The Use of Dendrochronology to Determine Avalanche Frequency Along the Avalanche Path East of Balu Peak, Within the Balu Pass Trail, Roger's Pass, BC.

1 The Use of Dendrochronology to Determine Avalanche Frequency Along the Avalanche Path East of Balu Peak, Within the Balu Pass Trail, Roger's Pass, BC. Allison Dick, Donald Mcfarlane, and Robyn McGregor Geography 477: Field Studies in Physical Geography Instructor: Dan Smith University of Victoria: Department of Geography December 2011

2 Abstract Glacier National Park in British Columbia is an area of steep mountainous terrain and heavy snowfall. This results in frequent avalanche activity which can lead to the destruction of property and the loss of human life. Research focused on determining the potential for avalanches and the frequency of avalanche events can help to mitigate this risk. Dendrochronology (the study of tree rings), can be used as a basis for this research. By analyzing a specific avalanche path along the Balu trail in Glacier National Park using dendrochronological techniques, an avalanche history could be determined. This information was then used in conjunction with weather station climate data to attempt to ascertain an avalanche activity pattern, and hopefully aid in the prediction of future avalanche events.

3 Table of Contents Abstract...2 Table of Contents...3 List of Figures...3 List of Tables...4 Acknowledgements...5 1. Introduction...(6) 1.1 Introduction...(6) 1.2 Background - Glacier National Park...(7) 1.3 Avalanche Characteristics...(8) 1.4 Avalanche Risk...(10) 1.5 Dendrochronology...(11) 2. Site Area...(13) 2.1 The Balu Pass Tail...(13) 2.2 Study Site...(13) 3. Methodology...(15) 3.1 Data Collection and Analysis...(15) 3.2 Dendrochronology...(15) 3.3 Weather and Climate...(16) 3.4 Event-response Index...(17) 4. Results...(17) 4.1 Tree Ages...(17) 4.2 Event Response Index...(18) 4.3 Snowpack...(19) 4.4 Precipitation and Temperature...(20) 4.5 Tables of Results...(22) 4.6 Slope Reconstruction...(25) 5. Discussion...(25) 6. Conclusion...(26) 7. References...(28) List of Figures Figure 1.3.1 Typical Avalanche Debris in Run Out Zone...(10) Figure 1.5.1 Sub Alpine Fir with External Scars...(11) Figure 1.5.2 Tree disc with evidence of reaction wood...(12)

4 Figure 2.2.1 View of Avalanche Path...(14) Figure 3.2.1 Example of Increment Borer...(15) Figure 4.1.1 Tree Ages...(18) Figure 4.2.1 Event Response Index...(19) Figure 4.3.1 Snow depth vs annual potential avalanche evidence...(19) Figure 4.4.1 Precipitation as rain vs annual potential avalanche activity...(20) Figure 4.4.2 Temperature Variations vs Annual Potential Avalanche Activity...(21) Figure 4.6.1 Generalized Model of Slide Site Terrain...(25) List of Tables Table 4.5.1 Minor, Moderate, and Major events as determined by Lab Analysis of Tree Ring Samples (22) Table 4.5.2: Maximum and Minimum Temperatures and Precipitation by Year...(23) Table 4.5.3: Snowpack data...(24)

5 Acknowledgements: First and foremost, we would like to thank Dr. Dan Smith and Dr. James Gardner for an inspiring and valuable field school experience. We would also like to express our sincere gratitude to Kara Pitman for her guidance and assistance throughout the research phase of this report, and Bethany Coulthard and Jodi Axelson for their technical expertise and unending support. We would also like to thank Parks Canada for providing a pristine backdrop for our research endeavours.

