Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Snow glide and full-depth avalanche occurrence, Cascade Mountains, British Columbia Clarke, Jennifer A. 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1995-0004.pdf [ 7.49MB ]
Metadata
JSON: 831-1.0098908.json
JSON-LD: 831-1.0098908-ld.json
RDF/XML (Pretty): 831-1.0098908-rdf.xml
RDF/JSON: 831-1.0098908-rdf.json
Turtle: 831-1.0098908-turtle.txt
N-Triples: 831-1.0098908-rdf-ntriples.txt
Original Record: 831-1.0098908-source.json
Full Text
831-1.0098908-fulltext.txt
Citation
831-1.0098908.ris

Full Text

SNOW GLIDE A N D FULL-DEPTH A V A L A N C H E OCCURRENCE, C A S C A D E M O U N T A I N S , BRITISH C O L U M B I A  by  JENNIFER A. C L A R K E B . E . S . , University of Waterloo,  1992  A THESIS S U B M I T T E D IN P A R T I A L F U L F I L M E N T O F THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in  T H E F A C U L T Y OF G R A D U A T E STUDIES Department of Geography  We accept this thesis as conforming to the renuired standard  T H E UNIVERSITY OF BRITISH C O L U M B I A November, 1994  © Jennifer Clarke, 1994  In  presenting  degree at the  this  thesis  in  University of  partial  fulfilment  of  of this thesis for  department  or  publication of  by  his  or  her  of  DE-6 (2788)  It  this thesis for financial gain shall not  (SeaSRAfW  The University of British Columbia Vancouver, Canada  Date  representatives.  Hovc.~s.be-/10. 19?^  for  an advanced  Library shall make it  agree that permission for extensive  scholarly purposes may be  permission.  Department  requirements  British Columbia, I agree that the  freely available for reference and study. I further copying  the  is  granted  by the  understood  that  head of copying  my or  be allowed without my written  11  Abstract Snow Glide and Full-Depth Avalanche Occurrence, Cascade Mountains, British Columbia  Snow glide is the translational slip of the entire snow pack over a sloping ground surface. It is thought that rapid rates of snow glide precede the release of full-depth avalanches. The nature of avalanches that release at the ground makes them difficult to predict and difficult to control using explosives. The aim of this research is to determine the relationship between rapid snow glide and full-depth avalanche occurrence and to examine climate factors affecting both processes. Data collected from an instrumented site along the Coquihalla Highway i n the Cascade Mountains of British Columbia were used for analysis during two winter seasons (199293, 1993-94). Glide is influenced by the nature of the interaction between the roughness of the ground and the snow pack, and by the distribution of water at the interface. The presence of water at the interface affects the material properties o f snow and the friction conditions. The impact of freewater on glide is influenced by the volume and rates of water input. Higher glide rates and fulldepth avalanche release are the almost immediate responses to contributions of free-water. The data show that the most significant contributor is rainfall, which is common i n the study area throughout the winter season. The supply of free-water from snow melt due to radiative and thermal sources of energy become more significant i n the spring. Water inputs increase the thickness of the saturated layer at the base of the snow pack, allowing greater amplitudes of roughness to be overcome. By drowning or partially drowning the roughness elements, a thin film of water reduces the shear resistance of the snow pack to downslope movement. Inputs of water at rates higher than transmissions rates w i l l increase pore pressures and decrease shear stress encouraging further downslope movement. Failure of the snow pack at the ground is translational, most often occurring 12-24 hours after a rainfall event, but sometimes much later when avalanche release would not be expected. Although there is no threshold glide velocity associated with avalanche release, it can be concluded that snow glide is a good indicator of active periods of full-depth avalanche occurrence. However, results from this study show that rainfall rates and snow melt rates may be more accurate predictors of avalanche occurrence i n the study area.  iii SNOW GLIDE AND F U L L - D E P T H A V A L A N C H E O C C U R R E N C E . C A S C A D E M O U N T A I N S . BRITISH C O L U M B I A TABLE OF CONTENTS  Abstract Table of Contents List of Tables List of Figures Acknowledgement Dedication  ii iii v vi viii ix  Chapter 1: 1.1 1.2 1.3  1 1 3 5  INTRODUCTION Overview Problem Statement and Rationale Research Objectives  Chapter 2: S T U D Y A R E A 2.1 Introduction 2.2 Regional Geography 2.2.1 Vegetation 2.2.2 Hydrology 2.3 Regional Climate 2.4 The Coquihalla Avalanche Area  6 6 9 9 11 11 15  Chapter 3: 3.1 3.2 3.3 3.4  19 19 20 23 28 28 30 32 34 39  Chapter 4: 4.1 4.2 4.3 4.4  41 41 42 46 48 49 51 55  SNOW GLIDE General Characteristics of Snow Glide Development of the Snow Glide M o d e l Snow Glide Instrumentation and Measurement Snow Glide Characteristics at the Study Site 3.4.1 Seasonal Nature of Snow Glide 3.4.2 Diurnal Nature of Snow Glide 3.4.3 Spatial Variation of Snow Glide on a Slope 3.5 Snow Glide and Full-Depth Avalanche Release 3.6 Discussion FREE WATER AT THE SNOW-GROUND INTERFACE Introduction Effect of Water on Snow Rheology Effect of Water on Snow Pack Separation and Friction at Interface Water F l o w at the Snow/Ground Interface 4.4.1 Movement Through the Saturated Zone 4.4.2 Experimental Results 4.5 Water Pressures in the Snow Pack  iv T A B L E O F C O N T E N T S (cont.) 4.6 Effect of Rain-on-Snow Events of Glide and Full-Depth Avalanche Occurrence 4.7 Discussion  60 70  Chapter 5: E N E R G Y A N D S N O W M E L T 5.1 Introduction 5.2 Energy Conditions at the Field Site 5.2.1 Instrumentation and Data Collection 5.2.2 Topographic and Site Influence 5.3 Contribution of Energy to Snow Melt 5.3.1 Energy Balance Approach 5.3.2 Approach Used i n the Study 5.3.3 A i r Temperature as a Surrogate Variable 5.4 Energy, Snow M e l t , and Full-Depth Avalanche Occurrence 5.5 Discussion  73 73 74 74 76 80 80 86 88 91 99  Chapter 6: D I S C U S S I O N A N D C O N C L U S I O N S  106  REFERENCES  110  Appendix A List of Full-Depth Avalanche Occurrences Selected for Analysis  116  LIST O F T A B L E S  Table Table Table Table Table  2.1: 2.2: 2.3: 2.4: 2.5:  Temperature Summary Precipitation Summary Summary of Avalanche Occurrences Full-Depth Avalanche Occurrences on Paths i n the Study Area Study-Area Avalanche Path Characteristics  13 14 16 17 18  Table 3.1: Different Constitutive Equations for Snow Glide Table 3.2: Summary of Glide Rates Table 3.3: Possible Full-Depth Avalanche Trigger Mechanisms Table Table Table Table  4.1: 4.2: 4.3: 4.4:  21 30 39 45 52 53  Conditions at the Snow/Ground Interface - Coquihalla Glide Site Conditions for Water F l o w Tests Water F l o w at the Interface - Experimental Results Correlation of Avalanche Occurrence with R a i n Events for Selected Time Periods  Table 5.1: Radiation Instrumentation at the Study Site Table 5.2: Correlation of Glide Velocity with M a x i m u m A i r Temperatures Selected Time Periods Table 5.3: Full-Depth Avalanches Triggered by Radiation Exposure  70 76 for 93 94  LIST O F F I G U R E S  Figure 2.1: Regional Location of Study Area Figure 2.2: Location of the Study Area Figure 2.3: Smooth Bedrock Slopes of Y a k Peak  7 8 10  Figure 3.1: Glide Shoe Instrumentation Figure 3.2: Placement of Glide Shoes on the Slope Figure 3.3: Seasonal Nature of Snow Glide a) 1992-93 Season b) 1993-94 Season Figure 3.4: Diurnal Nature of Snow Glide a) 1992-93 Season b) 1993-94 Season Figure 3.5: Schematic Showing Distribution of Snow Cover at the Glide Site for Two Different Scenarios Figure 3.6: Full-Depth Avalanche Release on Y a k Peak Figure 3.7: Schematic of Assumed Basal Shear Stress, Glide Velocity, and Geometry Definitions for the One-Dimensional Glide Crack M o d e l Figure 3.8: Glide Rates and Full-Depth Avalanche Occurrence a) 1992-93 Season b) 1993-94 Season  25 27 29  Figure Figure Figure Figure Figure Figure Figure Figure Figure  41 44 44 48 54 56 57 58  Figure Figure Figure Figure Figure Figure  4.1: 4.2: 4.3: 4.4: 4.5: 4.6: 4.7: 4.8: 4.9:  Effect of Water Layer on Surface Roughness Slush Rounded Polycrystal Increase i n Glide Velocity with Change i n Water Layer Thickness Photo of Dye Conditions for Test #2 Thickness of Saturated Layer with Distance Downslope from Divide Pore Pressures with Increasing Saturated Layer Thickness Critical Pore Pressure with Changing Density and Friction Angle Full-Depth Avalanche Occurrences and Rain-On-Snow Events 1992-93 Season 4.10: Full-Depth Avalanche Occurrences and Rain-On-Snow Events 1993-94 Season 4.11: Snow Glide, Avalanches, and Climate Conditions for February 25 - M a r c h 15, 1993 4.12: Snow Glide, Avalanches, and Climate Conditions for M a r c h 17 - 27, 1993 4.13: Snow Glide, Avalanches, and Climate Conditions for December 8 - 18, 1993 4.14: Snow Glide, Avalanches, and Climate Conditions for December 30 - January 15, 1993/94 4.15: Snow Glide, Avalanches, and Climate Conditions for February 18 - M a r c h 6, 1994  31  33 34 35 37  61 62 64 65 67 68 69  vii LIST O F F I G U R E S (cont.) Figure Figure Figure Figure Figure  Figure  Figure Figure Figure Figure Figure Figure Figure  5.1: 5.2: 5.3: 5.4: 5.5:  Influence of Slope Angle and Aspect on Radiation Exposure Valley Geometry - Incidence of Radiation Snow Cover on Zopkios Ridge at Two Different Times Radiation Recrystallization Seasonal Summary of Net Radiation Equivalent M e l t Rates a) 1992-93 Season b) 1993-94 Season 5.6: Snow Glide and A i r Temperatures at the Glide Site a) 1992-93 Season b) 1993-94 Season 5.7: Snow Glide, Avalanches, and Climate Conditions for December 10 - 24, 1992 5.8: Snow Glide, Avalanches, and Climate Conditions for January 23 - February 6, 1993 5.9: Radiation Conditions for January 23 - February 6, 1993 5.10: Snow Glide, Avalanches, and Climate Conditions for January 1 6 - 3 1 , 1994 5.11: Radiation Conditions for January 16 - 31, 1994 5.12: Snow Glide, Avalanches, and Climate Conditions for M a r c h 1 4 - 3 1 , 1994 5.13: Radiation Conditions for M a r c h 14 - 31, 1994  77 78 79 83 89  92  96 97 98 100 101 102 103  Vlll  Acknowledgements  I would first of all like to thank B i l l Golley of the Ministry of Transportation and Highways Avalanche Section for his invaluable assistance i n the field. B i l l Golley helped install instruments and accompanied me on many ski tours (providing ski tips on the way) to maintain instruments and conduct snow profiles. Meteorological and avalanche occurrence data from the Summit weather station was provided by the Ministry of Transportation and Highways. Thanks also to other staff at the Coquihalla Avalanche office for their generous cooperation and assistance. I owe many thanks to my friends and colleagues who helped i n the field and i n the office. In particular: Laurent M i n g o , Darren H a m , Craig Nistor, M i k e Smith, Scott Davidson, Magdelena Rucker, and Robin Jones. Thanks to my supervisor, Dave M c C l u n g , for his valuable academic advice and provision of research equipment and resources necessary for this study. Radiometers were thankfully borrowed from T i m Oke ( U B C ) and from Fes de Scally (Okanagan University College). Thanks also to M i k e Bovis for critically reviewing this manuscript and for the systematic eradication of three-letter words i n this paper.  Dedication  To M y M o m  L i k e a Diamond Glint on Snow  1 Chapter 1: I N T R O D U C T I O N  1.1 Overview  The term snow glide is used to describe the translational movement of a snow pack over a relatively smooth surface.  Snow on alpine slopes is always moving, or creeping, due to the  force of gravity and by snow metamorphism, or the rearrangement of snow grain within the snow pack. Settlement of the snow pack results i n densification, while shear deformation promotes failure. The snow pack can separate entirely from the ground, forming folds, after extreme deformation.  Prerequisites for snow glide have been formulated from field  observations (in der Gand and Zupancic, 1966; M c C l u n g , 1975; M c C l u n g , 1981; M c C l u n g , 1987).  These are as follows:  1) the snow ground interface must be at 0° C . This allows for the presence of free-water at the interface; a necessary factor for snow glide. If temperatures at the interface drop below 0° C then the snow pack is frozen to the ground.  2) slopes must be greater than 15° for roughnesses typical of alpine terrain.  3) a fairly smooth ground interface is required for snow glide.  Bare rock or grassy  vegetation provides a suitably smooth surface for snow glide.  The process of snow glide has always been associated with the release of avalanches,  full-depth  also called glide avalanches. Glide avalanches are wet slab avalanches that  2 release at the ground. Wet snow requires large amounts of energy to propagate shear fractures for failure, which explains why these avalanches are difficult to control using explosives.  A tensile fracture, or glide crack, i n the snow pack, initiated by rapid gliding, is  a prerequisite for full-depth avalanche occurrence ( M c C l u n g and Schaerer, 1993). Formation of this fracture requires an increase i n glide speed i n a downslope direction from the fracture location.  Early investigations of snow glide include a primarily descriptive investigation by i n der Gand and Zupancic (1966). Later studies help characterize the mechanics of gliding and fulldepth avalanche initiation ( M c C l u n g , 1981; Lackinger, 1987; M c C l u n g , 1987).  The  significance of free-water at the snow/ground interface, is addressed by M c C l u n g and Clarke (1987). Lackinger (1987) presents glide rates and associated climate conditions for a period before full-depth avalanche release but, since instrumentation was placed on the avalanched slope, data collection stops after the first avalanche occurrence.  Currently, no studies are  known that investigate the direct effect of meteorological conditions on rates of glide and full-depth avalanche release on a continuous basis. This study uses continuously recorded rates of snow glide, measured on a smooth rock slope i n the Cascade Mountains, to identify influencing factors and to investigate their relation to avalanche release on adjacent slopes i n the study area.  Data were collected at a study site located along the Coquihalla Highway i n the Cascade Mountains of British Columbia, approximately 250 kilometres est of Vancouver.  Smooth,  south-facing granite exposures characterize the ridge frequented by full-depth avalanches. Data collected during two winter seasons (1992-93, 1993-94) include instrumented rates of  3 snow glide and radiation. Meteorological data and avalanche records were made available by the British Columbia Ministry of Transportation and Highways - Avalanche Section from the Summit Weather Station.  1.2 Problem Statement and Rationale  This research focuses primarily upon the process of snow glide and its relationship to the occurrence of full-depth avalanches. Downslope movement of the entire snow pack over the smooth ground surface is influenced by a complex combination of factors.  Most important  are those factors that significantly influence the supply of free-water to the snow/ground interface.  The response of snow glide and avalanche release to changing climate conditions  is examined i n this study. Using measured glide rates, meteorological and snow pack conditions, and a record of avalanche occurrences, this study w i l l examine this response i n greater detail. A predictive relationship between snow glide and full-depth avalanche occurrence is the focus of interest for this study.  Data were collected over two winter seasons (1992-93, 1993-94) from a unique site located along the Coquihalla Highway on the slopes of Y a k Peak and Zopkios Ridge; part of the British Columbia Cascade Mountain Range.  The Coquihalla Highway is an important  transportation link between the coast and the interior of British Columbia. A staff of forecasters, working for the Ministry of Transportation and Highways, is responsible for the monitoring and control of avalanches along the route. The study area was selected for the high frequency of full-depth avalanches thought to be triggered by snow glide.  Full-depth avalanches lie beyond the norm of traditional techniques for avalanche prediction and control. Lackinger (1987) reports that on December 31, 1974 in Vorarlberg, Austria attempts to release a full-depth avalanche by blasting along a glide fracture were unsuccessful.  The slab later released and was said to have been triggered by a combination  of water emanating from the slope and a recent snow fall.  Twelve people were killed.  Intervals of rapid gliding are thought to precede the release of full-depth avalanches.  These  avalanches are potentially large i n magnitude. Their complex nature makes them difficult to predict and difficult to trigger using explosives.  Snow glide may also displace and/or cause  damage to transmission towers, ski lifts, and other standing structures on slopes.  The  process of snow glide is responsible for potentially hazardous snow sluffs from rooftops. The associated hazard potential would indicate a need to better understand snow glide and the mechanisms contributing to full-depth avalanche release.  Since it is suspected that water is the primary mechanism for increasing snow glide, the nature of water flow through the snow pack is examined i n this study. Time periods during two study seasons are used to determine the typical response to inputs of free-water.  Inputs  of free-water are supplied mostly by rainfall, which is common in the study area even i n mid-winter, and from snow melt events. complex factor.  The supply of water from snow melt is a more  The influence of exposed rock slabs in the study area makes radiation a  very important parameter for snow melt. O n clear days when the air temperature remains below zero, energy inputs can still warm exposed rock, causing snow melt that is transported directly to the snow/ground interface.  5 1.3 Research Objectives  In an examination of snow glide and the contributing mechanisms for movement, this study aims to determine a relationship to full-depth avalanche occurrences. The objectives for research are:  1. to further understand the process of snow glide and to determine how regional and local geographical and climatological conditions influence the process of snow glide.  2. to determine the temporal and spatial characteristics of snow glide at the study site.  3. to examine the effect of free-water contributions on rates of snow glide and full-depth avalanche release.  4. to determine how energy conditions contribute to the production of free-water, and to determine the relative importance of contributions from rainfall and from snow melt.  5. to determine the relationship between rapid rates of snow glide, climate conditions, and the release of full-depth avalanches, and to determine whether snow glide measurements accurately represent conditions on adjacent slopes.  It is hoped that this research w i l l expand current understanding of the nature of snow glide. It is also hoped that this research could also provide information and methods that could be incorporated into current avalanche forecasting models.  6 Chapter 2: S T U D Y A R E A  2.1 Introduction  The study area is located i n the Cascade Mountain Range of southern British Columbia at 49°31.5' N and 121°05' W (Figure 2.1).  The Cascade Mountains extend north into  British Columbia from western Oregon and Washington. To the west, the Cascade Mountains are separated from the Pacific Range of the Coast Mountains by the Fraser River.  O n the east the Cascades i n British Columbia are flanked by, and merge with, the  Kamloops Plateau.  The study area includes slopes adjacent to the Coquihalla Highway (British Columbia Highway #5) which are prone to snow gliding (Figure 2.2).  Chosen for this study are  those slopes on Y a k Peak and Zopkios Ridge which are characterized by exposed granite bedrock.  It is on these smooth slopes that full-depth avalanches commonly occur and  large magnitude events affect transportation along the Coquihalla Highway.  