snow avalanche order form - british columbiathis handbook addresses snow and avalanche phenomena in...

PLEASE SUBMIT THIS FORM TO:

Government Publication ServicesPO Box 9452 Stn Prov GovtVictoria, BC V8W 9V7Telephone: 250 387-6409 or 1 800 663-6105Fax: 250 387-1120Email: [email protected]

Queen’s Printer # 7610002926 or LMH55

Pricing $28.95ea.

(Total: $37.40) includes Shipping and Applicable Taxes

Contact Government Publication Services forvolume discount pricing when ordering 15 or morecopies

Snow Avalanches in recently harvested areas can damage new forest plantations, destroydownslope resources, and endanger public safety. Snow avalanches can be triggered by forestworkers or winter recreationists in steep cutblocks. This Land Management Handbook reviewssnow avalanche science as it applies to forestry. It presents risk assessment methods for usein forestry planning, outlines harvest designs, and describes silviculture strategies to reduce therisk of snow avalanches. Approaches for managing avalanche risks to forest workers are alsosummarized. With an extensive bibliography and list of Internet resources, this publication willbe a valuable reference for natural resource managers, recreation planners, land developers,and anyone who manages work sites in winter.Company Name:

Company Division or Branch:

Recipient’s Name: Position/Title:

Mailing Address:

City: Province: Postal Code:

Phone Number:

( )

Fax Number:

( )

E-mail Address:

Payment Method:

Certified or Business Cheque Enclosed Money Order

Visa and/or MasterCard # _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Expiry _ _ / _ _

Please enclose this form along with your certified/business cheque or moneyorder made payable to the “Minister of Finance”.

Check out our website at www.publications.gov.bc.ca

Snow AvalancheOrder Form

Queen’s Printer

Government Publication Services

mailto:[email protected]

SnowAvalanche

MANAGEMENT IN FORESTED TERRAIN

LANDMANAGEMENT

HANDBOOK5 5

Snow A

valancheM

AN

AG

EM

EN

T IN

FO

RE

ST

ED

TE

RR

AIN

Ministry of ForestsForest Science Program

Printed in British Columbia on recycled (10% post-consumer) paper

Snow avalanches inrecently harvested areascan damage new forestplantations, destroydownslope resources, andendanger public safety. This Land ManagementHandbook reviews snowavalanche science as it applies to forestry. Itpresents risk assessmentmethods for use in forestryplanning, outlines harvestdesigns, and describes

silviculture strategies to reduce the risk of snowavalanches. Approaches for managing avalanche risks toforest workers are also summarized. With an extensivebibliography and list of Internet resources, thispublication will be a valuable reference for naturalresource managers, recreation planners, land developers,and anyone who manages work sites in winter.

B.C. Government Publication Services, Publisher

www.publications.gov.bc.ca/

Please visit our web site for information on acquiringcopies of this Land Management Handbook, and otherpublications published by the Government of British Columbia.

9 780772 648815

ISBN 0-7726-4881-6

Snow Avalanche

MANAGEMENT IN FORESTED TERRAIN

Ministry of ForestsForest Science Program

Peter Weir

ii

National Library of Canada Cataloguing in Publication DataWeir, Peter.

Snow avalanche management in forested terrain

(Land management handbook ; 55)

Includes bibliographical references: p.

ISBN 0-7726-4881-6

1. Avalanches - Control. 2. Avalanches - British Columbia -

Control. 3. Forests and forestry - Environmental aspects - British

Columbia. 4. Forest management - British Columbia. I. British

Columbia. Ministry of Forests. Research Branch. II. Series.

TA714.W44 2002 551.57’848 C2002-960254-8

The use of trade, firm, or corporation names in this publication is for theinformation and convenience of the reader. Such use does not constitutean official endorsement or approval by the Government of British Colum-bia of any product or service to the exclusion of any others that may alsobe suitable. Contents of this report are presented as information only.Funding assistance does not imply endorsement of any statements or in-formation contained herein by the Government of British Columbia.

Citation

Weir, Peter. 2002. Snow avalanche management in forested terrain.

Res. Br., B.C. Min.For., Victoria, B.C. Land Manage. Handb. No. 55.

www.for.gov.bc.ca/hfd/pubs/Docs/Lmh/Lmh55.htm

For more information on Forest Science Program publications, visit our

web site at www.for.gov.bc.ca/hfd/pubs/index.htm

© 2002 Province of British Columbia

When using information from this or any Forest Science Program report,

please cite fully and correctly.

ABSTRACT

Snow avalanches are a common phenomenon in most mountain rangesof British Columbia and forest damage is a natural occurrence. Forestharvesting on steep slopes in areas of high snow supply can create newavalanche start zones. Snow avalanches starting in recently harvestedareas can damage new plantations, destroy downslope forest resources,and lead to soil loss and site degradation. Snow avalanches can be trig-gered by forest workers or winter recreationists in steep cutblocks; that is,in areas that were not prone to avalanching prior to harvest.

This handbook addresses snow and avalanche phenomena in a forestrysetting and presents a risk assessment procedure suitable for incorpora-tion in the terrain stability field assessment process. The handbookoutlines harvest design and silvicultural strategies to reduce the risk ofavalanche damage resulting from forest harvesting. Strategies for manag-ing avalanche risks in winter are presented. An extensive bibliography isincluded, along with links to relevant publications, data sources, and re-sources available on the internet.

iii

PREFACE

Recent Forest Renewal BC-funded research undertaken at the Universityof British Columbia shows that there are approximately 10 000 clearcutforest blocks in British Columbia that have been significantly affected byavalanching. The majority of these cutblocks have been affected by natur-al snow avalanches running in from above, but 10% of the cutblockssurveyed were found to have generated avalanches that had damaged newplantations within the blocks or forest resources downslope. Clearcut for-est harvesting on steep terrain, particularly in areas of high snow supply(in coastal British Columbia and Vancouver Island mountains where themean annual maximum snow accumulation exceeds 1000 mm waterequivalent and in the interior mountain ranges where mean annual max-imum snow accumulation exceeds 700 mm water equivalent) can createnew avalanche start zones.

On January 20, 2000, a group of 26 forest and heliski industry representa-tives, provincial government officials, and consultants with expertise insnow avalanches, forestry, and terrain mapping convened in Revelstoke,B.C. to identify issues to be addressed in a Land Management Handbookproposed by the British Columbia Ministry of Forests. The resulting drafthandbook, Forest Management of Snow-Avalanche-Prone Terrain, was cir-culated to interested parties and then debated at an open workshop attended by 36 people in Revelstoke on March 16, 2000. The documentwas amended following receipt of written submissions.

Key issues identified and discussed at the two meetings included:(1) The need to establish whether snow avalanches represent a signifi-

cant threat to downslope forest resources or restocked cutblocks.While the avalanche phenomenon is not regarded as a province-wideissue for the forestry sector, the hazard is considered significant onsteep terrain in areas of high snow supply and in regions where theclimate is conducive to development of weak layers in the snow-pack. Also pertinent to this discussion was the need to define whatdegree of resource loss attributable to snow avalanches is tolerable.

(2) The need to provide guidance for the scheduling of winter opera-tions to avoid unacceptable risks and to establish appropriateavalanche safety programs where winter harvesting may exposeworkers to avalanche hazards. It was noted that the prescription of“winter cable harvest,” a recommendation commonly advanced tominimize soil or stream disturbance, could markedly increase theavalanche hazard faced by workers.

v

Also recognized was the need to tailor the nature and scope ofavalanche safety and control programs to the avalanche environ-ment. Current industry programs and response to the avalanchehazard are quite varied.

(3) The need to assign responsibility for public safety should forest har-vesting expose downslope residents, facilities, utilities, or transportcorridors to significantly increased involuntary risk. A subsidiaryissue discussed was the potential for forest harvesting to elevate theavalanche risk faced by winter recreationists or commercial opera-tors travelling on forest roads passing through steep cutblocks thatmay be subject to snow avalanches.

This handbook, representing the culmination of all the workshops, dis-cussions, and expert input, is primarily intended for foresters, foresttechnicians, engineers and engineering technicians, geoscientists, harvestsupervisors, and others responsible for managing forests and risk inavalanche-prone terrain. It can be used in identifying avalanche hazardsand in planning silvicultural operations so that cutblock configurationand harvesting systems minimize the avalanche hazard. It also identifiesknowledge gaps, setting out areas for future research and method devel-opment.

Highlighted in the handbook are issues to be considered by managersand avalanche technicians who deal with both worker and public safety.As well, resource protection issues are stressed for the benefit ofavalanche technicians who may be called on to assess snow stability and,where avalanche danger exists, to perhaps mitigate the danger by artifi-cially triggering unstable snow with explosives. Forest operations withoutin-house avalanche management capability will find guidance as to whenspecialists should be called in to undertake snow stability or avalanche assessments. Overall, the handbook emphasizes practical aspects of thesubject matter, summarizes the present state of knowledge, and identifiesthe best industry practice. Some operators working in avalanche-proneterrain in British Columbia have developed avalanche risk managementprograms that demonstrate it is possible to proactively “manage withresidual risk.” This involves a subtle, but important, difference frommore conservative risk avoidance strategies.

There are no known controlled trials of cutblock design in British Columbia that clearly point to optimum strategies to minimize theavalanche hazard. Avalanching is seldom general or widespread, but it can occur repeatedly in some cutblocks. Examples of cutblocks that

vi

have produced or endured destructive avalanches are presented in thehandbook to illustrate avalanche-prone configurations. Observations ofadjacent cutblocks harvested in areas of high snow supply indicate subtletopographic features, the presence of road fillslopes, and surface rough-ness elements that may alter avalanche susceptibility.

The heliski industry, in partnership with forest companies, has demon-strated that recreational and commercial opportunities can be created byforest harvesting that avoids using avalanche-prone cutblock configura-tions. At the same time, it is known that commercial heliski, snowcat,and wilderness skiing opportunities can be adversely affected by the cre-ation of large clearcuts above prime recreational terrain.

In 1996, the B.C. Ministry of Forests commissioned a review of the inter-national snow and avalanche literature related to forestry practices insnow avalanche terrain. Key findings and case studies from this revieware cited here. In North America, research into avalanche issues related tooperational forestry is a very young science. Forest Renewal BC funded a4-year research program at the University of British Columbia to investi-gate snow avalanche activity in the province’s forests and develop adecision support system for forested and harvested terrain. Research out-comes, expected by 2003, will update the guidance in this handbook.Given the large avalanche observation databases that exist in British Columbia, it is anticipated that avalanche assessment procedures willevolve to become the most quantitative of all geotechnical assessmentsundertaken in the forest sector.

Where Forest Practices Code Regulations and Occupational Health andSafety Regulations are mentioned, they are current in British Columbiaas of March 2001 unless otherwise noted. The names of the province’sministries and agencies are also those as of March 2001.

Each photographer identified retains the copyright of the photographs inthis text. See Photo Credit section starting on page 181.

vii

ACKNOWLEDGEMENTS

The encouragement and support of the following people and organiza-tions was greatly appreciated:• Dr. D. McClung (Chair, Avalanche Research Group, University of

British Columbia)• S. Chatwin, P. Geo., Forest Practices Board, who originated this project.• Contributors and reviewers of the preliminary draft, including:

B. Allen; F. Baumann, P. Eng.; J. Bay; M. Deschenes, E.I.T.; A. Freeland; Dr. B. Jamieson, P. Eng.; Dr. P. Jordan, P. Geo.; D. Kelly,L.L.P. Geo.; P. Kimmel; H. Krawczyk; K. Marr; B. Runciman, R.P. Bio.and S. Walker

• L. Redfern, R.P.F. and Crestbrook Forest Industries • Staff of the B.C. Ministry of Transportation and Highways Snow

Avalanche Program, Victoria• D. Rowe, Workers’ Compensation Board (), Prevention Division,

Northern Region and other representatives• Staff of the B.C. Ministry of Forests, Nelson Forest Region and Colum-

bia Forest District• Participants attending the Revelstoke project initiation meeting and

subsequent workshop• The Canadian Avalanche Association technical committee and the

Division of Engineers and Geoscientists in the Forestry Sector guide-lines committee members who reviewed the final draft of thishandbook.