6 1. Introduction 1.1 Introduction Snow avalanches are an inherent natural process in mountainous environments. Avalanche potential is influenced by terrain and climatic processes, and they can be triggered naturally or by human activity. In British Columbia, Canada, Glacier National Park is an area that receives a great deal of snow throughout the winter months, as well as many backcountry visitors (Parks Canada, 2011b). The steep and rugged mountains that are found in the park contribute to its renowned natural beauty, however they also contribute to the danger associated with backcountry travel during the winter months. The Balu Pass Trail is a highly popular area used by backcountry travelers throughout the winter. The terrain surrounding the trail has been classified as complex avalanche terrain by Parks Canada, and this poses a great deal of risk for any area users. There are numerous avalanche tracks along the trail, and a specific one was chosen to investigate avalanche activity by using dendrochronological analysis. Dendrochronology, the study of tree dating and tree ring analysis, is commonly used as a tool to determine avalanche frequency and magnitude. Past avalanche events are recorded in the tree ring succession as distinct darker and wider rings. These rings can be dated and then cross referenced against climate data to determine the climactic influences necessary in the development of avalanche conditions. The purpose of this study was to investigate the relationship between climatic data and avalanche frequency for a specific avalanche track along the Balu Pass Tail in Glacier National Park. This study sought to determine if dendrochronology can be used as a predictive method for avalanche risks, based upon the trends and cycles of avalanche activity in previous years.

7 1.2 Background - Glacier National Park Glacier National Park, established in 1886 with the completion of the transcontinental Canadian Pacific Railway, lies within the Selkirk and Purcell Ranges of the Columbia Mountains of southeastern British Columbia. Within Glacier National Park Rogers Pass National Historic site, which commemorates this significant and perilous link of the Canadian Pacific Railway, is located at the summit of the Trans Canada Highway corridor. The highway was completed in 1962, and since then recreational tourism has increased dramatically throughout the year (Parks Canada, 2011a). Glacier National Park was the second national park to be created in Canada, and the first in British Columbia. The park itself is a rugged and steep mountainous landscape capped by numerous glaciers, and covers a total area of 1,350 km2. Glacier National Park is a diverse environment that has three distinct life zones that are mostly governed by elevation. The zones are characterized as, from lowest to highest elevation; interior rainforest, subalpine life zone, and no forest zone (Parks Canada, 2011b). These distinctly different zones create the diverse mosaic of habitats that are home to a wide variety of organisms, including many threatened and endangered species. The climactic conditions of the park are characteristic of the Columbia Mountains. High annual precipitation, heavy rains in the summer and deep snow pack in the winter, is caused by the wet and mild westerly air masses that are intercepted by the Columbia Mountains. Like any other steep and mountainous environment with a deep winter snow pack, there is a high risk and occurrence of avalanches. Along the Trans Canada Highway, avalanches are so abundant that a special team, the Rogers Pass Mobile Avalanche Control Program, was created to manage slopes that could be hazardous to those using the highway. The Rogers Pass Avalanche Control Program is the largest of its kind in the world, and this can attest to the severity of risk that is posed by the terrain and its deep snow pack. Parks Canada avalanche forecasters release a public avalanche bulletin every day to report the current avalanche risk within the given area (Parks Canada, 2011b). Many recreational tourists use the park trails throughout the year, and in recent decades ski touring has become an increasingly popular winter activity in the park. The park tails cross and intersect

8 numerous avalanche paths, and this poses a high level of risk for winter users. Much of the park s terrain is classified as complex avalanche terrain; this means that avalanches will happen often and can be of disastrous magnitudes (Parks Canada, 2011c). Though avalanche and snowpack information for the area is well documented, the terrain is not controlled like it is in ski resorts, and avalanche fatalities have occurred in the past. Having an understanding of the frequency and magnitude of avalanches on given slopes is a key tool that can be used to make decisions in the backcountry. There currently is a lack of information regarding frequency and magnitude of avalanches on specific slopes that intersect trails within the park. For this reason tree ring data was collected on the Balu Pass Trail, a popular ski touring trail, and dendrochronological analysis was used to determine the frequency of avalanches on a specific slope. 1.3 Avalanche Characteristics A snow avalanche is a rapid down slope release of a mass of snow. They can be highly destructive to natural and manmade features, and dangerous to those who travel through, or spend time within, avalanche zones. Of particular interest, due to their destructive capabilities, are slab avalanches (Jamieson, 2004). Slab avalanches are responsible for the majority of avalanche related fatalities within North America. According to the Canadian Avalanche Centre (2011), these occur on slopes that are generally between 25 and 45 degrees and can be formed by wet or dry snow, storm snow events, or consistent deposition of snow via wind. In each case, a mass of snow is sheared from an underlying surface due to the presence of a weak layer (or weak interface) within the snowpack. These shearing points are most often caused by the development of weak interfaces (sun crusts and rain crusts) or a weak layer (faceting and surface hoar). Sun crusts form when an increase in solar radiation results in a surficial melting of the snow layer. As temperatures decrease, this melted layer refreezes and acts as a destabilizing point for the snow mass. Similarly, increased temperatures may result in precipitation falling as rain. This saturates the snow surface, which then freezes, creating a shear and icy surface. Surface hoar occurs during clear and calm nights; snow radiates a great deal of heat that accumulates during the day