The geography and climate conditions at this study site are particularly favourable for snow glide.  The smooth granite slopes, combined with moist snow pack conditions,  create an optimal situation for snow glide.  Full-depth avalanches that have been released  by snow glide, are a frequent and common occurrence i n the study area.  0  5.0  10.0 SCALE  15.0 km  8  9 2.2 Regional Geography  The Cascade Mountains are composed of Palaeozoic and Mesozoic sedimentary and volcanic rocks strongly folded, metamorphosed and intruded by granitic batholiths (Holland, 1976).  Such a batholithic structure underlies Zopkios Ridge.  Summits of  peaks and ridges attain approximately uniform elevations; quite possibly the result of the dissection of a late Tertiary erosion surface (Holland, 1976).  Lower slopes i n the study  area are mantled with colluvial material, while till deposits f i l l the valley bottom.  Yak  Peak and Zopkios Ridge have concave-shaped sheets of bedrock that dip sharply from 25° near the bottom, to near vertical approaching the summit (Figure 2.3).  2.2.1  Vegetation  Vegetation i n the study area can be classified as montane to sub-alpine. In the subalpine, lower sections are composed of closed forest stands. Vegetation in upper sections of the subalpine are a transition between the forested stands and alpine tundra. Trees i n the upper zone are sparsely spaced and are stunted and shrub-like.  Vegetation on the upper slopes of the study area are sparse due to steep slopes and shallow bedrock.  Interior Douglas F i r (pseudotsuga menziessi) is present i n the valley  and at higher elevations.  Sub-alpine Englemann Spruce (picea engelmanni) and Sub-  alpine F i r (abies lasiocarpa)  are common ( M O H , 1978).  Forest stands are dissected by  avalanche paths. O n the disturbed slopes, avalanche-track alder {alnus crispa) and  10 willow species (salix spp.) dominate. The flexible nature of these shrubs enables them to withstand large down-slope snow forces.  Vegetation limits the ability for snow glide,  therefore, the upper slopes, being bare, are that much more prone to this process.  Figure 2.3: Smooth Bedrock Slopes of Yak Peak behind Summit Weather Station  11 2.2.2 Hydrology  Water flows from the slopes of the study site into either the Boston Bar Creek or the Coquihalla River, both of which are in the lower Fraser River drainage basin.  The  Water Survey of Canada (1991) collects discharge data from the Coquihalla River (below Needle Creek) downstream from the study area (Mean Annual Discharge: 3.14 m s ). 3  -1  The discharge data show that maximum instantaneous discharges often occur during the winter season, which would indicate the hydrological magnitude of a rain-on-snow event or a period of rapid melt.  2.3 Regional Climate  The climate of the study area dictates the nature of the snow cover and it is also a very important control on snow glide.  The nature of the snow cover varies throughout the  winter season depending upon meteorological conditions. The regional climate of the Cascades is particularly conducive to those conditions necessary for snow glide to occur.  Geographical factors that most strongly influence mountain climates are outlined by Barry (1992) as: latitude, continentality, altitude, and topography.  The continentality of the  study area is very important i n terms of snow conditions. The annual range of mean monthly temperatures indicates whether the area has a maritime or a continental climate. The study area has a maritime snow climate as described i n M c C l u n g and Schaerer (1993) . These snow climates are characterized by heavy snowfalls and m i l d temperatures.  In a maritime snow climate, avalanches usually occur during, or  12 immediately following, storm events and result from failures in new snow.  Full-depth  avalanches that release by gliding at the ground surface have a much more complicated response to climate conditions. Due to higher air temperatures in the area, deep instabilities do not persist as they do in a continental climate regime.  In the study area,  rain is likely to fall at any time during the winter.  Altitude influences the amount of incoming radiation, temperatures, and wind exposure. The Summit weather station is situated at 1230 m a.s.l.  The glide site is located at 1450  m and the radiation site at 1600 m. Differences in elevation between the instrumented locations are not significant enough to affect the analysis, although there is the possibility that precipitation as rain at the Summit station may occur as snow at the glide site, or on the slopes above it.  Moist, warm air from the coast moves up the Fraser Valley and as it is pushed eastward through the Cascades, air moves to higher elevations and precipitates as snow.  Large-  scale movement of air through the Coquihalla and Boston Bar Creek valleys is influenced by the relief and orientation of the Cascade Mountains. Meteorological data available from nearby monitoring stations in Hope, 50 k m south-west of the study area (49° 2 2 . 5 ' N 121° 2 6 ' W ) , and i n Merritt, approximately 60 k m north-east (50°5.5'N 121° 4 7 ' W ) .  The  climate at Hope (elevation 46 m) approximates that of a coastal or maritime location, while Merritt, located on the periphery of the Interior Plateau (elevation 580 m), has a more continental climate (Table 2.1).  The temperature summary indicates that Merritt is significantly cooler than Hope during  13 the winter months. Temperatures during the two study periods approximated those at Merritt.  O n average, winter temperatures are lower than those at Hope.  It is  documented that temperatures during the winter months (Nov-Mar) rise to 0° C an average of 63 times annually (1973-1977)(Ministry o f Transportation and Highways ( M o T H ) , 1980). The 1992-93 season was generally cooler than the following season. Monthly mean temperatures, recorded at the glide site, were lower for all but November and A p r i l .  T a b l e 2 . 1 : Temperature S u m m a r y f r o m G l i d e Site  Nov  Minimum Maximum -1.2 -4.8  Dec Jan Feb Mar  -9.1 -8.9 -7.0 -3.4  -5.5 -4.3  Apr  -1.1  3.8  -1.7 2.0  Avg -4.0  Minimum Maximum -2.5 -5.8  Avg -5.0  -7.9 -7.6 -5.5  -3.5 -1.8 -6.8  1.0 -2.4  -2.2 -1.0 -5.4  -1.6  -2.2  3.7  -0.3  0.7  -1.4  2.6  0.1  0.6  Merritt*  Hope*  1993-94 Temperatures  1992-93 Temperatures  Minimum Maximum Minimum Maximum 4.6 -3.8 7.6 1.7 0.2 -7.4 4.1 -0.9 -2.7 -10.8 2.3 -3.1 3.0 -6.4 6.9 -0.2 8.1 -3.5 10.1 1.0 3.9  14.6  0.2  13.5  * Hope and Merritt data from Climate Normals (1951-1980)  Precipitation characteristics at Hope, nor Merritt, are not representative o f conditions i n the study area (Table 2.2). The study area receives about the same number o f rain days as Merritt, and about a quarter o f those experienced at Hope during the winter months. A t Hope, a large proportion of the total winter precipitation is rain, whereas at Merritt, a greater proportion o f the total is made up of snow.  A t the study site snow makes up  more than two-third o f the total. The amount o f rainfall measured at the study site for two seasons almost doubles that of Merritt. Annual average winter precipitation (OctM a y ) between 1973 and 1978 is 1293 m m at the Summit weather station (Minstry of the  14 Environment ( M o E ) , 1985). In 1992-93 and 1993-94 winter precipitation measured between November and A p r i l was 801 m m and 1011 m m respectively.  T a b l e 2.2: P r e c i p i t a t i o n S u m m a r y f r o m S u m m i t W e a t h e r S t a t i o n  Merritt*  Hope*  1993-94 S e a s o n  1992-93 S e a s o n Rain  Snow  Snow  D a y s of  (mm)  (mmWE)  Rain  (mm)  (mmWE)  Rain  (mm)  (mmWE)  Rain  (mm)  (mmWE)  Rain  Nov Dec Jan  21  159  6  25  119  3  208.5  15.3  18  19.6  15.8  5  1  183  1  38  120  3  249.1  40.3  18  12.2  31.6  4  22  125  2  53  140  7  184.1  72.6  15  12  38.6  3  Feb Mar Apr Total  0  12  0  36  192  2  165.2  30.4  16  10.5  16.7  3  42  124  6  62  143  4  131.2  16.1  17  7.4  9.1  4  34  78  11  41  42  6  103.3  1.5  16  8.4  1.9  4  120  681  26  255  756  25  1041.4  176.2  100  70.1  113.7  23  D a y s of Rain  Snow  D a y s of Rain  Snow  D a y s of Rain  * M e a n m o n t h l y p r e c i p i t a t i o n f r o m C l i m a t e N o r m a l s (1951-1980)  L o c a l climate, which can be quite different from the regional climate, is influenced by topography.  A t the scale of this study, the influence of topography on snow conditions is  important. The relief, shape, and orientation o f the landscape with respect to prevailing winds is important for overall climate conditions. Slope angle and aspect cause extreme differentiation i n climate on a local scale. A t the study site, steep south-facing slopes are exposed to incoming solar radiation throughout the year, moderating air and snow temperatures. aspect.  A l l avalanche paths included i n the study have a south to south-west  The importance of aspect is evident from two seasons o f observations.  Full-  depth avalanches are much less frequent on an avalanche path having the same bare granite, steeply-sloped surface but north-east aspect. Glide cracks may form but release seems to be much rarer than on southerly aspects. If a slab releases o f a N E aspect, it is usually very late i n the season when daily mean air temperatures are w e l l above zero.  15 2.4 The Coquihalla Avalanche Area  The study area is situated within the Coquihalla Avalanche Area which is situated along the Coquihalla Highway (Highway N o . 5) near the summit of the route. The area includes approximately 17 kilometres of the highway which are affected by avalanches.  Six years of detailed avalanche records were collected by the "Avalanche Task Force" of the Ministry of Highways ( M o H , 1978) before construction of the highway i n order to fully assess the avalanche hazard. There are 108 avalanche paths in the Coquihalla Avalanche Area, of which 67 affect the highway.  O f these, at least 10 paths are prone to  full-depth avalanches that release by gliding.  Avalanche occurrences i n the Coquihalla Avalanche area are monitored by the Avalanche Section of the Ministry of Transportation and Highways.  Table 2.3 summarizes the  avalanche occurrences from the two seasons of record. O f all the avalanche occurrences recorded by the Ministry of Transportation and Highways approximately 2 0 % of the total number are artificially triggered.  Artificially released occurrences are not included i n the  data set for this study. A large proportion of all occurrences within the Coquihalla Avalanche Area are naturally occurring. W i t h i n the study area, comprising of 9 paths on Zopkios Ridge, a significant proportion of all the avalanches are full-depth avalanches (51% i n 1992-93 and 8 5 % in 1993-94). Twenty-eight occurrences in 1992-93 and 78 occurrences i n 1993-94 available for analysis.  Several requirements were used to select full-depth avalanches suspected to have been  16 triggered by snow glide.  These include: the release level of selected avalanches i n this  study must be recorded as being at the ground, not i n new or old snow; the moisture content of the avalanche deposits must be moist to wet; starting locations of the occurrences of interest are at the top to the middle of the avalanche path, where bare bedrock surfaces generally prevail. Full-depth avalanches are not likely to initiate on the lower sections of the track because of the vegetative cover.  Table 2.3: Summary of Avalanche Occurrences 1992-93  1993-94  189  305  # Artificially Triggered  44  61  # Naturally Triggered  145  244  Total # Naturally Occurring Avalanches in Study Area  51  92  # Full-Depth Avalanche Occurrences  26  78  Total # Avalanche Occurrences in Coquihalla Avalanche Area  Table 2.4 shows the occurrences at each of the selected paths in the study area. Using this record, "active" periods of avalanche activity are identified. F r o m this table, the frequency of avalanches at individual paths is apparent. Some avalanche paths have a greater frequency of avalanches than others.  Sometimes more than one avalanche  occurrence are recorded for an individual path on the same day.  This indicates that more  than one occurrence may occur within a specified path, or that perhaps several occurrences may be recorded during the same day. Individual avalanche path characteristics are noted i n Appendix A .  Individual path characteristics, outlined in Table 2.5, indicate that paths have almost  17  Table 2.4: Full-Depth Avalanche Occurrences on Paths in the Study Area  1992-93 Season "action" periods  A v a l a n c h e Path Number 46.3  Dec. 1 0 - 2 4  46.5 12/12  46.8  47.05  47.15  47.35  47.55  47.7  12/15 12/23 1/23  Jan. 23 - Feb. 6  1/25  1/25 1/28  1/30 1/30 1/31  1/25  1/25  1/30  1/30  1/28  1/31 1/31  1/31  1/30 1/30 1/31  3/7  3/16  Mar. 16 - 27  3/23 3/25  3/25  Total # A v a l a n c h e s  1993-94 Season  A v a l a n c h e Path Number  "action" periods  46.3  46.5  46.8  Dec. 8 - 1 8  12/10 12/10 12/11  12/11  12/11  47.05  12/11  1/2  Dec. 30 - Jan. 15  1/4 1/4  1/4 1/4  1/2 1/3 1/4 1/4 1/4  47.15  47.35  47.55 12/10  12/11 12/11 12/13 12/15  12/11  47.7 12/10  12/15  1/4  1/6 1/7 1/9 1/11  1/11 1/13  1/12 1/13  1/13  1/13  1/13 1/20 1/22  1/22 2/28  Feb. 1 8 - Mar. 6 3/1 3/2  3/2  3/2  1/26 2/28 2/28 2/28 2/28 3/1 3/1  3/2  3/2  3/2 3/9  3/9  3/9  3/9 3/9  3/10  Mar. 7 - 29  3/14 3/14 3/15  3/14 3/14 3/15 3/15 3/15 3/15  3/14  3/13  3/14 3/15  3/28 3/29 3/30  Total # A v a l a n c h e s :  TO"  T6~  T5~  18 identical relief, or vertical fall.  Starting zone inclinations vary between 34° and 46° and  are located in bedrock bluffs at the top of the path. Inclinations of the avalanche track vary from 24° to 32° and run-out distances generally reach at least as far as the highway.  Table 2.5: Study Area Avalanche Path Characteristics Path No.  Start Zone Incline  Track Incline  Runout Incline  Vertical Fall  46.3  39°  29°  -  1220 m  46.5  37°  26°  4°  1225 m  46.8  37°  31°  21°  1280 m  47.05  39°  24°  15°  1250 m  47.15  45°  28°  25°  1250 m  47.35  42°  32°  -  1220 m  47.55  46°  31°  12°  1250 m  47.7  39°  28°  14°  1250 m  47.9  34°  28°  8°  1250 m  19  Chapter 3: S N O W G L I D E  In this chapter, the characteristics and nature of snow glide are described.  Theoretical  development of a constitutive equation for snow glide is presented and parameters important to the variation i n glide velocity are identified. F i e l d instrumentation and data acquisition for snow glide and the characteristics of snow glide at the study site, determined from these measurements, are also presented.  3.1 General Characteristics of Snow Glide  The nature of snow glide on a smooth surface was first investigated in the laboratory by Haefeli (Bader, et a l . , 1939).  Snow glide over smooth rock, grassy vegetation, and bamboo-  covered slopes is described around the world by authors in Switzerland (in der Gand and Zupancic, 1966), Austria (Lackinger, 1987), Japan (Endo, 1983), and i n the Cascade Mountains of northwest U S A and Canada ( M c C l u n g , 1975; M c C l u n g , et a l . , 1991; M c C l u n g , et a l . , 1994). Research on the process of snow glide is somewhat limited due to the difficulties of installing field instrumentation.  Glide is possible on alpine terrain steeper than about 15° ( M c C l u n g et a l . , 1994).  Glide  velocity increases with slope angle and becomes more rapid i n zones of high longitudinal stress gradients such as on convex shaped slopes.  Everything else equal, snow on south-  facing slopes, i n the northern hemisphere, exposed to greater radiant heating, tends to glide faster than on northern aspects. A reappearance of glide cracks i n identical positions on a slope was observed by Lackinger (1987) and led to the belief that glide is gradient-  20 dependent.  A s snow glide is the movement of snow over the ground, it follows that the mechanics of snow glide are not only dependant on slope angle and snow pack properties, but are greatly influenced by the interaction of the snow pack with the ground surface. ground surface such as bare rock or grass is necessary for glide.  A relatively smooth  Ground roughness, which  may include colluvium or vegetation, provides resistance to glide.  3.2 Development of the Snow Glide Model  Snow grains at the glide interface and several c m above it are well-settled, well-rounded and saturated over most of the winter. To develop a comprehensive and realistic model for snow glide, an understanding of the mechanics of snow and, in particular, wet snow is necessary. Properties of wet snow are particularly important since the movement of an intact snow pack across a wetted surface is being modelled.  Modelling the mechanics of snow is a complicated task, as it behaves as a nonlinear viscoplastic material (Mellor, 1977). In general, the tangential drag, or shear resistance, of the snow pack is considered as the down slope component of the normal stress on the interface roughness elements (r=pgHsin^). for snow glide (Table 3.1).  There are many different constitutive equation proposed  Different theories arise from the uncertainty associated with the  rheological description of snow and from the mechanisms thought to contribute to snow glide.  Snow is highly temperature sensitive and compressible, causing it to behave  differently from many granular materials.  21 Table 3.1: Different Constitutive Equations for Snow Glide Haefeli (1939) - dry friction  T  Salm (1977) - adhesion & viscous shear of water f i l m  r  in der Gand and Zupancic (1966) - "thick" lubrication layer & dry friction  =  M °z  = c' + (njs)  s  u  T , = 0?i/d) U + n o  z  M c C l u n g (1981) - stagnation depth as apparent boundary layer Salm (1977) - combination of glide and creep  T  S  = (Vd')  7"sa = (V<*' P  T  u  + yJ&)  U  = c + a tan<£ where c = c' + (rjjd' + r; /5)U  Lackinger (1987) - M o h r - C o u l o m b mechanics  s  z  r  w  M c C l u n g (1987) - effect of free-water on material properties and friction W h e r e : T is t h e s h e a r r e s i s t a n c e ;  s  r  s  = U/2(1-J»)D* m  is t h e c o e f f i c i e n t o f d r y f r i c t i o n w h i c h is a f u n c t i o n o f g l i d e v e l o c i t y ,  S  t e m p e r a t u r e , a n d n o r m a l s t r e s s (<r); c ' is a d h e s i o n ; r; is t h e v i s c o s i t y o f w a t e r a t 0° C; 5 is t h e t h i c k n e s s o f t h e 2  w  l u b r i c a t i n g w a t e r f i l m (S = 10" c m ) ; ij is t h e v i s c o s i t y o f t h e t h i c k l u b r i c a t i o n l a y e r , t h e t h i c k n e s s b e i n g d 6  L  (d=0.8 c m ) ; r\ is t h e s h e a r v i s c o s i t y o f s n o w ; d ' is t h e s t a g n a t i o n d e p t h w h i c h is a f u n c t i o n o f g e o m e t r y , s n o w s  properties,  l u b r i c a t i o n l a y e r , a n d the glide m e c h a n i s m ;  t  is t h e a p p a r e n t s h e a r r e s i s t a n c e ; a n d ,  <j>  r  is t h e  r e s i d u a l f r i c t i o n a n g l e , v is t h e v i s c o u s P o i s s o n R a t i o , D * is t h e s t a g n a t i o n d e p t h m o d i f i e d b y a b a s a l w a t e r film. F r o m : L a c k i n g e r (1987)  Constitutive equations developed for snow glide may be used to illustrate the mechanisms that contribute to fluctuations in glide velocity.  Salm (1977) states that the shear stress  between a snow pack and a relatively smooth surface is explained by a stress-independent cohesion and a Newtonian viscosity,  i n der Gand and Zupancic (1966) propose that shear  stresses are composed of dry friction and Newtonian viscous friction i n a thick lubrication layer.  Their proposition implies that the wet boundary layer at the snow/ground interface is  quite thick (0.8 cm) while field observations show that this is not always the case.  W h e n the  snow pack conforms to a fully saturated glide interface, it was originally thought that the single mechanism for movement is by creep deformation around surface roughness elements  ( M c C l u n g , 1981).  This is based on results by Nye (1969) on the sliding of temperate  glaciers over their beds.  Lackinger (1987) applies the Mohr-Coulomb law to snow  mechanics which requires a residual friction angle for snow which is not currently known and w i l l not be constant.  