SATELLITE IMAGE ATLAS OF GLACIERS OF THE WORLD
GLACIERS OF NORTH AMERICAÑ
GLACIERS OF CANADA
GLACIERS OF THE CANADIAN ROCKIES
By C. Simon L. Ommanney1


Figure captions and tables follow References Cited.
Manuscript approved for publication, 7 March 2002.
FN1: International Glaciological Society, Lensfield Road, Cambridge CB2 1ER, England, U.K. (formerly with the National Hydrology Research Institute [now part of the National Water Research Institute], Environment Canada, 11 Innovation Boulevard, Saskatoon, Saskatchewan S7N 3H5, Canada).

SATELLITE IMAGE ATLAS OF GLACIERS OF THE WORLD
Edited by RICHARD S. WILLIAMS, Jr., and JANE G. FERRIGNO
U.S. GEOLOGICAL SURVEY PROFESSIONAL PAPER 1386ÐJÐ1 

The Rocky Mountains of Canada include four distinct ranges from the U.S. border to northern British Columbia: Border, Continental, Hart, and Muskwa Ranges. They cover about 170,000 km2, are about 150 km wide, and have an estimated glacierized area of 38,613 km2. Mount Robson, at 3,954 m, is the highest peak. Glaciers range in size from ice fields, with major outlet glaciers, to glacierets. Small mountain-type glaciers in cirques, niches, and ice aprons are scattered throughout the ranges. Ice-cored moraines and rock glaciers are also common

Contents
Abstract	
Introduction	  
	Figure 1.	Mountain ranges of the southern Rocky Mountains	  
	2.	Mountain ranges of the northern Rocky Mountains	  
	3.	Oblique aerial photograph of Mount Assiniboine, Banff 
National Park, Rocky Mountains			  
	4.	Sketch map showing glaciers of the Canadian 
Rocky Mountains			  
	5.	Photograph of the Victoria Glacier, Rocky Mountains, 
Alberta, in August 1973			  
	Table 1.	Named glaciers of the Rocky Mountains cited in the chapter	  
	2.	Estimate of glacierized area of the Rocky Mountains of 
Canada			  
	3.	Number and types of glaciers within the Nelson River basin 			  
Border Ranges (B)	  
Clark Range (Wilson and Lewis Ranges) (B1)	  
Macdonald Range (B2)	  
Galton Range (B3)	  
Continental Ranges (SouthÑBorder Ranges to Kicking Horse Pass)	  
Front Ranges (South) (FS)	  
Lizard Range (FS1)	  
Taylor Range (FS2)	  
Flathead Range (FS3)	  
Blairmore Range (FS4)	  
Livingstone Range (FS5)	  
High Rock Range (Tornado Group) (FS6)	  
Wisukitsat and Greenhills Ranges (FS7)	  
Harrison Group East (FS8)	  
Highwood Range (FS9)	  
Misty Range (FS10)	  
	Figure 6.	August 1985 photograph and sketch map, showing 
historical positions of the Rae Glacier, Misty Range, 
Rocky Mountains			  
Elk Range (FS 11)	  
Italian Group (FS12)	  
Joffre Group (French Military Group) (FS13)	  
Opal Range (FS14)	  
Fisher Range (FS15)	  
Kananaskis Range (FS16)	  
Spray Mountains (British Military Group) (FS17)	  
Rundle Peaks (FS18)	  
Goat Range (FS19)	  
Sundance Range (FS20)	  
Park Ranges (South) (PS)	  
Quinn Range (PS1)	  
Harrison Group West (PS2)	  
The Royal Group (PS3)	  
Blue Range (PS4)	  
Assiniboine Park Group (PS5)	  
Mitchell Range (PS6)	  
Ball Range (PS7)	  
Vermilion Range (PS8)	  
Bow Range (PS9)	  
Horseshoe Glacier	  
Wenkchemna Glacier	  
Victoria Glacier	  
	Table 4.	Average annual surface movement of Victoria Glacier		  
Cathedral Glacier	  
Ottertail Range (PS10)	  
Kootenay Ranges (K)	  
Continental Ranges (Central Kicking Horse Pass to Yellowhead 
Pass)	  
	Figure 7.	Annotated Landsat 2 MSS image of several ice fields and 
outlet glaciers in the Rocky Mountains, including the 
Columbia, Wapta, and Waputik Icefields			  
	8.	Reduced segment of the 1:250,000-scale topographic map 
(Golden , 82N) of part of the Rocky Mountains from Waputik
 Icefield to Lyell Icefield			  
Front Ranges (Central) (FC)	  
Fairholme Range (FC1)	  
Ghost River Area (FC2)	  
Clearwater Groups (FC3)	  
Drummond Glacier	  
	Figure 9.	Variations of glaciers in the Rocky Mountains	  
Ram River Glacier	  
	Figure 10.	Photograph of the Ram River Glacier, Rocky Mountains, 
Alberta			  
	11.	Mass-balance measurements of the Ram River Glacier, Rocky 
Mountains, Alberta: 1965Ð1975			 
Palliser Range (FC4)	  
Bare Range (FC5)	  
Vermilion Range (FC6)	  
Sawback Range (FC7)	  
Slate Range (FC8)	  
Murchison Group (FC9)	  
Hector Glacier	  
Ram Range (FC10)	  
Cline Range (FC11)	  
Bighorn Range (FC12)	  
First Range (FC13)	  
Cataract Group (FC14)	  
Le Grand Brazeau (FC15)	  
Southesk Group (FC16)	  
Queen Elizabeth Ranges (FC17)	  
Maligne Range (FC18)	  
Nikanassin Range (FC19)	  
Fiddle Range (FC20)	  
Miette Range (FC21)	  
Jacques Range (FC22)	  
Colin Range (FC23)	  
Park Ranges (Central) (PC)	  
Waputik Mountains (PC1)	  
Waputik Icefield	  
	Figure 12.	High-angle oblique aerial photograph of the Waputik Icefield 
looking to the northwest			  
Balfour Glacier	  
	Figure 13.	High-angle oblique aerial photograph of the Balfour Glacier, 
an outlet glacier of the Waputik Icefield			  
Wapta Icefield	  
Bow Glacier	  
	Figure 14.	Photograph of the terminus of the Bow Glacier, Rocky 
Mountains, Alberta, in September 1973			  
Peyto Glacier	  
	Figure 15.	Photograph of the Peyto Glacier, Rocky Mountains, Alberta,  
in July 1967			  
	16.	Mass-balance measurements of the Peyto Glacier, Rocky 
Mountains, Alberta			  
Yoho Glacier 	  
Van Horne Range (PC2)	  
Conway Group (PC3)	  
Mummery Group (PC4)	  
Barnard Dent Group (PC5)	  
Freshfield Glacier	   
	Figure 17.	High-angle oblique aerial photograph of Freshfield Glacier	  
Waitabit Ridge (PC6)	  
Blackwater Range (PC7)	  
Forbes Group (PC8)	  
Lyell Group (PC9)	  
Southeast Lyell Glacier	   
	Figure 18.	Terrestrial photograph of Southeast Lyell Glacier		  
	19.	High-angle oblique aerial photograph of West Alexandra 
Glacier and South Alexandra Glacier			  
Unnamed Range (Ñ)	  
Vertebrate Ridge (PC10)	  
Kitchen Range (PC11)	  
Columbia Icefield Group (PC12)	  
	Figure 20.	Reduced segments of two 1:250,000-scale topographic maps
(Brazeau Lake, 83C; Canoe River, 83D) of part of the Rocky 
Mountains in the area around the Columbia and 
Clemenceau Icefields			  
	21.	High-angle oblique aerial photographs of the Columbia 
Icefield, the largest ice field in the Rocky Mountains			  
	Table 5.	Characteristics of glaciers in the vicinity of Athabasca 
Glacier			  
Castleguard Glaciers	  
	Figure 22.	High-angle oblique aerial photograph of Castleguard 
Glacier IV, the principal southern outlet glacier of the 
Columbia Icefield			  
Saskatchewan Glacier	  
	Figure 23.	High-angle oblique aerial photograph of Saskatchewan 
Glacier, a major eastward-flowing outlet glacier from the 
Columbia Icefield			  
	Table 6.	Changes in the area and volume of the snout of Saskatchewan 
Glacier, 1965Ð1979			  
Columbia Glacier	  
	Figure 24.	High-angle oblique aerial photograph of Columbia Glacier, 
a major outlet glacier draining the northwest section of the 
Columbia Icefield			  
Athabasca Glacier	  
	Figure 25.	High-angle oblique aerial photograph of Athabasca Glacier, an 
outlet glacier of the Columbia Icefield			  
	Figure 26.	Photographs of the receding terminus of the Athabaska
Glacier, Jasper National Park, Alberta			  
	Table 7.	Recession and volume changes of Athabasca Glacier, 
1870Ð1970			  
	8.	 Changes in the area and volume of the snout of Athabasca 
Glacier, 1959Ð1979			  
	9.	Depth measurements of Athabasca Glacier	  
Dome Glacier	  
Boundary Glacier	  
Hilda Glacier	  
Winston Churchill Range (PC13)	  
Chaba Group (PC14)	  
Clemenceau Icefield Group (PC15)	  
Fryatt Group (PC16)	  
Whirlpool Group (PC17)	  
Unnamed Ranges (Ñ)	  
Cavell Group (PC18)	  
Angel Glacier	  
	Figure 27.	Terrestrial photograph of Angel Glacier, Cavell Group, 
Alberta, in August 1953			  
Cavell Glacier	  
Portal-MacCarib Group (PC19)	  
Trident Range (PC20)	  
Fraser-Rampart Group (PC21)	  
Bennington Glacier	  
Para Glacier	  
Meadow-Clairvaux Group (PC22)	  
Selwyn Range (PC23)	  
Continental Ranges [NorthÑYellowhead Pass to Muskwa
Ranges (M)]	  
	Figure 28.	Segment of the 1:250,000-scale topographic map (Mount 
Robson, 83E) of Mount Robson and environs			  
	29.	Annotated Landsat 1 MSS image showing several icefields and 
outlet glaciers in the Mount Robson are and environs			  
Front Ranges (North) (FN)	  
Boule Range (FN1)	  
Bosche Range (FN2)	  
De Smet Range (FN3)	  
Hoff Range (FN4)	  
Berland Range (FN5)	  
Persimmon Range (FN6)	  
Starlight Range (FN7)	  
The Ancient Wall (FN8)	  
Unnamed Ranges (Ñ)	  
Park Ranges (North) (PN)	  
Victoria Cross Ranges (PN1)	  
Unnamed Ranges (Ñ)	  
Treadmill Ridge (PN2)	  
Robson Group (PN3)	  
Rainbow Range (PN4)	  
Robson Glacier	  
	Figure 30.	High-angle oblique aerial photograph of Mount Robson and 
Mount Robson Glacier on 22 August 1964			  
Whitehorn Group (PN5)	  
Unnamed Ranges (Ñ)	  
Resthaven Group (PN6)	  
	Figure 31.	High-angle oblique aerial photograph of an outlet glacier from 
Resthaven Icefield on 8 August 1961			  
Unnamed Ranges (Ñ)	  
Other Unnamed Ranges (Ñ)	  
Sir Alexander Peaks (PN7)	  
Hart Ranges (H)	  
Dezaiko Range (H1)	  
Misinchinka Ranges (H2)	  
Pioneer Range (H3)	  
	Figure 32.	Photograph of Parsnip Glacier, Hart Ranges, northern Rocky 
Mountains, British Columbia			  
Murray Range (H4)	  
Muskwa Ranges (M)	  
	Figure 33.	Annotated Landsat 1 MSS image mosaic showing part of the 
glacierized northern Rocky Mountains from Mount Lloyd 
George to Mount Roosevelt			  
	34.	Terrestrial photograph of an unnamed glacier in the Muskwa 
Ranges, northern Rocky Mountains			  
Great Snow Mountain Area (M4)	  
Odyssey Icefield	  
Great Snow Icefield	  
Lloyd George Group (M6)	  
Lloyd George Icefield	  
	Figure 35.	Segment of 1:250,000-scale topographic map (Ware, 94F), 
showing the glacierized area around Mount Lloyd 
George, Muskwa Ranges, northern Rocky Mountains			  
Allies Group (M7)	  
Battle of Britain Range	  
Tower of London Range	  
Italy Range	  
Allied Leaders Range	  
North and West of Allies Group	  
References Cited	  


Abstract
The Rocky Mountains of Canada include four distinct mountain ranges (Border, Continental, Hart, and Muskwa Ranges) extending from the U.S. border to northern British Columbia, a distance of 1,350 kilometers. The Rocky Mountains encompasses about 170,000 square kilometers, are about 150 kilometers wide, and have an estimated glacierized area of 38,613 square kilometers. Mount Robson, at 3,954 meters above sea level, is the highest peak. Within the Continental Ranges are the Front, Park, and Kootenay Ranges; these include some of the most spectacular and heavily glacierized mountains in North America. Major glaciers, such as the Waputik, Wapta, Freshfield, Mons, Lyell, and Columbia Icefields are located in these mountains. The Columbia Icefield (325 square kilometers) is the largest ice field in the Rocky Mountains; its three largest outlet glaciers are the Columbia, Athabasca, and Saskatchewan Glaciers. Glaciers range in size from ice fields, with major outlet glaciers, to glacierets. Small mountain-type glaciers in cirques, niches, and ice aprons are scattered throughout the ranges. Ice-cored moraines and rock glaciers are also common.

Introduction
To the early explorers, the Rocky Mountains were a seemingly impenetrable barrier blocking access to the Pacific and inhibiting the coastal inhabitants in their movements to the east. The chain covers about 170,000 km2 and stretches in a continuous series of parallel ranges 150-km wide, from the British Columbia/Alberta/United States border some 1,350 km to northern British Columbia, where the Liard River, in cutting through the chain, acts as a convenient boundary. The same mountain mass continues into the Yukon Territory, but the rocks of the mountains, named here the Mackenzie Mountains, are younger geologically. To the east, the limits of the Rocky Mountains are not sharp. From the heights of the Front Ranges, one passes through the Foothills and out into the Interior Plains. To the west, the boundary is marked by one of the worldÕs greatest physiographic features, which can even be seen from the MoonÑthe Rocky Mountain Trench. The trench continues through the Yukon Territory and Alaska as the Tintina Trench. The highest elevation in the Rocky Mountains is Mount Robson, at 3,954 m above sea level (asl). The lowest elevation, 305 m, is in the north at the junction of the Liard and Toad Rivers.
Gadd (1986) gives an excellent account of the geology of the Rocky Mountains. The mountains were formed by strong compressive forces generated to the west of the Rocky Mountain Trench. During a 75-million-year period that began about 120 million years ago, a series of thrust faults forced the Precambrian, Paleozoic, and Mesozoic sedimentary rocks eastward along a fault plane over the soft Cretaceous shales of the Great Plains. The compressive forces produced a series of folds, which give the Rocky Mountains their characteristic parallel ranges. During the last 2 million years, intense glaciation in the region has eroded the mountains into big, rugged peaks separated by deep, wide valleys, creating a striking east-facing scarp.
There are four distinct ranges within the Rocky Mountains region, known as the Border, Continental, Hart, and Muskwa Ranges (figs. 1 and 2).2 Included in the Continental Ranges are the Front, Park, and Kootenay Ranges. The distinctions between the ranges are based largely on geologic structure and physiography. The entire region is characterized by limestones, quartzites, and argillites, which form massively bold peaks, best expressed in the Park Ranges of the Continental Ranges. The Park Ranges contain some of the most spectacular and heavily glacierized mountains in North America (fig. 3). Flat to gently dipping beds of quartzite or limestone have produced castellated peaks, subsequently modified by glacial oversteepening; large talus cones have developed postglacially. Summit elevations decline northward and southward from the Park Ranges; in particular, the Hart Ranges are considerably lower than the other three ranges. Elevations increase farther north in the Muskwa Ranges, and the castellated limestone and quartzite peaks reappear around Churchill Peak (3,200 m) and Mount Lloyd George (3,000 m) (Slaymaker, 1972).
FN2: The names in this section conform to the usage authorized by the Secretariat of the Canadian Permanent Committee on Geographic Names (CPCGN); URL address: [http:// geonames.nrcan.gc.ca/english]. The Website is maintained by the Secretariat through Geomatics Canada, Natural Resources Canada, and combines the GPCGN server with the Canadian Geographical Names Data Base (CGNDB). Variant names and names not listed in the GPCGN/CGNDB are shown in italics. See also Canadian gazetteers for British Columbia (CPCGN, 1985) and for Alberta (CPCGN, 1988).


