NINE DECADES OF GLACIER CHANGE IN THE CANADIAN ROCKY MOUNTAINS by Christina Louise Tennant B.Sc., University of Northern British Columbia, 2007 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN NATURAL RESOURCES AND ENVIRONMENTAL STUDIES UNIVERSITY OF NORTHERN BRITISH COLUMBIA APRIL 2012 © Christina Louise Tennant, 2012 1+1 Library and Archives Canada Bibliotheque et Archives Canada Published Heritage Branch Direction du Patrimoine de I'edition 395 Wellington Street Ottawa ON K1A0N4 Canada 395, rue Wellington Ottawa ON K1A 0N4 Canada Your file Votre reference ISBN: 978-0-494-87544-5 Our file Notre reference ISBN: 978-0-494-87544-5 NOTICE: AVIS: The author has granted a non­ exclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distrbute and sell theses worldwide, for commercial or non­ commercial purposes, in microform, paper, electronic and/or any other formats. L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'lnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conformement a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these. While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. Canada Abstract Glaciers adjust their physical properties to climate, but long records of glacier change needed to examine this relation are limited. I used Interprovincial Boundary Commission Survey maps of the Alberta-British Columbia (BC) border (1903-1924), aerial photographs (1948-1993), BC Terrain Resource Information Management data (1982-1987), and satellite imagery (1999-2009) to determined glacier changes in the Canadian Rocky Mountains over the past nine decades. From 1919-2006, total glacierized area in the Canadian Rocky Mountains decreased by 40 ± 7%; glaciers smaller than 1.0 km2 experienced the greatest percentage loss. Slope, mean, median, and minimum elevations are negatively correlated with area loss. From 1919-2009, glaciers of the Columbia Icefield retreated 1149.9 ± 34.1 m, shrank 22.5 ± 5.0 %, thinned 49.4 ± 25.2 m w.e., and lost 14.30 ± 2.02 km3 w.e. Debriscovered ice thinned 30-60% less than bare ice. Glacier changes correlated with temperature and precipitation lagged two to ten years. ii Table of Contents Abstract ii Table of Contents iii List of Tables v List of Figures vii List of Appendices x Acknowledgements xi 1. Introduction 1 2. Glacier Change in the Canadian Rocky Mountains, 1919 to 2006 4 2.1 Abstract 4 2.2 Introduction 4 2.2.1 6 2.3 2.4 2.5 Study Area Methods 9 2.3.1 Interprovincial Boundary Commission Survey Maps 9 2.3.2 Data Collection 12 2.3.3 Error Analysis 13 2.3.4 Glacier Change Analysis 16 2.3.5 Climate Data 16 Results 18 2.4.1 Glacier Properties 18 2.4.2 Area Change 20 2.4.3 Area Change with Properties 23 2.4.4 Climate 26 Discussion 28 2.5.1 Area Change 28 2.5.2 Area Change with Glacier Properties 30 2.5.3 Area Change with Climate 31 2.6 Conclusions 33 3. Glacier Change of the Columbia Icefield, 1919-2009 35 3.1 Abstract 35 3.2 Introduction 36 iii 3.2.1 3.3 42 3.3.1 42 Data Sources and Preparation 3.3.1.1 Interprovincial Boundary Commission Survey Maps 42 3.3.1.2 Aerial Photographs 42 3.3.1.3 Satellite Imagery 44 Data Collection 45 3.3.2.1 Planimetric Data 45 3.3.2.2 Elevation Data 47 3.3.3 Glacier Change Analysis 47 3.3.4 Error Analysis 48 3.3.5 Climate Data 51 Results 53 3.4.1 Glacier Change 53 3.4.2 Glacier Change over Time 58 3.4.3 Glacier Change by Type 58 3.4.3.1 3.4.4 3.5 37 Methods 3.3.2 3.4 Study Area Debris Cover 64 Climate 65 Discussion 69 3.5.1 Glacier Change 69 3.5.2 Glacier Change by Type 72 3.5.2.1 3.5.3 Debris Cover 74 Climate 3.5.3.1 76 Regional and Global Comparison 78 3.6 Conclusions 80 4. Conclusions 81 References 83 Appendices 90 iv List of Tables Table 2.1 Data used to rectify the IBCS maps and to assess glacier change from 1919 to 2006 11 Table 2.2 Error estimates for area and area change 16 Table 2.3 Pearson's correlation coefficients (r values) of glacier properties and climate variables with absolute and relative area change by flowshed (n = 506) 24 Table 2.4 Pearson's correlation coefficients (r values) of climate variables with absolute and relative area changes and rates, by period (n = 3) 27 Table 3.1 Properties of glaciers of the Columbia Icefield by flowshed (values calculated from 2009 data) 40 Table 3.2 Data used for glacier change analysis (AP = aerial photograph) 41 Table 3.3 Error estimates used in the glacier analysis 49 Table 3.4 Glacier changes for flowsheds (FS) of the Columbia Icefield. Mean and total values represent icefield-wide changes 54 Table A.l Rectification information for IBCS maps used in glacier change analysis 90 Table A.2 Aerial photograph data and photogrammetry results 91 Table B.l Correlations with absolute area change in the Canadian Rocky Mountains by period (n = 3), for a lag of zero to 19 years 92 Table B.2 Correlations with rates of absolute area change in the Canadian Rocky Mountains by period (n = 3), for a lag of zero to 19 years 93 Table B.3 Correlations with relative area changes in the Canadian Rocky Mountains, by period (n = 3), for a lag of zero to 19 years 94 Table B.4 Correlations with rates of relative area changes in the Canadian Rocky Mountains, by period (n = 3), for a lag of zero to 19 years 95 Table B.5 Correlations with length change rates in the Columbia Icefield, by time period (n = 10), for a lag of zero to 19 years 96 Table B.6 Correlations with area change rates in the Columbia Icefield, by time period (n = 10), for a lag of zero to 19 years 97 Table B.7 Correlations with elevation change rates in the Columbia Icefield, by time period (n = 10), for a lag of zero to 19 years 98 v Table B.8 Correlations with volume change rates in the Columbia Icefield, by time period (n = 10), for a lag of zero to 19 years 99 Table C.l Raw length measurements (m) 100 Table C.2 Estimated length measurements (m) 101 Table C.3 Raw area measurements (km2) 102 Table C.4 Estimated area measurements (km2) 103 Table C.5 Raw elevation change data (m w.e.) 104 Table C.6 Estimated elevation change data (m w.e.) 106 Table C.l Raw volume change data (xlO6 m3 w.e.) 107 Table C.8 Estimated volume change data (xlO6 m3 w.e.) 109 vi List of Figures Figure 2.1 Glaciers in the central and southern Canadian Rocky Mountains. Glaciers encompassed by the flowsheds (orange) are the focus of this study and include glaciers mapped by the Interprovincial Boundary Commission Survey between 1903 and 1924 7 Figure 2.2 Walter and Lieth climate diagram for the central and southern Canadian Rocky Mountains, showing monthly mean temperature (red), monthly mean precipitation (blue vertical striping), frost periods (blue boxes), and probable frost periods (cyan boxes). Precipitation greater than 100 mm is plotted at a reduced scale of 10:1. Absolute maximum and minimum temperatures are given on the left (black). Data are monthly temperature and precipitation values from ClimateWNA (Wang et al., 2012), compiled over the period 19192006 from the center and mean elevation of each glacier 8 Figure 2.3 (A) Example of a raw IBCS map that was rectified using 30 GCPs collected from (B) previously rectified Landsat imagery and (C) TRIM hillshading to produce (D) a rectified map from which glacier extents and contours were extracted 10 Figure 2.4 Example of (A) a glacier on the IBCS maps and (B) digitized extents and contours 13 Figure 2.5 Errors and problems associated with the glacier extents: (A) offset 1919 glacier contours; (B) mismapped 1919 extents; (C) cut off glaciers; (D) unedited 2001 extents; (E) mismapped 1985 and 1919 extents; and (F) shadow and cloud cover 14 Figure 2.6 Size class distribution of glaciers in the Canadian Rocky Mountains for the years 1919 and 2006, by percent glacier number and area 18 Figure 2.7 Log-log plot of 2006 glacier area versus 1919 glacier area. Points are separated into groups based on an increase (disintegrating), decrease (disappearing), or no change (same) in the number of ice masses within a flowshed. The solid line is a linear model fitted to the data described by Eq. 2.1. The one-to-one line is dashed 19 Figure 2.8 Total (A) absolute and (B) relative area change by size class of glaciers in the Canadian Rocky Mountains between 1919 and 2006. Error bars are shown representing the mean error for each size class 21 Figure 2.9 (A) Absolute and (B) relative rates of area change for the Canadian Rocky Mountains, over three periods from 1919 to 2006. Boxes represent the first and third quartiles with the horizontal black line as the median. The whiskers represent the data extremes (5th and 95th percentile) and the circles are outliers Vll 21 Figure 2.10 Regional (A) maximum, (B) mean, and (C) minimum temperature anomalies calculated for the average annual, ablation, and accumulation seasons for the periods 19191985, 1985-2001, and 2001-2006. (D) Precipitation anomalies calculated from the precipitation totaled over the hydrologic year (annual), and ablation and accumulation seasons for the same periods. The climatic mean is based on the period 1919-2006. Additional periods, 1919-1946 and 1946-1985, are included to show the change in climate over the winter of 1945/1946 25 Figure 3.1 The Columbia Icefield, Canadian Rocky Mountains. Glaciers encompassed by the orange flowsheds are the focus of this study. The image is a SPOT 5 scene from 30 August 2009 38 Figure 3.2 Walter and Lieth climate diagram for the Columbia Icefield showing monthly mean temperature (red), monthly mean precipitation (blue vertical stripes), frost periods (blue boxes), and probable frost periods (cyan boxes). Precipitation greater than 100 mm is plotted at a reduced scale of 10:1. Absolute maximum and minimum temperatures are given on the left (black). Data are monthly temperature and precipitation values from ClimateWNA (Wang et al., 2012), compiled over the period 1919-2009 from a range of elevations on a 1 km grid over the icefield 39 Figure 3.3 Problems encountered in mapping glacier extents: (A) glacier cut off by the edge of the IBCS map sheet; (B) inconsistent 1919 glacier extents; (C) snow cover from late lying snow in the 1979 aerial photographs; and (D) debris on glaciers 46 Figure 3.4 (A) Systematic bias with slope in the 1966 elevation data. I fit a linear model (dashed line) to the data to remove the bias (B) 50 Figure 3.5 (A) Absolute elevation differences from check patches between an individual year and the 1986 reference data. (B) Relative elevation differences from check patches between two sequentially differenced datasets. Boxes represent the first and third quartiles with the horizontal black line as the median. The whiskers represent the data extremes (5th and 95th percentile) and the circles are outliers 52 Figure 3.6 Area and elevation change of the Columbia Icefield, 1919 to 2009 55 Figure 3.7 Rates of (A) length, (B) area, (C) elevation, and (D) volume change over each period from 1919 to 2009 56 Figure 3.8 (A-B) Absolute and relative length, (C-D) absolute and relative area, (E) elevation, and (F) volume change of glaciers by watershed 57 Figure 3.9 Changes in length, area, and elevation on glaciers of the Columbia watershed, 1970-1979 59 Figure 3.10 (A-B) Absolute and relative length, (C-D) absolute and relative area, (E) elevation, and (F) volume change of glaciers by size class 61 Figure 3.11 (A-B) Absolute and relative length, (C-D) absolute and relative area, (E) elevation, and (F) volume change of glaciers by type 62 Figure 3.12 (A-B) Absolute and relative length, (C-D) absolute and relative area, (E) elevation, and (F) volume change of glaciers by type of debris cover 63 Figure 3.13 Elevation change rates of debris-covered and bare ice for glaciers with (A-B) debris-covered sides and (C-F) debris-covered termini 66 Figure 3.14 Annual, ablation, and accumulation season temperature and precipitation anomalies at the Columbia Icefield for each period between 1919 and 2009. The climatic mean is based on the period 1919-2009 67 Figure 3.15 Peyto cumulative mass balance 1966 to 2007 (Demuth et al., 2009), compared with cumulative geodetic balance of glaciers of the Columbia Icefield from 1966 to 2009.. 72 Figure D.l Area and elevation change for the periods (A) 1919-1948, (B) 1919-1955, (C) 1919-1966, and (D) 1919-1979 110 Figure D.2 Area and elevation change for the periods (A) 1948-1955, (B) 1948-1966, (C) 1955-1966, and (D) 1955-1970 111 Figure D.3 Area and elevation change for the periods (A) 1966-1970, (B) 1966-1979, (C) 1970-1974, and (D) 1970-1979 112 Figure D.4 Area and elevation change for the periods (A) 1974-1979, (B) 1979-1986, (C) 1979-1993, and (D) 1986-1993 113 Figure D.5 Area and elevation change for the periods (A) 1986-1999/2001, (B) 19931999/2001, (C) 1993-2009, and (D) 1999/2001-2009 114 Figure E.l Rates of elevation change for glaciers with debris-covered sides: (A) Athabasca, (B) Saskatchewan, (C) Castleguard IV, and (D) Columbia 115 Figure E.2 Rates of elevation change for glaciers with debris-covered termini: (A) FS1, (B) Stutfield, (C) Kitchener, (D) Dome, (E) Boundary, and (F) Hilda 116 Figure E.3 Rates of elevation change for glaciers with debris-covered termini: (A) FS17, (B) Manitoba, (C) FS23, (D) FS24, and (E) FS25 117 ix List of Appendices Appendix A: Ancillary information and rectification results for Interprovincial Boundary Commission Survey (IBCS) maps and aerial photographs 90 Appendix B: Pearson's Correlation coefficients of climate variables with glacier changes in the Canadian Rocky Mountains 92 Appendix C: Raw and estimated glacier measurements and change data for each year or period of available imagery in the Columbia Icefield. Bold values were used in the estimation of missing glacier measurements indicated by a grey box 100 Appendix D: Area and elevation changes for each period used in the glacier change analysis between 1919 and 2009 110 Appendix E: Rates of elevation change between the years 1948 and 1993 on debris-covered and bare ice for debris-covered glaciers. Due to limited coverage of the 1948 photographs, I calculated elevation change for Boundary (FS7), Columbia (FS18), and FS23 between 1955 and 1993, and for Manitoba (FS19) between 1966 and 1993 115 x Acknowledgements First, I would like to thank my supervisor Brian Menounos. If it wasn't for him picking me up out of undergraduate obscurity and showing me the wonderful world of glaciers, I wouldn't be where I am today, completing my master's degree. At times it seemed like I would be working on my thesis forever, but his constant support and encouragement kept me going, especially his assurance that I was actually going to finish my thesis. I would also like to thank my committee members Roger Wheate and John Clague, as well as my external examiner, Brian Luckman, for their quick, critical, and informative revisions that strengthened my thesis. Research funding was provided by the Natural Sciences and Engineering Research Council of Canada (Canadian Graduate Scholarship), the University of Northern British Columbia (Graduate Entrance Scholarship), and the Western Canadian Cryospheric Network (funded by the Canadian Foundation for Climate and Atmospheric Sciences), without which I would not have been able to pursue a master's degree. The many aerial photographs in my thesis were scanned by Tyler Sylvestre and Callin Smith. Rectification and digitization of the Interprovincial Boundary Commission Survey maps were tackled by a team of students including Joanne Lee, Lyssa Maurer, Brenden McBain, and Natalie Saindon. Cardinal Systems provided licensing and support for their VR Mapping photogrammetry software. I specifically want to thank Jason Price who answered all of my many questions in a timely manner as I learned the software. Finally, I would like to thank my parents (both sets), family, and friends for providing support and encouragement throughout my graduate experience. Specifically, I want to thank the following people: my mom for always being available to listen and talk, celebrating with me during the highs and encouraging me during the lows; Amanda and Garrett for making sure I got out and had some fun every once in a while; Bruce and Janet for providing me with a second home every other weekend (and holidays), complete with laundry facilities and home cooked meals, which I greatly appreciate; my aunt and uncle, Diana and Richard, for opening up their home to me while I was doing an exchange at Simon Fraser University and teaching me that bacon does make it better; and my office mates, Lyssa and Natalie for sharing ideas and interests over a hot cup of tea. xi 1. Introduction Glaciers are prominent features in the Canadian Rocky Mountains. They serve as reservoirs of fresh water, supplementing summer flow in rivers when other sources have diminished (Henoch, 1971; Barry, 2006; Granshaw & Fountain, 2006; Stahl & Moore, 2006; Moore et al., 2009). Meltwater released from glaciers is important to aquatic and alpine ecosystems, and is used in irrigation, hydropower generation, and consumption (Granshaw & Fountain, 2006; Stahl & Moore, 2006; Moore et al., 2009). The Canadian Rocky Mountains contain a triple water divide, distributing water to the Pacific, Arctic, and Atlantic oceans. Globally, mountain glaciers constitute only 3-4% of the cryosphere, but significantly contribute to sea level rise, accounting for 27% of the change in sea level between 1988 and 1998 (Arendt et al., 2002; Rignot et al., 2003; Kaser et al., 2006). Glaciers are also sensitive indicators of climate change, adjusting their physical properties in response to changes in temperature and precipitation. Long records of climate and glacier change are required to understand their relationship (Dyurgerov & Meier, 2000; Barry, 2006). Glacier mass balance records provide a direct link to climate, as mass balance is the response to seasonal and annual meteorological conditions (Dyurgerov & Bahr, 1999; Pelto, 2006). The annual net mass balance indicates whether a glacier gained or lost mass during the year (Dyurgerov & Meier, 2000). Only a few such records exceed 20 years, and they are spatially limited due to the time, expense, and access needed to collect the data (Dyurgerov & Meier, 2000; Andreassen et al., 2002; Berthier et al., 2004; Barry, 2006). Glacier change can also be measured from maps, photographs, and satellite imagery, reducing the time and money spent collecting the data, and increasing the spatial and temporal representativeness of glacier monitoring (Andreassen et al., 2002; Berthier et al., 1 2004). These methods are less direct than surface mass balance measurements for assessing the health of a glacier because changes in length, area, and elevation are delayed responses to climate conditions and can occur over decades (Dyurgerov & Bahr, 1999; Dyurgerov & Meier, 2000; Barry, 2006; Pelto, 2006). The majority of the early records from the Canadian Rocky Mountains are measurements of terminus retreat or elevation and volume estimates limited to ablation areas, and only to a small number of glaciers (Field, 1948; Meek, 1948; Henoch, 1971; Kite & Reid, 1977; Reynolds, 1996). More recently area, elevation, and volume change measurements have been spatially expanded to cover the majority of glaciers in the Canadian Rocky Mountains (DeBeer & Sharp, 2007; Schiefer et al., 2007; Bolch et al., 2010), but these studies are temporally limited. My thesis contains two papers intended for publication. In Chapter 2,1 temporally extend the central and southern Canadian Rocky Mountains portion of the western Canada glacier inventory of Bolch et al. (2010) back to the early 1900s using Interprovincial Boundary Commission Survey (IBCS) maps. These maps were created by phototopographic methods during the survey of the Alberta-British Columbia (BC) border between 1903 and 1924 (Interprovincial Boundary Commission, 1917), and provide a valuable dataset for early 20th century glacier cover. I calculate changes in glacier cover from 1919 to 2006, and explore possible causes by relating changes in area to glacier properties and climatic variables. Chapter 3 presents a more spatially focused study of glacier activity during the past 90 years. I calculate changes in length, area, elevation, and volume for glaciers that drain the Columbia Icefield in the Canadian Rocky Mountains. Data sources for this chapter include 2 the IBCS maps from 1919, aerial photographs from 1948 to 1993, and satellite imagery from 1999 to 2009. I also explore the relations between glacier changes and climate using temperature and precipitation anomalies. Chapter 4 is a summary of my findings. Ancillary information about the data sources, rectification results, glacier change data for the Columbia Icefield, and climate correlations, are included in appendices. 3 2. Glacier Change in the Canadian Rocky Mountains, 1919 to 2006 2.1 Abstract I used Interprovincial Boundary Commission Survey (IBCS) maps of the AlbertaBritish Columbia (BC) border (1903-1924), BC Terrain Resource Information Management (TRIM) data (1982-1987), and Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper (ETM+) imagery (2000-2002 and 2006) to document planimetric changes in glacier cover in the central and southern Canadian Rocky Mountains between 1919 and 2006. Total glacierized area decreased by 590 ±100 km2 (40 ± 7%), with 17 of 523 glaciers disappearing and 124 glaciers fragmenting into multiple ice masses. Fourteen of the glaciers that disappeared were less than 0.5 km , and glaciers smaller than 1.0 km experienced the greatest relative area loss (64 ±17%). Variation in area loss increased with small glaciers, suggesting local topographic setting controls the response of these glaciers to climate change. Absolute area loss negatively correlates with slope and minimum elevation, and relative area change negatively correlates with mean and median elevations. Similar rates of area change were observed for the periods 1919-1985 and 1985-2001, at -6.3 ± 0.9 km2 a'1 (-0.4 ±0.1% a"1) and -5.0 ± 0.5 km2 a'1 (-0.3 ±0.1% a"1), respectively. The rate of area loss significantly increased between 2001 and 2006, -19.3 ± 2.4 km2 a"1 (-1.3 ± 0.2% a"1), with continued high minimum and accumulation season temperature anomalies and variable precipitation anomalies. 