses than at road centers (Table 3-4).
Table 3-4. Summary of Pearson’s and Spearman’s correlation coefficients of gravel shoulder surface
vegetation with environmental variables. Significant statistical relationships are shown with bolded p
values (<0.05).
Vegetation variables
Tree cover
Tree cover
Tree cover
Shrub cover
Woody cover
Woody cover
Forb cover
Fern and Fern Allies cover
Graminoid cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-Vascular cover
Vascular cover
Vascular cover
Vascular cover
Exotic cover
Total plant cover
Total plant cover
Species Richness
Species Richness
Shannon-Wiener Diversity
index

Environmental
variables
Light
eFFP
PAS
PAS
Light
AHM
RH
PAS
AHM
RH
RH
Eref
EXT
Eref
PAS
TSA
Light
RH
PAS
Light
TSA
RH

Pearson's
-0.555
0.403
0.319
0.378
-0.558
-0.469
-0.209
0.255
0.410
-0.335
-0.221
0.207
-0.406
-0.290
0.460
0.062
-0.555
-0.304
0.456
-0.464
0.098
-0.340

p value
0.001452
0.027300
0.085420
0.039440
0.001347
0.008888
0.267700
0.174600
0.024570
0.070520
0.255800
0.273200
0.026200
0.293500
0.010560
0.744800
0.001440
0.102900
0.010560
0.009772
0.605500
0.065870

Spearman's
-0.339
0.366
0.391
0.399
-0.295
-0.401
-0.362
0.380
0.154
-0.349
-0.225
0.471
-0.106
0.419
0.527
0.086
-0.261
-0.340
0.534
-0.273
0.464
0.332

p value
0.066730
0.047010
0.032450
0.028960
0.114100
0.027980
0.049140
0.038170
0.416200
0.058780
0.232100
0.008677
0.579000
0.021050
0.002747
0.652100
0.163200
0.065970
0.002357
0.144700
0.009844
0.073050

TSA

0.175

0.355800

0.113

0.553500

Light (canopy openness, %), eFFP (day of the year on which frost-free period ends), PAS (precipitation as
snow, mm/year), AHM (annual heat-moisture index, (MAT+10)/(MAP/1000)), RH (mean annual relative
humidity, %), Eref (Hargreaves reference evaporation, mm), EXT (extreme maximum temperature over 30
years, oC), and TSA (time since road abandonment, years).

The best fit road models (Table 3-5) shared five predictor variables with the best fit
shoulder models (Table 3-6). There were three temperature-related predictor variables unique
to the road: DD1040, bFFP, and (Eref). There were two temperature related predictor
variables unique to the shoulder (EXT and eFFP) and one climate moisture variable (PAS)
(Table 3-6). All models in the Tables 3-6 and 3-7 include parsimonious predictor variables,
47

meaning that sometimes a simpler model was favored over a more complex model but is still
a good representation of the data for the individual growth form assessed (Appendix 4).
Table 3-5. Best fit linear models of environmental predictors for the abundance of plant growth forms
at road centers on abandoned road segments (n=30). Values are coefficients that can be used in a
predictive equation.
Response variables
Vascular plant cover
Total plant cover
Exotic plant cover
Graminoid cover
Bryophyte cover
Lichen cover
Forb cover

p value
0.004616
0.001791
0.000042
0.024570
0.003207
0.024800
0.026160

R2
0.329
0.298
0.580
0.168
0.271
0.167
0.347

Intercept
109.64
136.38
26.31
13.63
25.67
1.08
25.31

RdSub

TSA
32.08
31.84

-14.01

Light

RH

AHM

-7.68

-25.05

Eref
-17.65

DD1040

bFFP

8.33
14.10
1.44
-4.49

7.12

-5.80

-8.66

Abbreviations: RdSub (0 for gravel and 1 for asphalt substrates, respectively), TSA (time since road
abandonment, years), Light (canopy openness, %), RH (mean annual relative humidity, %), AHM (annual heatmoisture index), Eref (Hargreaves reference evaporation, mm), DD1040 (degree-days above 10°C and below
40°C), and bFFP (day of the year the frost-free period begins).

Table 3-6. Best fit linear models of environmental predictors for the abundance of plant growth forms
on the gravel shoulder surface (n=30). Values are coefficients that can be used in a predictive
equation.
Response variables p value
Vascular plant cover 0.008232
Total plant cover
0.011260
Exotic plant cover
0.016650
Graminoid cover
0.046150
Bryophyte cover
0.100700
Lichen cover
0.025400
Shrub cover
0.009085
Tree cover
0.027330

R2
0.299
0.208
0.321
0.204
0.093
0.166
0.219
0.162

Intercept
124.98
118.61
8.23
5.88
10.85
0.3833
39.43
-244.09

RdSub
-27.79

TSA

-5.73
-4.52

-2.85

Light

RH

AHM

PAS
14.16
18.68

EXT

eFFP

-4.46
2.68
4.51
-0.3266

-19.92
1.12

Abbreviations: RdSub (0 for gravel substrate and 1 is for asphalt at road center), TSA (time since road
abandonment in years), Light (% canopy openness), RH (% mean annual relative humidity), AHM (annual heatmoisture index ((MAT+10)/(MAP/1000))), Eref (Hargreaves reference evaporation in millimeters), PAS
(precipitation as snow, mm/year), EXT (extreme maximum temperature over 30 years), and eFFP (day of the
year on which FFP ends).

From the best fit models there were predictor variables that were most strongly correlated
with each plant response group. The most important variable for vascular plant cover on the
road centers was TSA (time since road abandonment). On the road center, exotic plant cover
was most strongly correlated with RH (relative humidity). The highest correlation for total
plant cover in the shoulder model was PAS (precipitation as snow).

48

3.5 Plant Community Ordination
Plot positions in multivariate species space portray the compositional differences
(multidimensional distances) among locations, as shown in Figure 3-22. Axis 1 accounts for
38.7% of the variation, and Axis 2 accounts for 18.7% of the variation, for a total of 57.4% in
plant community composition explained by the distance matrix, measured by Sorensen/BrayCurtis measure (Bray and Curtis 1957). Axis 3 is not shown in Figure 3-22 but included
12.9% of the variance explained, which, if combined with Axis 1 and 2, would total 70.3% of
the variation in species composition among sites. The distribution of points is guided by plant
community composition with similar plots placed close together in the biplot. The resulting
statistically significant (randomization test p= 0.004) three-dimensional ordination solution
shows the distribution of both road and shoulder sample sites, with similar plant communities
placed closer together. Axis 1 and Axis 2 represent collapsed gradients of compositional
differences in the data matrix. To test for environmental drivers of plant community
composition trends, a total of 18 climate and environmental variables were listed in the
secondary environment matrix. The distance matrix axis scores in Figure 3-22 were
significantly correlated with eight climate variables and three other environmental variables
but only those with r>0.5 are shown as red vectors in the biplot. The remaining seven
variables that showed significant correlation with NMS axis scores are listed in Table 3-7
with r<0.5 and therefore are less informative of overall plant community differences (Peck
2016, McCune and Grace 2002).

49

Figure 3-22. NMS (nonmetric multidimensional scaling) ordination of 60 averaged plots (R = road
plot, S = shoulder plot) and 254 plant species sampled on 30 abandoned road segments. Green
triangles indicate asphalt road plots, with the centroid of all asphalt plots denoted by a green cross;
red triangles indicate gravel road plots, with the centroid of all road gravel plots denotes by a red
cross. The climate and environmental variables correlated with the ordination axes are the following:
DD1040 (degree-days above 10°C and below 40°C), RH (relative humidity), PAS (precipitation as
snow), AHM (annual heat-moisture index), bFFP (day of the year the frost free period begins), EXT
(extreme maximum temperature over 30 years), and eFFP (day of the year on which FFP ends), soil
EC (electrical conductivity), soil pH, and light (canopy openness). Both axes are significant as per the
Monte Carlo Test (p = 0.004 and final stress = 15.61).

Axis 1 appears to represent the climate gradient with wet coastal sites on the left of
the biplot (Figure 3-22) and drier interior sites more towards the right. Axis 2 appears related

50

to the less significant road substrate and time variables (r<0.5 in Table 3-7) with legacies of
the disturbance history such as time since road abandonment (TSA), rooting depth, and
penetrability (compaction) being more pronounced at high Axis 2 values. All variables with
r>0.5 in Table 3-7 are represented in the biplot (Figure 3-22).
Table 3-7. Pearson's correlation coefficients of climate and environmental variables for Axis 1 and
Axis 2.

Variable
RH
eFFP
bFFP
AHM
DD1040
pH
EC
PAS
Light
EXT
Penetrability
CN Ratio
Total N
L Depth
Eref
Total C
TSA
Rooting Depth

Axis 1 (R2 = 38.7)
-0.830
-0.815
0.783
0.748
-0.674
0.652
-0.640
-0.635
0.565
-0.476
0.325
0.285
-0.260
-0.249
-0.236
-0.191
0.028
-0.064

Axis 2 (R2 = 18.7)
-0.317
-0.368
0.274
0.446
-0.257
0.320
-0.336
-0.381
-0.080
-0.176
0.158
0.127
0.091
-0.085
-0.177
0.003
0.225
0.242

Strongly correlated species associated with Axis 1 (“species loadings”) are
summarized in Table 3-8. The highest positive correlation with Axis 1 is Fragaria
virginiana. Positively correlated trees with Axis 1 are lodgepole pine (Pinus contorta) and
white spruce (Picea glauca); positively correlated shrubs are Sitka alder (Alnus viridis ssp.
sinuata), and saskatoon (Amelanchier alnifolia); positively correlated forbs include the native
American vetch (Vicia americana) and the exotic tall hawkweed (Hieracium piloselloides)

51

and the exotic white sweet-clover (Melilotus albus). Also correlated with high Axis 1 scores
are the exotic grass, smooth brome (Bromus inermis), and rock moss (Niphotrichum
ericoides) that is commonly found on disturbed sites. Negatively correlated trees with Axis 1
are red alder (Alnus rubra) and western hemlock (Tsuga heterophylla); negatively correlated
shrubs are elderberry (Sambucus racemosa), thimbleberry (Rubus parviflorus) and devil’s
club (Oplopanax horridus). Native forbs negatively correlated with Axis 1 include sweetscented bedstraw (Galium triflorum) and goatsbeard (Aruncus dioicus), and the exotic
mountain sweet cicely (Osmorhiza berteroi). One fern (Athyrium filix-femina) and a leafy
moss (Plagiomnium ciliare) are also negatively correlated with Axis 1 values.
Table 3-8. Pearson's correlation coefficients of plant species abundance from Axis 1 scores from
NMS ordination.

Axis 1 Associated Species
Negative (-)
Positive (+)
Alnus rubra
0.712 Fragaria virginiana
Tsuga heterophylla
0.492 Pinus contorta
Sambucus racemose
0.414 Bromus inermis
Galium triflorum
0.411 Vicia americana
Aruncus dioicus
0.384 Niphotrichum ericoides
Rubus parviflorus
0.375 Picea glauca
Oplopanax horridus
0.331 Melilotus albus
Plagiomnium ciliare
0.315 Hieracium piloselloides
Athyrium filix-femina
0.303 Alnus viridis ssp. sinuata
Osmorhiza berteroi
0.294 Amelanchier alnifolia

0.579
0.503
0.456
0.382
0.365
0.354
0.348
0.345
0.338
0.326

Of the species associated with Axis 2 (Table 3-9), the most positive correlation is
exhibited by smooth brome (Bromus inermis), and exotic grass. Positively correlated exotic
forbs associated with Axis 2 are common dandelion (Taraxacum officinale), Canada
goldenrod (Solidago canadensis), and oxeye daisy (Leucanthemum vulgare). There was one
positively correlated native forb, northern bedstraw (Galium boreale). There was one native

52

tree species negatively correlated with Axis 2, mountain hemlock (Tsuga mertensiana). Red
huckleberry (Vaccinium parvifolium) was the only negatively correlated native shrub. The
only negatively correlated native grass was poverty oatgrass (Danthonia spicata). Mosses
negatively correlated with Axis 2 were shaggy rock moss (Niphotrichum ericoides) and redstemmed feathermoss (Pleurozium schreberi).
Table 3-9. Pearson's correlation coefficients of plant species abundance with Axis 2 scores from NMS
ordination.

Axis 2 Associated Species
Negative (-)
Positive (+)
Niphotrichum ericoides
0.551 Bromus inermis
Pleurozium schreberi
0.365 Taraxacum officinale
Danthonia spicata
0.339 Galium boreale
Tsuga mertensiana
0.311 Solidago canadensis
Vaccinium parvifolium
0.310 Leucanthemum vulgare

0.462
0.458
0.428
0.421
0.416

4.0 Discussion
This study aimed to provide an understanding of vegetation recovery on abandoned
roads across a climate gradient in northwestern British Columbia. The research provides a
plant species list for thirty road ecosystems that have developed on abandoned Highway 16
segments over approximately 16 to 57 years. Research on natural or unassisted plant
recovery on anthropogenic disturbances in northwestern British Columbia is sparse. This
study has increased the knowledge base on ecosystem recovery after disturbance in the
Coastal Western Hemlock (CWH), Interior Cedar Hemlock (ICH), and Sub-Boreal Spruce
(SBS) biogeoclimatic zones. Ecosystem recovery is a passive type of ecosystem restoration
where natural processes allow for plant establishment and growth after a site has been
degraded, damaged, or destroyed (SER. n.d., Meffe et al. 1997). Vegetation recovery can be
measured different ways depending on research objectives and tools available, but in this
53

study plant percent cover was recorded inside small stratified random plots along each
abandoned road segment. Plant cover data were separated into growth form groups
(Appendix 1) as response variables to compare significantly correlated predictors which were
inserted into multiple regression models. This method is one of several potential ways to
summarize multi-species data. Other methods could include grouping plant species by shadetolerance, nutrient preferences, and other indicator values (Klinka et al. 1989). Plot size and
layout varies depending on whether in situ methods are used (Novak and Prach 2003,
Rehounkova 2006, Salemaa 2008, Tanentzap et al. 2009, Cutler 2011), or larger area ex situ
plots are used with desktop methods such as remote sensing (Hezhen et al. 2021).
Before sampling, it was hypothesized that increased time since road abandonment would
accelerate primary succession, thereby increasing plant cover. It was also hypothesized that
there would be more vascular plant cover on the wetter coastal locations compared to the
drier interior locations further east along the climate gradient sampled. The third hypothesis
was that plant cover, and species richness and diversity would be higher on gravel surfaces
compared to the more impenetrable asphalt road surfaces. Key findings showed that time
since road abandonment is more important to vascular and total plant cover at abandoned
road centers than climatic factors (Table 3-3). However, vascular, and total plant cover on
road shoulders (Table 3-4) and plant community composition, are more reflective of climatic
factors and are strongly driven by the coast-to-interior climate gradient (Figure 3-22, Table 37). There wasn’t a significant difference in plant diversity between gravel versus asphalt road
centers (Table 3-2).