Glaciers in the Rocky Mountains (fig. 4, table 1) are typically of the small mountain type, lying in cirques, niches, or along some of the upturned strata as ice aprons. Scattered at intervals along the range are larger ice fields, with major outlet and valley glaciers. Glacierets, ice-cored moraines, and rock glaciers occur at the limits of contemporary glacierization. In some situations, the avalanching of hanging glaciers has created regenerated glaciers lower on the mountain; part of the nourishment of the well-known Victoria Glacier is of this form (fig. 5). 
Unfortunately, the precise area of permanent snow and ice in the Rockies is not known. Henoch (1967) made a determination for the major Canadian drainage areas, but because the Columbia Icefield drains to the Pacific Ocean, to Great Slave Lake, and into the Nelson River system that empties into Hudson Bay, it is hard to determine what proportion of the total area should be attributed to the Rocky Mountains alone (table 2). 
The nature of the ice cover within part of the Rocky Mountains is indicated in table 3 (Ommanney, 1972a), which shows the distribution of various glacier types within the Nelson River drainage basin, stretching from Waterton Lakes National Park to the southern part of the Columbia Icefield. Major ice fields,3 such as the Waputik, Wapta, Freshfield, Mons, Lyell, and Columbia Icefields,3 are not adequately reflected in table 3. Because of the nature of their subglacial topography and ice cover, all of the constituent glaciers can be identified individually.

FN3: Ice field is used in glaciological terminology to refer to Òan extensive mass of land ice covering a mountain region consisting of many interconnected alpine and other types of glaciers, covering all but the highest peaks and ridgesÓ (Jackson, 1997, p. 316). The UNESCO (1970) definition is slightly different. ÒIce masses of sheet or blanket type of a thickness not sufficient to obscure the subsurface topography.Ó In Canada, icefield is used as a synonym for glacier in many glacier place-names and in the text to refer to such place-names, but is not necessarily used in the formal glaciological sense.


Hydrometric stations in the Park Ranges and Muskwa Ranges typically show two prominent stream-discharge peaks that correspond to snowmelt (June) and glacier melt (JulyÐAugust), in addition to a minor peak in the fall. In the Hart Ranges, the glacier-melt peak is generally absent, whereas there is neither a fall-rain peak nor a glacier-melt peak (Slaymaker, 1972) in the Border Ranges farther south.
The glaciers and terrain of the major mountain ranges and their subsidiary mountain ranges, mountains, mountain groups, mountains peaks, and mountain ridges will be discussed here in turn, beginning with the Border Ranges at the southern end of the Rocky Mountains. All major ranges are identified by an alphanumeric code in figures 1 and 2.4 The Continental Ranges, the largest mountain mass, are subdivided into a southern, central, and northern part because they cover such a large area. Within each subdivision, the mountains of the Front and Kootenay Ranges and Park Ranges are considered in turn. The terminology has been drawn from maps and alpine guides (Putnam and others, 1974; Boles and others, 1979); however, some of the names are not endorsed by the Canadian Permanent Committee on Geographical Names (CPCGN).2 Those place-names not yet approved by the CPCGN are shown in italics.

FN4: The major mountain ranges are coded with letters; subsidiary mountain ranges, mountains, mountain groups, mountain peaks, and mountain ridges are coded with numbers. It is noteworthy that many of the subsidiary mountain ranges have never been given formal names; in the subheadings in the text, they are simply referred to as Òunnamed range.Ó On figures 1 and 2 they are unnumbered. Glaciers and ice fields are described within the subsidiary mountain ranges, etc. The letters and numbers are keyed to figures 1 and 2, which delineate each major and subsidiary mountain ridge on a Landsat image mosaic base map.



Border Ranges (B)4
The Border Ranges (fig. 1, B) mark the southern limit of the Rocky Mountains in Canada. They include the Galton, Macdonald, Clark, Wilson, and Lewis Ranges. No substantial ice bodies exist here. The glacierets that are located in the Border Ranges are of little significance for water supply but may be important for some plant and animal life. Abandoned cirques, tarns, horns, and other remnants of mountain glaciation testify to the previous presence and areal extent of extensive glaciers.

Clark Range (Wilson and Lewis Ranges) (B1)4 
The Clark Range (fig. 1, B1) lies along the Continental Divide that forms the border between the provinces of British Columbia and Alberta in the southern Rocky Mountains. The range rises to an elevation of almost 3,000 m asl in Mount Blakiston (2,919 m) and has a mean peak elevation of about 2,500 m asl. The Wilson and Lewis Ranges are outliers within Waterton Lakes National Park.
Although no glaciers are shown on the National Topographic System (NTS) maps for this area nor are any observable on the satellite images, interpretation of aerial photographs for the Canadian Glacier Inventory has revealed some permanent snow and ice, or glacierets, in sheltered areas of the park and farther north in the range around Mount Haig (2,611 m).

Macdonald Range (B2)
The Macdonald Range (fig. 1, B2), lies west of the Clark Range and is separated from it by the Flathead Basin; it also has no glaciers, according to current maps. No inventory has been completed for this area, so it is not known whether glacierets exist. Any accumulation of perennial ice is unlikely, because the mean elevation of the range is lower than that of the Clark Range and, at this latitude in the Rocky Mountains, permanent snow and ice tend to remain only on the sheltered eastern and northern slopes rather than on the more exposed western flanks.

Galton Range (B3)
At an even lower mean elevation, the Galton Range (fig. 1, B3) forms the western flank of the Border Ranges. The range abuts the Rocky Mountain Trench, which in this area includes the Kootenay River and Lake Koocanusa. Because the maximum height is 2,230 m asl and the mean height is just over 2,000 m, it is reasonably certain that not even permanent snow patches exist here.

Continental Ranges (SouthÑBorder Ranges to Kicking Horse Pass)
Because the Continental Ranges extend more than 725 km and contain thousands of glaciers, several of which have been studied and are quite well known, they have been divided into three sectionsÑsouth, central, and north.
The northern boundary of the southern area is marked by the Trans-Canada Highway and the main line of the Canadian Pacific Railroad (CPR), both of which follow the Bow River from the Foothills, through Banff to the Kicking Horse Pass and its famous spiral tunnel, and then along the Yoho River to the west. This coincides with the northern limit of the Kootenay Ranges. The southern limit has been taken as the beginning of the Border Ranges, which were discussed previously. The slightly warmer climate south of Kicking Horse Pass means that generally this group of mountain ranges contains fewer and smaller ice fields and glaciers than those to the north, but there is ample evidence of previous glacierization.
The series of parallel ranges running from east to west have been grouped and are discussed according to the area within which they fallÑthe Front, Park, and Kootenay Ranges.

Front Ranges (South) (FS)
The Front Ranges (fig. 1, FS) constitute a series of parallel mountain ranges rising from the Foothills up to and beyond the Continental Divide that marks the provincial boundary. The outer, or eastern, group of ranges consists of the Blairmore, Livingstone, Highwood, Misty, Opal, and Fisher Ranges and ends in the north at the Rundle Peaks. The next parallel group of ranges is made up of the Flathead, Taylor, High Rock, Wisukitsat, Greenhills, Elk, and Kananaskis Ranges. The final section of the Front Ranges is made up of the Lizard Range, the Harrison, Italian and Joffre Groups, Spray Mountains, and the Sundance Range. It is this inner and higher set of ranges in which several glaciers are found, particularly in the Spray Mountains and in the Italian and Joffre Groups. The divide follows the central range as far as the Elk Range then swings over and passes out into the Park Ranges.

Lizard Range (FS1)
The Lizard Range (fig. 1, FS1) is the southernmost element of the Front Ranges and is situated on the western flank of the Rocky Mountains, south of the Kootenay Ranges. Elevations here are <2,300 m, and there are no glaciers.

Taylor Range (FS2)
Just to the west of the Continental Divide, the mountains in the Taylor Range (fig. 1, FS2) are somewhat higher in elevation, up to 2,445 m, but the orographic-precipitation regime is insufficient to sustain any permanent ice masses.

Flathead Range (FS3)
The elevation of the mountains increases within the Flathead Range (fig. 1, FS3), averaging about 2,500 m asl, but with higher peaks such as Mount Ptolemy (2,815 m) and Mount Darrah (2,745 m). Mount Ptolemy has a few rock glaciers in its vicinity (for example, Canadian glacier inventory glacier No.*4*5AA16 and 17), and there are some glacierets farther south in the headwaters of the Carbondale River. None, however, is shown on the published topographic maps.

Blairmore Range (FS4)
Moving toward the prairies and almost in the Foothills, elevations drop to close to 2,000 m in the Blairmore Range (fig. 1, FS4), and no glaciers of any size exist.

Livingstone Range (FS5)
Just to the north of the Blairmore Range, the Livingstone Range (fig. 1, FS5), a long, sinuous, limestone range, almost 100-km long, is the first of five major ranges that constitute the next group in the Continental Ranges. Some peaks, such as Centre Peak and Mount Burke, rise to about 2,500 m. Although no glaciers are visible on the Landsat images or any NTS maps, a group of four rock glaciers has been identified at the southern end of the range near the Blairmore Range.

High Rock Range (Tornado Group) (FS6)
Situated parallel to the Livingstone Range and to its west, the High Rock Range (fig. 1, FS6), also known as the Tornado Group, rises from 2,550 m in Crowsnest Mountain in the southern part of the range and extends past Tornado Mountain (3,100 m) to several 3,000-m peaks in the northern part. A small ice apron and rock glacier (*4*5BL5 and 6; Ommanney, 1989) occur at the foot of Mount Cornwell; they are remnants of the glaciers observed there during the 1916 survey (Interprovincial Boundary Commission, 1924). Small glaciers, glacierets, ice-cored moraines, and rock glaciers are mostly found at the foot of the higher peaks on the north- and east-facing slopes, though none are shown on NTS maps. This may explain DentonÕs (1975) conclusion that there was no glacier on Tornado Mountain.

Wisukitsat and Greenhills Ranges (FS7)
Two small ice-free ranges, the Wisukitsat and Greenhills Ranges (fig. 1, FS7), lie between the High Rock Range and the next major parallel feature of the Continental Ranges, to the west, the Harrison Group East. Both are fairly low, with peaks less than 2,700 m and 2,400 m in elevation, respectively, and are not known to be glacierized.

Harrison Group East (FS8)
The last major mountain block in this discussion of the Front Ranges (South) is the Harrison Group East (fig. 1, FS8), which is almost 100 km long and about 15 km wide. It is bounded to the east by Elk River and to the west by Bull River. Summit elevations are lowest in the southern part (<2,500 m) and rise to just over 3,000 m northward. No detailed studies have been made of the ice cover in this range, so it is not known whether glacierets and rock glaciers can be found here. It would seem likely, as well-developed glaciers are found just to the north around Mount Abruzzi in the Italian Group.

Highwood Range (FS9)
Moving farther north again, the next major eastern outlier of the Rockies is the Highwood Range (fig. 1, FS9). It has elevations varying from 2,782 m (Mount Head) to more than 2,800 m in the northern part of the range. Small glaciers are found at the headwaters of tributaries of Sheep River.

Misty Range (FS10)
Lying at the headwaters of Sheep River and cut off to the west by Highwood River, the Misty Range (fig. 1, FS10) rises to more than 3,000 m asl in Mist Mountain (3,138 m) and Mount Rae (3,219 m). The latter is the location of Rae Glacier [*4*5BJÐ4] (fig. 6) (Gardner, 1983), which was studied briefly by a group from the University of Saskatchewan (Lawby and others, 1994). There are three other glaciers near Rae Glacier, as well as many other small ones at the head of Mist Creek. Gardner (1983) refers to several small cirque and niche glaciers above 2,900 m in elevation on shaded and leeward slopes.

Elk Range (FS 11)
The Elk Range (fig. 1, FS11) is a northerly extension of the High Rock Range. Mean maximum elevations vary from 2,600 to more than 2,800 m. Again, small glaciers are scattered along the base of the range in sheltered north- and east-facing basins.

Italian Group (FS12)
The Italian Group (fig. 1, FS12) is really a continuation of the Harrison Group. It is centered on Mount Abruzzi, a major peak rising to 3,265 m asl, with other peaks exceeding 3,000 m in its vicinity. Many small glaciers are located here in sheltered north- and east-facing cirques. Most are about 1 km long, with snouts terminating around 2,450 m. Abruzzi Glacier (4 km2), which is the largest, is some 2 km wide, 2.5 km long, and descends to 2,500 m. This is the southernmost group of glaciers in the Canadian Rocky Mountains, if one relies solely on the existing topographic maps. However, glacier inventory studies have revealed small permanent ice masses all the way south to the U.S./Canada border. Just over the border in Glacier National Park, Montana, 37 named, small mountain glaciers have been documented within the park, and two outside. [See section ÒGlacier Retreat in Glacier National Park, Montana,Ó in the Glaciers of the Western United States (JÐ2) part of this volume.] Glacier National Park, not to be confused with the Canadian Glacier National Park in the Selkirk Mountains of central British Columbia, near Revelstoke, is one of two contiguous national parks, north and south of the border; the other is Waterton Lakes National Park, Alberta.

Joffre Group (French Military Group) (FS13)
Bounded to the east by Elk River and to the west by the Palliser River, the Joffre, or French Military Group (fig. 1, FS13), is a northwesterly extension of the Italian Group and has peaks rising to a maximum in Mount Joffre (3,449 m) and more and larger glaciers than in the previously discussed mountain ranges. The Mangin and PŽtain Glaciers are about 5 km2 in area and 4.5 and 3.5 km in length, respectively. The Foch Glacier, Elk, Castelnau, Lyautey and Nivelle Glaciers are 1.5 to 2 km in length, and there are several other unnamed glaciers nearby of comparable size. Only the Nivelle Glacier is located on the western side of the main range. The average lowest elevation of the ice varies from 2,400Ð2,600 m. The equilibrium line altitude (ELA) likely lies between 2,600 and 2,700 m.

