Landscape Analysis of Fuel Treatment Longevity and Effectiveness in the 2006 Tripod Complex Fires

Project Description

The 2006 Tripod Complex fires, which burned more than 70,000 ha and involved more than 380 past harvest and fuel treatment units, offered a relatively unique opportunity to assess fuel treatment efficacy under extreme fire weather conditions. A secondary objective was to evaluate other drivers of fire severity such as landform, weather, vegetation, and past disturbances, including wildfires and a recent mountain pine beetle outbreak. We evaluated drivers of burn severity in two study areas that are centered on early progressions of the wildfire complex.

A common interpretation of weather-driven fire events is that bottom-up controls, including fuels and topography, are superseded by climatic factors and are relatively unimportant. However, even during extreme weather, landform, vegetation and fuels clearly influenced patterns of fire severity and spread in the Tripod Complex fires. Fuel treatments that included recent prescribed burning of surface fuels were particularly effective at mitigating fire severity. In contrast, units that were mechanically thinned from below and those with sanitation cuts in which small trees were cut and piled tended to burn at moderate to high severity.

Purpose and Scope

Under a warming climate and increased fire hazard, managers of dry forests face numerous challenges in strategizing for and implementing fuel reduction treatments. Existing studies of fuel treatments in dry forests generally agreed that mechanical thinning followed by prescribed burning is the most effective at reducing surface and crown fuels and increasing forest resilience to wildland fire. However, little was known about the duration of treatment effectiveness and if treatments can remain effective in extreme fire events.

Methods

• Validation plots were sampled across a range of severity classes during the summers of 2007 and 2009. Forty-four plots were collected in the summer of 2007.
• We supplemented this dataset with an additional 55 plots in the summer of 2009 to ensure adequate representation in each burn severity class. Needles on scorched trees were still present in 2009 and allowed for comparable composite burn index (CBI) observations.
• Prior to analysis, we assembled data layers of predictor variables. A geospatial treatment layer, including harvest type and date and prescribed burn type and date, was compiled within the Tripod Complex perimeter and verified using hard-copy records.
• Models were constructed to predict severity based on progression order, landform variables, heat load index, weather variables, vegetation variables, and past fuel treatment.

Implementation

• Differenced normalized burn ratio (dNBR) and relative differenced normalized burn ration (RdNBR) images used in this analysis were calculated based on virtually cloud-free, pre-and post-burn Landsat TM images taken one year before and one year after the 2006 Tripod Complex fires.
• Composite burn index (CBI) data were collected to validate that dNBR and RdNBR represented burn severity in the field and to compare the two indices.

Key Findings

1. Fuel treatments and past wildfires mitigated fire severity under extreme fire weather.
2. Our ability to predict fire severity is limited by a number of missing variables that are generally unavailable for large fire events.
3. Fire severity, as represented by dNBR values, is highest at mid- to high-elevations between 1500 and 2000 m in both study areas.
4. MaxTemp and AvgWind are important predictors in some models, suggesting that broadly summarized weather by progression interval was still able to represent finer-scale fire-weather relationships
5. Vegetation cover and type are both important predictors of burn severity