54

4.1 Time Since Road Abandonment
The disturbed abandoned roads in this research describe vegetation recovery using the
time since abandonment (TSA) predictor (Tables A6-10 and A6-13), which was the most
significant predictor for tree cover, forb cover, overall vascular cover, and total plant cover at
road centers. Vascular plant cover on the road was also significantly influenced by the road
substrate. This makes sense because more time was needed for plant establishment where
asphalt remained, compared to gravel shoulders that had deep rooting depths suitable for
encroaching woody and herbaceous vegetation (Tables 6-4 and 6-5). Soil depth on asphalt
roads was measured as rooting depth (litter plus soil depth penetration) and varied between
sites from 0 to 3.6 centimeters (Table 6-4), which is thin for vascular plant establishment and
growth within the study’s timeline (16 to 57 years depending on site). In an abandoned road
soil and plant recovery study in Puerto Rico, soil depth on old paved roads ranged from 1 to
16 centimeters after 30 to 60 years, which significantly increased soil pools on paved roads,
creating a suitable rooting medium for vegetation recovery (Heyne, 2000).
Shannon-Wiener diversity index results indicate that TSA was one of the strongest
correlates but wasn’t statistically significant (Table 3-4). However, species richness on gravel
shoulders (Table 3-4) showed a significant positive rank correlation (rho=0.46, p=0.0098)
with TSA.
The ‘best fit’ vascular plant cover linear model included TSA (positive relationship)
and Hargreaves reference evaporation (Eref) (negative relationship) as significant variables
in generating Figure 4.1. The model is not an exact fit along the 1:1 line in Figure 4-1, as it
tends to slightly under-predict vascular cover values over 100%. Percent cover over 100%
exists because of the overlap of vegetation from different canopy and ground cover layers in
each quadrat sample, which were then summed.
55

Figure 4-1. Road vascular plant cover model (in Table 3-5 with 0.329 R2 and a p value of 0.0046)
predicted versus observed values of percent cover. This model slightly under-predicts vascular cover
for levels above 100% cover.

4.2 Substrate Differences as an Environmental Driver
Plant cover and diversity differences associated with road substrate differences were
compared using paired t-tests (Table 3-2 and Tables A2-1 and A2-2) for road centers and
gravel shoulder (n=30), asphalt road and gravel shoulder (n=11), and gravel road and
shoulder (n=19). Bryophyte cover (and hence nonvascular cover) showed the most
significant differences (lowest p values) between road substrates, especially when comparing
asphalt road center with adjacent gravel shoulders (Table 3-2). This makes sense because
during sampling it was observed that surface water would pool on the asphalt under moss,
creating a favorable wet growing environment that would not be there if the road substrate
was well drained gravel. Therefore, there was significantly more bryophyte cover on the
asphalt roads compared to gravel. Running water was also observed between the pavement
and the soil developed on abandoned roads in Puerto Rico (Heyne 2000).
56

In another study of lightly used forest roads in Vermont, USA, microtopography of
the road surface and surface water flow direction affected plant community composition,
with differences observed between road centers (elevated and designed to shed water),
shoulders (well drained), and ditches where aquatic plants persist in pooled water (Neher et
al. 2013).
Woody vegetation and vascular plant cover showed significantly different mean
values between the asphalt road and gravel shoulder. In particular, trees and shrubs
dominated the gravel shoulders where deeper rooting was possible. Impermeable asphalt
roads made it difficult for larger woody roots to establish. In other studies, high soil bulk
density was concluded to have inhibited vascular plant growth, especially deeply rooted
woody vegetation (Bolling 1996).
Paired t-test analysis comparing gravel and asphalt road surfaces revealed there was
no significant effect of substrate on Shannon-Wiener differences in diversity (Table 3-2).
Intuitively, it could be thought that gravel substrate with greater rooting depths would have
more woody plant cover and therefore more diversity, but as observed in the field, there was
higher diversity on asphalt substrate (Table 3-2). There was a richness of vascular and nonvascular plant species on asphalt roads and plant sizes were smaller, but that doesn’t mean
fewer species overall. Larger woody plant species were recorded more on shoulders, but
there was often fewer herb species under the woody vegetation’s shade. Plant species
colonized both substrates equally well in overall numbers or richness, even if the species
were vastly different from one site or climate regime to another. There were some plant
species more adapted to road asphalt and some more adapted to gravel shoulders. The road
ecosystem is linear and mostly uniform from one end to another so richness and diversity
along that line would be roughly the same, especially since quadrat data were bulked for the
57

road and for the shoulder at each site. Species diversity or evenness across the road stayed
roughly the same except where there was something different on the road surface that could
have changed moisture, shade, soil development, etc. Some examples of infrequent changes
on substrate included erosion where material washed away sections of shoulders, coarse
woody debris from a fallen tree on the road which holds moisture, soil or sediment pooling
where captured by a rock, and the odd pavement crack.
Further study of other environmental or climate factors need to be explored to explain
plant diversity on abandoned road segments in northwestern BC. For example, in Puerto
Rico, a study of soil and vegetation recovery on abandoned roads identified that some soil
parameters (pH and bulk density) and adjacent forest vegetation influenced plant
establishment (Heyne, 2000), suggesting that similar analyses in this study could provide
additional information on plant species recovery and species evenness.
4.3 Trends Along the Climate Gradient
There was a visible contrast in the field between coastal CWH biogeoclimatic sites
(e.g., Figures 3-10 and 3-11) with high plant cover on roads and the interior sites in the ICH
and SBS biogeoclimatic sites) with less cover (e.g., Figures 3-13 and 3-15), likely due to
coastal climate being wetter and more productive. Species richness and the Shannon-Wiener
diversity index were calculated to describe the number of plant taxa and plant species
evenness at each site (Table A1-8). The coastal CWH sites were generally less diverse than
drier interior sites in the ICH and SBS biogeoclimatic zones, possibly due to wetter coastal
sites creating more opportunity for dominance by a few fast-growing species compared to
drier and less productive interior sites. Richness was mostly site-specific, meaning that the
abundance of plant species was likely based on non-climate factors such as seed dispersal
from adjacent disturbed lands, and continued human disturbance on abandoned roads, which
58

was not tested. Other factors more strongly correlated with plant diversity than climate
variables were time since road abandonment and road substrate.
4.3.1 Correlation Trends Along the Climate Gradient
The Shannon-Wiener diversity index results showed that extreme maximum
temperature over 30 years (EXT) was the climate variable most strongly correlated with
diversity, but this negative relationship wasn’t statistically significant (Table 3-3). Graminoid
cover was positively correlated with the with the annual heat-moisture index (AHM) climate
variable, although grass species were recorded across the three biogeoclimatic zones in this
research. Grasses were identified on dryer interior sites and wetter coastal sites, but the
species were different, reflecting different habitat moisture and temperature requirements.
Forb cover was lower on the road than the shoulder and negatively correlated with mean
annual temperature (MAT) and degree-days above 10°C but below 40°C (DD1040), which
was evident with greater forb cover at drier interior locations. Numerous other plant recovery
studies cite temperature as one of the main drivers of plant growth on disturbed sites (Aplet
1998, Rehounkova 2006, Cutler 2011, Hamilton 2021).
Non-vascular cover (lichen and bryophytes) and exotic plant cover were more
strongly correlated with climate indices such as Hargreaves reference evaporation (Eref) and
mean annual relative humidity (RH). Non-vascular cover was positively correlated to Eref,
an estimate of potential evapotranspiration (Hargreaves and Allen 2003). This result makes
sense because of the ability of bryophytes to adapt to changes in available moisture and
withstand drought periods between rainfall events. Exotic plant cover was most strongly
correlated with interpolated estimates of RH, which means the exotic vegetation in this study
was more prominent on the wetter coastal sites. This does not mean that there weren’t exotic
species that spanned the climate gradient; for example, Hieracium species (invasive
59

hawkweeds) were found across the full range of study sites. Initially, this was somewhat
surprising, because drier interior sites were bordered by more agricultural and residential land
dominated by drought-tolerant exotic vegetation, but coastal sites were often bordered by
other disturbances such as forestry cut blocks, new stretches of Highway 16, the CN rail line,
and gravel pits, which are also capable of introducing exotic species. As with temperature,
moisture climate indices have been found to commonly influence plant growth on disturbed
sites (Aplet 1998, Rehounkova 2006, Hamilton 2021, Hezhen 2021).
4.3.2 Model Trends Along the Climate Gradient
Percent cover vegetation data from the road and shoulder were run in multiple or
single regression models (Tables A4-1 and A4-15) using predictors from the strongest
correlation coefficients (Tables 3-4 and 3-5). Some response variable plant groups produced
null (void) models due to no tested predictors having a detectable influence on vegetation
recovery on the road or shoulder. This explains why plant groups listed in Tables 3-6 and 3-7
are different with some data missing, e.g., herb cover was very low on gravel shoulders
compared to woody cover, so herb cover resulted in a null model with no predictors on the
shoulders.
The best fit road models (Table 3-5) shared five predictor variables with the shoulder
models (Table 3-6). There were three temperature-related predictor variables unique to the
road: DD1040 (degree-days above 10°C and below 40°C) contributing a negative coefficient
for forb cover, bFFP (day of the year the frost-free period begins) having a positive
coefficient for lichen cover, and Eref (Hargreaves reference evaporation, mm) with a
negative coefficient for vascular plant percent cover and a positive coefficient for bryophyte
percent cover. There were two temperature related predictor variables unique to the shoulder:
EXT (extreme maximum temperature over 30 years) having a negative effect on lichen cover
60

and eFFP (the day of the year on which frost-free period ends) having a positive effect on
tree cover. For the shoulder there was one climate moisture variable PAS (precipitation as
snow, mm/year), with positive coefficients for vascular and total plant covers.
4.4 Adjacency Effects
Several additional findings were noted during fieldwork and following data analysis.
A visual observation from fieldwork included that the forest vegetation on either side of a
road segment often appeared to have different composition and structure. This could be due
to the road having split that ecosystem in two, thereby altering surface water drainage and
collection in different locations from the original ecosystem. Historical anthropogenic
disturbances (timber harvest, land use conversion, private rangeland, right of ways, etc.)
adjacent to abandoned roads could also have altered plant community structure and
composition over time. There were a few sites without adjacent forests (where the adjacent
feature was a new highway, rail line or waterbody instead of forest).
It was originally thought that the forest would be a seed source for vegetation
recovery on the road surface but there were no significant p values or correlations (Table A33) of adjacent tree cover with any plant groups, species richness, or the Shannon-Wiener
diversity index. As discussed in Chapter 2, basal area of forest trees could have been a better
indicator of forest influence on road vegetation recovery but there were errors with the field
data, and it could not be used in the analysis. Those data gaps limited some ability to
determine forest influence on road plant recovery. From the environmental correlations,
adjacent forest tree percent cover was not a significant correlate with cover of any of the
plant groups studied, and therefore was not included as a predictor in the multiple regression
models, thereby limiting further understanding of forest influence on road vegetation. In

61

other primary succession studies, the presence of forest cover adjacent to abandoned roads
was a key driver in road plant recovery (Novak and Prach 2003, Neher et al. 2013).
Another visual observation in the field indicated a large amount of woody vegetation
in the ditch and on the shoulder compared to the road surface, a pattern also evident from the
paired t-test in Table 3-2 where there is higher woody cover on the gravel shoulder. This is
presumably due to greater moisture availability in the low-lying ditches, looser gravel
shoulder where greater root penetration was possible than on the road surface which was
more compacted (Tables A1-4 and A1-5). The woody vegetation in the ditch often leaned
heavily over the road, where more light was available than at the road edge. Ditches on either
side of the road segments often supported plant species (Table A1-1) more suited to wet
environments, reflecting the ability of the ditch to hold water runoff from the road and any
adjacent hillsides during precipitation events. The impenetrable asphalt road surfaces in this
study moved surface water onto shoulders and into ditches during rainfall, but in some cases,
water would pool in the bryophyte-dominated road cover.
4.5 Plant Community Patterns and Correlations
In the NMS (nonmetric multidimensional scaling) ordination analysis, the biplot (Figure
3-22) shows patterns of plant community similarity and correlations with the strongest
environmental predictors. A drier climate, and mostly gravel roads and shoulders are
represented on the right side of Axis 1 in the biplot, for which the dominant environmental
correlations are with soil pH, light (canopy openness), AHM (annual heat moisture index),
and bFFP (day of the year the frost-free period begins). Wetter climate and most of the
asphalt roads are located on the left side of Axis 1 where the most dominant environmental
correlates are EC (soil electrical conductivity), PAS (precipitation as snow), RH (relative
humidity), eFFP (day of the year on which FFP ends), DD1040 (degree-days above 10°C and
62

below 40°C), and EXT (extreme maximum temperature over 30 years). The higher PAS on
the coastal sites reflects the larger coastal snowpack compared to the interior, and the
oscillating winter temperatures may be responsible for the higher soil electrical conductivity
(EC) values due to greater road salt application for vehicle safety when the abandoned roads
were active. The biplot (Figure 3-22) shows younger to older sites moving up along Axis 2,
along with shorter times since abandonment (TSA). There was a similar but less pronounced
positive relationship with the environmental correlate, rooting depth on Axis 2.
Other environmental correlates may be important in explaining vegetation cover and
plant community composition. As in this study, most investigations of early succession or
vegetation recovery after disturbance measure factors similar to the soil or climatic factors
measured in this research such as soil pH (Auerbach et al. 1997, Heyne 2000, Rehounkova
and Prach 2006), soil compaction or bulk density, total soil nitrogen (Heyne 2000), substrate
or stoniness (Aplet et al. 1998, Favero-Longo 2012, Auerbach et al. 1997), snow cover or
PAS (Cutler 2011, Favero-Longo 2012) – all of which can be important to plant growth and
vegetation development. Most studies considered climatic factors such as mean annual
temperature and precipitation. Factors not measured in this research include plant
physiological traits (Diaz 1999), slope aspect and angle (Bochet and Garcia-Fayos 2004),
average wind speed (Cutler 2011), heavy metal concentrations in soil (Neher et al. 2013), and
availability of microsites for plant growth (Howard 1985). Consideration of these and many
other factors, such as the nature and frequency of post-abandonment disturbance, might have
further explained some of the variation in plant cover and community composition.

