The LSim forest landscape model: A case study of application on the Deschutes National Forest

What is the LSim model? 

Forest landscape models (FLMs) are important tools used to address a wide range of forest management policy tradeoffs on public and private forests. Several recent studies have used FLMs to examine the effects of forest and fuels management on future wildfire activity, carbon, water yield, resiliency, and other forest metrics. Studying longer-term (e.g. > 20 years) dynamics between management activities and natural disturbances can reveal ecosystem tipping points, feedbacks, and unintended consequences of management that are not possible to observe otherwise. Most recently, applications of FLMs have provided insights on the potential effects of management actions on future fire and forest composition under a range of climate change scenarios. Many of these studies in the U.S. have been conducted on portions of the large (76 million ha) national forest network where wildfires are increasingly impacting ecosystem services and burning into adjacent developed areas. 

Researchers developed the new forest landscape model called LSim, which integrates the Forest Vegetation Simulator (FVS) together with the large-fire simulation model FSim. The resulting LSim is a strategic forest landscape model that enables research on long-term management and wildfire feedbacks. It can simulate spatially coordinated forest management over time under a background of large stochastic wildfires with models that have undergone decades of field application. This is in contrast to other FLMs that have yet to be used to guide site specific management activities as part of forest and fuels management on national forests. The LSim model provides a platform to simulate detailed prescriptions developed by silviculturists in the field as part of forest landscape management projects. 

Case Study: LSim Application on Deschutes National Forest 

In this study, researchers applied the LSim model to the Deschutes National Forest in central Oregon to demonstrate how LSim can be used to examine a wide range of wildfire and forest management issues on western U.S. landscapes. Specifically, researchers used the LSim model to understand how accelerated forest restoration might affect future wildfire area burned, fire severity, fire exposure to the wildland urban interface (WUI), and commercial timber production. Researcher are also applying the model to national forests in the Blue Mountains in eastern Oregon to study the effect of climate change on future fire. A third study area is in northern Arizona examining alternative restoration and fire management strategies through time on fire behavior.  

Key Findings

• Restoration treatments over time had a large effect on fire severity, reducing potential flame length by 45% for the study area within the first 20 years, whereas reductions in area burned were relatively small.
• Although wildfire contributed to reducing flame length over time, area burned was only 16% of the total disturbed area (managed and burned with prescribed fire) under the 3x management intensity.
• Interactions among spatial treatment scenarios and treatment intensities were minimal, although inter-annual and among-replicate variability was extreme, with the former coefficient of variation in burned area exceeding 200%.
• Researchers also observed simulated fires that exceeded four times the historically recorded fire size. Fire regime variability has manifold significance since very large fires can homogenize fuels and eliminate clumpy stand structure that historically reduced fire size and maintained landscape resiliency. 
• Finally, researchers discuss specific research needs to better understand future wildfire activity and the relative influence of climate, fuels, fire feedbacks, and management to achieve fire resiliency goals on western U.S. fire frequent forests.

Diagram showing the major components of the LSim model: large fire event simulator (FSim) coupled with the Forest Vegetation Simulator (FVS). Fire effects are calculated using the Fire and Fuels Extension to FVS (FFE) using fire behavior (flame length) calculated by FSim.

Detailed diagram of the LSim model components showing the functionality in the two main submodels, the Forest Vegetation Simulator (FVS) and FSim, and the data translator within the parent LSim model. For each simulation cycle, FVS loops through each stand in the landscape to affect fire mortality, growth, management, and natural mortality. The resulting landscape fuels are translated to the binary format read by FSim. FSim then loops through each day in the fire season and simulates fires based on daily predicted weather. At the end of the fire season the wildfire intensity grids are read by FVS and tree mortality is predicted using the functionality in the FVS Fire and Fuels Extension.

Variability in area burned (ha) among five simulated replicates for each treatment priority and under the mid-range treatment intensity (2×; 72,000 ha per 5-year cycle). See Table 2 for scenario descriptions. Also shown is the maximum and mean historical area burned per year between 1990 and 2012.

Mean annual area burned (as a percentage of the land base) across 30 replicates for different treatment priorities and treatment levels on manageable and non-manageable lands. Grey bands correspond to 95% confidence intervals.