9 and in doing so the snow surface becomes very cold. This warm supersaturated air sits above the snow surface. The water vapour then condenses as it cools onto the snow surface and forms fine feather-like icy crystals. This is simply the winter equivalent of dew. Faceting, which is a process that builds angular grains (facets) in the snowpack that bond poorly to one another and other grains present in adjacent layers, are formed by strong temperature gradients in the snowpack. When a strong temperature gradient is present in the snowpack (>1 degree / 10cm), water vapour moves rapidly from warm grain surfaces to colder ones. Since snow is an excellent insulator the temperature at the bottom of the snowpack is relatively warm (at or near 0 degrees C), and cools towards the snow surface. When there is a large temperature gradient within the snowpack, water vapour from the warmer layers wants to move from areas of high concentration to areas of low concentration (the colder snow layers). As the water vapour diffuses rapidly through the snowpack, it changes rounded crystals into faceted ones. In other words, this changes the affected snow from strong cohesive snow, to weak non cohesive snow (Temper, 2010). In each case, the snow mass is destabilized until an avalanche is triggered. Triggers can be natural (mass becomes too great and gravity triggers release of avalanche) or human induced (snowmobilers highlining, or backcountry skiers traversing across the potential avalanche zone). Once triggered, the snow mass then travels down an avalanche path until velocity is lost and mass is deposited (Canadian Avalanche Centre, 2011). The avalanche path is defined by three zones: the starting zone, the track zone, and the run-out zone (Johnston & Jamieson, 2011). Snow is accumulated in the starting zone until the unified mass reaches a failing point and begins to travel down slope through the track zone. As the snow mass descends, it gains velocity and debris before reaching the run-out zone (an area of lower slope angle) where the snow and debris is deposited (Canadian Avalanche Centre, 2011). The avalanche path chosen for this study is confined in its run-out zone by an uphill slope on the opposite side of the valley. The result is a dense collection of debris deposited at the valley bottom.

10 Figure 1.3.1 Typical Avalanche Debris in Run-out Zone 1.4 Avalanche Risk The study of avalanches in populated and recreational areas is important in order to protect the safety of all area users. In Glacier National Park, backcountry skiers and recreational users must be aware of the inherent avalanche risks, and rely on reports from Parks Canada and the Canadian Avalanche Association to help them make informed decisions. While these organizations provide detailed and accurate reports on conditions, tragedies do still occur. In 2003, 17 students were caught in an Avalanche along the Balu trail in Glacier National Park during a back country skiing expedition. 7 did not survive (Canadian Broadcasting Corporation, 2003). It was an eye opening event, and one that highlighted and reaffirmed the risks associated with recreational use in mountainous areas. According to the CBC news article (2003), the students and trip coordinators were aware of the avalanche risk rating of considerable,

11 indicating a possibility of avalanches, yet due to their considerable expertise within the group, decided to proceed. 1.5 Dendrochronology One of the many tools available to those interested in studying avalanches is determining the magnitude and frequency of events as recorded by changes in tree ring characteristics. Dendrochronology utilizes either an increment borer, which extracts a sample core of the tree rings without harming the individual, or by sawing discs from the base of the tree. The latter, unfortunately, does result in the death of the individual tree. The rings may be later analyzed in a lab setting to determine the age and climactic controls acting upon the tree. There are two features that are of particular interest in using dendrochronology to study avalanche frequency and magnitude: scars and reaction wood. Scars are formed when debris or snow is dragged along a tree in an event such as a snow avalanche, leaving a marked scar on the tree s bark. By backdating on a core or a disc taken at the scar sight from the scar to the pith, one can determine the date of the scar in relation to the tree s total age. Figure 1.5.1 Sub Alpine fir with Scars