ABOUT THE AUTHOR

Peter Weir, P. Geo., M. IPENZ, is a geoscientist and avalanche specialistwith over 20 years experience in the transportation and forestry sectors.He has worked in highway avalanche forecasting and control programs inCanada and New Zealand, and currently is a partner in a consulting com-pany, Geoscience Solutions Group, which specializes in services to theforest industry. In 2001–02, he was chair of the Division of Engineers andGeoscientists in the Forest Sector of the Association of Professional Engi-neers and Geoscientists of British Columbia ().

viii

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii

About the Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii

1 Management of Snow Avalanche–prone Forest Terrain . . . . . . . . . . . 11.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Avalanche Risk Management in British Columbia Forests . . . . . 3

Avalanche Risk in Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2 Snow and Avalanche Phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.1 Seasonal Snow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Snow Accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Energy Exchanges over Snow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2.2 Avalanche Phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Snow Avalanche Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Slab Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Snow Avalanche Path Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . 14Creep and Glide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17Avalanche Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18Snow Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19Avalanche Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21Avalanche Probability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

2.3 Avalanche Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28Avalanche Speed and Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

2.4 Identification of Snow Avalanche Terrain . . . . . . . . . . . . . . . . . . . 312.5 Runout Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332.6 Role of Forest in Avalanche Protection . . . . . . . . . . . . . . . . . . . . . . 342.7 Hydrological Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352.8 Ecological Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Grizzly Bear Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38Habitat Protection Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

3 Avalanche Risk Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453.1 Protection Forests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

Avalanches in British Columbia Forests . . . . . . . . . . . . . . . . . . . . . . 473.2 Avalanche Risk Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

Risk Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48Assessing Risk to Forest Cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50Assessing Risk above Transportation Corridors,

Facilities, or Essential Resources . . . . . . . . . . . . . . . . . . . . . . . . 51Consequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52Risk Management Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53Responsibility for Risk Management . . . . . . . . . . . . . . . . . . . . . . . . 55

3.3 Ownership of Risk in Forest Operations . . . . . . . . . . . . . . . . . . . . . 56

ix

CONTENTS

3.4 Logging above Highways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573.5 Avalanche Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

Air Photo Interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63Field Checking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64Avalanche Mapping for Land Use Planning . . . . . . . . . . . . . . . . . . . 66Map Use and Interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

3.6 Documenting Existing Avalanche Paths . . . . . . . . . . . . . . . . . . . . 753.7 Runout Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

Dynamics Modelling Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75Probabilistic Modelling Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . 77Statistical Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

3.8 Impact Pressures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

4 Mitigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 854.1 Harvest Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

Slope Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 864.2 Silvicultural Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

Increasing Surface Roughness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91Modification of Local Scale Climate . . . . . . . . . . . . . . . . . . . . . . . . . 93Retention of Timbered Margins Adjacent to Avalanche Runout . . 94Species Vulnerability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

4.3 Effects of Harvesting on Heliski, Snowcat, and Wilderness Skiing Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

4.4 Stream Crossing Location and Road Layout . . . . . . . . . . . . . . . . . 95

5 Managing Avalanche Risks in Winter . . . . . . . . . . . . . . . . . . . . . . . . . . 1015.1 Field Observations and Recording Systems . . . . . . . . . . . . . . . . . . 1015.2 Class I Data: Avalanche Observations . . . . . . . . . . . . . . . . . . . . . . . 1025.3 Class II Data: Snowpack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Snow Profile Interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106Numerical Avalanche Forecasting . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

5.4 Class III Data: Weather Observations . . . . . . . . . . . . . . . . . . . . . . . 108Permanent Weather Stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

5.5 Mountain Weather Forecasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1125.6 Avalanche Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

Passive and Active Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114Consider the Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115When, and When Not, to Trigger Avalanches . . . . . . . . . . . . . . . . . 115Consider the Runout, Consider the Consequences . . . . . . . . . . . . . 117Helicopter Bombing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

5.7 Winter Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119Harvesting Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119Snowmobiles in Forest Operations . . . . . . . . . . . . . . . . . . . . . . . . . . 120Removal of Avalanche Debris on Roads . . . . . . . . . . . . . . . . . . . . . . 121

5.8 Avalanche Rescue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122Rescue Transceivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122Alternative Rescue Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126Rescue Caches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126

x

Rescue Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1275.9 Information Exchanges and Avalanche Bulletins . . . . . . . . . . . . . 127

6 Case Studies of Risk Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1296.1 Wardance Slope at Big Sky, Montana . . . . . . . . . . . . . . . . . . . . . . . 1296.2 Protection Role of Forests above the Coquihalla

Highway, B.C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1306.3 Avalanche Triggered in Cutblock Kills

Recreationist near Salmo, B.C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1316.4 Ranch Ridge, near Hills, B.C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1316.5 Airy Creek in the Slocan Valley, B.C. . . . . . . . . . . . . . . . . . . . . . . . 1326.6 Nagle Creek Cutblocks, near Mica Creek, B.C. . . . . . . . . . . . . . . . 1336.7 Avalanche in a Block with Active Logging,

Southern Interior, B.C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135Day One – Accidental Triggering . . . . . . . . . . . . . . . . . . . . . . . . . . . 135Day Two – Equipment Damaged during Avalanche Control . . . . . 137Incident Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139Development of Safe Work Procedures for Snow Stability Assessment in Active Harvest Areas . . . . . . . . . . . . . . . . . . . . . . . . . 139

1 Occupational health and safety regulations . . . . . . . . . . . . . . . . . . . . . . . 1432 Safe forest operations in avalanche-prone terrain . . . . . . . . . . . . . . . . . 1453 Online resources and information sources . . . . . . . . . . . . . . . . . . . . . . . 1544 Avalanche site identification form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1585 Avalanche assessment checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1606 Snow stability rating system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

Photo Credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188

xi

xii

1 Influence of forest cover on the climate parameters that affect snow stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2 Snow avalanche types described according to release mechanism . . . 123 Conversion of units used in measuring slope angle . . . . . . . . . . . . . . . 164 Classification of snow avalanche size . . . . . . . . . . . . . . . . . . . . . . . . . . 195 Terminology used to describe frequency of mountain slope hazards . . . 236 Encounter probability for various return period avalanches

in given periods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Vegetative indicators of snow avalanche frequency . . . . . . . . . . . . . . . 268 Ranking of avalanche start zone wind exposure . . . . . . . . . . . . . . . . . 329 Differences in snow accumulation and stability between

forest openings and closed canopy coniferous forest . . . . . . . . . . . . . 3510 a) Species whose habit includes snow avalanche–prone terrain by

British Columbia biogeoclimatic zone . . . . . . . . . . . . . . . . . . . . . . . . . 39b) Key to biogeoclimatic units referenced in Table 10a . . . . . . . . . . . . 42

11 Estimate of avalanche likelihood from site-specific observations and analysis of climate data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

12 Risk ratings for expected avalanche size and expected avalanche frequency for forest harvest resulting in damage to forest cover . . . . 51

13 Risk ratings for expected avalanche size and frequency for forest harvest when downslope transportation corridors, facilities, or essential resources may be affected . . . . . . . . . . . . . . . . . . 52

14 Examples of risk management strategies . . . . . . . . . . . . . . . . . . . . . . . . 5315 Appropriate risk responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5416 Mapping methods that may be applicable for use in

snow avalanche–prone terrain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6017 Fraser Valley Regional District snow avalanche planning response . . 7218 Avalanche runout summary statistics: mean values . . . . . . . . . . . . . . 7919 Avalanche runout summary statistics: extremes . . . . . . . . . . . . . . . . . 7920 Relation between impact pressures and potential damage . . . . . . . . . 8221 Scale effects on snow accumulation . . . . . . . . . . . . . . . . . . . . . . . . . . . 8622 Basic snow grain classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10523 Snowpack hardness test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10524 Typical observations of weather, snow, and avalanche data . . . . . . . . 11025 First avalanche start zone: geometry and dimensions . . . . . . . . . . . . 13726 Second avalanche start zone: geometry and dimensions . . . . . . . . . . . 139

xiii

1 Significant forest damage caused by a snow avalanche at Ningunsaw Pass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 Snow avalanches that run into cutblocks or onto roadson Duffey Lake Road near Lillooet . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

3 Resource loss in standing forest near Nagle Creek . . . . . . . . . . . . . . . . 64 Some forms of new snow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Coast Range snowpacks accumulate as a series of snow

wedges truncated by melt at low elevations . . . . . . . . . . . . . . . . . . . . . 96 Typical layering in a winter snowpack . . . . . . . . . . . . . . . . . . . . . . . . . 97 Mountain barriers in southern British Columbia impinging

on the prevailing synoptic-scale westerly circulation . . . . . . . . . . . . . 98 Surface hoar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Loose snow avalanches start small but grow in mass

as they move downslope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1210 Failure surfaces in a slab avalanche . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1311 Slab avalanches exhibit distinct fracture lines . . . . . . . . . . . . . . . . . . . 1312 Slab avalanche nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1413 Steep road fillslopes may act as preferred crown locations . . . . . . . . . 1414 Start zone, track, and runout zone of a confined

snow avalanche path at McLean Point . . . . . . . . . . . . . . . . . . . . . . . . . 1515 Unconfined avalanche path on an open slope in the alpine zone . . . 1516 Slab avalanche dependence on start zone slope declination . . . . . . . . 1517 New snow avalanche paths cut through mature forest below

a steep clearcut near Nagle Creek . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1618 Snow avalanche runout destroying 20 ha of old-growth forest . . . . . 1619 Wildfire on steep slopes has created mid-slope

openings capable of generating large snow avalanches . . . . . . . . . . . . 1620 Snow creep and glide deformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1721 Snow glide avalanche on a smooth grassy surface at Galena Pass . . . 1822 Forty years of ranked snow water equivalent values

from Glacier National Park . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2023 Expected snow water equivalents for various return periods . . . . . . . 2024 Relationship of mean annual maximum snow accumulation with

elevation for three sub-regions of the Columbia Mountains . . . . . . . 2125 Snow avalanche areas affecting British Columbia highways . . . . . . . . 2226 Recent snow avalanche damage in mature forest . . . . . . . . . . . . . . . . . 2227 Vegetative trim lines and subtle colour differences indicate

relative frequency of snow avalanche activity . . . . . . . . . . . . . . . . . . . . 2228 An indication of snow avalanche frequency from tree ring pattern . . . . 2329 Snow avalanches down the centre path at Fish Lake . . . . . . . . . . . . . . 2330 Development in the runout zone has eliminated clues that might

otherwise be used to assess snow avalanche frequency . . . . . . . . . . . . 23

31 Probability of not experiencing one avalanche of a 5-, 10-, or 30-yearreturn period as a function of time . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

32 Vegetative indicator: J-shaped tree trunk indicates probable snowavalanche impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

33 Rubble lodged in tree trunk indicates moderate impact pressure . . . 2634 Flagged tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2735 Tree trunk snapped by a snow avalanche at Greenslides . . . . . . . . . . . 2736 Impact scars from entrained material . . . . . . . . . . . . . . . . . . . . . . . . . . 2737 Rubble perched on logging slash deposited by snow avalanche . . . . . 2738 Trim lines and debris in valley bottom, Serpentine Creek . . . . . . . . . 2739 Rubble perched in alder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2740 Large area of mature timber destroyed by natural avalanching . . . . . 2841 Snow avalanches carrying detritus downslope continue

to build a fan in the runout zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2842 Steep conical fan built by successive snow avalanches

carrying colluvium down from an alpine start zone . . . . . . . . . . . . . . 2843 An avalanche tarn is an impact feature formed in the

runout zone of a large snow avalanche path . . . . . . . . . . . . . . . . . . . . . 2844 Speed in a large dry avalanche shows rapid acceleration . . . . . . . . . . . 2945 Speed and density profiles for a large flowing avalanche . . . . . . . . . . . 2946 Large, mixed-motion avalanche with an airborne component . . . . . 2947 Large blocks of debris remained intact in a hard slab avalanche . . . . 2948 Snow plastered on this tree trunk indicates flow height of core of

avalanche . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3049 Maximum avalanche speed versus square root of slope distance . . . . 3050 Sequence of a powder avalanche . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3151 Large wet avalanche flowing from a confined gully out onto a fan . . . . . 3252 Wind exposure class 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3253 Avalanche run-up is likely to occur opposite avalanche paths

that descend into narrow valleys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3454 Trees lying upslope on the opposite side of the valley from

the avalanche path indicate the destructive forces . . . . . . . . . . . . . . . . 3455 An avalanche that started in a hard dry slab, but melt caused the mass

to flow in the runout zone in the mode of a wet snow avalanche . . . 3456 A large wet avalanche that released in spring . . . . . . . . . . . . . . . . . . . . 3457 Significant forest damage caused by a snow avalanche at

Ruby Mountain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3658 Conceptual stream flow model for a subalpine stream . . . . . . . . . . . . 3659 Stream undercuts debris and melts snow . . . . . . . . . . . . . . . . . . . . . . . 3760 Rapid breach of a snow avalanche dam caused Mobbs Creek, near

Trout Lake in the Kootenays, to damage a trout spawning channel . . . . 3761 Barge torn from its moorings by a snow avalanche from Mt. Rainey . . . 3762 Snow avalanche paths often contain a complex mosaic of plant

communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

xiv

63 Avalanche paths dominated by alder, willow, and herbaceous sites often contain valuable spring bear foods . . . . . . . . . . . . . . . . . . . 38

64 Willow, groundsel, and a range of herbaceous species often occur in runout zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

65 Snow avalanche path dominated by closed-canopy conifer shrub . . . 4366 Village protected by forest reserved from logging in 1937 . . . . . . . . . . 4567 Avalanche start zone defence structures . . . . . . . . . . . . . . . . . . . . . . . . 4668 Snow support structures built in an opening above a subdivision . . . 4669 Dense forest that was probably destroyed by fire when the

railway was built in 1885 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4770 Projections from growth model . . . . . . . . . . . . . . . . . . . . . . . . . 4871 Small avalanches may run regularly in long narrow gullies . . . . . . . . 4872 Conceptual model of avalanche susceptibility in a regenerating

opening in steep terrain with regular, high snow supply . . . . . . . . . . . 4973 Damage in a regenerating cutblock from a small avalanche . . . . . . . . 4974 Avalanche consequences for traffic will be very high

because of the water body below . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5375 framework for risk assessment and reduction . . . . . . . . . . . . 5476 Steps in the risk management decision-making process . . . . . . . . . . . 5577 Process of continuous risk assessment . . . . . . . . . . . . . . . . . . . . . . . . . 5578 Typical ratios of non-injury industrial incidents to

injurious accidents and to fatalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5679 Managing the avalanche risk at a logging operation at

Doctor Creek . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5780 Winter operations in potential avalanche terrain on private

land at Enterprise Creek . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5881 Steep terrain in a recently harvested cutblock on private land . . . . . . 5882 Avalanches initiating in the steep harvested terrain above

the Trans-Canada Highway in the Kicking Horse Canyon . . . . . . . . . 5883 Harvesting was proposed in an even-aged pine forest where

red trees indicate insect attack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5984 Method used by MoF and MoT to identify proposed harvest

blocks that may generate snow avalanches . . . . . . . . . . . . . . . . . . . . . . 5985 Stereopair of Nagle Creek avalanche path . . . . . . . . . . . . . . . . . . . . . . 6386 Slope profile of the Nagle Creek avalanche path plotted

from 20-m contours . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6587 Avalanche mapping for Val d’Isère, France . . . . . . . . . . . . . . . . . . . . . 6888 Risk map produced by simulation of avalanche runout . . . . . . . . . . . 6989 Ministry of Transportation avalanche mapping showing

distribution of avalanche paths above highway at Ningunsaw Pass. . . . . 7090 Terrain mapping for an area in the Interior of British Columbia

that has both major and minor avalanche paths . . . . . . . . . . . . . . . . . 7391 A landslide initiating below an old road created an opening . . . . . . . 7592 Typical page from an avalanche atlas . . . . . . . . . . . . . . . . . . . . . . . . . . 7693 Deterministic application of the avalanche dynamics model . . . . 77

xv

94 Terrain parameters used in runout calculation . . . . . . . . . . . . . . . . . . 7895 a) Radar speed measurements of the frontal pulse of a

large avalanche . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83b) Impact pressures measured at two heights on a tower located in the path shown in Figure 95a . . . . . . . . . . . . . . . . . . . . . . . . 83

