Forest Investment Account (FIA) - Forest Science Program
FIA Project Y103004

    Thresholds for post-wildfire flood, erosion, and mass wasting processes
Project lead: David Scott (University of British Columbia)
Contributing Authors: Covert, S. Ashley; Jordan, Peter; Curran, Mike P.
Subject: Forest Investment Account (FIA), British Columbia
Series: Forest Investment Account (FIA) - Forest Science Program
Flooding, erosion, and mass movement following wildfires is an example of a threshold for watershed hydrologic function being exceeded, which has significant risk implications for downslope and downstream values, and for management of forest land affected by wildfire.

Changes to the soil following fire, such as water repellency and loss of forest floor, can create a threshold in hydrologic response, whereby overland flow generates exceptionally high peak runoff on a watershed scale. Normally, forest soils rarely or never generate extensive overland flow, and hydrologic response is dominated by shallow and deep subsurface flow. Widespread overland flow often causes severe erosion, especially where the forest floor and soil structure have been altered. In steep areas, debris flows and landslides can result.

In British Columbia, large-scale flooding and erosion had not been documented following wildfires until after the 2003 fire season. This contrasts with the western US, where severe flooding and erosion are common after wildfires, and where considerable resources are devoted to assessing post-wildfire risks, and applying treatments to affected areas. Climate change may increase the the likelihood of these events in BC.

Following the severe 2003 fire season, major erosion and mass movement incidents were observed on 5 fires: the October 2003 flood/erosion event in Kelowna; the August 2004 debris flows on Kuskonook Creek near Creston, which destroyed several houses; the August 2004 debris flood event on Lamb Creek near Cranbrook; the July 2004 debris flows/floods at the Cedar Hills fire near Falkland; and in October 2005, 15 debris flows and landslides occurred in the Mt Ingersoll fire near Burton. Some of these events were documented in Jordan et al (2004 and 2006) [1,2], and several involved substantial property damage and risks to the public.

Since then, BC Forest Service staff have worked on developing an interim procedure for post-wildfire risk assessment, and a number of critical knowledge gaps have been identified. (See “Objectives” below.)

The fundamental question addressed by this research project is: What vegetation and soil burn severity and extent is necessary to generate sufficient overland flow in a watershed to cause flooding and erosion, and what are the roles of water repellent soils and forest floor alteration in these processes?

The proposed 3-year project seeks to address these knowledge gaps by a research program which emphasizes field data collection and analysis, and which integrates the methods of soil science, hydrology, and geomorphology. Work will be done by MOFR staff scientists, with cooperation from our partners at UBC and Selkirk College, and MOFR Protection Branch fire centres. Specific tasks to be undertaken include:
- Further field investigation of the events which followed the 2003 fires (and to a much lesser extent, 2004-2005 fires), to better document and analyse the processes which caused the flooding, landslides, and debris flows.
- In new fires which occur during the study: establish plots to measure relevant soil parameters, and measure erosion, in burned areas of various severity classes, and reference plots in adjacent unburned areas.
- Refine field procedures for measuring burn severity, water repellency, and forest floor properties at burned and unburned sites, in terms of the hydrologic function and antecedent moisture content of the mineral soil, duff, and litter layers.
- Rainfall simulation experiments to investigate infiltration and overland flow processes in the study plots.
- Monitor rainfall, runoff, erosion, and sediment transport, in small instrumented watersheds.
- Monitor the effectiveness of treatments to reduce erosion, which were applied on two of the 2007 fires, and measure erosion and runoff in treated and untreated areas.
- Data analysis will be done using appropriate statistical, laboratory, and GIS methods, with which the MOFR research team has good in-house resources and experience.

This field work program depends on wildfires which occur during the course of the study. In most years, at least several high-risk fires occur in the region which are likely to be suitable for the study, and are of interest because of risks to downslope values.