Opal Range (FS14) 
The toe of an L-shaped range, the Opal Range (fig. 1, FS14), provides part of the initial buttress of the Rocky Mountains, which face the Foothills to the east, and has the upper part of the ÒLÓ tucked in behind the more northerly Fisher Range. Along with the following two ranges, it forms part of what has been called the Kananaskis Range. The western limits are marked by the broad valley of the Kananaskis River. There is a slight east-to-west gradient in maximum peak elevations from 2,900 m to slightly more than 3,000 m. The peaks to the west resemble the Sawback Range. The sharp and jagged peaks were created by erosion of the nearly vertical, steeply dipping beds of the Rundle Limestone. No glaciers are shown on the published maps, but the glacier inventory identified some small ice aprons, glacierets, and rock glaciers in parts of the range.

Fisher Range (FS15)
Fisher Range (fig. 1, FS15) marks the northern limit of the easternmost ranges of the Rocky Mountains considered in this part of the Continental Ranges. Elevations are similar to those in the Opal Range, rising to a maximum of just over 3,000 m asl at Mount Fisher, where the glacier inventory identified two small cirque glaciers, the only permanent ice here.

Kananaskis Range (FS16)
Although there are no glaciers large enough for skiing, this is the area of Mount Allan in the Kananaskis Range (fig. 1, FS16), which includes the area chosen for CanadaÕs winter Olympic downhill skiing events in 1988. The range lies across the Kananaskis River valley and is bounded on the west by Smith-Dorrien Creek with mountains rising slightly to just over 3,100 m asl. Numerous small ice aprons and cirque glaciers have been identified here, as well as several small rock glaciers.

Spray Mountains (British Military Group) (FS17)
Otherwise known as the British Military Group, after the association of peak and glacier names, the Spray Mountains (fig. 1, FS17) contain several glaciers along the divideÑa continuation of those found farther south in the Joffre Group. The largest, about 4 km2, 3 km long, and descending to 2,225 m, is the Haig Glacier. Beatty Glacier, just to the south of the Haig Glacier, is only about 1.5 km long and terminates a little higher at 2,380 m. Other glaciers vary in length from 0.5 to 3 km and have average snout elevations of about 2,500 m asl. The ELA in this range is estimated to be about 2,650 m on the east side and 2,670 m on the west side.

Rundle Peaks (FS18)
Paralleling the Trans-Canada Highway on its southern margin, as the highway enters the Rocky Mountains and curves northward toward Banff and Banff National Park, are the Rundle Peaks (fig. 1, FS18). Peak elevations are less than 3,000 m and include the Three Sisters above the town of Canmore, Alberta, and Mount Rundle, all of which afford the scenic backdrop to the Banff townsite from Mount Norquay. There are no glaciers in this range.

Goat Range (FS19)
West of the Rundle Peaks is the small Goat Range (fig. 1, FS19), stretching from the Spray Lakes Reservoir, which can be seen clearly on Landsat images, to the southern end of Sulphur Mountain in Banff. The lower part of the range has elevations comparable to those of the Rundle Peaks, approximately 2,900 m, but, unlike the latter, it does have several small cirque glaciers and ice aprons around Mount Nestor (2,960 m).

Sundance Range (FS20)
As viewed from space, the Sundance Range (fig. 1, FS20) looks like a giant tuning fork; it extends south-southeast from the Banff area before splitting into the two tines that reach to the head of the Spray Lakes Reservoir. Mean peak elevations vary from 2,800Ð2,950 m asl. Although not shown on NTS maps, the southern part, from Cone Mountain to Fatigue Mountain (2,959 m), contains numerous small glaciers (*4*5BC61Ð78, 84Ð91; Ommanney, 1989) that lie along benches and terraces eroded in the tilted strata. Some climbers include this range with the Assiniboine Park Group [and the Blue and Mitchell Ranges, as part of the Assiniboine Park Group] in the Park Ranges. Such trans-range mountain groups make the delineation of meaningful groups of mountain ranges extremely difficult, because differentiation may be based on physiographic, geologic, or other considerations.

Park Ranges (South) (PS)
The Park Ranges (fig. 1, PS) lie between the Kootenay and Front Ranges. They are generally higher and more heavily glacierized than the Front Ranges (South). The northern limit of this area is composed of the Ball, Vermilion, Bow, and Ottertail Ranges; the latter includes the Washmawapta Icefield. In the vicinity of Mount AssiniboineÑthe ÒMatterhornÓ of the Rocky MountainsÑare found the Assiniboine Park Group and Royal Group, as well as the Blue and Mitchell Ranges. In the southern part, the ranges narrow into the western part of the Harrison Group and terminate in the Quinn Range. Glaciers are concentrated in the northern part of the Park Ranges (South) and also around Mount Assiniboine.

Quinn Range (PS1)
The Quinn Range (fig. 1, PS1) is about 50 km long; it is nestled between the Harrison Group East of the Front Ranges and the Van Nostrand Range of the Kootenay Ranges. The range rises to a maximum height of 3,300 m asl at Mount Mike, and maximum elevations are generally close to 3,000 m, yet no glaciers are shown on the maps.

Harrison Group West (PS2)
The Bull and White Rivers separate the western and eastern sections of the Harrison Group, the Harrison Group East (fig. 1, FS8), and the Harrison Group West (fig. 1, PS2). The western part of the Harrison Group West is somewhat higher than the eastern part, having several peaks >3,000 m asl; Mount Harrison rises to 3,359 m. Despite the fact that Mount Harrison is higher than Mount Abruzzi and about 40 km farther north, there are apparently no glaciers in the vicinity.

The Royal Group (PS3)
The centerpiece of the Royal Group (fig. 1, PS3) is Mount King George (3,422 m), an impressive landform of towers and massive walls, but there are several other peaks that also rise above 3,000 m. The Princess Mary, King George, and Prince Albert Glaciers are located on the flanks of Mount King George. They vary in area from 0.5 to 1.5 km2 and have termini at about 2,450 m asl. More than 2 km long, the Tipperary Glacier is located on Mount Cradock; another small glacier (Albert Glacier) fills a cirque north of Mounts Queen Elizabeth and King Albert. Part of The Royal Group, with elevations ranging up to 2,950 m asl, lies to the west, separated from the rest of the group by the Albert and Cross Rivers.

Blue Range (PS4)
The small mountain block of the Blue Range (fig. 1, PS4) straddles the Continental Divide south of Mount Assiniboine, from which it is separated by Aurora and Owl Creeks. Elevations do not reach 2,900 m even along the border, and no glaciers are shown on current maps, even though glacier inventory work on the eastern end of the range revealed numerous cirques having small ice bodies and large moraines.

Assiniboine Park Group (PS5)
The dominant peak in the southern Rockies and the highest in the southern Continental Ranges, Mount Assiniboine (3,618 m) (fig. 3) rises as a majestic horn above its surrounding glaciers, which lie on shelves around the main mountain core, the Assiniboine Park Group (fig. 1, PS5). The Indian name means Òstone-boiler,Ó after the practice of using hot stones for cooking. Of the dozen glaciers on Mount Assiniboine, the largest is about 2.5 km2 in area. The average elevation of the glacier tongues ranges from 2,450 to 2,550 m. On the east side of the group, the equilibrium line probably lies at about 2,650 m. Rock-glacier forms in the area have been studied by Yarnal (1979). Although a popular stop for tourists and climbers, no scientific studies have been done on the glaciers in this area. The Assinboine Park Group is well-defined on all sides, limited by the Mitchell River and the Aurora, Bryant, and Owl Creeks.

Mitchell Range (PS6)
The main part of the Mitchell Range (fig. 1, PS6) lies along the east side of the Kootenay River south of the Simpson River. This part has peaks rising to >2,900 m. The western slopes of the Mitchell Range are deeply incised with well-developed cirque basins that probably show evidence of recent glaciation, if not some permanent ice. The range spreads out eastward in a series of four connected blocks ending at Simpson Ridge, where a small glacier (0.5 km2) is located beneath Nestor Peak; its snout is at 2,485 m.

Ball Range (PS7)
As one drives from Banff northward to the Kicking Horse Pass, the Ball Range (fig. 1, PS7) can be seen where it forms part of the eastern section of the Park Ranges. Limited on three sides by the broad valleys of the Vermilion and Bow Rivers, the Ball Range has peaks varying from 2,800 to 3,100 m that reach a maximum elevation at Mount Ball (3,312 m). The six glaciers plotted on the NTS maps are all located on this mountain. They are less than 1 km long and have lower elevations differing by as much as 1,000 m (2,057Ð3,050 m). The largest has an area of about 0.75 km2. Stanley Glacier (0.6 km2 in area) is the only one lying on the western side of the range.

Vermilion Range (PS8)
Not to be confused with the Vermilion Range of the Front Ranges, 45 km to the northeast, this Vermilion Range (fig. 1, PS8) forms a major ridge to the west of the Continental Divide between the Vermilion and Beaverfoot Rivers, rising to maximum elevations of >3,000 m. Although maps include the Washmawapta Icefield in the Ottertail Range, the Vermilion Range should probably be considered as ending at Wolverine Pass. Six glaciers lie in cirques along the eastern slope of the range, the largest being Tumbling Glacier about 1 km2 in area. Average snout elevations are at about 2,100 m asl and lengths vary from 0.2 to 1.5 km.

Bow Range (PS9)
The Bow Range (fig. 1, PS9) is the focal point of the visit of most tourists to Banff National Park. The range marks the northern limit of the area considered as the southern Continental Ranges. It is some 21 km long, 19 km wide, and has peaks that are amongst the highest of the mountain ranges discussed so farÑMounts Allen (3,301 m), Lefroy (3,423 m) and Victoria (3,646 m), and Deltaform (3,424 m) and Hungabee Mountains (3,492 m). Glaciers occur on both sides of the main range and include several debris-covered and rock-glacier forms. Many glaciers exceed 1 km in length. On the western side, the lowest elevation of glaciers is about 2,600 m asl, but several termini end at lower elevations; on the east, the glaciers tend to be at even lower elevations. The popular Lake Louise is dammed by an early Holocene moraine formed by the Victoria Glacier; it probably formed during the Eisenhower Junction glaciation, some 10,000 years before present (B.P.) (Kucera, 1976). Ironically, Moraine Lake is dammed by a rock slide from the Tower of Babel (Kucera, 1976) rather than by a moraine. Kucera (1976) and Gardner (1978b) provide popular accounts of the glaciers and landforms in this area. Several glaciers that have been the subject of specific studies and comments are discussed below.

Horseshoe Glacier
Horseshoe Glacier, with an area of 4.3 km2, is about the same size as Victoria Glacier (3.5 km2 in area) but has not been studied in detail. It is fed by snow avalanches from Ringrose Peak, Mount Lefroy, Hungabee Mountain, and The Mitre at the head of Paradise Valley. It extends some 1.3 km from 2,800 m to 2,220 m asl, through a heavily debris-covered ablation zone that makes up over half its area and retards its rate of retreat. The glacier was described by the Vauxes and Sherzer in the early part of the century (Sherzer, 1907, 1908; G. and W.S. Vaux, 1907b). The snout, which is an ice cliff, calves directly into a proglacial lake dammed by deposits left by the retreating glacier. The lake is fed by glacier meltwater from a terminus that has receded 945 m from its ÒLittle Ice AgeÓ maximum (Gardner, 1978b).

Wenkchemna Glacier 
Wenkchemna Glacier, located in the Valley of the Ten Peaks at the head of Moraine Lake, is 3.7 km2 in surface area, of which two-thirds is covered in moraine. Wenkchemna Glacier extends more than 4 km, from a number of independent ice streams that flow from the mountain wall below the Wenk-chemna Peaks at about 2,700 m to a tongue at 1,900 m asl. These streams turn down valley to create the 1-km-wide ice-and-debris tongue that is estimated to be from 30 to 100 m thick. The glacier surface is irregular and hummocky, and thaw pits have formed where surface lakes have penetrated the underlying ice and drained. The comparative inactivity of the glacier, and the effectiveness of the debris as a sediment trap, means that the meltwater stream is clear, and a delta is not forming at the head of Moraine Lake. Around the margin and terminus are arcuate ridges up to 3 m high that are characteristic of rock glaciers. The early observations made here by Sherzer (1907, 1908), by the Vaux family (G. and W.S. Vaux, 1907b; by G. Vaux, 1910; M.M. and G. Vaux, 1911), by Field and Heusser (1954), along with more recent observations by Gardner (1977, 1978a), all show very little change in the glacier limits since 1903. Gardner reported the glacier had thinned by up to 50 m. Although several people have observed the debris-covered ice terminus encroaching on the surrounding forest, Kucera (1976) is of the opinion that Wenkchemna Glacier is shrinking.

Victoria Glacier
Victoria Glacier (fig. 5), lying at the head of Lake Louise, is probably one of the most frequently photographed glaciers in the Rocky Mountains, although some visitors may not recognize the debris-covered tongue for what it is. The glacier is visible and easily accessible from Chateau Lake Louise by a good trail that passes beside the lake.
The continuous ice stream flows northward from Abbot Pass (2,923 m) for about 2 km before turning sharply to the northeast, where it degenerates into an indistinct debris-covered ice tongue after another 2 km. The Abbot Pass basin contains less than 20 percent of the accumulation area. The rest, some 1.8 km2 in area, lies in a broad apron that stretches from Popes Peak past Mount Victoria toward Abbot Pass and avalanches 300 m or more to form a reconstituted ice mass where the Abbot Pass ice stream changes direction. It is mainly an interrupted valley glacier; of its 3.5 km2 area, 24 percent is debris-covered. Lefroy Glacier (1.3 km2) flows from the basin between Mount Lefroy and The Mitre and is separated from Victoria Glacier by a band of moraine.
The earliest known record is a 1897 photo by William Hittel Sherzer (Collie, 1899). The following year, long-term studies were initiated by the Vaux family of PhiladelphiaÑGeorge Jr., William, and MaryÑwho carried out observations in 1898, 1899, 1900, 1903, 1907, 1909, 1910, and 1912 (G. and W.S. Vaux, 1901, 1907a, b, 1908; G. Vaux, 1910; M.M. and G. Vaux, 1911; M.M. Vaux, 1911, 1913; Cavell, 1983). These observations were interspersed with those of Sherzer, who returned to the area in 1904 and 1905 on behalf of the Smithsonian Institution (Sherzer, 1905, 1907, 1908). Studies in the inter-war years were sparse, apparently limited to surveys in 1931 and 1933 by the Alpine Club of Canada (Wheeler, 1932, 1934). The reasonably good historical record led to the selection of this glacier by the Calgary office of the Dominion Water and Power Bureau (DWPB) in 1945 for its network of glaciers being assessed for their contribution to runoff. The position of the terminus and changes in its areal extent were measured, and a set of plaques were placed on the ice surface to measure velocity (McFarlane, 1945, 1946a, 1947; McFarlane and May, 1948; Meek, 1948a, b; McFarlane and others, 1949, 1950; May and others, 1950; Carter, 1954). The surveys continued every year until 1950, then biennially until 1954, when the snout was so covered in debris that it was almost impossible to identify the toe (Ommanney, 1971). Recession from various surveys is shown in figure 9; the average value is about 13.5 m aÐ1. Average velocities, measured upstream of the junction with the Lefroy Glacier, are given in table 4.
Osborn (1975) reported on the penetration of the ÒLittle Ice AgeÓ moraines by advancing rock glaciers. Luckman and others (1984) extended their studies of the Holocene to this area, and some investigations have been initiated on sedimentation in Lake Louise (Hamper and Smith, 1983).