96 Typical engineered structures used in Europe to prevent avalanche initiation or arrest any small events . . . . . . . . . . . . . . . . . . . 85

97 The Total Chance Harvest Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8798 The length of fetch for wind-transported snow is greatly

reduced in small openings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8899 Group selection using small cable or helicopter harvesting systems

is appropriate in steep terrain in areas of high snow supply . . . . . . . . 88100 Tree patches retained in steeper terrain alter the radiation balance

over the snowpack, intercept snowfall, and alter local wind flow . . . 88101 Silviculture prescription specifies retention of high stumps . . . . . . . . 89102 Polygon-based 1:5000 scale mapping produced during

a Terrain Stability Field Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . 89103 Block layout strategies to reduce avalanche susceptibility . . . . . . . . . . 90104 Logs retained by high stumps used to inhibit snow creep

and glide in a gully . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93105 Avalanche initiated at a convex change in slope . . . . . . . . . . . . . . . . . 93106 Cutblock boundary set too close to gullies where avalanches run . . . 94107 Narrow, vertically oriented cutblocks set out for heliskiing in the

Bugaboos are designed to minimize avalanche susceptibility . . . . . . . 95108 A poor bridge location directly in line with the runout of a

large snow avalanche path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96109 Bridge span destroyed by a large snow avalanche . . . . . . . . . . . . . . . . 96110 Alternative bridge design for a high-frequency avalanche

path containing a stream that has permanent flow . . . . . . . . . . . . . . . 97111 Rock ford used to cross the track of a large avalanche path . . . . . . . . 97112 Forest roads that cross high in an avalanche track will be

affected more often . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98113 Use of this switchback section of road in winter prolongs

exposure to avalanches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98114 Avalanche path on the public road to Shames ski area, Terrace . . . . . 99115 Oversteepened road fillslopes that dissect large clearcuts in

steep terrain are common points for avalanche initiation . . . . . . . . . 99116 Causal chain considered in avalanche prediction . . . . . . . . . . . . . . . . 102117 Sample page from field notebook of basic avalanche observations . . . . . 103118 Snow profile studies conducted at fracture lines are used

to identify failure layers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103119 Shovel shear test is used to locate, but not rate, a weak layer . . . . . . . 104120 The Rutschblock test requires a slope of 25° (47%) or more . . . . . . . 104121 Snowpack observations conducted at safe sites at the

elevation of avalanche start zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

xvi

122 Standard field notebook set out for a full snow profile observation . . . . 106123 Plot of a snow profile produced by application . . . . . . . . . . . . . . . 107124 Screen from a knowledge-based expert system designed to

assist with snow profile interpretation . . . . . . . . . . . . . . . . . . . . . . . . . 107125 A simple, temporary weather observation plot . . . . . . . . . . . . . . . . . . 109126 Mountaintop remote weather station used for avalanche forecasting . . . 111127 Lower-elevation weather station adjacent to a road in

avalanche terrain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112128 Time series plot of weather data used in avalanche prediction . . . . . . 112129 Spatial and temporal scales of various weather

forecasting techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113130 Predictive skill associated with meso-scale forecasting technique . . . 114131 Explosive charges are delivered by helicopter to start zones to

trigger avalanches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118132 Local experience helps in the identification of preferred targets . . . . 118133 An operator sheltering behind bulldozer was crushed when an

avalanche overturned the machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119134 Loaded logging truck hit by a Size 3 avalanche . . . . . . . . . . . . . . . . . . . 119135 Avalanche debris removal in Galena Pass near Trout Lake . . . . . . . . . 121136 Typical rescue transceivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122137 Searchers probe a Size 5 avalanche for a victim who was

not wearing a transceiver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123138 Large, blocky avalanche deposit 500 m downslope of the

spring snowline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125139 Searching with a detector device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126140 Avalanche debris removal in Bear Pass near Stewart . . . . . . . . . . . . . . 126141 A slope logged to create a ski run has potential to generate

avalanches that might run out into high-value buildings below . . . . 129142 Supporting structures in start zone designed to inhibit avalanches . . . . 130143 Ten-year-old trees offer little protection in a

1.25 m deep snowpack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130144 Two wet snow avalanches originated in adjacent clearcuts . . . . . . . . . 131145 The first of two snow avalanches originating in a clearcut

stopped just before Highway 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132146 Mixed deposit of snow, trees, and colluvium . . . . . . . . . . . . . . . . . . . . 132147 Airy Creek avalanche originated in a clearcut and ran

into a community watershed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133148 The block in the foreground avalanched but the two adjacent

clearcuts, which have similar topography, were not affected . . . . . . . 133149 Fracture occurred at upper convex slope change and below

the road . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133150 Timber damage and soil erosion represent substantial

resource losses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134151 Damage to immature trees in the start zone . . . . . . . . . . . . . . . . . . . . . 134

xvii

152 image of the area surrounding Mica Dam showing forest damaged by avalanches at Nagle Creek . . . . . . . . . . . . . . . . . . . 134

153 A natural avalanche occurred in the cutblock . . . . . . . . . . . . . . . . . . . 135154 New path resulting from slab avalanche triggered by explosives in

clearcut above . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

xviii

Chapter 1 Management of snow avalanche-prone forest terrain 1

Management of snow 1avalanche-proneforest terrain

1.1 BACKGROUND

When the U.S. Forest Service ap-pointed snow rangers at LittleCottonwood Canyon, Alta, Utah,in the 1950s, the stage was set formanagement of snow avalanchehazards in forests of North Ameri-ca. The snow rangers’ prime rolewas promoting safety for land useplanning and the winter recre-ational use of public lands. Monty Atwater and Ed LaChapelle (who laterbecame Professor of Atmospheric Sciences at the University of Washing-ton in Seattle) developed the science of snow avalanche forecasting andexplored methods for the control of snow avalanches. Prof. LaChapelle’s1980 analysis of conventional avalanche forecasting remains current 20years later and sets a foundation for innovations still to be realized. In thelate 1960s, the U.S. Forest Service moved its research effort to FortCollins, Colorado, where meteorologist Pete Martinelli, Jr. assembled ateam of snow and avalanche researchers. In 1972, Martinelli and Dr. RonPerla published the first definitive avalanche handbook for forest landmanagement in North America, the blue-covered U.S. Department ofAgriculture Avalanche Handbook.

Beginning in the 1950s, the National Research Council of Canada ()and Parks Canada became involved in snow avalanche work at RogersPass, B.C., when the Trans-Canada Highway was constructed throughthe Selkirk Mountains. Parks Canada developed the first snow avalancheforecasting and control program in Canada and laid the foundation forthe weather, snowpack, and avalanche observations standards in use today.

A Swiss-trained engineer, Peter Schaerer, P. Eng., was involved in the design of snow avalanche defence structures along the Trans-CanadaHighway and subsequently developed the ’s snow avalanche researchprogram at Rogers Pass. Peter Schaerer went on to participate in the B.C.

Ministry of Transportation Snow Avalanche Task Force set up in response to a snow avalanche that killed seven people in a roadside cafénear Terrace in 1974. Throughout the 1970s and 1980s, Peter Schaererpublished more than 90 papers and chapters documenting research intomany aspects of the snow avalanche phenomenon. In 1981, he became thefounding president of the Canadian Avalanche Association (), an or-ganization that now has more than 300 members.

A graduate of Professor LaChapelle’s program at the University of Wash-ington, Dr. David McClung worked on snow avalanche dynamics researchat Rogers Pass and avalanche runout studies while Dr. Perla set up a coldlaboratory in Canmore, Alberta, to study snow metamorphism. In 1993,McClung and Schaerer published a completely revised edition of theAvalanche Handbook. This comprehensive (and now red-covered) hand-book is regarded as the most definitive text in the subject area.

Today Dr. McClung, P. Eng/P. Geo., is chair of Snow and Avalanche Re-search in the Geography Department at the University of BritishColumbia and is involved in collaborative research with the British Co-lumbia forest industry. He is internationally regarded as one of theworld’s foremost snow avalanche researchers and many of his papers arecited in this handbook.

Dr. Bruce Jamieson, P. Eng., of the University of Calgary, is a past presi-dent of the and current chair of the Technical Committee. Hehas undertaken extensive field research to develop and refine snow stabil-ity tests for field technicians. His investigations focus on the nature ofpersistent weak layers that often release slab avalanches in the interior ofBritish Columbia.

The majority of the world’s snow and avalanche research is conducted incontinental and intermountain climates where wet snow and rain-on-snoware uncommon in winter. Dr. Howard Conway of the Geophysics De-partment at the University of Washington in Seattle has worked in theCascade Range and Olympic Peninsula to extend the science of wet snowavalanche prediction. His research is applicable to the management ofsnow avalanche risk in British Columbia’s Coast Mountains and CascadeRange.

In recent years, several tertiary educational institutions, including the Geography and Civil Engineering departments at the University of BritishColumbia and the Civil Engineering and Geophysics departments at theUniversity of Calgary, have developed courses in snow and avalanche

2 Snow Avalanche Management in Forested Terrain


science. These efforts have helped ensure there is a pool of knowledgeablegraduates for furthering this work.

In 1998, the Columbia Mountains Institute of Applied Ecology () rana workshop in Revelstoke, B.C. on forestry and avalanches. Issues per-taining to forest harvesting and habitat were discussed.

The , headquartered in Revelstoke, has developed an intensive two-tier, short course, with a training syllabus that focuses on day-to-daysnow stability assessment and snow avalanche forecasting. The andthe British Columbia Institute of Technology originally established thetraining system in the 1970s. One course of instruction is aimed specifi-cally at the transportation and resource industries. Extensive practicalfield experience underpins the formal training. The prerequisite entry cri-teria for the ’s Level 2 course are 100 logged days (i.e., typically twowinters) of relevant experience within an established snow avalancheprogram.

The contracted to the Canadian National Search and Rescue Secre-tariat to produce two documents on avalanche risk and hazard mapping.One is a technical guidebook that is a standard reference for registeredprofessionals involved in avalanche risk assessment in the forest sector. A training course is available for professionals who work with this riskstandard.

1.2 AVALANCHE RISK MANAGEMENT IN BRITISH COLUMBIA FORESTS

From 1995 to 2002, the role of professionals in the forest sector who un-dertook planning, mapping, and terrain assessments was governed by theForest Practices Code of BC Act (1995). Major changes to the Forest Prac-tices Code were being proposed at the time of publication.

The Joint Practice Board () of the Association of B.C. ProfessionalForesters () and the Association of Professional Engineers andGeoscientists of B.C. () have published a skill-set documentdefining appropriate course work, training, and background experiencefor a qualified registered professional undertaking snow avalanche assess-ments (Joint Practice Board 2002). The has issued a LimitedLicence in Geoscience allowing a person so licensed to define snowavalanche hazards, avalanche path boundaries, and the probable limits oftravel of future snow avalanches, to identify terrain that could producesnow avalanches, and to estimate snow avalanche return periods. Theseareas are all considered to be in the realm of Professional Geoscience.

The (1995) gives Ministry of Forests (MoF) District Managers discre-tion in requiring that a Forest Development Plan () “will adequatelymanage and conserve the forest resources.” It requires that licensees notcarry out a forest practice that could result in “inordinate soil disturbance,or other significant damage to the environment.” District Managers haveconsiderable discretion in asking for information to be included in an or in a Silviculture Prescription (), and in requiring that TerrainStability Field Assessments (s) be undertaken for proposed cut-blocks. For example, the Columbia Forest District routinely requests thatavalanche hazard assessments be undertaken in cutblocks proposed forharvest.

Within a cutblock, the licensee is responsible for any damage caused byavalanches or another hazard (e.g., insects, disease, fire) until the planta-tion is declared a “free-growing stand” (as defined in the ). If forestadjacent to a cutblock that has not reached the free-growing stage is dam-aged by any hazard (including avalanches, landslides, fire escape, orwindthrow), it is common to amend the to allow the licensee to sal-vage the damaged timber, at the discretion of the District Manager andwith the agreement of the licensee. In such instances, the licensee is re-sponsible for reforesting the affected area. Salvage harvesting may also berequired. After a cutblock is declared free growing, the Crown is respon-sible for any subsequent damage.

The Occupational Health and Safety Regulations (2001) of the B.C. Work-ers’ Compensation Board (B.C. regulation 296/97, parts 21.85, 26.17, and26.18) relate specifically to snow avalanche control and safety (Appendix 1).Should there be a conflict between the Workers’ Compensation (Occupa-tional Health and Safety) Amendment Act and a Forest Practices Code Actregulation, then the Occupational Health and Safety Act, or the regulationmade under that Act, takes precedence.

Transport Canada has the mandate to regulate the use of aircraft for car-riage and delivery of explosives onto snow slopes.

Avalanche Risk in Context

Snow avalanches are a common natural hazard in most mountain rangesin British Columbia. It is estimated that more than 300 000 large snowavalanches occur in the province each winter, primarily in forested zones(Stitzinger et al. 2000), but only a very minor proportion of those actuallydamage property or injure people. Many snow avalanches occur naturally



in remote areas where there are few people and little developed property.Snow avalanches typically kill about 12 people in Canada each winter, al-though this has increased to about 20 fatalities in recent years. Most ofthese deaths involve backcountry recreationists. Statistics produced bythe indicate that for every fatality caused by an avalanche five morepersons are caught but not killed.

Extensive forest damage is a common natural occur-rence in mountainous regions of British Columbiaand the Pacific Northwest (Figure 1). A survey ofavalanching in the province’s forests indicates thatapproximately 10 000 clearcuts are significantly af-fected by avalanching (McClung 2001a). Most ofthese cutblocks have been affected by natural snowavalanches running in from above (Figure 2); about10% of cutblocks surveyed have generated avalanch-es that have damaged downslope forest resources orinfrastructure. Across the province, the latter casemay represent a very small proportion of all cut-blocks. However, this balance may change in thefuture as more harvesting occurs on steeper, higher-elevation slopes in high-snowfall areas. Damagefrom avalanches initiating in clearcuts in both theinterior and Coast Mountains of British Columbiainvolves loss of both timber and soil resources (Figure 3).