2.2 Introduction Glaciers adjust their form in response to climate. Mass balance provides the most * direct link with climate, because it is a response to meteorological conditions on a seasonal 4 or annual time scale (Dyurgerov & Bahr, 1999; Pelto, 2006). Changes in glacier area are an indirect link with climate because they are a delayed response to long-term climate change (Dyurgerov & Bahr, 1999; Dyurgerov & Meier, 2000; Barry, 2006; Pelto, 2006). Area change measurements are less intensive and more extensive than mass balance measurements, but long records are still needed to investigate the relation between glaciers and climate. Glaciers in the Canadian Rocky Mountains constitute an important freshwater resource. Glacier runoff flows into four major watersheds, the Mackenzie, Nelson, Fraser, and Columbia river basins and drains into the Arctic, Atlantic, and Pacific oceans. The contribution of meltwater to total discharge may be low, but glacier runoff supplements the summer flow and regulates stream temperature, which is important for aquatic ecosystems, irrigation, industry, hydro power and human consumption (Henoch, 1971; Barry, 2006; Granshaw & Fountain, 2006; Stahl & Moore, 2006; Moore et al., 2009; Marshall et al., 2011). As glaciers retreat, there is an initial increase in runoff followed by a decline as volume is lost, which has already occurred for glaciers in the Canadian Rocky Mountains (Moore et al., 2009; Marshall et al., 2011). Bolch et al. (2010) completed a glacier inventory of western Canadian glaciers for the period 1985-2005. Although this inventory encompasses all of the Canadian Rocky Mountains, it is temporally limited. In the Canadian Rocky Mountains, early 20th century glacier measurements tend to be limited to terminus positions of a few easily accessed glaciers (Field, 1948; Meek, 1948; Heusser, 1956). During the Interprovincial Boundary Commission Survey (IBCS) of the AlbertaBritish Columbia (BC) border between 1903 and 1924, maps containing glacier extents and 5 contours were created through photo-topographic methods using oblique terrestrial photographs taken from mountain ridges (Interprovincial Boundary Commission, 1917; Wheeler, 1920). These maps, once properly corrected for topographic distortion and systematic bias, provide an important method to extend the glacier inventory of the Canadian Rocky Mountains back in time. The objectives of this study are to: a) calculate changes in glacier cover from 1919 to 2006 for the central and southern Canadian Rocky Mountains; b) relate changes in area to glacier properties such as slope, aspect, and mean, median, and minimum elevations, and; c) compare changes in glacier cover to changes in climate. 2.2.1 Study Area The central and southern Canadian Rocky Mountains trend north-northwest and form the Continental Divide between Alberta and BC (Figure 2.1). These mountains extend from the USA border in the south, to Williston Lake in the north, and are bounded by the Rocky Mountain Trench to the west and the Rocky Mountain Foothills to the east. Differential weathering and erosion of uplifted, resistant Paleozoic carbonates and weaker Mesozoic sandstones and shales, and recurrent alpine and continental glaciations, have formed the high relief (1000 to 4000 m above sea level, asl) of the Canadian Rocky Mountains (Heusser, 1956; Osborn et al., 2006). Some of the highest peaks in the Canadian Rocky Mountains are Mt. Robson (3954 m asl) and Mt. Columbia (3747 m asl). 6 120 VV Brttish Columbia Aberta Mackenzie 0 250 500 km Nelson Freser •* * Legend • Town I6CS Map Colurnbi'a - Provincial Border WMarttttd Laka Glacier Flowihed MkxviPmk x- 0 25 50 Provincial Park IIS'W Figure 2.1 Glaciers in the central and southern Canadian Rocky Mountains. Glaciers encompassed by the flowsheds (orange) are the focus of this study and include glaciers mapped by the Interprovincial Boundary Commission Survey between 1903 and 1924. 7 Most subalpine regions of the Canadian Rocky Mountains lie within the Engelmann Spruce-Subalpine Fir biogeoclimatic zone; trees give way to alpine tundra above ca. 2250 m asl (Heusser, 1956; BC Ministry of Forests, 1998). Cold temperatures, long winters, short summers, and high amounts of precipitation, with abundant snowfall, are typical of these zones (BC Ministry of Forests, 1998). Mean annual temperature and total annual precipitation are -2.TQ and 1299 mm, respectively (Figure 2.2) (Wang et al., 2012). Canadian Rocky Mountains (2465 m) 1919-2006 -2.7'C 1299 mm 300 mm 100 40 - 20 - 13.1 10 - -10 -17.8 -20 -1 J F M A M J J A S O N D Figure 2.2 Walter and Lieth climate diagram for the central and southern Canadian Rocky Mountains, showing monthly mean temperature (red), monthly mean precipitation (blue vertical striping), frost periods (blue boxes), and probable frost periods (cyan boxes). Precipitation greater than 100 mm is plotted at a reduced scale of 10:1. Absolute maximum and minimum temperatures are given on the left (black). Data are monthly temperature and precipitation values from ClimateWNA (Wang et al., 2012), compiled over the period 1919-2006 from the center and mean elevation of each glacier. 8 Maritime polar air masses dominate the Canadian Rocky Mountains west of the Continental Divide; cyclonic storms cross the region from the North Pacific between September and June (Heusser, 1956; Henoch, 1971; Hauer et al., 1997). East of the Divide, continental polar air masses dominate, particularly during winter (Heusser, 1956; Hauer et al., 1997). In 2005, the central and southern Canadian Rocky Mountains had ca. 1800 km2 of glacier cover (Bolch et al., 2010). Glacier types include valley, cirque, icefield outlet, avalanche-fed, debris-covered, land-terminating, and lake-terminating glaciers (Heusser, 1956; Denton, 1975; Bolch et al., 2010). The main icefields from north to south are Resthaven, Reef, Hooker, Chaba, Clemenceau, Columbia (the largest), Lyell, Mons, Freshfield, Wapta, and Waputik. This study focuses on glaciers in the central and southern Canadian Rocky Mountains that were mapped during the 1903-1924 IBCS of the Alberta-BC border. 2.3 Methods 2.3.1 Interprovincial Boundary Commission Survey Maps The IBCS maps consist of 54 maps at a scale of 1:62,500 and with a contour interval of 100 feet, produced from 1903-1924. Scanned digital copies of the maps were obtained from Library and Archives Canada (LAC). Of the 54 maps, map sheets #9-36, 38, and 39 (30 maps total) contained glaciers and were used in this analysis. 9 G0022 00023 - •(HI024 I i: V 00(125 1 0001? (Mia. * (.(*>24 <30016 G002S 000*2 00035 G0022 Ci0023 000 16 G0034 (KI026 C.0039 ^iXn4gKu003J^HKj0033 . (HI036 (30026 IKW3P (HKGO 13003ft mm GftHT .1 V UU(WI (XXI17 ooom GOO is (30038 > 00041 00020 00042 4;G0i»M G0040 (30021 (KKI31 fG(XX17 fit**? (30007 G0038 00043 WXMA-J 00044 (HX)23 00022 (40023 (50024 (30024 7 • : 0002! 00043 „ (30020 (XXI20 00007 G0OI7 00041 (HX41 00042 UXS.1 • GO036 OW39 GOO.V 00039 G0024>. » ' *GOOS>7 ~ «t c»0m ' CKXMfi " * ! +- '• X ,>30031 V W :: . .. ..." «0f©> —*• ' 00029,* C.U44 : (30044 Figure 2.3 (A) Example of a raw IBCS map that was rectified using 30 GCPs collected from (B) previously rectified Landsat imagery and (C) TRIM hillshading to produce (D) a rectified map from which glacier extents and contours were extracted. 10 Table 2.1 Data used to rectify the IBCS maps and to assess glacier change from 1919 to 2006. Year (Date) 1982-87 1985 (July 31) 2001 (Sept 14) 2004 (Aug 18) 2004 (Sept 28) 2006 (Aug 18) 2006 (Aug 19) 2006 (Aug 26) Data Source (path/row) BC TRIM + Alberta DEM hillshading Landsat 5 TM (45/23) Landsat 7 ETM+ (44/24) Landsat 5 TM (47/22) Landsat 5 TM (46/23) Landsat 5 TM (43/24-25) Landsat 5 TM (44/24) Landsat 5 TM (45/24) Resolution 25 m 30 m 30 m 30 m 30 m 30 m 30 m 30 m Data Source IBCS BC TRIM/Landsat 5 TM Landsat 7 ETM+ Landsat 5 TM Scale/Resolution 1:62,500 1:20,000/30 m 30 m 30 m Data Source IBCS contours BC TRIM + Alberta mass points SRTM Resolution 100 m 25 m 90m Extents Years (Median) 1903-1924(1919) 1982-1987 (1985)" 2000-2002 (2001)" 2006 (2006)" DEMs Years (Median) 1903-1924 (1919) 1982-1987 (1985)" 1999 (1999) a For more detailed information on these data sources see Bolch et al. (2010). The extents are currently on the GLIMS website (www.glims.org). I rectified the IBCS maps in PCI Geomatica OrthoEngine v.10.2 using a 5th order polynomial transformation model that adjusts the positions of mapped features based on ground control points (GCPs). I collected 25-40 GCPs per map (Figure 2.3) from previously rectified Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper (ETM+) imagery, and a shaded relief model (hillshading) derived from the BC Terrain Resource Information Management (TRIM) digital elevation model (DEM; Table 2.1). Mountain peaks and ridges were the main source of GCPs as they were easy to identify on the maps and represent stable, vegetation-free areas. As the polynomial transformation did not use elevation data, I focused on spatially distributed GCPs rather than an altitudinal distribution. However, GCPs from the valleys were needed to provide control for glacier termini and map edges, so I used the centers of small lakes, as shorelines may have changed over ca. 80 years. The average root 11 mean square errors (RMSE) in the easting (x) and northing (y) were 12.2 and 11.5 m, respectively, with a minimum value of 2.3 m and a maximum value of 24.9 m (Appendix A). I also visually checked the features on the maps against the TRIM hillshading and Landsat imagery. 2.3.2 Data Collection I digitized glacier extents and contours from the rectified IBCS maps using ArcGIS v.9.3. There are no glacier boundary lines on the maps, although the contour lines of the glaciers are blue, rather than brown. Therefore, I digitized around the blue contours to delineate glacier extents (Figure 2.4). Where maps overlapped, I digitized glacier extents and contours from the map that exhibited the least offset from the Landsat imagery. I also generated a DEM from the contours to extract glacier properties from these old maps. I compared the IBCS glacier extents with glacier polygons in the Landsat-based glacier inventory of western Canada (Bolch et al., 2010), available from the Global Land Ice Measurements from Space (GLIMS) program (www.glims.org). Extents were available for the periods 1982-1987, 2000-2002, and 2006 (Table 2.1). All glacier extents were clipped to the IBCS map extents and separated into glacier flowsheds modified from Bolch et al. (2010). These flowsheds were based on drainage basins and, depending on a specific period, may contain one or several glaciers. I used the median year of the glacier coverage (1919,1985,2001, and 2006) to define approximate acquisition dates for the glacier data as a whole, although individual area change was calculated based on the actual year for a given flowshed. For the IBCS data, where multiple maps may have been used for a given flowshed, I used the average date of the two maps (rounded to the nearest whole year). 12 Figure 2.4 Example of (A) a glacier on the IBCS maps and (B) digitized extents and contours. 2.3.3 Error Analysis Mapping and printing errors (21% of flowsheds) were evident in the IBCS maps (Figure 2.5 A-C). In some cases, glacier contours were shifted in relation to the land contours, making the location of the glacier margin difficult to determine. I placed the glacier margin halfway between the shift of the land and glacier contours. I found one instance of a missing terminus of a large glacier (Figure 2.5 B), where the cartographer estimated the feature due to incomplete photographic coverage (Wheeler, 1988). In some cases, glaciers extended beyond the limit of the map sheet (6%). These situations pose a problem in calculating area change for a given flowshed, and so I removed them from the analysis. 13 Edited Unedited (E) 1 I 11919 I 1985 | | | 2001 12006 Figure 2.5 Errors and problems associated with the glacier extents: (A) offset 1919 glacier contours; (B) mismapped 1919 extents; (C) cut off glaciers; (D) unedited 2001 extents; (E) mismapped 1985 and 1919 extents; and (F) shadow and cloud cover. 14 When comparing IBCS glacier extents with those from 1985,2001, and 2006,1 noticed missing glaciers in one or more of the glacier inventory extents. In these cases, I manually digitized the extents from the original Landsat imagery that had been used to create the glacier inventory (see Bolch et al. (2010) for list of Landsat imagery). I also assembled and edited a ca. 2001 glacier extent for the central Canadian Rocky Mountains, as no finished inventory for this epoch existed in the data originally assembled by Bolch et al. (2010) (Figure 2.5 D). I inspected each glacier extent overlain on the imagery and manually modified the extents where I noticed mismapped glaciers. For the 1985 extents, I did not have complete Landsat imagery coverage to check the TRIM data, so any observed errors (4%) or missing glaciers (8%) could not be corrected (Figure 2.5 E). Some glacier extents (5%) could not be added or corrected due to shadows and cloud cover (Figure 2.5 F). To reduce the potential bias imposed by these mapping errors, I did not compare planimetric change of flowsheds where problematic glaciers exist and removed them before area change analysis was conducted. Of the original 937 flowsheds, 414 (44%) were removed, leaving 523 flowsheds for analysis. The error terms for glacier extents from the glacier inventory of western Canada are 3-4% (Bolch et al., 2010). However, I had to modify the extents for snow cover, shadows, and missing glaciers, so I calculated new error terms for all extents using a buffer method similar to Granshaw & Fountain (2006) and Bolch et al. (2010). I calculated an error term for the 1919 extents using a buffer equal to the estimated horizontal error (30 m), incorporating the RMSE of the map rectification (16 m), a digitizing error equal to half the width of a contour line (7.5 m), and half the mean offset between overlapping maps (25 m). For the glacier inventory extents, I used a buffer equal to half the resolution of the data 15 (Table 2.1). As the 1985 extents were from TRIM data and Landsat imagery, I used a buffer equal to half of the combined resolution (11m). Between each period, I calculated a RMSE term using the error estimates from the two years that make up a period. Calculated error terms are listed in Table 2.2 and range from 7.8 to 18.8% for the individual years and 9.8 to 16.6% for the periods. Table 2.2 Error estimates for area and area change. Year Error' (%) Period 18.8 1919 1919-1985 7.8 1985-2001 1985 11.5 2001-2006 2001 13.3 2006 1919-2006 8 Mean error estimate calculated from individual glacier buffers. b Mean RMSE calculated from the error estimates of the two differenced extents. Error* (%) 14.4 9.8 12.5 16.6 2.3.4 Glacier Change Analysis I calculated the area and number of ice masses within each flowshed. I used the zonal statistics tool in ArcGIS v9.3 on the DEMs from the IBCS contours (1919), TRIM data (1985), and the Shuttle Radar Topography Mission (SRTM; 1999), with the flowsheds as zones. From the zonal statistics, I determined the mean, median, and minimum elevations of ice within each flowshed. I also calculated zonal statistics on slope, aspect, and incoming solar radiation (insolation, watt hours per square meter, W h m"2) surfaces derived from the DEMs. I calculated absolute (km2) and relative (%) area change and rates between successive years, and compared them with the properties mentioned above. 2.3.5 Climate Data I obtained monthly minimum, mean, and maximum temperatures and monthly total precipitation from 1901 to 2006 at the center of each glacier, based on latitude, longitude, and elevation, from the ClimateWNA v.4.62 program (http://www.genetics.forestry.ubc.ca/ 16 cfcg/ClimateWNA/Climate WNA.html). This program extracts climate data at specific locations from downscaled PRISM and historical data using anomalies (Wang et al., 2012). Anomaly surfaces are derived from interpolated historical data created by Mitchell & Jones (2005) and a baseline reference grid (2.5 arc min) of monthly climate data for the 1961-1990 normal period generated by PRISM (Daly et al., 2002). Bilinear interpolation and lapse-rate elevation adjustments are used to integrate the historical anomaly surfaces and the baseline grid and downscale the climate data at a specific location (Mbogga et al., 2009; Wang et al., 2012). More detailed information on the methodology behind the datasets can be found in Wang et al. (2012). I calculated total annual precipitation based on the water year, October to September, as well as total ablation (May to September) and accumulation (October to April) season precipitation. Total precipitation was used because snow is common throughout the year at high elevation. I determined minimum, mean, and maximum annual temperature, in addition to ablation and accumulation season temperature. I compared area change with annual and seasonal total precipitation and minimum, mean, and maximum temperature, by flowshed. Also, I averaged the temperature and precipitation data over all of the flowsheds to produce one value for the region for each year. I derived temperature and precipitation anomalies from the 1919-2006 annual and seasonal means and grouped them into three periods: 1919-1985, 1985-2001, and 2001-2006. I choose these three periods to match the periods of area changes and rates in order to test for correlations. Also, to determine a possible delayed response between area change and climate, I lagged the start and end years of the periods of anomalies up to 19 years. 17 2.4 Results 2.4.1 Glacier Properties In 1919, the Canadian Rocky Mountains contained 569 ice masses in 523 flowsheds inventoried by this study, with a total area of 1470 ± 280 km2. Each flowshed represents total ice draining an area, which may include multiple ice masses (e.g. an avalanche-fed glacier with a separate accumulation area or disconnected tributary glaciers). For the 1919 data, however, the glacierized area within a flowshed is considered to represent one glacier. • number 1919 B area 1919 • number 2006 • area 2006 0.05-0.1 0.1-0.5 0.5-1.0 1.0-5.0 5.0-10.0 >10.0 Area (km2) Figure 2.6 Size class distribution of glaciers in the Canadian Rocky Mountains for the years 1919 and 2006, by percent glacier number and area. Glacier areas ranged in size from 0.06 ± 0.01 km2 to 50 ± 9 km2 (Freshfield Glacier); •y the mean glacier area was 2.9 ± 0.5 km . I separated glaciers into six size classes based on 18 the 1919 areas: 0.05-0.1 km2,0.1-0.5 km2,0.5-1.0 km2,1.0-5.0 km2, 5.0-10.0 km2, and> 10.0 km2. The 1.0-5.0 km2 class contained the most glaciers (37% of the total), but "i 0 combined, glaciers smaller than 1.0 km contained 49%; the > 10.0 km class had the greatest area (41%; Figure 2.6). The glaciers occurred between 1410 and 3860 m asl and had a mean elevation range of 620 m. Both mean and median elevations of the glaciers were 2470 m asl, and these values increase to the southeast. The mean slope was 19°, with a range from 6° to 46°. The majority of glaciers faced north and northeast (45%); few glaciers had a southwest aspect (5%). S g -• o m 6 w b o o o 0.01 0.05 0.50 5.00 50.00 1919 Area (km2) Figure 2.7 Log-log plot of 2006 glacier area versus 1919 glacier area. Points are separated into groups based on an increase (disintegrating), decrease (disappearing), or no change (same) in the number of ice masses within a flowshed. The solid line is a linear model fitted to the data described by (Eq. 2.1. The one-to-one line is dashed. 19 By 2006, glacierized area in the Canadian Rocky Mountains had decreased to 880 ± 120 km2, a loss of 590 ± 100 km2 since 1919. The number of ice masses increased to 724 due to the disintegration of 124 glaciers (Figure 2.7). Despite this increase, 13 ice masses disappeared from flowsheds still containing glaciers in 2006, and 17 glaciers disappeared completely. Fourteen of the 17 glaciers that disappeared were smaller than 0.5 km2. Since 1919, mean glacierized area per flowshed has decreased to 1.7 ± 0.2 km . There also has been a shift in the number of glaciers by size class to smaller glaciers (Figure 2.6). In 2006 the 0.1-0.5 km2 class contained the greatest percentage (45%) of glaciers. Elevation data were not available for 2006, so elevation properties, slope, and aspect, are based on data from 1999. Glaciers occurred at elevations between 1570 and 3660 m asl, a decrease in the elevation range since 1919. However, the decrease in maximum elevation may be due to errors in the uncorrected 1919 DEM. Mean and median elevations increased to 2530 m asl, mean slope increased to 21°, and the distribution of glaciers by aspect remained unchanged. 2.4.2 Area Change From 1919 to 2006, glaciers lost a total area of 590 ± 100 km2 (40 ± 7%) at a rate of -6.8 ± 1.1 km2 a'1 (-0.5 ± 0.1% a"1). The 1.0-5.0 km2 class had the greatest absolute area loss (242 ± 30 km2; Figure 2.8). Relative area loss was greatest for the 0.5-1.0 km2 class (67.8 ± 12%); glaciers smaller than 1.0 km2 lost on average 64 ± 17% of their area (Figure 2.8). Glaciers in the 0.05-0.1 km2 class decreased least in total area, whereas large glaciers (> 10.0 km2) lost the smallest area when expressed as a percentage. 20 100 005-01 0.1-05 0.5-1.0 1.U.0 5.0-10.0 >100 Size class (ion2) Size class (km2) Figure 2.8 Total (A) absolute and (B) relative area change by size class of glaciers in the Canadian Rocky Mountains between 1919 and 2006. Error bars are shown representing the mean error for each size class. Rowsheds = 523 508 Flowsheds * 523 508 508 508 in 2. o T o 5 in 5 - 1919-1985 1985-2001 2001-2006 1919-1985 1985-2001 2001-2006 Period Period % Figure 2.9 (A) Absolute and (B) relative rates of area change for the Canadian Rocky Mountains, over three periods from 1919 to 2006. Boxes represent the first and third quartiles with the horizontal black line as the median. The whiskers represent the data extremes (5th and 95th percentile) and the circles are outliers. 21 The median absolute (relative) rates of area change for the periods 1919-1985 and 1985-2001 are similar - -0.0065 ± 0.0009 km2 a'1 (-0.50 ± 0.07% a1) and -0.0047 ± 0.0005 km2 a"1 (-0.45 ± 0.04% a"1), respectively (Figure 2.9). Rates of area change for the period 2001-2006 are significantly higher - -0.0200 ± 0.0025 km2 a'1 (-1.67 ± 0.21% a"1). Total absolute (relative) rates of area change are -6.3 ± 0.9 km2 a'1 (-0.43 ± 0.06% a"1), -5.0 ± 0.5 km2 a"1 (-0.34 ± 0.03% a"1), and -19.3 ± 2.4 km2 a"1 (-1.31 ± 0.16% a"1) for the periods 19191985,1985-2001, and 2001-2006, respectively. These patterns in the absolute and relative rates of area change are consistent across the different glacier size classes. As expected, a strong relationship (r2 = 0.87, p < 2.2e-16) exists between areas of glaciers in 1919 and 2006 (Figure 2.7): 1919Area = j0°«°log(2^™)+ 0337 (Eq 2 ]) Eq. 2.1 can thus be used to estimate glacier extents in regions of the Canadian Rocky Mountains not covered by the IBCS. Using the 2006 area of each glacier in the central and southern Canadian Rocky Mountains in the glacier inventory created by Bolch et al. (2010), I calculated a total glacierized area of 3160 ± 150 km2 for 1919. The total glacierized area of this region in 2006 was 1770 km2, resulting in an area change of -1390 ± 80 km2 (-44 ± 3%) from 1919 to 2006. Most variability in Figure 2.7 is associated with glaciers with areas between 0.5 and 5 km2. I correlated the model residuals with possible explanatory factors, such as elevation, slope, aspect, or climate. Mean slope had the highest correlation, with an r value of -0.39 (p < 0.01), followed by mean and median elevations (r = -0.19, p < 0.01), maximum ablation season temperature (r = -0.18, p < 0.01), and mean ablation season temperature (r = -0.17, p < 0.01). There are no significant correlations between latitude or longitude and residuals, 22 and no observable trends between residuals and aspect. Because slope has a moderate correlation, I included it in the model, which improved the relation (r2 = 0.90, p < 2.2e-16) and reduced the residual standard error from 0.20 to 0.18. 