63

4.6 Research Applications and Limitations
Applications of this research stem from it being a vegetation recovery study on a severe
anthropogenic disturbance initially void of vegetation and seeds. The abandoned road
substrate was constructed and compacted similar to events that occur in nature. For example:
glacial forefields lack seeds in the soil and can also have heavily compacted gravels, newly
cooled lava flows are similar to asphalt road surfaces, and landslides remove vegetation and
can expose bare rock and sterile soil parent material. Other vegetation studies in northwestern
BC that mention non-vascular plant cover as abundant during early succession on barren
substrates include the Burton and Burton (2022) documentation of vegetation on the Nass
valley lava beds. Abandoned asphalt road surfaces are like lava rock because of their hard
and impermeable surface that absorb and reradiate heat in direct sun.
Studies of primary succession following glacial recession (Favero-Longo et al. 2012) also
mention higher nonvascular cover on impenetrable rock, which is like an asphalt road
surface. Like the glacial recession research, the abandoned gravel or asphalt road habitats
showed differences in dominant plant groups between substrates but not necessarily
represented by the highest percent cover; increased non-vascular cover on asphalt versus
higher woody cover on gravel. Some road ecology studies (Heyne 2000, Bochet and GarciaFayos 2004) mention woody vegetation growing on asphalt where small pockets of soil
become mounded and vegetation is protected from wind erosion and can establish in road
cracks, beside curbs, and against rocks, thus allowing woody cover establishment in these
microsites.
There is little research in northern BC that identifies which annual climate predictors
are strongly correlated with vegetation recovery after disturbance. One example is on the
Caribou-Chilcotin grasslands where the AHM (annual heat moisture index) predictor was
64

positively correlated with grazed and ungrazed (disturbance type) grassland graminoid
percent cover (Hamilton et al. 2021). This same AHM predictor was also reflected here, in
which graminoid percent cover also was positively correlated with AHM. The CaribouChilcotin grasslands arid grassland plant communities had more native grass species
compared to cooler and wetter sites with more exotic plant cover (forbs and graminoids). In
the research reported here, exotic plant cover was also highly correlated with wetter coastal
sites.
Limitations of this research were related to a field sampling error, a large data set not
yet fully explored, and a sample size limited by the site selection criteria along Highway 16.
Comparable forest data were needed to see if seed sources from adjacent ecosystems
impacted road vegetation recovery but there were errors in measuring tree basal area, which
would have been a better indicator than forest tree cover (Tables A3-3 and A3-6). During
sampling, percent cover vegetation was collected in ditches at each abandoned road segment,
but no analysis of ditch data was undertaken. An assessment of exotic plant cover there could
have been useful for invasive plant management on roads or linear disturbances in
northwestern BC. A larger sample size (here n=11) of road segments with intact asphalt had
been anticipated for comparison with gravel roads (sampled n=19) and would have been
desirable, but it still may not have impacted the results.
It may be important to mention that the weather was different between the two
summer fieldwork seasons. Vegetation sampling was conducted in July and August 2018,
which was hot and dry with the mean monthly temperature in Terrace being 19.2 ℃ in July
and 18.6 ℃ in August. Total rainfall in Terrace for July and August 2018 combined was 31.8
millimeters. The 2019 summer was cool and wet, with the mean temperature being 17.5 ℃ in
July and 16.9 ℃ in August in Terrace. Total rainfall in Terrace for July and August 2019
65

combined was 148.8 millimeters (ENVCAN 2022). The differences in moisture and
temperature between field data collection years could have affected percent plant cover of
some or all species between seasons.
5.0 Conclusions and Recommendations
This study shows how drivers of vegetation recovery are multifactorial and
multidimensional. This research analyzed a range of predictors against plant recovery groups
on abandoned sections of Highway 16: time since abandonment, road substrate type and
multiple annual climate variables. Despite the many environmental factors measured, the best
statistical models explained 17% to 58% of variance (Table 3-5) in the cover of different
plant groups growing on road centers. Some measure of the maritime-coastal / continentalinterior climatic gradient emerged as driving factors to most vegetation attributes tested. The
nonmetric multidimensional scaling (NMS) analysis identified several environmental and
climate variables that influenced vegetation recovery on the abandoned road segments, which
explain 57.4% in plant community composition (Table 3-7) of the plant community
composition. The disturbed road segments were undergoing a range of trajectories in early
primary succession (less than 60 years), and it will take additional decades during later
succession for environmental factors to sort out the relative abundance of plant species best
suited for each site and climate in the long run.
It was not the purpose of this study to support or refute the utility of the
chronosequence approach. As warned by Johnson and Miyanishi (2008), sample sites are
typically influenced by many environmental factors other than time since disturbance, and
this was accentuated by the climatic gradient considered here. The climate and substrate
factors limit interpretations of the successional patterns and processes reflected in this
66

relatively modest sampling program. However, it is safe to conclude that ecosystem recovery
is not being driven by early seral plant species creating suitable habitat for later seral species
in a sequential order, as late-successional species characteristic of CWH, ICH, and SBS
biogeoclimatic zones were found on sites of all ages (Meidinger and Pojar 1991). Vegetation
recovery on abandoned roads is comparable to primary succession on glacial till, where early
species colonize a site based on the availability and dispersal of plant propagules from
adjacent lands, which affects recruitment and timing of later arriving species (Johnson and
Miyanishi 2008). Multiple factors contributed to vegetation development of the abandoned
roads that didn’t follow a pattern across all the sites or even across the climate gradient.
Interactions between plants and the unique context of each road segment ecosystem were
clearly influential to vegetation recovery. Many factors not tested, such as microsite-level
heat and moisture availability, seed rain from adjacent vegetation and land uses, competition
between plants, and the type and level of repeated human use (Table 2-1) may further explain
the variation observed in the field.
There is room to explore additional environmental and historical variables related to
plant recovery in future research, such as post-abandonment disturbance regimes not well
characterized in this study. Future research on ecosystem recovery on abandoned roads could
include investigating impacts of air pollution on old road vegetation from nearby highway
vehicle traffic, studying exotic species introduction from adjacent anthropogenic linear
corridors (new highway or rail line), and exploring exotic ditch vegetation correlations with
road surface vegetation. Specific post-abandonment disturbance activities on sampled roads
post-abandonment were recorded during sampling (Table 2-1) but did not lend themselves to
quantitative analysis, which is unfortunate because it was hoped they could help explain the
67

current vegetation and inferred trajectories of succession. A disclimax is more likely to be
maintained than directional secondary succession where repeated disturbances occur on the
road ecosystems despite road abandonment.
Recommendations for assisting vegetation recovery on abandoned sections of primary
or secondary highways, forest roads, and industrial roads include:
•

Removing asphalt road substrate and ripping the compacted roadbed to decrease soil
compaction in order to improve root penetration and increase vascular and woody
plant cover;

•

Early detection and rapid response to managing invasive or exotic plants along
abandoned roads and adjacent ecosystems to prevent their escape into native
ecosystems;

•

The use of native seed mixes and transplant species as listed in Table A1-1 to assist in
successfully colonizing plant species that could be considered for use in forestry or
industrial restoration projects; and

•

Adding coarse woody debris or organic amendments to road surfaces in order to help
trap sediments and organic matter, and store water for plant growth.

The 30 abandoned road segments in this study provide opportunities for education and
demonstration of plant recovery on linear disturbances within easy walking access of
Highway 16. There could be value in protecting representative examples of abandoned road
segments as long-term study sites at which to further explore primary succession in later
stages.

68

Literature Cited
Anonymous. 1999. Forest Road Deactivation Practices in the Pacific Northwest. BC Ministry of
Forests, Info Flip Brochure. Flip Productions Limited, Nanaimo, BC.
Aplet, G.H., H.R. Flint, and P.M. Vitousek. 1998. Ecosystem development on Hawaiian lava flows:
biomass and species composition. Journal of Vegetation Science 9:17-26.
Auerbach, N., M.D. Walker, and D.A. Walker. 1997. Effects of roadside disturbance on substrate and
vegetation properties in arctic tundra. Ecological Applications 7(1):218-235.
Banner, A., W. Mackenzie, S. Haeussler, S. Thompson, J. Pojar, and R. Trowbridge. 1993. A Field
Guide to Site Identification and Interpretation for the Prince Rupert Forest Region. Land Management
Handbook 26. Victoria, BC.: British Columbia Ministry of Forests.
Bochet, E., and P. Garcia-Fayos. 2004. Factors controlling vegetation establishment and water erosion
on motorway slopes in Valencia, Spain. Restoration Ecology 12(2):166-174.
Bolling, J.D. 1996. Ecological Succession on Abandoned Roads in the Lake Mead
National Recreation Area. M.Sc. thesis in Biological Sciences. University of
Nevada, Las Vegas.
Bolling, J.D., and L.R. Walker. 2000. Plant and soil recovery along a series of abandoned desert
roads. Arid Environments 46:1-24.
Bonwick, G., D. Bustard, G. Hazelwood, K. Higgins, R. Kellerhals, M. Kent, M. Miles and D. St.
Thomas. 1992. Highway 16: Exchamisks River to Andesite Creek, Environmental Impact
Assessment. Report prepared for B.C. Ministry of Transportation and Highways. Terrace, British
Columbia. 259 p.
Bray, J.R., and J.T. Curtis. 1957. An ordination of the upland forest communities of southern
Wisconsin. Ecological Monographs 27(4):325-349.
Burnham, K.P., and D.R. Anderson. 2002 Model selection and multimodel inference: a practical
information-theoretic approach. 2nd ed. New York, Springer-Verlag.
Burton, P.J., and C.M. Burton. 2022. Stop-and-go primary succession: Episodic vegetation
development on a recent lava flow in northwestern British Columbia. Poster presented at the 2022
joint meeting of the Ecological Society of America and the Canadian Society for Ecology and
Evolution. Abstract published on-line at
https://www.eventscribe.net/2022/ESA/fsPopup.asp?Mode=posterinfo&PosterID=502572. [viewed
25 September 2022].
Cardoso, G.C., J.W. Atwell, E.D. Ketterson, and T.D. Price. 2007. Inferring performance in the songs
of dark-eyed juncos (Junco hyemalis). Behavioural Ecology 18:1051–1057.
Causton, D.R. 1988. Introduction to Vegetation Analysis. Published by the Academic Division of
Unwin Hyman Ltd. London, UK. 342 p.
Clements, F.E. 1916. Plant Succession: Analysis of the Development of Vegetation. Carnegie
Institute of Washington Publication Sciences, 242:1-512.

69

Coffin, A.W. 2007. From roadkill to road ecology: A review of the ecological effects of roads.
Journal of Transport Geography, 15(5):396–406.
Costech Analytical Technologies Ltd. 2022. ECS 4010 – CHNSO Dual Furnace. Valencia, California,
USA. Available online at https://costechanalytical.com/analytical-instruments/elementalmicroanalysis-ecs-4010-chnso-dual-furnace/.
Cutler, N. 2011. Vegetation-environment interactions in a sub-arctic primary succession. Polar
Biology 34:693-706.
Diaz, S., M. Cabido, M. Zak, E.M. Carretero, and J. Aranibar. 1999. Plant functional traits, ecosystem
structure and land-use history along a climatic gradient in central-western Argentina. Journal of
Vegetation Science 10(5):651-660.
Duddy, D. 2020. Personal Communication. Interviewed in person to locate missing legal road
surveys, gazettes and helped with technical information. [May 2020] Legal Land Survey Technician.
Terrace, British Columbia.
ENVCAN. 2022. Historical weather daily data report for Terrace, BC. Environment and Natural
Resources, Government of Canada. Ottawa, Canada. Available online at
https://climate.weather.gc.ca/climate_data/daily_data_e.html?StationID=51037&timeframe=2&Start
Year=1840&EndYear=2022&Day=5&Year=2018&Month=7#.
Favero-Longo, S., M.R. Worland, P. Convey, R.I. Lewis-Smith, R. Piervittori, M. Guglielmin, and N.
Cannone. 2012. Primary succession of lichen and bryophyte communities following glacial recession
on Signy Island, South Orkney Islands, Maritime Antarctic. Antarctic Science 24(4):323-336.
FLNRO. n.d. Forest Service Roads – Engineering. B.C. Ministry of Forest, Lands and Natural
Resource Operations, Victoria, B.C. Available at
https://www.for.gov.bc.ca/dck/Engineering/dck_engineering.htm
Forman, R.T.T., and M. Godron. 1986. Landscape Ecology. John Wiley and Sons, New York, USA.
619 p.
Forman, R.T.T., and L.E. Alexander. 1998. Roads and Their Major Ecological Effects. Annual
Review of Ecology and Systematics, 29(1): 207–231.
Forman, R.T.T., D. Sperling, J. Bissionette, A. Cleavenger, C. Cutshall, V. Dale, L. Fahrig, R. France,
C. Goldman, K. Heanue, J. Jones, F. Swanson, T. Turrentine, and T. Winter. 2003. Road Ecology –
Science and Solutions. Island Press, Washington. 481 p.
Hamilton, N., P. Sanborn, and P.J. Burton. 2021. Effects of grazing on plant communities and
successional processes vary along an aridity gradient at a northern temperate grassland. Plant Ecology
223:151-170.
Hargreaves, G.H., and R.G. Allen. 2003. History and evaluation of Hargreaves evapotranspiration
equation. Journal of Irrigation and Drainage Engineering 129(1):53-63.

70

Heyne, C.M. 2000. Soil and Vegetation Recovery on Abandoned Paved Roads in a Humid Tropical
Rain Forest, Puerto Rico. Masters Thesis. University of Nevada, Las Vegas, USA. UNLV
Retrospective Theses & Dissertations. 1146.
Hezhen, L., N. Scovronick, S. Yang, X. Ren, L. Shi, Y. Fu, M. Cai, and Y. Luo. 2021. A balance
exists between vegetation recovery and human development over the past 30 years in the Guizhou
Plateau, China. Ecological Indicators 133.
Hobbs, R.J., S. Arico, J. Aronson, J.S. Baron, P. Bridgewater, V.A. Cramer, P.R. Epstein, J.J. Ewel,
C.A. Klink, A.E. Lugo, D. Norton, D. Ojima, D.M. Richardson, E.W. Sanderson, F. Valladares, M.
Vilà, R. Zamora, and M. Zobel. 2006. Novel ecosystems: theoretical and management aspects of the
new ecological world order. Global Ecology and Biogeography, 15:1–7.
Howard, L.F. 1985. Plant Colonization on Abandoned, Elevated Highway in New York City.
University of Hawaii Press. Yearbook of the Association of Pacific Coast Geographers 47:89-104.
Johnson, E.A., and K. Miyanishi. 2008. Testing the assumptions of chronosequences in succession.
Ecology Letters 11:419-431.
Kalra, Y.P., and D.G. Maynard, 1991. Methods manual for forest soil and plant analysis. Forestry
Canada, Northwest Region, Northern Forestry Center, Edmonton, Alberta. Information Report NORX-319E. 116 p.
Klinka, K., V.J. Krajina, A. Ceska., and A.M. Scagel. 1989. Indicator Plants of Coastal British
Columbia. UBC Press, Vancouver. 288 p.
Klinkenberg, B. 2021. E-Flora BC: Electronic Atlas of the Flora of British Columbia [eflora.bc.ca].
Lab for Advanced Spatial Analysis, Department of Geography, University of British Columbia,
Vancouver, B.C.
Lindenmayer, D.B., and J. Fischer. 2006. Habitat Fragmentation and Landscape Change. CSIRO
Publishing and Island Press, Australia. 328 p.
Logan, M. 2010. Biostatistical Design and Analysis Using R: A Practical Guide. Wiley Blackwell
Publishing, West Sussex, UK. 546 p.
Lloyd, R.A., K.A. Lohse, and T.P.A. Ferre. 2013. Influence of road reclamation techniques on forest
ecosystem recovery. Frontiers in Ecology and the Environment, 11(2):75-81.
Lugo, A., and H. Gucinski. 2000. Function, Effects and Management of Forest Roads. Forest Ecology
and Management, 133(3):249-262.
Mackinnon, A., J. Pojar, and R. Coupé. 1992. Plants of Northern British Columbia. B.C. Ministry of
Forests and Lone Pine Publishing, Edmonton, AB. 344 p.
McCune, B., and L. Geiser. 2009. Macrolichens of the Pacific Northwest. 2nd edition. Oregon State
University Press, Oregon, USA. 464 p.
McCune, B., and M.J. Mefford. 2016. PC-ORD, Multivariate analysis of ecological data, Version 7
software for Windows edition.