At this stage, it is not possible to rank the effects ofsnow avalanches with debris flows and landslides inBritish Columbia forests, but snow avalanches maydominate in the interior, while debris flows andlandslides may dominate on the coast (P. Jordan,MoF, pers. comm.).

Natural hazards that threaten human life are com-monly rated more highly than damage to economicresources or the environment (Smith 1992). Al-though snow avalanches regularly kill people in theprovince and cause substantial forest damage, theypose a less serious problem than earthquakes, vol-canic eruptions, floods, and severe storms. The firstrecorded snow avalanche fatalities in the forestry

Significant forestdamage caused by a snow avalanche atNingunsaw Pass in northern BritishColumbia.

Snow avalanches thatrun into cutblocks or onto roads arerelatively common in British Columbia,such as these on the Duffey Lake Roadnear Lillooet.

sector in British Columbia involved three forest workers who werecaught on a forest road blocked by avalanches in the Flathead Valley,near Fernie, while travelling home for Christmas in 1971 (Stethem andSchaerer 1979).

Today, greater concern is being raised about the potential for snowavalanches to be triggered by forest workers or winter recreationists insteep cutblocks (i.e., areas that were not prone to avalanching before harvest). This handbook advocates a risk-based approach to the manage-ment of such events on snow avalanche-prone forest terrain (seeAppendix 2).


Resource loss in standing forest near Nagle Creek caused by a snow avalanche that initiated in a mid-slopeclearcut 200 m above the damaged area. Clearcut harvesting on steep slopes at higher elevations in high snowfall areas ofBritish Columbia is becoming more common, so an increased incidence of forest damage can be expected.

Chapter 2 Snow and avalanche phenomena 7

Snow and avalanche 2phenomena

2.1 SEASONAL SNOW

The shape of new snow crystals is largely a function oftemperature and humidity in the atmosphere at the timeof formation (Figure 4). Snow crystals are modified dur-ing their descent through the lower atmosphere and maybecome rimed. In extreme cases, they form “graupel,”where the underlying crystal is no longer recognized.

The form and properties of new snow can change veryrapidly at the surface, particularly when temperaturesare warm (i.e., close to 0°C). If significant wind accom-panies a snowfall, then snow crystals will be broken intofragments and packed to form a dense, hard surfacelayer. Rimed snow and graupel can form layers in asnowpack that have properties quite different from thosefound in other forms of new snow.

The physical properties of snow grains in the snowpack depend princi-pally on temperature and the change in temperature across a layer, aswell as its hardness, failure toughness, and strength (McClung andSchweizer 1999). A layer of snow at 0°C (i.e., snow at its melting point)has markedly different properties than a layer that has a sub-zero temperature.

Some forms of new snow.

a) Stellar snow crystal. b) Clusters of columns with plates. c) Rimmed needles.

Snow metamorphism, the set of microscopic processes controlled largelyby temperature, determines the manner in which snow grains and theirbonds change in shape and strength through time (Colbeck 1997). An understanding of snow metamorphism is required by anyone intendingto undertake day-to-day snow stability evaluations—a prerequisite foridentifying periods of high snow avalanche danger.

There is no universally applicable field test to determine the strength of a snowpack, snow layer, or interface between two adjacent layers, al-though a number of snow stability tests exist. Often several parametersthat correlate with strength must be measured in situ, but care must betaken when extrapolating the results because of the high spatial variabili-ty that characterizes mountain snowpacks. The ’s 1995 ObservationGuidelines and Recording Standards for Weather, Snowpack, andAvalanches presents a range of field tests.

A full discussion of the properties of snow as they relate to avalanche initiation is beyond the scope of this handbook. Readers are referred toMcClung and Schaerer’s 1993 Avalanche Handbook for a comprehensivediscussion of the subject.

Snow Accumulation

The seasonal snowpack accumulates layer by layer through the early- tomid-winter period, with more snow often falling at higher than at lowerelevations. Numerous studies also show that more snow accumulates inopenings than in adjacent forest areas (e.g., Golding 1982; Heatherington1987).

Many maritime storms have highly variable freezing levels and oftenbring precipitation in the form of rain to lower slopes and as snow abovethe freezing level (i.e., the elevation of the 0°C isotherm). In maritimesnow environments, the mountain snowpack accumulates as a series ofstacked wedges (Weir and Owens 1980).

Some storms may have high freezing levels and bring rain up to the high-est mountaintops, leaving crusts within the snowpack at high elevationsand melting back the snowline at lower elevations (Figure 5).

In eastern British Columbia, the freezing level is generally below the baseof the mountains during winter. The mountain snowpack accumulates insuccessive layers (Figure 6), with increased accumulations occurring athigher elevations. Rain crusts are less common and the snowline rarelyrecedes up the mountain.



Maritime storms from the Pacific Ocean produce the highest precipitationin the Coast Range, where mid-elevation sites receive about 3000 mm peryear (Schaefer 1978). The greatest snow accumulations tend to occuraround the winter 0°C isotherm; (i.e., the average elevation of the freezinglevel) (D. McClung, University of BritishColumbia, pers. comm.). Precipitation totals and snow accumulation lessen onmountain chains to the east of the CoastRange (Figure 7).

In the interior of British Columbia, theSelkirk Range features a transitional snowenvironment, intermediate between the

Coast Range snowpacks accumulate as a seriesof snow wedges truncated by melt at low elevations.

Typical layering in a wintersnowpack. The strength of weaker layers andbonding between layers often controls thestability. A very thin layer of surface hoarburied within the snowpack often forms acritical failure layer.

Mountain barriers in southern British Columbia impinging on the prevailing synoptic-scale westerlycirculation: Vancouver Island Mountains; Coast Mountains / Northern Cascade Range; Columbia Mountains group(Monashee, Selkirk, and Purcell); and Rocky Mountains (flanking the Prairies). (Source: Chilton 1981).

Elev

atio

n

127° 126° 125° 124° 123° 122° 121° 120° 119° 118° 117° 116° 115°

3000

2700

2400

2100

1800

1500

Elev

atio

n (m

etre

s)

maritime Coast Range snowpack and the continental snowpack found in the Rocky Mountains (McClung and Schaerer 1993, p. 18).

Between snowfalls, the snow surface may be eroded by wind, subject todeposits of rime ice, and affected by sun (particularly on south-facingslopes; north aspects are affected by sun only in spring). Surface hoarcrystals commonly grow on the snow surface in the interior of BritishColumbia. New snow that falls without wind is often cohesionless, butwithin a few minutes, hours, or days (depending on the environment),bonds may develop between snow grains to form cohesive snow. Cohe-sion is an important determinant of the type of avalanching that mayoccur at or below the snow surface.

Energy Exchanges over Snow

WinterDuring winter, there is often a net loss of energy from the snow surface in open areas (by long-wave radiation). This leads to a cooling of the near-surface layers, particularly on shaded aspects under clear sky conditions.Surface hoar develops when atmospheric water vapour sublimates onto a

cold snow surface (Figure 8). A strong temperaturegradient in the upper layers of the snowpack (i.e.,identified by a notable change in temperature withdepth) controls processes that reduce the snow’sstrength and promote the growth of faceted grains.Energy exchanges are damped under continuouscloud cover or under a forest canopy.

Snow accumulation, snowpack structure, and sta-bility are markedly different under closed-canopyconiferous forest compared to that in openings(Table 1). Wind, temperature, and radiative energyexchanges under deciduous forest (e.g., larch) areintermediate between these two extremes.

SpringIn spring, the snowpack’s energy balance changessignificantly and there is often a net positive inputof solar (short-wave) radiation to the snow’s sur-face. The surface may melt by day, producingliquid water that percolates down through thesnowpack. At night, the surface refreezes when energy is lost by long-wave radiation to the atmos-phere.


a) Predominantly two-dimensionalgrowth. When the snow surface is cooledby long-wave radiation loss, surface hoargrows as a result of the sublimation of at-

mospheric water vapour onto the surface.

b) If surface hoar is buried, it mayconstitute a weak layer in the snowpack

that can persist for days or even weeks.

Surface hoar.


Influence of forest cover on the climate parameters that affect snow stability (after Frey and Salm 1990)

Parameter In forest openings Under closed-canopy forest

Wind Modified by terrain and forest Wind speed markedly reduced margins. Cornice formation within and below the canopy. Wind common at ridges. Winds may shakes the canopy, causing snow to scour the snow surface. fall, which disturbs the surface of

the snowpack below.

Precipitation Accumulation of snow Canopy interception and sublima-water equivalent (mm/hr) tion losses reduce accumulation onequal to precipitation rate the ground by typically 30% (snowunless modified by wind. water equivalent).

Air temperature Strong air temperature Lesser air temperature gradients gradients develop immediately develop immediately above theabove the snow surface. snow surface.

Radiation Receipt of short-wave radiation Snow surface is shaded by canopy.is affected only by topographic Long-wave radiation balanceshadowing and forest margins. damped and energy losses fromLosses of long-wave radiation snow surface reduced. promote surface cooling, Surface hoar formation reducedfavouring surface hoar and generally any surface hoarformation and upper-level destroyed by snow falling from thefaceting. canopy.

Melting and refreezing, occurring in the near-surface layers, producerapid but predictable diurnal changes in snowpack strength. In spring,once they become wet, layer boundaries within the pack become muchless distinct and the snowpack increases in density, hardness, andstrength. However, free water, produced by surface melt or from rain onfine-grained new snow, can lead to rapid loss of strength and subsequentsnow avalanche activity (Conway and Wilbour 1999).

2.2 AVALANCHE PHENOMENA

Snow Avalanche Classification

Avalanches in seasonal snow can be broadly classified by release type (i.e., mode of failure) as either loose snow avalanches or slab avalanches(Table 2). Further classification is based on depth of failure, mode offlow, and moisture of the avalanche mass.

Snow avalanche types described according to release mechanism

Visible Name characteristic Description

Loose Releases at • Movement starts at a point on or near the surface, thensnow a point spreads out in a triangular form.avalanche • Failure involves a local loss of cohesion at the snow

surface.

Slab Releases • Snow fails at a weak layer at depth in the pack.avalanche along a • A shear failure propagates under a cohesive slab.

fracture line • A tensile failure at the crown leaves a characteristic fracture line.

Loose Snow AvalanchesLoose snow avalanches start at a point and leave a characteristic invertedV-shape on the slope (Figure 9). The snow mass set in motion by loosesnow avalanches can have sufficient impact to damage poorly located facilities, such as mountain cabins, or to break small trees.

A local loss of cohesion at the snow surface (i.e.,commonly new snow or old wet snow) on a steepslope, or an impact from snow falling from trees orrock bluffs, is often all that is required to trigger aloose snow avalanche.

Loose snow avalanches are frequently small, butonce in motion they can trigger slab avalanches ifunstable snow exists downslope. Loose snowavalanches that involve wet snow can be very destructive if they are confined in a gully.

Slab AvalanchesA snow slab is a layer of cohesive snow that overliesa weaker layer that may fail. A slab avalanche oc-

curs when a fracture propagates under one or more layers of cohesivesnow that have accumulated over a weak layer, setting the unsupportedslab in motion.

When a slab avalanche releases, it leaves a set of readily identifiable fail-ure surfaces in the surrounding snowpack (Figure 10). A distinct fractureline is evident and can be anywhere from a few metres to a few kilometreslong (Figure 11). The failure may involve only the near-surface layers, inolder snow lower in the pack; or it may occur at the ground, resulting ina full-depth snow avalanche.


Loose snowavalanches start small but grow in

mass as they move downslope.Widespread loose snow avalanching

may occur on steep slopes iftemperatures rise quickly following a

fall of new snow.


Slab Failure

Immediately before a slabavalanche releases, a shear failureoccurs under the slab at the bedsurface and at the two flanks. Next,a tensile failure occurs at the crownface and a compressive failure atthe stauchwall. The slab is then setin motion.

Occasionally, only some of thesesurfaces fail and the slab remainssupported by the remaining surfaces. A characteris-tic “whumpfing” sound generally indicatesoccurrence of a shear failure at the slab’s bed surface.Johnson et al. (2000) used geophones to record thespeed of “whumpf” propagations on surface hoarlayers.

Cracks that propagate out in front of snowmobiletracks or from a person on snowshoes can be indica-tive of imminent slab failure. It may be that the slabis held in place only at the flanks or stauchwall. Par-tial failures are indicators of high instability and aclear cause for concern.

Slab volume can be calculated according to the dimensions shown in Figure 12.

Very generally, slab dimensions are typically twice aswide as they are long. This relationship enables snowavalanche sizes to be scaled. For example, a small slabmeasuring 100 m wide and 50 m long (i.e., 0.5 ha)failing to a depth of 1 m will release 5000 m3 of snowand typically a mass of about 1500 t. A slab avalancheof 10 ha occurring in a clearcut or burn area may set10 000 t or more of snow in motion. Thus, if a frac-ture propagates across a cutblock or burnt area, thena large destructive snow avalanche may be generated.In a detailed study of avalanches initiating inclearcuts in British Columbia, McClung (2001a)showed the area damaged downslope to be generallytwice the estimated start zone area.

CrownTensile failure

Bed SurfaceShear failure

StauchwallCompressive failure

Longitudinal profile

Slab

Failure surfaces in a slab avalanche.

a)

b)

Slab avalanchesexhibit distinct fracture lines. Variationsin the thickness of the slab are due tocross loading of the slope with wind-transported snow.

When a shear fracture propagatesunder a slab, stress concentrationsare likely at convex breaks in slope,below ridges and cornices wherethe slab thickness tapers, and atanchor points such as rocks or

trees. Steep road fill slopes appear to be a commonlocation for avalanche release within clearcuts (Figure 13).

Snow Avalanche Path Nomenclature

Three separate components can be identified inmost avalanche paths: the start zone (i.e., analo-gous to the initiation zone in a debris flow orlandslide), the track, and the runout zone (Figure 14).