Assuming that snow properties are uniform on a planar slope, and that snow deforms as a linear, viscous material. The constitutive equation relating the tangential drag on the snow pack, or local basal shear stress, T , to the glide velocity, U , is modified by M c C l u n g and Clarke (1987) to:  [3-1]  T -  2(1  -v)D*  where D* is a length parameter equivalent to the stagnation depth, and n and v are the shear viscosity and viscous Poisson ratio of snow at the interface.  Changes i n snow material  properties and friction conditions at the interface due to the presence of free-water are taken into consideration. The shear and bulk viscosity (77) are functions of density, temperature, and snow type.  The Poisson ratio, v, contains the effects of the bulk viscosity as p=(3r]-  2/*)/2(3TI+/X) (Mellor, 1975).  Equation [3.1] assumes that material properties of the snow  are constant when, i n reality, material properties of the snow pack are temporally variable. The shear viscosity and viscous Poisson ratio are influenced by changing density and by changing water contents i n the snow pack ( M c C l u n g and Clarke, 1987); therefore it may be realistic to hold these variables constant only over a period of avalanche occurrences.  In general, the length parameter D* is a complicated function of interface geometry and water distribution at the interface.  For incompressible flow ^=0.5, which indicates that for the  23 same geometry and viscosity, incompressible flow gives a higher drag. The stagnation depth parameter incorporates changes in friction conditions at the interface due to the presence of free-water at the interface.  When D* is much less than the snow pack depth, tangential drag  is more likely controlled by creep processes.  When D* is greater than snow pack depth, the  snow pack moves forward by sliding due to lubrication of the snow/ground interface.  Observations by M c C l u n g and Clarke (1987) indicate that most of the fluctuations i n glide velocity may be explained by the presence of water at the interface.  When a rough interface  is partially drowned, the resistance to snow glide must come from the roughness of the portion of ground which is not drowned since portions of the drowned surface impose no resistance.  Water also affects the material properties of the snow pack at the interface.  Water reduces viscosity and hardness, resulting in lower shear resistance at the glide interface.  The constitutive equation for snow glide presented i n this chapter w i l l be used to  qualitatively examine the importance of various factors to fluctuations in glide rates.  3.3 Snow Glide Instrumentation and Measurement  In der Gand and Zupancic (1966) first developed snow glide instrumentation, upon which installations for this study were based. Lackinger (1987) used similar instrumentation on a 30-35° grassy, south-facing gully and a steeper (36-42°) slope interspersed with rocks. Measurements at the study site were established by D . M c C l u n g i n 1987 (see M c C l u n g et a l . , 1994).  Snow glide instrumentation is situated immediately adjacent to path #47.9 (see Location M a p  24 - Figure 2.2) at an elevation of 1450 m. The glide site is characterized by the same bare granite as the adjacent slopes.  The slope of the instrumented site (31°) is steep enough for  gliding to occur but it is not steep enough for avalanching. If a steeper site had been chosen, then measurements would perhaps only record snow glide up until the first avalanche. However, rates of snow glide at the site are considered representative of conditions on the upper slopes.  Snow glide measurements at the study site are meant to represent conditions on the adjacent slopes of Zopkios Ridge.  This is considered to be a fair representation since the character of  the smooth rock surface is the same. A n y differences in timing between the two may be attributed to a couple of factors.  The slope of the glide site is slightly less that some of the  avalanche start zones resulting perhaps i n a moderately slower glide rate. Perhaps of greater significance is the proximity of the upper slopes to exposed portions of bare rock.  Warming  of the exposed rock by incoming solar radiation w i l l result in higher rates of snow melt.  At  the instrumented glide site, the rock step, when exposed, w i l l simulate this effect to some degree. Results from this research w i l l ascertain the suitability of the instrumented location to represent snow glide conditions on adjacent slopes.  The snow glide instruments consist of a stainless steel "shoe" placed on the end of strong cord wound onto a drum inside a weather-proof enclosure (Figure 3.1).  The drum is geared  to a potentiometer that records changing resistance as the cord is pulled out. The resistance is calibrated at the beginning of the season to an equivalent distance of snow travel. Precision of the instrument is approximately +10%. data logger powered by two 12V batteries.  Measurements are recorded hourly by a  Data are downloaded into a computer via  25 telecornmunication and a modem.  The cord attached to a given shoe has a maximum displacement of approximately 3 m and it is not uncommon for some of the faster gliding shoes (Shoe #3 and Shoe #5) to run-out before the end of the season. In this study, jerky movements due to "stick-slip" behaviour as the snow pack passes over the ground sometimes occur. Yamada et al. (1991) improved on snow glide instrumentation by placing a gear box into a recess in the ground with the teeth exposed to the snow interface.  Gliding rotates the gear giving a displacement measurement  of the snow pack over the ground. This avoids the error related to the interaction of the glide shoe with the ground roughness.  Such instrumentation would be extremely difficult to  install at this site due to the hardness of the rock.  F i g u r e 3 . 1 : G l i d e Shoe Instrumentation  26 F i v e shoes were mounted on the slope, with relative placement designed to determine how glide varies with position on the slope. Figure 3.2 illustrates their placement with respect to a large joint plane that creates a step i n the slope. The step i n the rock, below which the snow pack separates to form an open crack, may simulate the conditions around a glide crack.  Shoe #1 is placed above the step and Shoes #2 and #3 are placed i n line directly  below the step. Shoe #4 is placed on one side of the centre alignment and Shoe #5 is placed on the other side, strategically placed to intercept a path of water.  This water, draining from  the slopes above, is visible even during the summer months.  Temperatures, recorded at the glide site using thermistors, include air temperature, temperature at the snow/ground interface, temperature at 1 c m depth i n soil, and temperature at the rock step face.  Measurements at the rock step would record the large temperature  variations due to radiative loading on the exposed rock surface.  These measurements are  valuable when the step is exposed but are often lost when the thermistor becomes buried i n snow.  Hourly recordings of glide displacements and temperatures are summed over 12 hours  for analysis. The two 12-hour time periods are divided into daytime (0600-1800) and nighttime (1800-0600) cases.  In addition to automated data collection, the glide site was visited several times over the winter season to survey a snow profile.  Snow pack observations were used to determine the  normal and shear loads on the slope and the water content (by volume) through the snow pack.  Observations of the snow pack near the interface show that a wet layer (between 1 and  3 c m i n thickness) is always present.  Volumetric water contents at the interface, measured  using a dielectric measuring device, are between 8-15%. M a x i m u m snow accumulations at  Figure 3.2: Placement of Glide Instruments on the Slope  Shoe # 4  ~ 2.0 m  I Shoe # 2 Interface Temp.  Shoe # 3  Top View  Shoe # 5 1 c m Soil Temp.  28 the field site reached depths of about 1.5 m during both seasons.  3.4 Snow Glide Characteristics at the Field Site  Characteristics of snow glide were determined at the same field site in the 1988-89 and 198990 winter seasons (refer to M c C l u n g et al, 1994). F i e l d measurements reported here for 1992-93 and 1993-94 augment this research. Measurements over two seasons show both steady and fluctuating components of snow glide. M c C l u n g et al. (1994) thought that steady glide rates ( < 1 mmd' ) might occur when separation of the snow pack from the ground is 1  minimal and down slope movement occurs by creep.  The fluctuating component of snow  glide (several cmd" ) can be attributed to local separation of the snow pack from the ground 1  surface since the same fluctuations could not be explained by variations in material properties.  3.4.1 Seasonal Nature of Snow Glide  Snow glide data over two winter seasons are presented in Figure 3.3 a) and b) for different glide shoes.  The general nature of snow glide through the season is apparent. During the  early season, heat storage in underlying rock, thin snow packs, and relatively low snow densities contribute to high early season glide rates. In the spring, or late season, observations confirm that higher rates of glide become apparent. This may due to the increase i n the supply of meltwater to the interface and rain-on-snow events which w i l l be examined further. Glide velocities during the mid-season range from periods having high rates of glide to periods of near zero velocities.  Figure 3.3: Seasonal Nature of Snow Glide a) 1992-93 Season b) 1993-94 Season  30 Table 3.2 summarizes glide rates at each shoe during two study seasons.  A n n u a l mean glide  rates at each shoe are higher during the 1993-94 season. Monthly mean glide rates are highest i n November.  In 1993-94, only rates during February were lower than i n the  previous season. These differences in glide rates may be attributed to differences i n meteorological conditions between the two seasons.  Both seasons show large fluctuations i n  glide velocity during the early period of record i n fall, and moderate fluctuations i n the spring.  During the mid-season period, periods of higher glide rates are apparent.  Table 3.2: Summary of Glide Rates  1992-93 M e a n Glide Rates (mm/day)  1993-94 M e a n Glide Rates (mm/day)  S h o e #1  S h o e #2  S h o e #4  Shoe #5  Avg  S h o e #1  S h o e #2  S h o e #3  S h o e #4  Nov  13.0  8.9  18.0  17.0  14.2  18.2  13.4  23.4  27.2  Dec Jan  6.0  5.3  7.8  11.0  7.5  3.5  11.0  12.1  3.4  1.0  9.4  6.9  5.2  5.4  15.0  17.2  Feb  1.9  4.8  6.1  9.6  5.6  1.2  3.4  5.4  7.6  4.4  Mar  1.9  6.6  6.9  1.5  4.2  3.8  15.0  38.6  2.4  9.3  21.5  -  11.1  2.3  4.9  10.3  -  19.0  Apr  -  Avg  4.8  6.0  11.6  9.2  5.7  10.5  17.6  19.7  17.3  Shoe  #5  Avg  25.6  21.6  15.2  15.5  11.5  21.4  20.6  15.9  5.0  3.4.2 Diurnal Nature of Snow Glide  In this study, data for two field seasons are used to determine diurnal variations i n rates of glide.  Linear regression of daytime (0600-1800) versus nighttime (1800-0600) glide rates  showed that there was little distinction between the two (Figure 3.4 a) and b)).  This  corresponds with observations by M c C l u n g et a l . (1994). Outliers identified i n the analysis belonged to early season dates, when thin snow packs would be most influenced by daytime vs. nighttime weather conditions and when glide rates are characterized by large fluctuations  Figure 3.4: Diurnal Variation in Glide Rates (measurements from Glide Shoe #4) a) 1992-93 Season b) 1993-94 Season  O  10  20  30  Nighttime Glide Velocity  0  10  20  30  Nighttime Glide Velocity  40  50  (mm/day)  40 (mm/day)  50  32 N o significant difference between daytime and nighttime snow glide was found when daytime glide rates were taken from 1000 to 2200. Despite incorporating a 4 hour lag time for snow pack warming, there is no support for the notion of faster glide rates during the warmer daytime periods.  Reasons for the lack of distinction between daytime and nighttime glide  rates may be a reflection of the sample size. Using the entire season for this purpose may not be appropriate.  Snow glide response is a complex function of several meterological  factors including rain events, which occur at any time during the day or night. The impact of diurnal patterns of air temperature and radiation conditions is more effectively examined in greater detail over shorter time periods.  3.4.3 Spatial Variation of Snow Glide on a Slope  Variation i n glide rates across the slope is observed.  The-snow pack does not move down  slope as a rigid slab, so a variation i n rates at different locations can be expected.  Local  variations i n ground roughness and i n water supply contribute to variations i n glide velocity across the slope.  The distribution of rates of snow glide on a slope w i l l reveal the most effective position for the glide shoes.  Glide rates increase with position down slope especially on convex-shaped  slopes ( M c C l u n g et a l . , 1991). Measurements from in der Gand and Zupancic (1966) indicate that glide velocities measured below a glide crack are much more rapid than at any other instrumented location on the slope.  If the rock step at the study site simulates a glide crack, or tensile fracture i n the snow pack,  33 then glide rates recorded at different shoe positions may indicate spatial variation on the slope with respect to a free uphill surface. Table 3.1 shows that glide rates at Shoe #1, located above the rock step, are, on average, lower than all others.  Rates are higher for  Shoes below the crack for a number of reasons: a free uphill surface provides an avenue for water percolation to the interface, and also provides less resistance to down slope movement. Figure 3.5 illustrates the distribution of snow cover at the glide site with respect to the rock step for two scenarios. The rock step can be exposed to warming from incoming solar radiation and water percolation from the slopes above.  After snow accumulations, and when  snow slides from the step to the pack below, the step may be closed and the supply of freewater may be slowed.  Figure 3.5: Schematic Showing Distribution of Snow Cover at the Glide Site for Two Different Scenarios  34 3.5 Snow G l i d e a n d F u l l - D e p t h Avalanche Release  Full-depth avalanches are slab avalanches that release at the ground.  Starting locations for  these avalanches in the study area is almost always near the top of the path where slopes are steep and bedrock is exposed. Peak.  Figure 3.6 illustrates a slab release from the slopes of Y a k  Note the fracture line across the slope.  fracture line  F i g u r e 3.6: F u l l - D e p t h Avalanche Release on Y a k Peak  Tensile fractures, called glide cracks, form perpendicular to the ground surface ( M c C l u n g , 1987) and develop at the crest of an avalanche.  Glide cracks precede full-depth avalanches  but avalanche release is not always immediate. During glide crack formation, rates of snow glide are high (Endo, 1983) and increase down slope from a glide crack, meaning that there is a longitudinal strain gradient (Yamada, et al, 1991; M c C l u n g and Schaerer, 1993). Stability of the snow pack after crack formation depends on interface topography and friction.  M c C l u n g (1987) presents an idealized one-dimensional model for glide crack initiation.  The  model describes a "breakdown" zone (x') down slope from the glide crack where a loss of basal shear strength and an increase in glide velocity is exhibited. There is also a "recovery" zone (x ), a region of constant glide stiffness, up-slope of the breakdown zone (Figure 3.7). 0  Figure 3.7: Schematic of Assumed Basal Shear Stress, Glide Velocity and Geometry Definitions for the One-Dimensional Glide Crack Model From: McClung (1987)  A dovinslope <  36 It is thought that intervals of rapid gliding precede, or signal, the release of full-depth avalanches but no threshold rate or predictive equation has been proposed.  Studies by  Lackinger (1987) showed that an increase in the frequency and magnitude of Micro-seismic emissions ( M S E ) signalled the impending release of a full-depth avalanche, approximately 3 hours prior to the event. The increase i n emissions may have been be attributed to an increase in glide velocity.  Yamada, et al. (1991), using glide rates from a fixed position on  the slope, determined rates of strain in order to apply prediction models.  This type of  analytical application is difficult for this study since distances between glide shoes are not fixed on the slope and, more importantly, since avalanches do no occur at the glide site, strain rates would not be the same as those on the avalanche prone slopes, nor would they indicate critical rates of strain associated with failure at the ground surface.  Figure 3.8 shows the glide velocities and full-depth avalanche occurrences for two winter seasons. In most cases avalanche release corresponds with higher rates of glide, or the increase i n rates of glide. threshold velocity.  However, avalanche release does not seem to be associated with a  N o avalanche occurrences were recorded i n the early part of the season  when glide rates are highest. Reasons for this may be that snow accumulations were not enough to cause avalanches, or that avalanches were not of a sufficient magnitude to be recorded.  In February of both seasons, very little glide activity occurred and there were also  no avalanches. It should also be noted that single-avalanche days are not necessarily small i n magnitude (see Appendix A ) .  This indicates that these days are still a concern and may not  be disregarded.  During the 1992-93 season, avalanches usually released when glide rates were higher than  37  Figure 3.8: Glide Rates and Full-Depth Avalanche Occurrence a) 1992-93 Season b) 1993-94 Season  38 10-15 m m d but not i n all cases. During the 1993-94 season, avalanches tended to occur 1  when glide rates were higher than 15 mmd" . Due to the variation in glide rates across the 1  slope and due to the remote representation of avalanche slopes, a threshold rate of glide could not be determined for the data set.  The glide measurements, recorded on a less steep slope with different boundary conditions, are the best indicator to indicate instability of the snow pack at the snow/ground interface. The record for this study shows peaks in glide rates after avalanche release i n many cases. This may indicate that the same process is present but is delayed a certain amount of time. Glide rates, i f instruments were placed higher up on the avalanche path, may peak with avalanche release.  F o r this study, other parameters which could help pinpoint times of  avalanche release must be identified. Rates of snow glide, recorded at the glide site indicate times of accelerated motion immediately prior to avalanche release.  This accelerated velocity  may indicate a change i n friction conditions that is taking place faster than the snow pack has time to which to adjust.  The record of full-depth avalanche occurrences can be sub-divided according to possible trigger mechanisms. Table 3.3 lists the frequency of various trigger mechanisms for both seasons.  A l l but a few avalanches are thought to have been triggered by the contribution of  free-water to the snow pack. For this reason, subsequent chapters of this study examine the impact of rain-on-snow events, warm, convective periods, and clear-sky days on snow glide and full-depth avalanche occurrence.  39 Table 3.3: Possible Full-Depth Avalanche Trigger Mechanisms 1992-93  1993-94  Loading (Snowfall Only)  3 (11.5%)  3 (3.8%)  Rain-on-Snow Events  8 (30.8%)  54 (69.2%)  Snow Melt (Warm Temps & Clear Skies)  15 (57.7%)  21 (26.9%)  3.6 Discussion  The snow pack moves over the ground surface by a combination of creep and glide. Mechanisms that influence glide include, slope angle, snow pack depth and density, and most importantly, the snow/ground interface conditions. The interface geometry, the presence of free water at the interface, and the nature of snow material properties at a saturated interface contribute to these interface conditions. A very thin film of water, which exists that the snow/ground interface, imposes no shear resistance, thus the area of water coverage affects the overall drag at the base of the snow pack. The effects of free-water on snow glide w i l l be further discussed i n Chapter 4.  