12 Unlike the external evidence of scars, reaction wood is recorded within the tree rings, and is therefore accessed using increment borers or by slicing discs from the tree. Reaction wood forms when an event causes the tree to bend, such as from the force of an avalanche. As the tree continues to grow, it attempts to realign and become once again erect. This results in a marked and noticeable change in the tree ring pattern reaction wood is characterized by wider and darker rings on the increment borer cores, and is shown as a distinctive pattern on tree discs (figure 1.5.2). Reaction wood can be caused by a multitude of events, such as rockfalls, high winds, or even heavy snowpack. Therefore, cross referencing samples for years of distinct reaction wood occurring in a controlled area is necessary to determine the cause of the reaction wood. Figure 1.5.2 Tree disc with evidence of reaction wood Both the discs and the cores are analyzed under microscope in a lab setting, and tree rings are counted to determine total age of the tree and years of scars or reaction wood evidence. As samples are cross referenced to each other, a chronology may begin to emerge that highlights the avalanche frequency and history along the chosen path.

13 2. Study Area 2.1 The Balu Pass Trail The trailhead for the Balu Pass Trail is located behind the parking lot of the Rogers Pass Discovery center, which makes this trail easily accessible and very popular among visitors. The Balu Pass Trail is a 6.4 km hike up to the Balu Summit, which passes through stands of Mountain Hemlock (Tsuga mertensiana) and Englemann spruce (Picea engelmannii), numerous avalanche run-out zones, and finally up into the alpine meadow and wetland area near the summit. The trail gets its name from the First Nations word baloo, meaning bear, and as the name suggests the terrain along this trail is prime bear habitat (Parks Canada, 2011d). The possibility of bear sightings throughout the spring, summer, and autumn months attracts many visitors in spite of the inherent dangers of bear sightings. This is not the only dangerous and exciting attraction that entices visitors to journey off the highway; in the winter months the deep snowpack and steep terrain of Balu Pass Trail attracts ever-increasing numbers of backcountry skiers and snowboarders. The terrain surrounding the trail has been classified by Parks Canada, as complex avalanche terrain. Complex avalanche terrain is described as, exposure to multiple overlapping avalanche paths or large expanses of steep, open terrain; multiple avalanche starting zones and terrain traps below; minimal options to reduce exposure. Complicated glacier travel with extensive crevasse bands or icefalls (Parks Canada, 2011c). Since the terrain surrounding the trail is so dynamic, steep, and complex the risk of avalanche activity throughout the Balu Pass area is high. 2.2 The Study Area The avalanche path that was selected for study is located approximately 4 km from the head of the Balu Valley Trail in Glacier National Park (51 17 25 N, 117 34 12 W) with an elevation at the trail of approximately 1800m (Google Earth, 2011). It was chosen for its ease of access to vegetation, as well as a clearly delineated avalanche path. The main vegetation on the avalanche path was Subalpine fir (Abies lasiocarpa) and Aspen.

14 Since the terrain surrounding the trail is classified as complex avalanche terrain, there are numerous avalanche tracks that intersect the trail and pose risk for winter trail users (Parks Canada, 2011c). The track that was chosen for this study because it intersects the trail on a steep slope, which is very exposed. The study site is also a run-out zone for the Bruins Pass area, which is very popular among backcountry skiers and snowboarders due to its slightly more manageable terrain. The trees that were sampled from the track were all Subalpine fur (Abies lasiocarpa). The height and age of the sampled trees varied; however the younger and smaller trees tended to be located closer to the center of the avalanche track, where as the older and larger trees tended to be further away from the center. There was a great deal of evidence from past avalanches on the living trees. This included extensive scarring, trunk deformation, and stripped off uphill branches on the main stem of trees. There was also an abundance of Alder shrubs throughout the study area. These features can be used in conjunction with a dendrochronological reconstruction of avalanche activity to assess the frequency of avalanches along the selected avalanche track. Figure 2.2.1 View of Avalanche Path

15 3. Methodology 3.1 Data Collection and analysis Field data collection was primarily focused on vegetative evidence, with some data compiled for a topographic reconstruction of the avalanche slope. Slope data was found using a digital range finder to determine the angle of slope in the accumulation, track, and run-out zone of the avalanche path. This data was later used to create a generalized model of the underlying terrain. 3.2 Dendrochronology In order to determine the frequency of large avalanche events, tree cores and discs were taken from vegetation along the North East section of the avalanche path. In the accumulation zone, tree core samples were taken from individuals with visible markings and scars, all of which were Subalpine firs. Samples were discarded if they did not reach the pith of the tree. Tree cores were taken with a 5mm increment borer. The samples were taken low to the ground (approximately breast height) to ensure the maximum presence of tree rings, in order to establish the most accurate dates Figure 3.2.1 Example of Increment Borer