2.4.3 Area Change with Properties I compared absolute and relative area changes in the Canadian Rocky Mountains to glacier attributes such as minimum, mean, and median elevations, surface slope, latitude, longitude, and insolation (Table 2.3) using Pearson's correlation. Throughout all periods, absolute area changes moderately correlate with minimum elevation (mean r = 0.43, p < 0.01) and mean slope (mean r = 0.40, p < 0.01), whereas relative area changes correlate with mean elevation (mean r = 0.18, p < 0.01), median elevation (mean r = 0.21, p < 0.01), and minimum elevation (mean r = -0.16, p < 0.01). There are weak significant correlations between both types of area change and latitude, longitude, and insolation, but they differ in strength between the periods. Absolute area changes are greatest for glaciers with north and northwest aspects; these aspects support the largest glacier area. There is no observable trend between relative area change and aspect. The correlations between rates of area change and the glacier properties mirror those between area change and glacier properties. 23 Table 2.3 Pearson's correlation coefficients (r values") of glacier properties and climate variables with absolute and relative area change by flowshed (n = 506). Glacier Property* 1914M985 " Abiotate Are# Change (Rate) 1985-2001 Latitude Longitude Insolation 0.425" (0.446^" -0.026 (-0.014) -0.050 (-0.038) 0.478" (0.486") 0.102* (0.079) -0.097* (-0.072) -0.170** (-0.170 ) Climate Variable* 1919-1985 Emin ^mcan Emod «mean J 0.379" (0.376 ) -0.001 (0.003) -0.029 (-0.026) 0.297** (0.297**) 0.017 (-0.002) -0.021 (-0.003) -0.061 (-0.063) Abnolute Area Change (Rate) 1985-2001 2001-2006 — 1919-ms Relative Area Change (Rate) 1985-2001 2001-2006 0.476 (0.461 ) -0.004 (-0.027) -0.017 (-0.039) 0.413" (0.406**) -0.16" (-014") 0.14" (0.18**) 0.16" (0.20**) -0.20" (-020") 0.17 (0.18") 0.20" (0.21") -0.12" (-0.14") 0.24** (0.20**) 0.27" (0.23**) -0.022 (0.021) 0.15" (0.15") -0.20" (-0.21**) -0.12" (-0.12**) -0.07 (-0.11*) 0.024 (-0.023) -0.098 (-0.101*) 0.13" (0.06) -0.13" (-0.07) -0.07 (-0.06) -0.21**(-0.13") 0.22" (0.13**) 0.17" (0.15**) 2001-2006 1919-1985 0.06(0.10*) 0.14" (0.14**) Relative Area Change (Rate) 1985-2001 2001-2006 r*Annual 0.091 (0.075) 0.081 (0.073) -0.14" (-0.16") 0.084(0.117") -0.13" (-0.17") -0.20 (-0.14 ) 0.051 (0.036) 0.058 (0.050) 0.059(0.091*) -0.15" (-0.19**) -0.14" (-0.16**) -0.19" (-0.13") Abl. Tjnax 0.101* (0.093*) 0.127" (0.110*) 0.105* (0.139**) -0.14" (-0.16**) -0.20" (-0.14**) -0.10*(-0.15**) Acc. Tnjax 0.083 (0.065) Annual Tmean 0.064 (0.060) 0.079(0.107*) -0.13" (-0.17 ) -0.19" (-0.14**) -0.15" (-0.16**) 0.029 (0.015) 0.053 (0.073) -0.17" (-0.20") -0.15" (-0.11*) 0.030 (0.028) -0.13" (-0.14") Abl. Tmean Acc. Tmean 0.133" (0.112*) 0.095* (0.088*) 0.100* (0.135**) -0.08 (-0.14**) -0.16" (-0.18**) -0.21" (-0.15") 0.101* (0.088*) 0.098* (0.084) 0.087 (0.124**) -0.12** (-0.16 ; -0.12" (-0.16**) -0.20" (-0.13**) Annual Tnu,, -0.14" (-0.17**) Abl. Tm,n 0.077 (0.062) 0.092* (0.076) 0.060 (0.106*) -0.11 *(-0.15**) -0.22" (-0.14") Acc. Tmin -0.12" (-0.15**) 0.098* (0.086) 0.103* (0.090 ) 0.089* (0.126**) -0.13** (-0.17 ) -0.20" (-0.13**) -0.178" (-0.182**) -0.102* (-0.099*) 0.03 (0.04) Annual P -0.171" (-0.175**) 0.16" (0.16**) 0.12" (0.09*) 0.030 (0.025) 0.013 (0.015) 0.03 (0.03) -0.037 (-0.028) 0.04 (0.03) Abl. P -0.07 (-0.05) 0.19" (0.21") 0.15 (0.10*) -0.168** (-0.160") -0.251" (-0.258**) -0.221" (-0.232") Acc. P 0.10* (0.10*) a Significant correlations are denoted by p < 0.05 and p < 0.01. b Emin is minimum elevation, Emean is mean elevation, and E med is median elevation. c Tmax is maximum temperature, Tmean is mean temperature, Tmjn is minimum temperature, and P is precipitation. Abl. is ablation season and Acc. is accumulation season. Annual Ablation (MJJAS) Accumulation (ONDJFMA) • mm <8191946 hb 1 $46.1SOS mmm 1919-1985 mm 1M5-20W Annual Ablation (MJJAS) Accumulation (ONDJFMA) Figure 2.10 Regional (A) maximum, (B) mean, and (C) minimum temperature anomalies calculated for the average annual, ablation, and accumulation seasons for the periods 1919-1985, 1985-2001, and 2001-2006. (D) Precipitation anomalies calculated from the precipitation totaled over the hydrologic year (annual), and ablation and accumulation seasons for the same periods. The climatic mean is based on the period 1919-2006. Additional periods, 1919-1946 and 1946-1985, are included to show the change in climate over the winter of 1945/1946. 25 2.4.4 Climate Temperature anomalies are ca. -0.1°C from the 1919-2006 mean for the period 19191985, with accumulation season (ca. -0.2°C) and all minimum temperature anomalies (ca. -0.3°C) slightly below average (Figure 2.10). Since 1985, all minimum temperature anomalies are ca. 0.7°C above the long-term average. Maximum ablation season temperature anomalies are ca. -0.4°C below average, but accumulation season temperature anomalies are ca. 0.4°C above average for the period 1985-2001. In the recent period, 20012006, all temperature anomalies are 0.5°C above average except for ablation season maximum temperature anomalies. Precipitation anomalies are negative (ca. -20 mm) for the period 1919-1985 and positive (ca. 25 mm) for the period 1985-2001. The period 2001-2006 had the most negative precipitation anomaly (ca. -68 mm) and the most variability. I examined the spatial and temporal relations between climate and glacier change. I correlated absolute and relative area changes and rates with epoch-averaged temperature and precipitation for each flowshed (Table 2.3). Absolute area changes and rates were correlated with accumulation season (mean r = -0.21, p < 0.01) and annual precipitation (mean r = -0.15, p < 0.05), as well as mean and maximum accumulation season temperature (mean r = 0.11, p < 0.05). Relative area changes and rates are correlated with all temperatures (mean r = -0.15, p < 0.05) and with accumulation season precipitation (mean r = 0.15, p < 0.05). I compared the total absolute and relative area changes and rates in the Canadian Rocky Mountains to epoch-averaged precipitation and temperature anomalies (Table 2.4). Absolute and relative area changes are strongly correlated with mean and minimum temperature anomalies, but only annual minimum temperature anomalies are significant (p < 26 0.05). The absolute and relative rates of area change are strongly correlated with maximum temperature anomalies, and annual and accumulation season precipitation. However, the only significant correlation is between the absolute rate of area change and accumulation season precipitation (p < 0.05). Table 2.4 Pearson's correlation coefficients (r values8) of climate variables with absolute and relative area changes and rates, by period (n = 3). Annual Tm„ 0.672 -0.943 0.670 -0.942 Abl. Tmax -0.375 -0.710 -0.376 -0.712 ACC. Tmax Annual Tmran 0.730 -0.912 0.729 -0.912 0.941 -0.676 0.940 -0.674 Abl. Tmean 0.977 -0.573 0.977 -0.572 ACC. Tmean Annual Tmm Abl. Tmin 0.897 -0.754 0.897 -0.752 0.999* 0.966 -0.420 -0.134 0.999* 0.966 -0.418 -0.132 ACC. Tmn, 0.981 -0.556 0.981 -0.554 Annual P 0.013 0.917 0.015 0.918 Abl. P 0.576 0.532 0.577 0.534 Acc. P -0.338 0.999* -0.336 0.999 a Significant correlations are denoted by p < 0.05. b T max is maximum temperature, Tmean is mean temperature, T,„jn is minimum temperature, and P is precipitation. Abl. is ablation season and Acc. is accumulation season. I correlated area changes and rates with lagged precipitation and temperature anomalies to examine area change response time (Appendix B). Temperature anomalies have significant (p < 0.05) positive correlations with absolute and relative area changes with a lag of up to a decade. Precipitation anomalies show strong, positive correlations with a lag of six years, but only the ablation season anomaly with a lag of seven years is significant (p < 0.05). Absolute and relative rates of area change have strong negative correlations with temperature anomalies with a lag of 12 to15 years, and significant (p < 0.05) correlations with mean temperature anomalies. There are no significant correlations between rates and 27 precipitation anomalies, but strong positive correlations are evident for the first two years and again with a lag of 18 years with annual and accumulation season anomalies. 2.5 Discussion 2.5.1 Area Change Glaciers in the Canadian Rocky Mountains, analyzed by this study, decreased 590 ± 100 km2 (40 ± 7%) between 1919 and 2006. The errors reflect uncertainties in the glacier extents, primarily from the IBCS maps where extents may have been incorrectly mapped due to snow cover, or estimated where photographic coverage was incomplete or the perspective was poor (Wheeler, 1988). However, shadows, debris cover, snow cover, and clouds introduce uncertainties in all years. I compared my results to those of Bolch et al. (2010) from the southern and central Canadian Rocky Mountains, because I used a subset of their glacier inventory data. A perfect comparison was not expected because I modified and edited the glacier extents, but I found my area changes comparable to those of Bolch et al. (2010) within error. Between 1985 and 2005, area changes from Bolch et al. (2010) are -17.5 ± 4.1% and -14.8 ± 4.1% for the central and southern Canadian Rocky Mountains, respectively, similar to my estimates of -16.7 ± 1.9% for the combined region over the same period. The rates for Bolch et al. (2010), -0.47± 0.36 and -1.21 ± 0.96% a"1, are also similar to my estimates, -0.34 ± 0.03 and -1.31 ± 0.16% a"1, for the periods 1985-2000 and 2000-2005, respectively. Direct comparisons are difficult to make with other inventories, as most values are calculated for different periods. Interpretation differences, mapping errors, and the number of glaciers may also contribute to discrepancies between inventories. Other glacier area 28 change estimates for the Canadian Rocky Mountains range from -15% to -25% from ca. 1950 to 2000 (Luckman & Kavanagh, 2000; DeBeer & Sharp, 2007), and -22 to -36% from 1975 to 1998 (Demuth et al., 2008). Jiskoot et al. (2009) calculated rates of area change of -9% per decade and -19% per decade for the Clemenceau and Chaba icefields, respectively, from 1985 to 2001. Glaciers on the north and south coasts of British Columbia show lower relative area changes (-8 to -10% from 1985-2005) than the Canadian Rocky Mountains, primarily due to the larger sizes of glaciers and possible influence of a maritime climate (Bolch et al., 2010). Most inventories for the Northern Hemisphere determine glacier change after 1950 to the end of the 20th century, and range from 7 to 32% area loss (Kaab et al., 2002; Paul, 2002; Granshaw & Fountain, 2006; Bolch, 2007). Two studies that determined glacier area change since the early 1900s report area losses of 23% between 1930 and 2003 for Jotunheimen, Norway (Andreassen et al., 2008), and 49% and 35% between 1850 and the mid-1970s for glaciers in the New Zealand and European Alps, respectively (Hoelzle et al., 2007). Patterns of area loss are more easily compared than area loss values. In the Canadian Rocky Mountains, small glaciers (< 1.0 km2) lost the greatest percentage of their area. This result is consistent with those of the majority of the studies mentioned above (Kaab et al., 2002; Paul, 2002; Granshaw & Fountain, 2006; Demuth et al., 2008). Granshaw & Fountain (2006) argue that this difference is due to a high area-to-volume ratio, so for the same ablation rate, small glaciers should shrink faster. Another possible explanation is that small glaciers have a higher perimeter-to-area ratio, which makes them increasingly susceptible to radiation from the surrounding terrain (Demuth et al., 2008). 29 DeBeer & Sharp (2007) found that glaciers in the 1.0-5.0 km2 class had the highest absolute area loss due to the large number of glaciers in this size class and high individual area loss, similar to my results (Figure 2.8). However, they found that small glaciers (< 0.5 km2) experienced little or no change. The smaller glaciers tended to be located in more sheltered locations, at high elevations, or at sites with reduced insolation or inputs from avalanching or wind (DeBeer & Sharp, 2007; Demuth et al., 2008). This sheltering effect may be present in my data, because although I observed large percentages of area loss for small glaciers, the 0.5-1.0 km2 class lost the highest percentage, not the smallest class (0.050.1 km2). Another trend apparent in my data and other studies (Kaab et al., 2002; Granshaw & Fountain, 2006; DeBeer & Sharp, 2007; Andreassen et al., 2008) is an increase in variability with smaller glacier sizes; the scatter can be greater than the differences between glacier size classes. It likely arises from non-climatic factors such as local topography, hypsometry, and glacier type (e.g. debris-covered), which may have a stronger influence on a glacier's response than regional climate (Kaab et al., 2002; Granshaw & Fountain, 2006; DeBeer & Sharp, 2007; Hoffman et al., 2007; Andreassen et al., 2008; Jiskoot et al., 2009). Correlating the residuals from (Eq. 2.1, which represent the variance in the model of glacier area with glacier properties, showed that slope (r = -0.39) could be a factor in explaining the scatter. Adding slope to the model increased the r2 value from 0.87 to 0.90 (p < 2.2e-16). 2.5.2 Area Change with Glacier Properties Glacier size is the main property related to area change, but moderate correlations were also found between area changes and slope and mean, median, and minimum elevations. Absolute area loss increased with lower slopes and lower minimum elevations. 30 These properties are typically associated with larger glaciers that extend down into valleys. Small temperature changes with elevation can influence large areas of the glacier due to the low slopes and may increase the area over which melting occurs. Correlations with slope were also noted by Andreassen et al. (2008) and Jiskoot et al. (2009). Relative area loss increased with decreasing mean and median elevations. However, Bolch et al. (2010) did not find any correlation between area loss and median elevation for glaciers in western Canada. Their data, however, included all glaciers in Alberta and BC, from different climates (i.e. maritime and continental), which may allow glaciers to exist at different elevations, obscuring any regional trends. Granshaw & Fountain (2006) noted a weak correlation with median elevation, as did Andreassen et al. (2008). North- and northeast-facing glaciers decreased the most, a finding similar to Andreassen et al. (2008), but these aspects also contain the greatest glacier area. 2.5.3 Area Change with Climate Temperature and precipitation explain some of the spatial glacier changes in this study. Accumulation season precipitation has the highest correlation with absolute area change by flowshed, indicating the importance of precipitation, not just temperature on area loss. Annual and minimum ablation season temperatures are weakly correlated with percent glacier shrinkage, in agreement with previous work (Bitz & Battisti, 1999; Hoffman et al., 2007). The period 1919-1985 contains at least two different climate conditions (Figure 2.10). Around the 1920s some glaciers were still near their Little Ice Age maximum extents (Field, 1948). From the 1920s to the 1950s, the climate was warm and dry, and glaciers retreated at high rates (Field, 1948; Heusser, 1956; Luckman & Kavanagh, 2000; Pelto, 2006; Hoffman 31 et al., 2007; Andreassen et al., 2008). For the second half of the period, the climate was cooler and wetter, resulting in slowed retreat or minor advances of many glaciers in the region (Henoch, 1971; Luckman et al., 1987; Luckman & Kavanagh, 2000; Hoffman et al., 2007; Andreassen et al., 2008). The change in climate over the winter of 1945/1946 corresponds with a shift in the Pacific Decadal Oscillation (PDO) (Bitz & Battisti, 1999; Demuth et al., 2008). There was another shift in the PDO, in 1976/1977, which may have influenced the last decade of the period and persisted into the period 1985-2001. Unfortunately, these different climates are not captured in the area change or climate data, as aggregated here over both climate conditions from 1919 to 1985 produced low rates of area loss and temperature and precipitation anomalies near the 1919-2006 mean. Rates of area loss for the period 1985-2001 are lower than those for the period 19191985, despite positive temperature anomalies (ca. 0.5°C). Warmer temperatures and higher retreat rates during this period are mentioned in other studies (Pelto, 2006; Hoffman et al., 2007). However, the accumulation season precipitation anomalies were also positive, and the increased precipitation may have offset some area loss due to the warmer temperatures. A contributing factor to the low rates of area loss for the period 1985-2001 may have been the snow cover in some of the 2000-2002 Landsat imagery resulting in an overestimation of glacier area (Bolch et al., 2010). Glaciers also may have downwasted more than retreated over this period. Area change is not an immediate response to a change in climate. Some glaciers may have been responding to the cooler and wetter climate near the end of the period 1919-1985. 32 The rate of glacier shrinkage was much higher between 2001 and 2006 than in the previous periods. A combination of positive temperature anomalies (> 0.5°C above average) and negative precipitation anomalies (ca. -68 mm) may be responsible. The high rate may also be a continued response from the warm temperatures of the 1980s and 1990s. I correlated area changes and rates over the periods with lagged temperature and precipitation anomalies. Area changes are correlated with temperature anomalies lagged one to ten years and precipitation anomalies lagged six years. These values are comparable to other studies that have shown terminus changes responding to temperatures lagged a few years and precipitation lagged five to ten years (Salinger et al., 1983; Sigurdsson et al., 2007; Beedle et al., 2009). However, the rates of area change do not correlate with temperature and precipitation anomalies lagged the same number of years as area change. My correlations are limited to three periods, which may be a factor in the low number of significant correlations. 2.6 Conclusions I used IBCS maps from 1919 and a portion of the western Canada glacier inventory from 1985, 2001, and 2006 to determine area change of glaciers in the central and southern Canadian Rocky Mountains. The main uncertainties and errors arise from IBCS map errors (e.g. snow patches, shapes, and incorrect termini) and glaciers cut off at map boundaries. Although large errors are associated with the IBCS maps, they are a useful resource for early 20th century glacier extents and elevation, and over long periods, glacier changes are significantly larger than the errors in the maps. Other sources of error in the most recent 33 datasets include late-lying snow, shadows, and debris cover, which hinder glacier delineation. Area change was influenced predominately by glacier size, with large glaciers losing the most absolute area and small glaciers losing the most percentage of their area. Variability increases with smaller glaciers, suggesting local non-climatic factors modulate the response of these glaciers to climate. Although glacier properties, such as elevation and slope, correlate with area change, future research should focus on other properties related to glacier type and source of nourishment (i.e. outlet, cirque, avalanche-fed, and debris-covered glaciers). Temperature and precipitation are both important influences on area change, spatially correlating with individual glacier area change, and temporally correlating with area change from 1919 to 2006. 34 3. Glacier Change of the Columbia Icefield, 1919-2009 3.1 Abstract I determined length, area, elevation, and volume changes of the Columbia Icefield using Interprovincial Boundary Commission Survey (IBCS) maps from 1919, eight sets of aerial photographs from 1948 to 1993, and satellite data from 1999 to 2009. Over the period 1919-2009, glaciers on average retreated 1149.9 ± 34.1 m and shrank by 2.38 ± 0.24 km2. Total area loss was 59.60 ± 1.19 km2 (22.5 ± 5.0 %) and mean elevation change was -49.4 ± 25.2 m water equivalent (w.e.), resulting in a total volume loss of 14.30 ± 2.02 km3 w.e. Large outlet glaciers experienced the greatest absolute ice loss, while small detached glaciers lost the most relative length and area. The percentage of debris-covered ice decreased slightly from 1948 to 2009, but thinning rates of debris-covered ice were 30-60% lower than those for clean ice. Between 1919 and the mid-1970s, periods of ice loss were influenced by positive temperature anomalies and negative precipitation anomalies, whereas periods of slower loss or ice gain were influenced by negative temperature anomalies and positive precipitation anomalies. Different rates of glacier change were observed between the mid-1970s and 2009, when temperature and precipitation anomalies had competing influences. The complex response of these glaciers to variable climate likely arises from differing response times of the glaciers and the importance of both precipitation and temperature on long-term changes in glacier mass. 35 3.2 Introduction Glaciers adjust their length, area, thickness, and volume in response to temperature and precipitation changes over periods of years to decades (Dyurgerov & Bahr, 1999; Dyurgerov & Meier, 2000; Barry, 2006; Pelto, 2006). They are also important freshwater resources, as ice lost or gained during these adjustments is stored and later released for use in aquatic and alpine ecosystems, irrigation, industry, hydropower, and consumption, and eventually contributes to sea level rise (Granshaw & Fountain, 2006; Stahl & Moore, 2006; Moore et al., 2009). The Columbia Icefield, in the Canadian Rocky Mountains, plays a significant role as a water resource because it is situated on a triple water divide. Meltwater from the glaciers flows into the Athabasca, Saskatchewan, and Columbia watersheds, which drain, respectively, to the Arctic, Atlantic, and Pacific oceans. The Columbia Icefield also has economic importance as a major tourist attraction in Jasper and Banff national parks, with millions of people visiting the icefield each year (Parks Canada Agency, 2011). The environmental and economic importance of the Columbia Icefield, in addition to its relative ease of access, has drawn people to document the changes of its outlet glaciers for more than 100 years. Mountaineers and surveyors were the first to measure retreat of Athabasca, Saskatchewan and Columbia glaciers (Field, 1948; Meek, 1948; Heusser, 1956; Denton, 1975). They recorded changes in glacier length, and in some cases thickness and width. However, only a few studies estimated area and elevation change, primarily limited to the ablation areas of the glaciers (Konecny, 1963; Kite & Reid, 1977; Reynolds, 1996; Reynolds & Young, 1997). The Water Survey of Canada monitored Saskatchewan and Athabasca glacier from the 1940s to the 1980s by mapping the glaciers and calculating changes in terminus position, elevation, and volume (Reid & Charbonneau, 1979; Luckman, 36 1986; Reynolds, 1996). With the advent of satellite remote sensing, more glaciers of the Columbia Icefield are being monitored (DeBeer & Sharp, 2007; Schiefer et al., 2007; Demuth et al., 2008; Bolch et al., 2010), but few of these studies use data that span more than several decades. Long-term time series in glacier length, area, and volume are needed to properly evaluate the response of these glaciers to changes in climate. The primary objective of this study is to use Interprovincial Boundary Commission Survey (1BCS) maps from 1919 in conjunction with aerial photographs from 1948 to 1993 and satellite imagery from 1999 to 2009, to extend, spatially and temporally, measured glacier changes to other glaciers of the Columbia Icefield. I use sequential analysis of glacier extents and elevation data to calculate length, area, elevation, and volume changes of glaciers of the Columbia Icefield, including accumulation areas. I compare the glacier changes to temperature and precipitation anomalies from 1919 to 2009. 