71

Moore, G.D. 1994. Resource Road Rehabilitation Handbook: Planning and Implementation
Guidelines (Interim Methods). Watershed Restoration Technical Circular No. 3. Province of British
Columbia, Watershed Restoration Program.
MOTI. 2018. Abandoned highway 16 segments reconnaissance survey prior to and during field
research. Interviews and emails with British Columbia Ministry of Transportation and Infrastructure
staff (Grant Watson, Daena Bilodeau Cooper, and Marlene Keehn).
Meidinger, D., and J. Pojar. 1991. Ecosystems of British Columbia. Special Reports Series No.
6. British Columbia Ministry of Forests, Victoria, BC.
Meffe, G.K., C.R. Carroll, and S.L. Pimm. 1997. Community and ecosystem level conservation:
species interactions, disturbance regimes, and invading species. Principles of Conservation Biology
(eds G.K. Meffee and C.R. Carroll). Sinauer Associates, Sutherland. 236-267 p.
McCune, B., and J.B. Grace. 2002. Analysis of Ecological Communities, Third edition. MjM
Software Design, Glenden Beach, Oregon. 300 p.
Neher, D., D. Asmussen, and S.T. Lovell. 2013. Roads in northern hardwood forests affect adjacent
plant communities and soil chemistry in proportion to the maintained roadside area. Science of the
Total Environment 449:320-327.
Nolan, K.A., and J.E. Callahan. 2006. Beachcomber biology: The Shannon-Wiener Species Diversity
Index. Tested Studies for Laboratory Teaching, (M.A. O'Donnell, Editor). Proceedings of the 27th
Workshop/Conference of the Association for Biology Laboratory Education (ABLE) 27:334-338.
Novak, J., and K. Prach. 2003. Vegetation succession in basalt quarries: pattern on a landscape scale.
Applied Vegetation Science 6:111-116.
Peck, J. 2016. Multivariate Analysis for Ecologists: Step-by-Step, Second edition. MjM Software
Design, Glenden Beach, Oregon. 192 p.
Pojar, J., and A. Mackinnon. 1994. Plants of Coastal British Columbia. B.C. Ministry of Forests and
Lone Pine Publishing, Vancouver, BC. 527 p.
Prach. K., and L.R. Walker. 2018. Differences between primary and secondary plant succession
among biomes of the world. Journal of Ecology 107(2):510-516.
Rehounkova, K., and K. Prach. 2006. Spontaneous vegetation succession in disused gravel-sand pits:
role of local site and landscape factors. Journal of Vegetation Science 17:583-590.
Royer, F., and R. Dickson. 2006. Weeds of Canada and the Northern United States. The University of
Alberta Press, Edmonton, AB and Long Pine Publishing, Vancouver, BC. 434 p.
R Studio Team. 2022. R Studio: Integrated Development for R Studio, PCB, Boston, MA. Available
online at http://www.rstudio.com/.
Salemaa, M., J. Derome, and P. Nojd. 2008. Response of boreal forest vegetation to the fertility status
of the organic layer along a climate gradient. Boreal Environmental Research 13 (suppl. B):48-66.

72

Schofield, W.B. 1992. Some Common Mosses of British Columbia. Royal British Columbia
Museum, Victoria, BC. 394 p.
SER. n.d. What is ecological restoration? The Society for Ecological Restoration. Available online at
https://www.ser-rrc.org/what-is-ecological-restoration/ (viewed October 30, 2022).
Smith, R.L., and T.M. Smith. 1998. Elements of Ecology, Fourth edition. The Benjamin/Cummings
Publishing Company, Menlo Park, California, USA. 555 p.
Symonds, R.E.M., and A. Moussalli. 2011. A brief guide to model selection, multimodel inference
and model averaging in behavioral ecology using Akaike’s information criterion. Behavioral Ecology
Sociobiology 65:13-21.
Tanentzap, A.J., L.E. Burrows, W.G. Lee, G. Nugent, J.M. Maxwell, and D.C. Coomes. 2009.
Landscape-level vegetation recovery from herbivory: progress after four decades of invasive red deer
control. Journal of Applied Ecology 46(5):1064-1072.
Terry, R.D., and G. Chilingar. (1955). Comparison charts for visual estimation of foliage cover.
Society of Economic Paleontology and Minerology. Sedimentary Petrology, 25(3):229-234.
Tilman, D. 1988. Plant Strategies and the Dynamics and Structure of Plant Communities. Princeton
University Press, New Jersey. 360 p.
Trammell, T.L.E., and M.M. Carreiro. 2011. Vegetation composition and structure of woody plant
communities along urban interstate corridors in Louisville, KY, U.S.A. Urban Ecosystems,
14(4):501–524.
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.R., D. Wardle, R. Bardgett, and B.D. Clarkson. 2010. The use of chronosequences in
studies of ecological succession and soil development. Ecology 98:725-736.
Walker, L. 2012. The Biology of Disturbed Habitats. Oxford University Press, New York. 319 p.
Wang, T., A.D. Hamann, D. Spittlehouse, and C. Carroll. 2016. Locally downscaled and spatially
customizable climate data for historical and future periods for North America. Locally Downscaled
and Spatially Customizable Climate Data for Historical and Future Periods for North America. PLoS
ONE 11(6): e0156720.
Whitson, T.D., L.C. Burrill, S.A. Dewey, D.W. Cudney, B.E. Nelson, R.D. Lee, and R. Parker. 2009.
Weeds of the West. 10th edition. The Western Society of Weed Science, Western United States Land
Grant Universities Cooperative Extension Services, and The University of Wyoming, USA. 628 p.
Wickham, H., M. Averick, J. Bryan, W. Chang, L.D. McGowan, R. François, G. Grolemund, A.
Hayes, L. Henry, J. Hester, M. Kuhn, T.L. Pedersen, E. Miller, S.M. Bache, K. Müller, J. Ooms, D.
Robinson, D.P. Seidel, V. Spinu, K. Takahashi, D. Vaughan, C. Wilke, K. Woo, and H. Yutani. 2019.
“Welcome to the tidyverse.” Journal of Open Source Software, 4(43), 1686.
Wickham, H, R., L. François, and H.K. Müller. 2022. dplyr: A Grammar of Data Manipulation.
https://dplyr.tidyverse.org, https://github.com/tidyverse/dplyr.

73

Appendix 1. Species and vegetation summary.
All plant species were organized into plant groups for use in multi-factorial models. E-Flora
BC (Klinkenberg 2021) was used to categorize each plant. Plant groups are as follows:
•
•
•
•
•
•
•
•
•
•
•
•

Tree cover (single-stemmed woody plants capable of achieving heights >5 m)
Shrub cover (multi-stemmed woody plants <5 m in height)
Woody cover (the sum of tree and shrub cover)
Forb cover (broadleaf herbs)
Fern and fern allies cover (vascular plants that reproduce by spores, such as ferns,
horsetails, and club mosses)
Graminoid cover (grasses, rushes, and sedges)
Vascular plant cover (all plants except non-vascular such as bryophytes and
lichen)
Bryophyte cover (mosses and liverworts)
Lichen cover (ground cover lichens)
Non-vascular cover (bryophytes and lichens)
Exotic plant cover (species exotic to British Columbia, both domesticated and
invasives)
Total plant cover

74

Abies sp.
Acer glabrum
Alnus incana ssp. tenuifolia
Alnus rubra
Betula papyrifera
Malus fusca
Malus pumila*
Picea glauca
Picea sitchensis
Picea sp.
Pinus contorta
Populus balsamifera ssp.
trichocarpa
Populus tremuloides
Prunus emarginata
Salix c.f. scouleriana
Thuja plicata
Tsuga heterophylla
Tsuga mertensiana
Xanthocyparis nootkatensis
Shrubs:
Alnus viridis ssp. sinuata
Amelanchier alnifolia
Arctostaphylos uva-ursi
Corylus cornuta

Plant Species
Trees:

Frequency
0.08
0.14
1
19.5
1.9
0
0
1.3
3.1
0.33
4.4
6.5
0.74
0
0.28
2.0
5.2
0.18
0
2.7
0.33
1.4
0

2
5
1
16
7
0
0
4
10
1
8
17
3
0
2
12
16
1
0
6
3
3
0

Mean
Cover %

Road, n=30

6
4
2
0

19
3
1
0
17
21
1
1

4
4
1
17
9
0
0
4
12
1
9

Frequency

75

2.2
0.54
0.5
0

6.6
1.4
0.01
0
7.3
12.2
0.4
0.08

0.68
1.5
1.4
16.9
2.3
0
0
2.0
5.5
0.79
3.7

Mean
Cover %

Shoulder, n=30

5
9
2
1

22
8
1
0
21
20
1
1

6
14
1
22
12
0
1
5
18
4
8

Frequency

1.5
2.1
1.4
0.017

7.9
1.3
0.03
0
8.1
10.3
0.33
0.08

1.1
1.7
0.33
21.1
2.8
0
0.08
1.3
7.7
0.32
4.6

Mean
Cover %

Ditch, n=30

1
5
3
0

15
4
0
0
21
21
0
0

9
13
2
16
13
1
0
5
12
4
8

Frequency

2.2
1.6
0
0
4.2
10.7
0
0

2.3
0.36
0.09
3.5
1.5
0
0
1.2
3.1
0
3.6

Mean Basal
Area, m2/ha

Forest, n=28

0.45
2.1
0.45
0

4.6
4.4
0
0
12.0
11.4
0
0

2.9
4.2
0.4
7.5
2.2
0.04
0
2.2
0.8
0.59
4.3

Mean
Understory
Cover %

Table A1-1. Species list and overall representation of vegetation species encountered across the study area, organized by growth form in each of the sampled habitats. Values denote frequency
as a percentage of the locations at which a species occurred and mean percent cover across all locations (including those with zero cover of the species). Overstory cover by species was not
evaluated for adjacent forest sites, for which overstory species dominance is indicated by basal area instead. The shaded area denotes N/A.

Cornus sericea
Dryas drumondii
Gaultheria shallon
Juniperus communis
Lonicera involucrata
Menziesia ferruginea
Oplopanax horridus
Paxistima myrsinites
Prunus virginiana
Ribes bracteosum
Ribes lacustre
Ribes laxiflorum
Rosa acicularis
Rosa nutkana
Rosa woodsii
Rubus idaeus
Rubus leucodermis
Rubus parviflorus
Rubus pubescens
Rubus spectabilis
Salix bebbiana
Salix glauca var. acutifolia
Salix lucida
Salix sp.
Salix sitchensis
Sambucus racemosa
Shepherdia canadensis
Sorbus sitchensis
Spiraea douglasii
Spiraea pyramidata
Symphoricarpos albus
Vaccinium alaskaense
Vaccinium caespitosum
Vaccinium membranaceum

7
0
0
0
5
0
2
0
0
1
4
2
4
3
0
3
0
14
2
10
0
1
2
1
3
8
6
1
1
0
2
0
0
1

1.9
0
0
0
1
0
0.39
0
0
0.01
0.19
0.12
1.8
0.11
0
0.26
0
5.8
0.53
3.1
0
0.03
0.03
0.03
0.24
1.6
0.93
0.05
0.03
0
0.38
0
0
0.01

9
0
0
0
7
1
4
1
1
0
4
3
2
3
1
3
0
15
2
7
1
0
2
0
2
11
3
0
0
0
3
1
0
1
76

1.8
0
0
0
0.78
0.01
0.53
0.13
0.07
0
0.13
0.66
0.33
0.42
0.17
0.26
0
7.4
0.11
2.5
0.03
0.01
0.1
0
0.67
6.0
0.73
0
0
0
0.29
0.07
0
0.05

17
2
0
1
8
4
9
7
0
1
10
5
6
8
1
5
1
21
1
9
2
3
6
0
2
12
6
2
3
0
5
0
0
4

4.4
0.22
0
0.02
0.75
0.5
0.7
0.78
0
0.03
1.6
0.25
2.0
0.86
0.33
0.35
0.03
5.4
0.03
2.6
0.33
0.9
1.5
0
0.18
4.0
0.68
0.03
0.23
0
1.4
0
0
0.21

9
1
1
0
4
7
10
7
0
1
7
0
5
5
0
3
0
17
1
6
2
2
5
0
0
7
5
1
1
1
5
1
1
6

1.5
1.6
0.09
0
0.2
2.0
0.98
3.2
0
0.004
0.43
0
1.1
0.43
0
0.28
0
3.2
0.05
2.3
0.14
0.29
0.69
0
0
1.9
0.46
0.02
0.41
0.09
0.66
0.11
0.13
0.36

Aruncus dioicus
Asarum caudatum
Astragalus cicer*
Calypso bulbosa
Capsella bursa-pastoris
Castilleja miniata
Cerastium fontanum
Chimaphila umbellata
Circaea alpina
Cirsium arvense*
Cirsium palustre
Claytonia sibirica
Clintonia uniflora
Comandra umbellata
Collinsia parviflora
Cornus canadensis
Crepis tectorum
Cynoglossum boreale
Delphinium glaucum
Epilobium angustifolium
Epilobium brachycarpum
Epilobium ciliatum
Euphrasia nemorosa
Eurybia conspicua
Fragaria vesca
Fragaria virginiana
Galeopsis tetrahit*
Galium boreale

Forbs:

Vaccinium ovalifolium
Vaccinium parvifolium
Vaccinium sp.
Vaccinium vitis-idaea
Viburnum edule
9
0
0
1
0
3
4
2
1
0
2
1
0
1
1
4
2
0
0
4
1
4
1
2
1
8
7
2

2
3
1
0
4
1.9
0
0
0.002
0
0.04
0.21
0.07
0.06
0
0.17
0.03
0
0.01
0.01
0.45
0.17
0
0
0.11
0.002
0.28
0.02
0.15
0.03
0.59
0.13
0.02

0.03
0.03
0.01
0
0.11
8
0
0
1
0
2
1
2
2
0
0
1
1
0
0
6
0
1
0
7
0
1
1
2
0
9
6
1

2
1
1
0
1

77

1.4
0
0
0.002
0
0.19
0.01
0.19
0.02
0
0
0.003
0.03
0
0
0.45
0
0.2
0
0.38
0
0.01
0.01
0.16
0
0.48
0.53
0.02

0.09
0.01
0.03
0
0.05
13
1
0
0
1
5
1
4
4
2
2
0
6
0
0
7
1
1
1
10
0
1
1
2
0
7
4
1

3
5
1
0
6
2.3
0.02
0
0
0.01
0.27
0.01
1.8
0.18
0.25
0.07
0
0.43
0
0
0.71
0.05
0.03
0.03
1.4
0
0.004
0.03
0.28
0
1.4
0.95
0.04

0.48
0.36
0.03
0
0.67
6
1
1
0
0
4
0
7
0
0
0
0
12
0
0
13
1
1
0
6
0
1
0
3
0
6
0
3

3
10
1
1
6

0.22
0.02
0.01
0
0
0.32
0
2.1
0
0
0
0
2.8
0
0
1.9
0.02
0.02
0
0.44
0
0.05
0
0.29
0
1.08
0
0.16

0.36
1.4
0.02
0.01
0.36

Galium triflorum
Geocaulon lividum
Geranium erianthum
Geranium richardsonii
Geum macrophyllum
Goodyera oblongifolia
Hieracium albiflorum
Hieracium aurantiacum*
Hieracium gracile
Hieracium maculatum*
Hieracium c.f. praealtum
Hieracium piloselloides
Hieracium sabaudum
Hieracium sp.
Hieracium umbellatum
Heracleum maximum
Heuchera micrantha
Hypericum perforatum
Lathyrus nevadensis
Lathyrus ochroleucus
Leucanthemum vulgare*
Lindernia dubia
Linnaea borealis
Lysichiton americanus
Maianthemum dilatatum
Maianthemum racemosum
Maianthemum stellatum
Medicago lupulina*
Medicago sativa*
Melampyrum lineare
Melilotus albus*
Melilotus officinalis*
Mentha arvensis
Monotropa hypopithys