Cathedral Glacier
The Cathedral Glacier (0.84 km2 in area) is poised like a Sword of Damocles above one of CanadaÕs main transportation routes through the Rocky Mountains. Situated on Cathedral Mountain on the south side of Kicking Horse Valley, between approximately 2,410 m and almost 3,000 m elevation, Cathedral Glacier has, at least five times in 1925, 1946, 1962, 1978, and 1982, generated mudflows that blocked the Canadian Pacific Railroad (CPR) line and even buried the Trans-Canada Highway. Both Jackson (1979a, b, 1980) and Holdsworth (1984) speculated on the possible cause of these jškulhlaups. It is thought that the surface topography of the glacier, particularly a giant snowdrift ridge, permits development of a surface lake at 2,960 m asl, which is recharged by normal snowmelt and rain. A pulse discharge through the glacier and down a narrow ravine picks up speed and unconsolidated till to create a mudflow that heads towards the CPR tracks in the spiral tunnel section and the Trans-Canada Highway. The regional firn line at 2,890 m asl is situated below the lake level. This is one of two major glacier hazards identified in the Rockies, the other being associated with Hector Glacier and Peyto Glacier.

Ottertail Range (PS10)
The Ottertail Range is a northwesterly extension of the Vermilion Range and is similar to it in many respects. The highest peak is Mount Goodsir at 3,562 m. Glaciers in the northern part of the range, including the Hanbury Glacier, Goodsir Glacier, East Goodsir Glacier, and Sharp Glacier, also lie in east- and north-facing basins, except for the West Washmawapta Glacier and Washmawapta Icefield, which fills a large basin below Limestone Peak. The Washmawapta Icefield is only about 4 km2 in area and hardly warrants its name. Glaciers terminate at about 2,500 m asl on this side of the range but are generally lower (some 2,300 m) on the eastern side. Hanbury Glacier is about the same size as the Washmawapta Icefield.

Kootenay Ranges (K)
South of Kicking Horse Pass, the Kootenay Ranges (fig. 1) form the western limit of the Rocky Mountains. The physiographic transition farther westward is very sharp, because the mountains drop down steeply into the Rocky Mountain Trench and the Kootenay and Columbia River systems. The range is narrow in the north, where it consists of the Beaverfoot (fig. 1, K5), Brisco (fig. 1, K4), and Stanford Ranges (fig. 1, K3), having peak heights varying from 2,500Ð2,700 m. Farther south it broadens into the 90-km-long Hughes Range (fig. 1, K1), which joins Lizard Range to form the southwestern limit of the Continental Ranges. The mountains here are slightly higher, up to 2,800 m, and are higher still in the eastern, parallel Van Nostrand Range (2,905 m) (fig. 1, K2). However, none of the topographic maps show any evidence of perennial ice features in the Kootenay Ranges.

Continental Ranges (CentralÑKicking Horse Pass to Yellowhead Pass)
This central region, extending 300 km between Lake Louise and Mount Robson, is the most heavily glacierized part of the Continental Ranges. The mountains in this region trend northwest to southeast and are divided into the Front Ranges and the Park (or Main) Ranges (fig. 1). The former consist of Devonian and younger limestones, dolomites, and shales, all of which dip steeply to the west, whereas the latter are made up of nearly horizontal beds of Cambrian and older quartzitic sandstones, limestones, dolomites, and shales (Gardner, 1972). At the northern end is Mount Robson (3,954 m), the highest mountain in the Rockies. Most peaks in the Front Ranges have elevations between 2,800 m and 3,280 m asl, whereas those in the Park Ranges vary in elevation between 3,125 m and 3,600 m. Local relief is generally on the order of 1,400 m to 1,900 m.
The ranges become progressively drier from west to east toward the prairies. Hence the eastern ranges do not contain many glaciers or ice fields. Westward, the first large glacier is the Bonnet Glacier, on the east side of Castle Mountain and hidden from the view of the average tourist by the 18-km-long bulk of that impressive mountain. The glaciers take a variety of forms, including ice fields such as the Columbia Icefield, outlet glaciers such as the Athabasca and Saskatchewan Glaciers, valley glaciers such as Robson Glacier, and mountain glaciers of the cirque, niche, and ice-apron type such as the Angel Glacier. Located north of Kicking Horse Pass, the Waputik Icefield is the first of a long series of ice fields that straddle the Continental Divide (figs. 7, 8). The second is the Wapta Icefield, source of the Yoho and Peyto Glaciers, followed by the Campbell and Freshfield Icefields, which send a magnificent valley glacier to feed the Howse River. Next come the Mons and Lyell Icefields, southeast of the largest ice field in the Rocky Mountains, the Columbia Icefield (fig. 4). This last ice field culminates in the Snow Dome, which is the hydrographic apex of the continent and drains into three oceans. Located immediately northwest of the Columbia Icefield, the Continental Divide is capped by the Clemenceau, Chaba, and Hooker Icefields (fig. 4). The only ice fields off the main divide are the Wilson Icefield and Brazeau Icefield, which lie just east of the valleys of the North Saskatchewan and Athabasca Rivers.
The glaciation level is lowest in the western parts of the Park Ranges and through the Yellowhead Pass (about 2,600 m asl) but rises in the vicinity of Mount Robson to more than 2,900 m and climbs eastward toward the Front Ranges to more than 3,100 m (¯strem, 1973b).
In this part of the Continental Ranges, moraines of the ÒLittle Ice AgeÓ are evidence of the most significant regional Holocene glacial event in the Rocky Mountains. The best developed moraines are those from the early 1700Õs, when about one-third of the glaciers showed maximum advance, and from the mid- to late-nineteenth century, when major readvances built moraines close to or beyond that earlier extent (Luckman, 1986).
Rock glaciers are distributed throughout the area. Luckman and Crockett (1978) reported on those in the southern half of Jasper National Park. The 119 rock glaciers identified range in area from 0.035 km2 to 1.57 km2 and lie between 1,710 and 2,670 m asl; that is 400 to 600 m below the glaciation level. Their distribution seems to be controlled strongly by lithology, and they are predominantly oriented to the north. In Banff National Park, Papertzian (1973) found no evidence for lithological control of the 80 rock glaciers there. They range in area from 0.011 km2 to 1.26 km2 and lie between 1,737 and 2,743 m asl. ¯strem and Arnold (1970) mapped both rock glaciers and ice-cored moraines in southern British Columbia and Alberta without distinguishing between them. An intermediate form that is quite common is the debris-covered glacier, such as Dome Glacier.
Descriptions of Rocky Mountain glaciers date from the time when Athabasca Pass was a major fur-trade thoroughfare, and they became more common once the Canadian Pacific Railroad line was completed. Despite the fairly long history of formal and informal glacier study, data for the whole range are sporadic. This is especially true for glaciers of the Front Ranges and for ice masses away from the main transportation routes such as the Clemenceau Icefield. However, virtually all of the detailed information on glaciers in the Rockies comes from the central Continental Ranges, primarily through investigations at the Peyto, Saskatchewan, and Athabasca Glaciers.

Front Ranges (Central) (FC)
The Front Ranges (Central) (fig. 1, FC) are a continuation of the series of parallel ridges described from the region to the south, although the parallelism may be less pronounced. Mountain outliers are found to the east in the Bighorn (fig. 1, FC12), Nikanassin (fig. 1, FC19), and Fiddle Ranges (fig. 1, FC20). The southern section, starting with the Fairholme Range (fig. 1, FC1) just north of the Bow River, continues northwest of Lake Minnewanka, with the Ghost River area bordering the Foothills, and extends westward through the Palliser (fig, 1, FC4), Bare (fig. 1, FC5), Vermilion (fig. 1, FC6), Sawback (fig. 1, FC7), and Slate Ranges (fig. 1, FC8). Between the Red Deer River and the North Saskatchewan River, the Front Ranges are more blocky in outline, with only two large groups, Clearwater (fig. 1, FC3) and Murchison Groups (fig. 1, FC9), and the Ram Range (fig. 1, FC10) as a northeasterly outlier. Between the Ram Range and the Athabasca River are a series of mountain blocks and groups whose boundaries are not very distinct. From south to north these are Cline Range (fig. 1, FC11), First Range (fig. 1, FC13), the Cataract Group (fig. 1, FC14), Le Grand Brazeau (fig.1, FC15), the Southesk Group (fig. 1, FC16), Queen Elizabeth Ranges (fig. 1, FC17), and Maligne Range (fig. 1, FC18), with the parallel Colin (fig. 1, FC23), Jacques (fig. 1, FC22), and Miette Ranges (fig. 1, FC21) rounding out this section in the north. The characteristics of the ranges and their glaciers will be discussed in turn.

Fairholme Range (FC1)
The Fairholme Range (fig. 1, FC1) is a large S-shaped feature that is clearly visible in satellite images. It extends some 25 km from where the Trans-Canada Highway passes through the towns of Exshaw and Canmore to Lake Minnewanka. Several peaks lie between 2,800 and 3,000 m asl, but the range is essentially free of ice.

Ghost River Area (FC2)
The Ghost River area (fig. 1, FC2), including the irregular mountain mass lying north of Lake Minnewanka and east of the Ghost River, has a few tiny cirque glaciers around Mount Oliver. These are the easternmost glaciers in the central Front Ranges. Elevations range to a maximum of slightly more than 2,900 m.

Clearwater Groups (FC3)
Probably the largest mountain block in this section of the Continental Ranges is the Clearwater Groups (fig. 1, FC3). They adjoin the northern part of the Ghost River Area and extend northwestward, encompassing the headwaters of the Clearwater River to end where the North Saskatchewan River cuts through the Front Range. The Clearwater Groups are bounded to the west by the Pipestone and Siffleur Rivers, and most of the glaciers are concentrated in the part of the group just to the east of these two river valleys. In the southwestern sector is found Drummond Glacier. There are about a dozen glaciers in the headwaters of the Clearwater River, which range in area from 0.5 km2 to 3 km2 and which generally terminate at elevations between 2,450 and 2,550 m. The equilibrium line altitude (ELA) probably lies between 2,600 and 2,650 m asl. Average peak elevations rise to well over 3,000 m, with a maximum at 3,373 m. To the north, the glaciers flow into the Escarpment and Ram Rivers, the latter having as its source the glacier of the same name.

Drummond Glacier
About 8.5 km2 in area and 3.5 km in length, Drummond Glacier flows from just over 3,000 m to 2,375 m; it is the largest glacier in the Clearwater Groups. It is part of a small ice field about 13 km2 in area just west of Mount Drummond at the headwaters of the Red Deer River. Historical photographs taken in 1884, 1906, 1917Ð20, 1930, and 1939 and 1963, were used by Brunger and others (Brunger, 1966; Nelson and others, 1966; Brunger and others, 1967) to reconstruct its recession (fig. 9). The University of Calgary group also made measurements of ablation and surface movement from 1962 to 1965.

Ram River Glacier
The Ram River Glacier (fig. 10) was the most easterly, and hence most continental, of the five selected by the Canadian Government as representative glacier basins for the International Hydrological Decade (IHD) in an east-west transect of the cordillera. It was the smoothest, the smallest (1.89 km2), the most compact, and, because of a mean elevation of 2,750 m, at the highest elevation of those studied. Possibly because of this, it is also the least dynamic. It lies in a cirque dominated by high, steep cliffs. A standard mass-balance measurement program was carried out here from 1965 to 1975, and the results (fig. 11) have been published (Young and Stanley, 1976a). A base map at a scale of 1:10,000 was published in 1967. The mean specific winter balance for the decade was 0.88 m water equivalent (w.e.), the annual balance was Ð0.43 m, and the mean equilibrium line was 2,838 m asl.

Palliser Range (FC4)
Surrounded by Lake Minnewanka and Cascade and Ghost Rivers, and south of Red Deer River, Palliser Range (fig. 1, FC4) is a westerly extension of the Ghost River Area. The maximum elevation in the range is 3,162 m at Mount Aylmer, with other peaks ranging from 2,900 to 3,100 m, which is roughly the elevation of the glaciation level. Three small glaciers have been identified in the central part of the range.

Bare Range (FC5)
Slightly lower in height, 2,750Ð2,950 m, the Bare Range (fig. 1, FC5) is a 20-km extension of the Palliser Range through which flows the Panther River. Because it is below the regional glaciation level, no glaciers have been observed here.

Vermilion Range (FC6)
The Vermilion Range (fig. 1, FC6) lies west of Cascade River; it is bounded to the south by Banff and to the north by the Red Deer River. Elevations through its 60-km extent are on the order of 2,850Ð2,950 m. Just south of Prow Mountain is a cluster of small glaciers along the headwaters of a tributary of Red Deer River.

Sawback Range (FC7)
Farther westward, the last major mountain block in this transect before reaching the Park or Main Ranges, is the Sawback Range (fig. 1, FC7). It adjoins the Vermilion Range on the east and has a sharp boundary on the west created by the Bow River valley, through which runs the Trans-Canada Highway. To the north it is limited by the Red Deer River valley. Elevations increase here above the regional glaciation level to more than 3,200 m in Mounts Douglas and St. Bride and in Bonnet Peak. Ice cover lies predominantly along the east side of the range in a series of almost continuous masses north of Bonnet Peak. The largest of these is Bonnet Glacier, 3.3 km long and about 6 km2 in area with a snout at 2,560 m asl. Other glaciers from 0.5 to 1.2 km in length also terminate below 2,600 m asl.

Slate Range (FC8)
Bounded by Pipestone River and Baker Creek, the Slate Range (fig. 1, FC8) lies at the northwest corner of the Sawback Range. Peaks rise to slightly more than 3,000 m in elevation. The ice cover is concentrated in a small ice field around Mount Richardson. Glaciers range from 0.5 to 1.0 km in length, and their lower limits are from 2,410 to 2,570 m asl.

Murchison Group (FC9)
The Murchison Group (fig. 1, FC9) is more than 50 km long and generally 10 km wide, although becoming broader northward. It is bounded on the west by the broad valley of the Mistaya and Bow Rivers, on the east by the Siffleur and Pipestone Rivers, and in the north by the North Saskatchewan River trench. Many peaks rise to more than 3,000 m asl, and the range tends to climb northward to a maximum of 3,210 m in Mount Loudon, well above the regional glaciation level. About 20 glaciers are found here. They average just over 1 km in length, with termini at 2,500 m asl. The largest are the Hector and Molar Glaciers, 3.0 km and 2.0 km long, respectively. The ELA probably lies at about 2,750 m asl. This range includes the smallest named ice field in the Rocky Mountains, the Murchison Icefield, which is about 0.3 km2 in area.

Hector Glacier
Hector Glacier lies in the southern part of the Murchison Group. It flows from Mount Hector, at about 3,350 m, northward for 3 km to 2,430 m asl. In the 1960Õs, this glacier was heavily crevassed and split into several tongues. Brunger (1966) and Brunger and others (1967) used historical photographs to reconstruct the recession of the western snout (fig. 9). In the late summer of 1938, a large ice mass separated from the glacier and fell into the Molar Creek valley, uprooting trees and destroying everything in its path. The glacier traveled more than 3 km and spread a broad carpet of ice over the valley up to 60-m-deep. Old-timers in the district reported no similar occurrence during the previous 40 years (B.C. Mountaineer, 1939). This represents the second major known glacier hazard in the Rocky Mountains, in addition to the repeated mudflows from the Cathedral Glacier, previously discussed.