16 In the track zone, discs were sawed from Subalpine firs, with an emphasis placed on trees that exhibited scarring, potential reaction wood, and trees that had branches stripped off on the uphill side of the main stem. The discs were taken throughout the length, as well as laterally from the center of the avalanche path, for potential determination of the magnitude of the avalanche events. The samples were then taken back to the University of Victoria Tree Ring Lab for further analysis. The tree core samples were glued and mounted to boards, sanded (using six different grades of sandpaper), and analyzed under a microscope to determine the age of the tree as well as dates of avalanche evidence in the form of reaction wood and scars. Rings were counted inward from the bark with an error margin of +/- two years. Scars and dark, wide rings (reaction wood) were noted as years of potential avalanche activity. Similarly, discs were sanded and years were counted inward from the bark to determine total age, year of reaction wood, and year to which a scarring event occurred. All findings were compiled into a spreadsheet for graphing and cross referencing of the results. 3.3 Weather and Climate For weather and climate data, nearby weather stations (Mt. Abbot Station (2A14) and Glacier Station (2A02) Water Stewardship Division of British Columbia) were used to determine the historical snowpack data in the area. These sites were chosen based on their proximity to the study site. Data was graphed for the years of 1959 2011, as these stations did not contain recorded climactic data earlier then 1959. Snow depth data was retrieved from each station and plotted against the number of times reaction wood and scars were recorded for each year, in the hopes of discovering an emerging trend. Similarly, climate data was collected from the Glacier National Park, Rogers Pass weather station (Environment Canada) where temperature and total precipitation fallen as rain (mm) was recorded for each month from 1965 2007. The monthly rainfall, and minimum and maximum temperatures for November through February for each year were averaged and graphed against the number of times reaction wood and scars were present within the sample set for each year.

17 Precipitation, snow pack, temperature, and temperature variation are considerable factors in determining seasonal avalanche risk. These parameters were chosen in an attempt to highlight the main climactic controls that determine the conditions for avalanche formation. In winter seasons with high fluctuations of temperatures, rain crusts, and sun crusts are likely to occur. Likewise, in seasons of uncharacteristically cold temperatures, faceting and surface hoar are likely to occur. It is therefore necessary to study many weather and climactic phenomena when determining avalanche risk. 3.4 Event-response Index An event-response index (ERI) was determined by dividing the total number of reactions recorded using dendrochronology within a year by the number of samples collected within that year, multiplied by 100 (Schweingruber, 1988). This index was created to help delineate the avalanche event s magnitude and frequency (Johnson, & Smith, 2010). The minimum ERI value used to determine the high magnitude events is a value determined by the studier with respect to the site understudy (Butler, et al., 1987). Snow avalanche frequency was determined by subtracting the date of the earliest high-magnitude avalanche event (based on the minimum ERI value) by the sampling year, and divided by the total number of high-magnitude avalanche events (Johnson, & Smith, 2010). 4. Results 4.1 Tree Ages Our sampling resulted in 111 years of usable data with 150 events of scars and, or reaction wood recorded within the samples (with an error of +/- two years). The age of the trees within the study area ranged from 45 to 350 years of age

Age (years) 18 Age of Tree Core and Ring Samples from Avalanche Site, Balu Trail, Roger's Pass, BC 350 300 250 200 150 100 50 0 Field Samples from lowest altitude to highest altitude Figure 4.1.1 Tree Ages 4.2 Event Response Index High magnitude events were determined by using a minimum index value of 25%. This value was determined by taking the number of years that displayed a higher incidence of reaction wood and scarring, divided by the number of samples and multiplied by 100. From the minimum ERI value, it was established that four or more incidence of reaction wood or scars within the sample set show a moderate to major event occurring, while detection of three reaction wood or scars within the sample set show a minor event, both having an error of + / two years. The moderate to major events necessitate attention because they would be responsible for doing the largest amount of damage, while potentially being the most fatal. By subtracting the date of the earliest high-magnitude avalanche event (based on the minimum ERI value) by the sampling year, divided by the total number of high-magnitude avalanche events (Johnson, & Smith, 2010), the recurrence interval for this study area was determined to be approximately 9.5 years.