The seasonal nature of snow glide at the study site confirms the general pattern of rapid glide rates and large fluctuations of glide rates i n the early season and the late season.  Results  show that despite different temperature and precipitation conditions of the two study seasons, the pattern observed from previous seasons ( M c C l u n g , et a l . , 1994) is still maintained.  The diurnal variations i n snow glide is subdued when the entire season is considered i n the analysis. Diurnal variations of temperature and radiation occur on time scales that are too  40 small to influence snow glide. impact on glide rates.  Perhaps it is the long-term trends i n warming which have an  This w i l l be discussed further in Chapter 5. It may also be possible  that, due to the complex combination of factors that contribute to variations i n glide, no diurnal pattern would be apparent.  Data show that higher rates of snow glide do correspond with periods of avalanche activity but that times of avalanche release are not necessarily coincident with peak rates. release does seem to coincide with times of accelerated glide.  Avalanche  Major cycles of avalanches  generally occur when glide rates at the glide site are greater than 10-15 mmd" . 1  41 Chapter 4: F R E E - W A T E R A T T H E S N O W / G R O U N D I N T E R F A C E  4.1 Introduction  Variations i n steady glide can be attributed to the presence of free-water at the snow/ground interface.  Free-water affects the material properties of snow and also promotes partial  separation of the snow pack from the ground. A t the interface, where components of roughness provide resistance to down slope movement, water acts to promote rapid glide by drowning some of the roughness elements, thus reducing the effective roughness of the ground surface (see Figure 4.1).  1-  WATER (  N e  t>RA&)  Figure 4.1: Effect of Water Layer on Surface Roughness  M c C l u n g and Clarke (1987) show that, depending on the geometry chosen, glide speeds increase exponentially with linear increases in water film thickness. In this chapter, the effect of water flow on snow glide w i l l be examined. It is thought that the nature and pattern of water flow across the slope beneath the snow pack w i l l influence the spatial nature of glide.  This phenomenon may partly explain the topographic influence of glide avalanche  42 release as suggested by Lackinger (1987). In this chapter, the possibility of pore pressure development within the saturated zone at the snow/ground interface is investigated.  F i e l d observations of water flow w i l l be used to make general conclusions as to how changing free-water conditions affect snow glide.  Several time periods during two seasons of  data collection w i l l be focussed upon for more detailed analysis of the influence of free-water contributions on snow glide and the release of full-depth avalanches.  4.2 Effect of Water on Snow Rheology  The introduction of free water to a snow pack affects the material properties of snow.  When  free water is present at a smooth, impermeable interface, the effect is particularly relevant to snow glide.  The mechanical and physical properties of wet snow differ significantly from  those of dry snow.  When snow is wetted, inter-granular bonds are weaker and the pack has  a much lower cohesion. Wet, loose-snow avalanches are often initiated with the added loading of rainfall and the loss of cohesion i n the snow from percolating rainfall ( M c C l u n g and Schaerer, 1993).  The overall effect of higher normal loads is to increase shear stress at the base of the snow pack resulting in increased rates of glide.  This increase i n load must be quite large i n order  to activate full-depth avalanches. Results from two seasons indicate that there are very few occurrences that appear to have been triggered by loading (see Table 3.3).  The large temperature variations that drive metamorphism i n dry snow do not exist i n wet  43 snow.  Snow grains evolve in wet snow as water passes through the snow pack. Rapid  melting at the contact points between grains results i n an overall coarsening of grains. Wakahama (1965, 1968) conducted the first quantitative observations of grain growth i n wet snow.  The rate and extent of grain growth is influenced by the free water content.  When  water contents are greater than 15% (by volume), ice grains are completely separated by a continuous path of water.  Such water-saturated snow is classified as slush (Figure 4.2).  In  slush, smaller snow grains are quickly melted. Small grains disappear and relatively large particles grow by solid mass-exchange (Raymond and Tusima, 1979).  Snow having water  contents between 8-15% contains air bubbles that reduce the area available for heat conduction and act to retard grain growth (funicular regime). A t these lower water contents ice grains may be clustered i n well-bonded rounded polycrystals (Figure 4.3).  G r a i n growth is accompanied by a loss of shear and compressive strength (Male, 1980; Kobayashi et a l . , 1992) and this loss of strength is especially noted i n snow overlying impervious layers or over the ground (Colbeck, 1974). A t the glide site, snow profile observations indicate that snow grains reach some sort of equilibrium diameter of 1.0 to 2.0 m m over the entire season. This indicates that snow remains i n a funicular regime where air retards the growth of grains at water saturations between 8 and 1 5 % . A t the Coquihalla site depth hoar is rare, so large crystal grain sizes at the base of the snow pack are rarely, i f ever, present.  Since snow grains throughout the snow pack do not increase significantly i n  size, the decrease in shear strength due to grain coarsening w i l l not be very significant.  A decrease i n snow hardness with increasing water content is documented by Izumi and Akitaya (1985). They show that as water content increases from 0 to 2 2 % i n snow having  44  Figure 4.3: Rounded Polycrystal From: Colbeck (1987)  45 dry densities of 250-500 kg m , the snow hardness, measured using Kinosita's hardness 3  gauge, is reduced by a factor of 2 for coarse-grained snow and by a factor of 3 for finegrained snow.  Their results indicate that significant decreases i n snow hardness occur on  low density, fine-grained snow but the hardness of higher density snow (approx. 500 kg m~ ) 3  was shown to be nearly independent of water content.  Since snow at the interface, at the  study site, consists of this saturated, high density snow, the hardness may only be slightly reduced by changing water conditions.  A t the study site, snow pit observations over two seasons show that the snow/ground interface was always at 0° C and was always wet to varying degrees (Table 4.1). at the field site indicate a relatively high hardness at the interface.  Conditions  Values for hardness may  be indicative of the lowest centimetre or two and not the very thin f i l m of slushy wet snow adjacent to the interface. mechanics of glide.  A very stiff layer sitting above a soft slushy f i l m may influence the  The thickness of the slushy weak film w i l l limit the degree of ground  roughness that can be moved across.  If amplitudes of ground roughness are greater than the  thickness of the slush layer, there w i l l be a greater resistance to glide.  Table 4.1: Conditions at the Snow/Ground Interface Coquihalla Glide Site Hardness  Thickness (cm)  Grain Size (mm)  Temp. (°C)  Density* (kg/m )  Moisture  01/16/93  1.0  0.5  0  340  -  P  03/28/93  2.0  1.0  0  480  6.47  4F  01/30/94  1.0  1.0  0  380  11.2  P  02/09/94  10.0  1.0  0  500  13.5  P  03/06/94  0.5  2.0  0  370  9.6  P  Date  * Density is an average of bottom 10 cm due to size of sampling tube  3  (%)  46 The shear and bulk viscosity of snow increases i n the lower layers due to densification, but decreases i n the presence of free water (Haefeli, 1967). Investigations by Kobayashi et al. (1992) on the mechanical properties of slush determined that the breaking strength and compressive, or bulk, viscosity rapidly decrease with increasing water content.  The shear  viscosity of slush is quantified by Kobayashi and Izumi (1991) using a Coulette-Hatschek viscometer.  W i t h an increasing water content, the shear viscosity of snow decreases.  under slow rates of shear, has a higher viscosity than snow under more rapid shear.  Snow, In  relation to steady glide, slower rates of shear along the thin water-saturated layer are subject to viscous effects higher than that for water at 0° C . W i t h increasing rates of glide, the viscosity is lower and the resistance to glide due to viscous effects is lower.  This describes a  pseudoplastic behaviour.  If the bulk and shear viscosities of snow decrease similarly, the net effect on the viscous Poisson ratio, which is later incorporated into the equation for snow glide, may be small. This effect would result in a slightly decreased shear strength or a higher glide velocity.  An  extension to include a reduced viscosity in the lowest layer i n the form of a multi-layered viscosity model has not yet been attempted.  4.3 Effect of Water on Snow Pack Separation at the Snow/Ground Interface  Water at the snow glide interface reduces the friction between the snow pack and the ground due to a combination of material effects and lubricating effects of a thin f i l m of water. Water at the snow/ground interface drowns out some of the roughnesses which may lead to conditions of snow pack separation. These effects combine to induce higher rates of snow  47 glide.  A n effective coefficient of friction is described as the ratio of tangential to normal stress. Experiments by Haefeli revealed that friction between a block of snow and a glass surface decreased with increasing normal stress (Bader, et a l . , 1939).  W i t h a decrease i n friction  due to the changing nature of the properties of snow, water at the ground surface also affects the interaction with the snow pack. Water at the base of the snow pack may collect in low pressure zones i n the troughs of roughness. movement is small.  In these zones resistance to down slope  Expressions that relate the drag of a linear incompressible material to  the auto-correlation of bed topography are formulated by M c C l u n g and Clarke (1987).  They  also determined, mathematically, that it is the smallest elements that contribute the highest resistance to down slope creep (Based on field observations, it may be the larger elements that resist glide). It is assumed that no drag is imposed from drowned portions of the bed, since shear stress is given as (/i U/5 ) where 6 is equal to the water f i l m thickness. Their w  w  W  results show that there is an exponential increase i n glide speed (glide gain) with a linear increase in water layer thickness for a specific geometry (Figure 4.4).  The mean water layer  thickness (h) is divided by the mean square deviation of glide surface elevation (VJRJ, which is a measure of the strength of topographic fluctuations, to get a dimensionless expression.  The model of M c C l u n g and Clarke (1987) presumes that a substantial fraction of the bed must be covered with water to significantly reduce drag and to precipitate rapid gliding.  The  enhancement of glide rates is presumed to depend not only on the water thickness but the effect of separation may, i n fact, depend on the location of the separation and contact points (Lliboutry, 1968) which makes the area of coverage more important than the thickness.  48 Using water layer thickness as the only parameter reducing drag, implies dependence of drag on geometry.  The area drowned by a depth of water is a function of the ground roughness  (r = A/X). It, therefore, becomes important to look at the areal coverage of a water film to determine the effect on glide.  Glide Gain (U /U°) h  0.5  l.O  J.5  z.o  2.5  Dimensionless Water Layer Thickness (h/^R,,)  Figure 4.4: Increase in Glide Velocity with Change in Water Layer Thickness From: McClung and Clarke (1987)  4.4 Water Flow at the Snow/Ground Interface  Information on water flow through the snow pack is generally used by hydrologists to construct snow-melt hydrographs and to predict discharge of meltwater from the snow pack. In this study, the nature of water flow at the snow/ground interface is of interest because of its possible influence on gliding. Determination of the characteristics of water flow at the study site may provide information on conditions at the interface, where interaction between the snow pack and the ground surface is relevant to the process of snow glide. information w i l l later be used to determine the existence of pore pressures.  This  49 4.4.1 Movement Through the Saturated Zone  Modelling water movement through the saturated zone of a snow pack uses the principles of saturated flow derived from Darcy's Law.  Flows through the snow pack at the study site  are considered to be laminar and Darcian. Assuming a strip of hillside with a unit width and a constant (and small) slope angle (/3) Colbeck (1974) gives a continuity equation for the flow of meltwater as it reaches an impermeable boundary as:  \i where: p g |8 \i <j)  dx  [4.2]  intrinsic, saturated permeability of snow pack density of water at 0° C acceleration due to gravity slope angle (sin 0=0) viscosity of water at 0° C - porosity of snow pack  Equation [4.2] shows that the thickness of the saturated layer (h) varies with distance down slope (x) as the input (I) varies in space and time. The thickness of the saturated layer increases down slope but an assumption is made that it doesn't change faster than the change i n slope (jS).  It is also assumed that the rainfall or meltwater input is spatially uniform  across a planar slope.  This assumption is valid at the scale of an individual h i l l slope.  The  equation also contains the assumption that the saturated permeability is horizontally, and vertically, constant. This assumption is not valid for most snow packs but may hold for the short distances of this study. Equation [4.2] is equivalent to the kinematic wave equation; the basic equation governing flow in saturated soils.  The wave speed of a slug of water moving down slope through the saturated layer is derived  50 from the above equation as:  ^ P . „ k sinp  [4.3]  S  where k is the hydraulic conductivity. The first part of the equation is constant and describes the nature of the fluid passing through the matrix. To solve for equation [4.3] the intrinsic permeability and the porosity must be determined.  Permeability, unique to a porous medium, is independent of the fluid passing through it, and it depends on the geometrical structure of the material. Snow has a complicated structure so permeability can only be described i n statistical terms.  The specific permeability of fine-  grained compact snow is empirically derived from results on dry snow by Shimizu (1970) and is given as:  k  = 0.077 d  2  s  exp(-7.8  s  where G is the specific gravity of snow and d is the mean grain size diameter. s  uncertain how the relation holds for wet snow.  [4.4]  G) It is  Dunne et al. (1976) state that due to grain  growth i n wet snow, the ratio of saturated permeability (k ) to unsaturated permeability s  is equal to the square of the ratio of the particle sizes.  (k ) u  This analogy is irrelevant for this  study since the same grain growth is not experienced during the time scale of the experiment. Values for perjtneability determined for the snow at the snow/ground interface at the study site compare favourably with those published i n the literature (see Table 4.2).  The total porosity of snow is the fraction of the bulk volume that is occupied by voids. Voids which contribute to fluid flow are such that they connect both ends of the material  51 across which the pressure difference is applied. These voids are then termed "effective". The effective porosity is the fraction of the bulk volume of the material occupied by all effective voids. F o r snow, total porosity and effective porosity are almost the same. K n o w i n g the snow pack density from field measurements, and knowing that it is equal to the total mass of ice, water, and air (p = fp, + fj) + f p ) , where the f coefficients describe s  a  w  w  the volume fractions of ice, air, and water, porosity can be determined from free-water measurements made in the field. Measurements of percentage free-water (by volume) were made using a dielectric moisture meter.  Porosity, determined for snow at the snow/ground  interface at the study site, ranges from 0.25 to 0.36 (see Table 4.2).  4.4.2 Experimental Results  Several tests were conducted at the glide site in order to determine the nature of water flow at the base of the snow pack. The purpose of these tests was to determine whether or not water moves down slope i n established channels or as a film.  This information would help  establish how a slug of meltwater expands laterally down slope. F r o m information gained from this experiment, lag times between periods of snow melt, or rainfall, and periods of increased glide velocities w i l l be estimated. F l o w speeds measured i n this test are used later to determine i f positive pore pressures could develop at the base of the snow pack.  The experiment was conducted in the spring of 1994. Three tests were conducted at different locations on the glide site rock slab. O n that day, snow depth was 0.66 m .  Air  temperature was -3.0° C , snow surface temperature was -2.0° C , and skies were overcast which meant that melt water flow at the base of the snow pack was at least somewhat  52 minimized.  Conditions for the three tests are outlined in Table 4.2. A pulse of dye  (Rhodamine) was injected at the base of the snow pack and was traced a small distance down slope. Time of arrival was used to determine the particle speed or flow celerity.  Table 4.2: Conditions for Water Flow Tests Local Slope Angle  Saturated Zone Thickness  Water Content (by volume)  Density (bottom cm)  38°  2 cm  16 %  630 kg/m  Test #2  31°  3 cm  Test #3  41°  2 cm  Test  Porosity (calc.)  Permeability (calc.)  3  0.36  8.8xl0 m  2  720 kg/m  3  0.27  4.4xl0- m  2  630 kg/m  3  0.35  8.8xl0 m  2  1 0  #r >  17 % 6 %  10  1 0  Test results are shown i n Table 4.3. Results indicate that measured speeds for two out of three tests are within one order of magnitude of those derived using Colbeck's (1974) equation [4.3] but differ by a factor of four. This difference, once the uncertainties of the situation are considered, is acceptable. Two out of three tests showed that water flows at the base of the snow pack can be modelled with some degree of certainty using Colbeck's equation.  Two types of water flow at the interface were observed. Tests #1 and #3, indicated a slower, more uniform, flow.  Slower flow rates, measured in Since flows are slow through a  dense matrix, it is thought that the flow can be characterized by film flow.  Over the small  distance travelled down slope, a certain amount of lateral spreading was observed.  Test #2  indicates a faster, channel flow at the snow/ground interface. Figure 4.5 illustrates the dye conditions for Test #2, down slope from the dye injection site, where the saturated zone is  53 apparent.  The conditions for this test may indicate flow along a topographically-established  drainage network.  Water is channelized and is quickly conducted down slope.  lateral spreading was observed.  Very little  Channel flow beneath a snow pack involves a different  geometry and a different flux-concentration relationship than Equation [4.2] which is why Colbeck's theoretical equation does not work.  Table 4.3: Water Flow at the Interface - Experimental Results Experimental Results: From Glide Site (31° slope)  Rates of Flow in Literature:  Test#l: 0.216 cm/s Theoretical*: 0.941 cm/s  0.11 cm/s (10° slope)  Test #2: 6.600 cm/s ** Theoretical*: 0.556 cm/s  0.04 cm/s (calc.)  