3.2.1 Study Area The Columbia Icefield straddles the Continental Divide along the Alberta-British Columbia (BC) border (Figure 3.1). The local relief is ca. 2700 m, ranging from 1000 to 3700 m above sea level (asl). Major mountains in the vicinity of the icefield include Mt. Columbia (3747 m asl), Mt. King Edward (3475 m asl), Mt. Stutfield (3450 m asl), the Twins (3684 and 3559 m asl), Mt. Kitchener (3505 m asl), and Mt. Castleguard (3077 m asl) (Heusser, 1956). The lower part of the region lies within the Engelmann Spruce-Subalpine Fir biogeoclimatic zone; this zone gives way to alpine tundra above ca. 2250 m asl (Heusser, 1956; BC Ministry of Forests, 1998). 37 f-T^r/v Figure 3.1 The Columbia Icefield, Canadian Rocky Mountains. Glaciers encompassed by the orange flowsheds are the focus of this study. The image is a SPOT 5 scene from 30 August 2009. The climate is characterized by long winters and short summers, with cold temperatures and high amounts of precipitation falling primarily as snow (BC Ministry of Forests, 1998). Mean annual temperature and total annual precipitation of the Columbia Icefield are respectively, -4.0°C and 1277 mm (Figure 3.2). The icefield is subject to both maritime polar air masses from the west, bringing cyclonic storms between September and June, and continental polar air masses from the east, particularly during winter (Heusser, 1956; Henoch, 1971; Hauer et al., 1997). This study focuses on 25 glaciers of the Columbia Icefield based on glacial drainage basins (Figure 3.1). Glaciers include individual ice bodies and outlet glaciers, some of which 38 have significant icefalls and are fed by avalanching. About 70% of the glaciers are debris covered; 50% have debris-covered termini and the other 20% have debris-covered sides. The main outlet glaciers draining the icefield are Columbia on the west, Saskatchewan and Athabasca on the east, Castleguard IV on the south, and Stutfield on the north. Total ice cover was 205 km2 in 2009, with a mean glacier area of 8.2 km2. Glacier mean elevation is 2580 m asl and mean slope is 21°. The majority of glaciers face north to east. Detailed properties for each of the glaciers are listed in Table 3.1. Columbia Icefield (2586 m) 1919-2009 -4°C 1277 mm 13.2 -19.3.20 J J F M A M J J A S O N D Figure 3.2 Walter and Lieth climate diagram for the Columbia Icefield showing monthly mean temperature (red), monthly mean precipitation (blue vertical stripes), frost periods (blue boxes), and probable frost periods (cyan boxes). Precipitation greater than 100 mm is plotted at a reduced scale of 10:1. Absolute maximum and minimum temperatures are given on the left (black). Data are monthly temperature and precipitation values from ClimateWNA (Wang et al., 2012), compiled over the period 1919-2009 from a range of elevations on a 1 km grid over the icefield. 39 Table 3.1 Properties of glaciers of the Columbia Icefield by flowshed (values calculated from 2009 data). Flowched Watershed Glader Type Debris Cover Type Debris Cover <%) UqA («> ATM (km1) Mean CkwatiM (•M0 Mwfian Elevation (•ad) Elevation (mari) Mutnan Elevation {MtMl 1 Athabasca Detached, Avalanche-fed Terminus 63 1716 0.87 2322 2386 1829 2 (Stutfield) Athabasca Icefall Outlet, Avalanche-fed Terminus 3 Athabasca Detached, Avalanche-fed Terminus 18 7053 20.34 2730 3000 58 2090 2.29 2414 2310 4 (Dome) Athabasca Icefall Outlet, Avalanche-fed Terminus 31 5713 8.07 2557 2353 Jtaat* (•) Slope O Aspect 2997 1168 24 E 1716 3641 1924 22 N 2113 3104 992 19 N 1985 3499 1513 20 NE 5 (Athabasca) Athabasca Icefall Outlet Sides 11 8486 18.45 2661 2705 1941 3630 1689 18 NE > (Little Athabasca) Athabasca Detached Clean 4 2344 2.33 2744 2712 2258 3301 1043 25 N 7 (Boundary) Saskatchewan Detached Terminus 22 2637 1.15 2707 2715 2292 3300 1008 24 NE 8 Saskatchewan Detached Terminus 100 1293 0.24 2426 2409 2358 2526 168 18 E 2062 2796 734 25 NE 9 (Hilda) Saskatchewan Detached Terminus 60 3007 1.38 2369 2328 10 (Saskatchewan) Saskatchewan Outlet Sides 7 12248 38.32 2513 2583 1776 3534 1758 16 E 11 (Castleguard i) Saskatchewan Outlet Clean 0 3077 2.22 2627 2598 2325 3020 695 16 E 12 (Castleguard II) Saskatchewan Detached Clean 5 795 0.65 2624 2609 2482 2809 327 19 SE 13 (Castleguard III) Saskatchewan Outlet Clean 2 1299 1.49 2609 2606 2463 2762 299 12 SE 14 (Castleguard IV) Saskatchewan Outlet Sides 3 9959 16.02 2429 2428 1973 3031 1058 15 SE 15 Columbia Icefall Outlet Clean 1 10686 22.22 2460 2504 1620 3268 1648 18 SW 16 Columbia Outlet Clean 0 5547 9.42 2540 2527 1810 3122 1312 16 SE 17 Columbia Icefall Outlet Terminus 4 3262 6.60 2684 2685 1767 3416 1649 22 SW 18 (Columbia) Athabasca Icefall Outlet Sides 10 5831 32.15 2738 2811 1511 3567 2056 22 W 19 (Manitoba) Athabasca Detached, Avalanche-fed Terminus 25 3483 3.94 2508 2628 1707 3115 1408 24 N 20 Athabasca Detached, Avalanche-fed Terminus 37 958 0.27 2328 2245 2139 2893 755 29 NE 21 Athabasca Detached, Avalanche-fed Terminus 65 2478 2.09 2201 2163 2018 3009 991 13 N 22 Athabasca Icefall Outlet Clean 8 2550 5.37 2904 2997 2232 3604 1372 27 SE 23 Athabasca Detached, Avalanche-fed Terminus 29 2682 1.50 2666 2543 2141 3374 1233 24 W 24 Athabasca Detached, Avalanche-fed Terminus 26 4543 5.94 2738 3047 1776 3685 1910 23 NW 25 Athabasca Icefall Outlet, Avalanche-fed Terminus 10 2978 2.13 2985 3063 2251 3446 1195 31 NW Utm lfJs§f!81Bl 24 4269 8 22 2579 2598 2022 3218 1196 21 tsiilsi Total U 205.45 Table 3.2 Data used for glacier change analysis (AP = aerial photograph). Accumulation Area Contrast Flombeds Covered Snowline (m) 1:62,500 - 25 - 1:20,000 Poor 21 - 1:70,000 Poor 20 2450 1:60,000 Good 22 2490 1:80,000 1:80,000 Excellent Fair 23 13 2690 2650 1:70,000 Fair 25 - 11 1:60,000 Good 23 2560 22 1:50,000 Fair 24 2670 Data Source Year IBCS Map 1918-1920 Federal AP 1948 10-Sep A11655 9 Federal AP 1955 6-Aug 18 Federal AP 1966 22-Aug A14895 A19684 20 A19685 6 Provincial AP Federal AP 1970 19-Aug 1-Sep 15BC5394 A23885 13 7 A25165 17 A25177 15BC86085 5 A27990 1974 ID 10-Sep 21-23 A11711 3 26 19-Sep A11729 23 Federal AP 1979 9-Jul Provincial AP 1986 1993 15-Aug 9-Sep Federal AP #ofImase* Scale/Resolution Date SRTM 2000 Feb N52W117 1 90 m - 23 - Landsat 5 TM 2001 14-Sep 44/24 1 30 m - 25 2760 SPOT 5 2009 20-Aug ; 1 2.5 m Good 25 2890 30-Aug 1 3.3 Methods 3.3.1 Data Sources and Preparation 3.3.1.1 Interprovincial Boundary Commission Survey Maps I used digital copies of the IBCS maps obtained from Library and Archives Canada (LAC). Terrestrial photographs of the Columbia Icefield, taken during the survey of the Alberta-BC border, primarily in 1919, were used to create the maps through phototopographic methods (Interprovincial Boundary Commission, 1917). The icefield spans three map sheets (#21-23) produced in 1918-1920 (Table 3.2). The maps have a scale of 1:62,500 and a contour interval of 100 ft. I rectified the maps in PCI Geomatica OrthoEngine v.10.2 using a 5th order polynomial transformation and 34-36 ground control points (GCPs) on mountain ridges and small lakes from Landsat imagery and Terrain Resource Information Management (TRIM) data. The average root mean square error (RMSE) in the easting (x) and northing (y) of the three map sheets are, respectively, 17.5 and 13.5 m. Total RMSE is 22.1 m (individual RMSE values are listed in Appendix A). 3.3.1.2 Aerial Photographs I included aerial photographs from 1948,1955,1966, 1970,1974, 1979,1986, and 1993 to supplement the geodetic data contained in the interprovincial maps (Table 3.2). Digital copies of diapositives or negatives of the photographs photogrammetrically scanned at a resolution of 12-14 |xm were obtained from the Canadian National Air Photo Library and the British Columbia Ministry of Forests, Lands and Natural Resource Operations. The photographic scales range from 1:24,000 to 1:80,000. Contrast in the snow-covered accumulation areas was poor for the 1948 and 1955 photographs, but satisfactory for other years. 42 Most of the photographs were taken near the end of the ablation season, with dates ranging from 6 August to 18 September, except the aerial photographs acquired on 9 July 1979. Those photographs have late-lying snow which can obscure glacier extents. The photographic coverage is complete for most years, with only a few of the smaller glaciers of the Columbia Icefield missing in some years. Detailed data for the photographs are listed in Appendix A. The 1986 aerial photographs are TRIM aerial diapositives with triangulation points (PUG points) projected in NAD83 UTM zone 11. I used 61 PUG points as ground control points (GCPs) to rectify these photos. I built stereo models using the VR Mapping photogrammetry software suite from Cardinal Systems. Pass points were collected between photographs to tie the images together. I ran a block adjustment to rectify the photographs and produce an exterior orientation file to create my reference stereo models. I collected 2236 GCPs on stable bedrock or stationary boulders, free of vegetation and snow, from the 1986 stereo models to rectify the other seven years of photography (Schiefer & Gilbert, 2007; Barrand et al., 2009). The GCPs were distributed over a range of elevations and across the stereo imagery. Where possible, the same set of GCPs was used to enhance coregistration of the stereo models and minimize error on slopes where a small horizontal offset could result in a large elevation difference (Kaab & Vollmer, 2000; Schiefer & Gilbert, 2007; Schiefer et al., 2007). However, due to differences in photographic coverage, not all GCPs were used for every year of photography. Pass points were collected, bundle adjustments run, exterior orientations were created, and stereo models built for the remaining years of photography. The average RMSE values for the rectified aerial photographs in the easting (x), northing (y), and elevation (z) were 0.51, 0.54, and 0.10 m, respectively. The average 43 RMSE for the stereo models was 0.32 m. The RMSE from the rectification of each year of photography and the resulting stereo models are listed in Appendix A. 3.3.1.3 Satellite Imagery For glacier extents around the year 2000,1 used a rectified Landsat 5 Thematic Mapper (TM) image with a resolution of 30 m. The image was captured on 14 September 2001 (Table 3.2) and obtained from the BC Ministry of Forests. I also used the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) from February 2000, obtained from the USGS website (http://dds.cr.usgs.gov/srtm/). I reprojected the DEM to NAD83 UTM zone 11 and resampled it to a resolution of 100 m. As radar can penetrate winter snow cover over glaciers, the DEM is representative of the ablation surface at the end of the 1999 ablation season (Rignot et al., 2001; Berthier et al., 2006; VanLooy et al., 2006; Schiefer et al., 2007). Some areas of the icefield, primarily steep terrain, have no data due to radar layover and shadowing (Granshaw & Fountain, 2006; Berthier et al., 2007; Farr et al., 2007). The terminal lake of Columbia Glacier also has no data. The most recent imagery covering the Columbia Icefield available for this study were two SPOT 5 (Satellite Pour l'Observation de la Terre) images with a resolution of 2.5 m, acquired on 20 August and 30 August 2009. These images form a stereo pair from which elevation data were extracted (Table 3.2). The images have good contrast in accumulation areas and no snow cover on the glaciers. Using PCI Geomatica OrthoEngine v.10.3,1 rectified the SPOT 5 imagery using the same set of GCPs and method used to rectify the aerial photographs. Tie points (same as pass points) were collected between the images and a least squares bundle adjustment was run. The RMSE in the easting (x) and northing (y) were, respectively, 1.1 and 1.3 m. 44 I created epipolar pairs from the two SPOT 5 images. A 20 m DEM was automatically extracted from the epipolar pairs using a correlation-based image-matching algorithm. This algorithm identifies similar patterns of pixels between images and determines the elevation based on parallax (Berthier et al., 2007; Schiefer & Gilbert, 2007). The model failed in areas of low contrast, primarily those with snow cover and shadows (Kaab & Vollmer, 2000; Berthier et al., 2007; Schiefer et al., 2007). A correlation coefficient image was created along with the DEM, containing values ranging from 0 to 100 (perfect correlation). I used this image to remove pixels with a score less than 70, a threshold delimiting areas of poor contrast (Berthier et al., 2010; Tennant et al., 2012). A 2.5 m resolution orthoimage was also created for 2009 using the rectified SPOT 5 images and DEM. 3.3.2 Data Collection 3.3.2.1 Planimetric Data Using ArcGIS v.9.3,1 digitized glacier extents from the rectified 1919 IBCS maps, the 2001 Landsat 5 TM image, and the 30 August 2009 SPOT 5 orthoimage. I reprojected the extents from the IBCS maps and the Landsat 5 TM image from BC Albers to NAD83 UTM zone 11 to match the other data sets. I digitized glacier extents from the stereo models of the aerial photographs using the VR Mapping photogrammetry suite. The snow line and debris cover were also digitized where present. There were two main problems with the 1919 extents: some glaciers along the east side of the icefield were cut off by the edge of the map and some of the extents appeared inconsistent with subsequent extents (Figure 3.3 A-B). I used the 1919 oblique terrestrial 45 photographs in conjunction with the 1948 and 1955 aerial photographs to complete the missing extents and modify the shapes of the inconsistent glaciers. Figure 3.3 Problems encountered in mapping glacier extents: (A) glacier cut off by the edge of the IBCS map sheet; (B) inconsistent 1919 glacier extents; (C) snow cover from late lying snow in the 1979 aerial photographs; and (D) debris on glaciers. Late-lying snow in the 1979 photographs and supraglacial debris in all imagery hindered delineation of some glaciers (Fox & Nuttall, 1997; Hoffman et al., 2007) (Figure 3.3 C-D). For debris-covered ice, I used elevation changes between the stereo models from the aerial photographs to help determine the extent of glacier ice. The glaciers were divided into 25 flowsheds (FS) modified from the glacier drainage basins in the western Canada glacier inventory by Bolch et al. (2010) (Figure 3.1). I also digitized glacier flow lengths for the main glaciers, but not tributary glaciers, in each 46 flowshed. Lengths were measured from the highest point along the edge of the flowshed down to the center of the glacier terminus, perpendicular to the contours. 3.3.2.2 Elevation Data I digitized contours from the IBCS maps, reprojected them, and created a DEM with a resolution of 100 m. There were no contours for the termini of Hilda (FS9) and Saskatchewan (FS10) glaciers, thus only estimates of elevation change could be made for these glaciers. I manually digitized elevation points from the stereo models using the VR Mapping photogrammetry suite. Elevation points were collected on a 100 m grid in the same location for all years. A 200 m grid was used in the accumulation area and on snow cover where there is low contrast and elevations are more difficult to determine. I extracted elevations from the IBCS, SRTM, and SPOT 5 DEMs at the same points as on the aerial photographs. The elevation points were also divided into the flowsheds for analysis. 3.3.3 Glacier Change Analysis 1 determined changes in length, area, and elevation by differencing sequential data for each glacier flowshed over ten periods: 1919-1948, 1948-1955,1955-1966,1966-1970, 1970-1974, 1974-1979, 1979-1986, 1986-1993,1993-2000, and 2000-2009. Where data are missing due to lack of coverage, I differenced data from one year with the next year of available data, creating a new period (e.g. 1955-1970). In order to create a comparable dataset for each flowshed, I calculated the rate of glacier change over the new period (e.g. 1955-1970) and used it to estimate the change over the two encompassed periods (e.g. 19551966 and 1966-1970; see Appendix C for raw data and calculations). I calculated elevation change for each elevation point. Elevation changes greater than three standard deviations were considered errors and removed from analysis. The elevation 47 changes were weighted based on the 100 m or 200 m grid, as the data of the 200 m grid represent a larger area than the data of the 100 m grid. The changes were converted to water equivalent (w.e.) values assuming a density of 550 kg/m for points in the accumulation area and 900 kg/m3 for points in the ablation area (Berthier et al., 2007; Schiefer et al., 2007). The density-depth relation is assumed constant through time based on Sorge's Law (Bader, 1954; Andreassen et al., 2002; Cox & March, 2004). The elevation separating the accumulation and ablation areas was determined from the mean snowline elevations of the years constituting the period. The glacier-wide (mean) elevation changes, adjusted for density (w.e.) thus represent the geodetic balance of the glaciers. I calculated mean length and elevation change of the entire Columbia Icefield as the mean value of all the flowsheds for each period. I summed the area change in all of the flowsheds for each period to determine the total change in the icefield. I calculated the volume change for each flowshed by multiplying the mean elevation change by the largest area of the flowshed over the period to account for glacier advance. Total volume change for the Columbia Icefield was determined by summing volume change from all of the flowsheds for each period. 3.3.4 Error Analysis Error estimates for each year of data are summarized in Table 3.3. Error in length measurements was based on the combined mean horizontal RMSE and one-half of the resolution of each year of data. I added an additional error term for offset between the IBCS maps (24 m) and snow cover in the 1979 photographs (5 m). I calculated error in glacier extents as the area of buffers surrounding each glacier, with a distance equal to the length error. This method is similar to the one proposed by Granshaw & Fountain (2006) and Bolch 48 et al. (2010). Errors in length and area change ( E A ) were calculated by combining the error terms from the early year (E\) and the later year (£2) of a period using Eq. 3.1: E& = ^ (E\2 + Ei + ... + En) 3 Table 3.3 Error estimates used in the glacier analysis. Year Length Error (m) Area Error (km1) Absolute Elevation Error (ra w.e.) 1919 34.0 12.29 36.5 1948 0.9 0.48 1.1 1955 0.8 0.43 2.4 1966 0.7 0.40 1.2 1970 0.8 0.45 2.1 1974 0.7 0.47 1.0 1979 5.1 3.06 0.9 1986 1.5 0.87 reference 1993 1999/2001 0.7 15.0 0.43 7.66 1.1 5.9 2009 2.1 1.30 6.4 I determined systematic and random elevation errors by analyzing elevation change between two years on stable, non-vegetated, non-glaciated areas where no elevation change was expected (Schiefer & Gilbert, 2007). Depending on the image coverage, I collected 1426 check patches of 25 points (5x5 points on a 10 m grid) (Figure 3.1). Some patches had high elevation changes, but were located on steep slopes where small horizontal errors can result in large vertical errors (Cox & March, 2004; Schiefer & Gilbert, 2007). I looked for systematic errors (bias) in elevation, slope, aspect, latitude, and longitude, arising from rectification inaccuracies that had to be corrected for valid results (Schiefer et al., 2007). I created linear models describing these biases and removed them from the elevation data where present (Figure 3.4). The standard deviation (± la) of the check patches represented 49 the random error in the elevation data (VanLooy et al., 2006). I used Eq. 3.1 to combine the systematic (Ft) and random errors (Ei) for a total error term. •"T 1 1 1 1 T 1 1 1 1 0 10 20 30 40 0 10 20 30 40 Slope (•) Slope (*) Figure 3.4 (A) Systematic bias with slope in the 1966 elevation data. I fit a linear model (dashed line) to the data to remove the bias (B). Absolute error (Table 3.3), or elevation accuracy with the datum (NAD83), was determined by differencing elevation data with the 1986 reference data (Figure 3.5 A). The elevation data were not adjusted to account for different acquisition times of the imagery or emergence, because no mass balance data were collected on the Columbia Icefield to compare with or model adjustments, and the values are typically negligible for glacier-wide measurements (Andreassen et al., 2002; Cox & March, 2004; Berthier et al., 2010). Ablation adjustments from Gulkana Glacier, Alaska (Cox & March, 2004), and Mer de Glace, French Alps (Berthier et al., 2004), are no greater than -0.5 m, which falls within my calculated error for the elevation data. For relative error (elevation change error), I applied the same methods, but between two subsequent years rather than using the 1986 reference data (Figure 3.5 B). I included a 0.5 m w.e. error term (Ej) to account for changes in the density profile 50 (Andreassen et ah, 2002; Cox & March, 2004) and a 5 m w.e. error term (£4) for the accumulation area elevation points to represent the uncertainty due to low contrast (VanLooy & Forster, 2008). I calculated volume change error using standard error propagation methods (Eq. 3.2): Ew = V (Arf Ea2 + A2 W) - (Eq. 3.2) where E&y is the volume change error, AH is mean elevation change, E&h is the elevation change error, A is the area of extent used to calculate volume change, and EA is the area error. 