SNAMP Publication #2: Simulating fire and forest dynamics for a landscape fuel treatment project in the Sierra Nevada
Collins, B.M., S.L. Stephens, G.B. Roller, and J.J. Battles
- A fire behavior model was run to assess the ability of planned forest thinning treatments to reduce fire spread and intensity in the Tahoe National Forest.
- The model used conditional burn probabilities to assess how effective these treatments will be at reducing landscape-level fire behavior.
- The model simulated varying intensities of thinning treatments, using different upper diameter limits, and their effects at 10-year intervals, for 30 years into the future.
- All the simulated thinning intensities effectively reduced burn probabilities for approximately 20 years following implementation.
- Residual forest stand structure varied with different thinning scenarios, but likelihood of burning did not vary with different thinning scenarios.
- Results suggest that at the landscape scale the effectiveness of fuel reduction treatments relies more on treating surface fuels and thinning ladder fuels than on thinning trees based on diameter limits.
The Model Experiment:
This article reports on a fire behavior modeling experiment focused on the Last Chance project area within the Tahoe National Forest in the northern Sierra Nevada where forest fuel reduction treatments are planned. The planned treatments were evaluated for their efficacy at reducing landscape-level fire behavior based on conditional burn probabilities. Conditional burn probability is the probability of a pixel burning under specified fire weather, given that a fire ignites in the analysis area. In a fire behavior model, conditional burn probabilities are computed by dividing the total number of times each pixel burned by the total number of fires simulated in the model. In this model, pixels were 60 meters in size, and fires were simulated a total of 5,000 times using FlamMap software.
In the model runs, the proposed fuel treatments were varied based on the upper limit of tree diameter removed. Researchers investigated three different diameter-limited thinning scenarios: 30.5 cm (12 in.), 50.8 cm (20 in.), and 76.2 cm (30 in.). Thinning scenarios were evaluated for differences in residual forest stand structure and modeled landscape-scale burn probabilities. Model runs simulated a time period from treatment implementation to 30 years in the future. The model was also run for a scenario where no treatment was implemented.
The planned fuel treatments reduced conditional burn probabilities substantially across the landscape relative to the scenario with no simulated treatments. The reduction in burn probability resulting from the treatments in the model was still evident approximately 20 years after implementation of the simulated treatment. Different thinning scenarios (based on different tree removal diameter limits) used in the model resulted in different residual forest stand structure, however, there was no real difference in modeled landscape-level burn probabilities detected. Several environmental inputs to the model were varied to ensure the outcome was robust, including the fuel model, vegetation in-growth and regeneration over time, and wind speed.
1. Research findings predict that the Last Chance project, as planned, will be effective in reducing modeled burn probabilities.
2. The Last Chance project does not conform to the idealized chevron pattern of a strategically placed landscape area treatment (SPLAT), but is still effective at reducing the intensity of fire behavior across the landscape, even in untreated areas.
3. Although differences in residual forest structure among the different thinning scenarios were predicted, no clear differences in the probability of fires with flame length greater than two meters was predicted. This suggests that the diameter of trees removed has less effect on burn probabilities than removal of surface and ladder fuels.
4. This research was done in a modeled environment, and as such the analysis has limitations. Models may under-represent crown fire propagation and spotting, and thus may not be able to capture differences in the reduction of crown fire potential among thinning scenarios.
5. Predicting changes in fire intensity and subsequent effects through modeling of post fuel treatment project conditions improves the ability to evaluate whether or not a landscape fuel treatment may achieve its objectives.
Full Reference: Collins, B.M., S.L. Stephens, G.B. Roller, and J.J. Battles. 2011. Simulating fire and forest dynamics for a landscape fuel treatment project in the Sierra Nevada. Forest Science 57: 77-88.
The full paper is available at: http://cnr.berkeley.edu/stephens-lab/Articles.htm
For more information about the SNAMP project and the FFEH team, please see: http://snamp.cnr.berkeley.edu/teams/fire-forest-health