When a slab avalanche initiates, the released snowslab rapidly accelerates in the start zone to form anavalanche that may gain additional mass down-slope. In the track, the avalanche achieves itsmaximum speed and there may be additional massgain. In the runout zone, the avalanche rapidly de-celerates and deposition occurs. Slope gradient

typically decreases down the path from the start zone to the track to therunout zone.

The start zone, track, and runout zone are readily identifiable wherever asnow avalanche becomes confined in a gully, but these components areoften less apparent on open slopes (Figure 15).


Flank Flank

Stauchwall

Crown Face

d

Plan View

L

W

Slab avalanche nomenclature. Estimatesof width (W), length (L), and depth (d) enable

calculation of slab volume as L x W x d, where: L is the length of the slab that fails (i.e., the

distance from crown face at the topdownslope to the stauchwall)

W is the width of the slab (i.e., averagedistance from flank to flank)

d is the average depth of the crown face (After Mears 1992)

Avalanches withinClearcuts

A survey of snowavalanche

occurrences withinclearcuts in the

interior and coastalBritish Columbiaregions indicates

that there is a clearpreference for slab

avalanches torelease in concaveterrain (i.e., bowls)

(McClung 2001a).

Road prism

Typicalfracturelocation

Cut and fill

Steep road fillslopes may act as preferredcrown locations.


It is possible for smaller snow avalanches to initiatein the track of a path that spans a large elevationalrange. Small avalanches may stop within the startzone on a large path.

Studies of a large number of slab avalanches haveshown start zones to feature a characteristic rangeof slope gradients (Figure 16). In a study of slabavalanches initiating in logged areas in British Columbia, McClung (2001a) found a mean startzone angle of 37° (78%) with a standard deviation of5° (n = 77). Harvested start zones ranged in slopefrom 30 to 50° (58–120%, respectively); refer to Table 3 for conversions from degrees to percent.

Slopes steeper than 55° (140%) are typically too steepto accumulate sufficient snow to produce large slabavalanches. New snow tends to slough off extremelysteep slopes as very small avalanches (sluffs) duringand immediately after storms.

When a cutblock is harvested onsteep terrain in an area with a largesnow supply, the area may poten-tially function as one large startzone (Figure 17). See Appendix 3for a more detailed discussion of asnow avalanche that occurred incutblock 83D007-27 at Nagle Creeknear the Mica Dam, north of Rev-elstoke, B.C.

Start zone, track,and runout zone of a confined snowavalanche path at McLean Point,Highway 16 near Terrace. Largerevents run out into the Skeena River.

Unconfinedavalanche path on an open slope in thealpine zone, running out into forestbelow.

25° 30° 35° 40° 45° 50°41% 58% 70% 84% 100% 143%

50%

40%

30%

20%

10%

0%

Slab

ava

lanc

hes

Bed surface declination

Slab avalanche dependence on startzone slope declination (n = 200). (After Perla 1977)

Start Zone Angles in Clearcuts

New data from cutblock surveys show that many avalancheevents initiate on slopes of around 30° (≈ 60%) (McClung2001a). The survey showed no relationship between snowavalanche size and start zone angle for clearcuts on slopesranging from 30 to 50°. This finding is significant as it indicatesno preferred slope angle for avalanching in blocks proposed forclearcut cable harvesting.

Although a large snow avalanche can cut a trackthrough a uniform forested slope, greater damageis likely if the mass of flowing snow is channelledby a local depression or if it enters an incised gully(Figure 18). Once trees are entrained, the avalancheis likely to be more destructive (see the Airy Creekexample, Chapter 6.5).

In densely forested mid-slope locations, it is oftendifficult to define the runout zone because the areamay not have experienced avalanching for severalhundred years. However, much of the steep terrainin high-snowfall areas of British Columbia wasprobably prone to snow avalanching prior toforests becoming established in the area after thelast glaciation, about 10 000 years before present. Itis also likely that many steep, forested slopes inareas of high snow supply have experiencedepisodes of avalanching following severe wildfires(Figure 19).


New snowavalanche paths cut through

mature forest below a steepclearcut near Nagle Creek,

destroying 12.5 ha of forest.

In 1982, a snowavalanche on the right-hand path ran

out farther than in the recent past,destroying 20 ha of old-growth forest.

The Greenslides avalanche path on Mt.Cartier, east of Revelstoke, is often active

in late spring.

Conversion of units used in measuring slope angle

Degrees Percent Percent Degrees

5.0 9 5 37.5 13 10 6

10.0 18 15 912.5 22 20 1115.0 27 25 1417.5 32 30 1720.0 36 35 1922.5 41 40 2225.0 47 45 2427.5 52 50 2730.0 60 55 2932.5 64 60 3135.0 70 65 3337.5 77 70 3540.0 85 75 3742.5 92 80 3945.0 100 85 4047.5 109 90 4250.0 119 95 4452.5 130 100 4555.0 143 105 4657.5 157 110 4860.0 173 115 49

Wildfire on thesesteep slopes has created steep mid-slope

openings capable of generating largesnow avalanches. Fire has increased the

risk to traffic on the road in the valleybottom.


Creep and Glide

On horizontal ground, snow deformation is termed “settlement”; onsloping ground, deformation produces both settlement and “snowcreep.” On smooth slopes, snow creep and glide produce very slowmovements in the snowpack; rates are in the order of millimetres per day (McClung et al. 1994).

Snow creep involves slow deformation within the snowpack, with fasterrates occurring at temperatures close to 0°C. Shear creep occurs whensnow grains within a layer on a slope undergo shear deformation due togravity (McClung and Schaerer 1993, pp. 63–67) as opposed to settlement,which occurs perpendicular to the slope (Figure 20).

Snow glide, a very slow translational slip of the entire snowpack acrossthe ground, occurs preferentially over very smooth surfaces (e.g., onsmooth grass slopes or planar rock slabs).

Creep and glide processes tend tobe most rapid at the start of winterbecause the ground is warm andthe early snow is of low density(viscosity is proportional to densi-ty). Creep and glide may bereactivated in early spring whenthe snowpack first becomesisothermal at 0°C and liquid waterpercolates to the base of the snow-pack.

Harvesting in the Runout Zone

Harvest planners and operators should consider the risk posed by snow avalanches that mightinitiate on steeper terrain above a block when planning to harvest on less steep terrain (e.g., ona fan) by ground-based methods.

W

W

Vu

u

Uu

Snow creep and glide deformationwhere: w is the slope-perpendicular creep velocity (settlement), u is the slope-parallel velocity, uu is the glide velocity, (u-uu) is the slope-parallel glide velocity, and v is the resultant velocity vector. (Source: SFISAR 1990)

Creep and glide can predispose a slope to avalanching (Figure 21). Whileglide cracks in the snowpack may serve as indicators of likely full-depthavalanching, meteorological parameters (e.g., temperature and net radia-tion) are better predictors (Clarke and McClung 1999). Glide avalanchesare difficult to forecast without on-site instrumentation. Generally, snowglide does not occur in clearcuts because of the rough ground surface.

A full-depth avalanche may occur when snowglides over smooth rock or wet ground. Glideavalanches generally occur when meltwater istrapped at the base of snowpack. A glide crackmay be evident for minutes to days before fail-ure occurs. Snowmelt and or rain-on-snowevents often provide the liquid water that pre-disposes smooth slopes to full-depth avalanches(Clarke and McClung 1999).

Avalanche Size

In Canada, snow avalanche size is classified according to its destructive potential (Table 4).The classification system embodies the conceptsof magnitude, exposure, and vulnerability,which are necessary components of any risk assessment (discussed in Part 2). The avalanchesize classification system uses a logarithmic

scale of destructive potential. It is therefore similar to the modified Mer-calli Scale used for ranking the shaking intensity of earthquakes.

The size of an avalanche is determined by the amount of snow that re-leases in the start zone, plus the net amount entrained in the track. Thesize of the start zone (in hectares), the depth of slab failure, and the me-chanical properties of the snow govern the initial volume of snowreleased. Hardness of the slab may control how far a fracture will propa-gate and thus what proportion of the start zone releases.

If a 10-ha area in a steep cutblock were to fail to a depth of 1 m, and givena typical slab density of 300 kg/m3, then approximately 30 000 t of snowwould be set in motion. By definition, this would generate a Size 4 snowavalanche.


Snow glide avalancheon a smooth grassy surface at Galena Pass.Note the other glide cracks surrounding the

glide failure.


Note that many fatalities occur in small avalanches that typically involve100–1000 tonnes of snow (Size 2.5 and 3). Jamieson and Geldsetzer (1996)give a broad review of recent snow avalanche accidents in Canada and include detailed summary statistics.

Classification of snow avalanche size (source: McClung and Schaerer 1981)

Typical

Path ImpactMass length pressure

Sizea Destructive potential (t) (m) (kPa)

1 The avalanche is too small to injure a person. <10 10 1

2 The avalanche could bury, injure, or kill a 100 100 10 person.

3 The avalanche could bury and destroy a car, 1 000 1 000 100damage a truck, destroy a small building, or break a few trees.

4 The avalanche could destroy a railway 10 000 2 000 500locomotive, large truck, several buildings, or a forest with an area up to 4 ha.

5 The avalanche could destroy a village or a 100 000 3 000 1 000forest with an area of 40 ha.

a Half sizes may be used to describe avalanches that are intermediate between classes.

Notes:

• The classification is designed to describe avalanches in seasonal snow. Events much greater

than a Size 5 have occurred when avalanches of glacial ice, permanent and seasonal snow, soil,

and meltwater have fallen from steep mountainsides.

• Measurements of slab width, length, and depth, and assumption of an average slab density,

allow snow avalanche mass to be calculated. A 1-ha area that releases to a depth of 1 m will typ-

ically produce a Size 3 avalanche. This calculation is often the only way of estimating size of a

powder avalanche that may blast across an area, destroying vegetation but leaving little mass

behind.

• Little is known about penetration of avalanches into mature forest, but typical impact pres-

sures associated with a Size 3 event (100 kPa) are sufficient to break mature trees.

• Different sizing systems are employed elsewhere in the world (including the United States), so

care must be taken when reading literature or discussing snow avalanche magnitude with prac-

titioners from other countries. Refer to Appendix D of McClung and Schaerer (1993) for a

discussion of other avalanche size classification systems.

Snow Supply

Snow supply is the primary determinant of avalanche frequency (Smithand McClung 1997; McClung 2000). Snow supply can often be estimatedby reference to snow survey records (Claus et al. 1984). Historical snow

accumulation data for British Columbia are available on the Internet(Appendix 4). In high snowfall areas along transportation routes, snow-fall records often enable reliable estimates of snow supply to be made(the B.C. Ministry of Transportation’s Snow Avalanche Program possess-es large winter climate data sets).

Risk analysis for forestry applica-tions requires that snowaccumulation data be analyzed toestablish the magnitude of snowsupply in avalanche start zones(i.e., annual, 10-, 30-, and 100-yearreturn period snow water equiva-lents) for any given elevation.McClung (2001a) has undertaken aregional analysis of snow accumu-lation versus elevation for BritishColumbia and showed that this re-lationship can be described by theextreme value statistical distribu-tion developed by Gumbel (1958).Analyses typically use the Hazenplotting position method (Watt etal. 1989, p. 55; Stedinger et al. 1993)(Figures 22 and 23).

Avalanche-prone areas broadly include the steep mid to higher el-evations of the coastal BritishColumbia and Vancouver Islandmountains where the mean annualmaximum snow accumulation ex-ceeds a threshold of 1000 mmwater equivalent, and the moun-tain ranges in the interior wheremean annual maximum snow ac-cumulation exceeds 700 mm waterequivalent. High-risk areas arecharacterized by an annual snowaccumulation that exceeds 1000 mm (interior) and 1900 mm(coast) (Stitzinger 2001; McClung


1200

1000

800

600

400

200

0-2 -1 0 1 2 3 4 5

Reduce variate

Snow

wat

er e

quiv

alen

t (m

m)

Glacier Park (1250 m) 1937–2000

SWE = 119.5x + 609

R2 = 0.96

Plot of 40 years of ranked snow waterequivalent (SWE) values from Glacier National Park, interior

British Columbia fitted to the Gumbel distribution. Fromthis example, the location (u) and scale (b) parameters from

the fitted regression equation, SWE= u + b{-ln[-ln(Hazenplotting position)]}, are used to predict the snow supply in

an avalanche start zone of similar elevation for variousreturn periods (Figure 23).

1200

1000

800

600

400

200

00 20 40 60 80 100

Return period (years)

Snow

wat

er e

quiv

alen

t (m

m)

Glacier Park (1250 m)

Expected snow water equivalents forvarious return periods predicted by Gumbel analysis of datain Figure 22 (logarithmic trend line fitted). The return period

for the critical snow supply associated with destructivesnow avalanches can be determined from this analysis (a

threshold of 700 mm is used for the Interior of BritishColumbia and 1000 mm for the Coast).


and Stitzinger 2002). Major snow avalanches occur only when criticalcombinations of weather and snowpack conditions exist. The winter of1998/99 featured about a 50-year return period snowpack and producedlarge avalanches in clearcuts (D. McClung, University of British Colum-bia, pers. comm.).

Because snow accumulation is strongly dependent on elevation, it is important to consider the elevation where proposed forest harvestingmay be undertaken. Regional analyses of the mean maximum snow accu-mulation are required to determine the snow supply in avalanche-proneterrain. McClung (2001a) presents curves for sub-regions of the ColumbiaMountains (Figure 24). It is appropriate to calculate the return period ofwinters when a threshold snowsupply will be exceeded at the elevation of the cutblock underconsideration.

Avalanche Frequency

Analysis of historical snowavalanche records often enables reliable estimates of avalanche frequency to be made. The B.C.Ministry of Transportation andParks Canada’s Rogers Pass snowavalanche programs possess largedata sets of snow avalanche occur-rence. However, such data must beinterpreted with care because ex-plosive-based control programs are suggested to cause about a three-foldincrease in snow avalanche frequency (Martinelli 1974).

The proximity of a designated snow avalanche area on a highway in BritishColumbia is an indicator that the combination of moderate to high snowsupply and avalanche terrain exists in the area (Figure 25 shows designat-ed avalanche areas). An open-ended snow avalanche hazard index is usedto rate the avalanche hazard on British Columbia highways, based on thefrequency and magnitude of avalanching as well as on traffic volumes inthe area (Schaerer 1989). The formula that underpins the ratings preventsthe hazard index from being applied in adjacent forested areas.