3.3.5 Climate Data I extracted annual, ablation (May to September), and accumulation (October to April) season (Letreguilly, 1988) minimum, mean, and maximum temperatures, as well as precipitation from 1919 to 2009 from the ClimateWNA v.4.62 program (http://www.genetics.forestry.ubc.ca/cfcg/ClimateWNA/Climate WNA.html). This program extracts climate data at specific locations from anomaly surfaces derived from interpolated historical data created by Mitchell & Jones (2005) and a baseline reference grid (2.5 arc min) of monthly climate data for the 1961-1990 normal period generated by PRISM (Daly et al., 2002). Bilinear interpolation and lapse-rate elevation adjustments are used to integrate the historical anomaly surfaces and the baseline grid and downscale the climate data at a specific location (Mbogga et al., 2009; Wang et al., 2012). More detailed information on the methodology behind the datasets can be found in Wang et al. (2012). 51 Data source IBCS Map AP 1919 1948 1955 1966 AP AP 1970 1974 AP SRTM i_ SPOT L. 1979 1993 1999 2009 19791986 19861993 19931999 19992009 Year 19191948 19481955 19551966 19661970 19701974 19741979 Period Figure 3.5 (A) Absolute elevation differences from check patches between an individual year and the 1986 reference data. (B) Relative elevation differences from check patches between two sequentially differenced datasets. Boxes represent the first and third quartiles with the horizontal black line as the median. The whiskers represent the data extremes (5th and 95th percentile) and the circles are outliers. 52 I extracted data on a 1 km grid encompassing the Columbia Icefield. I derived temperature and precipitation anomalies from the 1919-2009 mean and grouped them into ten periods: 1919-1948, 1948-1955,1955-1966,1966-1970, 1970-1974,1974-1979,19791986, 1986-1993, 1993-2000, and 2000-2009. I correlated the anomalies with rates of length, area, elevation, and volume change over the ten periods, and lagged the start and end years of the anomaly periods up to 19 years to determine a possible delayed response between glacier change and climate. 3.4 Results 3.4.1 Glacier Change Changes in glacier length, area, elevation, and volume for the period 1919-2009 indicate substantial recession and mass loss (Table 3.4). All glaciers retreated between 1919 and 2009 (Figure 3.6). The glaciers retreated a mean distance of 1149.9 ± 34.1 m (25.0 ± 1.0%) at a rate of -12.8 ± 0.4 m a"1 (-0.28 ± 0.01 % a"1) along their flow lines. Columbia Glacier (FS18) experienced the largest absolute retreat - 3723.4 ± 34.1 m - but it terminates in a lake. Castleguard II (FS12) had the greatest relative length loss - 69.4 ± 1.3%. FS17 lost the least absolute (450.4 ± 34.1 m) and relative (12.1 ± 0.9%) length. In 1919, the area of the Columbia Icefield was 265.05 ± 12.29 km2. By 2009, the icefield lost 59.60 ± 1.19 km2 (22.5 ± 5.0%) at a rate of-0.66 ± 0.01 km2 a"1 (-0.25 ± 0.06% a'1). Mean glacier area change was -2.38 ± 0.24 km2 (-33.5 ± 6.1%) with a mean rate of 0.026 ± 0.003 km2 a"1 (-0.37 ± 0.07 % a'1). Saskatchewan Glacier (FS10) experienced the greatest absolute area loss - 10.09 ± 0.57 km2 - and Castleguard II (FS12) lost the greatest 53 relative area - 73.9 ± 7.5%. The least absolute and relative area losses were at FS8 (0.29 ± 0.08 km2) and Stutfield Glacier (FS2; 10.3 ± 0.6%). Table 3.4 Glacier changes for flowsheds (FS) of the Columbia Icefield. Mean and total values represent icefield-wide changes. A Length (m) A Length A Am dan*) A Area <%) A Elevation (»*•«•) A Volume {km3 w.e.) 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 -588.8 ±34.1 -1086.3 ±34.1 -970.8 ±34.1 -918.4 ±34.1 -1319.0 ±34.1 -1006.1 ±34.1 -539.7 ± 34.1 -637.8 ±34.1 -858.3 ± 34.1 -2321.8 ±34.1 -1118.5 ±34.1 -1802.0 ±34.1 -1413.5 ±34.1 -2090.8 ±34.1 -1574.6 ±34.1 -918.7 ±34.1 -450.4 ±34.1 -3723.4 ±34.1 -1111.6 ± 34.1 -703.5 ±34.1 -517.0 ±34.1 -835.9 ±34.1 -816.1 ±34.1 -731.5 ±34.1 -69L8±34J -25.5 ±1.5 -13.3 ± 0.4 -31.7 ± 1.1 -13.8 ±0.5 -13.5 ±0.3 -30.0 ±1.0 -17.0 ±1.1 -33.0 ± 1.8 -22.2 ±0.9 -15.9 ±0.2 -26.7 ±0.8 -69.4 ± 1.3 -52.1 ± 1.3 -17.4 ±0.3 -12.8 ±0.3 -14.2 ±0.5 -12.1 ±0.9 -39.0 ± 0.4 -24.2 ±0.7 -42.3 ± 2.0 -17.3 ±1.1 -24.7 ± 1.0 -23.3 ±1.0 -13.9 ±0.6 -18.9 ±0.9 -0.64 ±0.12 -2.34 ±0.15 -1.73 ±0.25 -2.16 ±0.23 -2.97 ±0.21 -0.96 ± 0.09 -0.64 ±0.10 -0.29 ± 0.08 -1.13 ±0.16 -10.09 ±0.57 -0.88 ± 0.09 -1.84 ±0.19 -1.66 ±0.14 -5.68 ±0.32 -4.84 ±0.26 -2.65 ±0.16 -2.64 ±0.20 -6.85 ± 0.44 -2.03 ± 0.29 -0.46 ±0.12 -1.19 ±0.23 -1.35 ±0.18 -1.08 ±0.19 -2.64 ±0.32 -085^15 -42.3 ± 8.0 -10.3 ±0.6 -43.0 ±6.2 -21.1 ±2.2 -13.9 ±1.0 -29.2 ± 2.8 -35.9 ±5.4 -55.3 ± 15.2 -45.1 ±6.4 -20.8 ± 1.2 -28.3 ± 2.9 -73.9 ±7.5 -52.7 ±4.4 -26.2 ± 1.5 -17.9 ± 1.0 -22.0 ± 1.4 -28.6 ± 2.2 -17.6 ± 1.1 -34.0 ±4.9 -62.7 ± 16.0 -36.3 ± 7.0 -20.1 ± 2.7 -41.9 ±7.5 -30.8 ± 3.7 -28£±5Jl -36.6 ±28.7 -37.1 ±22.7 -47.7 ±27.1 -45.4 ±26.2 -54.8 ± 54.8 -51.6 ±22.1 -51.9 ±22.9 -39.0 ± 28.5 -79.3 ± 27.8 -61.6 ±24.6 -28.7 ±25.8 -44.2 ± 25.5 -37.9 ±25.1 -48.8 ± 27.7 -42.8 ± 25.9 -33.3 ± 25.3 -84.0 ±23.1 -58.5 ± 23.4 -43.8 ± 23.8 -39.3 ± 28.4 -56.9 ±28.7 -56.0 ±21.0 -11.8 ±24.3 -92.3 ± 24.4 -51.8 ±21.2 -0.05 ±0.04 -0.83 ± 0.52 -0.18 ±0.11 -0.44 ±0.27 -1.97 ±0.52 -0.17 ±0.08 -0.09 ±0.04 -0.02 ± 0.02 -0.17 ±0.07 -2.79 ± 1.21 -0.09 ±0.08 -0.10 ±0.06 -0.11 ±0.08 -1.01 ±0.60 -1.12 ± 0.71 -0.39 ±0.31 -0.76 ±0.22 -2.21 ± 0.92 -0.26 ±0.15 -0.03 ± 0.02 -0.19±0.10 -0.37 ±0.15 -0.03 ±0.06 -0.78 ± 0.23 ^M5±0X)7 Mil -3723.4 ±34.1 -69.4 ±1.3 -10j©9±0J7 •73.9 ±7.5 -92.3 ±24.4 -2.79 ±121 " -59.60 ±1.19 • -14.30 ± 2.02 FS 1 2 3 4 5 6 7 Total Glaciers of the Columbia Icefield thinned, on average, 49.4 ± 25.2 m w.e. from 1919 to 2009, at a rate of -0.55 ± 0.28 m w.e. a*1. FS24 experienced the most thinning (92.3 ± 24.4 m w.e.), and FS23 thinned the least (11.8 ± 24.3 m w.e.). Spatially, thinning was greatest at lower elevations, near the center flowline, and close to the 2009 terminus position (Figure 3.6). Thickening near some of the icefalls is likely due to positional uncertainties that 54 amplify elevation change (Koch et al., 2009). Accumulation areas had few usable elevation points over the period 1919-2009. This is not a major issue, however, because I calculated elevation change between 1919 and 2009 by summing mean elevation change through each period, for which there are more reliable elevation points (Appendix D). Total volume change of the Columbia Icefield was -14.30 ± 2.02 km w.e. at a rate of -0.16 ± 0.02 km3 w.e. a"1. Mean volume change of individual glaciers was -0.57 ± 0.40 km3 w.e., and the mean rate of ice loss was -0.0064 ± 0.0045 km3 w.e. a"1. Saskatchewan Glacier (FS10) lost the greatest volume (2.79 ± 1.21 km3 w.e.), whereas FS8 lost the least volume (0.02 ± 0.02 km3 w.e.). 1919 to 2009 I 11919 Exttnt I 1 iOOfl Extent Elevation Chang* (m «u.| •• -200 to -160 H -150 to -100 I I -100 to-40 I I-SO too I 0 to 50 Figure 3.6 Area and elevation change of the Columbia Icefield, 1919 to 2009. 55 Period 19191948 19481955 19551966 19661970 19701974 19741979 19791986 19861993 19932000 20002009 19191948 19481955 19551966 19661970 19701974 19741979 19791986 19861993 19932000 20002009 Period Figure 3.7 Rates of (A) length, (B) area, (C) elevation, and (D) volume change over each period from 1919 to 2009. 56 Attttfenca »4FS« $Mfc*dw«an SMfcttchawn CohmtM •effS« 14 CoAirtota 14 <01 iCi n e? I ? u S R Athabnu #dFS- Saskatdwwsn Saskatchfwan CoMnba 14 Figure 3.8 (A-B) Absolute and relative length, (C-D) absolute and relative area, (E) elevation, and (F) volume change of glaciers by watershed. 57 3.4.2 Glacier Change over Time Some of the highest rates of glacier retreat and shrinkage, and the highest rates of elevation and volume loss, occurred in the earliest period, 1919-1948 (Figure 3.7). After 1948, ice loss decreased to minimum values during the periods 1970-1974 and 1974-1979. Rates of glacier change in these two periods are near zero, as ice loss decreased and some glaciers advanced, increased in area, thickened, and gained volume. From 1979 to 1993, rates of elevation and volume loss again increased. The rates of retreat and area loss also increased, but only for the period 1979-1986, before they decreased again from 1986 to 2000. The highest rates of retreat and area loss occurred during the period 2000-2009; rates of thinning and volume loss, however, began to decrease around 1993. 3.4.3 Glacier Change by Type I compared glacier changes over the period 1919-2009 between watersheds, size classes (< 1, 1-5, 5-15, 15-30, and 30-50 km2), glacier types, and type of debris cover as listed in Table 3.1. Glaciers that flow into the Columbia River watershed lost the least relative length and area, but the most absolute area and volume (Figure 3.8). The glaciers of the Saskatchewan River watershed display the most variability and the greatest absolute retreat and relative area loss. Elevation changes are similar for all three watersheds. Columbia River watershed glaciers show an increase in length from 1955 to 1979, as well as increases in area, elevation, and volume during the 1970s (Figure 3.9), while glaciers in the other watersheds mainly experienced decreases in ice loss. 58 Wfill m m m »* m mm** m m m mm 7 m mim m * m m m a mmmmm 1049 Extant i 11070 Extant DwKow Chans* fm WJ.) >30 to-20 20 to-10 I ]-10to0 OtolO 10 to 20 Figure 3.9 Changes in length, area, and elevation on glaciers of the Columbia watershed, 1970-1979. Absolute retreat, area, and volume loss are greatest for large glaciers (30-50 km2). Relative retreat and area loss have the opposite trend and are greatest for small glaciers (< 1 km2). However, the largest glaciers do not show the least relative retreat and area loss (Figure 3.10). There is a small increase in thinning over time in the case of larger glaciers, but there is high variability in the mid-sized glaciers (5-30 km2) that overwhelmed most changes between size classes. I classified glacier types into five groups: outlet glaciers, outlet glaciers with icefalls, avalanche-fed outlet glaciers with icefalls, detached glaciers, and avalanche-fed detached glaciers. All three groups of outlet glaciers have greater losses in absolute length, area, and volume, but lower losses in relative length and area (Figure 3.11) than the two detached glacier groups. Elevation change is similar among the different glacier types, except for 59 icefall outlet glaciers, which have the greatest elevation loss, and outlet glaciers without icefalls, which have the lowest elevation loss. The icefall outlet glaciers that are avalanchefed lost the least absolute length and area among the outlet glacier groups, and nonavalanche-fed icefall outlet glaciers have the highest volume losses. Detached glaciers that are avalanche-fed lost more absolute area, but less elevation, than non-avalanche-fed detached glaciers, resulting in similar values of volume change. Glaciers with debris-covered sides have the greatest absolute length, area, elevation, and volume losses, but the lowest relative length and area losses (Figure 3.12). Glaciers with debris-covered termini have lower absolute length and area losses, but greater elevation loss compared to glaciers with no debris cover (bare ice). 60 #<*r$» 2 11 t-5 #«fFS» 2 #«fFS* (km) * of FS • 2 2 <1 15-30 2 Area i km ? # ' I I' » D«tri*-covercd fiaraka i i—r C* Etevatfoo (m asl) Elevation (m asl) JO) Docna (FS4J 0<9 °Q i S « o J £ i i i i 1 T5 t S _ a * DC •—» Dtbrtxcrvered —— taraice » I 1 I I 1 I I I I I I 'I I 1 I I 1 I 8 8 8 8 8 8 8 8 8 8 * - Elevation (m asl) i i § M i ~!—r i g W ft I OeWs-ccvtred 8aralca § «*> Elevation (m asi) (I) FS24 T $ O £ f n * I81 8i*f i 5 j j t"! • j I • i ;• • IT B ffl 0 i< I t I i 1 11* *« I * i a» 9 Oebffet-covwBd Baratea 1 I 1 I I I I I » I I I I I I I I 1 Bevafion (m asl) OIMkomckI Baraks Elevation (m asl) Figure 3.13 Elevation change rates of debris-covered and bare ice for glaciers with (A-B) debriscovered sides and (C-F) debris-covered termini. 66 Annual Ablation (MJJAS) Accumulation (ONDJFMA) & 8. % Ui Annual Ablation (MJJAS) Accumulation (ONDJFMA) Figure 3.14 Annual, ablation, and accumulation season temperature and precipitation anomalies at the Columbia Icefield for each period between 1919 and 2009. The climatic mean is based on the period 1919-2009. 67 Ablation season precipitation anomalies across the periods of study are near the 19192009 mean, except for the periods 1919-1948 and 2000-2009, which have negative anomalies (Figure 3.14 D). Accumulation season precipitation anomalies show greater deviation from the mean. Precipitation anomalies increase to positive anomalies up to 1974, and then decrease significantly during the period 1974-1979. Positive precipitation anomalies increase again from 1986 to 2000, before decreasing to negative values in the period 2000-2009. Annual precipitation anomalies are similar to accumulation season anomalies, with a more pronounced decrease in the most recent period. I correlated rates of length, area, elevation, and volume change with the temperature and precipitation anomalies lagged zero to 19 years (Appendix B). Mean annual and accumulation season temperature anomalies lagged two to five years have the largest significant correlations (p < 0.05) with rates of length and area change. The only significant ablation season temperature anomaly correlation (r = -0.661, p < 0.05) is for the rate of length change at a lag of two years. Annual and accumulation season precipitation anomalies lagged two to eight years with length change rates and five to seven years with area change rates have the largest significant correlations (p < 0.05). Length change rates have larger correlation coefficients and greater significance (p < 0.01) than area change rates. Rates of elevation and volume change correlate significantly (p <0.05) with maximum ablation season temperature anomalies at a lag of 12 to 13 and 16 to 18 years. More significant correlations (p < 0.01) occur with precipitation anomalies. The largest correlations are with precipitation anomalies lagged one to nine years for elevation change rates and with precipitation anomalies lagged two to ten years for volume change rates. 68 3.5 Discussion 3.5.1 Glacier Change Glaciers of the Columbia Icefield retreated a mean distance of 1149.9 ±34.1 m and decreased in area by 2.38 ± 0.24 km2 on average between 1919 and 2009. Over this period, total area loss was 59.60 ± 1.19 km2; combined with a mean elevation change of -49.4 ± 25.2 m w.e., the icefield lost 14.30 ± 2.02 km3 w.e. The greatest thinning occurred at mid-glacier 1919 positions, along the center flowline (Figure 3.6). At the lowest elevations near the 1919 terminus, thinning was limited by the ice available for melting; this ice disappeared as the glaciers retreated (Koch et al., 2009; Berthier et al., 2010; Tennant et al., 2012). There has been apparent thickening in the accumulation areas and at the terminus of some glaciers. Certain areas in the accumulation area may be more favourable for snow accumulation than others, and, at the terminus, sediment and debris may be deposited within the former 1919 glacier extent as the glaciers retreated. However, errors arising from poor contrast, lower point density, and errors in the accumulation area contours make this thickening suspect (Fox & Nuttall, 1997; Andreassen et al., 2002; Berthier et al., 2010). Most instances of thickening are in icefalls. These steep areas are subject to large elevation errors if there is a small horizontal shift between the elevation data. The IBCS maps and imagery were rectified using a different set of GCPs, which may account for horizontal shifts between the data (Andreassen et al., 2002). Previous studies on glaciers of the Columbia Icefield focused primarily on the retreat of the main glaciers - Athabasca, Columbia, and Saskatchewan - and some of the smaller glaciers, such as Dome, Manitoba, Stutfield, and Boundary glaciers, from the mid-19th century to the 1970s. My values of retreat are within 30-40 m of the data published by 69 Henoch (1971), Denton (1975), the Water Survey of Canada in Luckman (1986), Reynolds & Young (1997), and Robinson (1998). The rate of retreat measured on Athabasca Glacier from 1919 to 1948 (Kite & Reid, 1977) and Saskatchewan Glacier from 1953-1963 (Denton, 1975) were 20 m a"1 and 25 m a'1, respectively, which are comparable to my calculated rates of 19.8 ± 1.2 and 28.9 ±0.1 m a"1 over similar periods. There are some large discrepancies (> 100 m) in retreat between my results and data presented by Robinson (1998) and Jones & Rowbotham (2000/01). The differences are probably due to the measurement techniques (field versus images), the location of the measurements (as change is different along the terminus), and the mapping of the terminus from which my measurements were taken. Snow cover, shadows, and especially debris cover hinder the delineation of glacier termini, which will differ among researchers (Fox & Nuttall, 1997; Bolch et al., 2010). Despite the differences in retreat values, the trends are similar, for example the high rates of retreat prior to 1950, followed by reduced rates of retreat or a slight advance. There are fewer published estimates of area loss. Henoch (1971) concluded that glacier area in the North Saskatchewan River watershed decreased 10% from 1948/1951 to 1966. My total area change for this watershed over the same period is only 4%, but the average area loss is 7%. Demuth et al. (2008) calculated an area loss of 22% in the North Saskatchewan River watershed between 1975 and 1998. Again my values are lower, 10%, over a similar period 1974-1993. Contributions from other glaciers in the watershed with greater area change may account for the greater area loss. Bolch et al. (2010) calculated a rate of area loss of 0.67 ± 0.19% a"1 for the entire southern Canadian Rocky Mountains from 1985 to 2005, comparable to my rate of 0.6% a"1 for the Columbia Icefield from 1986 to 2009. 70 Elevation and volume change estimates are only available for the Saskatchewan, Athabasca, and Columbia glaciers. My thinning values tend to be lower, and my volume loss values higher, than published data (Konecny, 1963; Kite & Reid, 1977; Reid & Charbonneau, 1979; Reynolds, 1996; Reynolds & Young, 1997). This difference is probably due to the different area being studied. Previous studies are based on measurements from the ablation area only, whereas I included the accumulation area, which reduces glacier-wide thinning values. Although my thinning values are less than those of previous researchers, they are applied over a larger area, resulting in greater volume loss, especially because the accumulation area is proportionally larger than the ablation area. Reynolds (1996) calculated a total volume loss of 0.23 km3 in the ablation area of Athabasca Glacier (excluding debris cover) between 1919 and 1979. My total volume loss is 0.97 ± 0.45 km3, but dividing by the ^ 1 respective areas, the rates are similar - 0.07 and 0.05 ± 0.03 km a" , respectively. Denton (1975) reports downwasting on Columbia Glacier of -90 to -105 m between 1919 and 1948, similar to thinning values I observed in the ablation area on Columbia Glacier over a similar period. The only long term mass balance data from the Canadian Rocky Mountains are from Peyto Glacier (Demuth et a!., 2009). I compared these data to the geodetic balance of glaciers of the Columbia Icefield. Most of the glaciers show a similar pattern to Peyto Glacier, but only Saskatchewan Glacier has a similar balance (Figure 3.15). The geodetic balances of the other Columbia Icefield glaciers are generally less negative. 71 m CM o rt Peyto - Stuttield -- Athabasca Boundary * Saskatchewan Columbia — Columbia Iceftold 19661970 19701974 19741979 19791986 19861993 19931999 19992009 Period Figure 3.15 Peyto cumulative mass balance 1966 to 2007 (Demuth et al., 2009), compared with cumulative geodetic balance of glaciers of the Columbia Icefield from 1966 to 2009. 3.5.2 Glacier Change by Type Most glacier changes between watersheds are similar, except the glaciers of the Columbia River watershed have higher area and volume losses, probably due to their southfacing aspects. Individual glacier change depends on a variety of factors such as climate, topography, glacier size, type of glacier, source of nourishment, and debris cover (Meek, 1948; Gardner, 1978; Pelto, 2006; Bolch, 2007). Some of these factors complicate a glacier's response to climate. Debris-covered or avaianche-fed glaciers for example, can be disconnected from climatic inputs and will respond differently from clean or outlet glaciers 72 (Granshaw & Fountain, 2006; Hoffman et al., 2007). Large glaciers have higher absolute losses than smaller glaciers. They have more ice available for melt and tend to be directly connected to the icefield. Small glaciers had greater relative length and area changes, probably because they do not have large unchanging accumulation areas and do not exist at as high elevations as some of the larger glaciers (Granshaw & Fountain, 2006; Bolch, 2007; Bolch et al., 2010). However, the largest glaciers do not have the lowest relative losses, suggesting that they are retreating and shrinking faster than some of the glaciers a class or two smaller. Outlet glaciers of the Columbia Icefield show greater ice losses than the detached glaciers. An important component is glacier size; the outlet glaciers tend to be the largest glaciers with the greatest absolute ice losses. The detached glaciers are generally smaller, more heavily debris-covered, and are nourished through avalanching or possibly wind-blown snow rather than directly from the icefield. Within the outlet glacier group, icefall outlet glaciers lost more ice than the non-icefall outlet glaciers. This difference may be caused by differences in glacier dynamics related to icefalls. Luckman (1986) noted a slower and smaller response on Dome Glacier than on neighboring Athabasca Glacier, which he attributed to differences in the ice source and debris cover. Glaciers with debris-covered sides experienced the greatest ice loss, but they are also the largest glaciers and are outlet glaciers, which have experienced the greatest ice losses. Glaciers with debris-covered termini lost less length and area, but more elevation than clean ice glaciers. This behaviour has been reported for other glaciers in the Canadian Rocky Mountains (Gardner, 1978) and Tasman Glacier in New Zealand (Kirkbride & Warren, 1999). In their study on Stutfield Glacier, Osborn et al. (2001) observed that, as the ice 73 beneath the debris cover stagnates, terminus retreat slows. Meanwhile, supraglacial ponds and streams expose ice and enhance melt, increasing thinning. 