1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 Snow Depth (cm) Number of Reaction Wood and Scars 19 Figure 4.2.1 Event Response Index 4.3 Snowpack The results show that the events are generally following a similar trend when compared to snow pack depth throughout the years, where heavier snowfall events commonly resulted in larger avalanche events. 350 300 250 200 150 100 50 Relationship Between Average Annual Snow Depths and Reaction Wood / Scars Found in Each Year Along the Balu Trail, Roger's Pass, BC, 2011 5 4 3 2 1 0 0 Number of reaction wood and Year scars Average Snow Depth from Mt Abbot and Glacier weather stations Figure 4.3.1 Snow depth vs annual potential avalanche evidence

Precipitation fallen as Rain (mm) Impact Scars and Reaction Wood (# of times) 20 4.4 Precipitation and Temperature The results from the precipitation and temperature data follow very closely to that of the snow pack depth data. The moderate to major events occurred during years with increased amounts of rainfall, although a considerable amount of variability is present from 1970 to 1980 40 35 30 25 Average Precipitation Fallen as Rain from November to February, 1965-2007 in Relation to the Number of Times Reaction Wood or Scars were Present 6 5 4 20 3 15 10 5 2 1 0 0 Year Average Rainfall (mm) Number of reaction wood and scars recorded Figure 4.4.1 Precipitation as rain vs annual potential avalanche activity

21 The moderate and major events also occurred during the years with larger temperature variations, and temperatures rising above freezing for those years Figure 4.4.2 Temperature Variations vs Annual Potential Avalanche Activity Two other minor events were detected based on the number of times reaction wood and scars were present in 1906 and 1939, and one major event was detected in 1954. There is less evidence of avalanche activity in the earlier years due to 35% of our samples being under 80 years of age. All event magnitudes and dates from 1900 to 2011 are displayed in table 4.5.1, although event certainty is not as high from 1900 to 1958 due to the lack of weather information available for those years. With our sample size, a definitive trend is not clearly displayed but based on our data and climatic analysis the major events occur within 10, 14, and 16 years, which with further research and added sampling, a definitive trend could emerge.

22 4.5Tables of Results Order of Events Year # of Reaction wood and scars present within the year Major, Moderate, or Minor Events Time between events (years) Time between Moderate to Major Events (years) 1 1906 3 Minor 0-2 1939 3 Minor 33-3 1954 5 Major 15 0 4 1968 4 Moderate to 14 14 Major 5 1971 3 Minor 3-6 1974 3 Minor 3-7 1976 3 Minor 2-8 1979 3 Minor 3-9 1984 4 Moderate to 5 16 Major 10 1988 4 Moderate to 4 4 Major 11 1989 5 Major 1 1 12 1993 3 Minor 4-13 1998 3 Minor 5-14 1999 4 Moderate to 1 10 Major 15 2000 3 Minor 1-16 2004 3 Minor 4 - Table 4.5.1: Minor, Moderate, and Major events as determined by Lab Analysis of Tree Ring Samples

Average Maximum and Minimum Temperatures and Rainfall by Year Average Max. Temp. Average Minimum Average Rainfall Year ( o C) 23 Temp. (mm) 1965 3.6-19.7 21.5 1966 2.2-19.15 1.2 1967 1.275-19.33 2.3 1968 2.225-25.98 20.3 1969 0.15-21.13 8.6 1970 1.95-23.5 0.0 1971-0.125-22.5 9.5 1972 0.425-26.13 0.2 1973 1.675-23.2 5.4 1974 2.075-19.45 11.2 1975 2.075-26.13 33.7 1976 1.275-20.43 9.0 1977 1-22.78 0.0 1978 0.4-25.03 19.5 1979-0.5-23.25 11.1 1980 2.625-24.13 36.1 1981 2-20.63 5.7 1982 0.5-24.63 3.4 1983 1.125-19.75 16.2 1984 0.25-24.5 15.5 1985-0.75-28.63 1.3 1986 2.125-20.38 17.7 1987 2-20.13 12.9 1988 2-22.88 14.2 1989 2-22 34.4 1990 1.375-23.5 30.5 1991 1.5-17.38 10.6 1992 1.125-18.5 4.4 1993 0.375-24.13 2.4 1994 0.875-20.88 1.7 1995 2.25-22.88 20.7 1996 0.75-28.63 9.3 1997 1.5-19.25 4.1 1998 2-22 6.5 1999 1.25-16.13 24.3 2000-0.125-17.38 0.0 2001 1.25-19.13 17.2 2002 1.25-18.25 17.4 2003 0.375-19.25 2.5 2004 2-20.63 13.9 2005 2.375-21.5 36.2 2006 1.25-19 15.6 Table 4.5.2: Maximum and Minimum Temperatures and Precipitation by Year