Away from public highways , the frequency of large magnitude (> Size 3)avalanches is difficult to judge because of a lack of data. Estimates canoften only be given to an order of magnitude. In low-snowfall areas in

0 500 1000 1500 2000 2500

Elevation (m)

Mea

n m

axim

um s

now

am

ount

(mm

w.e

.)

1800

1600

1400

1200

1000

800

600

400

200

0

Region 1: R2 = 0.74Region 2: R2 = 0.84Region 3: R2 = 0.95

Relationship of mean annual maximumsnow accumulation (SWE) with elevation for three sub-regions of the Columbia Mountains, interior BritishColumbia. Each point is produced by Gumbel analysis oflong-term data from a single snowcourse. (McClung 2001a)

the province—typically low total precipitation areasin the lee of the interior mountain ranges and thecoastal mountain ranges that are commonly sub-ject to intense rain rather than snowfall—availablerecords may not capture storms that producedlarge snow avalanche events.

Vegetative and geomorphic clues must be used todetermine areas that were affected by large snowavalanches in the past, and to estimate the potentialfor major snow avalanches in the future (e.g., Fig-ures 26–30). Dead mature trees, evident amongstyoung regrowth, do not necessarily indicate a high-frequency avalanche path. Instead, trees may indi-cate that a large-magnitude, low-frequency eventoccurred in the recent past. The absence of vegeta-tive clues (e.g., where fans have been cleared for


NOTE: Areas are classified as High avalanche hazard (red), Moderate hazard(yellow), Low hazard (blue), and Very Low hazard (white) based on traffic

volumes, exposure, proximity of adjacent paths, and frequency andmagnitude of avalanches. Although much of the Coast Mountain Range,

from Squamish through Bella Coola to Telegraph Creek, is avalanche-prone,there are few public roads in these areas. (BC MoT Snow Avalanche Programs)

Recent snowavalanche damage in mature forest.Dating the trees will give a first-order

estimate of snow avalanche frequency.

Vegetative trimlines and subtle colour differences

indicate relative frequency of snowavalanche activity. Note that the

avalanche path in the centre of thephoto, from Wood Arm, Kinbasket

Reservoir, has the potential todestroy additional areas of forest.

Snow avalanche

areas affecting British Columbia

highways.

C A N A D A

U S A


farming or developed for habitation) increases the uncertainty associatedwith estimates of snow avalanche frequency or runout potential.

Several approaches should be used when attempting to establish the fre-quency and magnitude of snow avalanching in forested areas. Generally,few, if any, observational data are available in remote forest areas, butroad maintenance personnel (especially grader oper-ators) may have valuable personal knowledge as tothe frequency of snow avalanches running out ontoroads in an area. Reports of snow avalanches affect-ing bridges or blocking mainline roads are very useful.

In British Columbia, the Earth Science Task Force ofthe Resources Inventory Committee () has devel-oped a method for assessing landslide hazard,consequence, and risk (Gerath et al. 1996; Table 5).According to the criteria in Table 5, snowavalanche frequency would generally be rated as“high” or “very high.” In snow avalanche work, it isgenerally not possible to discriminate between ’s“moderate” and “low” relative fre-quency classes.

Snow avalanche frequency is deter-mined both by climate and terrain.The Coast Range has a higher snowsupply, which may increase the frequency of avalanching. In theinterior of British Columbia, higher

Terminology used to describe frequency ofmountain slope hazards (after Gerath et al. 1996; Hungr1997)

Range of annual:

Return Probability of Relative period occurrence term T (years) (Pa)

Annual 1/1 Pa ≈1.0

Very high 1/20 to 1/2 0.05 < Pa < 0.5

High 1/100 to 1/20 0.01 < Pa < 0.05

Moderate 1/500 to 1/100 0.002<Pa < 0.01

Low < 1/500 Pa < 0.002

Development in the runoutzone has eliminated clues thatmight otherwise be used toassess snow avalanchefrequency in these avalanchepaths at Three Valley Gap westof Revelstoke.

Snow avalanches regularlyrun down the centre path atFish Lake, between Kaslo andNew Denver. Less frequentevents may affect other partsof the face. Most of theseslopes were affected by fireand may continue to besusceptible to avalanching.Note that snow avalanchesalso run down the slopebelow the photographer.

An indication ofsnow avalanche frequency can beobtained from tree ring pattern, whichmay show effects of avalanche impact.It may be possible to date some of thescars.

elevations are more prone to avalanche initiation than lower elevations.Snowpacks in the Interior are dominated by surface hoar and facetedweak layers, which also make this region susceptible to avalanching (McClung 2001a). In the long term, the two effects probably offset oneanother, but one regime may dominate in any given winter. There is agreater chance of a high-intensity rain-on-snow event occurring in thehigh-snowfall areas of the Coast Range, leading to wet snow avalanchingduring the winter.

Avalanche Probability

If snow avalanches are assumed to be rare independent events, then theirfrequency (or arrival rate at some point of interest such as a forest roador bridge location) can be described using a statistical distribution knownas the Poisson process (McClung 2000). LaChapelle (1966) and Smithand McClung (1997) used the Poisson process to calculate an encounterprobability for snow avalanches (i.e., the chance that at least oneavalanche will reach or exceed a certain point in its path in a given period).

Encounter probability describes the chance that at least one snowavalanche of a specified return period of T years will occur within anygiven interval of N years, as follows:

E = 1 – (1 – 1)NT

where:N is the number of years consideredT is the avalanche return period (in years)

T = 1Pa

Pa is the annual probability of snow avalanche occurrence (ranging be-tween 0.0 and 1.0)

Mears (1992) used this method to illustrate encounter probabilities forvarious return periods and lengths of observation record (or period ofoccupation of a path). Table 6 confirms that there is a good chance ofwitnessing a 30-year return period event in a human lifetime, but only aslim chance of witnessing a 100-year event. Encounter probability can beused to demonstrate that the length of record should be at least twice aslong as the return period to have at least 90% confidence in experiencingthe event in question.



The method can also be used tocalculate the probability of not experiencing at least one snowavalanche in a given period of observation or exposure. For in-stance, given that a 100-year returnperiod snow avalanche has an en-counter probability of 0.39 in any50-year period, then the probabili-ty of non-encounter is (1–0.39) or0.61. The probability of not experi-encing at least one avalanchedecreases with time for any givenreturn period (Figure 31).

Most continuous records of snowavalanche observations in NorthAmerica are too short (10–40years) to confidently contain anobservation of long return periodsnow avalanches. Mears (1992)considers that this has led to a poorperception of the snow avalanchehazard in many quarters, as well asthe use of poor information in landuse planning.

Given the general lack of data inBritish Columbia and elsewhere inNorth America, snow avalanchefrequency can often only be esti-mated to an order of magnitude. Mears (1992) recommends that adistinction should just be drawn between 10-year and 100-year return pe-riod avalanches, where the return period describes a range of time. Withthis uncertainty, a “10-yearavalanche” may have anactual return period of3–30 years, while a “100-year avalanche” mayhave an actual return pe-riod of 30–300 years.

Encounter probability for various return period avalanches in given periods (afterMears 1992)

Length ofReturn record Probabilityperiod (or occupation) of

T (years) N (years) encounter

10 10 0.6510 30 0.9610 50 0.9930 10 0.2930 30 0.6430 50 0.8230 100 0.97

100 30 0.26100 50 0.39100 100 0.63100 200 0.87100 300 0.95

0 10 20 30 40 50

1.0

0.8

0.6

0.4

0.2

0

T = 30

T = 10

T = 5Prob

abili

ty o

f non

-enc

ount

er

Years

Probability of not experiencing oneavalanche of a 5-, 10-, or 30-year return period (T) as afunction of time.

Order of magnitude estimates of snow avalanchereturn period:

10-year avalanche return period ⇒ 3 < T < 30 years100-year avalanche return period ⇒ 30 < T < 300 years

(Mears 1992)

Vegetative IndicatorsWhere no snow avalanche occurrence records are available (i.e., mostforest situations), vegetative clues may be used to estimate the avalanchefrequency (Table 7). Dendrochronological interpretation of incrementcores from a few trees (Figure 32), supplemented with diameter at breastheight (dbh) data, may assist in establishing the date of the most recentdisturbance at any point in the track or runout zone (Burrows and Bur-rows 1976; Hétu 1990; Jenkins 1994; Boucher et al. 1999).

Vegetative indicators of snow avalanche frequency (after Mears 1992; McClung and Schaerer 1993)

Frequency: at leastone event in period Vegetative indicators

1–10 years Track supports grasses, shrubs, and flexible species (e.g., alderand willow). Patches of bare soil and shrubs. No trees higher than 1–2 m. No dead wood from large trees except at edges or distal end of runout zone.

10–30 years Predominantly pioneer species. Dense growth of small trees andyoung trees of climax species similar to adjacent forest. Brokentimber on ground at path boundaries.

30–100 years Mature pioneering species (non-coniferous) of uniform age andyoung trees of local climax species. Old and partially decomposeddebris.

More than 100 years Mature, uniform-age trees of climax species. Increment core datauseful.

Moss and lichen may be polished offthe uphill sides of tree trunks and treesmay display a J-shaped swept butt. If asnow avalanche entrains talus and rub-ble, then rocks may be embedded intrees (Figure 33).

In some paths, snow avalanches mayrun between scattered trees, evidencedby broken lower limbs on all but thedownhill side of a tree, a process knownas flagging (Figure 34). Note, however,that trees on very windy sites, whererime accumulation is common, may besimilarly flagged. Where more destruc-tive snow avalanches occur, tree trunks


Vegetativeindicator: J-shaped tree trunk

indicates probable snowavalanche impact. An incre -

ment borer can be used toobtain a core, which may

show scars and reaction wood.

Rubblelodged in tree trunk

indicates moderate impactpressure.


may be snapped, shattered, or scarred on the upslopeside (Figures 35 and 36). In the runout zone, rubblemay remain perched on stumps, in live vegetation,or on woody debris after snow avalanche debris hasmelted (Figures 37–39). This evidence can help dis-tinguish between the effects of windthrow and snowavalanches. Trees destroyed by avalanche blast canbe difficult to distinguish from those dropped bywindthrow (Figure 40).

Geomorphic IndicatorsChannelized wet snow avalanches and debris flowsare alike in many respects, parts of a continuum ofgeomorphic processes, separated by a phenomenonknown as “slushflow”—the rapid mass movement ofwater-saturated snow (Hestnes et al. 1994). Slush-flows are considered to be relatively common insmall creeks in high-rainfall regions of coastal BritishColumbia and frequently block or wash out culvertsand bridges (R. Gee, pers. comm.).

Although the cumulative effects of successive snowavalanches can involve movement of substantial vol-umes of surficial material, it should be noted that

Flagged tree. A snowavalanche initiating in aclearcut damaged the foliageto the full height of this tree,Akolkolex River.

Tree trunk snapped by a snowavalanche at Greenslidesnear Revelstoke indicatesimpact pressures of 50–100 kPa.

Impact scars from entrainedmaterial.

Rubble perched on loggingslash deposited by a recentsnow avalanche.

Trim lines and debris in valley bottom, Serpentine Creek,North Thompson.

Rubble perched in alder.

many of the gullies in which snow avalanches occurtoday may owe their geomorphological origin tomore active episodes of mass movement immedi-ately following the last glaciation (Butler andMalanson 1990).

In alpine areas, slope angle, slope segmentation,and longitudinal and sedimentological sorting canbe used to differentiate rockfall talus from snowavalanche landforms (Jomelli and Francou 2000).Figures 41–43 show a variety of geomorphic indica-tors, including avalanche tarns and cones.

2.3 AVALANCHE DYNAMICS

At high-consequence or other critical sites, it willbe necessary to consider avalanche dynamics. Thesubject is both technical and complex. The applica-tion of runout models or calculation of impactpressures is in the realm of professional engineer-ing and geoscience.

Avalanche Speed and Motion

Avalanches move by sliding or flowingover the surface; some may becomeairborne. A single avalanche can ex-hibit all three types of motion.

Dry Snow AvalanchesAvalanche speed varies with the size ofthe avalanche and the moisture con-tent of both the snow that fails and thesnow that is entrained downslope. Drysnow avalanches on steep slopes canaccelerate rapidly and quickly attainspeeds of 25 m/s or more (90 km/h) inthe first 100 m downslope (Figure 44).In recent field measurements at ahighly instrumented site in Switzer-land, frontal speeds approaching 90 m/s (325 km/h) were recorded in anavalanche with a flow height of 10 m(Dufour et al. 2000).


Large area of maturetimber destroyed by natural

avalanching, at South Rice Brook,tributary to the Kinbasket Reservoir. Note

avalanche tarn on right.

Snow avalanchescarrying detritus downslope continue to

build a fan in the runout zone, UpperAllan Creek, North Thompson. Note trimlines and destruction of immature trees

in the foreground.

Steep conical fan built by

successive snow avalanchescarrying colluvium down

from an alpine start zone,Serpentine Creek, North

Thompson.

An avalanche tarn is an

impact feature formed in therunout zone of a large snow

avalanche path. Materialdeposited in the tarn by

avalanches has been subse-quently gouged out, building

the berm on the downslopeedge. The berm has been

stabilized by alder and otherpioneering vegetation.


There can be considerable variation in speed anddensity within a large, mixed-motion avalanche(Figure 45). The flowing component involves a high-density core of granular material (McClung 2001c).At speeds of more than 10 m/s (35 km/h), a low-density dust cloud develops above the flowing component, resulting in a mixed-motion avalanche(Figure 46) (McClung and Schaerer 1993, p 105).

After failure, a snow slab often breaks up into blocksthat may disintegrate into dense granular particles.The rate and degree of disintegration depend on thehardness of the slab and geometry of the path (Figure 47).

Immediately after an avalanche, vegetation damageand snow deposited on trees can reveal informationabout the depth of the flowing and airborne compo-nents of mixed-motion avalanches (Figure 48).