Ram Range (FC10) 
Guarding the southern side of the pass through the Front Ranges created by the valley of the North Saskatchewan River, Ram Range (fig. 1, FC10) is a small, L-shaped block on the northern end of the Clearwater Group. Elevations average about 2,500 m with a maximum of 2,844 m. A few tiny ice masses are found in sheltered north-facing valleys.

Cline Range (FC11)
Bounded by the North Saskatchewan River to the south and the Cline River to the north, Cline Range (fig. 1, FC11) contains numerous small cirque glaciers, generally less than 1 km2 in area and terminating on average at about 2,400 m asl. The Wilson Icefield, 12 km2 in area, which rises to 3,261 m on Mount Wilson, lies above the regional glaciation level at 3,000 to 3,100 m in elevation. It is located in the southwestern corner of the range and contains about a dozen outlet glaciers, which range in length from 2 to 4 km and have termini that flow down to 2,000 m. The highest peak is Mount Cline (3,361 m), with small glaciers on its north and south slopes. One of these was, for a short time, subject to some commercial exploitation as the Ice Age Company mined it for ÒpureÓ freshwater and ÒgourmetÓ ice for sale in Alberta (Brugman, 1989; Rains, 1990). An application to expand the operation led to an environmental impact assessment (Ice Age Co., 1989) and non-renewal of the mining licence. Although there are several other peaks with elevations of more than 3,000 m, average peak elevations are about 2,800 m.

Bighorn Range (FC12)
The Bighorn Range (fig. 1, FC12) is about 5 km broad at its widest part and 44 km long. It lies well outside the main body of the Rockies and could be considered part of the transition to the Rocky Mountain Foothills. It has no glaciers, and its peaks average less than 2,500 m in elevation.

First Range (FC13)
The next major mountain block in the Front Ranges is the First Range (fig. 1, FC13). The Brazeau River marks its northern boundary and the Cline River its southern one. The 40x25 km block is cut by Job and Coral Creeks, with the section to the west being known as the Job Creek Peaks. Peak elevations average between 2,750 and 2,900 m asl, rising to a maximum of 3,150 m. Detailed aerial photographic analysis in this area revealed more than 40 small glaciers in the western section, with a particularly heavy concentration at the head of Job Creek. About the same number have been plotted in the main First Range. Most lie in the headwaters of the Bighorn River and Littlehorn Creek and in the eastern basin of Job Creek. The glaciers are too small to see on Landsat images and are not shown on current topographic maps.

Cataract Group (FC14)
Nestled in behind Job Creek Peaks, and divided from the Park Ranges by the valley of the North Saskatchewan River, is the Cataract Group (fig. 1, FC14). Peak elevations here average more than 3,000 m and have a maximum in Mount Stewart (3,312 m). Increased elevation and proximity to the source of moisture mean that the area covered by glaciers is now a little denser. Almost 70 small glaciers can be found in this group, of which the Huntington (1.5 km2) and Coleman Glaciers (2 km2) are amongst the largest. The regional equilibrium line altitude is thought to lie at about 2,550 m asl. Large proglacial moraine areas and rock glaciers are common.

Le Grand Brazeau (FC15)
Stretching some 50 km from the Rocky Mountain Foothills to the Park Ranges is the mountain block referred to as Le Grand Brazeau (fig. 1, FC15), not be to be confused with the Brazeau Range, which is a small feature 75 km to the east in the Foothills. Officially the name is applied only to that part of the mountain block centered on Poboktan Mountain (3,323 m), but climbers have used the wider application. Peaks rise to more than 3,000 m asl with several reaching about 3,200 m. The regional glaciation level is at about 2,900 m. As in the mountain group to the south, glacier density increases westward, and there is a predominance of rock glaciers, debris-covered glaciers, and large expanses of proglacial moraine. West of Poboktan Mountain, the glaciers drain to the Arctic Ocean, and to the east they contribute to the Nelson River system that flows into Hudson Bay. Most of the 25 larger ice bodies terminate between 2,350 and 2,500 m asl. Their lengths range from 1 to 2.5 km, but some extend to 3.5 km, and one is 4.2 km long. Cornucopia Glacier is probably typical of those in the area; it is 2.5 km long and has a snout at 2,550 m asl. It forms part of Brazeau Icefield, which lies at the junction with the Queen Elizabeth Ranges and is the largest ice field in the Front Ranges, having an area of 40 km2.

Southesk Group (FC16)
North of Southesk River and south of Rocky River lies the 25- by 30-km-wide Southesk Group (fig. 1, FC16). There are a few tiny glaciers in the headwaters of Ruby and Thistle Creeks and the Cairn River. To the west, glaciers become larger, and three are more than 1 km in length, terminating at about 2,400 m asl. Peak elevations tend to lie below 3,000 m, although Mount Balinhard rises to 3,130 m and is the site of North Glacier.

Queen Elizabeth Ranges (FC17)
Extending northwestward from Brazeau Icefield for more than 50 km are the Queen Elizabeth Ranges (fig. 1, FC17). Whereas the largest ice-covered area is that around Mount Brazeau (3,470 m), others are found around Maligne Mountain (3,193 m) and Mount Unwin (3,268 m). Coleman (1903) visited the area in 1902 and described the ice field as rising into two white mounds to the south and sinking away to dirty surfaces of ice in the valleys to the east. All glacier tongues show signs of recession. Kearney (1981) dated the moraines here and in the vicinity of Mary Vaux and Center Glaciers. Peak elevations decline northward from more than 3,200 m to about 2,500 m. As a result, most of the glaciers are found around the upper part of Maligne Lake. Almost 20 glaciers average 1 to 2.54 km in length and have lower snout elevations of about 2,300 m asl. Coronet Glacier (3.5 km long), an outlier of Brazeau Icefield, is one of the largest, exceeded only by the 5-km-long glacier flowing north from Mount Brazeau.

Maligne Range (FC18)
Somewhat lower and lying between the Maligne and Athabasca Rivers, the Maligne Range (fig. 1, FC18) extends for more than 60 km in a northwestward orientation and marks the western limit of this section of the Front Ranges. Average peak elevations are about 2,600 m. About two dozen small cirque glaciers and ice aprons are located in sheltered north- and east-facing basins, the largest being those on the slopes of Mount Kerkeslin (2,956 m).

Nikanassin Range (FC19)
The ranges in the northern section of the central Front Ranges (FC) begin to decline in height and break up into a series of more isolated, parallel ridges in the region of the Nikanassin Range. Elevations in the Nikanassin Range (fig. 1, FC19) are less than 2,500 m and there are no glaciers.

Fiddle Range (FC20)
An extension northward of the Nikanassin Range, Fiddle Range (fig. 1, FC20) has an even lower elevation and likewise no glaciers.

Miette Range (FC21)
West of and parallel to the Fiddle and Nikanassin Ranges lies the Miette Range (fig. 1, FC21). It is slightly higher than these two ranges, rising to a maximum of 2,795 m asl. Some tiny permanent ice masses may exist in north-facing cirques, but all would be too small to be visible from space or to be shown on topographic maps.

Jacques Range (FC22)
To the west of the Miette Range lies the Jacques Range (fig. 1, FC22), which has an unnamed extension of the range to the southeast. Mountain elevations in the Jacques Range are comparable to those in Miette Range; it is unlikely that there are any glaciers here.

Colin Range (FC23)
The northwesternmost parallel range in this transect of the Front Ranges is the Colin Range (fig. 1, FC23). Mountain elevations here rise to 2,600 m asl. No glaciers are plotted on any of the topographic maps, and because the highest peaks lie below the regional glaciation level, it is not expected that any glaciers will be found here.

Park Ranges (Central) (PC)
The central section of the Park, or Main, Ranges (fig. 1, PC) consists of three more-or- less parallel sets of mountains between the valley of the Athabasca River on the east and the Rocky Mountain Trench on the west. The southern part of the central section (PC) is separated from the southern section (PS) by Kicking Horse Pass with the Canadian Pacific Railroad (CPR) and the Trans-Canada Highway. The northern limit of the central section is marked by Yellowhead Pass, which is the route of the Canadian National Railroad (CNR) from Edmonton to the west. This part of the Park Ranges contains the greatest concentration of glaciers in the Rocky Mountains, including all of the main ice fields (figs. 7, 8). Moving northward through the inner chain, one passes the Waputik Mountains, with the Wapta and Waputik Icefields; the Conway, Barnard Dent, and Forbes Groups, with the Freshfield Icefield, Campbell Icefield, and Lyell Icefield; the Columbia Icefield and the Winston Churchill Range, and thence through the Fryatt, Cavell, and Portal-MacCarib Groups to the Trident Range. The central chain includes the Van Horne Range, the Chaba and Clemenceau Icefield Groups with their extensive ice covers, and the Whirlpool, Fraser-Rampart, and Meadow-Clairvaux Groups. The westernmost chain is largely unnamed, apart from the large block of the Selwyn Range at the northern end. All are discussed below, with particular emphasis being given to the areal coverage of glacier ice and those glaciers that have been studied in the most detail.

Waputik Mountains (PC1)
Extending northward from Kicking Horse Pass is a triangular, elevated area of peaks and ridges bounded on the east by the Bow and Mistaya Rivers and the mass of the Front Ranges, and on the west by the valleys of the Amiskwi and Blaeberry Rivers. The Waputik Mountains (fig. 1, PC 1) contain the subsidiary President and Waputik Ranges as well as the two southernmost major ice fields of the Rockies, the Waputik and Wapta Icefields (figs. 7, 8). An inventory of the glaciers in this region was completed by Stanley (1970) as a pilot study for the world inventory of perennial snow and ice masses. He found more than 100 glaciers that covered an area of 146 km2. They ranged in elevation from 2,100 to 3,200 m and had an average snowline in the vicinity of 2,400 m. The Waputik Range lies east of the Waputik Icefield (fig. 8) between Bow River and Bath Creek. It rises to about 2,750 m and contains one major ice mass from which drains the Waputik Glacier (3 km2). Just west of the Yoho River is the President Range. Peaks here are on the order of 3,000 m asl, and a number of small glaciers are nestled around them, including the Emerald Glacier and the President Glacier. The Emerald Glacier is a small ice apron whose northern section is almost detached; it covers an area of only 0.6 km2. Part of the tongue has a continuous supraglacier debris cover and part is relatively debris free. Batterson (1980), Rogerson and Batterson (1982), and Rogerson (1985) determined rates of advance for the push moraine of Emerald Glacier (fig. 9).
President Glacier flows from the north slope of The President toward Little Yoho River. Its present snout is at 2,353 m asl, but it formerly extended downslope an additional 2 km. The President Glacier had one advance about 1714 and a second about 1832 (Bray, 1964). In 1937, McCoubrey (1938) noted a marked shrinkage of ice (32Ð36 m) on the left side of the glacier as compared to the earlier survey by Roger Neave in 1933 (Wheeler, 1934). Later, Bray (1965) discussed the relationship between solar activity and glacier variation here.

Waputik Icefield
The main feature in the southern section of these mountains is the Waputik Icefield (figs. 7, 8, and 12). It is some 53 km2 in area, straddles the Continental Divide between Mount Bosworth and Balfour Pass, and is the source of a number of quite large glaciers. The Waputik Icefield can easily be identified on Landsat images (fig. 7). Bath Glacier (4.3 km2) has a large ice apron that extends 7 km southward from the slopes of Mount Daly and can be seen from the Trans-Canada Highway. Most of the ice field drains westward through Daly Glacier (13.7 km2 in area) with much of the remainder flowing northward in Balfour Glacier (5.9 km2 in area). Glaciers terminate at about 2,100 m, and the equilibrium line altitude lies close to 2,450 m asl.

Balfour Glacier
At the time of the initial photographs by Wilcox (1900), Balfour Glacier (fig. 13) was a compound valley glacier draining about 14 km2 of the northeast sector of the Waputik Icefield (figs. 7, 8). The glacier is now about half that size (8.8 km2). Its main stream is a northwesterly flowing mass that no longer coalesces with the glaciers draining the ice aprons north and east of Mount Balfour. McFarlane (1945) did not include this glacier in the Dominion Water and Power Bureau (DWPB) network because of the high cost of visiting it.
Studies of proglacial Hector Lake established a chronology of proglacial-lake sedimentation back to 1700 (Smith, 1978; Leonard, 1981, 1985; Smith and others, 1982). Most sediment in the lake is provided by nival and glacial meltwater from Balfour Creek. Sediment input varies with inflow discharge and is controlled mainly by melting rates (Smith, 1978).
Leonard (1986a, b) documented the changing glacial outwash sedimentation and glacial activity over a period of about 1,000 years. He concluded that the very regular rhythmic laminations were indeed true varves and could be correlated with the climate record from Lake Louise. Multiple cores were used to assess lake-wide sedimentation characteristics likely related to changes in total sediment input. The maximum ice stand occurred about 1847. Recession rates averaged about 10 m aÐ1 from the late 1840Õs to 1900 and then increased fourfold from 1900 to 1948. Since 1948, recession has been almost negligible. The major moraine-building episodes of 1700Ð1720 and 1840Ð1860 were periods of persistent high sedimentation rates as was that of the very rapid ice recession period, 1910Ð1945. The two earlier periods were probably because of high glacial erosion rates and the later period almost certainly due to high sediment availability.

Wapta Icefield
Wapta Icefield (fig.8) lies northwest of the Waputik Icefield; it is about 80 km2 in area and has been one of the focal points of glaciological research in the Rocky Mountains. It is drained on the Alberta side by the Vulture (4.9 km2 in area), Bow (5.1 km2 in area) and Peyto (12.6 km2 in area) Glaciers and, on the British Columbia side, by the Yoho (20.9 km2 in area) and Ayesha (3.2 km2 in area) Glaciers, as well as by the Glacier des Poilus (12.8 km2 in area). The Glacier des Poilus is the southwestern extension of the Wapta Icefield and occupies two large basins. It flows from almost 3,000 m down to 2,240 m and has an equilibrium line altitude at about 2,450 m asl. Bigras (1978) reported on the sediment transport, discharge, volumetric change, and geomorphology of this glacier. He concluded that it had lost 4.5x106 m3 of ice during the last century. Average recession since 1951 is shown in figure 9.
Crowfoot Glacier (1.5 km2 in area), part of a separate outlier glacier (5 km2 in area) which lies east of the Wapta Icefield (figs. 7, 8), is frequently photographed by motorists traveling on what Parks Canada now refers to as the Icefields Parkway. Leonard (1981) investigated lichen growth curves here and found them to be 35 percent lower than those reported by Luckman (1977), thus casting doubt on some regional growth curves. Elevations of mountain peaks rise to more than 3,100 m, and the regional ELA lies at about 2,440 m. Northward, the range narrows. There are significant ice accumulations around Mistaya Mountain, namely, the Delta (2.2 km2 in area), Barbette (4.5 km2 in area) and Parapet (1.1 km2 in area) Glaciers, and also around Mount Sarbach (1.5 km2 in area). Howse Peak (3,289 m), the highest in these mountains, is located in this part of the Park Ranges (Central).