Fujino (1971)  Test #3: 0.157 cm/s Theoretical*: 0.650 cm/s  0.2 - 1.0 cm/s 0.03 cm/s  Male and Gray (1981) Colbeck (1974)  * theoretical values calculated using Equation  0.28 cm/s (25° slope) Dunne and Leopold (1978)  [4.3]  * * see text for discussion o f flow conditions  Channel and network formation beneath the snow pack should prevent the build up of excessive water pressure (Kattelmann, 1987).  This would imply that when a snow pack  drainage system is well-established, the effect of meltwater or rainfall inputs is less influential on glide rates, depending on the rate of water movement or discharge through the snow pack. The thickness of a saturated layer would not increase down slope as it would for a slower flow speed.  It is the slower f i l m flows that w i l l most affect glide rates, since there is more lateral divergence of the flow down slope since a larger surface area beneath the snow pack could be covered by a thin layer of water.  Channelized flow w i l l still affect glide rates since an  area of the basal snow surface is separated from the ground but, since channel flow rapidly  54 directs excess free-water down slope, there is little spreading of this flow.  The spatial  variability associated with glide is influenced by the spatial nature of water drainage at the base of the snow pack. Large-scale gliding, which incorporates the snow pack across the whole slope w i l l necessitate a larger areal coverage of the slope.  F i g u r e 4.5: Photo of Dye Conditions for Test #2  A t film-flow speeds of 0.2 c m s , time to travel 100 m, an approximate distance down slope 1  from a ridge for full-depth avalanche initiation, is approximately 14 hours. A t 6.6 c m s , the 1  speed associated with channel flow, time to travel 100 m is 25 minutes. Using results from  55 the experiment, one can predict response times of approximately 12 hours between a meltwater flux (rain or melt) and the commencement of more rapid glide velocities. Using continuously recorded rates of glide this response w i l l later be examined.  4.5 W a t e r Pressures i n the Snow P a c k  Water pressures at the base of the snow pack, if present, would support a fraction of the normal weight, thus reducing the resistance to shear. The development of pore pressures i n the bottom layers of a snow pack would then affect the constitutive equations for glide.  It is  very important to determine if these conditions can develop and what would be required for them to be significant.  Using a solution of equation [4.2], one can model the changing thickness of the saturated layer (h) with distance down slope from a divide (x).  Mx)  From:  = i  [4.5]  where I is an input rate (ms" ), and C is the celerity of water flow through the saturated 1  layer ( m s ) . 1  s  Using a porosity of 0.35, deduced from known water contents and densities,  and a range of flow celerities measured on a slope of 30°-40°, the thickness of the saturated layer (m) can be modelled.  Figure 4.6 shows the development of a saturated wedge down  slope from a divide (or perhaps a glide crack) for various input rates. The plot shows that flow rates associated with channel flow (0.066 m s ) do not allow the development of a 1  significant saturated layer thickness. The saturated layer thickness is shown be sensitive to input rates. Input rates of 15-25 m m / 6 hours are comparable to the size of average rain-  56 on-snow events i n the study area, while 100 mm / 6 hours may indicate a more extreme combination of rain and snow melt.  O n steeper slopes, flow celerity would be higher and a  thick saturated layer would be less likely to develop. Figure 4.7 shows that the pore pressures associated with the thickness of the saturated layer, calculated from u = h cos /37 , 2  w  are not particularly sensitive to changing slope angle.  F i g u r e 4.6: Thickness o f Saturated L a y e r w i t h Distance D o w n Slope f r o m Divide  1000:  0.001 -  n—; I  n  1—i—r~r 10  i I  Distance Downslope (m) Input Rate=15mm/6h  Input Rate=25mm/6h  m 100  1000  Input Rate=100mm/6h  In order to determine the critical pore pressure associated with failure at the ground surface, a M o h r - C o u l o m b approach is utilized. This approach is complicated by the fact that the internal angle of friction for snow is not known and must be assumed. (1987) the critical pore pressure for failure can be found from:  F r o m Keefer, et al.  57  u* = y z c o s p 2  (1-  t a n  P )  [4.6]  tanar Figure 4.8 indicates the changing critical pore pressures required for a 1.0 m deep snow pack o f varying densities and friction angles.  F o r this calculation a depth averaged snow  pack density is used. The plot shows that density has a larger effect on critical pore pressures within a snow pack having higher internal angles o f friction. Higher critical pore pressures are associated with higher friction angles.  Using a range o f friction angles (36° -  45°) for a snow pack having a depth-averaged density of 500 k g m , the critical pore 3  pressures range from 200-1500 N m " . This would be equivalent to a saturated layer 2.0 to 2  15.3 c m thick. O n the other hand, since tan|8/tan is a small number on steep slopes, the pore pressures w i l l be fairly small.  F i g u r e 4.7: Pore Pressures w i t h Increasing Saturated L a y e r Thickness  s =-,.:  -  —  —-=.  -==^0£±,  s  «  1 0s  ========  o  °-  0.1  I =-  -  O.OO01  -  -  -  -  •  >  - --------  __  ^ • ^ ^ ^  ^  - - .-  —  "  -  "  1—i—i i i 11 II 0.001  1—i—i l i n n 0.01  1—i—i i i 11 II 0.1  1—i i i i i n i  i i i i 111 II  1  Thickness of Saturated Layer (cm) Slope Angle = 20'  Slope Angle = 30'  Slope Angle = 40'  10  i i i 100  58 The height of water needed to raise the water table to that critical height is calculated from (u/jv,)*^,  where n,, is the effective porosity. This height w i l l range from 7 m m of rainfall  at the lowest extreme, which is very possible to achieve, to 53.6 m m at the highest, a less common occurrence but still possible on days of combined rain and snow melt i n the spring.  F i g u r e 4.8: C r i t i c a l Pore Pressure w i t h C h a n g i n g Density a n d F r i c t i o n A n g l e 2500  CM  2000-  E z  For Slope Angle = 35 1500-  o 1000 a.  500  400  450  500  550 600 Density of Snow Pack (kg/m3)  Friction Angle = 36'—— Friction Angle = 38'  Friction Angle = 40'  650  700  Friction Angle = 45'  Others have rejected the possibility of pore pressure development at the base of the snow pack. M c C l u n g (1981) proposes that i n order for the porous medium effect to be significant, the half-wavelength of the roughness elements should be compared to the diffusivity (c) for water flow i n snow.  This would require:  U, c  2  [4.7]  for the porous media effect to be significant. The wavelength o f roughness elements at the study site varies from 0.5 m m to 0.5 m approximately.  The equation is satisfied only for  59 diffusivities close to that of clay ( 1 0 rrrV ). Values for snow are not exactly known but 6  1  may be up to 4 times that for clay. Therefore, steady glide is slow enough to permit dissipation of pore pressures.  Snow pack separation may be affected by the pressure of free-  standing water at the interface although this is not generally observed in the field.  M c C l u n g and Clarke (1987) suggest that fluid pressures at the base of the snow pack may increase due to compression resulting in an equal decrease in effective stress.  Water  pressure may be expressed as the fraction (k) of the applied normal stress following K a m b (1970). The condition for the onset of separation becomes:  _/2tanj3_^ T:  (1-k)  1  _  [ 4  8 ]  r  This shows that when fluctuations of total normal stress become comparable to the overburden pressure, separation of the snow pack from the ground becomes possible ( M c C l u n g , 1981).  F o r example, water pressures i n excess of two times the applied normal  stress would have to be present on a 31° slope having a smooth roughness (r=A/X) of 0.1 i n order for this condition to be satisfied: less water pressure.  Steeper slopes of the same roughness would require  Snow on steep slopes with a rougher surface (r=0.5) separates with  pore pressures only 10% of the normal stress. Wankiewicz (1978) measured diurnal pressure changes during fine-weather snow melt of approximately 300 N m ' i n amplitude i n a 2  ripe Coastal British Columbian snow pack. A 1.0 m snow pack at the study site with an average density of 400 k g m would have a normal pressure of approximately 2019 N m " on a 3  31° slope.  2  Diurnal pressure changes on the order of 300 N m " would only offset 15% of the  total normal pressure.  2  Diurnal meltwater pressure changes would not be significant enough  to decrease the resistance to glide.  The effect of pressure change due to an influx of rain or  60 meltwater would be larger but the nature of this pressure change is not known.  Using the kinematic-wave equation to examine pore pressure development within a snow pack, it can be concluded that as long as flow speeds are comparable to film-flow, high input rates may generate pore pressures at the base of a snow pack. In terms of failure prediction, the conditions are highly sensitive to the internal angle of friction, which may vary with time and space within a snow pack. Until the internal angle of friction is known, one would not be able to accurately predict times of snow slab failure. When parameters are better understood, the kinematic-wave model may be useful. Based on the nature of the snow pack, water pressures might be significant enough to promote gliding by themselves.  Critical pore  pressures may be attainable in the sloping snow pack. Based on the effect of a thin water layer on the roughness of the ground, this depth of water may be more influential to glide than the pressures generated by the water layer.  4.6 Effect of Rain-on-Snow Events on Glide and Full-Depth Avalanche Occurrence  Rainfall events, which are very common in the study area during the winter, have a significant effect on glide rates and full-depth avalanche release. in increased glide rates and failures at the ground surface.  Inputs of free-water result  Rain-on-snow increases the  normal load of the snow pack, but the primary mechanism for failure at the ground surface is the loss of friction between the base of the snow pack and the ground surface.  A summary of glide avalanches and rain-on-snow events is shown for the two study seasons in Figure 4.9 and Figure 4.10. F r o m these series it is apparent that a significant proportion  61 Figure 4.9: Full-Depth Avalanche Occurrences and Rain-on-Snow Events 1992-93 Season  9  15 21 27  November  3  111 •• 1111I 111 I1111 II i 9  15 21 27  January  December .  Avalanche  February Shoe #2  9  15  March  21  llllllllllllllllllllllllllllllllllllll 0 27 2 8 14 20 26 1  April  Shoe #4  2a E  ^  15-  10-  0 iflfiiillli:'JH " 'J^.'. . .''. . . . . . . . .''•''•'''''»'*'IIIIII11II11 111 lilt II11lll .,,,,,,,,,,,.„ ^^ 9 15 21 27 1,1  11J11  1  November  December  iJanuary  - • February  LA March  14 20 26 April  1  Figure 4.10: Full-Depth Avalanche Occurrences and Rain-on-Snow Events 1993-94 Season  63 of all full-depth avalanches release in association with the rain events, and that the response to rain-on-snow events is relatively quick. During the 1992-93 season, out of 26 full-depth avalanche occurrences, 8 are thought to have been triggered by rainfall.  During the 1993-94  season, out of 78 full-depth avalanche occurrences, 54 are thought to have been triggered by rainfall.  A small number of releases are attributed to snow loading, and the remaining full-  depth avalanches are thought to have been triggered by snow melt processes (Chapter 5).  In  order to determine the full effect of rain on glide rates and avalanche release, several cases are identified for more detailed analysis.  Figure 4.11 illustrates the climate, snow glide, and avalanche conditions for the period February 25 - M a r c h 15, 1993. Glide rates responded to increasing air temperature from M a r c h 1, with a slight but steady increase. Increases in glide rates during this period correspond to snow loading on M a r c h 2 and M a r c h 4. O n M a r c h 5 and 6, 15 m m of rain fell; a relatively small amount of rainfall.  Approximately 24 hours after the last rainfall, on  M a r c h 7, 1 full-depth avalanche was recorded.  Glide rates continued to increase past this  time and peaked on M a r c h 8. This timeperiod shows that avalanches were triggered by snowfall and a combination of snowfall and rain.  A n increase i n glide rates corresponded  with the avalanche activity but did not coincide with peak glide rates.  Conditions for M a r c h 14 to 27, 1993 (Figure 4.12) show that glide rates at Shoe #4 were responsive to a small rain event and warm temperatures on M a r c h 17 and to larger inputs (24 mm) on M a r c h 21-22. Avalanches on M a r c h 23 and 25 are thought to have been triggered by this latter rainfall.  A t this time, Shoe #1 was not displaced any significant distance and  Shoes #3 and #5 had run out so these measurements were excluded from the analysis.  64 Figure 4.11: Snow Glide, Avalanches, and Climate Conditions for February 25 - March 15, 1993 20-  Shoe #1 Shoe #5  Shoe #2 |'///A Avalanches  Shoe #4  20-  -15  -10 15  HI O  E E,  J  1 0  o  a a. o o a.  a.  E a  I—i—i—i—i—i—i—i—i25 26 27 28 1  I  I Snow  -i—I—r  3  4  —r  6 7 Date g | g Rain  8  "i—I—I—I—I—i—I—I—I—I—I—I—r -10 10  11 12 13 14  Max. Air Temp.  15  Figure 4.12: Snow Glide, Avalanches, and Climate Conditions for March 14 - 27, 1993 20-  15-  A  10-  <  e  > o TJ  O  / \  14  ~i  1—  15  T—i—r —r—r 17 16  i—r I r -19  18  Shoo #2  —I—T 20  1  21 March Shoe #4  22  23  /  1  24  U o 6 z  XA  26  25  ~~\ 1  T  27  Avalanche  30  25  — 20 m E E •A  ; 1-  10-  14  15  i—i—i—i—i—i—i—i—i—i—i—i—r—r 16 17 18 19 20 21 22 23 March I  I Snow  K B i Rain  ~l 24  1 1 1 1 1—H"-6 25 26 27  Max. Air Temp.  66 Figure 4.13, for the period of December 8 to 18, 1993, shows a dramatic number of avalanche (12 i n total) released on December 10 and 11 following 17 m m of rain on December 9. In this case, avalanche response to rainfall was between 12 and 24 hours. Glide rates during this time were consistently increasing for all glide shoes.  Avalanches on  December 13 and 15 are identified as snow melt triggered events and w i l l be discussed i n Chapter 5.  Conditions for December 30 to January 15, 1993-94 are illustrated i n Figure 4.14. During this time period, it is a combination of factors that contribute to avalanche release.  R a i n on  January 2, 3, and 4 (31 m m total) was responsible for one avalanche occurrence on the evening of January 2, and 6 occurrences on January 4. A n increase i n snow glide is apparent i n the afternoon of January 3. Glide rates peaked for Shoe #5 on January 5 while peaks at other glide shoes were less obvious.  Avalanche occurrences for the remainder of  the time period are thought to be related to snow loading and snow-melt.  Glide rates at the  end of this time period significantly increase in response to a dramatic increase in air temperatures but there are no more avalanches.  Figure 4.15 illustrates climate, snow glide and avalanche conditions for the period from February 18 to M a r c h 6, 1994. During this time 81 m m of rain fell i n 3 days from February 28 to M a r c h 2. A n immediate response in glide rates and avalanche occurrence to this significant input of water is indicated. A total of 14 full-depth avalanches were released i n the same three day period. Glide rates at Shoes #2 and #4 definitely increased i n response to the rainfall but, again, glide rates peaked after avalanche release.  One may pinpoint the  moment that glide rates accelerated to a time approximately 4 days prior to avalanche  67  Figure 4.13: Snow Glide, Avalanches, and Climate Conditions for December 8 - 18, 1993 30-  December  I  Shoo #1  Shoe #2  Shoo #4  Shoe #5  I Snow (mmWE) Egg Rain (mm)  Shoe #3 £  Avalanches  Max. Air Temp.  Figure 4.14: Snow Glide, Avalanches, and Climate Conditions for December 30 - January 15, 1993/94  1  |Snow  BH3 Rain  — Max Air Temp  Figure 4.15: Snow Glide, Avalanches, and Climate Conditions for February 18 - March 6, 1994  Date Shoe #1  18  19  20  21  |  22  | Snow  Shoe #2  23  24  25  Shoe #4  26 27 Date  1389 " ' a  n  28  Avalanches  1  2  3  Max. Air Temp  4  5  6  70 release.  This would indicate that, even though there were no avalanche releases, there was  some effect of a significant snow fall event (36 m m total) on February 23 on the glide rates.  Time-series analysis for periods during the two study seasons indicates that full-depth avalanches release i n response to rain-on-snow events.  There is a strong positive correlation  between avalanche occurrence and rainfall less than 12 hours and up to 48 hours after rain events (see Table 4.4).  Table 4.4: Correlation of Avalanche Occurrence with Rain Events and Snow Events for Selected Time Periods Time Period  Correlation (Aval. vs. Rain)  2 x S.E.  Lag Time  Correlation (Aval vs Snow)  2 x S.E.  Lag Time  Feb. 25 M a r . 15,1993  0.873  0.348  48 hours  0.854  0.354  60 hours  Dec. 8 -18, 1993  0.662  0.436  12 hours  0.573  0.425  < 12 hours  Dec. 30 Jan. 15, 1993  0.723  0.342  < 12 hours  -  -  Feb. 18 M a r . 6, 1994  0.963  0.342  < 12 hours  -  -  4.7 Discussion  Early season fluctuations and high glide rates may be due to a greater effect of free-water on the material properties of low density snow than on high density snow.  This is also  compounded by the effect o f a geothermal heat flux on a thin snow pack. In the spring, water has a less significant effect on the material properties of higher density snow.  A t this  71 time, free-water, generated by processes of melt, acts more to separate the snow pack from the ground and lubricate the interface causing fluctuations in rates of glide.  Experimental results on water flow speeds at the base of the snowpack indicate the presence of two types of flow: f i l m flow and channel flow.  Channel flow speeds were measured i n  the field as 6.6 cms" but varied significantly with snow pack conditions. Channelization of 1  water beneath the snow pack results in rapid drainage and little lateral divergence of the flow.  It is also more difficult for faster flow to generate pore pressures. It is suspected that  film flow speeds on the order of 0.2 c m s affect a larger area beneath the snow pack. 1  Due  to the decrease i n surface roughness by drowning, there is a decrease in shear resistance causing separation of the snow pack from the ground and increased rates of glide velocity.  F r o m previous studies, it is known that free-water affects rates of snow glide ( M c C l u n g and Clarke, 1987). A n examination of possible trigger mechanisms for full-depth avalanches i n the study area reveals that water inputs from rainfall and from snow melt account for all but a few of all full-depth avalanche occurrences. In this chapter, it is shown that contributions of free-water by snow melt and by rain-on-snow events are significant and have a profound influence on snow glide and the release of full-depth avalanches. Avalanche response to rainfall events ranges from an almost immediate response to a 48 hour lag i n response. Glide rates correspond accordingly by increasing significantly after rain-on-snow events.  Data shows that there is also a response to changes in air temperatures.  The time series  discussed i n this chapter also show a relation between high air temperatures and higher rates of snow glide.  In consideration of the effects of free-water contributions from rain events,  72 the effect of above freezing temperatures was not considered i n detail. The relationship between air temperatures and snow glide is further examined i n the discussion of snow melt in Chapter 5.  73 C h a p t e r 5: E N E R G Y A N D S N O W M E L T  5.1 I n t r o d u c t i o n  The production of meltwater is relevant to this study since the presence of free-water at the snow/ground interface affects rates of snow glide (Chapter 4).  This chapter describes how  energy sources, both radiative and thermal, contribute to snow melt. Equations used to calculate the contribution of energy to snow melt are outlined. Southerly slope aspects i n the study area are conducive to significant absorption of solar radiation. In this chapter, characteristic measurements of radiation and equivalent melt, taken over two seasons, are presented.  