24 Snowpack Data from Two Nearby Weather Stations Year Mt Abbot Snow Pack Data (cm) Glacier Snow Pack Data (cm) Average Snow Depth (cm) 1959 290 190 240 1960 257 142 199.5 1961 262 160 211 1962 282 183 232.5 1963 234 137 185.5 1964 290 175 232.5 1965 251 150 200.5 1966 315 206 260.5 1967 371 249 310 1968 274 157 215.5 1969 254 147 200.5 1970 229 142 185.5 1971 275 193 234 1972 312 221 266.5 1973 257 157 207 1974 386 201 293.5 1975 259 185 222 1976 320 188 254 1977 157 122 139.5 1978 229 145 187 1979 198 138 168 1980 230 150 190 1981 205 133 169 1982 305 187 246 1983 216 143 179.5 1984 269 160 214.5 1985 188 130 159 1986 282 194 238 1987 211 148 179.5 1988 258 140 199 1989 264 144 204 1990 319 192 255.5 1991 310 184 247 1992 259 187 223 1993 166 118 142 1994 227 153 190 1995 224 128 176 1996 307 200 253.5 1997 252 185 218.5 1998 254 156 205 1999 357 200 278.5

25 2000 284 157 220.5 2001 151 111 131 2002 282 143 212.5 2003 198 134 166 2004 243 126 184.5 2005 235 173 204 2006 244 141 192.5 Table 4.5.3: Snowpack data 4.6 Slope Reconstruction Using the digital rangefinder, it was determined that the slope varied from 14 to 28.8 degrees, placing a majority of the slope within the parameters of a potential slide site. Figure 4.6.1 Generalized Model of Slide Site Terrain

26 5. Discussion While it was hoped that a cyclical trend would emerge that would aid in the prediction of avalanche events, our results do highlight the relationship between multiple climactic controls and avalanche activity. The graphed results of temperature variation, snowpack, and precipitation show anomalous weather patterns leading to increased avalanche events. In years with high precipitation, and therefore high snowpack, multiple avalanche events are often recorded. Similarly, years with high temperatures and marked temperature fluctuations, avalanche activity appears to increase. In 1968, for example, temperature fluctuations were high, with a change of 28 degrees C between November and February. In that same year, 4 potential avalanche events were recorded. Similarly, 1989 saw 34.5 mm of rainfall, and a 24 degree C temperature fluctuation. 5 potential events were recorded for this year. Conversely, 1970 received 0mm of rain, and only one potential avalanche event. 2003 received low precipitation, fewer temperature fluctuations, and a low snowpack. As a result, one potential event was recorded. When recording reaction wood, it was important to make note that reaction wood is not only caused by avalanches. Rock falls, soil creep, extreme weather, and even a heavy snow pack can all result in reaction wood formation. It is difficult to discern avalanche activity, and assuming that each year of reaction wood recorded is an avalanche event leads to a potential for erred results. Additional sources of error included a lack of equipment in the field. GPS units and tape measures would have allowed for a more detailed spatial analysis of the distribution of scars and instances of reaction wood. This would aid in determining the extent of avalanche activity. In the lab setting, due to the inexpertise of the users, some error was introduced in counting rings, scars, and reaction wood incorrectly. 6. Conclusion Despite the previously mentioned errors, some conclusions may be drawn based upon our data and research. Our original research question asked can dendrochronology be used as a predictive method