Maximum avalanche speed is a function of the scaleof the path (Figure 49). McClung (1990) analyzed allknown avalanche speed measurements and estab-lished two relationships:

0 500 1000 1500 2000

60

50

40

30

20

10

0

Spee

d (m

/s)

Distance along path (m)

Speed in a large dry avalanche showsrapid acceleration to a maximum of 62 m/s (230 km/h) andvery rapid deceleration. (Source: McClung 1990, frommeasurements by H. Gubler of the Swiss Federal Institute forSnow and Avalanche Research)

Speed and density profiles for a largeflowing avalanche. (After Perla and Martinelli 1976)

Large, mixed-motionavalanche with an airborne component.The powder cloud may be in excess of 40 m high, while the dense flowing coreis probably about 5 m thick.

Large blocks of debris(> 1 m3) remained intact in a hard slabavalanche that slid 300 m down a steepopen slope.

Speed, m/s

u60

30

Density, k

g/m3

100200

ρρ

ρρ

u

Umax= 1.5 (slope distance ) 0.5

andUmax= 1.8 (vertical fall) 0.5, where U = the speed of the snow avalanche.

These relationships should be used to test the validity of any modelledavalanche speeds.

Powder AvalanchesPowder avalanches lack the dense core of a mixed-motion event. The snow is entirely airborne andheld in suspension by turbulent eddies. This typeof avalanche can be very destructive. Powderavalanches often form when the moving snowmass falls over steep cliffs or bluffs (Figure 50).Small powder avalanches are sometimes called“snow dust” events; while normally harmless, theseevents can present a serious traffic hazard whenvisibility is lost.

Powder avalanches can often only be distinguishedfrom high-speed, dry mixed-motion events by thedeposit. Fast-moving powder avalanches havesheared off steel bridge guard rails but left no morethan a few centimetres of snow in the vicinity (McClung and Schaerer 1993, p. 110). Mears (1992,p. 13) describes extensive damage in a subdivision

in Juneau, Alaska caused by a powderavalanche that entrained timber in itspath.

Air BlastIn some terrain, avalanches may buildup an air pressure wave that precedesthe flowing snow mass and powdercloud. Effects of air blast are difficultto distinguish from damage caused bypowder avalanches, unless the event iswitnessed.

Air blast is likely to be less destructivethan a powder avalanche because thedensity of the snow dust and air mix


80

70

60

50

40

30

20

10

00 10 20 30 40 50 60 70

Um

So

Maximum avalanche speed versus squareroot of slope distance (So) for avalanches from Canada (o),

and Switzerland and Norway (*). The envelope Umax = 1.5 (So) 0.5 is shown. (Source: McClung 1990)

Snow plastered on this tree trunk

indicates flow height of core ofavalanche. Flagged branches higher

in the tree indicate height ofturbulent, airborne component.


is about 10 times higher than the density of air alone (McClung andSchaerer 1993, p. 107).

Wet Snow AvalanchesWet snow avalanches generally have no dust cloud and are characterizedby their very high surface friction. Large wet snow avalanches can be verydestructive. They often entrain vegetation, soil, andboulders in the track and, if they run frequently, canbe powerful geomorphic agents (Figure 51).

Wet snow avalanches tend to follow depressions andgullies and can be deflected by changes in terrain andobstacles in the path. Like debris flows, wet snowavalanches sometimes flow in a slow, surging man-ner. They have been known to make right-angleturns and run down low-friction surfaces such asroads, causing damage in areas that had been consid-ered safe zones. Engineered earthworks in the runoutzone can redirect slow-moving wet snow avalanches.

Multiple wet snow avalanches may release in a rain-on-snow event. Individual releases may be small butcan coalesce to form large deposits.

Speeds of wet snow avalanches may be considerablyless than those of dry events, but the flow densitiesmay be twice as high at 150–200 kg/m3 (McClungand Schaerer 1993, p. 115).

2.4 IDENTIFICATION OF SNOW AVALANCHE TERRAIN

Based on field work and air photo interpretation, the U.S. Forest Service established guidelines foridentifying and evaluating snow-avalanche terrain(Martinelli 1974). A checklist is presented in Appendix 5.

Any clearcut or forest opening located below a ridgeor high plateau in the lee of the prevailing wind islikely to feature increased snow loading. This in turnincreases the susceptibility of the area to avalanchesafter forest harvesting.

a) A large slab avalanche descends oversteep cliffs and becomes airborne.

b) Larger snow particles fall out and apowder avalanche engulfs the valleybelow the cliffs.

c) A harmless powder cloud of snowparticles in suspension travels more than2 km down the valley.

Sequence of a powder avalanche.

Schaerer (1977) proposed a five-level index of wind exposure (Table 8),which can be used to assess the likelihood that the snowpack in a clearcutor burn area may be loaded by drifting snow, and hence prone toavalanching. An example of wind exposure class 5 is shown in Figure 52.

Analysis of 76 destructive avalanche occurrences inharvested blocks in British Columbia by McClung(2001a) showed: 1. a start zone mean slope angle of 37°, a standard

deviation of 5°, with a range of 30–50°;2. a prevalence of moderately concave slopes,

with cross-slope concavity being more pro-nounced than in the downslope direction;

3. ground roughness, vegetation height, and vege-tation coverage being potentially important ininhibiting initiation; and

4. a wind exposure index favouring classes W2and W3 (i.e., moderate but not extreme windexposure).

Within cutblocks, cross-loading of gullies by wind-transported snow canproduce localized snow accumulations that increase avalanche suscepti-bility. Because wind flow in mountainous regions is a complex subject, itis difficult to predict how wind flows may vary once trees are removed.Foresters considering windfirmness and windthrow potential share thisproblem (Stathers et al. 1994).


Large wetavalanche flowing from a confined

gully out onto a fan, Mt. Bunting,Bear Pass, near Stewart, B.C.

Ranking of avalanche start zone wind exposure (Schaerer 1972)

Windindex Start zone wind exposure

W1 Start zone completely sheltered from wind by surrounding dense forest.

W2 Start zone sheltered by an open forest or facing thedirection of the prevailing wind.

W3 Start zone on an open slope with rolls or other irreg-ularities where local drifts can form.

W4 Start zone on the lee side of sharp ridge.

W5 Start zone on the lee side of a broad, rounded ridgeor open area where large amounts of snow can bepicked up by the wind (Figure 52).

Wind exposure class 5. Cornice development indicates

extensive wind-transported snow (right to left).


Boundary layer wind flow models have been developed for forest areas(Oke 1983, p. 133; Greene et al. 1999), but it is unlikely that they can oper-ate at the scale of a cutblock or predict edge effects or local wind flowperturbations around small, micro-scale terrain features.

2.5 RUNOUT CHARACTERISTICS

Issues related to prediction of avalanche runout distance are central toanalysis of consequence associated with forest harvesting on steep terrain.

Avalanche runout distance is correlated with avalanche magnitude. Lar-ger avalanches generally run farther on any given path, with the watercontent of entrained snow being an important determinant.

Flow conditions along an avalanche path are controlled by the mech-anical properties of flowing snow, the nature of the sliding surface andterrain configuration (channelled vs. open slope), and surface roughnessof the path. These parameters, along with the mass of snow released, areimportant determinants of runout distance of extreme snow avalancheevents (Mears 1992).

In narrow valleys with high relief, avalanches may not only run out in thevalley floor but may also run up slopes on the other side of the valley(Figures 53 and 54).

An avalanche that initiates high on a mountain as a dry slab may entrainwet snow at lower elevations (Figure 55). If a large snow avalanche be-comes confined in a gully, then runout distances can be very long, withthe mass generally not decelerating until the path gradient decreases toaround 8–12° (14–21%) (Mears 1992).

Avalanche Susceptibility in Forest Openings

Any clearcut block, burn area, or other forest opening should be regarded as having thepotential to generate destructive snow avalanches if it has the following characteristics:

• a concave profile, either down or across a slope (typically bowls or gullies) with a gradient steeper than 30° (58%);

• an adequate supply of new snow;

• moderate exposure to wind; and

• an average depth of winter snow that is greater than the height of any rough surfacefeatures (e.g., projecting stumps or slash).

Wet snow avalanches tend to pile up and formdeep deposits as they come to rest, often where theavalanche discharges from a confined gully onto afan (Figure 56). The combination of high densityand depth can make wet avalanches very destruc-tive. The slow-moving mass will entrain trees, largeboulders, and soil from the path.

2.6 ROLE OF FOREST IN AVALANCHE PROTECTION

Snow accumulation, snowpack layering, the energybalance, and avalanche frequency can be markedlydifferent in closed-canopy forest compared to forest openings (Table 9). These differences canrepresent important controls on snow stability andsusceptibility to avalanching.

Research from Austria indicates that snowavalanches are more likely to occur in harvestedareas than in openings created by natural dieback,because much less woody material remains on theslope in harvested areas (Heumader 1999).

Studies from the Canadian Rockies show thatyoung trees with basal diameters exceeding 0.1 m,growing near the top of snow avalanche runoutzones, are generally uprooted or broken when


Avalancherun-up is likely to occur

opposite avalanche pathsthat descend into narrow

valleys, such as theCummings River.

Trees lying upslopeon the opposite side of the valley from

the avalanche path indicate thedestructive forces encountered in a large

avalanche.

This avalanchestarted in a hard dry slab, but melt,

induced during the avalanche’s longdescent in a broad gully, caused the

mass to flow in the runout zone in themode of a wet snow avalanche. Note

the superelevation of the deposit on thisoutside bend in the path. (Refer

McClung 2001c)

This large wet avalanche released in spring, flowed a long distance,and left a deep deposit. St. Mary Valley, Purcells.


impacted (Johnson et al. 1985). A theoretical European study indicated thattrees with diameters up to 0.3 m may be broken by snow avalanches run-ning as little as 30 m distance downslope (Gubler and Rychetnik 1990).

Differences in snow accumulation and stability between forest openings and closed canopy coniferous forest(after Frey and Salm 1990)

Variable In forest openings Under closed-canopy forest

Snow depth • Controlled by topography • Less snow accumulates under forest• Greater micro-scale variability

(loss due to canopy interception of snowfall)

Snow • Higher rates than under forest • Fewer faceted grains because of metamorphism • High rates of radiative energy reduced radiation losses and less

exchange and greater air air temperature variationtemperature variation can promote rapid changes in grain form

Layering of • Continuous layering over large • Layer formation disturbed bysnowpack areas dripping meltwater and snow

• Enhanced formation of weak falling from canopylayers (surface hoar and depth hoar)

Creep and glide • Disturbance by snow creep • Creep and glide insignificant(movements or glide depends on local because of rough surfaces< 1 cm/day) surface roughness

Avalanche type • Large avalanches possible • Small loose snow avalanches• Slab releases common common

• Large slab avalanche releases are rare but may occur in openings wider than one tree height

2.7 HYDROLOGICAL EFFECTS

Wet snow avalanching often coincides with rain-on-snow events (Figure 57) (Ferguson 2000). Snow avalanches can occur within a fewminutes to a few hours of the onset of rain (Conway and Raymond 1993;Carran et al. 2000). When widespread avalanching or a single very largeavalanche deposits debris in a mountain stream channel, the nature andtiming of stream discharge can be altered (Figure 58).

The hydrological effects of deep accumulations of avalanche snow de-posited well below the snowline often counteract each other (de Scally1992). For example, because the albedo (reflectivity) of snow avalanche

debris is lowered by entrained soil and because lower elevations provide awarmer ambient environment in spring, accelerated melt and consequentincreases in peak stream discharge are likely. However, if high loads ofsurficial materials are entrained in an avalanche, then debris will melt outonto the surface to provide a colluvial veneer that insulates the remainingavalanche snow, retarding further melt. Topographic shadowing of the

valley stream channels will also retard the melt of avalanche-deposited snow, compared to thesnowmelt rates that would occur on higher, sunny slopes.

Snow avalanches rarely dam streams for more thana few hours, although there are noted exceptions(de Scally 1996). Streams often rapidly undercutavalanche deposits (Figure 59).

When large numbers of trees and high volumes of soil are entrained by a snow avalanche, thenconsiderable volumes of woody debris, as well assediment, will be introduced to the channel system(Figure 60). Geomorphic effects may be similar to those found when lateral bank erosion occurs

in streams draining forested watersheds.

The incidence of other rapidmass movement processes (e.g.,debris flows) can be increasedwhen snow avalanches run downsteep gullies into a stream system,promoting channel disturbance(de Scally 1996).

Snow avalanches can generatesignificant waves when waterbodies are impacted. Given that a large avalanche typically trans-ports 10 000–100 000 tonnes ofsnow to the runout zone, debris

from such events can apply considerable loads to thick ice on frozenlakes. De Scally (1996) cites a report of a snow avalanche that struck a 1.7 m thick ice sheet over a small lake, causing a flood wave that drained70% of the unfrozen water in the lake.


Significant forestdamage caused by a snow avalanche at

Ruby Mountain in the North CascadeMountains of Washington. The

avalanche initiated above timberlineand descended into forest during a rain-

on-spring-snow event, following aperiod of sustained warming.

Conceptual stream flow model for asubalpine stream. Hydrograph separated to show (1) base

flow, (2) spring snowmelt, (3) avalanche snowmelt, and (4) rain. Avalanche deposits can extend the hydrograph

recession, maintaining mid-summer flows. (de Scally 1996)

1. Baseflow2. Normal snowmelt

(individual peaks not shown)3. Avalanche snow meltwater4. Quickflow from rainfall

1

3

24

J F M A M J J A S O N D

Dis

char

ge


Snow avalanches from Mt. Rainey above Stewart in northern British Columbia have plunged into the Portland Canal and created surges thathave torn barges from their moorings in the vicinity of the log loadingarea at the port (Figure 61).

2.8 ECOLOGICAL SIGNIFICANCE

Snow avalanche paths are diverse, often productive ecosystems. They provide habitat for a variety ofwildlife species. In forested environments, snowavalanche paths form ecotones (or edges) betweenmature forest and early seral communities. Depend-ing on slope, aspect, elevation, moisture andnutrient characteristics, and frequency of distur-bance, seral communities in snow avalanche pathsmay consist of sparsely vegetated colluvium, herb,shrub-herb, deciduous pole sapling, coniferous polesapling, or complex mosaics of these communitytypes (Figure 62). As determined by these site factors,early seral communities may proceed normallythrough successional sequences, persist for long pe-riods of time, or become self-perpetuating disclimaxcommunities no longer capable of reaching edaphicor climatic climax states.