The overall objective of this chapter is to determine the snow glide response to  snow melt by radiative and thermal conditions.  Factors governing the rate of meltwater production include the availability of energy and its exchange at the snow surface.  Short-term prediction of snow melt using an energy balance  approach has proven successful (Price and Dunne, 1976; Ishikawa, et a l . , 1985; M a l e and Granger, 1981).  Obled and Harder (1978) present an overview of radiation conditions and  snow melt i n a mountainous environment. Marks et al.(1992) determined the energy balance and climate o f an alpine region of the Sierra Nevada, but very few comprehensive energy balance measurements have been conducted in remote alpine environments.  In terms of energy and its effect on snow, radiation measurements have been included, with other meteorological elements, for research on snow conditions and avalanche forecasting. Radiation characteristics have been used i n a multi-variate and discriminant analysis approach  74 to avalanche forecasting (Bois et a l . , 1975; Obled and Good, 1980). Others have used radiation measurements to model evolution i n snow pack stratigraphy (Gubler, 1980) (Brun et a l . , 1989).  The purpose of measuring radiation for this study is to characterize the radiation inputs to the south-facing slopes of Zopkios Ridge, an area prone to glide avalanches. The investigation of radiation conditions is of interest i n this study because of the occurrence of full-depth avalanches during periods of clear, cold weather.  During these periods, when avalanches are  somewhat unexpected, their trigger is unknown. A n investigation of the radiation balance on the slope w i l l help to determine i f rates of snow melt are great enough to supply the snow/ground interface with significant amounts of water.  Radiation conditions w i l l be used  to determine periods of intense radiation exposure causing rapid melt conditions. Radiation data w i l l be used to correlate sunny time periods to rapid gliding and avalanche release.  5.2 Energy Conditions at the Study Site  5.2.1 Instrumentation and Data Collection  In remote alpine environments, measurement of radiation during the winter months is generally rare, the limitations being climate conditions, power supply, accessibility, and snow accumulation on sensors. Radiation measurements have not been conducted previously at this site.  During the first season of study (1992-93), radiation measurements were conducted at the  75 Summit weather station located in the valley at 1300 m (see Figure 2.2).  A t this location the  instruments were maintained on a daily basis by M o T H staff and a power supply was readily available.  The sky-view factor, calculated using fisheye-lens photographs, was calculated as  per Steyn (1980) to be 0.801. This is comparable with the sky-view factor measured at the glide site (0.77). This site, although logistically attractive, is affected by horizon shading. Needle Mountain, and its ridges across the valley from the study area, block sunlight for at least half of the day during the winter months at the Summit Weather Station.  F o r this reason, radiation instruments were relocated to a site halfway between N a k peak and the valley bottom at an elevation of 1600 m . (see Location M a p - Figure 2.2).  The sky-view  factor, although not calculated, would be similar to that at the glide site. Instruments were mounted on a 15 foot tower i n order to overcome snow accumulations of approximately 3 m. The new site was remote so access to the site was limited to times of low avalanche hazard. During periods between maintenance visits there was the possibility of lost data due to the accumulation of hoar frost, freezing rain, or snow on the domes of the instruments.  Without  access to an adequate power source no method for keeping the domes free of accumulations was achieved.  The tower was situated in a clearing on a ridge so that stronger winds would  lessen snow accumulations on the sensors.  Despite the location, there are periods within  otherwise continuous data runs that were lost due to snow accumulation on the instruments. F r o m the two different locations, it is determined that measurements at the Summit Weather station, although shaded at times, provide the highest quality of data. Too many days were obscured from the record from snow accumulations at the remote site.  Instrumentation used i n the study is outlined in Table 5.1. Readings were taken at 10 minute  76 intervals and then integrated into hourly averages.  This level of resolution was considered  sufficient to characterize radiation conditions at the site.  Radiation conditions at the site were measured on the horizontal. T o convert these measurements (S) to that of the direct beam radiation received on a 31° slope, the cosine law of illumination is used (Oke, 1987):  Radiation Rec'd on Slope =  ( S )  c  o  s  9  cosZ  [ 5 1 ]  =(S)(1.40) whenZ=45° =(S)(2.40) whenZ=72° where 6 is the angle of incidence between the normal to the slope and the solar beam. The solar zenith angle, which varies from 45° on A p r i l 3 and September 10 to 72° on December 22 (winter solstice), is the reciprocal of the solar altitude. The equation shows that radiation received on the slope is 1.4 to 2.4 times that measured on a horizontal surface.  Table 5.1: Radiation Instrumentation at the Study Site Precision  Instrument  Effective Range  Net Radiation  Micromet pyrradiometer  0.3 - 50.0 urn  ± 20 W i n "  Short-Wave Radiation  K i p p & Zonen pyranometer  0.3 - 3.0  + 10 W m  Long-Wave Radiation  Eppley (PIR) pyrgeometer  4.0 - 50.0 jian  + 10 W i n  Air Temperature  Campbell Scientific 107 Temperature Probe  -35° - 50° C  + 0.1° C  iim  5.2.2 Topographic and Site Influences  Radiation conditions vary with latitude, elevation, slope, and aspect, but the sensible and  2  2  2  77 latent heat contributions are not modified by topography except where wind speed is affected (Dunne and Leopold, 1978). By contrast, the effects of differential radiation loading on north versus south aspects is very pronounced in terms of snow conditions. Northerly aspects are shielded from incoming direct radiation and are therefore cooler.  Monteith  (1973), illustrates the striking contrast in daily totals of direct solar radiation for different aspects (Figure 5.1).  Slope angle influences the incoming energy flux density, as it affects  interception with the angle of inclination of the sun. South facing 30° - 40° slopes of Zopkios Ridge are subjected to the highest energy flux densities.  25 , - r  0  30 SLOPE  60  90°  ANGLE  Figure 5.1: Influence of Slope Angle and Aspect on Radiation Exposure From: Monteith (1973)  In all mountainous environments, energy balances are largely influenced by horizon shielding.  U s i n g a sun-path diagram selected for the appropriate latitude and the solar  declination appropriate to the winter season, the solar altitude and azimuth may be  78 determined.  Using these data, the variation in radiative loading throughout the winter season  is determined.  A t the glide site, the horizon shields elevations below the glide site.  The  valley in which the Summit weather station is situated is shaded during morning hours of the winter season.  Figure 5.2, a valley profile at an azimuth of 196°, illustrates how lower  portions of the valley are shaded at noon between November 9 through to February 9.  This  affects meterological and radiation measurements taken at this site during the first season of data collection and is the reason for instrument relocation i n the following season.  Ftofile Azimulh = 196'  Solar /Altitude (at Noon) = 18* (Pec. 22) VUnter Solstice = 45 CAx3/Sept.lO) -  Based on Sun-Ftrth Diagram for 50' N  Figure 5.2: Valley Geometry - Incidence of Radiation  O f interest for this study is exposure of the steep bedrock slabs on the upper slopes to radiation. The rock faces oriented towards the sun are subjected to intense radiative loading  79  80 on clear-sky conditions. Bare rock slabs, which have a much lower surface albedo than snow, warm quickly from absorbed radiation causing snow melt. Figure 5.3 illustrates the varying amounts of snow cover on the steep bedrock slabs. E v e n during mid-winter, snow on the steep slopes may melt or slough off, exposing large amounts of bare rock to atmospheric warming. Considering that the albedo of a clean snow surface is 0.98, and the albedo for a dark rock surface is approximately 0.30, this difference would significantly influence the radiation balance. This, i n turn, would affect snow melt.  Seasonal variation of radiative loading affects rates and timing of snow melt generation. During the winter months, lower temperatures, cloudy skies, and precipitation factors w i l l outweigh the influence of sunshine but i n spring, radiation becomes a greater contributor to snow melt.  5.3 Contribution of Energy to Snow Melt  Snow melt, expressed in terms of depth of water released from the snow pack, takes place only when the snow pack is isothermal at 0° C . To melt 1 gram of ice at 0° C , 80 calories of heat are required. This latent heat of fusion may be supplied by various meteorological sources.  Often, the amount of snow melt can be estimated by determining the relative  magnitude of these energy sources.  5.3.1 Energy-Balance Approach  The energy balance approach is an alternative method to snow melt measurement.  This  81 method is commonly used in watershed run-off models that base snow melt estimates on climate parameters.  Energy sources required for snow melt include radiative sources and  convective sources.  The amount of energy available for snow melt (Q ) is calculated as the sum of atmospheric M  energy sources:  Q  m  = Q*  + Q  H  + Q  + Q  E  G  [5.2]  + Q  p  These sources of energy include radiative sources of net radiation (Q*), and thermal sources of sensible (Q ) and latent heat (Q ), ground heat (Q ), and precipitation (Q ). H  E  G  P  * Short-wave Radiation  The incoming flux of shortwave radiation (K ), present only during daylight hours, is strongest at the snow surface but penetrates down into the snow pack. The amount of radiation penetrating the snow pack decays exponentially at a rate determined by the extinction coefficient.  A t a depth of 1 c m , 4 0 % of the incident solar radiation  remains. A t 10 c m , only 3-4% of the incident radiation remains ( O ' N e i l l and Gray, 1972). During clear, cold days incoming solar radiation penetrates the snow pack causing internal melt just below the surface which remains dry and cold by loss of long-wave radiation. This condition can cause large temperature gradients resulting i n some radiation recrystallization at the snow surface.  Radiation recrystallization, more  commonly experienced i n continental climates and higher altitudes, was witnessed at  82 the study site on M a r c h 6, 1993. Figure 5.4 illustrates the conditions on that day, where 0.5 c m of dry snow overlies a wet 2.5 cm layer.  In this case, radiation has  penetrated and affected snow to a depth of 3.0 cm. It should be noted that if this process continues and the dry layer is buried by subsequent snowfall, it w i l l remain as a zone of potential instability within the snow pack. Radiation penetration affects the rate and quantity of melt water released during a period of melt, but the overall significance of internal melting is probably small. Time for percolation through an isothermal snow pack must be considered.  The albedo, or reflectivity, of a snow-covered surface to short-wave radiation is characteristically high under clear skies. Albedo regulates surface short-wave radiation absorption and dominates the net radiation budget. thermal climate of the surface and adjacent air layers.  It therefore controls the  F r o m darker surfaces,  the  majority of incoming short-wave radiation is not reflected causing surface wanning and an intensification of snow melt.  The exposure of bare rock on the ridge causes  warming and snow melt, potentially resulting in a significant supply of free-water available to affect snow glide.  Long-wave Radiation  Long-wave radiation is emitted from the atmosphere (water vapour, carbon dioxide, ozone) and the surrounding terrain and is dependant upon air temperature and atmospheric vapour pressure.  Outgoing long-wave radiation, a function of the snow  surface temperature and its emissivity, is generally constant at the surface of a melting  83 snow pack (316 Wm~ ). Snow is almost a black body with respect to long-wave 2  radiation.  F i g u r e 5.4: R a d i a t i o n Recrystallization M a r c h 6, 1993 H S = 146.0 cm A i r Temp = -1.0° C Snow Pack Isothermal @ 0° C Sky = Clear  Net long-wave radiation (L*), active only at the snow surface, indicates the direction of energy flow.  When L* is positive, cloud base temperatures are warmer than the  snow surface and radiation flows toward the snow pack. Net long-wave radiation is negative during clear-sky conditions, indicating a loss from the snow pack. This can create energy deficit conditions (freezing of surface) which must be overcome before snow melt can occur.  84 Net Radiation  A s a contributor to snow melt, net radiation (Q*), the sum of net long-wave and net short-wave components, is the most important of all energy terms.  The high albedo  and high emissivity of the snow surface means that a large proportion of the energy incident to the snow surface is reflected back again resulting i n the low overall energy status of the snow pack.  Sensible and Latent Heat Fluxes  The sensible (Q ) and latent (Q ) turbulent heat fluxes result from turbulent eddies i n H  E  the wind stream that carry parcels of warm, moist air down to the snow surface. Controlling factors include wind speed, temperature, and water vapour density. Sensible and latent heat sources are not subject to the same topographic and vegetation modification as radiant energy sources unless wind speed is modified.  This results i n  a fairly uniform meltwater flux from the surrounding landscape and it also means that melt from these energy sources continues throughout the night. Periods of high sensible and latent heat fluxes are usually associated with convective, cloudy and rainy conditions. Both Q  H  and Q  E  should be corrected for atmospheric instability. If the surface of the  snow pack is cooler than the atmosphere, cooler air near the snow pack surface resists circulation and the turbulent exchange is reduced Therefore, when T > T a  is around 0°C the turbulent fluxes may become negligible.  s  or when T  a  85 * Ground Heat F l u x  The ground heat flux (Q ) is small and of minor overall significance to the snow pack G  energy balance. In most cases it is not included in the snow melt calculations. In terms of snow glide though, i n the early season, when the snow pack is thin, the ground heat flux may still be important. The ground heat flux, in combination with, m i l d air temperatures and the penetration of shortwave radiation into the thin snow pack may help explain seasonal variation i n glide rates (Chapter 3).  * Precipitation Heat F l u x  The amount of energy, released from rainfall is given as:  •  Q  P  = PT c a  [5.3]  wPw  where P is the depth of rainfall (cm), T is the air temperature (°C), c a  heat of water (calg^d ), and p 1  w  w  is the specific  is the density of water (gem ). Energy from rainfall 3  (Q ) is small and is generally neglected as an energy source. The direct contribution P  of free water to the snow pack from rainfall far outweighs the melt generated from rainfall heat energy. E v e n low-intensity rainfall events contribute much more water than extreme rates of snow melt (Dunne and Leopold, 1978).  The rate of release of melt water (cmd ), is calculated from: 1  86  M=  [5.4]  where the energy available for snow melt is divided by parameters describing the energy requirements for snow melt. Energy requirements (the latent heat of fusion) are adjusted by the density of water at 0° C , and the thermal quality of the snow pack (1-0), where 6 is the irreducible water content. The irreducible water content is the amount of water that a melting snow pack w i l l retain against drainage and is generally 3-5 % (by weight) (Male and Gray, 1981). M e l t rates can be negative at night during periods of energy loss by evaporation.  This negative balance is a heat deficit which must be overcome before any  further melt can take place. Using the energy balance approach, one can achieve a better than 10% accuracy i n snow melt prediction (Harding, 1986).  5.3.2 A p p r o a c h Used i n this Study  A modified version of the energy balance approach was used in this study to determine the magnitude of energy inputs contributing to snow melt. Turbulent fluxes require a detailed instrumentation program which becomes more difficult to implement in a remote alpine environment.  Due to the complicated nature of turbulent flux measurements, measurements  of Q  were unobtainable. This modified approach to snow melt estimation assumes  H  and Q  E  that net radiation is the dominant control of snow melt and turbulent fluxes are minor i n comparison.  Time periods of interest for this study are clear sky conditions when this  condition would be the case (Dunne and Leopold, 1978). During warm, convective periods, which are common during the winter season, the turbulent transfer of sensible and latent heat is high. The impact of these convective periods is more obvious and was addressed i n  87 Chapter 4.  In this study, net radiation is considered to be the most important radiative energy source. Jordan (1978), conducting research i n an alpine environment (1900 m), determined that 9 0 % of the total melt energy is derived from the net radiative flux. This was determined to be due to low air temperatures and little wind. Hendrie and Price (1978) conducted studies i n a leafless deciduous forest and determined that the turbulent energy exchanges could be ignored in the estimation of total daily snow melt. They also determined that on days free from rainon-snow, Q* dominates the energy balance. Wankiewicz (1976) assumed all energy from net radiation went into snow melt and found these values compared favourably with snow melt measurements from lysimeter data.  Other studies determined that the contribution of net radiation to snow melt was not as significant.  Braun (1980) determined that this quantity is 5 9 % +8% based on daily totals of  net radiation and lysimeter data at a sub-alpine (1300 m) site. A study by Martinec (1989) over a complete melt season (May 9 - July 15, 1985) in Weissflujoch, Switzerland (2540 m) determined that Q* accounted for 6 0 % of snow melt.  Seasonal summaries of radiation conditions, Figure 5.5 a) and b), show that the magnitude of radiation increases with time through the season. During the 1992/93 season, measurements were not started until January 23. M o r n i n g hours during this season i n mid-winter are obscured from the measurements due to horizon shielding, therefore measurements may underestimate radiation receipts.  During the 1993/94 season, snow accumulation on the  sensors, located at a remote site, resulted i n several periods of lost data which, unfortunately,  88 seemed to correspond with many clear-sky days. Despite this, seasonal summaries indicate that melt rates are quite significant in the spring.  Times of interest for this study are clear, cold weather conditions, since these are times when avalanche release would not normally be expected.  The hypothesis is that during cold times,  especially i n mid-winter, incoming solar radiation warms exposed rock on the upper slopes, generating significant quantities of snow melt. In the spring, incoming radiation contributes a significant amount of energy towards snow melt. During time periods of interest, the sensible and latent heat fluxes are small. O n clear, cool days, the net radiation equivalent melt was solely used to determine melt rates i n the study area.  Net radiation equivalent melt (cmd ) does not necessarily correspond with rates of meltwater 1  generation and run-off.  