Bow Glacier
Bow Glacier (fig. 14) flows northeast for 3 km from the Wapta Icefield (figs. 7, 8) and begins to descend steeply from about 2,600 m asl. Its terminus rests at just over 2,300 m asl, about 200 m above the recent maximum position. The lower part of the glacier is quite heavily crevassed where the ice flows out of the ice field over a series of ridges. The popular Num-Ti-Jah Lodge, just off the Icefields Parkway, affords an excellent view of the glacier across Bow Lake.
The glacier was first visited in 1897 (Stutfield and Collie, 1903) when the ice extended below the base of the main cliff. The terminus remained there until 1922 before retreating above the base of the cliff by 1933 (Wheeler, 1934). Further retreat from the 1933 position appears to have been comparatively minor. The glacier was photographed in 1945 by the DWPB but not investigated further, as it was then hanging over the cliff (McFarlane, 1945). Heusser (1956) concluded that the glacier retreated 1,100 m from 1850 to 1953. Aerial photographs taken in 1952 show a narrow lake developing between the glacier and cliff edge. By 1966, only a quarter of the present-day lake was visible, and by the 1980Õs, the entire lake could be seen and the glacier tongue was a few meters above it (Smith, 1981).
A number of studies have been carried out on sedimentation in Bow Lake and the reduction in sediment input caused by the development of the proglacial pond (Kennedy, 1975; Leonard, 1981, 1985; Smith, 1981; Smith and Syvitski, 1982; Smith and others, 1982). Leonard (1981, 1985) was able to explain up to 70 percent of the sedimentation rate variance when he compared the Bow Lake varve record to the ablation-season temperature record from Lake Louise. The sediment load carried by the stream flowing from the glacier was substantial enough to create a large delta at the western end of Bow Lake. Church and Gilbert (1975) have looked at the long profile and mean clast size of the outwash fan. In 1997 a group from the University of Alberta installed an automatic weather station at Bow Glacier and commenced a continuing study of sedimentation and runoff chemistry (M.J. Sharp, oral commun., 2000).

Peyto Glacier 
The first records of Peyto Glacier (fig. 15) were photographs taken by Wilcox in 1896 and by Thorington in 1923. Recession since 1923 was recorded by Thorington, Kingman, Dickson, and Vanderburg in 1933 (Wheeler, 1934) and again by Kingman in 1936 (McCoubrey, 1938). In 1945, it was selected as a representative glacier by the DWPB, and for the next 17 years the position of the snout, changes in the areal extent of Peyto Glacier, and ice velocities were measured (fig. 9). The glacier gradually retreated into a narrow gorge, which made it less representative. After 1962, the survey was terminated. Detailed reports, prepared by the DWPB as internal documents, were summarized by Ommanney (1972b), and some results have been published (McFarlane, 1946a; Meek, 1948b; McFarlane and others, 1950; Collier, 1958).
Heusser (1954, 1956) used photographic and botanical techniques to reconstruct the recent history of Peyto Glacier. The recessional moraine was dated to 1863 and four other moraines were identified. Retreat from 1865 to 1953 was determined to be about 1,009 m. Brunger and others (1967) concluded that an ecesis interval (colonization of flora and fauna in deglacierized terrain) of 25 years was more appropriate than the 12-year interval used by Heusser. They found glacier response lagged climate by about 15 years, close to the 20-year lag estimated by Collier (1958).
In 1965, Peyto Glacier was selected as one of the representative International Hydrological Decade (IHD) glacier basins in the Rocky Mountains by the Canadian Government. The 13.4 km2 glacier originates at an elevation of 3,185 m asl; its tongue is at about 2,100 m asl, and the mean elevation is 2,635 m. The measurement program was described by ¯strem and Stanley (1969). Young (1976) modified this methodology and proposed a grid square technique to produce accumulation, ablation, and mass-balance maps using associations between snow depth and terrain geometry (for example, surface slope, azimuth, and local relief). The technique could be used to assess the recurrence patterns of accumulation, bias in sampling networks, effects of different sized sampling networks, progress of melt throughout an ablation season, as well as being able to extrapolate to unvisited parts of the glacier (Young, 1974a). The assumption of a linear function for snow accumulation produced results that seemed realistic but were substantially different from those derived from the normal stake network (Young, 1974b).
A report on the glaciological, hydrological, and meteorological data collected for the IHD program (1965 to 1974) has been published (Young and Stanley, 1976b). A plot of the mass balance, including more recent data (Mokievsky-Zubok and others, 1985; IAHS/UNEP/UNESCO, 1988, 1993, 1999; IAHS/UNESCO, 1998) is given in figure 16. Young (1977a, 1981) found that the same annual balance was produced by different combinations of winter accumulation and summer ablation; the patterns did not necessarily repeat themselves. Fluctuations in meltwater discharge over a few days closely parallel the air-temperature curve, that from Lake Louise being a better predictor than the Peyto Glacier station. The elevation of the transient snowline was a good indicator of the health of the glacier. LetrŽguilly (1988) found mass balance to be almost entirely related to summer temperature. The correlation was best with data from the meteorological station in Jasper, some 200 km away, rather than with the closest station at Lake Louise. This is in line with TangbornÕs conclusion (1980) that mass balance here was likely most dependent on summer ablation. LetrŽguilly also found that the correlation coefficients of meteorological data with the ELA were as good or better than with the mass balance. Yarnal (1984) demonstrated that the mass balance of Peyto Glacier is also related to the 500-mbar patterns. Synoptic atmospheric pressure patterns having cyclonic circulation favor glacier accumulation, whereas anticyclonic types inhibit the buildup of the regional snowpack. Ablation is suppressed by synoptic patterns associated with cloudy days and (or) low temperatures and is enhanced by patterns associated with warm sunny days. Peyto Glacier accumulation appears to be associated with the large-scale patterns. According to Xie and Zhang (1986), the glacier has a reasonably stable regime with a coefficient of glacial stability of 0.33. Others, in analyzing global mass-balance data, which include those from Peyto Glacier, have concluded that representative mass-balance values might be obtained from observations at the ELA (Ohmura and others, 1986), at the weighted mean altitude (Valdeyev, 1986), or another single point (Konovalov, 1987). Attempts to develop appropriate indices for these and other glacier characteristics, using data from Peyto Glacier, are continuing (Bahr and Dyurgerov, 1999; Dyurgerov and Bahr, 1999).
The glacierÕs accessibility, only a 2- to 3-hour walk from Peyto Lookout off the Trans-Canada Highway, and the availability of semipermanent facilities soon led to the development of many other complementary studies, often in collaboration with Canadian and other universities.
A map of the glacier, at a scale of 1:10,000 with 10-m contours, was prepared as a base for glaciological research. A nine-color edition using a French bedrock-portrayal technique was published in 1970 (Sedgwick and Henoch, 1975), followed in 1975 by a Swiss-style eight-color map; bedrock portrayal and shaded relief added a three-dimensional effect (Henoch and Croizet, 1976).
A subsequent experiment produced an orthophotograph, stereomate, digital-terrain model, and contour map of part of the glacier (Young and Arnold, 1977). There was good agreement in the ablation area with elevations of the existing map (<1 m) but large discrepancies in the freshly snow-covered accumulation area. The accumulation area had thinned by about 20 m since 1966Ð76.
Holdsworth and others (1983) and Goodman (1970), using a radio-echosounder, measured depths of 40Ð192 m in the ablation area and 120Ð150 m in the accumulation area, indicating that the 1967 seismic survey by Hobson and Jobin (1975) was seriously in error, and the volume of 532 x106 m3 of water equivalent was wrong by at least a factor of 2Ð3.
Power and Young (1979a, b), using a modified University of British Columbia (UBC) model, compared simulated discharge for 1967Ð74 to that predicted using algorithms derived from mean daily temperature and then extended the results using a lapse rate of 0.65oCÐ100m with an arbitrary reduction of 15 percent melt applied to the accumulation area (Young, 1980). The model simulates snowmelt and icemelt processes reasonably well. The results confirm that the Peyto Glacier tends to compensate for changes in snowmelt runoff; for example, lower winter snowpacks produce high glacier-melt runoff. The response time of the basin and its various flow components is closely linked to the progression of the snowline upglacier, although summer hydrographs can vary markedly over the seasons. The difference between the calculated and measured discharge over an 8-year period averaged 5x106 m3, which could be accounted for by an average evaporation over the basin of 200 mm aÐ1 (Young, 1982). Gottlieb (1980) obtained reasonably good overall agreement using a degree-day approach and simplified water routing procedures, but his model did not seem to be able to predict peak flows.
Derikx (1973) considered a glacier analogous to a ground-water system and developed a black-box model to predict meltwater runoff in response to meteorological data. Although a linearized equation with simple boundary conditions is a very crude approximation of reality, the results showed a fair correspondence between the calculated and measured meltwater discharge. Derikx and Loijens (1971) refined the model by adding inputs distributed by different elevation bands and by separating the four hydrological components (exposed ice, snow-covered ice, firn, and rock). Daily discharge calculated by the model was 21 percent higher or lower than the measured value. In the warm, dry period of 2Ð27 August 1967, when the snowcover had been removed, Peyto Glacier contributed 40 percent of the Mistaya River streamflow. Applied to the other glaciers in the basin, this means glacier-melt contribution in August can reach 80 percent. Loijens (1974) observed that 46 percent of the annual flow in the river in 1967 occurred in a 7-week period and that 70 percent of this was supplied by glacier runoff and 31 percent by glacier icemelt. Average ice ablation contributed 15 percent to annual streamflow. Young (1977b) reported that 4 years of negative balance contributed some 37x106 m3 of water from storage, or 20 percent of total streamflow, and 4 positive years retained 13.8x106 m3. Hence he proposed a revision in the contribution to annual flow in the North Saskatchewan River at Saskatchewan River Crossing caused by the 1948Ð66 reduction in glacier volume. He suggested 8Ð10 percent compared to HenochÕs (1971) original estimate of 4 percent.
Prantl and Loijens (1977) and Collins and Young (1979, 1981) used hydrochemical characteristics to separate the various flow components and to study their temporal variation. Conductivity falls rapidly as dilute meltwater arrives at the gauge and rises steadily as chemically enriched meltwater routed more slowly through basal tunnels, cavities, and subglacier moraines reaches the portal.
Johnson and Power (1985, 1986) reported on a high-magnitude catastrophic event that substantially modified the proglacial area of Peyto Glacier. A major storm led to the overtopping of a drainage tunnel and the collapse of a large section of ice-cored moraine. The downstream area was subjected to alternate damming, flooding, and draining. Flood waves up to 26 m3 sÐ1 deposited an estimated 6,000 m3 of gravel in the valley, destroying the gauging facility. Other major flooding episodes occurred in 1984, again in association with extreme rainfall events.
Goodison (1972a) derived a snowmelt-rainfall recession curve and found that any particular dayÕs rainfall takes about 5 days to pass through a completely snow-covered basin. The response time shortens as the season progresses. Regression models developed using 1968 data accounted for 82.4 percent of the discharge variation in the 1969 hydrograph and provided a particulary good fit to the JuneÐJuly runoff. Another study found no statistically meaningful relationship between global radiation and ablation (Goodison, 1972b). ¯strem (1973a) came to a similar conclusion in the course of developing an operational model based on temperature, wind, and precipitation. Daily computed runoff values were within about 21 percent of the observed values. The negative correlation between glacier discharge and radiation has been observed for other glaciers where liquid precipitation, usually associated with low incoming short-wave radiation, is a dominating influence on glacier runoff. However, Munro (1975) found that short-term variations in the meltwater hydrograph were closely controlled by the net radiative flux; the sensible heat flux was also an important energy source. Munro and Young (1982) concluded that net shortwave radiation was the prime determinant of meltwater discharge. Estimates were a fair approximation of energy equivalents of ablation derived from stake measurements. An unexpectedly thin boundary layer, about 1 m deep, was attributed to katabatic control of flow (Munro and Davies, 1976). Such a finding has implications for the long-term effectiveness of the turbulent-transfer approach (Munro and Davies, 1977). Another factor in the long-term suitability of turbulent transfer theory for glacier-melt prediction is the finding by Stenning and others (1981) that katabatic layer development and characteristics are also subject to synoptic-scale influences.
Fšhn (1973) found fair agreement between energy- and mass-balance results. About 20 percent of daily snowmelt takes place internally as a result of the penetration of solar radiation. At the base of the snowpack, the water table fluctuated 10Ð50 mm throughout the first 9 days under conditions of continuous snowcover. At some times, lateral inflow of water made up for actual mass loss at the upper end of the snowpack.
By studying the weathering surface in a 5,320-m2 ablation-area site, Derikx (1975) concluded that the hydrological response time was extremely short and closely followed melt calculated from the hourly energy balance. Any delay seemed to be mainly the result of the channel network. Dye-tracer tests by Collins (1982) gave average flow-through velocities of 0.13Ð0.35 m sÐ1, with delays of up to 5 h at low flows and under 2 h during times of peak surface ablation, thus showing a strong dependence on discharge.
The glacier has fairly extensive ice-cored moraines. A simple model suggested that their ablation could be estimated from meteorological variables, if the surface temperature of the debris layer were available (Nakawo and Young, 1982).
Krouse (1974) demonstrated that the isotopic record retains characteristics of the winter precipitation record, but he and West (1972) have both attributed some of the large isotope fluctuations to wind drainage on the glacier and to the topographic shape of the ice surface (Krouse and others, 1977). This conclusion was supported by Foessel (1974), who found Òcool-air poolingÓ in the ablation area in response to dish-shaped terrain, and who developed an equation that proved very reliable in predicting daily mean temperatures at any elevation in the basin. He also found that seasonal temperature trends behaved in a cyclic manner in response to migration of synoptic weather systems over western Canada (in line with the findings of Yarnal (1984) in relation to mass balance).
The suspended sediment regime is very irregular seasonally and diurnally; pulses of sediment occur independently of discharge variations. Sediment concentrations averaged about 660 mg lÐ1 and ranged from 19 to 3,379 mg lÐ1, with one extreme event at 14,000 mg lÐ1. The total suspended sediment output in 1981 was in excess of 68,000 tonnes. Subglacier reorganization of outflow streams that change sediment availability seemed to be the main factor influencing sediment output (Binda and others, 1985). Downstream in the drainage basin of the Peyto Glacier, Smith and others (1982) estimated that over a 75-day measurement period in 1976 some 40x103 tonnes of material was transported to and deposited in Peyto Lake. In 1996, a special session of the annual Canadian Geophysical Union scientific meeting in Banff, Alberta, was devoted to past and present work on Peyto Glacier. Papers from this session will be published by the National Water Research Institute in its science report series (M.N. Demuth, oral commun., 2000).

Yoho Glacier 
Yoho Glacier is the largest southern outflow from the Wapta Icefield (figs. 7, 8). It flows 7 km from the center of the ice field at 3,125 m asl to a terminus at 2,150 m asl. The ELA on the glacier surface lies at about 2,450 m. The first description of Yoho Glacier was published by Habel (1898) when the ice was close to its maximum (Bray and Struik, 1963). At that time the Yoho Glacier had a magnificent ice fall that attracted visitors for many years. Subsequently, studies were conducted by W.H. Sherzer of the Smithsonian Institution (Sherzer, 1907, 1908), and the Yoho Glacier was included in a set of observations undertaken by the Vaux family (Vaux, G., Jr., and Vaux, W.S., 1907a, b, 1908; Vaux, G., 1910; Vaux, M.M. and Vaux, G., Jr., 1911; Vaux, M.M., 1911, 1913). These studies were extended by A.O. Wheeler and members of the Alpine Club of Canada (ACC), which held a number of field camps in that valley (Wheeler, 1907, 1908, 1909, 1910, 1911, 1913, 1915a, 1917, 1920a, b, 1932, 1934). The Yoho Glacier was inspected in 1945 by the DWPB, but by then had retreated up the valley and was a hanging glacier unsuitable for recording purposes (McFarlane, 1945, Meek, 1948a, b). Heusser (1956) concluded that a series of recessional moraines had formed in about 1865, 1880, and 1884. Parks Canada became interested in the hydrology of Yoho National Park, and an attempt was made to extend the record of glacier recession (Kodybka, 1982). For a short time, the National Hydrology Research Institute extended their Peyto Glacier program to include observations on the glaciers and streams around the Yoho Glacier. No report of that work has been published.

