WWETAC Projects
Project Title: Application of wildfire risk analysis to examine carbon flux from fuel treatments
Principal Investigators: Alan A. Ager, Western Wildland Environmental Threat Assessment Center; Jim Cathcart, Oregon Department of Forestry
Collaborators: Mark A. Finney, USDA Forest Service, Rocky Mountain Research Station; Andrew McMahan, USDA Forest Service, Forest Health Technology Enterprise Team; Fremont-Winema National Forest
Status: Ongoing
E-mail Contact: Jim Cathcart, jcathcart[at]odf.state.or.us
Key Issue: A number of recent studies have suggested short- and long-term carbon benefits from fuel treatments (e.g. Hurteau et al.2008, Finkral and Evans 2008). Specifically, stand-level models show that that fuel treatments reduce potential fire severity compared to non-treated stands, and therefore reduce carbon lost from emissions. It is assumed that a portion of the carbon removed by treatments is fixed in building materials etc. While informative, these analyses have the following limitations:
1. Wildfire probability is assumed to be 1.0, and thus carbon benefits are conditional on a fire burning the stand. No fire, no benefit. Actual wildfire probabilities from long term records for the western US average about 0.004 (1960 – 2006).
2. The carbon benefits outside the treatment units from reduced burn probability and severity are not considered. There is ample empirical and experimental evidence that offsite treatment effects are substantial for both likelihood and intensity.
Goals and Objectives: This study concerns the development and testing of a methodology for determining expected landscape-scale carbon dioxide emission offsets from fuels management activities. The study is part of the West Coast Regional Sequestration Partnership (WESTCARB) and is developing the methods on the 80,000 ha Drews Watershed of the Fremont-Winema National Forests in Lake County, Oregon.
Description of Work: The expected avoided carbon emissions from a hypothetical fuel treatment project in the study area is being calculated as follows:
(1) Estimate p(Fi), the probability of every stand burning at a given fire intensity i
(2) Estimate the loss of carbon for a given fire intensity, ΔCi. This is accomplished by burning each stand in FVS while varying the flame length.
(3) For each pixel, calculate Expected standing carbon, C – Σ(p(Fi) * ΔCi), where C is the carbon before the fire, and Σ(p(Fi) * ΔCi) is teh expected carbon given a random ignition on the landscape and a severe fire event.
(4) Simulate fuel treatments and re-calculate values described in #1, #2, and #3 above for the treated landscape. In this project we simulated fuel treatments on about 15% of the forested area.
(5) Expected carbon difference after fuel treatments, E(ΔC), is E(Cnotreat) - E(Ctreat).
A comparison of carbon between a typical treated and untreated stand after a wildfire (right) often shows a net carbon loss from treatment, largely because decomposition of burned trees is slow and non-merchantable removals as part of a fuel treatment constitute an emission. Over time, depending on decomposition rates in the dead tress and rates of regeneration the carbon loss could be reversed.
For the watershed as a whole, the total expected carbon was 3,144,622 tons for the treated landscape and 3,353,464 tons for the untreated. Thus the treatments resulted in a net loss of 208,842 tons over the 160,000 acres. The fuel treatment was applied on about 15% of the forested lands within the watershed.
Examination of carbon balance inside and outside treatments showed that the loss of carbon from treatment was confined to within the treatment areas. The image below shows treatment areas as hatched outline. Areas in red indicate carbon loss. Areas in blue and dark green show net carbon benefit. The image suggests extensive carbon benefits outside the treatment area, but little within.
Deliverables and Citation:
Cathcart, J., Ager, A.A., McMahan, A.M, Finney, M.A. (in press). Carbon offsets from fuel treatments using a risk analysis. Proceedings of the 2009 National Silviculture Workshop, Boise, ID. June 15-18. Rocky Mountain Research Station, General Technical Report RMRS-P-X
Project ID: FY08AA54