Several conditions are required for melt to occur. The snow pack  w i l l release water to snow melt only when it is isothermal at 0° C . It is, therefore, important to determine when the snow pack is isothermal. It is also necessary for the heat deficit generated by negative melt rates in the evening to be overcome. The situation at the study site somewhat modifies the requirements, since the exposure of bare rock can result in warming of the adjacent snow pack even when the surrounding air and snow temperatures are cold. F i e l d measurements indicate that the interface temperatures remain at 0° C , so that water may percolate down slope from the rock slabs to affect gliding and avalanches further down slope.  5.3.3 A i r Temperature as a Surrogate V a r i a b l e  Figure 5.5 Seasonal Summaries of Net Radiation Equivalent Melt Rates a) 1992-93 Season b) 1993-94 Season 60-,  iniimnrwiiwniirrivu111111luWrTTO1111T1 nTifWiuiiTfyi'nmmrrTfiTuins*ri*iIWIffi11'  -z° ^iiiiiiuiiimriiiiffiwwiuiffiiiim 9  15  21  9  15  1  21  27  2  ,  27  3  8  14  20  26  1  7  13  19  TPIIUIHIIIU'WI  11IWI1111ni?n11IWITIffiu11111111111111mnffiin11fD'PViwnifiin11111111111n1111mffft1111n11111uiTfWrfWW  9  27  3  November December January February Radiation Measurements Converted to that Received on 31" Slope indicates missing data Q* (Equ.M) E l Clear Sky  25  3  9  15  21  March  27  :  -  2 0  15  21  9  15  21  27  2  8  14  20  26  1  7  13  19  November December January February Radiation Measurements Converted to that Received on 31'Slope indicates missing data Q* (Equ.M) ta Clear Sky  25  3  9  15  21  March  27  90 A i r temperature is the most obvious and most available index of the amount of energy available for snow melt. A i r temperature is a key component of the sensible heat flux and incoming long-wave radiation and is a valuable indicator of the complexity of the local climate. It is often used for regional basin snow melt predictions but it may not be accurate enough for site specific melt predictions.  Above freezing, air temperatures are associated with snow melt. During the spring, avalanche forecasters recognize the relationship between higher air temperatures and spring avalanches. The relation between avalanches and radiative conditions is not as well defined. The release of glide avalanches during periods of cold, sunny conditions is rare but it does happen. It is this phenomenon which puzzles forecasters and is of interest i n this study. Therefore, air temperatures alone w i l l not provide the data necessary to determine conditions required for this type of avalanche release.  The degree-day method of calculating rates of  snow melt uses daily average air temperature, a reference temperature, and a temperature index, the latter two parameters being empirical factors that vary from site to site (Harding, 1986; M a l e and Gray, 1981; Dunne and Leopold, 1978).  Harding (1986) found that the equation accounted for only 2 1 % of the total measured melt. Reasons for the model's failure included the fact that a large proportion of the energy used to melt the snow pack came from net radiation, and air temperature may not necessarily be directly related to radiation. Temperature-index relations must also be calibrated for each region and vegetation type.  In this study, the Pearson correlation between air temperatures  and net radiation is significantly positive (r=0.450) but the relationship is not a particularly strong one.  91 A s noted i n Figure 5.6 a) and b) glide velocities somewhat follow the general trend of air temperatures.  Higher air temperatures generally relate to higher glide rates but this is not  always the case, since glide velocity changes usually lag behind fluctuations i n air temperature.  Larger fluxes of melt water move faster than smaller ones.  When larger  fluxes, catch up to smaller ones, the confluence may cause a sudden increase i n melt water discharge.  This means that extended periods of warm temperatures may affect glide rates  more than a single rain event, and can help explain why there is not an immediate response of glide rates to temperature fluctuations.  Cross correlation between glide velocity and air temperature is possible since both series are continuous.  F o r selected time periods during both seasons, the relation between the two is a  statistically significant, positive relation with lag times ranging between 12 to 24 hours (Table 5.2).  This means that changes i n snow glide rates lag 12 to 24 hours behind changes  in air temperature.  Results show that correlation between snow glide and air temperature is  slightly more significant during time-periods used as radiation examples i n the text.  Time  periods used i n Chapter 4, to illustrate the response to rain-on-snow events, indicate positive correlations, yet not quite as statistically significant.  5.4 E n e r g y . Snow M e l t , a n d the Impact on F u l l - D e p t h A v a l a n c h e O c c u r r e n c e  Response to snow melt on the upper rock slabs in the study area is of great interest to this study. The rock slabs are probably the most responsive to radiative energy sources since there is little to no surrounding vegetation and their orientation and slope amplify radiation fluxes.  92 Figure 5.6: Snow Glide and Air Temperatures at the Glide Site a) 1992-93 Season b) 1993-94 Season  November  December Shoe #2  January  - — Shoe #4  February Max. Air Temp  March Min. Air Temp  April  93  Table 5.2: Correlation of Glide Velocity with Maximum Air Temperatures for Selected Time Periods Example  Time Period  Correlation  2 x Standard Error  Time Lag  Radiation  Dec. 10 - 24, 1992  0.450  0.378  24 hrs  Radiation  Jan. 23 - Feb 6, 1993  0.620  0.372  12 hrs  Rain-on-Snow  Feb. 25 - M a r . 15, 1993  0.380  0.334  12 hrs  Rain-on-Snow  Dec. 30 - Jan 15, 1994  0.450  0.348  12 hrs  Rain-on-Snow  Feb. 18 - M a r . 6, 1994  0.460  0.360  12 hrs  Radiation  M a r . 1 4 - 3 1 , 1994  0.600  0.266  12 hrs  A l l but a few avalanche occurrences not triggered by rainfall, were triggered by the input of free-water from snowmelt.  A large proportion of glide avalanches triggered by snow melt,  released during warm convection periods characterized by higher air temperatures and high rates of snow melt. A much smaller sample of glide avalanches is unexplained by conventional avalanche forecasting knowledge.  These are full-depth avalanches that release  during sunny and cold conditions. These occurrences are somewhat infrequent but avalanche forecasters can often be caught by surprise.  During the two seasons of study, periods with characteristically sunny and cold conditions produced 7 glide avalanches. These time periods and associated climate conditions are listed in Table 5.3. It is this type of glide avalanche that is the most difficult to predict.  94 T a b l e 5.3: F u l l - D e p t h Avalanches Triggered by R a d i a t i o n E x p o s u r e  Date Dec. 12, 1992  Feb. 4, 1993 Mar. 25, 1993  Sky clear  clear clear to  Average Air Temp (C)  Date  -10.0  9  overcast  -4.5  22 mm snow  10  part. cl.  -6.5  4 mm snow  11  part, cl  -8.0  2  clear  -3.0  -  3  clear  -6.0  -  23  overcast  0.5  24  clear  13  overcast  14  -2.0 -2.0  partly cl. Dec. 15, 1993  overcast  -3.0  to clear Jan. 20, 1994  scatt.  -2.0  cloud  Previous Two Days Sky Temp. Precip.  Aval. Occurrence Path Size* 46.5  1.5  47.9  2.0  29 mm snow  46.5  2.0  -3.0  8 mm snow  46.8  1.5  -2.5  -  47.7  2.0  overcast  -2.0  trace  47.35  2.0  18  overcast  -1.0  -  47.35  2.5  19  overcast  0.0  -  -  •  * Using Canadian Snow Avalanche Size Classification System (McClung and Schaerer, 1993)  It is thought that incoming solar radiation on clear-sky days generates sufficient snow melt to affect glide rates and trigger full-depth avalanches. Inputs of radiation are now examined over selected time periods. Radiation measurements of interest for this study are net radiation, converted to the equivalent melt rate, and 12 hour sums o f incoming short-wave radiation. The latter highlights the daytime quantities of incoming solar radiation that are important contributors to snow melt.  Conditions for December 10 - 24, 1992 are illustrated i n Figure 5.7. Before the avalanche on December 12, glide rates were increasing despite below-freezing air temperatures. The starting location was at the top of the path where the snow pack might be thin. Warming of adjacent rock slabs may have had a significant effect on a thin snow pack. Unfortunately, instrumentation was not installed at the time of the December 12, 1992 avalanche. A n avalanche released o n December 15 due to loading (24 m m W E snowfall) and another released o n December 23 due to 46 m m W E snowfall over the previous 36 hours.  95 The period from January 23 to February 6, 1993 was characterized by a number of full-depth avalanches, probably triggered by different mechanisms (Figure 5.8).  R a i n on January 25-27  caused avalanches on January 25 and January 28. This rainfall, along with air temperatures that reached a high of 16° C on January 31, resulted in unstable snow conditions. A rapid increase i n snow glide indicated by 3 out of 4 shoes signals these conditions. A total of 11 full-depth avalanches were triggered on January 30-31 i n response to the rapid snow melt.  A  single avalanche occurrence on February 4 may have been triggered by snow melt from a combination of high air temperature and radiation. Records indicate that skies were clear from February 2-4. Radiation conditions for this time period (Figure 5.9) indicate that net radiation equivalent melt peaked on January 30, when most of the avalanches were released. Net radiation equivalent melt peaked again on February 3, 24 hours before the February 4 occurrence.  The 12 hour sum of incoming short-wave radiation which peaked on February 2  does not provide the same information on snow melt generation as net radiation. Incoming short-wave radiation provides us with a general indication of melt conditions only, whereas net radiation allows us to better estimate actual snow melt conditions.  Figure 5.10 illustrates conditions for January 16 to 31, 1994. O n January 20 maximum air temperatures rose from 1° C i n the morning (0600) to almost 14° C i n the evening (1800). This temperature change, although significant, was not reflected i n the glide rates.  Instead,  glide rates increased over several days after this time. This lack of response to rapid temperature fluctuations is evident throughout the two seasons of study. Skies on January 20 had scattered cloud cover. A peak in the daytime sum of incoming solar radiation reflects these conditions (Figure 5.11). Measurements of net radiation equivalent melt are not available due to snow accumulations on the sensors. It is the energy contributions on  96  Figure 5.8: Snow Glide, Avalanches, and Climate Conditions for January 23 - February 6, 1993  I  | Snow  Shoo #1  -  Shoo #2  Shoo #5  g  Avalanche  gm  Rain  — - Shoe #4  Max. Air Temp.  98  Figure 5.9: Radiation Conditions for January 23 - February 6, 1993  CM I  E c o .2 "•5 DC CO  a c  e  o o c  o E to CM  AA 23 24 25 H B  A 26 27 Q  * (  e c  28 29 30  i u  m e l t  A  )  +  D a  31  1  y  S  t i m e  u  2 m  - Avalanche Occurrence  o f  3 K  *  4  5  6  99 January 20 which trigger one avalanche in the evening of January 20. Energy inputs increase significantly at the end of January and air temperatures are also quite m i l d .  It is  puzzling that there are no corresponding increases in glide velocity.  Figure 5.12 illustrates the meteorological and glide conditions for the period between M a r c h 14 to 31, 1994. Avalanche releases on M a r c h 14-15 were triggered by rainfall and above freezing temperatures.  There was no response, in glide rates or avalanche occurrences, to  the snow storm around M a r c h 18. A dramatic increase in air temperature at the end of the month is reflected in the glide rates. During the previous 5 days maximum air temperature remained above 0° C . There was no precipitation at this time and skies remained clear.  The  daytime sum of incoming solar radiation (Figure 5.13) is consistently high during this time. M e l t rates are positive, yet fluctuating, which is probably due to instrumentation problems. F i e l d notes indicate some condensation problems on the net radiometer at this time.  Three  full-depth avalanche occurrences at the end of the month on M a r c h 28-30 were triggered by significant contributions of free-water from snow melt generated by both radiative and thermal sources of energy.  5.5 Discussion  The relationship of snow glide and full-depth avalanche release to radiation conditions was investigated at the study site. The slope and aspect of the study area is such that radiation conditions become a strong influence on snow melt. Instrumentation of radiation at two different sites each had associated problems.  During the first season, radiation instruments at  the Summit weather station were shaded for morning hours. During the second season, snow  Figure 5.10: Snow Glide, Avalanches, and Climate Conditions for January 16 - 31, 1994  Shoe #1  -—  Shoe #2  -• —• Shoe #3  Shoe #4  -—  Shoe #5  g 7Z\ Avalanche 7  -14  J-12  Mo  J  \\ I i  \  V \\  r-4  \l  V  /\  I /\ 1/ ^ 1 \  . 16  17  18  n , , , n , F ' l 'i , 'I . . . n • n  19  |  20  | Snow  21  22  23 24 January I Rain  25  26  27  >-2  n 28  Max Air Temp  29  30  31  Figure 5.11: Radiation Conditions for January 16 - 31, 1994  3000  0-^ i i i i i i i i i i i i i i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—r 16 17 18 1 9 2 0 21 22 23 24 25 26 27 28 29 30 31  January Daytime Sum of K*  A  Avalanche Occurrence  Figure 5.12: Glide Rates, Climate Conditions, and Avalanches for March 14 - 31, 1994  i r r  14  15  i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—r 16  17  18  19  20  Shoo #1  ~  UJ  21  22 23 March  Shoo #2  24  26  Shoe #4  27  i  28  29  I  I  30  I  i  r  31  Avalanches  60-  20  50-  15  40-  10 O  £ E, O  25  k  M i  ii  30  20  10  o  1  r i ii  14  15  '1  , 'i i, 'in i, 'ii i i i i—i i i i i i i i i i i i i -io 'i—i 'i i 'i—i 'i i 1 'i i  16  17  |  18  19  |Snow  20  21  1  22 23 24 March Rain  25  26  27  28  29  Max. Air Temp.  30  31  103 Figure 5.13: Radiation Conditions for March 14 - 31, 1994  7000  6000  -20  T—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i  14 15 16 17 18  i—i—r  19 20 21 22 23 24 25 26 27 28 March  fg| Q * (equ. melt) A  +  Daytime S u m of K+  - Avalanche O c c u r r e n c e  29 30  31  E  accumulations on domes resulted i n a loss of data. Access'difficulties to the remote site at times of high avalanche hazard made this a problem.  It is recommended that instrumentation  continue at the Summit weather station because of ease of maintenance. A correction factor for the morning hours may be determined.  F o r the purpose of estimating snow melt, net radiation (Q*) is the most important variable. Instrumentation for Q* is sensitive to snow accumulations and moisture conditions despite use of a tougher dome construction. W i t h limited resources, measurement of incoming solar radiation (K ) would be adequate to determine sky conditions and general estimation of the radiation balance.  F o r snow melt estimation, the ground heat flux and precipitation heat flux can be disregarded.  The turbulent heat fluxes, which may be significant contributors to snow melt,  were not measured in this study. A i r temperatures and precipitation are good indicators of convective climate conditions and are much easier to instrument. Use of Q* alone to calculate rates of snow melt gives a good indication of conditions on clear days only.  The  response of glide rates to air temperature is not always predictable but long term records provide at least an indication of response.  O n clear, cold days, incoming radiation warms exposed rock slabs causing snow melt at their periphery.  Percolation of the water down slope, at flow speeds determined in the previous  chapter, w i l l affect free-water conditions at the snow/ground interface further down slope. Except i n extreme circumstances, such as indicated i n Figure 5.11, glide rates are not as responsive to snow melt events as they are to rain-on-snow events.  The input volume of  105 rainfall is generally larger than that from snow melt and the areal distribution of rainfall is also larger, especially as snow melt may take place only adjacent to exposed rock slabs.  The  location of glide instrumentation in relation to Zopkios Ridge may explain why glide rates are not as responsive to snow melt as they are to rainfall, since the starting locations for avalanches triggered by melt are near the top of the path and glide rates are measured further down slope.  A visual survey of snow cover on Zopkios Ridge once a day would help to estimate the impact of rock warming on snow melt rates. A relative scale would provide a useful indication of changing radiative conditions. If a large area of rock is exposed, then there is a larger warming influence. A t the same time, if the slopes are bare it also indicates that there is little snow available for avalanche release at the upper starting zone locations.  Avalanche occurrences that are triggered by intense radiation inputs on clear cold days are generally uncommon. They are relatively rare and knowing radiation balances in great detail is not necessary. forecasters.  The few releases that do occur still remain a concern for avalanche  It is recommended that forecasters become aware of the impact of warming due  to radiation even when air temperatures are below freezing.  It sometimes occurs than an avalanche is triggered i n the evening after a sunny day. It is thought that surface cooling on a clear night caused the relatively dry snow at the surface to contract. This can result i n a rapid increase of tensile stress, causing the release of a slab already i n tension due to water lubrication of the previous days melt ( M c C l u n g and Schaerer, 1993).  106 C h a p t e r 6: D I S C U S S I O N A N D C O N C L U S I O N S  The smooth, steeply-dipping and south-facing bedrock exposures, combined with moist snow pack conditions, create favourable conditions for snow glide and full-depth avalanche release. A large proportion of all the avalanches occurring i n the study area are full-depth avalanches, creating a large data set with which to work with. The Coquihalla Avalanche Area afforded an accessible location, furnished with data collected by the Ministry of Transportation and Highways.  Results from this study extend previous knowledge on the process of snow glide. There are very few studies as comprehensive as this, which attempt to correlate high rates of snow glide with full-depth avalanche occurrence.  It was previously speculated that there would be a strong  correlation between the two. Rates of snow glide, climate conditions, and avalanche occurrence observations are examined on a continuous time scale over two seasons in an attempt to clarify this thought.  Snow glide characteristics determined in this study confirm observations by M c C l u n g et al. (1994) conducted at the same site.  Measurements show that seasonal patterns of glide are  apparent over a number of seasons. Rates are high i n the early season and i n the spring due to thinner snow packs and higher air temperatures. Glide rates fluctuate during the winter season in response to meteorological and snow pack comditions. Diurnal variations i n snow glide are not distinct enough to indicate a significant difference between daytime and nighttime snow movements.  107 The spatial nature of gliding also confirms that velocities tend to increase down slope from a crack, or break i n the snow cover.  Results indicate that glide measurements should be taken  below a free uphill surface or at a point on the slope below which the snow slab is i n tension. Tensile stress build downslope and glide speed increases with distance from the step. Glide rates measured at Shoe #1, placed above the rock step, do not give representative glide displacements and therefore should not be used in forecasting.  