27 for avalanche risks, based upon the trends and cycles of avalanche activity in previous years?. It was concluded that dendrochronology may be a very effective predictive technique, but the scope and limitations of this particular research project did not yield a definitive trend and therefore was unable to make predictions for the particular slide site. It was also hypothesized that analyzing multiple climactic controls is necessary to determine avalanche potential. It was found that there was no singular climactic control that lead to high avalanche event years, but rather each climate factor studied (precipitation, temperature, temperature variations, and snowpack) contributed to years of assumed avalanche events. Given the scope of Canada s mountainous regions, it is difficult to make detailed information available about specific and small scale backcountry areas. The resources necessary for dendrochronological studies at this small of a scale along the entire Balu Pass recreation area would be staggering. However, site specific data concerning weather conditions is readily available to the public. Our research highlights the relationship between climate and avalanche frequency, which helps to reinforce the need for backcountry users to be aware of the conditions of the area prior to setting out. This increased awareness allows backcountry users to enjoy the natural splendour and beauty of rugged mountain environments, while lessening the risk of loss of life.

28 7. References Butler, D. R., Malanson, G. P., and Oelfke, J. G. (1987). Tree-ring analysis and natural hazard chronologies: minimum sample sizes and index values. The Professional Geographer, 39, 41 47. Canadian Avalanche Centre. (2011). Online Avalanche Course. Retrieved November 5, 2011, from Canadian Avalanche Centre: http://www.avalanche.ca/cac/training/online-course Canadian Broadcasting Corporation. (2003, February 4). Teens were Caught in Powerful, Fast-Moving Avalanche. Retrieved November 05, 2011, from CBC: http://www.cbc.ca/news/canada/story/2003/02/02/revelstoke030201.html Historic Snow Survey Data for British Columbia. (n.d. a). Water Stewardship Division. Retrieved November 20, 2011, from http://a100.gov.bc.ca/pub/mss/stationdata.do?station=2a02 Historic Snow Survey Data for British Columbia. (n.d. b). Water Stewardship Division. Retrieved November 20, 2011, from http://a100.gov.bc.ca/pub/mss/stationdata.do?station=2a14 Jamieson, B. (2004). Between a Slab and a Hard Layer. Avalanche News, 70. Jenkins, M. J., & Hebertson, E. G. (2004). Using Vegetative Analysis to Determine the Extent and Frequency of Avalanches in Little Cottonwood Canyon, Utah. Department of Forest Resources,. Logan, Utah: Utah State University. Johnson, A. L., & Smith, D. J. (2010). Geomorphology of Snow Avalanche Impact Landforms in the Johnston, K. S., & Jamieson, B. (2011). Estimating Extreme Snow Avalanche Runout for the Columbia Mountains, British Columbia, Canada. Department of Geoscience. Calgary, Alberta: University of Calgary. Southern Canadian Cordillera. The Canadian Geographer, 54(1), 87-103. Metadata Canada's National Climate Archive (n.d.). Home Canada's National Climate Archive. Retrieved November 30, 2011, from http://climate.weatheroffice.gc.ca/climate_normals/station_metadata_e.html?stnid=1363 Parks Canada (2011 a). Glacier National Park of Canada. History. Retrieved November 23, 2011 from: http://www.pc.gc.ca/eng/pn-np/bc/glacier/natcul/natcul2.aspx Parks Canada (2011 b). Glacier National Park. The Columbia Mountains Natural Region. Retrieved Novemeber 23, 2011 from: http://www.pc.gc.ca/eng/pn-np/bc/glacier/natcul/natcul1.aspx Parks Canada (2011 c). Glacier National Park. Avalanche Terrain Ratings. Retrieved November 23, 2011 from: http://www.pc.gc.ca/eng/pn-np/bc/glacier/visit/a7.aspx Parks Canada (2011 d). Glacier National Park. Balu Pass Trail. Retrieved November 25, 2011 from: http://www.pc.gc.ca/eng/pn-np/bc/glacier/activ/activ2/activ6/i.aspx

29 Schweingruber, F. H. (1988). Tree Rings: Basics and Application of Dendrochronology. Dordrecht / Boston / London: Kluwer Academic Publishers. Stoffel, M., Bollschweiler, M., Butler, D. R., & Luckman, B. H. (2010). Tree Rings and Natural Hazards. Springer Science and Business Media. Temper, B. (2010). Staying Alive In Avalanche Terrain. Seattle: The Mountaineers Books