Van Horne Range (PC2)
Between the Amiskwi River and the Rocky Mountain Trench and south of Blaeberry River lies the Van Horne Range (fig. 1, PC2). Maximum elevations here are less than 2,900 m asl, which is the height of the regional glaciation level. Some of the deep, north- and east-facing cirques may contain small glaciers, but none is shown on the topographic maps. The range consists of a number of northwest- and southeast-trending ridges.

Conway Group (PC3)
The Conway, Mummery, and Barnard Dent Groups all form part of the same range that lies west of the Blaeberry River and is centered on the Freshfield Icefield (figs. 7, 8). The Conway Group (fig. 1, PC3) is the easternmost of the three groups. Many peaks rise above 3,000 m, reaching a maximum elevation of 3,260 m asl at Solitaire Mountain. The ice-covered area of more than 20 km2 contains several glaciers, of which the most notable are the Cairnes, Lambe, and Conway Glaciers, which range in length from 4 to 5 km with ELAÕs at about 2,400 m.

Mummery Group (PC4)
Southwest of the Conway Group, running parallel to and north of the Blaeberry River, is the Mummery Group (fig. 1, PC4). Elevations trend southwesterly from above 3,000 m at the apex of the Freshfield Icefield (fig. 8) downward to the 2,500-m range near the Rocky Mountain Trench. The main glaciological feature is Mummery Glacier, a southward-flowing extension of Freshfield Icefield. This 7-km-long glacier is joined near its tongue by a large glacier flowing eastward from Mount Mummery (3,320 m). Its ELA is about 2,450 m asl.

Barnard Dent Group (PC5)
The Barnard Dent Group (fig. 1, PC5) contains three major ice masses, including the bulk of the 78-km2 Freshfield Icefield (fig. 8). This ice field is situated in a northeasterly basin that flows into Howse River through a magnificent valley glacier. The Campbell Icefield (13 km2) lies to the west of the Continental Divide. A significant ice accumulation is also on Mount Alan Campbell. Summit elevations generally exceed 3,000 m, with the highest being that of Mount Freshfield (3,325 m). Apart from the Freshfield (11 km), Campbell (5.5 km), Pangman (4.2 km), and Niverville (2.5 km) Glaciers, most are 1 to 2 km long and terminate between 2,200 and 2,400 m. Waitabit Glacier, just to the south of the Freshfield Icefield, has been reduced to three small, disconnected ice masses that total slightly more than 1 km2. 

Freshfield Glacier
Field (1979) provided a comprehensive and illustrated report on Freshfield Glacier (figs. 7, 8, and 17) following the American Geographical Society (AGS) expedition there in 1953, and Ommanney (1984) has compiled a list of publications about work on Freshfield Glacier. The Freshfield Glacier was determined to be slightly over 14 km long when mapped by the Interprovincial Boundary Survey in 1917 (Interprovincial Boundary Commission, 1924). According to Heusser (1956), the glacier began to retreat in 1871. Short readvances took place in 1881 and 1905 (Heusser, 1956); in 1897, Collie (1899) reported on a small push moraine formed in that year. For most of the 20th century, retreat has been continuous and rapid, with a total recession of 1,640 m (fig. 9). The Freshfield Glacier was discovered by Hector (1861) in 1859, a time when the snout was within 100 m of the ancient terminal moraine. The Freshfield Glacier had a row of angular blocks running down the center of the ice stream. One of the largest boulders was used by all the early observers as a common reference point, thus minimizing errors (McFarlane, 1947). The glacier was observed by Stutfield and Collie (1903) in 1897 and 1902, by Hickson (1915) in 1913, by Palmer (1924a, b) in 1922, and by Thorington (1927, 1932, 1938, 1945) in 1926, 1930, 1934, 1937 and 1944.
The long historical record was one reason why the DWPB selected Freshfield Glacier for their studies of the water resources of mountainous rivers in 1945. The position of the snout and changes in its areal extent were measured, and a set of plaques was placed on the ice surface to measure velocity. The survey was abandoned after 1954 due to the expense, including logistics of accessing the glacier. Detailed reports were prepared by the DWPB as internal documents, but some results were published (McFarlane, 1947; McFarlane and May, 1948; Meek, 1948a, b; McFarlane and others, 1949, 1950; May and others, 1950; Carter, 1954). The AGS expedition used photographic and botanical techniques (Field and Heusser, 1954; Heusser, 1954; 1956) to identify glacier limits and variations. Some subsequent visits were made by the ACC (Gray, 1962) but little new scientific information was added. In 1922, Palmer (1924a) estimated the firn line at 2,400 m; today it is closer to 2,600 m. Because the ice field has a low gradient, any change of the firn limit will have far-reaching effects on the glacierÕs mass balance.

Waitabit Ridge (PC6)
Nestled in between the Mummery and Barnard Dent Groups is Waitabit Ridge (fig. 1, PC6). The maximum elevation is in the center of the Ridge in Robinson Peaks (2,925 m), but the average tends to be on the order of 2,600 m. Only one glacier is visible, a 1.5-km-long ice body flowing to the northwest in the section adjoining the Campbell Icefield.

Blackwater Range (PC7)
In the Blackwater Range (fig. 1, PC7), the elevations are lower than those of Waitabit Ridge, but there is a small glacier (<1 km2) in a south-facing cirque between Felucca and Blackwater (2,732 m) Mountains. The range runs parallel to the Rocky Mountain Trench, with one spur pointing eastward along Bluewater Creek. Northward is a small, unnamed range, running from Frigate Mount to the large artificial water body of Bush Arm. It rises to 2,800 m and has two small glaciers.

Forbes Group (PC8)
North and west of Forbes Creek and Howse River lies the Forbes Group (fig. 1, PC8), centered on the 3,612-m peak of that name. Elevations in the Forbes Group exceed 3,000 m, and several are greater than 3,200 m. The area is heavily glacierized and forms part of the Mons Icefield (<30 km2), which spreads out along the range between Mount Outram and Mons Peak. Much of the ice drains through Mons Glacier, which was joined with neighboring Southeast Lyell Glacier in 1902 (Outram,1905). In 1918, the two glaciers were 300 m apart. Between 1918 and 1953, Mons Glacier receded 1,100 m horizontally and 350 m vertically (Field, 1979). Part of the ice field drains westward from the Continental Divide. Several glaciers, including the East, West, and Sir James Glaciers, are on the order of 3 km long. Termini generally lie between 2,300 and 2,500 m asl, though the ice field ELA is thought to be about 2,450 m.

Lyell Group (PC9)
The Lyell Group (fig. 1, PC9) consists of two ranges separated by the valleys of Lyell and Arctomys Creeks. In the southern part of the Lyell Group glaciers are scattered in cirques along the ridge that runs from Mount Erasmus (3,265 m) up to the main body of Lyell Icefield (fig. 8). To the west, this ridge trends to the southwest and Rostrum Peak (3,322 m), along which there are numerous glaciers, 1- to 2-km long, in southeast-facing cirques. The focal point here is Lyell Icefield itself, with its outlet glaciersÑEast, Southwest, and Southeast Lyell Glaciers, that cover an area of about 50 km2. The bulk of the ice field spreads southward from Mount Lyell (3,500 m) before flowing east and west from the Continental Divide and pushing tongues of ice well below 2,000 m. The ELA is probably about 2,500 m asl.

Southeast Lyell Glacier
The Southeast Lyell Glacier (fig. 18) flows eastward from Lyell Icefield (figs. 7, 8). The accumulation area of the Southeast Lyell Glacier extends along the provincial boundary some 9 km between Mount Lyell and Division Mountain at elevations ranging from 2,440 to 3,050 m. As recently as 1902, the Southeast Lyell Glacier was connected with Mons Glacier (Outram, 1905), but, subsequently, the Mons Glacier receded so far up valley as to be scarcely visible (Field and Heusser, 1954). The glacier is strongly broken and crevassed where it travels steeply downward, and a prominent, broad, lateral moraine exists along the northern edge. In 1858, the Southeast Lyell Glacier so intrigued James Hector (1861) of the Palliser Expedition, the first European to investigate this area, that he left a detailed description of it that Thorington used to establish a datum from which to measure frontal recession (Gardner, 1972). Following HectorÕs visit, the positions of the terminus were recorded in 1902 by Outram (1905), in 1918 by the Interprovincial Boundary Survey (Interprovincial Boundary Commission, 1924), and in 1926, 1930, and 1944 by Thorington (1927, 1932, 1945). Heusser (1956) dated the moraines to 1841, 1855, 1885, 1894, 1902, and 1906. In 1947, the DWPB visited the glacier to see whether it might be suitable for inclusion in their network. It was found to have receded so much that what was left of the forefoot was too steep for their purpose (McFarlane, 1947), although it was photographed and a reference baseline established (Meek, 1948a, b). Glacier limits and variations for the Southeast Lyell Glacier were documented by an AGS expedition in 1953 using photographic and botanical techniques (Field and Heusser, 1954; Heusser, 1954, 1956). The AGS expedition concluded, on the basis of field observations, that the reason for the rapid recession since 1930 was a proglacial lake in which the terminus was situated for much of the 1930 to 1953 period. By 1953, the terminus was 60 m beyond the lakeshore and some 4 m above its surface. Available information of the retreat of the Southeast Lyell Glacier is plotted in figure 9.
North of Lyell Icefield is another mountain block running east from Mount Amery (3,329 m) across the ice-covered divide to Cockscomb Mountain (3,140 m). Ice flows east and west from the Continental Divide, pushing down to below 2,400 m in fairly large glaciers such as the Rice Glaciers group (2Ð3 km long) and to below 2,200 m in the Alexandra Glaciers (5.5 and 4 km long) (fig. 19), which were joined in 1918. This ice mass, which might more properly be called the Alexandra Icefield, covers an area of about 25 km2. Average peak elevations are usually well above 3,000 m asl. The western part of the Lyell Group consists of up to two dozen small cirque and niche glaciers, averaging 1 km in length and about 0.5 km2 in area, on either side of the ridges in that part of the range.

Unnamed Range (Ñ)
Between Bush River and Prattle Creek, south of the western extension of the Columbia Icefield, is an umbrella-shaped range with no name. Its axis is almost 40 km long and 5 km wide at its narrowest, whereas the arc of the umbrella frame is 25 km long. Peak elevations range between 2,500 and 2,800 m in the south, and as much as 3,100 m in the north. Some glaciers, as much as 1.5 km2 in area, are scattered along the ridge of the shaft, but the majority are concentrated in the northerly part of the range. On the eastern side, facing the headwaters of Bush River, is an elongated ice apron spread along 8 km of the range. Other glaciers in this sector take the form of cirque glaciers or small ice fields with outlet glaciers ranging from 2 to 4 km in length and about 4 km2 in area. Whereas some glaciers push their snouts below 2,000 m asl, some as low as 1,710 m, most terminate at about 2,300 m.

Vertebrate Ridge (PC10)
Vertebrate Ridge (fig. 1, PC10) extends to the northwest about 20 km; its main feature is a 7-km2 ice field centered on Stovepipe Mountain (2,804 m). A 4-km-long outlet glacier drains from the ice field northward to an elevation of 2,000 m. The southern part of the ice field is drained to the east by a glacier reaching to 1,800 m. Heights along the ridge are generally about 2,550 to 2,650 m in elevation.

Kitchen Range (PC11)
Parallel to Vertebrate Ridge, and marking the western edge of the Park Ranges in this section, is the 30-km-long Kitchen Range (fig. 1, PC11). Summit elevations range between 2,800 and 2,950 m. About eight glaciers are situated on either side of the main ridge, the largest being a 4-km2 ice mass east of Poker Mountain. Lower ice limits vary between 2,025 and 2,530 m, and glacier lengths are generally less than 1.5 km.

Columbia Icefield Group (PC12)
Seen from space, the Columbia Icefield (fig. 1, PC12) (figs. 7, 20), about midway between Lake Louise and Jasper, appears as an extensive snow-covered upland with comparatively little relief. Shaped like a stylistic ÒT,Ó it runs almost 40 km from east to west, and 28 km from northwest to southeast. Its area depends on how many of the peripheral ice masses are included, but an acceptable figure would be 325 km2. It is situated on a plateau 3,000Ð3,325 m in elevation that dips slightly to the south and culminates in the Snow Dome (fig. 21), a gently-sloping 3,520-m peak, completely covered with ice and snow, which is the hydrographic apex of the continent, draining into three oceans (Freeman, 1925; Lang, 1943). Harmon and Robinson (1981) provided a poetic and pictorial commentary on the beauties of the area. Numerous peaks more than 3,500 m asl fringe the Columbia Icefield, so that ice flowing to the margin of the ice field forms huge cliffs above high rock walls and avalanches to the base of the rock walls to form reconstituted glaciers (Boles, 1974). The ice discharges radially through outlet glaciers; the three largest (Columbia, Athabasca, and Saskatchewan Glaciers) flow in deeply incised valleys. There are many other small ones such as Stutfield, Dome, and the Castleguard Glaciers [I, II, III, IV; see footnote 2 in table 1] (Baranowski and Henoch, 1978). Several have been studied in some detail and will be discussed further. Not only is the Columbia Icefield (fig. 7) the largest ice field in the Rocky Mountains, but together with the Clemenceau and Chaba Icefields, it drapes the peaks, mainly along the Continental Divide, in a continuous glacier cover for more than 60 km.
A comprehensive inventory of the glaciers (some of which are listed in table 5) and of the landforms in the Columbia Icefield area, particularly near the Athabasca (figs. 7, 20) and Dome Glaciers, was undertaken by Baranowski and Henoch (1978) and Kucera and Henoch (1978). Detailed geomorphological maps were prepared at scales of 1:25,000 and 1:2,500. Unfortunately, little is known about the bulk of the ice field that lies in British Columbia.
Several federal government agencies and university departments undertook a joint project in 1977 to map the surface of the ice field and to calculate the amount of water held in its snow and ice (Canada, Energy, Mines and Resources, 1978). Preliminary findings showed that the part feeding the Athabasca and Saskatchewan Glaciers (figs. 7, 20) was thinner than previously thought (100Ð365 m). Surveyors used stereoscopic, vertical aerial photography and an Inertial Survey System (ISS) to fix additional ground control points on the ice. The aerial survey aircraft also carried a thermal infrared (IR) line-scanner from which radiometric temperature isolines at 1¡C were to be plotted. One of the thermal IR images generated was used for the cover of a new map of the Columbia Icefield (Canada, Environment Canada, 1980).
The accessibility of the Columbia Icefield, particularly the Athabasca and Saskatchewan Glaciers, and the availability of a fairly good historical sequence of observations, probably favored its selection as the site for a wide variety of glaciological studies. These have included investigations of glacier chemistry, glacier flow, depth measurement, photogrammetry, resistivity, sediment transport, and temperature, among others. Each will be discussed in the context of those parts of the Columbia Icefield on which the work was carried out.