It is a complex combination of factors that lead to increased rates of glide. Results confirm that contribution of free-water to the snow/ground interface is the principle mechanism controlling snow glide and full-depth avalanche release. Inputs of free-water to the snow/ground interface act to increase the thickness of the saturated layer at the base of the snow pack. This can affect the amplitude of roughness that can be overcome by downslope movement, implying that it is the larger amplitudes that resist glide. Downslope flow has the greatest effect on snow before flow is channelized. Slower flow speeds result i n larger areal coverage beneath the snow pack. Determination of critical pore pressures is highly uncertain due to the lack of knowledge on the friction conditions between a mechanically varying snow pack and a smooth rock surface. Response of snow glide to contributions of free-water is also a function of the snow pack conditions; existing water drainage, temperatures, and snow density.  Radiation conditions at the study site contribute to induce snow glide due to the favourable slope and aspect of the avalanche-prone slopes. Radiation instrumentation is best placed at the Summit weather station, located i n the valley bottom.  Despite periods of horizon shielding, the  instruments are more accessible for maintenance.  Rates of snow glide are not quite as  responsive to snow melt events as they are to rainfall events. This is due to the differing volume  108 of free-water generated and also due to the placement of the glide shoes. The rock crack at the glide site is not always exposed to solar warming, whereas bare rock is almost always present on the upper slopes.  This may explain the relative lack of response of snow glide to radiation  inputs. Bare rock exposures at the top of the avalanche path may be the reason why the majority of avalanche start zones are located there. Therefore, it would be useful to know the proportion of bare rock exposed on the ridge at the top of the paths at any given time.  Evidence from two seasons of data show that full-depth avalanches are primarily associated with rain-on-snow events.  One half to three-quarters of all avalanche occurrences can be attributed  to rainfall. A l l but a few of the remaining avalanche occurrences can be attributed to rapid snow melt.  Snow melt may be generated by high air temperatures caused by incoming radiation.  However, the role of the bare rock slabs is very important i n terms of snow melt. Warming to exposed radiation results in accelerated rates of snow melt near the top of the avalanche path. Full-depth avalanche occurrence is positively correlated with rainfall events less than 12 hours to more than 36 hours after rainfall. Correlation of avalanche release with snow melt events is similar but glide rates do not respond as dramatically as i n rain events.  Input volumes from  rainfall are generally much more significant than those from snow melt.  Snow glide instrumentation on a slope adjacent to avalanche paths gives a fairly good representation of glide conditions.  Peaks in glide velocity range i n amplitude, and do not  necessarily pin-point, or predict, times of full-depth avalanche release. Results indicate, though, that glide velocities often increase during times of avalanche occurrence. may not be used as a sole predictor of avalanche release.  Snow glide velocity  Rather, it is a complex combination  of climate, topography, and snow pack conditions that contribute to full-depth avalanches.  109 Sometimes glide rates attain a peak after the times of avalanche release.  Reasons for some  variance between peaks i n glide rate and timing of avalanche release may be related to the problems of extrapolating glide rates from the study site to the avalanche slopes. Measurements taken at the glide site may be slightly slower than on steeper avalanche-prone slopes nearby. The rock slab at the glide site is an isolated slab, flanked by vegetation on either side, so boundary conditions are different than on the upper slopes. Placement of glide instrumentation in the starting zone of the avalanche path would be useful, but extremely difficult to install.  It can be concluded that for the forecast of full-depth avalanches, rainfall and snow-melt estimation using net radiation (Q*) are very good indicators. Use of snow glide at the Coquihalla field site provides a good indication of snow pack stability at the snow/ground interface.  The  current understanding of the relationship between snow glide and full-depth avalanche occurrence has been improved but a predictive relation has not been established at this time. There is room for further work i n this area.  110 REFERENCES  Akitaya, E . and H . Shimizu (1987) Obervation of Weak Layers Temperature Science Series A , N o . 46, pp. 67-75.  i n Snow Cover,  Low  Ambach, W . (1974) The Influence of Cloudiness on the Net Radiation Balance of a Snow Surface with H i g h Albedo, Journal of Glaciologv, V o l . 13, N o . 37, pp. 73-84. Bader, H . , R. Haefeli, E . Bucher, J . Neherer, A . Eckel, C . Thams and P. N i g g l i (1939) D e r Schnee und seine metamorphose. Beitrage zur Geologie der Schweiz. Geotechnische Serie, Hydrologie, L i e f 3. (English Translation, U . S . Snow, Ice, and Permafrost Establishment, Translation 14, 1954) Barry, R . G . (1992) Mountain Weather and Climate, 2nd Edition, Routledge, London, 402 p. Bois, P . , C . Obled, and W . Good (1975) Multivariate Analysis as a Tool for Day-By-Day Avalanche Forecast, Snow Mechanics Symposium, pp. 391-403. Braun, L . N . (1980) Scale Dependence of Factors Controlling the Release of Water F r o m Snow and Ice Storage, unpublished M . S c . Thesis, University of British Columbia, Department of Geography, Vancouver. Brun, E . , E . M a r t i n , V . Simon, C . Gendre, and C . Coleou (1989) A n Energy and Mass M o d e l of Snow Cover Suitable for Operational Avalanche Forecasting, Journal of Glaciologv, V o l . 35, N o . 121, pp. 333-342. Colbeck, S . C . (1972) A Theory of Water Percolation i n Snow, Journal of Glaciologv, V o l . 11, N o . 3, pp. 369-385. Colbeck, S . C . (1974) Water F l o w Through Snow Overlying an Impermeable Boundary, Water Resources Research, V o l . 10, N o . 1, pp. 119-123. Colbeck, S . C . (1975) Grain and Bond Growth i n Wet Snow, I A H S Publication N o . 114, pp. Colbeck, S . C . (1977) Short-Term Forecasting of Water Run-Off F r o m Snow and Ice, Journal of Glaciologv. V o l . 19, N o . 81, pp. 571-587. Colbeck, S . C . (1980) Dynamics of Snow and Ice Masses, N e w Y o r k Academic Press, N e w Y o r k , 469 p. Colbeck, S . C . (1982) A n Overview of Seasonal Snow Metamorphism, Reviews of Geophysics and Space Physics. V o l . 20, N o . 1 pp. 45-61. Colbeck, S . C . (1989) Snow-Crystal Growth with Varying Surface Temperatures and Radiation Penetration, Journal of Glaciologv, V o l . 35, N o . 119, pp. 23-29.  Ill Colbeck, S . C . and E . A . Anderson (1982) The Permeability of a Melting Snow Cover, Water Resources Research. V o l . 18, N o . 4, pp. 904-908. Colbeck, S . C . and G . Davidson (1973) Water Percolation Through Homogeneous Snow, i n : The Role of Snow and Ice i n Hydrology. I A S H Publication N o . 107, V o l . 1, pp. 242-257. de Quervain, M . R . (1973) Snow Structure, Heat, and Mass F l u x Through Snow, The Role of Snow and Ice i n Hydrology. Proceedings of Banff Symposium, V o l . 1, I A H S . Dunne, T . and L . B . Leopold (1978) Water i n Environmental Planning. W . H . Freeman & C o . , New Y o r k , 818 p. Dunne, T . , A . G . Price, and S . C . Colbeck (1976) The Generation of Run-off F r o m Subarctic Snow Packs, Water Resources Research. V o l . 12, N o . 4, pp. 677-685. Endo, Y . (1983) Glide Processes of a Snow Cover as a Release Mechanism of an Avalanche Slope Covered with Bamboo Bushes (in Japanese), Institute of L o w Temperature Science. Series A . pp. 39-68. Fujino, K . (1971) Measurement of F l o w D o w n Speed of Meltwater i n Snow Cover II - F l o w of Meltwater i n Snow Cover on a Slope (in Japanese), L o w Temperature Science - Series A , pp. 151-158. F u k a m i , H . , K . Kojima, and H . Aburakawa (1985) The Extinction and Absorption of Solar Radiation W i t h i n a Snow Cover, Annals of Glaciology, N o . 6, pp. 118-123. Gamier, B . J . and A . Ohmura (1968) A Method of Calculating the Direct Short-wave Radiation Income of Slopes, Journal of Applied Meteorology, V o l . 7, pp. 796-800. Granger, R . J . and D . H . Male (1978) Melting of a Prairie Snowpack, Journal of Applied Meteorology. V o l . 17, pp. 1833-1842. Gubler, H . (1980) Simultaneous Measurements of Stability Indices and Characteristic Parameters Describing the Snow Cover and the Weather i n Fracture Zone of Avalanches, Journal of Glaciology. V o l . 26, N o . 94, pp. 65-73. Haefeli, R. (1967) Some Mechanical Aspects on the Formation of Avalanches, i n : Proceedings of the International Conference on L o w Temperature Science, ed: H . Oura, Hokkaido University, Japan, V o l . 1, Part 2, pp. 1199-1213. Harding, R . J . (1986) Exchanges of Energy and Mass Associated with a Melting Snow Pack, Modelling Snow M e l t Induced Processes (Proceedings of the Budapest Symposium, July, 1986) I A H S Publication N o . 155, pp. 3-15. Hendrie, L . K . and A . G . Price (1978) Energy Balance, Snow M e l t , and Run-Off i n a Deciduous Forest, i n : Proceedings on Modeling Snow Cover Runoff, eds. S . C . Colbeck and M . Ray, U S A r m y C o l d Regions Research Engineering Laboratory, Hanover, N H , pp. 211-  112 221. Holland, S.S. (1976) Landforms of British Columbia: A Physiographic Columbia Department of Mines and Petroleum Resources, 138 p.  Outline, British  i n der Gand, H . D . and M . Zupancic (1966) Snow Gliding and Avalanches, International Symposium on Scientific Aspects of Snow and Ice Avalanches, Davos, Switzerland, I A H S Publication N o . 69, pp. 230-242. Ishikawa, N . , K . K o j i m a , and H . Motoyama (1985) Predictions of Hourly and Daily Amounts of Snowmelt by Heat Balance or Bulk Meteorological Elements (in Japanese), L o w Temperature Science Series A , pp. 63-75. Izumi, K . (1987) Studies on the Hardness of Wet Snow and its Decrease Due to Solar Radiation, Annual Report of the Research Institute for Hazards i n Snowy Areas, Niigata University, N o . 9. Izumi, K . (1989) Effects of Solar Radiation on the Formation of Weak Wet Snow, Annals of Glaciologv. N o . 13, pp. 120-123. Izumi, K . and E . Akitaya (1985) Hardness of Wet Snow, Annals of Glaciologv, V o l . 6, pp. 267268. Jordan, R . P . (1978) The Snowmelt Hydrology of a Small Alpine Watershed, unpublished M . S c . Thesis, University of British Columbia, Department of Geography, Vancouver. K a m b , B . (1970) Sliding M o t i o n of Glaciers: Theory and Observations, Review of Geophysics and Space Physics. 8(4), pp. 673-728. Kattelmann, R. (1987) Some Measurements of Water Movement and Storage i n Snow, I A H S Publication N o . 162, pp. 245-254. Keefer, D . K . , R . C . W i l s o n , R . K . M a r k , E . E . Brabb, W . M . B r o w n III, S . D . Ellen, E . L . Harp, G . F . Wieczorek, C S . Alger, R . S . Zatkin (1987) Real-Time Landslide Warning During Heavy Rainfall, Science. V o l . 238, pp. 921-925. Kobayashi, S. and K . Izumi (1991) A Study on Slush F l o w Disaster, Japan-US Workshop on Snow Avalanche, Landslide, Debris F l o w Prediction and Control, pp. 197-205. Kobayashi, S., K . Izumi, Y . E z a k i , and M . Tan (1992) Mechanical Properties of H i g h Water Content Snow, Second International Conference on Snow Engineering, Santa Barbara, California, C R R E L Special Report 92-27, eds: W . Tobiasson and E . Wright, pp. 313321. Lackinger, B . (1987) Stability and Fracture of the Snow Pack for Glide Avalanches, i n : Avalanche Formation. Movement, and Effects (Proceedings of the Davos Symposium, Sept. 1986) I A H S Publication N o . 162, pp. 229-241.  113 Lliboutry, L . (1968) General Theory of Subglacial Cavitation and Sliding of Temperate Glaciers, Journal of Glaciology, V o l . 7, N o . 49, pp. 21-58.  M a l e , D . H . (1980) The Seasonal Snow Cover, in: Dynamics of Snow and Ice Masses, ed. S . C . Colbeck, New Y o r k Academic Press, New Y o r k , 468 p. M a l e , D . H . and R . J . Granger (1981) Snow Surface Energy Exchange, Water Research. V o l . 17, N o . 3, pp. 609-627.  Resources  M a l e , D . H . and D . M . Gray (1981) Snowcover Ablation and Runoff, i n : Handbook of Snow Principles, Processes, Management, and Use, eds: M a l e , D . H . and D . M . Gray, Pergamon Press, Toronto. M a r k s , D . , J . Dozier, and R . E . Davis (1992) Climate and Energy Exchange at the Snow Surface in the Alpine Region of the Sierra Nevada, 1. Meteorological Measurements and Monitoring, Water Resources Research. V o l . 28, N o . 11, pp. 3029-3042. M c C l u n g , D . M . (1975) Creep and the Snow-Earth Interface Condition i n the Seasonal Alpine Snowpack, i n : Snow Mechanics Symposium, Proceedings of the Grindelwald Symposium, A p r i l 1974, I A H S - A I S H Publication N o . 114, pp. 236-247 M c C l u n g , D . M . (1981) A Physical Theory of Snow Gliding, Canadian Geotechnical Journal, V o l . 18, N o . 1, pp. 86-94. M c C l u n g , D . M . (1987) Mechanics of Snow Slab Failure from a Geotechnical Perspective, i n : Avalanche Formation, Movement, and Effects (Proceedings of the Davos Symposium, Sept. 1986) I A H S Publication N o . 162, pp. 475-506. M c C l u n g , D . M . and G . K . C . Clarke (1987) The Effects of Free Water on Snow G l i d i n g , Journal of Geophysical Research, V o l . 92, N o . B 7 , pp. 6301-6309. M c C l u n g , D . M . , S. Walker, and W . Golley (inpress) Characteristics of Snow G l i d i n g on Rock, Annals of Glaciology, N o . 19. M c C l u n g , D . M . , J . D . Bennetto, S. Walker, and W . Golley (1991) Snow G l i d i n g on Steep Rock, Coquihalla, B . C . , Proceedings of the International Snow Science Workshop, pp. 93-94. M c C l u n g , D . M . and P. Schaerer (1993) The Avalanche Handbook, The Mountaineers, Washington, 271 p. M e l l o r , M . (1975) A Review of Basic Snow Mechanics, I A H S Grindelwald, pp. 251-291.  Publication N o .  119.  M e l l o r , M . (1977) Engineering Properties of Snow, Journal of Glaciology, V o l . 19, N o . 81, pp. 15-65.  114 Ministry of Highways (1978) Coquihalla (Hope-Merritt) Highway Preliminary Environmental Report, Ministry of Highways - Design and Surveys Branch, June, 96 p. Ministry of Highways (1980) Snow Avalanche Atlas - Coquihalla, Province of British Columbia, Ministry of Highways -Snow Avalanche Section - Maintenance Services, A p r i l , 257 p. Ministry of the Environment (1982) Canadian Climate Normals, Volume 2 - Temperature, Atmospheric Environment Service, Canada. Ministry of the Environment (1982) Canadian Climate Normals. Volume 3 - Precipitation. Atmospheric Environment Service, Canada. Ministry of the Environment (1985) Snow Survey Measurements - Summary (1935-1985). Water Management Branch, Province of British Columbia, 193 p. Monteith, J . L . (1973) Principles of Environmental Physics, Edward A r n o l d , London. M o r r i s , E . M . (1983) Modelling the F l o w of Mass and Energy W i t h i n a Snowpack Hydrological Forecasting, Annals of Glaciologv, N o . 4, pp. 198-203  for  N y e , J . F . (1969) A Calculation on the Sliding of Ice over a Wavy Surface Using a Newtonian Viscous Approximation, Proceedings of the Royal Society of London. A(311), pp. 445467. Nye, J . F . (1970) Glacier Sliding Without Cavitation i n a Linear Viscous Approximation, Proceedings of the Royal Society of London, Series A . V o l . 315, pp. 381-403. Obled, C . and Harder (1978) A Review of Snow Melt in the Maountina Environment, i n : Proceedings on Modeling Snow Cover Runoff, eds. S . C . Colbeck and M . Ray, U S A r m y C o l d Regions Research Engineering Laboratory, Hanover, N H , pp. 179-204. Obled, C . and W . Good (1980) Recent Developments of Avalanche Forecasting by Discriminant Analysis Techniques: A Methodological Review and Some Applications to the Parsenn A r e a (Davos, Switzerland), Journal of Glaciologv. V o l . 25, N o . 92, pp. 315-346. Oke, T . R . (1987) Boundary Layer Climates. 2nd Edition, Routledge, New Y o r k , 435 p. O ' N e i l l , A . D . J , and D . M . Gray (1972) Solar Radiation Penetration Through Snow, U N E S C O The Role of Snow and Ice in Hydrology, Proceedings of the Banff Symposium, I A H S , V o l . 1, pp. 227-241. Price, A . G . and T. Dunne (1976) Energy Balance Computations of Snowmelt i n a Subarctic Area, Water Resources Research, V o l . 12, N o . 4, pp. 686-694. Raymond, C . F . and K . Tusima (1979) Grain Coarsening of Water-Saturated Snow, Journal of Glaciologv, V o l . 22, N o . 86, pp. 83-105.  115 Salm, B . (1977) Snow Forces, Journal of Glaciology. V o l . 19, N o . 8 1 , pp. 67-100. Shimizu, H . (1970) A i r Permeability of Deposited Snow, L o w Temperature Science Series A . V o l . 22, pp. 1-32. Steyn, D . G . (1980) The Calculation of V i e w Factors from Fisheye-Lens Photographs - Research Note, Atmosphere-Ocean. V o l . 18, N o . 3, pp. 254-248. Wakahama, G . (1965) Metamorphism of Wet Snow (in Japanese), L o w Temperature Science Series A . V o l . 23, pp. 51-66. Wakahama, G . (1968) The Metamorphism of Wet Snow, International Association of Hvdrological Sciences (IAHS) Publication N o . 79. pp. 370-379. Wankiewicz, A . (1976) Water Percolation Within a Deep Snowpack - F i e l d Investigations at a Site on Mount Seymour. British Columbia, unpublished P h . D . Thesis, Department of Geography, Universiyt of British Columbia, 177 p. Wankiewicz, A . (1978) Water Pressure i n Ripe Snowpacks, Water Resources Research, V o l . 14, N o . 4, pp. 593-600 Water Survey of Canada (1991) Historical Streamflow Summary - British Columbia to 1990, Inland Waters Directorate, Water Resources Branch, Ottawa. Yamada, Y . , Y . Nohguchi and T . Ikarashi (1991) Snow Avalanche Release due to Instability of Snow Glide M o t i o n , Proceedings of the Japan-US Workshop on Snow Avalanche, Landslide, Debris F l o w Prediction and Control, Sept 30-Oct 2, 1991, Tsukuba, Japan, pp. 105-114.  APPENDIX A  List of Full-Depth Avalanche Occurrences Selected for Analy  Appendix A: List of Full-Depth Avalanche Occurrences Selected for Analysis S e a s o n : 1992-93 Month  Day  AM/PM  Path No.  12 12 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2  12 15 23 23 25 25 25 25 25 27 28 28 30 30 30 30 30 30 31 31 31 31 31 4  3 3 3 3  16 23 25 25  PM AM PM AM AM AM PM PM PM AM AM AM AM AM PM PM PM PM AM AM AM PM PM AM PM AM AM AM AM  46.6 46.5 46.8 4/.0fa 47.35 47.05 44.3 46.3 46.8 44.3 46.8 46.5 46.3 46.3 47.7 47.7 47.05 47.35 46.5 47.55 46.8 46.8 47.35 4/.y 4/.3b 46.5 47.7 46.5 46.8  12  1  Size* 1.b 2.0 1.5 1.5 2.5 2.0 2.0 2.5 2.5 1.5 2.0 2.0 1.5 . 1.5 1.5 1.5 2.0 3.0 2.0 2.0 2.0 2.5 2.0 y.u 2.0 2.0 1.0 2.0 1.5  (continued)  S e a s o n : 1993-94 Month  Day  AM/PM  Path No.  Size*  Month  Day  AM/PM  Path N o .  Size*  12 12  10 10 10 10 10 10 11 11 11 11 11 11 13 15 15 2  AM AM AM AM AM AM AM AM AM AM AM AM AM AM AM PM AM AM AM AM AM AM AM AM AM AM AM AM AM PM AM PM AM AM AM PM PM PM PM  46.8 47.35 47.7 47.55 46.3 46.3 46.3 46.5 47.35 46.8 46.8 47.55 47.35 47.7 47.35 46. b 46.8 46.8 46.8 46.8 46.8 47.9 47.35 46.5 46.5 46.3 46.3 47.7 47.7 47.35 47.35 46.8 47.05 47.05 47.35 46.3 46.3 46.5 47.35  2.0 2.5 2.0 2.5 3.0  1 1 1  22 22 26 28 28 28 28 28 1 1 1 2 2 2 2 2 2 9 y 9 9 9 10 13 14 14 14 14 14 14 15 15 15 15 15 15 28 29 30  AM AM PM AM AM AM AM PM AM AM AM AM AM AM AM AM AM AM AM AM AM PM AM AM PM PM PM PM PM PM AM AM AM AM AM AM HM AM AM  46.b 46.8 47.55 47.35 4/.bb 47.55 47.55 47.55 47.55 47.55 47.05 47.7 46.3 46.5 46.5 47.35 47.05 47.7  2.0 2.0 2.5 2.5 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 2.0 1.5  12  12  12  12 12  12 12 12 12  12 12 12 12  1 1 1  1 1 1  1 1  1 1  1 1 1  1 1 1 1 1 1  1 1 1 1 1  2  3 4 4 4 4 4 4 4 4 4 6 7 9 11 11 12 13 13 13 13 13 20  2.0 1.5 1.5 1.5 2.0 1.5 2.0 2.5 2.0 2.0 2.0 2.5 2.5 2.0 2.0 2.0 2.5 2.5 2.0 2.0 2.0 2.0 2.5 2.5 2.5 2.5 2.5 2.0 2.5 2.5 2.5 2.5 2.5 2.5  2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3  * - C a n a d i a n S n o w A v a l a n c h e Size C l a s s i f i c a t i o n ( M c C l u n g a n d Schaerer, 1993)  M.I  46.3 46.3 47.15 46.5 47.7 46.5 46.5 47.35 47.35 46.8 46.8 46.8 46.8 46.8 46.8 47.7 46.5 4/55 47.55 46.8  1.b 1.5 1.5 1.5 2.0 2.5 1.0 1.0 2.0 2.0 1.0 1.0 2.0 2.0 2.0 2.0 2.5 3.0 2.6 2.5 3.5  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0098908/manifest

Comment

Related Items