3.5.2.1 Debris Cover Percent debris cover increased slightly for glaciers with debris-covered termini from 1948 to 1993, but debris cover decreased for all debris-covered glaciers between 1993 and 2009. I had expected an increase in debris cover through a redistribution of debris from higher areas of thick debris (i.e. medial moraines) to lower areas of bare ice that rapidly melt (Kirkbride & Warren, 1999). However, I did not observe this redistribution of debris. Glacier flow apparently kept the bare ice clean by supplying bare ice from up-glacier locations, and retreat of the terminus, where the majority of the debris cover exists, removed debris from the glacier. Because the up-glacier extent of the debris cover changed little and the debris-covered terminus retreated, the debris-covered area decreased, reducing the percentage of debris cover. Thinning rates under debris cover differ according to the thickness of debris. Studies by Mattson et al. (1993), Nakawo & Rana (1999), Mattson (2000), and Brock et al. (2010) found that debris cover thinner than ca. 0.01 m enhances ablation by absorbing more shortwave radiation, due to its lower albedo, and transferring it to the ice below. Once debris cover reaches a critical thickness, however, it acts as an insulator, reducing ablation. On the Columbia Icefield, thinning rates were reduced 30 to 60% over the period 1949-2009. A 50% reduction in thinning rates was observed on Bering Glacier, Alaska, by Berthier et al. (2010) over the period 1962-2006, and a 40% reduction was observed on Khumbu Glacier, Himalayas, by Takeuchi et al. (2000) during the 1999 ablation season. Shortwave radiation is absorbed by the debris, but the thickness of the debris prevents the transfer of heat to the 74 ice below. The critical thickness ranges between 0.01 and 0.09 m and depends on the lithology, condition (wet or dry), continuity of the debris cover, and meteorological conditions. I do not have measurements of debris thickness, but the observed thinning rates provide insights into the possible thickness and distribution of the debris cover (Figure 3.13). Glaciers with debris-covered sides show the greatest difference in thinning rates between debris-covered and bare ice. The debris cover appears continuous in aerial photographs, and the significantly reduced melt implies a thick layer of debris. However, at higher elevations the rates between debris-covered and bare ice are similar and in some cases (e.g. Athabasca Glacier, FS5) thinning under debris cover has been greater than on bare ice. Brock et al. (2010) noticed a similar enhancement of thinning under debris cover compared to bare ice near the upper limit of the debris sheet (2400 m asl) on Miage Glacier, Italy, and attributed it to thin debris. The difference in rates is not as large on glaciers with debris-covered termini. From the aerial photographs, I could see that the coverage of debris differed considerably and supraglacial streams and ponds were present on some of the glaciers. The water can expose ice (Luckman, 1986), enhancing ablation especially on exposed ice cliffs where Nakawo & Rana (1999) measured ablation rates up to 12 times greater than on the surrounding debriscovered ice. Mass wasting and horizontal retreat of these cliffs can cause large elevation changes over time. Also, variable debris cover leads to differential ablation, creating areas of reduced or enhanced melt and possibly exposing new ice surfaces (Luckman, 1986). In calculating elevation changes over the variable debris cover, I would be combining the 75 reduced and enhanced thinning rates. On a few glaciers (e.g. FS20), enhanced melt is prominent and rates of thinning are greater under debris cover than on bare ice. 3.5.3 Climate Some of the highest rates of length, area, elevation and volume loss for glaciers of the Columbia Icefield occurred during the earliest period, 1919-1948 (Figure 3.7). Rapid retreat of Saskatchewan, Columbia, Athabasca, Dome, and some of their tributary glaciers have been documented by Field (1948) and Osborn & Luckman (1988) from 1920 to 1950. Glaciers also rapidly retreated in the Premier Range in the Columbia River watershed (Luckman et al., 1987). These changes reflect a period of warm, dry conditions after the end of the Little Ice Age, when glaciers reached their maximum extents between mid-18th and 19th centuries (Osborn et al., 2001). These conditions continued into the 1930s and 1940s (Luckman & Kavanagh, 2000), with high maximum temperature anomalies and low accumulation season precipitation anomalies. From 1948 to the 1970s, rates of ice loss decreased and some glaciers advanced and thickened. Slowed retreat, standstills, and some minor advances of the termini of glaciers of the Columbia Icefield and throughout the Canadian Rocky Mountains have been noted by other researchers (Henoch, 1971; Luckman, 1986; Luckman et al., 1987; Osborn & Luckman, 1988; Robinson, 1998; Luckman & Kavanagh, 2000; Jones & Rowbotham, 2000/01; Demuth et al., 2008; Moore et al., 2009). Robinson (1998) documented the advance of Kitchener and Columbia glaciers in the 1960s and 1970s, and Jones & Rowbotham (2000/01) noted a standstill of Boundary Glacier from 1966 to 1970. Luckman et al. (1987) observed an advance of most glaciers in the Premier Range, which mirrors behaviour of glaciers in the Columbia Icefield in the same watershed. The glaciers of the 76 Columbia Icefield not only advanced, but increased in area, thickened, and gained volume during the 1970s (Figure 3.9). Increased precipitation in the accumulation season, peaking in the period 1970-1974 and a drop in maximum and mean temperatures drove these changes (Henoch, 1971; Luckman et al., 1987; Luckman & Kavanagh, 2000). Glacier change rates are variable between 1979 and 2000, but the overall trend is one of ice loss. Elevation and volume loss increased until 1993 before starting to decrease, whereas length and area losses increased up to 1986 before decreasing. Luckman & Kavanagh (2000) and Demuth et al. (2008) also noted renewed recession, increased melt rates, reduced flow of mass from accumulation areas, and increased ice loss of glaciers in the Canadian Rocky Mountains. These observations correspond with my observed decrease in positive precipitation anomalies during the period 1974-1979, followed by an increase in 1986-1993. Temperature anomalies peaked during the period 1986-1993, which included some of the warmest years on record (Luckman & Kavanagh, 2000), after which accumulation season temperatures decreased. The highest retreat and shrinkage rates occurred during the period 2000-2009, and for some glaciers the rates were greater than during the period 1919-1948. Bolch et al. (2010) reported an increase in the rate of area loss from 2000 to 2005. Precipitation anomalies decreased to negative values in both accumulation and ablation seasons after 2000, but temperature anomalies decreased, especially in the accumulation season. Length and area could still be responding to the warmer temperatures of the previous periods (Demuth et al., 2008; Koch et al., 2009). The rates of elevation and volume loss decreased between 2000 and 2009. These rates are significantly lower than the rates between 1919 and 1948, even though temperature and precipitation anomalies are similar between the two periods. 77 However, elevation and volume changes may be responding to the positive precipitation anomalies of the previous periods because the rates are highly correlated with precipitation lagged two to ten years. Ablation season temperatures are known to have a strong influence on glacier change in a continental climate (Letreguilly, 1988; Bitz & Battisti, 1999; Pelto, 2006; Bolch, 2007; Hoffman et al., 2007; Koch et al., 2009; Moore et al., 2009). I correlated lagged temperature and precipitation anomalies with glacier changes in the Columbia Icefield and found that annual and accumulation season precipitation anomalies have higher and more significant correlations than temperature anomalies. The higher correlations occurred at lag times similar to those reported by Salinger et al. (1983), Sigurdsson et al. (2007), and Beedle et al. (2009). They determined that terminus changes are correlated with temperatures lagged a few years and precipitation lagged five to ten years. 3.5.3. J Regional and Global Comparison The general pattern of glacier change in the 20th century observed at the Columbia Icefield is one of rapid ice loss from the 1920s to the 1950s, followed by reduced ice loss in the 1960s and 1970s, and ending in renewed large losses during the 1980s and 1990s. This pattern is similar to those exhibited by most glaciers throughout British Columbia and elsewhere in western North America (McCabe & Fountain, 1995; Arendt et al., 2002; Menounos et al., 2005; Pelto, 2006; VanLooy et al., 2006; Hoffman et al., 2007; Schiefer et al., 2007; VanLooy & Forster, 2008; Koch et al., 2009) Some glaciers in the North Cascades of Washington State and in Alaska rapidly thinned in the 1990s at approximately double previous rates (Pelto & Hedlund, 2001; Arendt et al., 2002; VanLooy et al., 2006). 78 The similar glacier response throughout western North America suggests common large-scale forcing mechanisms. These mechanisms could include ocean-atmosphere circulation patterns, specifically the Pacific Decadal Oscillation (PDO) and the Pacific North American pattern (PNA). These patterns influence the temperatures and paths of winter storm systems that bring the majority of precipitation to western North America and have been linked to changes in mass balance (Bitz & Battisti, 1999; Dyurgerov & Meier, 2000; Demuth et al., 2008). Shifts in the PDO to negative values in 1945/1946 and to positive values in 1976/1977 have been linked to reduced rates of ice loss and renewed recession, respectively (Luckman et al., 1987; McCabe & Fountain, 1995; Dyurgerov & Meier, 2000; Pelto, 2006; Demuth et al., 2008). My rates of glacier change show moderate correlations (r = -0.24 to r = -0.47) with the PDO (Mantua, 2000), but none were significant (p < 0.05). The lack of significant correlation may be due to my choice of periods which are based on the available data for glacier changes rather than when climate changes occur. My rates of area and volume change are significantly correlated (respectively, r = -0.81, p < 0.01 and r = -0.70, p < 0.05) with the PNA (Mitchell, 2010), and while not significant, rates of length and elevation are also well correlated (respectively, r = -0.58 and r = -0.66). Similar trends in glacier change throughout the 20th century have been documented for glaciers in Europe, Asia, and South America. Since the end of the 1970s, rates of retreat, thinning, and negative mass balance of glaciers around the world have increased (Barry, 2006; Kaser et al., 2006; Bolch, 2007). In the Italian Alps, Valle d'Aosta glaciers experienced rapid retreat from 1930 to 1950 followed by a reduction in retreat from 1960 to 1980 (Vanuzzo, 2001), similar to glaciers of the Columbia Icefield. Again, there appears to be large-scale temperature forcing influencing long-term glacier response, even on a global 79 scale. However, the doubling of thinning rates in the 1990s documented in Patagonia (Rignot et al., 2003) and the Himalayas (Berthier et al., 2007) did not occur in the Columbia Icefield, indicating regional and local differences in climate and topography are still important influences on glacier change. 3.6 Conclusions I determined changes in glacier length, area, elevation, and volume in the Columbia Icefield from 1919 to 2009 using IBCS maps, aerial photographs, and satellite imagery. The IBCS maps and older aerial photography were integral in expanding the measurement record back to the early 20th century. This study identified moderate errors in the maps and photographs, but these errors are small compared to the large glacier changes observed over the length of the record. Over nine decades, the Columbia Icefield shrank by 59.60 ±1.19 km2, thinned by 49.4 ± 25.2 m w.e., and lost 14.30 ± 2.02 km3w.e. Changes in individual glaciers differed due mainly to the watershed (i.e. which one the glacier is in), glacier type, size, and debris cover. Size is the main factor, but significant differences stem from other factors, which must be taken into account when modeling glacier response to climate change. Rates of ice loss are strongly influenced by lagged annual and accumulation season precipitation and, to a lesser extent, by lagged temperature anomalies. The different glacier changes have variable response times to temperature and precipitation, indicating the importance of long glacier records. 80 4. Conclusions I used IBCS maps to determine area, elevation, and volume changes in glaciers in the central and southern Canadian Rocky Mountains over nine decades, from the 1920s into the 21st century. Glacierized area in the Canadian Rocky Mountains decreased 40 ± 7%, while at the Columbia Icefield, glaciers retreated 1149 ± 34.1 m, decreased 22 ± 5%, thinned 49.4 ± 25.2 m w.e., and lost 14.30 ± 2.02 km3 w.e. Differences in glacier response are related mainly to glacier size, but glacier type (e.g. outlet or debris-covered) is also important. For instance, debris cover can reduce thinning rates. Lagged precipitation and temperature anomalies ranging between two and ten years, are correlated with ice losses over the time. Periods of little or no ice loss occurred during or following periods of high precipitation and low temperature anomalies, while periods of increased ice loss occurred during or following periods of low precipitation and high temperature anomalies. Limitations of this study are mainly a product of data quality and availability. The IBCS maps contain mismapped snow patches, incorrect glacier extents, and cut-off glaciers that had to be identified and edited or removed. Shadows, snow cover, and poor contrast in the glacier accumulation areas are the main issues with the imagery and hindered the extraction of glacier extents and elevation data. Incomplete coverage of glaciers, primarily in the Columbia Icefield for some years of photography required that ice losses had to be estimated, and may not represent the actual glacier change over the period between photograph acquisitions. Glacier change in the Canadian Rocky Mountains could only be evaluated over three common periods, reducing the confidence in the correlations between area change and climate. Averaging over the period 1919-1985 obscures short-term glacier change and climate variability. Also, the periods represent the available data for calculating 81 glacier change and may not reflect climate patterns. Despite these limitations, I calculated substantial ice losses in the Canadian Rocky Mountains, determined valuable relations between glacier change and climate, and observed strong influences of glacier properties and type on ice loss. The data in this thesis spatially and temporally expand the record of glacier change in the Canadian Rocky Mountains, but more can be done. I did not make full use of the 1919 elevation data, which can provide elevation and volume change for other glaciers in the Canadian Rocky Mountains, besides the Columbia Icefield. It would be interesting to compare those changes with area-volume scaling results and to look for correlations between the types of glacier changes (i.e. length and volume). Future work should further explore the response of different glacier types to climate change for other glaciers in the Canadian Rocky Mountains. Debris cover and connectivity are important influences on ice loss and need to be better understood and accounted for in glacier models. A comparison of geodetic glacier changes to modeled changes would be useful validation for glacier modeling and may identify areas for which improvement is possible. Using the Columbia Icefield data in conjunction with the Peyto mass balance record and climate data may allow for the development of mass balance gradients for other glaciers in the Canadian Rocky Mountains and modeling their future response. 82 References Andreassen, L.M., Elvehey, H. & Kjollmen, B., 2002. Using aerial photography to study glacier changes in Norway. Annals of Glaciology, 34, pp. 343-348. Andreassen, L.M., Paul, F., Kaab, A. & Hausberg, J.E., 2008. Landsat-derived glacier inventory for Jotunheimen, Norway, and deduced glacier changes since the 1930s. The Cryosphere, 2, pp. 131-145. Arendt, A.A., Echelmeyer, K. A., Harrison, W. D., Lingle, C. S., & Valentine, V. B., 2002. Rapid wastage of Alaska glaciers and their contribution to rising sea level. Science, 297, pp. 382-386. D01:10.1126/science.1072497. Bader, H., 1954. Sorge's Law of densification of snow on high polar glaciers. Journal of Glaciology, 2, pp. 319-323. Barrand, N.E., Murray, T., James, T. D., Barr, S. L., & Mills, J. P., 2009. Optimizing photogrammetric DEMs for glacier volume change assessment using laser-scanning derived ground-control points. Journal of Glaciology, 55, pp. 106-116. Barry, R.G., 2006. The status of research on glaciers and global glacier recession: a review. Progress in Physical Geography, 20, pp.285-306. DOI: 10.1191/0309133306pp478ra. BC Ministry of Forests, 1998. Zone and Provincial Classification Reports. [Online] Available at: http://www.for.gov.bc.ca/hre/becweb/resources/classificationreports/provincial/index.html [Accessed 18 May 2011]. Beedle, M.J., Menounos, B., Luckman, B.H. & Wheate, R., 2009. Annual push moraines as climate proxy. Geophysical Research Letters, 36. L20501. D01:10.1029/2009GL039533. Berthier, E., Arnaud, Y., Baratoux, D., Vincent, C., & Remy, F., 2004. Recent rapid thinning of the "Mer de Glace" glacier derived from satellite optical images. Geophysiscal Research Letters, 31. L17401. D01:10.1029/2004GL020706. Berthier, E., Arnaud, Y., Kumar, R., Ahmad, S., Wagnon, P., & Chevallier, P., 2007. Remote sensing estimates of glacier mass balances in the Himachal Pradesh (Western Himalaya, India). Remote Sensing of Environment, 108, pp. 327-338. Berthier, E., Arnaud, Y., Vincent, C. & Remy, F., 2006. Biases of SRTM in high-mountain areas: Implications for the monitoring of glacier volume changes. Geophysical Research Letters, 33. L08502. D01:10.1029/2006GL025862. Berthier, E., Schiefer, E., Clarke, G. K., Menounos, B., & Remy, F., 2010. Contribution of Alaskan glaciers to sea-level rise derived from satellite imagery. Nature Geoscience, pp. 1-4. D01:10.1038/NGE0737. 83 Bitz, C.M. & Battisti, D.S., 1999. Interannual to decadal variability in climate and the glacier mass balance in Washington, western Canada, and Alaska. American Meteorological Society, 12, pp. 3181-3196. Bolch, T., 2007. Climate change and glacier retreat in northern Tien Shan (Kazakhstan/Kyrgyzstan) using remote sensing data. Global and Planetary Change, 56, pp. 1-12. Bolch, T., Menounos, B. & Wheate, R., 2010. Landsat-based inventory of glaciers in western Canada, 1985-2005. Remote Sensing of Environment, 114, pp. 127-137. D01:10.1016/j.rse.2009.08.015. Brock, B.W., Mihalcea, C., Kirkbride, M. P., Diolaiuti, G., Cutler, M. E., & Smiraglia, C., 2010. Meteorology and surface energy fluxes in the 2005-2007 ablation seasons at the Miage debris-covered glacier, Mont Blanc Massif, Italian Alps. Journal of Geophysical Research, 115. D09106. D01:10.1029/2009JD013224. Cox, L.H. & March, R.S., 2004. Comparison of geodetic and glaciological mass-balance techniques, Gulkana Glacier, Alaska, U.S.A. Journal of Glaciology, 50, pp.363-370. Daly, C., Gibson, W. P., Taylor, G. H., Johnson, G. L., & Pasteris, P., 2002. A knowledgebased approach to the statistical mapping of climate. Climate Research, 22, pp. 99-113. DeBeer, C.M. & Sharp, M.J., 2007. Recent changes in glacier area and volume within the southern Canadian Cordillera. Annals of Glaciology, 46, pp. 215-221. Demuth, M., Pinard, V., Pietroniro, A., Luckman, B., Hopkinson, C., Domes, P., & Comeau, L., 2008. Recent and past-century variations in the glacier resources of the Canadian Rocky Mountains: Nelson River system. Terra Glacialis, Special Issue: Mountain Glaciers and Climate Changes of the Last Century, pp. 27-52. Demuth, M.N., Sekerka, J. & Bertollo, S., 2009. Glacier mass balance observations for Peyto Glacier, Alberta, Canada (updated to 2007). [Online] Spatially referenced data set contribution to the National Glacier-Climate Observing System, State and Evolution of Canada's Glaciers Available at: http://pathwavs.geosemantica.net/WSHome.aspx?ws=NGP SECG&locale=en-CA. Geological Survey of Canada. Denton, G.H., 1975. Glaciers of the Canadian Rocky Mountains. In W.O. Field, ed. Mountain Glaciers of the Northern Hemisphere. Hanover, NH: United States Army Cold Regions Research and Engineering Laboratory, pp. 603-653. Dyurgerov, M.B. & Bahr, D.B., 1999. Correlations between glacier properties: finding appropriate parameters for global glacier monitoring. Journal of Glaciology, 45, pp. 9-16. Dyurgerov, M.B. & Meier, M.F., 2000. Twenieth century climate change: Evidence from small glaciers. Proceedings of the National Academy of Sciences of the United States of America, 97, pp. 1406-1411. 