13
0
2
0
9
2
1
1
1
3
0
7
2
1
2
1
1
0
4
2
8
0
4
0
2
1
4
3
2
1
3
1
1
0

0.41
0
0.05
0
1.4
0.14
0.01
0.01
0.04
0.5
0
1.3
0.3
0.07
0.04
0.05
0.01
0
0.27
0.35
1.3
0
0.08
0
0.40
0.05
0.22
0.43
0.43
0.01
0.09
0.01
0.13
0

10
0
2
0
4
2
1
0
2
3
0
6
3
1
2
2
0
0
6
1
7
0
3
0
1
2
4
0
1
0
1
0
0
0
78

0.35
0
0.05
0
0.28
0.06
0.02
0
0.05
0.19
0
0.29
0.41
0.01
0.08
0.3
0
0
0.8
0.01
0.24
0
0.2
0
0.6
0.07
0.36
0
0.01
0
0.02
0
0
0

17
1
1
1
4
3
3
0
2
2
0
5
0
0
0
6
1
1
7
2
9
0
4
1
3
9
5
1
2
1
2
0
0
0

1.3
0.08
0.08
0.05
0.19
0.09
0.02
0
0.06
0.33
0
0.45
0
0
0
0.7
0.05
0.004
0.83
0.13
0.63
0
0.3
0.33
0.25
0.47
0.58
0.2
0.22
0.05
0.07
0
0
0

11
1
1
3
1
8
0
0
0
0
1
2
0
1
1
2
0
0
6
2
4
1
6
1
0
5
9
1
2
0
0
0
0
1

1.4
0.09
0.13
0.16
0.08
0.17
0
0
0
0
0.09
0.07
0
0.01
0.004
0.1
0
0
0.79
0.7
0.34
0.04
0.07
0.1
0
0.37
0.56
0.07
0.3
0
0
0
0
0.04

Monotropa uniflora
Mycelis muralis*
Oenanthe sarmentosa
Orthilia secunda
Osmorhiza berteroi
Plantago major
Prosartes hookeri
Prunella vulgaris
Pyrola asarifolia
Ranunculus acris*
Ranunculus sp.
Ranunculus occidentalis
Ranunculus uncinatus
Sanicula marilandica
Senecio triangularis
Solanum dulcamara*
Solidago canadensis
Spiraea betulifolia
Spiranthes romanzoffiana
Stachys cooleyae
Stellaria calycantha
Streptopus amplexifolius
Streptopus lanceolatus
Symphyotrichum ciliolatum
Symphyotrichum foliaceum
Tanacetum vulgare*
Taraxacum officinale*
Tellima grandiflora
Thalictrum occidentale
Tiarella trifoliata
Trientalis europaea
Urtica dioica
Veratrum viride
Viola canadensis

0
7
0
0
2
4
1
6
4
4
0
1
1
0
1
0
3
0
1
0
1
0
1
1
0
3
10
3
1
4
1
1
0
0

0
0.83
0
0
0.03
0.09
0.02
0.72
1.1
0.16
0
0.01
0.01
0
0.01
0
0.14
0
0.01
0
0.01
0
0.05
0.03
0
0.32
1.2
0.11
0.02
0.12
0.01
0.01
0
0

1
4
0
2
2
1
0
0
5
0
0
0
0
0
0
1
2
2
0
0
0
0
4
3
1
2
7
0
2
3
0
0
0
1
79

0.01
0.08
0
0.05
0.03
0.01
0
0
0.72
0
0
0
0
0
0
0.002
0.07
0.07
0
0
0
0
0.39
0.04
0.07
0.77
0.19
0
0.02
0.09
0
0
0
0.05

1
9
1
3
1
1
4
1
7
0
0
0
0
1
0
0
3
0
0
2
0
0
7
4
3
3
11
4
4
4
0
1
1
0

0.05
0.29
0.08
0.09
0.004
0.02
0.38
0.03
1.2
0
0
0
0
0.1
0
0
0.22
0
0
0.06
0
0
0.25
0.3
0.2
0.25
0.63
0.32
0.5
0.28
0
0.03
0.02
0

1
2
0
2
0
2
3
0
6
0
1
0
0
1
0
1
4
0
0
0
0
1
8
3
2
0
7
1
4
4
0
0
1
1

0.05
0.39
0
0.13
0
0.21
0.83
0
0.94
0
0.004
0
0
0.09
0
0.004
0.38
0
0
0
0
0.18
0.91
0.46
0.23
0
0.67
0.05
0.45
0.42
0
0
0.05
0.05

Viola glabella
Trifolium hybridum*
Trifolium pratense*
Trifolium repens*
Vicia americana
Vicia cracca*
(unknown dicot)
Graminoids:
Agrostis capillaris
Agrostis exarata
Agrostis gigantea
Agrostis scabra
Agrostis sp.
Agrostis stolonifera*
Bromus anomalus
Bromus ciliatus
Bromus inermis*
Bromus vulgaris
Calamagrostis canadensis
Carex c.f. deweyana
Carex disperma
Carex kelloggii
Carex mertensii
Carex sp.
Cinna latifolia
Dactylis glomerata*
Danthonia spicata
Elymus glaucus
Elymus hirsutus
Elymus repens*
Festuca rubra
Festuca sp.
Festuca subulata
Hierochloe hirta

0
2.4
0
2.3
0.09
0
0.03
0
0.11
1.3
0.93
1.0
0.11
0.31
0
0.55
0
0.12
0
0.25
0
0.05
0.11
0.09
0.01
0.11
0.5
0
0.2
2.1
0.04
0
0.04

0
5
0
7
3
0
3
0
2
1
3
3
3
1
0
3
0
2
0
1
0
1
2
2
1
1
7
0
3
4
1
0
1

0
0
1
3
1
1
1
0
3
0
0
2
1
0
0
1
1
0
1
5
0
1
2
1
0
0

1
4
1
0
3
1
5

80

0
0
0.55
0.43
0.01
0.14
0.01
0
0.62
0
0
0.47
0.02
0
0
0.07
0.01
0
0.15
0.09
0
0.01
0.03
0.002
0
0

0.01
0.15
0.01
0
0.15
0.01
0.027
0
0
1
2
0
3
1
0
4
2
3
1
0
1
0
0
2
1
1
4
1
5
3
0
1
0

2
5
1
1
4
0
2
0
0
0.33
0.01
0
0.15
0.17
0
0.98
0.23
1.3
0.03
0
0.13
0
0
0.04
0.004
0.05
0.28
0.1
0.13
0.42
0
0.05
0

0.1
0.4
0.02
0.01
0.38
0
0.07
1
0
1
0
0
0
1
0
2
0
3
0
0
0
0
0
0
0
0
1
0
2
2
0
0
0

2
3
0
2
4
0
1

0.01
0
0.27
0
0
0
0.18
0
0.63
0
0.34
0
0
0
0
0
0
0
0
0.2
0
0.38
2.3
0
0
0

0.11
0.22
0
0.27
0.34
0
0.004

Lolium perenne*
Phalaris arundinacea
Phleum pratense*
Poa compressa
Poa palustris
Poa pratensis*
Podagrostis aequivalvis
Trisetum cernuum
Trisetum spicatum
Scirpus microcarpus
(unknown grass)
Ferns & Fern Allies:
Athyrium filix-femina
Blechnum spicant
Cryptogramma acrostichoides
Dryopteris expansa
Gymnocarpium dryopteris
Phegopteris connectilis
Polypodium glycyrrhiza
Polystichum braunii
Polystichum munitum
Pteridium aquilinum
Equisetum arvense
Lycopodium annotinum
Lycopodium clavatum
Lycopodium dendroideum
Bryophytes:
Amblystegium serpens
Antitrichia curtipendula
Aulacomnium palustre
Barbilophozia barbata
Bartramia ithyphylla
Brachythecium asperrimum
Brachythecium erythrorrhizon

0
0.07
0.05
0.74
0.11
4.4
0.03
0.02
0.02
0.06
0.13
2.6
0
0
0.25
0
0.13
0
0
0
0.05
0.68
0
0.05
0.03
0.01
0
0.04
0.83
0
0.02
0.01

0
1
3
2
5
8
1
1
1
1
1
9
0
0
4
0
1
0
0
0
2
4
0
1
1
1
0
1
3
0
2
1

0
0
2
1
1
0
1

8
0
0
4
0
1
0
0
0
3
7
0
0
0

1
0
0
1
2
6
0
0
0
0
0

81

0
0
0.02
0.01
0.01
0
0.01

1.6
0
0
0.87
0
0.02
0
0
0
0.22
0.56
0
0
0

0.003
0
0
0.27
0.03
0.11
0
0
0
0
0

0
0
2
0
0
1
0

18
1
1
9
2
2
2
1
4
3
12
2
1
1

0
2
3
1
2
4
0
1
1
1
0

0
0
0.45
0
0
0.17
0

3.7
0.37
0.12
2.0
0.08
0.18
0.03
0.02
0.13
0.6
2.3
0.07
0.02
0.01

0
1.2
0.32
0.01
0.04
0.47
0
0.01
0.08
0.13
0

0
1
1
1
0
0
0

9
1
0
8
11
1
1
0
1
3
6
1
0
2

0
0
2
0
3
1
0
0
1
0
0

0
0.004
0.04
0.18
0
0
0

1.5
0.63
0
1.0
0.54
0.04
0.01
0
0.18
0.2
1.3
0.09
0
0.09

0
0
0.21
0
0.21
0.04
0
0
0.09
0
0

Brachythecium frigidum
Brachythecium leibergii
Bryum calobryoides
Calliergonella cuspidata
Cephalozia macounii
Ceratodon purpureus
Climacium dendroides
Dicranum howellii
Dicranum pallidisetum
Dicranum polysetum
Dicranum scoparium
Dicranum sp.
Drepanocladus aduncus
Drepanocladus sordidus
Eurynchiastrum pulchellum
Heterocladium procurrens
Hygrohympnum eugyrium
Hylocomium splendens
Hypnum dieckii
Isothecium myosuroides
Leucolepis acanthoneuron
Meiotrichum lyallii
Mnium spinulosum
Niphotrichum canescens
Niphotrichum elongatum
Niphotrichum ericoides
Orthotrichum lyellii
Pellia sp.
Philonotis capillaris
Plagiochila porelloides
Plagiomnium sp.
Plagiomnium ciliare
Plagiomnium cuspidatum
Plagiomnium ellipticum

6
1
1
3
0
2
3
1
0
1
2
0
0
1
1
1
1
13
1
2
0
0
1
1
1
8
0
1
1
1
2
6
7
4

0.16
0.23
0.003
0.32
0
0.08
0.16
0.01
0
0.01
0.02
0
0
0.04
0.02
0.13
0.09
1.9
0.19
0.05
0
0
0.01
0.09
0.02
6.9
0
0.42
0.06
0.08
1.5
2.3
2.0
0.28

3
1
0
1
0
1
0
0
0
0
2
1
0
0
0
0
0
9
1
1
0
1
0
0
1
5
0
1
0
0
2
6
3
2
82

0.02
0.02
0
0.01
0
0.01
0
0
0
0
0.02
0.002
0
0
0
0
0
1.4
0.01
0.03
0
0.01
0
0
0.01
2.6
0
0.03
0
0
1.7
1.8
0.77
0.01

0
0
0
0
1
0
1
0
0
0
0
0
1
0
2
0
0
12
0
2
1
0
0
0
1
3
0
1
0
0
1
5
4
0

0
0
0
0
0.02
0
0.13
0
0
0
0
0
0.03
0
0.1
0
0
1.9
0
0.07
0.33
0
0
0
0.08
1.4
0
0.08
0
0
0.33
1.2
0.18
0

0
0
0
0
1
0
2
0
1
1
1
2
0
0
2
0
0
17
0
2
1
0
1
0
0
1
1
1
0
1
2
2
1
0

0
0
0
0
0.18
0
0.21
0
0.004
0.14
0.04
0.19
0
0
0.27
0
0
11.4
0
0.23
1.4
0
0.01
0
0
0.18
0.04
0.18
0
0.01
0.59
0.38
0.02
0

Plagiomnium insigne
Plagiomnium medium
Plagiothecium laetum
Pleurozium schreberi
Pogonatum contortum
Pohlia drummondii
Pohlia filum
Polytrichastrum alpinum
Polytrichum commune
Polytrichum juniperinum
Polytrichum piliferum
Porella cordaeana
Pseudoleskeella rupestris
Ptilium crista-castrensis
Ptychostomum
pseudotriquetrum
Rhizomnium glabrescens
Rhynchostegium aquaticum
Rhytidiadelphus loreus
Rhytidiadelphus subpinnatus
Rhytidiadelphus squarrosus
Rhytidiadelphus triquetrus
Sanionia uncinata
Sarmentypnum exannulatum
Sarmentypnum sarmentosum
Scapania bolanderi
Schistochilopsis incisa
Sciurohypnum latifolium
Sciurohypnum plumosum cfr.
Sphagnum girgensohnii
Thuidium recognitum
(unknown moss)
Lichens:
Bilimbia sabuletorum

0.23
0
0.01
0.82
0.003
0.003
0.06
0.03
0.13
0.38
0.03
0
0.23
0.03
0.13
0
0.003
0.53
0.02
0.95
3.2
0.1
0.04
0.02
0
0.02
0.18
0.64
0
0
0
0.01

3
0
1
6
1
1
1
2
1
6
1
0
1
2
1
0
1
4
1
2
12
3
1
1
0
1
3
1
0
0
0
1

0

0
0
0
5
0
2
9
3
0
0
0
1
0
0
0
0
0

1
1
1
6
0
0
0
1
0
2
1
0
1
2

83

0

0
0
0
0.05
0
0.17
1.5
0.01
0
0
0
0.002
0
0
0
0
0

0.1
0.01
0.01
0.22
0
0
0
0.02
0
0.04
0.1
0
0.02
0.01

0

0
1
0
4
0
1
13
0
0
0
0
0
0
0
1
1
0

1
2
1
2
0
0
0
1
0
2
0
0
0
2

0

0
0.08
0
0.8
0
0.75
1.4
0
0
0
0
0
0
0
0.08
0.1
0

1.1
0.03
0.02
0.23
0
0
0
0.08
0
0.05
0
0
0
0.37

0

0
1
0
6
0
1
13
1
0
0
1
0
0
0
1
1
1

1
1
0
4
0
0
0
2
0
1
0
0
0
2

0

0
1.2
0
0.6
0
0.18
3.8
0.07
0
0
0.02
0
0
0
0.18
0.05
0.02

0.07
0.07
0
0.52
0
0
0
0.25
0
0.05
0
0
0
0.32

Cladonia amaurocrea
1
Cladonia asahinae
0
Cladonia coccifera
1
Cladonia deformis
1
Cladonia ecmocyna ssp.
intermedia
1
Cladonia multiformis
1
Cladonia pyxidata
1
Cladonia scabriuscula
1
Cladonia squamosa
0
Cladonia stricta
1
Cladonia symphyicarpia
1
Cladonia verruculosa
0
Lobaria oregana
1
Lobaria pulmonaria
0
Peltigera aphthosa
0
Peltigera britannica
1
Peltigera canina
0
Peltigera conspersa
1
Peltigera kristinssonii
1
Peltigera lepidophora
1
Peltigera leucophlebia
1
Peltigera membranacea
0
Peltigera neopolydactyla
0
Peltigera rufescens
1
Stereocaulon tomentosum
3
(unknown lichen)
0
* exotic to British Columbia, as per Klinkenberg (2021)