Wildlife habitat values associated with snowavalanche paths have been summarized by Stevens(1995) and Quinn and Phillips (2000). Stevens(1995) identified 35 wildlife species on British

Within a minute of the snow avalancherunning into the channel, the stream had undercut thedebris and melted a considerable volume of snow.

Rapid breach of a snow avalanche damcaused Mobbs Creek, near Trout Lake in the Kootenays, toalter course and damage a trout spawning channel.

Barge torn from itsmoorings by a snow avalanche from Mt. Rainey (far side of the harbour), Stewart, B.C.

Snow avalanchepaths often contain a complex mosaic ofplant communities.

Columbia’s provincial “Red and Blue Lists” that utilize snow avalanchepath habitats in one or more of the province’s 15 biogeoclimatic subzones(Tables 10a and b). Note that while Alpine Tundra () is a non-forestedsubzone, snow avalanche habitats occur there.

Stevens (1995) defined snow avalanche path plant communities as“shrubland dominated by alders, or other shrubs where periodic snowand rock slides prevent coniferous forest establishment and where mois-ture is plentiful for much of the growing season; lower areas may supportrich herbaceous growth.” From a wildlife habitat perspective, snowavalanche paths should be considered at the landscape level. For example,in upper elevations of the Engelmann Spruce–Subalpine Fir () bio-geoclimatic zone, where there is little fire history, snow avalanche pathsprovide early seral habitats that are otherwise rare at a landscape level (R. Ferguson, R.P.Bio., pers. comm.). These habitats are used by foxsparrows and warbling vireos, among other bird species, some of whichare not found outside of avalanche paths in the biogeoclimaticzone. Forested margins of avalanche paths are of importance to cavity-nesting birds. Avalanches can shear the crowns from mature conifers onavalanche path margins, rendering these trees susceptible to rot.Avalanche-damaged trees, particularly large-diameter Douglas-fir, are extensively used by primary cavity-nesting birds in the biogeocli-

matic zone.

Grizzly Bear Habitat

Grizzly bears are known to use avalanche pathhabitats extensively, particularly during springwhen lush herbaceous communities provide anabundance of forage species including grasses,ferns, horsetails, clover, cow-parsnip, hedyserum,glacier lily, and spring beauty (Mowat and Ram-charita 1999).

The forested edges of avalanche paths provide se-curity (i.e., visual and escape) and thermal coverfor grizzly bears close to foraging sites (Mowat andRamcharita 1999). The runout zone and adjacentvalley-bottom streams also offer high-value foragefor grizzly bears and a variety of ungulate species(Figures 63–65).


Avalanche pathsdominated by alder, willow, and

herbaceous sites often contain valuablespring bear foods such as forbs,

horsetail, and fall berries. Ferns, ifpresent, occur only in small numbers.

The sites tend to be moist and are oftenassociated with surface water. They

occur along drainage channels in thecentre of avalanche tracks or in wet

runout zones. Aspects vary greatly, so itappears that moisture is the key

environmental factor. This site offersvery high grizzly bear habitat value

based on bear foods (Quinn and Phillips 2000).

Chapter 2

Snow and avalanche phenom

ena39

a Species whose habit includes snow avalanche-prone terrain by British Columbia biogeoclimatic zone (after Stevens 1995) 1

Biogeoclimatic zone / subzone AT SWB BWBS SBPS SBS CWH CDF BG PP IDF MS ESSF ESSFp ICH MH

Birds Status*

American kestrel Falco sparverius h,w,l

Golden eagle m,e,s s,b d,m,k v,d, d,k,h, m,s c h,w x,d x,d, x,v, x,d,m x,d,m d,k,m, w,lAquila chrysaetos k,c c,x m,w k,m w,c,x

Merlin h,w,lFalco columbarius

Northern goshawk m,e,s s,b d,m,k v,d, d,k,h, h,m,s c h,w x,d x,d, x,v, x,d,m x,d,m d,k,m,Accipiter gentilis k,c c,x m,w k,m w,c,x

Northern goshawk ssp. laingi Red h,m,s h,wAccipiter gentilis laingi

Northern shrike s,b d,m,k v,d, d,k,h, h,m,s c h,w x,d x,d, x,v, x,d, x,d, d,k,m,Lanius excubitor k,c c,x m,w k,m m,w m,x w,c,x

Red-tailed hawk h,w x,d x,d, x,v,Buteo jamaicensis w k,m

White-tailed ptarmigan m,e,s b k x,d,m h,w,lLagopus leucurus

White-tailed ptarmigan ssp. saxatilis Blue m hLagopus leucurus saxatilis

Wilson’s warbler v,d, d,k,h, h,m,s c h,w x,d x,d, x,v, x,d,m x,d,m d,k,m, h,w,1Wilsonia pusilla k,c c,x m,w k,m w,c,x

Yellow warbler s d,m,k v,d, d,k,h, h,m,s c h,w x,d x,d, x,v, d,k,m, w,lDendroica petechia k,c c,x m,w k,m w,c,x

1 Refer to Table 10b for biogeoclimatic units.

40Snow

Avalanche M

anagement in Forested T

errain

Table 10a Continued

Biogeoclimatic zone / subzone AT SWB BWBS SBPS SBS CWH CDF BG PP IDF MS ESSF ESSFp ICH MH

Mammals Status*

Bighorn sheep ssp. californiana Blue e d h,w x x,d x,v,m x,d x,dOvis canadensis californiana

Black bear m,e,s s,b d,m,k v,d, d,k,h, h,m,s c h,w x,d x,d, x,v, x,d, x,d, d,k,m, h,w,1Ursus americanus k,c c,x m,w k,m m,w mw w,c,x

Black bear ssp. emmonsi Blue s d mUrsus americanus emmonsi

Caribou (northern populations) e,s s,b d,m,k v,c k,c v x,d, x,d, c,xRangifer tarandus m,w m,w

Caribou (southeastern population) Blue e m,k c,x k d,w d,w c,xRangifer tarandus

Coyote m,e,s s,b d,m,k v,d, d,k,h, m,s h,w x,d x,d, x,v, x,d, x,d d,k,m, w,1Canis latrans k,c c,x m,w k,m m,w m,w w,c,x

Elk ssp. nelsoni m,e,s s,b d,m,k v d,h,x h x,d x,d x,v, x,d, x,d, d,k, h,lCervus elaphus nelsoni k,m m,w m,w m,w

Elk ssp. roosevelti Blue m h,m c wCervus elaphus roosevelti

Gray wolf m,e,s s,b d,m,k v,d, d,k,h, h,m,s c h,w x,d x,d, x,v, x,d, x,d, d,k,m h,w,1Canis lupus k,c c,x m,w k,m m,w, m,w w,c,x

Grizzly bear Blue m,e,s s,b d,m,k v,d, d,k,h, h,m,s x.d,w x,v, x,d, x,d, d,k,m, h,w,lUrsus arctos k,c c,x k,m m,w m,w w,c,x

Least chipmunk ssp. oreocetes Blue d dTamias minimus oreocetes

Least chipmunk ssp. selkiri Red d,w d,wTamias minimus selkirki

Chapter 2

Snow and avalanche phenom

ena41

Table 10a Concluded

B.C. biogeoclimatic zone / subzone AT SWB BWBS SBPS SBS CWH CDF BG PP IDF MS ESSF ESSFp ICH MH

Mammals Status*

Long-tailed weasel ssp. altifrontalis Red e m,h w w,lMustela frenata altifrontalis

Moose m,e,s s,b d,m,k v,d, d,k,h, m,s x,d, x,v, x,d, x,d, d,k,m, w,lAlces alces k,c c,x m,w k,m m,w m,w w,c,x

Mountain beaver ssp. rainieri Blue m mAplondontia rufa rainieri

Mountain goat m,e,s s,b d,m,k c,x h,m,s x x,d,w x,k,m x,d, x,d, m w,lOreamnos americanus m,w m,w

Mule deer ssp. columbianus m h,m,s c c,x h,w,lOdocoileus hemionus columbianus

Mule deer ssp. Hemionus m,e s d,m v,d, d,k,h, s h,w x,d x,d, x,v, x,d, x,d, d,k,m,Odocoileus hemionus hemionus k,c c,x m,w k,m m,w m,w w,c

Mule deer ssp. sitkensis s,b d h,m,s w w,lOdocoileus hemionus sitkensis

Thinhorn sheep ssp. dalli Red s s,b dOvis dalli dalli

Thinhorn sheep ssp. stonei Blue e s,b d,m,k x m m,xOvis dalli stonei

Vancouver Island marmot Red m m wMarmota vancouverensis

Wolverine ssp. luscus Blue m,e,s s,b d,m,k v,d, d,k,h, m,s x,d x,d, x,v, x,d, x,d, d,k,m, w,lGulo gulo luscus k,c c,x m,w k,m m,w m,w w,c,x

Wolverine ssp. vancouverensis Red m h,m c wGulo gulo vancouverensis


b Key to biogeoclimatic units referenced in Table 10a

Biogeoclimatic Biogeoclimatic zone Subzone zone Subzone

c = coastal Douglas-fir d = dry, warmCoastal Douglas-fir Interior Cedar– k = dry-moist, cool

Hemlock m = moist, warmw = wet, coolc = moist, coldx = very wet, cold

h = hyper-maritime v = very dry, coldCoastal Western m = maritime Sub-Boreal Pine– d = dry, coldHemlock s = sub-maritime Spruce k = moist, cool

c = moist, cold

h = hyper-maritime d = dry, hot-warmMountain w = windward maritime Sub-Boreal Spruce k = dry, coolHemlock l = leeward maritime h = moist, hot-warm

c = moist, mild-cool-cold

x = wet, cool

h = very dry hot d = dry, coolBunch Grass w = very dry warm Boreal White and m = moist, warm

Black Spruce k = wet, cool

x = very dry hot s = scrubPonderosa Pine d = dry hot Spruce-Willow-Birch b = forested

x = very dry x = very dryInterior Douglas-fir d = dry Engelmann Spruce– d = dry

m = moist Subalpine Fir m = moistw = wet w = wet

x = very dry, very cold m = above MHMontane Spruce v = very dry, cool Alpine Tundra e = above ESSF

k = dry, cool s = above SWBm = dry, mild

*Status (notes to Table 10a)

• The British Columbia provincial Red List includes any indigenous species or subspecies (taxa)

considered to be extirpated, endangered, or threatened in the province. Extirpated taxa no

longer exist in the wild in British Columbia, but do occur elsewhere. Endangered taxa are fac-

ing imminent extirpation or extinction. Threatened taxa are likely to become endangered if

limiting factors are not reversed. Red-listed taxa include those that have been, or are being,

evaluated for these designations.

• The British Columbia provincial Blue List includes any indigenous species or subspecies (taxa)

considered to be vulnerable in the province. Vulnerable taxa are of special concern because of

characteristics that make them particularly sensitive to human activities or natural events. Blue-

listed taxa are at risk, but are not extirpated, endangered, or threatened (BC MoELP 1999).

The importance of snow avalanche paths as wildlife habitat has been recognized in provincial and regional guidelines for land-use planningand resource development specifically for the protection of grizzly bearhabitat values. For example, the Kootenay–Boundary Land Use Plan() provides for the establishment of Avalanche Path ManagementZones (s) to maintain security cover in old and mature forest adja-cent to snow avalanche paths that provide grizzlybear habitat values (Government of British Colum-bia 1995; Kootenay Inter-Agency ManagementCommittee 1996). Modifications to the plan’s guide-lines have been recommended by Mowat andRamcharita (1999) and Quinn and Phillips (2000).Given the variable nature of snow avalanche pathvegetation communities, the plan provides for a professional biologist (R.P.Bio.) to undertake fieldassessments to determine the capability of individualsnow avalanche paths as grizzly bear habitat. Severalprojects have been undertaken to map and rankavalanche path habitats for grizzly bears at the local and landscape scales (Mowat and Ramcharita1999, p. 13).

Habitat Protection Guidelines

Quinn and Phillips (2000) reviewed the grizzly bearhabitat management recommendations from theKootenay–Boundary Land Use Plan for Tree FarmLicence 14 and recommended a higher level of hab-itat protection. The authors proposed that:• An of 100 m should be established around all

high-quality snow avalanche habitat regardless ofthe distance between paths. Where this habitat islocated only on a portion of the entire path (e.g.,on a runout zone), then only that portion re-quires an .

• Selective harvest retaining 70% of original basalarea may be permitted within s. Harvest op-erations should be scheduled to occur duringperiods of low forage use by grizzly bears (e.g.,spring and early summer operations should beavoided in low-elevation areas).


Willow, groundsel, anda range of herbaceous species often occurin runout zones where the slope becomesgentle and water accumulates. Aspectsare generally neutral. Willow dominatesother vegetation in this group, but oftenoccurs with subalpine fir or alder. Thisvegetation type provides high-quality,abundant browse for ungulates through-out the year (Quinn and Phillips 2000).

Snow avalanche pathdominated by closed-canopy conifershrub from just below the start zone tothe upper runout zone, yielding to analder-fern-meadowrue-valeriantransition, and a cowparsnip-fireweed-grass/sedge meadow. The meadow areais adjacent to a stream and offersvaluable, grizzly bear foraging habitat.The adjacent forest is very old ESSF andprovides excellent security cover andpotential grizzly bedding locations. Mostof the path above the runout zoneprovides little in the way of forage value.The runout portion of this path requires abuffer from forest harvesting (Quinn andPhillips 2000).

• In snow avalanche path complexes where there is less than 200 m between paths, no forest harvesting should occur between the paths.

• Road construction should avoid runout zones as well as high-qualityhabitat higher in the track.

• In areas with extensive high-quality snow avalanche habitat, short periods of disturbance followed by long rotations should be planned.Access should be strictly managed following these periods of operation.

Snow avalanche paths with low-quality habitat (e.g., north-facing, densealder, or continuous low conifer) should not require s.


Click for next page

snow avalanche order form - british columbiathis handbook addresses snow and avalanche phenomena in...

Documents