84 Farr, T.G., Rosen, P. A., Caro, E., Crippen, R., Duren, R., Hensley, S., et al., 2007. The Shuttle Radar Topography Mission. Reviews of Geophysics, 45. RG2004. DOI:10.1029/2005RG000183. Field, W.O..J., 1948. Glacier observations in the Canadian Rockies. Canadian Alpine Journal, 32, pp. 99-114. Fox, A.J. & Nuttall, A.M., 1997. Photogrammetry as a research tool for glaeiology. Photogrammetric Record, 15, pp. 725-737. Gardner, J.S., 1978. What glaciers do for western Canada. Canadian Geographical Journal, 96, pp. 28-33. Granshaw, F.D. & Fountain, A.G., 2006. Glacier change (1958-1998) in the North Cascades National Park Complex, Washington, USA. Journal of Glaeiology, 52, pp. 251-256. Hauer, F.R., Baron, J. S., Campbell, D. H., Fausch, K. D., Hostetler, S. W., Leavesley, G. H., et al., 1997. Assessment of climate change and freshwater ecosystems of the Rocky Mountains, USA and Canada. Hydrological Processes, 11, pp. 903-924. Henoch, W.E.S., 1971. Estimate of glaciers' secular (1948-1966) volumetric change and its contribution to the discharge in the Upper North Saskatchewan River Basin. Journal of Hydrology, 12, pp. 145-160. Heusser, C.J., 1956. Postglacial environments in the Canadian Rocky Mountains. Ecological Monographs, 26, pp. 263-302. Hoelzle, M., Chinn, T., Stumm, D., Paul, F., Zemp, M., & Haeberli, W., 2007. The application of glacier inventory data for estimating past climate change effects on mountain glaciers: A comparison between the European Alps and the Southern Alps of New Zealand. Global and Planetary Change, 56, pp. 69-82. Hoffman, M.J., Fountain, A.G. & Achuff, J.M., 2007.20th-century variations in area of cirque glaciers and glacierets, Rocky Mountain National Park, Rocky Mountains, Colorado, USA. Annals of Glaeiology, 46, pp. 349-354. Interprovincial Boundary Commission, 1917. Report of the Commission Appointed to Delimit the Boundary between the Provinces of Alberta and British Columbia: From 1913-1916, Parti. Ottawa, ON: Office of the Surveyor General. Jiskoot, H., Curran, C.J., Tessler, D.L. & Shenton, L.R., 2009. Changes in Clemenceau Icefield and Chaba Group glaciers, Canada, related to hypsometry, tributary detachment, length-slope and area-aspect relations. Annals of Glaeiology, 50, pp. 133-143. Jones, N.K. & Rowbotham, D., 2000/2001. Glaciological and historical analyses at the Boundary Glacier, Canadian Rocky Mountains. Western Geography, 10/11, pp. 30-42. 85 Kaab, A., Paul, F., Maisch, M., Hoelzle, M., & Haeberli, W., 2002. The new remote-sensingderived Swiss glacier inventory: II. First results. Annals of Glaciology, 34, pp. 362-366. Kaab, A. & Vollmer, M., 2000. Surface geometry, thickness changes and flow fields on creeping mountain permafrost: automatic extraction by digital image analysis. Permafrost Periglacial Processes, 11, pp. 315-326. Kaser, G., Cogley, J. G., Dyurgerov, M. B., Meier, M. F., & Ohmura, A., 2006. Mass balance of glaciers and ice caps: Concensus estimates for 1961-2004. Geophysical Research Letters, 33. L19501. D01:10.1029/2006GL027511. Kirkbride, M.P. & Warren, C.R., 1999. Tasman Glacier, New Zealand: 20th-century thinning and predicted calving retreat. Global and Planetary Change, 22, pp. 11-28. Kite, G.W. & Reid, I.A., 1977. Volumetric change of the Athabasca Glacier over the last 100 years. Journal of Hydrology, 32, pp. 279-294. Koch, J., Menounos, B. & Clague, J.J., 2009. Glacier change in Garibaldi Provincial Park, southern Coast Mountains, British Columbia, since the Little Ice Age. Global and Planetary Change, 66, pp. 161-178. DOI: 10.1016/j.gloplacha.2008.11.006. Konecny, G., 1963. Glacier surveys in western Canada. Photogrammetric Engineering, 30, pp. 64-82. Letreguilly, A., 1988. Relation between the mass balance of Western Canadian mountain glaciers and meteortological data. Journal of Glaciology, 34, pp. 11-18. Luckman, B.H., 1986. Landform Development in the Forefields of Athabasca and Dome Glaciers. Report to Parks Canada, Contract Number KJP 02960. Luckman, B.H., Harding, K.A. & Hamilton, J.P., 1987. Recent glacier advances in the Premier Range, British Columbia. Canadian Journal of Earth Sciences, 24, pp. 11491161. Luckman, B.H. & Kavanagh, T., 2000. Impact of climate fluctuations on mountain environments in the Canadian Rockies. Ambio, 29, pp. 371-380. Mantua, N., 2000. The Pacific Decadal Oscillation (PDO). [Online] Available at: http://iisao.washington.edu/pdo/ [Accessed 19 Feb 2012]. Marshall, S.J., White, E. C., Demuth, M. N., Bolch, T., Wheate, R., Menounos, B., et al., 2011. Glacier water resources on the eastern slopes of the Canadian Rocky Mountains. Canadian Water Resources Journal, 36, pp. 109-134. D01:10.4296/cwrj3602823. Mattson, L.E., 2000. The influence of a debris cover on the mid-summer discharge of Dome Glacier, Canadian Rocky Mountains. IAHSPublications, 264, pp. 25-33. 86 Mattson, L.E., Gardner, J.S. & Young, G.J., 1993. Ablation on debris covered glaciers: An example from the Rakhiot Glacier, Punjab, Himalaya. IAHS Publications, 218, pp. 289296. Mbogga, M., Hamann, A. & Wang, T., 2009. Historical and projected climate data for natural resource management in western Canada. Agricultural and Forest Meteorology, 149, pp.881-890. McCabe, G.J. & Fountain, A.G., 1995. Relations between atmospheric circulation and mass balance of South Cascade Glacier, Washington, U.S.A. Arctic and Alpine Research, 27, pp. 226-233. Meek, V., 1948. Glacier observations in the Canadian Cordillera. Canadian Geographical Journal, 37, pp. 190-209. Menounos, B., Clague, J.J., Gilbert, R. & Slaymaker, O., 2005. Environmenttal reconstruction from a varve network in the southern Coast Mountains, British Columbia, Canada. TheHolocene, 15, pp. 1163-1171. DOI:10.1191/0959683605hl888rp. Mitchell, T., 2010. Pacific/North American (PNA) index, 1948 - October 2010. [Online] Available at: http://iisao.washington.edu/data/pna/ [Accessed 19 Feb 2012]. Mitchell, T.D. & Jones, P.D., 2005. An improved method of cronstructing a database of monthly climate observations and associated high-resolution grids. International Journal of Climatology, 25, pp. 693-712. Moore, R.D., Fleming, S. W., Menounos, B., Wheate, R., Fountain, A., Stahl, K., et al., 2009. Glacier change in western North America: influences on hydrology, geomorphic hazards and water quality. Hydrological Processes, 23, pp. 42-61. DOI: 10.1002/hyp.7162. Nakawo, M. & Rana, B., 1999. Estimate of ablation rate of glacier ice under a supraglacial debris layer. Geografiska Annaler, 81A, pp. 695-701. Osborn, G. & Luckman, B.H., 1988. Holocene glacier fluctuations in the Canadian Cordillera (Alberta and British Columbia). Quaternary Science Reviews, 7, pp. 115-128. Osborn, G.D., Robinson, B.J. & Luckman, B.H., 2001. Holocene and latest Pleistocene fluctuations of Stutfield Glacier, Canadian Rockies. Canadian Journal of Earth Sciences, 38, pp. 1141-1155. DOI: 10.1139/cjes-38-8-l 141. Osborn, G., Stockmal, G. & Haspel, R., 2006. Emergence of the Canadian Rockies and adjacent plains: A comparison of physiography between end-of Laramide time and the present day. Geomorphology, 75, pp. 450-477. D01:10.1016/j.geomorph.2005.07.032. Parks Canada Agency, 2011. State of Canada's Natural and Historic Places 2011. Catalogue No: R61-63/201 IE-PDF. Parks Canada. 87 Paul, F., 2002. Changes in glacier area in Tyrol, Austria, between 1969 and 1992 derived from Landsat 5 Thematic Mapper and Austrian Glacier Inventory data. International Journal ofRemote Sensing, 23, pp. 787-799. DOI: 10.1080/01431160110070708. Pelto, M.S., 2006. The current disequilibrium of North Cascade glaciers. Hydrological Processes, 20, pp. 769-779. D01:10.1002/hyp.6132. Pelto, M.S. & Hedlund, C., 2001. Terminus behaviour and response time of North Cascade glaciers, Washington, U.S.A. Journal of Glaciology, 47, pp. 497-506. Reid, I.A. & Charbonneau, J.O.G., 1979. Glacier Surveys in Alberta -1977. IWD Report Series 65. Ottawa, ON: Environment Canada, Inland Waters Directorate, Water Resource Branch. Reynolds, J.R., 1996. Using Digital Elevation Models to Measure the Surface and Volumetric Change of Athabasca Glacier, Alberta, Canada, 1919-1979. MA thesis. Waterloo, ON: Wilfrid Laurier University. Reynolds, J.R. & Young, G.J., 1997. Changes in areal extent, elevation and volume of Athabasca Glacier, Alberta, Canada, as estimated from a series of maps produced between 1919 and 1979. Annals of Glaciology, 24, pp. 60-65. Rignot, E., Echelmeyer, E. & Krabill, K., 2001. Penetration depth of interferometric synthetic-aperture radar signals in snow and ice. Geophysical Research Letters, 28, pp. 3501-3504. Rignot, E., Rivera, A. & Casassa, G., 2003. Contribution of the Patagonia Icefields of South America to sea level rise. Science, 302, pp. 434-437. D01:10.1126/science.1087393. Robinson, B.J., 1998. Reconstruction of the Glacial History of the Columbia Icefield, Alberta. MSc thesis. London, ON: The University of Western Ontario. Salinger, M.J., Heine, M.J. & Burrows, C.J., 1983. Variations of the Stocking (Te Wae Wae) Glacier, Mount Cook, and climatic relationships. New Zealand Journal of Science, 26, pp. 321-338. Schiefer, E. & Gilbert, R., 2007. Reconstructing morphometric change in a proglacial landscape using historical aerial photography and automated DEM generation. Geomorphology, 88, pp. 167-178. Schiefer, E., Menounos, B. & Wheate, R., 2007. Recent volume loss of British Columbian glaciers, Canada. Geophysical Research Letters. L16503. DOI: 10.1029/2007/GL030780. Sigurdsson, O., Jonsson, T. & Johannesson, T., 2007. Relation between glacier-termini variations and summer temperature in Iceland since 1930. Annals of Glaciology, 46, pp. 170-716. DOI:10.3189/172756407782871611. 88 Stahl, K. & Moore, R.D., 2006. Influence of watershed glacier coverage on summer streamflow in British Columbia, Canada. Water Resources Research, 42. W06201. DOI:10.1029/2006WR005022. Takeuchi, Y., Kayastha, R.B. & Nakawo, M., 2000. Characteristics of ablation and heat balance in debris-free and debris-covered areas on Khumbu Glacier, Nepal Himalayas, in the pre-monsoon season. IAHS Publications, 264, pp. 53-61. Tennant, C., Menounos, B., Ainslie, B., Shea, J., & Jackson, P., 2012. Comparison of modeled and geodetically-derived glacier mass balance for Tiedemann and Klinaklini glaciers, southern Coast Mountains, British Columbia, Canada. Global and Planetary Change, 82-83, pp. 74-85. D01:10.1016/j.gloplacha.2011.11.004. VanLooy, J.A. & Forster, R.R., 2008. Glacial changes of five southwest British Columbia icefields, Canada, mid-1980s to 1999. Journal of Glaciology, 54,469-478. VanLooy, J., Forster, R. & Ford, A., 2006. Accelerating thinning of Kenai Peninsula glaciers, Alaska. Geophysical Research Letters, 33. L21307. DOI:10.1029/2006GL028060. Vanuzzo, C., 2001. The glacier retreat in Valle d'Aosta (Western Italian Alps) from the Little Ice Age to the second half of the 20th century: linear, areal, volumetric, and Equilibrium Line Altitude changes. Geografia Fisca EDinamica Quaternaria, 24, pp. 99-113. Wang, T., Hamann, A., Spittlehouse, D.L. & Murdock, T.Q., 2012. ClimateWNA - Highresolution spatial climate data for western North America. Journal of Applied Meteorology and Climatology, 51, pp. 16-29. D01:http://dx.doi.org/10.1175/JAMC-D-l 1043.1. Wheeler, A.O., 1920. The application of photography to the mapping of the Canadian Rocky Mountains. Canadian Alpine Journal, 2, pp. 76-87. Wheeler, J.O., 1988. A.O. Wheeler and mapping in the Canadian Cordillera. In B. Farrell & A. Desbarats, eds. Explorations in the History of Canadian Mapping; A Collection of Essays. Ottawa, ON: Association of Canadian Map Libraries and Archives, pp. 211-222. 89 Appendices Appendix A: Ancillary information and rectification results for Interprovincial Boundary Commission Survey (IBCS) maps and aerial photographs. Table A.l Rectification information for IBCS maps used in glacier change analysis. Map Sheet 9 Year RMSE RMSE v(m) RMSE #of GCPs Landsat used for reetificathn fpath/row-yynmdd) 8.9 13.9 11.7 11.8 12.7 12.2 18.6 15.8 16.3 16.1 27 31 28 30 28 13 1916 1916 1916 1916 1913 *(•) 8.4 12.3 10.7 11.3 10.0 14 1913 14.5 13.4 19.7 29 15 1916 13.7 14.1 19.7 33 16 1903,1906 15.5 14.8 21.4 31 17 1917 9.8 11.9 15.4 32 18 19 21* 22" 23' 24 1917 1918 1918 1918,1919 1919 1919,1920 1919,1921 15.2 11.4 14.8 19.6 24.9 7.9 2.3 11.9 11.5 12.7 15.1 14.1 11.2 7.9 19.3 16.2 19.5 24.8 28.6 13.7 8.2 33 28 40 36 34 34 25 25 1920,1921 10.2 10.0 14.3 30 26 1920 9.0 9.5 13.0 31 7.7 27 1920,1921 10.5 13.0 9.2 7.4 11.8 1921 28 9.2 29 1917 8.2 12.3 1922,1924 14.6 12.2 19.0 30 7.8 6.3 10.0 1922 31 10.3 12.7 1922,1924 16.4 32 13.6 11.7 18.0 33 1923 1923 16.1 16.2 22.9 34 10.0 1923 7.9 12.7 35 1923 6.0 7.4 9.5 36 7.1 8.5 1924 38 11.1 1924 11.0 10.3 39 15.1 Mean 1919 12.2 11.5 16.2 * Map sheets containing the Columbia Icefield. 30 30 27 34 30 36 33 36 28 28 27 27 43/24-060802,43/25-060802 43/24-060802,43/25-060802 43/24-060802,43/25-060802 43/24-060802,43/25-060802 43/24-060802,43/25-060802 43/24-060802,43/25-060802,44/24-010914, 44/24-060819 43/24-060802,43/25-060802,44/24-010914, 44/24-060819 43/24-060802,43/25-060802,44/24-010914, 44/24-060819 43/24-060802,43/25-060802,44/24-010914, 44/24-060819 44/24-010914,44/24-060819 44/24-010914,44/24-060819,45/24-060826 44/24-010914,44/24-060819,45/24-060826 44/24-010914,44/24-060819,45/24-060826 44/24-010914,44/24-060819,45/24-060826 44/24-010914,44/24-060819,45/24-060826 44/24-010914,44/24-060819,45/24-060826 44/24-010914,44/24-060819,45/23-850731, 45/24-060826 44/24-010914,44/24-060819,45/23-850731, 45/24-060826 45/23-850731,45/24-060826 45/23-850731,46/23-040928 45/23-850731,46/23-040928 45/23-850731,46/23-040928 45/23-850731,46/23-040928 45/23-850731,46/23-040928 46/23-040928 46/23-040928 46/23-040928 46/23-040928 46/23-040928,47/22-040818 47/22-040818 31 'ilPlllwlSliiSI?HillllillSlliSliiiiiliwtfSlll^Ss 10 11 12 20 90 Table A.2 Aerial photograph data and photogrammetry results. SHKC Year Federal 1948 10-Sep A11711 067-079,188-193,199-200,203-207 19-Sep All729 002-013,019-029 10-Sep All655 219-227 1:20,000 152.40 12.5 0.3 29 0.58 0.69 0.90 0.25 0.11 Federal 1955 6-Aug A14895 023-027,070-077, 115-119 1:70,000 152.83 14.0 1.0 34 0.44 0.46 0.63 0.09 0.34 Federal 1966 22-Aug A19684 001-007,026-033,066-070 1:60,000 153.00 12.5 0.8 31 0.41 0.37 0.55 0.11 0.23 A19685 105-110 Provincial 1970 19-Aug 15BC5394 001-004, 067-075 1:80,000 153.34 14.0 1.1 35 0.42 0.42 0.59 0.06 0.48 Federal 1974 1-Sep A23885 044-050 1:80,000 152.39 12.5 1.0 22 0.30 0.40 0.50 0.04 0.40 Federal 1979 9-Jul A25165 072-075,095-102,123-127 1:70,000 153.37 12.5 0.9 36 0.61 0.49 0.78 0.08 0.31 A25177 129-133 0.8 61 0.92 1.07 1.41 0.12 0.43 0.6 36 0.41 0.44 0.60 0.10 0.29 Mean RMSE 0S1 0.54 0.75 0.10 0.32 Provincial 1986 15-Aug 15BC86085 143-147,148-153 1:60,000 152.26 14.0 Federal 1993 9-Sep A27990 053-060,086-093, 123-128 1:50,000 152.85 12.5 Appendix B: Pearson's Correlation coefficients of climate variables with glacier changes in the Canadian Rocky Mountains. Table B.l Correlations" with absolute area change in the Canadian Rocky Mountains by period (n = 3), for a lag of zero to 19 years. Climate Variable* |pg Lag (Y«r) 0.941 0.974 0.980 0.923 0.928 0.993 0.903 0.971 0.952 0.941 1.000* 0.872 0.850 0.744 0.604 0.977 0.897 0.752 0.968 0.934 1.000** 0.872 0.918 0.685 0.884 0.858 0.595 0.516 0.438 0.897 0.929 0.952 0.880 1.000* 0.924 0.922 0.889 Annual 0.672 0.582 0.586 0.442 0.406 -0.539 0.648 -0.254 -0.285 -0.362 0.848 -0.332 1.000** -0.662 0.194 0.215 0.892 0.868 0.746 0.097 Abi. -0.375 -0.926 -0.978 -0.568 -0.880 -0.987 -0.907 -0.999* -0.940 -0.993 -0.998* -0.832 -0.766 -0.853 -0.644 Acc. 0.730 0.718 0.762 0.703 1.000** 0.531 0.360 0.180 0.070 0.484 0.993 0.768 0.675 0.999* 0.992 0.994* 0.998 0.988* 1.000* 1.000 0.995 0.998* 1.000** 0.9% 0.952 0.953 0.918 0.966 0.913 0.894 0.979 0.999* 0.994 0.996 0.968 0.914 0.967 0.970 0.934 0.904 0.914 0.918 0.981 0.999* 0.999* 0.948 1.000* 0.992 1.000** 1.000** 0.937 0.959 0.998* 0.938 0.946 0.883 0.718 Annual 0.013 0.107 0.040 -0.034 0.265 0.801 0.986 0.909 0.906 0.913 0.929 0.960 0.828 0.908 0.942 0.884 0.995 0.955 0.682 0.924 Abi. 0.576 0.770 0.384 0.131 0.650 0.912 0.998* 1.000* 0.968 0.910 0.992 0.993 0.971 Acc, -0.338 -0.218 -0.157 -0.158 -0.009 0.548 0.930 0.746 0.696 0.921 0.759 0.844 0.624 0.905 0.909 0.879 0.978 0.695 0.408 0.758 0.911 0.493 0.979 0.359 0.561 0.452 0.834 0.140 0.207 0.455 0.518 0.714 0.503 0.768 -0.672 0.912 0.892 0.027 0.761 0.509 0.068 0.442 -0.174 0.907 0.663 0.438 -0.678 0.952 0.777 0.559 -0.934 0.446 0.704 0.993 0.478 -0.279 0.012 0.924 0.459 0.995 0.756 0.917 0.001 0.616 1.000* 0.414 -1.000** -0.332 0.839 0.921 0.807 -0.594 0.373 -0.983 0.609 1.000* 0.222 -0.824 Significant correlations are denoted by * p < 0.05 and ** p < 0.01. Tmax is maximum temperature, Tmean is mean temperature, Tmin is minimum temperature, and P is precipitation. Abi. is ablation season and Acc. is accumulation season. Table B.2 Correlations3 with rates of absolute area change in the Canadian Rocky Mountains by period (n = 3), for a lag of zero to 19 years. Climate Variable11 I (Year) mHSpjlll P Acc. Annual Abl. Acc. Annual Abl. -0.420 -0.134 0.917 0.532 0.999* -0.676 -0.573 -0.754 -0.943 -0.710 -0.912 -0.556 -0.585 -0.975 -0.919 -0.270 0.024 -0.429 0.876 0.291 0.984 0.061 -0.700 0.011 0.704 0.972 -0.560 -0.649 -0.974 0.183 -0.892 -0.279 0.068 -0.422 0.906 0.317 -0.778 -0.927 -0.448 -0.192 -0.660 0.935 0.864 0.972 -0.711 -0.606 -0.998* -0.540 -0.375 -0.353 -0.413 0.787 -0.702 -0.690 -0.405 -1.000* -0.099 -0.524 0.450 0.926 -0.399 0.985 0.527 -0.411 -0.483 -0.270 0.243 0.026 -0.275 -0.004 0.576 0.560 0.721 -0.375 -0.698 0.048 0.114 0.001 0.990 0.739 -0.363 -0.467 -0.535 -0.442 -0.606 -0.721 0.080 0.994 0.436 0.838 -0.476 -0.378 -0.736 -0.412 -0.903 -0.155 -0.937 0.047 0.893 -0.440 -0.727 -0.741 -0.605 -0.931 -0.084 0.452 1.000* -0.038 -0.053 -0.773 0.998* 0.495 0.620 -0.383 -0.609 -0.109 -0.728 -0.734 -0.716 0.161 -0.463 -0.361 -0.805 0.947 0.442 -0.275 -0.600 -0.440 -0.701 -0.501 -0.894 -0.399 -0.789 -0.980 -0.651 -0.628 -0.971 -0.762 -0.191 -0.887 -0.691 -0.682 -0.492 -0.821 -0.646 -0.837 -0.814 -0.989 -0.794 -0.984 -0.297 -0.941 -0.743 -0.664 -0.594 -0.962 -0.904 -0.998* -0.902 -0.956 -0.152 -0.993 -0.720 -0.728 -0.774 -0.737 -0.733 -0.741 -0.968 -0.997* -0.967 -1.000* -0.831 -0.720 -0.920 -0.673 -0.566 -0.735 -0.980 -0.457 -0.887 -0.992 -0.997* -0.989 -0.933 -0.423 -0.982 -0.731 -0.921 -0.772 -0.762 -0.780 -0.998* -0.990 -0.947 -0.892 -0.650 -0.478 -0.739 -0.185 -0.998* -0.841 -0.416 -0.946 -0.998* -0.778 -0.881 -0.981 -0.995 0.030 -0.927 -0.493 -0.927 -0.094 -0.709 0.395 -0.997 -0.742 -0.896 -0.472 -1.000* 0.390 -0.923 -0.414 -0.965 0.411 -0.722 0.685 17 -1.000** 0.003 -0.985 0.548 -0.513 -0.857 -0.414 -0.967 -0.005 -0.826 0.308 -0.204 • 8 Significant correlations are denoted by * p < 0.05 and ** p < 0.01. b is maximum temperature, T^n is mean temperature, Tmin is minimum temperature, and P is precipitation. Abl. is ablation season and Acc. is accumulation season. Table B.3 Correlations8 with relative area changes in the Canadian Rocky Mountains, by period (n = 3), for a lag of zero to 19 years. Climate Variable* Tnnm Abl. Acc. Annual Annual Abl. Aec. 0.670 -0.376 0.999* 0.966 0.981 0.015 0.577 -0.336 0.940 0.977 0.897 0.729 0.993 0.999* 0.771 0.974 0.898 0.581 -0.927 0.717 0.914 0.109 -0.216 0.928 0.994 0.999* 0.385 0.980 0.754 0.585 0.895 0.042 -0.156 -0.978 0.761 0.952 0.133 0.923 0.967 0.947 0.998* 0.441 -0.569 0.702 0.980 -0.032 -0.156 0.879 0.651 0.927 0.934 0.999* 1.000* 0.988 0.404 -0.881 1.000* 0.266 -0.007 1.000* 0.993 0.993 0.925 -0.987 0.533 0.994 0.549 0.802 0.913 1.000* -0.537 1.000** 0.904 0.923 0.998* 0.930 1.000** 0.873 -0.253 -0.906 0.362 0.996 0.986 1.000* 0.889 1.000* 0.917 0.744 0.995 0.967 0.971 -0.284 -0.998* 0.182 0.908 1.000** 0.683 0.694 0.937 0.968 0.998* 0.072 0.905 0.650 -0.360 -0.940 0.913 0.952 0.920 0.960 0.883 -0.330 0.485 0.966 0.913 1.000** 0.942 -0.992 0.909 0.849 0.758 0.998* 0.993 0.969 0.996 0.992 1.000* -0.660 -0.998* 0.928 1.000* 0.857 0.843 0.993 0.951 0.937 0.891 -0.833 0.767 0.933 0.960 0.593 0.871 0.192 0.971 0.623 0.515 0.867 0.953 0.945 -0.767 0.673 0.903 0.827 0.849 0.213 0.745 0.905 0.917 0.437 0.910 0.095 -0.854 0.491 0.913 0.908 0.882 0.743 0.908 0.833 0.560 0.138 -0.646 0.358 0.917 0.942 0.716 0.451 0.603 0.979 0.878 0.767 0.891 0.025 -0.674 0.205 0.911 0.883 0.713 0.501 0.453 0.517 0.906 0.978 0.760 -0.176 -0.680 0.066 0.951 0.441 0.995 0.662 0.436 0.508 0.476 0.557 -0.281 -0.934 0.010 0.993 0.924 0.696 0.955 0.703 0.776 0.445 0.995 0.755 0.457 -0.334 -1.000** 0.999* 0.410 0.916 0.413 0.000 0.684 0.614 0.806 0.759 0.371 0.922 -0.825 -0.983 -0.595 0.999* 0.925 0.838 0.608 0.220 a Significant correlations are denoted by * p < 0.05 and ** p < 0.01. b Tjnax is maximum temperature, Tme is mean temperature, T j is minimum temperature, and P is precipitation. Abl. is ablation season and Acc. an is accumulation season. m n Table B.4 Correlations" with rates of relative area changes in the Canadian Rocky Mountains, by period (n = 3), for a lag of zero to 19 years. S8IIt Climate Variable* Annual Acc. Abl. Abl. Annual Acc." "i -0.674 -0.572 -0.712 -0.912 -0.418 -0.554 0.918 0.534 0.999 -0.752 -0.942 -0.132 -0.427 0.877 0.985 -0.584 0.062 -0.699 0.009 -0.919 -0.268 0.026 0.292 -0.974 -0.559 -0.891 -0.420 0.907 0.972 0.319 -0.648 0.181 -0.277 0.070 0.705 -0.973 -0.658 0.936 0.972 -0.709 -0.605 -0.777 -0.998* -0.541 -0.927 -0.447 -0.191 0.865 -0.688 -0.412 0.788 0.926 -0.701 -0.404 -0.101 -0.373 -0.523 -0.351 0.452 -1.000* -0.269 -0.273 -0.397 0.525 0.578 -0.409 -0.482 0.245 0.028 0.561 -0.003 0.985 -0.373 -0.696 0.049 0.116 0.003 0.737 0.722 -0.362 -0.465 -0.534 -0.440 0.991 -0.376 -0.902 -0.153 -0.719 0.082 0.994 0.434 0.839 -0.475 -0.605 -0.735 -0.410 -0.931 -0.082 -0.936 0.453 1.000* 0.045 0.894 -0.438 -0.726 -0.037 -0.740 -0.604 -0.772 0.162 0.622 -0.381 -0.107 -0.727 -0.733 -0.715 -0.051 0.998* 0.493 -0.608 -0.438 -0.699 -0.893 -0.360 -0.804 -0.398 0.440 -0.273 -0.461 -0.599 -0.499 0.947 -0.788 -0.971 -0.887 -0.681 -0.627 -0.820 -0.760 -0.980 -0.192 -0.649 -0.690 -0.491 -0.663 -0.961 -0.813 -0.989 -0.793 -0.299 -0.941 -0.645 -0.742 -0.836 -0.593 -0.984 -0.740 -0.903 -0.998* -0.902 -0.956 -0.993 -0.719 -0.727 -0.772 -0.736 -0.732 -0.154 -0.968 -0.997* -0.830 -0.719 -0.919 -0.671 -0.565 -0.734 -0.980 -0.968 -0.458 -1.000* -0.779 -0.991 -0.997* -0.989 -0.933 -0.425 -0.983 -0.886 -0.729 -0.921 -0.771 -0.761 -0.998* -0.990 -0.998* -0.947 -0.892 -0.946 -0.477 -0.738 -0.183 -0.842 -0.417 -0.649 -0.998* -0.980 -0.995 -0.928 -0.880 -0.926 0.397 -0.779 0.028 -0.492 -0.092 -0.708 -0.470 -0.999* -0.924 -0.895 -0.964 0.686 -0.997 -0.743 0.389 -0.413 0.412 -0.721 0.005 -0.985 -0.515 -0.856 -0.966 -0.003 0.310 -1.000* -0.206 0.546 -0.413 -0.825 " Significant correlations are denoted by * p < 0.05 and ** p < 0.01. b Tmax is maximum temperature, T„,,„ is mean temperature, T min is minimum temperature, and P is precipitation. Abl. is ablation season and Acc. is accumulation season. Table B.5 Correlations" with length change rates in the Columbia Icefield, by time period (n = 10), for a lag of zero to 19 years. Climate Variable* Lag (Yew) ML, : p Acc. "> Acc. Annual Abl. Annual Abl. -0.409 -0.068 0.500 -0.448 -0.072 -0.626 0.275 0.235 0.272 0.608 -0.759* -0.653* -0.200 0.203 -0.330 0.618 0.509 -0.526 0.655* -0.856** -0.776** -0.292 0.069 -0.374 0.743* 0.714* -0.661* -0.787** -0.234 0.142 -0.376 0.775** 0.247 0.778** -0.912** -0.331 -0.601 -0.147 -0.778** -0.252 0.113 -0.337 0.633* -0.109 0.697* -0.876** -0.521 -0.203 0.093 -0.484 -0.715* -0.260 -0.643* -0.356 0.807** 0.483 0.813** -0.122 -0.357 0.050 0.659* -0.398 -0.572 -0.353 -0.478 -0.158 -0.202 0.824** 0.819** -0.192 -0.284 -0.135 0.098 -0.290 -0.393 0.035 -0.379 0.068 -0.144 0.847** 0.221 0.867** 0.040 0.260 -0.341 0.005 -0.410 -0.113 0.035 -0.210 0.721* 0.140 0.788** -0.333 -0.305 -0.253 0.109 -0.394 -0.112 0.037 -0.175 0.572 0.061 0.631 0.141 -0.321 -0.187 -0.085 -0.093 0.449 -0.204 -0.334 -0.274 -0.103 0.436 0.227 -0.119 -0.039 -0.178 -0.146 -0.164 -0.055 -0.126 0.076 0.320 -0.140 0.395 -0.252 -0.267 -0.081 -0.054 -0.059 0.043 -0.196 0.011 -0.072 0.022 0.259 -0.154 0.338 -0.155 -0.169 -0.094 -0.045 -0.054 -0.066 0.052 -0.234 -0.268 -0.065 0.295 0.314 -0.198 -0.435 -0.365 -0.112 -0.459 -0.241 -0.127 -0.352 0.236 -0.036 0.270 -0.371 -0.242 -0.488 -0.485 -0.134 -0.318 -0.340 -0.170 0.085 -0.127 0.138 -0.242 -0.304 -0.301 0.105 -0.020 -0.153 0.235 -0.236 0.063 -0.287 -0.261 -0.225 -0.183 0.185 -0.395 -0.023 -0.322 -0.300 0.002 -0.358 -0.244 -0.023 -0.382 0.115 0.084 0.106 -0.277 -0.170 -0.250 0.018 -0.446 0.138 0.192 -0.413 -0.019 -0.411 0.027 0.051 -0.527 -0.509 -0.485 -0.400 -0.152 -0.453 0.015 0.015 -0.179 -0.077 0.004 -0.505 " Significant correlations are denoted by * p < 0.05 and ** p < 0.01. b Tmax is maximum temperature, T^n is mean temperature, T^,, is minimum temperature, and P is precipitation. Abl. is ablation season and Acc. is accumulation season. j -0.396 0.511 0.638* -0.594 0.338 -0.185 -0.466 -0.134 -0.007 -0.642* -0.506 -0.596 -0.598 '• ?