0
0
0
1
1
0
0
0
1
0
1
1
1
0
0
2
0
1
0
1
1
0
0
0
2
2

0.002
0
0.002
0.03
0.01
0.03
0.08
0.01
0
0.01
0.02
0
0.05
0
0
0.01
0
0.17
0.01
0.23
0.08
0
0
0.01
0.31
0

84

0.02
0
0
0
0.01
0
0.03
0.002
0.01
0
0
0.09
0
0.03
0
0.003
0.01
0
0
0
0.06
0.11

0
0
0
0.01
0
0
0
0
0
0
0
1
0
0
1
1
1
0
0
0
1
1
1
0
2
0

0
1
0
0
0
0
0
0
0
0
0
0.004
0
0
0.17
0.1
0.17
0
0
0
0.02
0.03
0.01
0
0.35
0

0
0.004
0
0
1
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
1
1
0
0
1
0

0
0
0
0

0.05
0
0
0
0
0
0
0
0
0.07
0.18
0
0
0
0
0
0.04
0.04
0
0
0.02
0

0
0
0
0

85

Site Name
Abbreviation Highway 16 km
Latitude
Longitude
Elevation (meters) BEC* Unit
Witset
MC
308
55.08203746 -127.3481765
425
ICHmc2
La Verge
LV
155
54.55582499 -128.4599286
55
CWHws1
Shames West
SW
123
54.41328903 -128.8956679
42
CWHws1
Delta Creek
DC
128
54.43502184 -128.8396496
62
CWHws1
Exstew
Ex
113
54.40093806 -129.0527434
23
CWHws1
SE of Exchamsiks Park
SE EP
96
54.33555062 -129.2666054
15
CWHvm1
Big Oliver West
BOW
190
54.81368129 -128.2830206
144
ICHmc2
30Mile
30M
100
54.33905626 -129.2014437
25
CWHvm1
Zymacord
Zm
137
54.48536771 -128.7366966
60
CWHws1
Exchamsiks Loop
EL
94
54.335257
-129.2924873
26
CWHvm1
Little Oliver
LO
189
54.80496833 -128.2701842
155
ICHmc2
Flint Creek
FC
210
54.9635563
-128.378019
183
ICHmc2
Telkwa
Tk
362
54.71717543 -127.0684217
505
SBSdk
Houston East
HE
419
54.4509263
-126.5774625
620
SBSdk
Priestly Hill
PH
523
54.12694097 -125.3487459
763
SBSmc2
Sheraton Mill
SM
513
54.17812838 -125.4749398
707
SBSmc2
Suskwa
Sw
291
55.20929378 -127.4398197
353
ICHmc2
Shames East
SE
123
54.4141382
-128.8904421
44
CWHws1
Cut Block
CB
130
54.44956651 -128.8166752
65
CWHws1
Kleanza
Kz
161
54.59511193 -128.3979014
122
CWHws1
Chimdemash West
CW
170
54.66013005
-128.37409
124
CWHws1
Chimdemash East Transect 1
CET1
170
54.65996725
-128.373891
114
CWHws1
Chimdemash East Transect 2
CET2
170
54.66215488 -128.3728807
113
CWHws1
St Croix
SC
175
54.6985018
-128.3217352
139
CWHws1
Legate
Lg
183
54.74985633 -128.2577209
128
ICHmc2
Rainbow Pass
RP
32
54.23974214 -130.0376976
145
CWHvh2
Research Farm Smithers
RFS
354
54.73859516 -127.1112104
511
SBSdk
Yellow Cedar Lodge
YCL
136
54.48171383 -128.7469531
99
CWHws1
Exchamsiks Provincial Park
EPP
94
54.33507129 -129.2985916
16
CWHvm1
Big Oliver East
BOE
191
54.81733882 -128.2906437
150
ICHmc2
*Biogeoclimatic ecosystem classification, as portrayed in maps at https://www.for.gov.bc.ca/hre/becweb/resources/maps/.

Road Substrate Abandonment (years) Disturbance Scale
gravel
26
4
gravel
55
1
asphalt
43
1
asphalt
43
4
asphalt
41
3
asphalt
46
4
gravel
49
0
asphalt
26
0
asphalt
45
2
asphalt
46
0
gravel
47
1
gravel
47
3
gravel
48
1
gravel
52
1
gravel
57
3
gravel
53
0
gravel
55
2
gravel
42
3
asphalt
43
0
gravel
51
0
gravel
52
2
gravel
52
0
gravel
52
0
asphalt
52
0
asphalt
46
1
gravel
41
4
gravel
35
1
asphalt
43
1
gravel
16
1
gravel
48
0

Table. A1-2. Study sites and characteristics of abandoned road segments (n=30) sampled in 2018 and 2019 along Highway 16 in northwestern British Columbia, Canada.

Site Name
Witset
La Verge
Shames West
Delta Creek
Exstew
SE of Exchamsiks Park
Big Oliver West
30 Mile
Zymacord
Exchamsiks Loop
Little Oliver
Flint Creek
Telkwa
Houston
Priestly Hill
Sheraton Mill
Suskwa
Shames East
Cut Block
Kleanza
Chimdemash West
Chimdemash East Transect 1
Chimdemash East Transect 2
St Croix
Legate
Rainbow Pass
Research Farm Smithers
Yellow Cedar Lodge
Exchamsiks Provincial Park
Big Oliver East

MAT MWMT MCMT TD
3.8
14.7
-9
23.7
5.9
15.9
-4.6
20.5
6.6
16.5
-3.8
20.2
6.2
16.3
-4.2
20.5
6.4
16.2
-3.8
20
6.5
16.2
-3.5
19.6
5.4
16.1
-6.3
22.3
6.3
16.1
-3.7
19.9
6.4
16.3
-4.1
20.4
6.7
16.3
-3.3
19.5
5.2
15.9
-6.4
22.2
5.4
16.3
-6.5
22.8
3.3
14.3
-9.4
23.7
3
14.3
-10.1 24.4
2.5
14
-10.6 24.5
2.6
13.8
-10.3 24.1
3.8
14.7
-8.9
23.5
6.6
16.5
-3.8
20.3
6.4
16.4
-4.1
20.5
5.6
15.6
-5.1
20.7
5.8
16.1
-5.2
21.4
5.8
16.2
-5.2
21.4
5.8
16.1
-5.3
21.4
5.4
15.9
-5.9
21.8
5.3
15.8
-6.1
21.9
6
14
-2.3
16.2
3.4
14.3
-9.4
23.6
6.2
16.2
-4.2
20.4
6.7
16.2
-3.3
19.6
5.4
16.1
-6.2
22.4

MAP
582
1233
1368
1347
1719
2178
931
2126
1227
2182
948
854
473
393
508
511
581
1366
1259
1122
1074
1071
1071
1065
1024
3563
483
1248
2173
927

MSP
255
301
334
325
395
519
224
492
295
523
228
214
202
173
220
214
259
333
303
278
258
256
256
233
222
870
206
301
522
221

AHM SHM
23.7 57.7
12.9 52.9
12.1 49.3
12
50.1
9.5
40.9
7.6
31.1
16.5 71.8
7.7
32.8
13.3 55.5
7.7
31.1
16.1 69.6
18.1 76.2
28.2 70.4
33.2 82.3
24.7 63.5
24.6 64.8
23.7 56.7
12.1 49.5
13
54.2
13.9 56.2
14.7 62.6
14.8
63
14.7 63.1
14.5 68.3
14.9 71.2
4.5
16.1
27.7 69.3
13
53.8
7.7
31.1
16.6
73

DD_0
892
502
417
462
428
402
642
428
447
381
652
656
955
1023
1104
1077
888
420
451
548
551
551
555
610
629
334
949
459
385
635

Table. A1-3. Summary of climate data per site (n=30) from ClimateNA (Wang 2016).
DD5
1184
1419
1537
1473
1479
1487
1415
1461
1500
1525
1384
1451
1110
1094
1027
1013
1176
1532
1506
1368
1448
1452
1447
1398
1380
1206
1115
1472
1518
1428

86

DD_18 DD18
5183
25
4436
49
4191
62
4327
55
4261
54
4215
54
4630
46
4281
51
4272
59
4143
58
4674
43
4611
51
5339
20
5446
18
5635
16
5615
15
5186
25
4200
62
4271
59
4551
44
4479
52
4474
53
4484
52
4606
45
4647
43
4363
30
5325
20
4323
55
4158
57
4608
49

NFFD
170
209
221
208
215
219
191
211
217
226
189
193
160
151
147
148
168
220
210
200
205
205
204
196
193
217
162
214
224
193

bFFP
148
136
132
139
135
135
146
138
133
132
146
143
154
161
166
165
151
133
138
141
138
138
139
143
144
138
152
134
133
145

eFFP
262
282
285
282
284
286
273
283
284
288
272
273
255
251
254
254
261
285
282
278
280
280
279
276
274
285
256
283
287
274

FFP
113
145
153
143
148
151
127
146
150
156
126
130
101
91
88
89
110
152
144
137
142
141
141
133
130
146
104
148
154
129

PAS
201
359
318
347
430
492
298
522
317
457
309
265
177
152
216
222
197
320
313
353
335
334
336
359
344
639
180
334
463
294

EMT
-38.3
-30.5
-28.9
-30.2
-29.3
-28.5
-33.6
-29.4
-29.5
-27.9
-33.8
-34
-39.7
-40.9
-41.8
-41.1
-38.2
-29
-30
-31.5
-31.5
-31.5
-31.6
-32.8
-33.2
-27.4
-39.6
-29.8
-28.1
-33.4

EXT
33.1
34.1
34.7
34.7
34.4
34.5
34.1
34.6
34.5
34.5
34
34.4
32.8
33.1
32.6
32.5
32.9
34.7
35
33.8
33.9
33.9
33.9
33.8
33.7
31.6
32.8
34.3
34.5
34.2

MAR
10.2
9.1
9.3
9.9
9.9
9.4
9.6
10.1
11
9.9
9.4
9.8
10.4
10.4
10.9
10.3
9.6
9.3
10.4
9.2
9
9
9
8.9
9
9.6
10.4
11.2
9.9
9.8

Eref
526
555
561
574
559
554
586
561
559
550
581
588
532
548
526
522
529
562
581
553
563
564
565
570
570
500
528
555
553
586

CMD
230
197
184
198
145
74
303
90
200
71
295
314
287
337
262
265
234
185
215
218
244
246
247
268
281
0
279
191
73
304

RH
64
70
71
68
70
71
66
69
71
72
66
66
61
59
60
60
63
71
68
69
69
69
69
68
67
72
62
71
71
66

CMI DD1040
14.62
454
79.97
608
91.71
679
88.71
643
127.5
641
173.98
641
45.15
610
168.29
629
78.1
659
174.3
663
47.68
589
36.59
633
3.65
405
-6.34
397
8.69
363
9.71
352
14.54
449
91.42
677
78.86
664
68.47
575
62.03
630
61.53
633
61.52
630
60.72
597
56.8
585
319.19
420
5.03
407
80.59
641
173.29
658
44.62
619

Site
Name
Wt
LV
SW
DC
Ex
SE EP
BOW
30M
Zm
EL
LO
FC
Tk
HE
PH
SM
Sw
SE
CB
Kz
CW
CET1
CET2
SC
Lg
RP
RFS
YCL
EPP
BOE

CWD
(%)
0
7
0
0
0
3
1
4
4
0
0
0
0
2
0
0
0
0
0
8
0
0
4
0
0
0
0
1
0
0

MWD
(%)
0.4
14.6
12.6
5.6
17.6
17.4
18.4
37.6
19
27
21
10.2
1
4.2
18.2
6.2
4.4
5
0.2
2.8
0
3.8
9.2
0.8
0
0.8
0
0
5.6
1.6

FWD
(%)
0.8
19.6
11.8
13.2
17
15
13.4
17.2
16
25.6
21
10
3
1.6
14.4
2.2
1.8
23
3
1.4
3.4
2.6
8
0.8
3
2.6
0
11.4
4.8
1.6

Litter
(%)
14.6
48
86
27.6
90.4
40
47
12.8
39
90
70.6
51
62
52
48
83
46
33.8
6.4
52.6
15.2
39.6
45.2
35.6
31.8
29.8
28
65.2
87.6
55

Rock
(%)
0
0.6
0
1.6
0
0.6
0
0
0
0
0
0
0.4
0
2.6
0
1
0
0.2
0
0
0
0.4
0
0.2
12.4
3.2
0.6
0
0

Bare Ground or Soil
(%)
4
0.2
0
3.6
4.6
0
1
1
0
0
3
7.8
2.8
7.8
3.2
0.4
0.4
10
0
0
5
0
0
0.1
0
0
6.6
0
0
0
87

Table. A1-4. Summary of road plot environment data averaged for five quadrats per site (n=30).

Asphalt
(%)
0.2
0
0
19
0.2
15.8
0
13.4
0.4
0
0
0
0
0
0.4
0
0
0
0
0
0
0
0
3.1
0
0
0
0
0
0

Gravel
(%)
80
0.2
0
1
0
0
0
0
0
0
0
0
9.6
0
40
0
2
0
0
0
0.8
0
0
0
0
12.6
61
1
0
0

Rooting depth
(cm)
2.8
7.2
3.4
0
0
0
5.8
0
0
0
6.6
4.6
7.4
21.2
3.6
9
7
1.3
1.04
5.3
4.4
2.8
4.8
0.1
4.6
4.6
4
9.4
0
2.2

Litter Depth
(cm)
0.2
2.4
2.8
3.6
2.8
1.3
2.8
4.6
1.8
3.4
3
2.8
1.4
1.4
1
4
1.8
6.9
0.4
2.7
1.2
3.2
2.8
1.6
4.2
4.4
0.2
2.4
3.4
2.2

Compaction
(PSI)
215
205
55
0
0
0
350
0
0
0
245
300
265
270
285
295
270
263
20
165
275
225
250
0
270
300
250
245
0
215

Light
(%)
91.5
9.5
6.4
18.2
9.6
33.1
20.9
39.2
12.1
4.0
12.4
13.1
19.7
37.6
30.4
16.1
8.6
9.1
56.6
24.9
20.2
18.9
14.6
19.7
42.3
30.5
91.3
18.7
23.6
76.0

Site
Name
Wt
LV
SW
DC
Ex
SE EP
BOW
30M
Zm
EL
LO
FC
Tk
HE
PH
SM
Sw
SE
CB
Kz
CW
CET1
CET2
SC
Lg
RP
RFS
YCL
EPP
BOE

CWD
(%)
0
2
1
0
3
7
0
11
0
10
0
0
0
0
1
1
4
0
0
6
2
0
1
2
0
23
0
1
1
0

MWD
(%)
0
4.4
6.2
4.4
10.4
4.6
17.0
5.2
21.0
11.0
7.0
4.0
5.4
4.2
22.4
9.0
8.0
6.0
0.6
10.8
1.4
14.6
8.4
10.0
2.4
21.2
0
5.8
0.4
0

FWD
(%)
0
23.6
3.6
12.4
11.4
19.0
13.6
11.0
21.0
16.0
12.4
7.4
14.0
2.0
21.0
5.6
6.2
6.5
1.9
1.3
30.0
4.6
8.8
2.8
9.2
3.8
0
5.0
12.0
3.4

Litter
(%)
11.8
53.0
90.0
75.6
74.6
69.0
82.0
32.4
96.0
80.6
45.4
40.0
64.6
24.0
83.6
57.6
38.6
61.3
36.6
71.0
59.8
50.0
63.8
35.0
67.0
53.0
75.0
88.4
85.8
62.2

Rock
(%)
0
0.2
0
0.4
2.4
14
0
0.4
0
0
6.6
0
0
4
6
0
0
0
0
0
0
0.2
2.8
0
0.4
0
2.2
0
0
0
88

Bare Ground or Soil
(%)
3.4
0.2
0
0
1.4
14
0
4
2.6
0
0
0
4
10.2
0.4
0
0
2.5
0
0
1
0
0
0
0
0
1.2
0
0
0

Asphalt
(%)
1.4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

Table. A1-5. Summary of shoulder plot environment data averaged from five quadrats per site (n=30).