Castleguard Glaciers
The southern limit of the Columbia Icefield is marked by a group of glaciers (collectively known as the Castleguard Glaciers [I, II, III, IV; see footnote 2 in table 1] (fig. 7)) at the head of Castleguard River. Livingston and Field visited them in 1949 and numbered them I to IV from northeast to southwest. Castleguard Glacier IV (fig. 22), the principal southern outlet glacier of the Columbia Icefield was called South Glacier by Ford (1983) and his colleagues from McMaster University working on the Castleguard Cave system. Castleguard Glacier I, situated at 2,300Ð2,700 m asl, was first photographed by the Interprovincial Boundary Survey in 1918 and subsequently by others in 1919, 1923, 1924, and 1949. Until 1924, it abutted a massive moraine which probably formed near the turn of the century (Denton, 1975). Castleguard Glaciers II and III have shrunk enormously since the 1920Õs, as has Castleguard Glacier IV. Ice fronts have receded an average of 500 m since the end of the ÒLittle Ice Age.Ó Sporadic observations by the McMaster group indicate a reduced average recession of the Castleguard Glaciers of about 20 m from 1967 to 1981. This is a key alpine karst locality, because the cave and glacier systems are still in contact. Recent work has focused on the karst cave system (Ford, 1975, 1983), subglacial chemical deposits (Hallet, 1976a, b, 1977; Hallet and Anderson, 1980), and the interaction between glaciers and the limestone bedrock (Smart, 1983, 1984, 1986).

Saskatchewan Glacier
The next major outlet glacier from the Columbia Icefield, the Saskatchewan Glacier (figs. 7, 20, 23) is about 13 km long and some 30 km2 in area. It declines gradually from east to northeast, without ice falls, to its terminus at 1,800 m. Several tributary glaciers were formerly active in providing nourishment, although by the late 1980Õs only one of these supplied the Saskatchewan Glacier. The ELA lies almost at the junction of the ice field and outlet tongue and ranges from 2,440 to slightly more than 2,530 m. Based on ice discharge, Meier (1960) computed the annual accumulation to be 1 m water equivalent (w.e.) and its gradient 13 mm mÐ1 below the firn limit, indicating a high degree of activity. Average ablation ranged from 1 m aÐ1 near the ice field through 2.6 m aÐ1 to 4 m aÐ1 at the snout. The ice was 442 m thick 8 km upglacier and, because of the valleyÕs marked U-shape and very steep walls, was as much as 305 m thick, even quite close to the margin (Meier, 1960).
In 1953, the American Geographical Society expedition used photographic and botanical techniques to determine the history of the Saskatchewan Glacier (Field and Heusser, 1954; Heusser, 1954, 1956). It withdrew from its terminal moraine in 1893 and by the time of their visit had retreated 1,364 m. The rate of recession from 1948 to 1953 was quite fast at 55 m aÐ1.
In 1945, the Saskatchewan Glacier was investigated by the DWPB Calgary office. At this time the toe was very irregular and the surface of the glacier very rough. The position of the snout and changes in its areal extent were measured every year, and a set of plaques was placed on the ice surface to measure velocity; in 1950, the measurement interval became biennial. Detailed reports were prepared by the DWPB as internal documents, but some results were published (Lang, 1943; McFarlane, 1946b; Meek, 1948a, b; McFarlane and others, 1950; Collier, 1958). In the mid-1960Õs, following recommendations made at the Glacier Mapping Symposium, the Water Survey of Canada began to use terrestrial photogrammetry to determine volumetric change, with results as shown in table 6 (Campbell and others, 1969; Reid, 1972; Reid and Charbonneau, 1972, 1975, 1979, 1981; Reid and others, 1978). Snout and plaque surveys were continued in the intervening years by the Calgary office of the Water Survey of Canada (WSC) (Warner and others, 1972; Canada, Environment Canada, 1976, 1982), but all were terminated by 1980 (fig. 9).
Meier (1960) measured the velocities on the surface and at depth, the surface and bedrock topography, and the ablation and flow structures in a project designed to test theories of glacier flow. Summer velocities were generally greater than annual ones, though there were significant velocity fluctuations in the short-term and there was even intermittent backward movement. Maximum surface velocities of 117 m aÐ1 at the firn limit decreased unevenly downglacier to 3.5 m aÐ1 at the snout. The flow law of ice was determined by analysis of a 140-m-deep vertical-velocity profile and a surface transverse-velocity profile. Three main classes of structural ice features were distinguished: (1) primary sedimentary layering, (2) secondary flow foliation, and (3) secondary cracks and crevasses. Meier (1958) also studied crevasse patterns as part of this general study of flow. Preliminary results showed that crevasse formation was preceded by a buildup in extension rate. Crevasses then formed so as to relieve the extension rate on intercrevasse blocks. Intense deformation resulting in pure shear preceded an extending crevasse. No crevasses deeper than 30 m were observed.
RigsbyÕs (1958, 1960) fabric diagrams did not show preferred orientations as strong as those for other temperate glaciers. Any strong patterns observed were thought to be in the more extensively metamorphosed, presumably older, ice flowing from depth. Melt recrystallization probably changed the strong orientation of crystals from a single maximum with optic axes normal to the foliation plane to 3Ð4 maxima, none of which necessarily coincided with the pole of the foliation plane.
Sharp and Epstein (1958) and Epstein and Sharp (1959) analyzed the oxygen-isotope ratios in different firn strata and found differences that reflected the elevation of accumulation, seasonal influences, differences among individual storms, and subsequent diagenetic changes. Ice along the centerline showed an irregular trend to lower ratios downglacier, which was thought to reflect ice transport along flow lines from different parts of the accumulation area.
McPherson and Gardner (1969) observed large cross-valley topographic highs, composed of till, in front of the Saskatchewan Glacier, which were old landforms emerging from beneath the ice. They might have been interpreted as end moraines. If so, they were not disturbed when overridden during the neoglacial advance.

Columbia Glacier
The Columbia Glacier (figs. 7, 20, 24), 8.5 km in length and about 16 km2 in area, is the major outlet glacier flowing from the northwest section of the ice field, draining ice as well from the western slopes of Snow Dome and the eastern slopes of Mount Columbia (3,747 m). Columbia Glacier drops rapidly from the plateau area over a major ice fall, which creates a series of very well-defined ogives, to flow in a 0.6- to 0.7-km-wide glacier to an elevation of about 1,500 m. Its ELA lies at about 2,140 m. Habel (1902), Schaffer (1908), Palmer (1924Ð1926, 1925), Field (1949), and Field and Heusser (1954) recorded the location of the terminus in 1901, 1907, 1920, 1924, 1948, and 1953. The Interprovincial Boundary Survey photographed the glacier in 1919 (Interprovincial Boundary Commission, 1924). Heusser (1954) established a chronology of recessional moraines and dated the outer one at 1724 and others at 1842, 1854, 1864, 1871, 1907, 1909, and 1919; he concluded that the glacier had retreated 394 m between 1724 and 1924. A plot of the recession data is given in figure 9; it assumes no change from 1725 to 1850. Baranowski and Henoch (1978) observed that the Columbia Glacier had advanced as much as 1 km from 1966 to 1977. Since the Columbia Icefield map was published, the glacier has advanced farther to completely fill the large proglacial lake, a distance of some 800 m (Parks Canada, oral commun., 1986). The glacier is not being surveyed regularly, so it is not known whether this advance is continuing.

Athabasca Glacier
The most-visited glacier in Canada is without a doubt the Athabasca Glacier (figs. 20, 25). Situated only 1 km from the Icefields Parkway, which passes through Banff and Jasper National Parks, it is one of the primary destinations for tourists and tours. It has also become the focus of Parks Canada interpretive program and provides, through the Brewster snowmobiles, one of the few easy opportunities for the general public to get up on the ice. However, most would probably identify it as the Columbia Icefield. Its accessibility and the need for information about it have led to numerous scientific studies, which are summarized below.
The 6.5-km-long glacier leaves the ice field at 2,800 m, descends in a series of three ice falls as it passes over successive rock thresholds and continues as a gentle 1-km-wide tongue with a slope of 3Ð7¡ to its terminus at 1,925 m asl. It cuts through the axis of a gentle anticline and is flanked by walls of limestone, dolostone, and shale. Crevasses are well developed in the lower two ice falls and extend almost across the entire width of the glacier. Kucera (1987) measured 15 of the largest crevasses and found them to be no deeper than 36 m. Part of the glacier front is formed by moraine-covered ice that continues up valley, forming about two-fifths of the glacier along the northwest side. Beno”t and others (1984) published a key to the photointerpretation of features on and around the Athabasca and Dome Glaciers. Elements that are important for the analysis of remote-sensing imagery, such as the characteristics of various features, their texture, and reflectivity are all discussed. This study complements an earlier glaciological and geomorphological investigation by Kucera and Henoch (1978). Howarth (1983) and Howarth and Ommanney (1983) reported some difficulty in interpreting Landsat scenes for this area.
It is surprising that no systematic annual mass-balance studies have been carried out on the Athabasca Glacier. However, geochemical studies within the 2,600Ð2,700-m elevation band were used by Butler and others (1980) to estimate a net annual mass balance here of 1.5 and 2.4Ð2.7 m w.e., respectively, for 1976Ð77 and 1977Ð78. Average snowfall on the ice field is estimated to be more than 7 m, and the ELA is at about 2,600 m asl. Holdsworth and others (1985) obtained an 11.5-m core from the top of Snow Dome to evaluate the site for a deep core drilling. Preliminary analysis indicated a lack of evidence of regular seasonal variations in the stable-isotope data but evidence of percolation and homogenization of the isotopic and chemical constituents. It is clear that the overall mass-balance trend during the 20th century has been strongly negative. In 1870, the glacier was about 1.5 times its present total volume (1,013x106 m3) and 2.5 times its area (6x106 m2 vs. 2.6x106 m2). The average rates of decrease in volume have declined: 3.2x106 m3 aÐ1 for 1870Ð1971 to 2.5x106 m3 aÐ1 for 1959Ð1971 (Kite and Reid, 1977). Ice volume has also been reconstructed by Mayewski and others (1979) as shown in table 7.
Observations of the retreat of the terminus of the Athabasca Glacier have been carried out for many years and are shown in table 8. Figure 26 shows the position of the terminus of Athabasca Glacier in 1952 and 1977. Hermann Woolley and J. Norman Collie visited and named Athabasca Glacier in 1898, at which time it coalesced with the Dome Glacier. Athabasca Glacier was photographed in 1908 (Schaffer, 1908) and again in 1919 (Interprovincial Boundary Commission, 1924). Field photographed it in 1919. In 1945, the DWPB commenced a series of annual surveys from their Calgary office. The position of the snout and the changes in the areal extent of Athabasca Glacier were measured every year, and a set of plaques was placed on the ice surface to measure velocity. Detailed reports were prepared as internal documents (McFarlane, 1945, 1946b, 1947; McFarlane and May, 1948; MacFarlane and others, 1949; May and others, 1950; Carter, 1954; Carter and others, 1956; Fowler and others, 1958; Chapman and others, 1960; Davis and others, 1962; Davies and others, 1964, 1966, 1970; Glossop and others, 1968), but some results were published (McFarlane, 1946a; Meek, 1948a, b; McFarlane and others, 1950; Collier, 1958).
Following experimental aerial photogrammetric surveys of the Athabasca Glacier in 1959 and 1962 (Reid, 1961), the University of New Brunswick examined the use of terrestrial photogrammetry for measuring the melt of the Athabasca Glacier (Konecny, 1963, 1966; Reid and Paterson, 1973). From 1959 to 1962 the surface was reduced an average of 1.7 m, and the glacier lost 3.936x105 m3, representing a 35 percent contribution to average annual streamflow (Reid and Paterson, 1973). The Water Survey of Canada then switched to a program of terrestrial surveys every 2 years (Campbell and others, 1969; Reid, 1972; Reid and Charbonneau, 1972, 1975, 1979, 1981; Reid and others, 1978) for the measurement of volumetric change (table 8); snout and plaque surveys were continued in the intermediate years by the Calgary office (Warner and others, 1972; Canada, Environment Canada, 1976, 1982). Paterson (1966) showed that the difference between individually surveyed markers and map heights was less than three times the theoretical error, or about 15 percent of the contour interval. Additional surveys were made of the Athabasca Glacier in August 1977 as part of a program to remap the Columbia Icefield and surrounding areas and to test the relatively new orthophoto-mapping process. Young and others (1978) reported that there was little to choose amongst the various techniques, although the orthophoto mapping might be faster and cheaper and produced a digital-terrain model in computer-compatible form of use for further computations. Errors in the three methods ranged from about 1Ð2 m on the lower glacier to about 10Ð20 m on the upper.
Some other records are also available from Field, who revisited the glacier in 1948, 1949, 1953 and 1963 (Denton, 1975). He and Heusser, using photographic and botanical techniques (Field and Heusser, 1954; Heusser, 1954, 1956), were able to develop the history further. Athabasca Glacier reached its maximum about 1714 and began to withdraw from different parts of its end moraine in 1721 and 1744. It readvanced in the first half of the 19th century, reaching almost to its maximum extent. Recession began between 1841 and 1866 and has continued with minor fluctuations marked by moraines formed about 1841, 1900, 1908, 1925, and 1935.
The glacier has been used as a test area for a variety of depth-measurement techniques, some of which have been designed to provide information for physical studies of glacier flow.
Kanasewich (1963) used gravity techniques to obtain ice-thickness values along eight transverse profiles, whereas Paterson and Savage (1963a) used seismic waves to determine depths for 2- km downglacier from the lowest ice fall. During the summer of 1959, five holes were drilled into the glacier with a prototype hotpoint drill (Stacey, 1960). Rossiter (1977) tested a 1Ð32 MHz radio interferometry depth-sounding technique on Athabasca Glacier. High scattering levels above 8 MHz were attributed to water-filled cavities within the glacier on the order of 3Ð6 m in width, having typical separations of 10Ð30 m. Strangway and others (1974) found that at 1, 2, and 4 MHz the ice had a dielectric constant of about 3.3, corresponding to that for ice near 0¡C. Measured depths were not always consistent with previous seismic, gravity, and borehole results. Radio-echosounding has featured prominently in depth investigations. A sophisticated mobile system linked to fixed transponders (Goodman, 1970, 1975) provided results that agreed within 14 m with the seismic and borehole measurements of Savage and Paterson (1963). Repetitive soundings at individual locations revealed the presence of intraglacier structures that appeared to be related to changing hydrological or glaciological conditions within the glacier (Goodman, 1973). Waddington and Jones (1977), using a 1Ð5 MHz radio-echosounder, sampled the accumulation area of the Athabasca Glacier from below Castleguard Mountain northward to the ice falls and westward to the divide. They found depths ranging from 100 to 365 m. Contour maps of ice thickness and basal elevations have been produced by Trombley (1986), based on 300 depth measurements of the lower 3.6 km of the glacier with a portable radio-echosounder. A summary of some of the results, based on Paterson (n.d.) and Trombley (1986) is given in table 9. 
Measurements show that in the ablation area, below the lowermost ice fall, the longitudinal and transverse profiles of ice thickness are geometrically regular and simple. A bedrock depression and two bedrock rises influence the flow. The ice is thickest (>320 m) in the deepest part of the depression, which appears to be a relic valley carved by what is now a small hanging glacier. Removal of the glacier would create a chain of paternoster lakes (Kanasewich, 1963).
In an interesting variant