-r Table B.6 Correlations" with area change rates in the Columbia Icefield, by time period (n = 10), for a lag of zero to 19 years. Climate Variable" Tana mmsm iglllflp Abl. Acc. | Annual Acc. Annual Abl. -0.395 -0.632* -0.822** 0.327 0.219 -0.574 0.027 -0.712* -0.062 0.356 -0.325 0.243 0.379 0.429 -0.355 -0.505 -0.646* -0.335 -0.566 -0.376 -0.101 -0.410 0.440 -0.562 0.536 -0.689* -0.755* -0.187 -0.559 0.541 0.412 -0.661* -0.426 -0.673* -0.311 -0.485 0.038 0.582 -0.692* -0.084 -0.791** -0.762* 0.056 -0.873** -0.469 -0.118 -0.643* 0.549 3 -0.376 0.471 -0.747* -0.715* 0.356 -0.813** -0.523 -0.107 -0.583 0.359* 0.202 -0.762* -0.667* -0.133 0.650* 0.291 0.673* -0.666* -0.027 -0.759* 0.142 -0.771** -0.492 -0.670* 0.514 0.653* -0.568 -0.572 -0.567 -0.089 -0.576 -0.426 -0.149 -0.496 0.655* -0.205 -0.347 -0.334 -0.069 0.684* 0.098 0.714* -0.395 0.140 -0.404 0.337 -0.397 -0.349 -0.069 -0.392 -0.379 -0.317 0.485 -0.199 0.591 -0.381 -0.310 0.124 -0.157 -0.348 0.446 -0.232 -0.338 -0.261 -0.184 -0.284 0.348 -0.315 -0.103 -0.341 0.151 -0.223 -0.097 -0.066 0.217 0.144 0.217 -0.231 -0.398 -0.039 -0.166 -0.253 -0.012 -0.212 -0.375 0.047 -0.085 0.172 -0.031 -0.261 0.183 0.145 -0.220 0.222 0.000 0.202 0.065 0.170 0.040 -0.020 0.163 0.209 0.004 -0.039 -0.216 -0.036 -0.331 -0.105 -0.297 -0.014 0.058 -0.017 0.128 0.310 0.002 -0.099 -0.162 -0.030 0.043 0.104 -0.136 0.004 -0.237 0.029 0.286 -0.168 -0.214 -0.219 0.036 -0.339 0.114 -0.262 -0.188 -0.080 0.303 -0.115 -0.319 -0.219 -0.110 -0.406 0.003 -0.268 0.029 -0.241 0.307 0.500 -0.084 -0.058 -0.106 -0.116 0.056 -0.071 -0.191 -0.138 -0.152 -0.384 -0.147 -0.179 0.041 -0.182 -0.085 0.173 -0.138 -0.193 -0.092 -0.200 -0.121 0.145 0.070 -0.145 -0.317 -0.075 0.022 -0.221 0.191 -0.157 -0.144 -0.243 -0.313 -0.046 -0.429 -0.344 -0.317 -0.332 -0.135 -0.253 -0.377 -0.304 -0.326 0.057 -0.340 -0.315 a Significant correlations are denoted by * p < 0.05 and ** p < 0.01. b is maximum temperature, T^,, is mean temperature, Tmio is minimum temperature, and P is precipitation. Abl. is ablation season and Acc. is accumulation season. Table B.7 Correlations3 with elevation change rates in the Columbia Icefield, by time period (n = 10), for a lag of zero to 19 years. P lJMK ImmSm, Abl. Acc. Annual tbl. 0.362 -0.360 -0.007 0.343 -0.099 0.383 0.422 -0.132 0.262 -0.313 -0.452 0.173 0.563 0.687* 0.658* -0.183 -0.218 -0.572 -0.623 -0.435 0.179 0.005 0.647* 0.112 0.531 -0.163 0.740* -0.423 0.205 0.070 -0.153 -0.547 0.102 0.767** -0.535 0.667* 0.605 -0.471 0.282 -0.040 -0.189 0.843** 0.781** 0.070 0.461 -0.515 0.551 -0.260 0.568 -0.541 0.239 -0.152 -0.220 0.855** 0.813** 0.106 0.394 0.399 -0.678* -0.255 -0.438 0.229 0.565 -0.095 0.368 -0.215 0.848** 0.820** 0.794** 0.011 -0.538 -0.313 -0.323 0.263 0.779** 0.750* 0.150 -0.044 -0.080 0.785** 0.283 -0.412 -0.345 0.533 -0.263 0.328 0.621 0.050 0.730* 0.231 0.536 -0.019 -0.313 0.784** 0.652* -0.234 0.595 -0.183 0.444 0.678* 0.667* 0.185 0.470 0.069 0.494 0.300 -0.217 -0.200 -0.176 0.436 0.373 0.586 0.214 0.669* 0.665* 0.330 0.679* 0.103 -0.201 -0.076 0.443 0.509 0.594 0.243 0.593 0.089 -0.171 0.567 0.277 -0.150 0.535 -0.102 0.330 0.526 0.280 0.551 0.526 0.610 0.225 -0.159 -0.100 0.465 0.449 -0.044 0.595 0.246 -0.127 0.098 0.332 0.778** 0.150 0.471 0.476 0.469 0.347 -0.091 0.108 0.423 0.430 -0.056 0.615 0.310 0.201 -0.082 0.465 0.462 0.662* 0.378 -0.192 0.448 0.099 0.302 0.616 0.023 0.449 0.389 0.503 -0.027 0.206 0.590 -0.148 0.517 0.051 0.159 0.336 0.401 0.353 0.340 0.468 0.107 -0.158 0.305 -0.155 0.141 0.388 0.395 0.343 0.581 0.227 0.672* 0.073 -0.087 0.462 0.255 0.092 0.370 0.470 0.215 0.021 -0.091 0.325 0.264 -0.149 0.662* 0.171 0.548 0.201 0.353 0.399 0.240 -0.027 0.314 0.248 0.144 0.719* -0.060 0.600 -0.226 0.265 0.329 0.510 0.064 0.101 0.008 -0.200 0.196 0.139 0.158 -0.180 0.461 a Significant correlations are denoted by * p < 0.05 and **p< 0.01. b Tmax is maximum temperature, T^,, is mean temperature, T^,, is minimum temperature, and P is precipitation. Abl. is ablation season and Acc. is accumulation season. Table B.8 Correlations" with volume change rates in the Columbia Icefield, by time period (n = 10), for a lag of zero to 19 years. Climate Variable* Lag (Year) Wis Warns 'aen Annual Acc. | Annual Abl. Acc. Abl. 0.369 -0.151 0.476 0.449 -0.176 0.298 -0.357 0.012 -0.528 0.166 0.529 -0.440 0.603 0.576 0.066 0.628 -0.147 -0.613 -0.379 0.581 -0.157 0.133 -0.542 0.211 0.700* 0.731* 0.655* -0.151 -0.121 -0.547 -0.575 -0.400 0.542 0.140 0.050 0.223 0.746* -0.175 -0.022 0.812** -0.467 0.560 -0.520 -0.311 0.624 0.091 0.437 0.311 -0.243 0.770** -0.158 0.799** -0.254 -0.584 0.326 -0.704* 0.591 0.106 0.426 0.248 -0.288 -0.151 -0.042 -0.376 0.808** -0.623 -0.525 0.816** 0.860** 0.572 0.321 0.207 -0.187 -0.139 0.069 0.818** -0.532 -0.440 -0.444 0.805** 0.840** 0.527 0.212 0.212 -0.060 0.663* 0.141 -0.443 -0.269 -0.399 -0.141 0.854** 0.803** 0.624 0.517 0.263 0.757* 0.414 0.380 0.060 -0.368 -0.289 -0.326 0.551 -0.058 0.208 0.768** 0.607 0.757* 0.005 0.764* 0.587 0.398 -0.348 -0.196 -0.301 0.434 0.272 0.108 0.596 0.673* 0.549 0.638* 0.080 0.507 0.435 -0.294 -0.230 -0.190 0.596 0.2% 0.175 0.638* 0.604 0.244 0.544 -0.177 0.270 -0.223 -0.113 0.542 0.494 0.405 0.492 0.357 0.539 0.169 0.602 0.713* 0.558 -0.097 -0.003 -0.114 0.505 0.325 0.368 0.499 0.226 0.636* 0.680* 0.531 -0.058 0.106 -0.031 0.340 0.486 0.437 0.399 0.487 0.005 0.522 -0.037 0.626 0.603 0.096 -0.201 0.439 0.427 0.291 0.183 0.449 0.122 0.454 0.309 0.175 -0.135 0.081 -0.121 0.537 0.507 0.342 0.233 0.462 0.434 0.324 0.730* -0.049 0.175 -0.084 0.268 0.149 0.613 0.420 0.196 0.377 0.324 0.642* -0.090 0.029 0.060 -0.123 0.217 0.246 0.414 0.427 0.576 0.670* 0.025 -0.049 -0.193 0.611 0.252 0.128 0.313 0.416 0.205 0.376 0.293 0.515 0.040 0.204 0.402 0.130 -0.236 -0.024 -0.169 0.186 0.104 0.292 0.318 a Significant correlations are denoted by * p < 0.05 and ** p < 0.01. b Tmax is maximum temperature, T^an is mean temperature, Tmi„ is minimum temperature, and P is precipitation. Abl. is ablation season and Acc. is accumulation season. Appendix C: Raw and estimated glacier measurements and change data for each year or period of available imagery in the Columbia Icefield. Bold values were used in the estimation of missing glacier measurements indicated by a grey box. Table C. 1 Raw length measurements (m). Year 2009 2154.8 FS2 m FS8 2059.3 2040.0 2062.1 1893.5 1863.0 1814.4 1791.7 1779.6 1716.4 7570.0 7290.2 7304.2 7293.6 7303.2 7248.8 704 .3 7081.9 7133.4 7052.9 2701.8 2260.8 2236.1 2244.5 2211.5 2214.7 2223.1 2188.7 2113.1 2089.5 6269.7 6180.0 6114.6 6063.4 6018.3 5993.1 5935.5 5837.5 5960.8 5712.8 9230.3 9063.4 8866.7 8772.8 8806.1 8793.0 8693.5 8645.5 8587.6 8485.9 2900.2 2830.7 2605.2 2603.9 2615.0 2615.7 2617.7 2622.7 2601.9 2343.5 3007.4 2990.3 2928.1 2927.3 2928.0 2825.0 2637.1 1539.4 1654.2 1673.9 1594.8 524.6 1554.1 1532.9 3764.2 3784.0 3782.2 3683.7 3674.7 3379.5 3281.3 3165.9 2980.4 1293.1 3006.6 FS10 14083.2 13867.0 13548.8 13468.6 13290.0 13095.3 12790.1 12672.2 12247.6 3187.6 1728.6 ; 3231.5 1065.0 3330.8 tm\ 3260.1 1859.3 3307.2 966.0 3282.0 864.1 3237.5 904.2 3077.0 795.2 1695.8 1466.1 1396.7 1598.6 1487.6 11391.4 11126.7 11064.3 10965.2 107633 1354.0 10689.4 10736.7 10810.2 10907.1 5602.7 10810.8 5744.5 10696.0 5707.4 10686.0 5622.7 10842.9 5812.7 1328.8 10176.0 10840.4 1298.7 mm 2225.6 11575.5 5658.5 5547.0 3436.3 3426.4 3306.4 3261.5 5831 2 •H FSK 1212.2 5763.2 3226.4 IIBiiiii 3490.4 3455.6 8359.3 8382.6 8606.7 8229.6 7348.7 3306.6 6497.1 3760.4 3753.2 3697.8 3677.9 1151.6 1143.5 3613.2 1132.4 3525.4 1005.5 3483.3 957.7 2609-3 2477.7 8636.1 •M 1275.6 fsa 9958 9 10907.8 5784.2 2645.6 2577.5 2902.3 2900.0 2905.1 2870.4 2915.0 2867.5 2749.1 2638.8 2550.4 3430.9 3360.8 3354.3 3138.2 3160.7 3116.1 3115.1 3076.6 2681.6 4653.6 4703.2 4746.6 4755.5 4676.9 4612.1 4669.7 4543.0 3646.7 3516.7 3382.6 3335.8 3331.9 3315.5 3446.5 2978.0 4444.6 4268.5 FS24 Mm 4688.5 100 Table C.2 Estimated length measurements (m). Year 1919 1941 19SS 19(6 197* 1974 HI 1979 1986 1993 2001 2009 1716.4 2 54.8 2059.3 2040.0 2062.1 1893.5 863.0 1814.4 1791.7 1779.6 7570.0 7290.2 7304.2 7293.6 7303.2 7248.8 7041.3 7081.9 7133.4 7052.9 2701.8 2260.8 2236.1 2244.5 2211.5 2214.7 2223.1 2188.7 2113.1 2089.5 FS4 6269.7 6180.0 6114.6 6063.4 60 8.3 5993.1 5935.5 5837.5 5960.8 5712.8 m 9230.3 9063.4 8866.7 8772.8 8806.1 8793.0 8693.5 8645.5 8587.6 8485.9 PSt 2900.2 2830.7 2605.2 2603.9 2615.0 2615.7 2617.7 2622.7 2601.9 2343.5 3007.4 2990.3 2928.1 2927.3 2928.0 2716.0 | 2825.0 2637.1 rss mma sHSS 1654.2 1673.9 1594.8 1524.6 1554. 3764.2 3784.0 3782.2 3683.7 3674.7 FS10 14083.2 13867.0 13548.8 3260.1 3187.6 1859.3 1728.6 2225.6 16958 1466.1 FSI4 11575.5 11391.4 Qgjg Bit 10736.7 10810.2 5622.7 mm mm FS12 1532.9 3379.5 1486.2 3165.9 1539.4 1293.1 3281,3 2980.4 3006.6 13468.6 13290.0 13095.3 12790.1 12672.2 12247.6 3231.5 3330.8 3307.2 3282.0 3237.5 3077.0 1065.0 1212.2 966.0 864.1 904.2 795.2 1396.7 1598.6 1487.6 1354.0 1328,8 1298,7 11126.7 11064.3 10965.2 10763.3 10689.4 10176.0 9958.9 10907.1 10907. 10842.9 10810.8 10696.0 10840.4 10686.0 5602.7 5763.2 5784.2 5812.7 5744.5 5707.4 5658.5 5547.0 8636.1 3490.4 8359.3 3455.6 8382.6 3753.2 8606.7 3426.4 8229.6 3306.4 7348.7 3306.6 6497.1 3261.5 5831.2 3483.3 3226.4 FSIS 3697.8 3677.9 3613.2 3525.4 1275.6 1151.6 1143.5 1132.4 1005.5 957.7 2577.5 2609.3 2477.7 2867.5 2749.1 2638.8 2550.4 3116.1 3115.1 3076.6 2681.6 3760.4 rsu 2798.6 2902.3 2900.0 2905.1 2870.4 2645.6 2915.0 3476.1 3430.9 3360.8 3354.3 3138.2 3160.7 4653.6 4703.2 4746.6 4755.5 4676.9 4612.1 4669.7 4543.0 3646.7 3516.7 3382.6 3335.8 3396.6 3331.9 3315.5 3446.5 2978.0 4444.6 4268.5 4770.5 3633.2 4688.5 MMI I estimated missing length measurements (NA) using the following example calculations with data from FS7, 1979 to 1993 (black box): 1979 length, Lig: 2753.8 m 1993 length, £93: 2716.0 m 1986 length, Lze- NA Length change 1979 to 1993, AL79.93 = £93-1,79 = 2716.0 - 2753.8 m = -37.8 m Year difference 1979 to 1993,179-93 = 1993 - 1979 = 14 a Rate, ALr = AZ79.93 / 779.93 = -37.8 m /14 a = -2.7 m a"1 Year difference 1979 to 1986, 7 9-86= 1986 - 1979 = 7 a 1986 length = L19 + (AZ,R) (779-86) = 2753.8 m + (-2.7 m a"1) (7 a) = 2734.9 m 7 Year difference 1986 to 1993, 786-93 = 1993 - 1986 = 7 a Check = Igfi + (IR) (786.93) = 2734.9 m + (-2.7 m a1) (7 a) = 2716.0 m 101 Table C.3 Raw area measurements (km2). I estimated missing area measurements (NA) using the following example calculations with data from FS7,1979 to 1993 (black box): 1979 area, ^79: 1.39 km2 1993 area, ^93:1-28 km2 1986 area, A^: NA Area change 1979 to 1993, AA79.93 =^93-^79 = 1.28 - 1.39 km2 = -0.11 km2 Year difference 1979 to 1993, 779-93 = 1993 - 1979 = 14 a Rate, AAr = A479-93/ *79-93 = -0.11 km2 / 14 a = -0.0078 km2 a"1 Year difference 1979 to 1986, 779-86= 1986 - 1979 = 7 a 1986 area = A-79 + (A/fR) (779-86) = 1.39 km2 + (-0.0078 km2 a1) (7 a) = 1.34 km2 Year difference 1986 to 1993,3^86-93 = 1993 - 1986 = 7 a Check = AS6 + (^r) (*86-93) = 1.34 km2 + (-0.0078 km2 a"1) (7 a) = 1.28 km2 102 Table C.4 Estimated area measurements (km2) Year 19W IMS vm 1955 1966 1970 1986 1993 2001 128 1.23 1.17 1.06 1.03 1.02 0.87 21.58 21.31 21.18 20.98 20.87 20.81 20.34 1974 2009 3.28 3.04 2.91 2.80 2.71 2.72 2.29 9.58 9.31 9.28 9.03 8.89 8.68 8.07 20.13 19.82 19.65 19.35 19.13 18.90 18.45 2.86 2.75 2.72 2.68 2.63 2.60 2.33 1.53 1.44 1.40 1.28 1.29 1.15 0.37 0.36 0.35 0.30 0.24 2.30 2.14 2.04 1.66 0.32 1.57 1.55 1.38 44.38 43.60 42.85 41.99 40.88 39.99 38.32 2.42 2.47 2.41 2.36 2.22 1.07 1.11 0.81 0.87 0.65 1.39 1 1.38 1.93 JH 1.77 1.67 1.81 1.69 1.65 1.49 19.69 19.16 18.76 18.14 17.70 16.66 16.02 23.83 24.06 24.05 23.84 23.06 22.80 22.22 10.26 10.42 10.31 10.42 9.97 9.87 9.42 36.60 7.62 36.04 7.50 35.80 7.55 35.11 7.05 34.28 6.93 33.04 6.60 32.15 4.93 4.71 4.51 4.27 4.16 3.94 0.43 0.39 0.31 0.27 2.57 §HH 2.09 5.85 5.71 2.38 5.97 2.38 0.49 FS24 2.70 Total i 265.05 231.94 103 5.37 1.81 1.75 1.96 1.50 6.77 6.65 6.63 5.94 2.50 2.50 2.38 2.13 Table C.5 Raw elevation change data (m w.e.). Period 1948 19191955 1919. 1966 19191979 1944- 194»- 19551955 1966 1966 19551970 19661978 19661979 197* 1974 19701979 19741979 FS4 -38.20 I979l 1996 19791999 t9M1993 1986- -3.48 -4.11 -1.88 -3.82 -1.96 2.16 -3.16 -5.97 -5.47 -6.31 0.91 -1.68 -3.10 -4.57 1.51 -2.75 -2.00 -3.42 -0.73 -54.79 -51.56 -3.35 -0.03 -51.87 1.27 -6.25 -12.15 0.95 -36.59 -37.14 -47.69 -45.35 -8.65 0.89 -0.26 -39.04 -79.31 -61.64 -28.69 -44.18 -3.40 0.50 -6.17 -48.75 -5.03 -3.93 2.52 2.91 -0.12 -3.74 -4.82 0.01 -9.67 -2.34 -0.51 -4.90 12.11 -42.76 -33.32 -84.03 -58.47 -4.37 -3.56 -5.38 -1.91 -3.32 -0.50 0.31 -2.19 -2.25 -1.90 -39.58 -6.44 -5.58 -2.90 -38.56 -45.82 -3.08 8.52 -19.32 -56.00 -8.72 -3.93 -1.21 2.76 -5.90 -4.73 15.01 -11.78 -92.33 3.16 -43.78 -54.72 mH< -86.04 " -49.91 -3.93 3989 -2.76 -4.29 MM "53 5° 1999- -1.12 -3.67 FS16 1993- -5.78 -5.62 O 1993. •ma |mq i«mi« iiinn _J22—liii—fS5GL»^j!!SJ^— -7.40 1.45 -1.33 -2.48 0.08 -1.44 -1.38 -1.62 -0.75 10.18 -6.40 -56.87 -1.88 -2.35 -3.13 -51.83 -0.31 -1.88 0.38 5.94 -6.96 -37.93 -39.34 -2.29 -49.40 I estimated missing elevation change data (NA) using the following example calculations with data from FS7,1979 to 1993 (black box): Elevation change 1979 to 1993, A//79.93: -5.62 m w.e. Elevation change 1979 to 1986, NA Elevation change 1986 to 1993, AH^y- NA Year difference 1979 to 1993, 779-93 = 1993 - 1979 = 14 a Rate, AHR = A//79-93 / 779-93 = -5.62 m w.e. /14 a = -0.40 m a"1 Year difference 1979 to 1986, *79-86 - 1986 - 1979 = 7 a Elevation change 1979 to 1986 = (AHr) (779-86)= (-0.40 m a"1) (7 a) = -2.81 m w.e. Year difference 1986 to 1993, 786-93 = 1993 - 1986 = 7 a Elevation change 1979 to 1986 = (A//R) (786-93) = (-0.40 m a"1) (7 a) = -2.81 m w.e. Check = A//79.86 + A/fg6-93 = -2.81 + -2.81 m w.e. = -5.62 m w.e. 105 Table C.6 Estimated elevation change data (m w.e.). Period 1919Wt FS4 O ON 19191966 1919. 1979 19481955 19481966 19551966 19661970 -3.76 0.39 0.01 -26.63 -0.76 -0.87 -1.57 -34.39 1.53 -0.07 -3.78 -3.66 -2.79 -3.30 0.04 -1.60 -44.25 1.21 -0.72 -39.80 -1.95 -26.80 -38.20 19661979 19701974 19701979 19741979 19791986 19791993 19061993 19061999 19931999 1995M09 19992009 19192009 -3.93 1.51 -36.59 -37.14 -47.69 -45.35 -54.79 -0.04 -3.48 -4.11 -1.96 -6.31 -4.57 -3.42 -0.73 -51.56 -1.68 0.18 -3.35 -0.03 -51.87 2.32 2.02 -26.92 -0.84 -3.13 -47.12 -3.12 -3.01 FS10 -26.97 -7.38 2.06 FS12 -27.60 -37.84 PS 16 19551970 -17.34 -42.33 FSB 191*1955 -1.10 -4.29 0.95 2.16 0.91 1.27 -6.25 -39.04 -6.44 -12.15 -79.31 mm -5.58 -3.08 -61.64 2.10 -8.65 8.52 -28.69 -1.55 •1 0.49 0.89 -0.26 -44.18 -27.58 -30.71 -2.57 -3.34 -0.21 1.17 0.50 -6.17 -48.75 -4.14 -1.21 -3.93 2.52 -42.76 -3327 -29.15 3.44 -1.93 -0.33 2.91 -9.67 -33.32 0.01 -2.34 -84.03 1.70 IBj 0.35 -3.02 12.11 -19.32 -58.47 -2.94 -1.59 -4.68 2J0 -75.26 -39.58 -3.22 -2.50 -38.36 -45.82 -0.30 -54.72 -7.40 -53.50 0.17 -0.84 0.63 86.04 -49.91 2.69 -0.06 1.07 0.62 -1.83 -56.00 -8.72 2.76 15.01 -11.78 3.16 -43.78 -92.33 -39.34 -2.29 -6.40 -3.13 -6.96 -56.87 -51.83 -37.93 Table C.7 Raw volume change data (xlO6 m3 w.e.). Period m9~ 1919- 1948 1955 19191W6 1919- 1979 -26.32 -603.86 -5.10 -16.62 -138.23 5.15 -0.69 -274.11 -906.86 O -J 194fr> 1955 -818.26 1948- 1966 19991966 0.50 -18.76 1995- 1979 19661978 1966- 197ft- 197ft- 19741979 1974 1979 1979 1979I9S6 19791993 19861993 19M1999 19931999 -58.49 -39.47 -8.51 -11.18 -16.70 -5.34 -10.24 -10.79 -28.87 -49.36 -56.15 -32.93 -87.34 -67.19 0.73 -145.81 3.54 -2.05 -31.68 -0.12 -71.52 -3.07 -2.56 0.26 -4.32 -14.24 -0.32 -1.17 0.83 0.40 -118.33 -1305.81 -7.31 -330.56 -6.92 -207.73 4.33 -140.60 -85.47 -94.04 4.91 -2.15 -86.92 -666.32 -5.54 -67.15 -81.54 2.06 -23.14 -900.28 82.00 -46.41 -8.02 -8.99 -10.10 153.56 -76.48 mm -228.08 -226.06 5.10 0.52 m -20.82 0.72 0.84 -12.42 -42.14 -45.99 -91.3 -89.17 -69.57 -36.98 85.42 67.21 0.12 -351.71 -695.74 -22.98 -1543.45 -66.57 107.56 -230.II -101.69 -5.15 -0.15 -1.86 -19.06 -179.52 -14.52 15.89 -28.03 -738. 2 -148.51 -298.91 -12.31 -33.46 -359.66 JH99 -85.86 -35.04 -0.40 1999- -3.57 0.01 -33.45 -12.41 flH 19933909 -0.12 -0.16 1.67 -2.71 -13.08 19.97 -12.56 -15.63 -20.78 14.14 -16.58 19192009 I estimated missing volume change data (NA) using the following example calculations with data from FS7, 1979 to 1993 (black box): Volume change 1979 to 1993, AF79.93: -7.82 xl06m3 w.e. Volume change 1979 to 1986, AF79.86: NA Volume change 1986 to 1993, A^6-93: NA Year difference 1979 to 1993, F79.93 = 1993-1979 = 14 a Rate, AVr = AF79.93 / 779.93= -7.82 xlO6 m3 w.e. /14 a = -0.56 xlO6 m3 w.e. a"1 Year difference 1979 to 1986, r79-86 = 1986 - 1979 = 7 a Volume change 1979 to 1986 = (AF R) (^79-86) = (-0.56 xlO6 m3 w.e. a"1) (7 a) = -3.91 xlO6 m3 w.e. Year difference 1986 to 1993, y86.93 = 1993 - 1986 = 7 a Volume change 1979 to 1986 = (AF R) (*86-93) = (-0.56 xlO6 m3 w.e. a"1) (7 a) = -3.91 xl06m3w.e. Check = AF79-86 + AFfj6-93 = -3.91 xlO6 + -3.91 xlO6 m3 w.e. = -7.82 xlO6 m3 w.e. 108 Table C.8 Estimated volume change data (xlO6 m3 w.e.). Period 191919S5 1M91*66 1»1»1979 FS2 IMS. 194* 195S1955 1964 1966 -18.76 -16.62 U» 1970 INC- 1946- 1970- 197®. 19741970 1979 1974 1979 1979 -35.04 -67.19 -818.26 1979- 19M1993 1993 198*1999 *9991999 3009 3009 »9I*2009 -39.47 -33.45 -12.41 FS4 19791986 -10.24 -10.79 -28.87 -16.70 -49.36 -31.68 -32.93 -59.88 -11.18 -7.82 -330.56 O SO 207.73 -140.60 -76.48 -153.56 -226.06 81.54 -23.14 -12.42 -42.14 -91.31 41.85 -45.99 94.04 -86.92 -666.32 -67. 5 900.28 -351.71 -46.41 23.64 -695.74 12.97 -1543.45 -50.22 -107.56 -22.98 -57.46 -230.11 -33.46 -36.98 -31.31 -66.57 -10 .69 -8.00 -2.70 -14.52 -0.15 -19.06 | -179.52 -359.66 -298.91 1.03 -1.22 8.72 -28.03 1.41 -1.67 -1.36 19.97 -738.12 7.51 2.62 0.17 -10.12 -11.35 -12.71 -15.63 -20.78 -186.80 -153.29 -4.85 5.56 -0.81 -4.70 0.94 14.14 -16.58 -105.63 148.51 5.89 -0.16 1.67 -13.08 ••• -32.14 -12.56 Appendix D: Area and elevation changes for each period used in the glacier change analysis between 1919 and 2009. (A) 1919 to 1948 (B) [—iw> Extent 1 11988 Extent Etevatton Chang*(m w.*.) Mi -100 to -75 Mi-76 to-80 03-60 to-2S Oafto# l—111» Extent 1—11»4« Extent (RC WT.) -20010 *156 1SO to-100 iii«to28 OtoSO (C) 1919 to 1966 I—I itlt Extent I—Mf Extent Etovation Chang* to-20 (gg -20 to -10 -26 to-10 rji-lOtoO c:].ioioo •iotoio •M10t»20 (C) ••14to 20 1955 to 1966 •1H6BHWH niMIErtwH Bwatton Chang* (m w.«.) •§-IS to-10 ra CD -6too B*«o9 1948 to 1966 (D) 1955 to 1970 I—11t« Extent {—lUTOEidant Eltvatfon Chang* (m w.*.) H <30 to <20 HB-20 to-10 : ; IQtP 0 Miotoie ••4010 39 Figure D.2 Area and elevation change for the periods (A) 1948-1955, (B) 1948-1966, (C) 1955-1966, and (D) 1955-1970. (A) 1966 to 1970 (B) 01M* Extort [—HtTOEoant Elation Chang* (m wi) 1966 to 1979 WM®»0-S •1*» Ertwtt I—I l«7t Extent Etovation Chang* (m ».*) •l-40»o-M •i-^to<2e •B^tos •18 to 10 Cn-ia»o •HO to 10 (C) 1970 to 1974 I |1»70Ejdflt •1l74ExlW< EtevaMon Chang* (m w.t.) Bi-14 to-10 r~i»>too ••I to 10 (D) 1970 to 1979 I—11*79 Ertatrt m Wimwl Etavatton Chang* (m w.*.) to-20 •• -20 to Extent I—Ht»> Extent Elevation Chang* {m w.*.) (•i-20tO.1» •|*1StO-10 Cslll -16 to -9 C~l*«teO mmttos 0 1 t * Figure D.4 Area and elevation change for the periods (A) 1974-1979, (B) 1979-1986, (C) 1979-1993, and (D) 1986-1993. (A) 1986 to 1999 (B) 1993 to 1999 I—11—3 Extant I—12001 Extort Starvation Chang*(m w.a.) ••-49 to-30 • 1m Extent I—12001 Extent PAYATTON CHANG* (M W.«.) flB-Mto4C •H-UTO-15 r;;:;] -1§too r:>i#t©o ••1ft to 30 •00to 19 ••19to 30 (C) 1993 to 2009 £•IMS Extort I—laooo Extort EtovaHon Chang*(m w.«.) ' 1 (D) 1999 to 2009 I—12001 Extort I—12000 Extort Etovattoa Chang*(m w.a.) 17 m-woio-Tt WB-ntn-m mm-Mto-as r^)-2»t»o •MO to 25 Figure D.5 Area and elevation change for the periods (A) 1986-1999/2001, (B) 1993-1999/2001, (C) 1993-2009, and (D) 1999/2001-2009. Appendix E: Rates of elevation change between the years 1948 and 1993 on debriscovered and bare ice for debris-covered glaciers. Due to limited coverage of the 1948 photographs, I calculated elevation change for Boundary (FS7), Columbia (FS18), and FS23 between 1955 and 1993, and for Manitoba (FS19) between 1966 and 1993. (B) Saskatchewan (FS10) (A) Athabasca (FS5) (0 o « 6 « » > 5 -I a » • t I yi E. 8, « S 9 Debris-covered Bare ice * » « T — Debris-covered —— Bare ice * a I i I I i I I I I I I 1 I i § § 8 8 8 8 1 I I I I I I Elevation (m asl) Elevation (m asl) (C) Castteguard IV (FS14) (D) Columbia (FS18) O o (0 ) 2 S i M f * t . O o 9 J "• • i l l 5 IP ! ' U> i ^5 o Debris-covered Bare ice -T- 8 -r-*i—r~ § 8 ' o i » l 9 I -<|i »« « Debris-covered Bare ice I I I I » I I I I I'l I 1 I I I I I I I I I s i l Ml Wl Ml Nl l l l l l i—r Bevation (m asl) Elevation (m asl) Figure E.l Rates of elevation change for glaciers with debris-covered sides: (A) Athabasca, (B) Saskatchewan, (C) Castleguard IV, and (D) Columbia. 115 (B) Stutfltld (FS2) (A|«1 2- • ' 8 I? t « & de i H* 1 i 1 * *«• a f ? i * * E f 1 * or18 oT § tt —• Oe))H» O * ' * * 1 T T a §„ & "* 9 "T ,2 • ^5 I 1 I I I § 1 <3 I I «* i, •B4> * S? Dtfcrtvcowad Bevation (m asl) Elevation (m asl) Figure E.2 Rates of elevation change for glaciers with debris-covered termini: (A) FSl, (B) Stutfield, (C) Kitchener, (D) Dome, (E) Boundary, and (F) Hilda. 116 (B) Manitoba{FS19) (A) FS17 ' (0 €iO 4> i £ iI m ' - f Oebri^covvrtd Sar«(c« i—r-r i » "5 5 1 Q: §M § •• 11 J1, 1• • • OeMKMnd —j—j—T-T—T-J 1 1 1 ! 1 1 1 1 1 1 g I 8 § CN CN N FN Etevation (m as)) Elevation (m asl) (C) PS23 (D)FS24 » »I '« 3- I I T I I f I ! I I ! I f I I I ! I I •-» OeWxevwed — Bare let -T-T T- § § § Bevaiton (m asl) Elevation ;m asl) (E1PS2S i£ °o is I ? M ? •• I ' ' a> DeM*cov«rari I I I I f ! I I 11 I Baltics i i 3 S Bevation (m aid) Figure E.3 Rates of elevation change for glaciers with debris-covered termini: (A) FS17, (B) Manitoba, (C) FS23, (D) FS24, and (E) FS25. 117