Gravel
(%)
27.8
0
0
0
0.4
0
0
7
0.6
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
21.6
0
0
0

Rooting
Depth (cm)
7.9
14
13.7
21
16.6
25.2
8.8
14.4
22.6
17.6
10.2
9.6
10.9
31.6
17
16.2
17.8
9
8.2
8.1
9
7.6
8.6
17.2
10.8
11.4
5.2
0
10.2
2.6

Litter Depth
(cm)
1.1
1.8
6.1
5.8
8.4
4
4.6
9.6
5
5
4.4
3.8
4.7
2.4
4.6
8.2
3.4
2.3
2.5
3.7
3.8
4
3.2
3.2
5
4
1.4
3.4
3.4
3.6

Compaction
(PSI)
195
235
240
160
200
225
295
245
200
185
265
270
260
255
275
210
200
206
160
203
275
245
190
205
255
240
270
0
260
275

Light
(%)
80.4
11.1
3.9
11.6
4.7
16.9
13.3
18.4
4.6
4.6
16.5
8.0
24.6
32.3
10.7
8.7
9.8
18.1
38.9
22.8
17.1
17.8
11.8
11.7
20.9
19.4
82.6
23.6
13.6
65.2

Table. A1-6. Summary of bulked road plot soil characteristics at each site (n=30).

Site Name
Witset
La Verge
Shames West
Delta Creek
Exstew
SE of Exchamsiks Park
Big Oliver West
30Mile
Zymacord
Exchamsiks Loop
Little Oliver
Flint Creek
Telkwa
Houston East
Priestly Hill
Sheraton Mill
Suskwa
Shames East
Cut Block
Kleanza
Chimdemash West
Chimdemash East Transect 1
Chimdemash East Transect 2
St Croix
Legate
Rainbow Pass
Research Farm Smithers
Yellow Cedar Lodge
Exchamsiks Provincial Park
Big Oliver East

pH
7.2
4.7
5.5
4.8
4.6
5.0
5.2
4.6
4.8
3.8
5.3
6.7
6.7
6.7
6.5
6.8
5.9
5.7
5.3
5.4
4.8
5.6
5.2
5.0
5.9
4.7
6.8
5.4
5.3
5.0

EC
(mS/m)
35.4
156.8
113.2
146.3
156.2
137.3
131.4
157.3
153.8
199.9
134.2
46.4
47.5
43.4
57.7
40.5
99.5
119.3
128.9
116.5
153.3
118.7
235.5
137.4
195.6
175.8
51.0
117.7
129.0
141.0

Total C
(%)
0.1
3.1
19.0
11.9
43.9
35.9
3.7
26.5
29.6
44.9
1.4
4.3
2.2
4.5
1.7
3.2
2.1
0.1
0.2
0.4
0.4
0.5
0.4
3.0
0.4
0.4
0.1
2.8
0.1
0.2

Total N
(%)
1.9
0.2
1.1
0.8
2.5
2.2
0.2
1.8
2.0
2.7
0.1
0.3
0.1
0.3
0.1
0.1
0.1
1.5
2.4
5.8
4.8
6.1
6.7
45.1
5.8
5.4
2.2
42.7
2.0
4.6

C:N ratio
0.04
15.5
17.3
14.9
17.6
16.3
18.5
14.7
14.8
16.6
14
14.3
22
15
17
32
21
0.07
0.06
0.07
0.08
0.08
0.06
0.07
0.06
0.07
0.05
0.07
0.06
0.05

EC = electrical conductivity in milliSiemens per meter, Total C = percent total carbon, Total N = percent total
nitrogen, C:N ratio (carbon to nitrogen ratio).

89

Table. A1-7. Summary of bulked shoulder soil plot characteristics at each site (n=30).

Site Name
Witset
La Verge
Shames West
Delta Creek
Exstew
SE of Exchamsiks Park
Big Oliver West
30Mile
Zymacord
Exchamsiks Loop
Little Oliver
Flint Creek
Telkwa
Houston East
Priestly Hill
Sheraton Mill
Suskwa
Shames East
Cut Block
Kleanza
Chimdemash West
Chimdemash East Transect 1
Chimdemash East Transect 2
St Croix
Legate
Rainbow Pass
Research Farm Smithers
Yellow Cedar Lodge
Exchamsiks Provincial Park
Big Oliver East

pH
6.2
4.8
5.1
5.2
5.2
4.9
5.2
5.1
4.3
4.7
5.4
6.6
6.6
6.7
6.2
7.0
6.2
6.0
6.0
5.6
5.0
5.1
3.2
4.2
3.9
5.2
6.6
6.1
5.2
5.5

EC
(mS/m)
75.2
151.1
136.0
128.7
122.3
142.5
131.6
141.0
177.4
154.4
122.7
45.4
54.4
39.7
82.2
13.6
88.7
101.6
89.9
114.0
149.9
149.4
239.4
182.0
95.2
155.3
48.3
84.3
126.0
108.5

Total C
(%)
0.2
4.4
6.2
3.6
6.2
8.1
4.9
4.4
5.4
4.3
2.4
4.6
3.3
3.5
2.7
2.2
3.1
0.1
0.1
0.4
0.4
0.4
0.6
2.1
0.4
1.2
0.3
0.4
0.2
0.3

Total N
(%)
4.4
0.3
0.4
0.2
0.3
0.5
0.2
0.3
0.4
0.3
0.1
0.3
0.1
0.3
0.1
0.1
0.2
1.3
2.3
5.0
5.3
5.7
11.7
30.2
8.3
19.5
3.7
7.1
3.1
6.0

C:N ratio
0.05
14.7
15.5
18
20.7
16.2
24.5
14.7
13.5
14.3
24
15.3
33
11.7
27
22
15.5
0.08
0.05
0.07
0.07
0.07
0.05
0.07
0.05
0.06
0.07
0.06
0.06
0.05

EC = electrical conductivity in milliSiemens per meter, Total C = percent total carbon, Total N = percent total
nitrogen, C:N ratio (carbon to nitrogen ratio).

90

Table A1-8. Total plant species richness and mean Shannon-Wiener diversity index of the road,
shoulder, ditch, and forest vegetation at each abandoned road segment of Highway 16.
Site Name
Witset
Witset
Witset
Witset
La Verge
La Verge
La Verge
La Verge
Shames West
Shames West
Shames West
Shames West
Delta Creek
Delta Creek
Delta Creek
Delta Creek
Exstew
Exstew
Exstew
Exstew
SE of Exchamsiks Park
SE of Exchamsiks Park
SE of Exchamsiks Park
SE of Exchamsiks Park
Big Oliver West
Big Oliver West
Big Oliver West
Big Oliver West
30Mile
30Mile
30Mile
30Mile
Zymacord
Zymacord
Zymacord
Zymacord
Exchamsiks Loop
Exchamsiks Loop
Exchamsiks Loop
Exchamsiks Loop
Little Oliver
Little Oliver
Little Oliver
Little Oliver
Flint Creek
Flint Creek
Flint Creek
Flint Creek
Telkwa
Telkwa
Telkwa

Plot Type
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
No Forest
Road
Shoulder
Ditch
No Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch

91

Richness
12
10
34
49
15
13
8
11
18
11
19
17
26
17
31
24
11
21
34

Shannon-Wiener index
0.9315
0.3929
1.1937
1.3554
0.8004
0.8907
0.6850
0.8903
0.9133
0.7511
0.9905
0.9380
1.0823
0.9871
1.1346
1.1750
0.5417
0.9054
1.2990

39
24
35

1.2900
1.0765
1.2539

17
18
16
29
25
35
41
13
25
11
13
17
14
14
25
15
21
14
31
25
37
29
38
23
24
19
18

0.6934
0.8158
0.7645
1.1535
1.1757
1.1942
1.3912
0.9353
0.9979
0.7487
0.7698
0.9666
0.7642
0.8589
1.0200
0.8703
0.9606
0.8001
1.2825
1.0108
1.2282
1.1484
1.3173
1.0787
0.8686
0.7235
1.0241

Telkwa
Houston East
Houston East
Houston East
Houston East
Priestly Hill
Priestly Hill
Priestly Hill
Priestly Hill
Sheraton Mill
Sheraton Mill
Sheraton Mill
Sheraton Mill
Suskwa
Suskwa
Suskwa
Suskwa
Shames East
Shames East
Shames East
Shames East
Cut Block
Cut Block
Cut Block
Cut Block
Kleanza
Kleanza
Kleanza
Kleanza
Chimdemash West
Chimdemash West
Chimdemash West
Chimdemash West
Chimdemash East Transect 1
Chimdemash East Transect 1
Chimdemash East Transect 1
Chimdemash East Transect 1
Chimdemash East Transect 2
Chimdemash East Transect 2
Chimdemash East Transect 2
Chimdemash East Transect 2
St Croix
St Croix
St Croix
St Croix
Legate
Legate
Legate
Legate
Rainbow Pass
Rainbow Pass
Rainbow Pass
Rainbow Pass
Smithers Research Farm
Smithers Research Farm

Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder

92

9
24
24
22
29
26
21
30
30
28
29
30
43
43
29
34
34
17
8
24
22
7
19
25
12
18
12
17
28
33
26
36
34
24
19
22
22
15
12
17
17
21
21
25
15
36
25
37
14
35
17
34
30
13
12

0.6185
1.0612
1.0473
1.2366
1.2552
1.0250
0.9538
1.2278
1.2894
1.0531
1.0158
1.4122
1.5022
1.2742
1.0416
1.3986
1.3885
0.9583
0.7584
0.9864
1.0227
0.1529
0.8845
1.1006
0.7937
1.0637
0.7485
0.8612
0.5759
0.9126
0.9347
1.0720
1.2545
0.8597
0.8701
0.8478
0.7892
0.7261
0.7282
0.7100
0.7207
0.9976
0.7077
1.0319
0.6512
1.1608
1.0070
1.3133
0.9127
1.1748
1.0504
1.2228
1.1710
0.8289
0.7517

Smithers Research Farm
Smithers Research Farm
Yellow Cedar Lodge
Yellow Cedar Lodge
Yellow Cedar Lodge
Yellow Cedar Lodge
Exchamsiks Provincial Park
Exchamsiks Provincial Park
Exchamsiks Provincial Park
Exchamsiks Provincial Park
Big Oliver East
Big Oliver East
Big Oliver East
Big Oliver East

Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest
Road
Shoulder
Ditch
Forest

93

23
21
12
8
21
15
20
19
27
19
22
21
41
18

1.1707
1.0655
0.6942
0.6907
1.0451
0.9976
0.9339
0.8504
1.0846
1.0023
0.8465
0.9556
1.2123
0.9872

Appendix 2. Road and shoulder vegetation comparisons.
Paired t-tests and Wilcoxon tests were run to determine which plant growth groups were
more frequent on gravel or asphalt road and shoulder surfaces. Table 6-8 summarizes road
and shoulder vegetation across sites. Table 6-9 summarizes road and shoulder vegetation
across gravel sites only.
Table A2-1. Mean (standard deviation) and results of paired t- tests or Wilcoxon* tests comparing
plant cover organized by growth form on road surfaces (which may consist of asphalt or gravel
substrates) and adjacent shoulders within abandoned road segments (n=30). Statistically significant
differences are shown by bolded p values (<0.05).

Variables
Tree cover
Shrub cover
Woody cover
Forb cover
Fern and Fern Allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-Vascular cover
Vascular cover
Exotic cover
Total plant cover
Shannon-Weiner index

Road Center
Cover Mean, %
46.61 (36.5)
23.1 (18.7)
46.61 (36.5)
22.5 (27.1)
3.8 (7.5)
13.6 (20.3)
39.9 (40.5)
25.7 (27.1)
1.1 (3.5)
26.7 (27.3)
109.6 (61.6)
17.4 (28.0)
136.4 (58.3)
0.9 (0.2)

Shoulder
Cover Mean, %
62.78 (31.9)
26.8 (20.9)
62.8 (31.9)
11.5 (11.9)
3.3 (5.6)
3.0 (5.5)
17.8 (15.9)
10.9 (14.8)
0.4 (0.8)
11.2 (14.8)
107.4 (42.3)
4.6 (7.9)
118.6 (40.9)
0.9 (0.2)

p value
0.0078
*0.4427
0.0111
*0.0219
0.6633
*0.0019
*0.0019
*0.0008
0.7223
*0.0003
0.8392
*0.0166
*0.0078
*0.0327

* denotes Wilcoxon tests used where data were non-normal.

Table A2-2. Mean (standard deviation) and results of paired t- tests or Wilcoxon* tests comparing
plant cover organized by growth form on gravel road and shoulder surfaces within abandoned road
segments (n=19). Significant statistical relationships with bolded p values (<0.05).

Variables
Tree cover
Shrub cover
Woody cover
Forb cover
Fern and Fern Allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-vascular cover
Vascular cover
Exotic cover
Total plant cover
Shannon-Wiener index

Road Center
Gravel
Cover Mean, %
45.9 (33.8)
23.9 (18.2)
69.8 (37.5)
27.0 (32.1)
5.4 (9.1)
14.1 (23.2)
46.4 (47.8)
23.0 (25.1)
1.7 (4.3)
24.7 (25.7)
116.2 (70.3)
20.3 (33.3)
140.9 (69.6)
1.0 (0.2)

* denotes Wilcoxon tests used where data were non-normal.

94

Shoulder Gravel
Cover Mean, %
57.0 (33.5)
19.5 (15.3)
76.5 (41.4)
10.7 (9.0)
3.5 (6.6)
2.3 (5.3)
16.5 (13.3)
12.7 (14.5)
0.6 (1.0)
13.3 (14.4)
93.0 (42.1)
3.7 (7.5)
106.4 (40.5)
0.9 (0.2)

p value
0.0983
*0.2145
0.4126
*0.0329
*0.0831
*0.0144
*0.0053
*0.0365
*0.5541
*0.0166
0.1105
*0.0280
0.0233
0.0225

Appendix 3. Vegetation correlations with hypothesized environmental drivers.
Pearson’s and Spearman’s correlation coefficients of plant group response variables for TSA
(time since road abandonment, Light (percent canopy openness), and forest tree percent
cover are in Tables A3-1 to 3-6. Road plots are first, then shoulder plots follow. Variables
with significant p values under 0.05 were used as predictor variables in multiple regression
models.
Table A3-1. Pearson’s and Spearman’s correlation coefficients of road surface vegetation by plant
group and time since road abandonment in years. Significant statistical relationships with bolded p
values (<0.05).