Work in year 1 (2007 field season) focused on establishing sites in new fires, and on work in earlier burns. Field methods and instrumentation were tested and applied on the 2007 Sitkum, Springer, and Pend d’Oreille fires. Additional study sites were established in year 2, and measurements will continue at all sites in year 3.

A related study by D Scott of UBC has been completed. This was focussed on assessing rehabilitation methods, and will complement the present study as it collected soil and hydrologic data from the 2003 fires. We plan to conduct cooperative work on the 2007 fires in year 3. There are at least two substantial research programs working on this subject in the US. A USFS team led by P Robichaud is studying the effectiveness of rehabilitation treatments applied to mitigate post-fire hazards [3,4], and a USGS team led by S Cannon is studying post-fire debris flows [5,6]. We plan to expand our existing co-operation with both these groups, and our intent is to complement rather than duplicate their work. Most of their research and most documented erosion incidents have been in more southern regions where the climate (drier, with higher-intensity rainfall) and soils (mostly non-glacial) are different from BC (although some research sites have been in areas such as north Idaho which are similar to southern BC). An expected outcome of this project will be improved understanding of how post-fire hydrologic and geomorphic processes differ in areas of contrasting ecosystems and climate.

Understanding of post-fire changes in hydrology, and the risks presented by these changes to communities and infrastructure, is much more advanced in the US than in Canada. In the US, substantial resources are invested in research, and in risk assessment and in mitigation of post-fire risks. In Canada, there is almost no such investment. Although the proposed research project is modest compared to the US research programs, we expect it will complement the US work by improving the awareness and understanding of post-fire risks in BC, and ultimately to make a contribution to improved public safety.

The forest management issues addressed by this project include:
- Risk assessment of areas downslope/downstream from wildfires that could be affected by post-wildfire flooding and erosion.
- Treatment of burned areas to mitigate hydrologic hazards, and planning of reforestation following wildfire.
- Climate change and mountain pine beetle infestation, which are likely to increase the potential for higher-severity wildfire and consequent hydrologic impacts.
- Planning and practice of salvage logging following wildfires, in terms of its role as a treatment which might mitigate or exacerbate post-wildfire hydrologic hazards.

(No major changes have been made from the original project description. Minor updates have been made to the description above, to reflect work done in 2007 and 2008.)

(Bracketed numbers in sections 3 and 5 indicate references in the documents attached to the original proposal and the year 2 work plan.)
Related projects:  FSP_Y081004FSP_Y092004


Journal article: Environ. Eng. Geoscience 15(4): 217-234. Debris Flows and Floods Following the 2003 Wildfires in Southern British Columbia (8.7Mb)
Executive Summary (97Kb)
Presentation to EGU: Post-Wildfire Erosion and Mass Movement in British Columbia: Site-Scale Soil Changes and Catchment-Scale Processes (5.8Mb)
Geophysical Research Abstracts Vol. 12, EGU2010-PREVIEW, 2010 (40Kb)
Geophysical Research Abstracts Vol. 12, EGU2010-PREVIEW, 2010 (39Kb)
EGU Poster: Rainfall Simulator (2.3Mb)
EGU Poster: Risk analysis procedure (11.7Mb)
Geophysical Research Abstracts Vol. 12, EGU2010-14991, 2010 (36Kb)
Presentation : Post-Wildfire Mass Movement and Erosion Events in BC 2003-2009 (12.0Mb)
Presentation: Effects of Burn Severity on Soil Characteristics and Erosion Response in South Western BC (5.0Mb)
Presentation: Mitigative Treatments for Hillslope Soils on the 2007 Springer Fire (14.7Mb)
Forrex article: Streamline 13(1): 5-9. A Portable Rainfall Simulator: Techniques for Understanding the Effects of Rainfall on Soil Erodibility (0.6Mb)

To view PDF documents you need Adobe Acrobat Reader, available free from the Adobe Web Site.

Updated August 19, 2010 

Search for other  FIA reports or other Ministry of Forests and Range publications.

Please direct questions or comments regarding publications to