Vegetation variables
Tree cover
Shrub cover
Woody cover
Forb cover
Fern and fern allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-vascular cover
Vascular cover
Exotic cover
Total plant cover
Species richness
Shannon-Wiener diversity index

Environmental
variable
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA

Pearson's
0.232
0.346
0.370
0.365
0.245
0.192
0.386
0.021
0.196
0.046
0.497
0.224
0.546
0.261
0.041

p value
0.216900
0.061030
0.044260
0.047290
0.191100
0.308900
0.035000
0.913800
0.298900
0.811200
0.005233
0.234500
0.001790
0.163100
0.163100

Spearman's
0.396
0.356
0.360
0.406
0.285
0.055
0.350
0.103
0.127
0.136
0.513
0.119
0.615
0.327
0.065

p value
0.030140
0.053520
0.050460
0.026040
0.126700
0.772100
0.057680
0.589600
0.502700
0.472500
0.003725
0.530500
0.000299
0.077590
0.732800

Table A3-2. Pearson’s and Spearman’s correlation coefficients of road surface vegetation by plant
group and light (percent canopy openness). Significant statistical relationships with bolded p values
(<0.05).
Vegetation variables
Tree cover
Shrub cover
Woody cover
Forb cover
Fern and Fern Allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-vascular cover
Vascular cover
Exotic cover
Total plant cover
Species Richness
Shannon-Wiener diversity index

Environmental
variable
Light
Light
Light
Light
Light
Light
Light
Light
Light
Light
Light
Light
Light
Light
Light

95

Pearson's
-0.592
-0.215
-0.635
-0.181
-0.200
-0.056
-0.186
-0.048
0.241
-0.017
-0.539
0.005
-0.577
-0.199
-0.121

p value Spearman's p value
0.000564
-0.643
0.000127
0.253900
-0.162
0.391100
0.000166
-0.592
0.000563
0.338900
0.031
0.338900
0.288900
-0.320
0.085140
0.770000
0.151
0.425100
0.324600
-0.043
0.821700
0.801400
0.050
0.792700
0.200100
0.119
0.532400
0.930600
0.085
0.656800
0.002109
-0.461
0.010390
0.980600
0.240
0.201400
0.000839
-0.382
0.037470
0.291400
0.009
0.963200
0.523400
0.070
0.713700

Table A3-3. Pearson’s and Spearman’s correlation coefficients of road surface vegetation by plant
group and nearby overstory tree cover. There were no significant p-values <0.05 for the association of
any vegetation attributes with forest tree cover adjacent to the abandoned road segments.
Vegetation variables
Tree cover
Shrub cover
Woody cover
Forb cover
Fern and Fern Allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-vascular cover
Vascular cover
Exotic cover
Total plant cover
Species Richness
Shannon-Wiener diversity index

Environmental
variable
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover

Pearson's
0.264
-0.104
0.176
-0.011
-0.221
0.089
-0.003
0.130
-0.283
0.092
0.109
0.098
0.158
0.214
0.211

p value
0.174600
0.597300
0.369600
0.956500
0.258400
0.651100
0.986800
0.511000
0.144700
0.643100
0.579600
0.619200
0.423400
0.275200
0.280500

Spearman's
0.266
0.253
0.334
0.264
0.210
0.284
0.267
-0.064
0.099
-0.005
0.408
0.327
0.470
0.334
0.103

p value
0.470400
0.879100
0.365900
0.569500
0.982600
0.203100
0.484600
0.373400
0.148300
0.560400
0.297100
0.219000
0.241900
0.222300
0.307800

Table A3-4. Pearson’s and Spearman’s correlation coefficients of shoulder surface vegetation by
plant group and time since road abandonment in years. Significant statistical relationships with bolded
p values (<0.05).
Vegetation variables
Tree cover
Shrub cover
Woody cover
Forb cover
Fern and Fern Allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-vascular cover
Vascular cover
Exotic cover
Total plant cover
Species Richness
Shannon-Wiener diversity index

Environmental
variable
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA
TSA

96

Pearson's
0.075
0.042
0.078
-0.092
0.243
-0.163
-0.042
-0.131
0.088
-0.126
0.062
-0.268
0.019
0.098
0.175

p value
0.6935
0.8249
0.6813
0.6291
0.1957
0.3886
0.8239
0.4896
0.6426
0.5063
0.7448
0.1527
0.9225
0.6055
0.3558

Spearman's
0.104
0.049
0.084
0.153
0.205
-0.037
0.152
0.090
-0.084
0.111
0.086
-0.082
0.150
0.332
0.113

p value
0.5855
0.7973
0.6581
0.4193
0.2768
0.8475
0.4228
0.6360
0.6587
0.5579
0.6521
0.6683
0.4284
0.0731
0.5535

Table A3-5. Pearson’s and Spearman’s correlation coefficients of shoulder surface vegetation by
plant group and light (percent canopy openness). Significant statistical relationships with bolded p
values (<0.05).
Vegetation variables
Tree cover
Shrub cover
Woody cover
Forb cover
Fern and Fern Allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-vascular cover
Vascular cover
Exotic cover
Total plant cover
Species Richness
Shannon-Wiener diversity index

Environmental
variable
Pearson's
Light
-0.555
Light
-0.265
Light
-0.558
Light
-0.063
Light
-0.226
Light
0.315
Light
-0.016
Light
0.305
Light
0.009
Light
0.305
Light
-0.555
Light
0.280
Light
-0.464
Light
-0.199
Light
-0.356

p value
0.001452
0.156200
0.001347
0.741700
0.229200
0.090100
0.931100
0.100700
0.963100
0.100700
0.001440
0.133300
0.009772
0.292600
0.053620

Spearman's
-0.339
-0.136
-0.295
-0.093
-0.302
0.199
-0.026
0.124
0.228
0.133
-0.261
0.143
-0.273
-0.103
-0.130

p value
0.066730
0.474500
0.114100
0.626200
0.105300
0.291800
0.891800
0.513500
0.225400
0.482700
0.163200
0.449500
0.144700
0.587500
0.492300

Table A3-6. Pearson’s and Spearman’s correlation coefficients of shoulder surface vegetation by
plant group and nearby overstory tree cover. There were no significant p-values <0.05 for the
association of any vegetation attributes with forest tree cover adjacent to the abandoned road
segments.
Vegetation variables
Tree cover
Shrub cover
Woody cover
Forb cover
Fern and fern allies cover
Graminoid cover
Herb cover
Bryophyte cover
Lichen cover
Non-vascular cover
Vascular cover
Exotic cover
Total plant cover
Species richness
Shannon-Wiener diversity index

Environmental
variable
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover
Forest tree cover

97

Pearson's
-0.073
0.016
-0.049
0.199
-0.107
0.191
0.179
0.008
0.215
0.019
0.018
0.286
0.025
0.348
0.349

p value Spearman's
0.311100
-0.112
0.442800
0.084
0.686400
0.011
0.896500
0.254
0.538200
0.056
0.999000
0.269
0.749900
0.312
0.834300
-0.056
0.365800
0.087
0.872900
-0.011
0.774100
0.027
0.772400
0.192
0.725200
0.109
0.841400
0.391
0.841000
0.298

p value
0.5823
0.4426
0.8389
0.7131
0.9364
0.7218
0.6772
0.8124
0.2121
0.8060
0.8280
0.6923
0.6465
0.9823
0.9284

Appendix 4. Multi-factorial model selection.
To analyze plant group responses on abandoned roads, multiple regression (multi-factorial)
models of select response variables (with highly correlated and significant p values of
predictor variables) to obtain a list of “better fit” models of the data. The models with the
lowest Akaike information criterion (AICc) score, a delta of zero and the highest weight are
the “best-approximating model,” but any model with a delta <2 also qualifies as a “better fit”
model for the data (Symonds and Moussalli 2011).
Model selection tables for road surface and plant groups are in Tables A4-1 to A4-7. Only
one model exists from the total plant cover and bryophyte cover data, therefore there weren’t
other models with which to compare AICc scores for these two plant groups. There are no
model selection tables for shrub cover and tree cover on the road because neither were
correlated with the road surface. During sampling, there were very few shrubs or trees on the
compacted road surface.
A general description of the model selection terms in Tables A4-1 to A4-15 are as follows: df
is the degrees of freedom (number of parameters in the model, minus 1); logLik is the
maximum likelihood scale used, AICc represents how well the model fits the data thereby
ranking all the models against each other, delta assess the relative strength of each model,
and Akaike weights are all the models summed to one with the best fit model being the
heaviest weight (Symonds and Moussalli 2011). Cardoso et al. (2007) provide complete
descriptions of table terms.
Table A4-1. Model selection table of total vascular plant cover on the road surface.
Vascular plant cover model selection table from dredge (delta<2)
df logLik AICc delta weight
Eref + TSA
4 -160 329
0 0.594
TSA
3 -161 329.7
0.76 0.406
Eref (Hargreaves reference evaporation in millimeters) and TSA (time since road abandonment in years).

Table A4-2. Model selection table of total plant cover on the road surface.
Total plant cover component models from dredge <2 delta
df
logLik AICc delta weight
TSA
3 -159 324.3
0
1
TSA (time since road abandonment in years).

Table A4-3. Model selection table of total exotic plant cover on the road surface.
Exotic plant cover component models from dredge <2 delta
df
logLik AICc delta weight
RdSub+Light+RH
5 -129 270.5
0 0.351
Light+RH
4 -131 270.8
0.29 0.304
RH
3 -132 271.8
1.37 0.177
RdSub+RH
4 -131 272
1.48 0.167
RdSub (road substrate), Light (% canopy cover), and RH (% mean annual relative humidity).

Table A4-4. Model selection table of total bryophyte cover on the road surface.
Bryophyte cover component models from dredge <2 delta
df
logLik AICc delta weight
Eref
3 -136 279.5
0
1
Eref (Hargreaves reference evaporation in millimeters).

Table A4-5. Model selection table of total forb cover on the road surface.
Forb cover component models from dredge <2 delta
df
logLik AICc delta weight
DD1040+TSA
4 -135 279.7
0 0.306
RH
3 -137 280.1
0.4 0.251
DD1040
3 -137 280.3
0.6 0.226
RH+TSA
4 -135 280.4
0.69 0.217
RH (% mean annual relative humidity), TSA (time since road abandonment in years), RdSub (road substrate),
DD1040 (degree-days above 10°C and below 40°C).

98

Table A4-6. Model selection table of total graminoid cover on the road surface.
Graminoid cover component models from dredge <2 delta
df
logLik AICc delta weight
AHM
3 -130 266.3
0 0.654
AHM+RdSub
4 -129 267.6
1.27 0.346
AHM (annual heat-moisture index) and RdSub (road substrate).

Table A4-7. Model selection table of total lichen cover on the road surface.
Lichen cover component models from dredge <2 delta
df
logLik AICc delta weight
bFFP
3
-77 160.9
0
1
bFFP (day of the year the frost-free period begins).

Model selection tables for the shoulder surface and plant groups are in Tables A4-8 to
A4-15. Only one model exists from the shoulder forb cover, therefore there weren’t other
models with which to compare AICc scores, meaning there is no shoulder forb table below.
All shoulder’s road substrate was gravel therefore this variable reflects the growing medium
for plant growth overall, not in comparison to another substrate.
Table A4-8. Model selection table of total vascular plant cover on the shoulder surface.
Vascular plant cover model selection table from dredge (delta<2)
df logLik AICc delta weight
RdSub+PAS
4 -149 307.8
0 0.281
RdSub+PAS+TSA
5 -148 308.1
0.3 0.242
PAS
3 -151 308.7
0.86 0.183
RdSub
3 -151 308.9
1.13
0.16
PAS+TSA
4 -150 309.3
1.48 0.134
RdSub (road substrate), PAS (precipitation as snow, mm/year), and TSA (time since road abandonment in
years).

Table A4-9. Model selection table of total plant cover on the shoulder surface.
Total plant cover component models from dredge <2 delta
df
logLik AICc delta weight
PAS
3 -150 306.8
0 0.319
RdSub+PAS
4 -149 307.2
0.4 0.261
PAS+TSA
4 -149 308.2
1.41 0.158
RdSub+PAS+TSA
5 -148 308.5
1.77 0.132
RdSub
3 -151 308.6
1.79
0.13
PAS (precipitation as snow, mm/year), RdSub (road substrate), and TSA (time since road abandonment in
years).

Table A4-10. Model selection table of total exotic plant cover on the shoulder surface.
Exotic plant cover component models from dredge <2 delta
df
logLik AICc delta weight
RdSub+RH+TSA
5 -98.3 209.2
0 0.517
RH+TSA
4 -100 210.3
1.15 0.291
RdSub+RH
4 -101 211.2
1.99 0.192
RdSub (road substrate), RH (% mean annual relative humidity), and TSA (time since road abandonment in
years).

Table A4-11. Model selection table of total bryophyte cover on the shoulder surface.
Bryophyte cover component models from dredge <2 delta
df
logLik AICc delta weight
Light
3 -121 249.7
0 0.325
Eref+Light
4 -120 250.1
0.36 0.272
Null
2 -123 250.2
0.46 0.259
Eref
3 -122 251.3
1.63 0.144
Light (% canopy cover), Eref (Hargreaves reference evaporation in millimeters), and Null (bryophyte cover
alone).

99

Table A4-12. Model selection table of total graminoid cover on the shoulder surface.
Graminoid cover component models from dredge <2 delta
df
logLik AICc delta weight
AHM+RdSub
4
-90 189.7
0
0.32
AHM+RdSub+TSA
5 -88.8 190
0.32 0.273
AHM
3 -92.2 191.3
1.57 0.145
Null
2 -93.5 191.4
1.68 0.138
AHM+TSA
4
-91 191.6
1.89 0.124
AHM (annual heat-moisture index), RdSub (road substrate), and TSA (time since road abandonment in years).

Table A4-13. Model selection table of total shrub cover on the shoulder surface.
Shrub cover component models from dredge <2 delta
df
logLik AICc delta weight
RdSub
3
-129 265.9
0 0.492
RdSub+PAS
4 -129 266.8
0.9 0.313
RdSub+Light
4 -129 267.7
1.85 0.195
PAS (precipitation as snow, mm/year), RdSub (road substrate), and Light (% canopy cover).

Table A4-14. Model selection table of tree cover on the shoulder surface.
Tree cover component models from dredge <2 delta
df
logLik AICc delta weight
eFFP
3
-143 293.5
0 0.663
eFFP+TSA
4
-143 294.9
1.35 0.337
eFFP (day of the year on which FFP ends) and TSA (time since road abandonment in years).

Table A4-15. Model selection table of total lichen cover on the shoulder surface.
Lichen cover component models from dredge <2 delta
df
logLik AICc delta weight
EXT
3 -32.7 72.3
0 0.539
3 -33.48 73.9
1.57 0.246
RdSub
4 -32.28 74.1
1.84 0.215
EXT+RdSub
EXT (extreme maximum temperature over 30 years), RdSub (road substrate).

100