A gridded forest insect disturbance data set derived from USDA aerial detection surveys and its application in the community land model

Insect outbreaks are major disturbances of forests across North America. Bark beetle and defoliator outbreaks can cause widespread tree mortality, which leads to effects on biogeochemical (e.g., carbon and nitrogen) and biophysical (e.g., albedo, leaf area, transpiration) processes. Such effects modify land-atmosphere exchanges of energy, mass, and momentum, and are important for climate, weather, and atmospheric chemistry.

Aerial detection surveys (ADS), also know as aerial sketch mapping, are conducted by the USDA Forest Service to map forest disturbances caused by insect outbreaks and other disturbance vectors. These data, collected by the respective Forest Service units, have been consolidated for the entire western US since 1997 as part of the Forest Health Monitoring program (FHM). The ADS data are polygons containing an estimate of the area affected, disturbance agent, host tree species, and the number of trees killed within the polygon. We converted the ADS polygon shapefiles into a 1-km2 gridded dataset for use in multiple applications. For each host type and damage agent (e.g., mountain pine beetle damage to lodgepole pine), we calculated the actual percent area affected within each grid cell for each year as well as the number of trees killed within that grid cell.

As an example of the use of this data set, we employed the gridded ADS datasets to investigate the impact of bark beetles on carbon cycling using the National Center for Atmospheric Research (NCAR) Community Land Model with prognostic Carbon and Nitrogen (CLM-CN). CLM-CN is part of NCAR's Community Climate System Model and provides lower boundary conditions of energy, momentum, and scalars (such as water, carbon, and nitrogen) to its coupled Community Atmosphere Model (CAM). We modified the standard version of CLM-CN to include a mechanistic representation of insect outbreaks, and conducted simulations with and without insect mortality from 1997 to 2008.

Preliminary results from point simulations indicate an initial reduction in net primary production (NPP), an increase in heterotrophic respiration (HR), and a decrease in net ecosystem production (NEP) following an insect outbreak. Cumulative HR increased by 14.6 Tg C, and cumulative NEP decreased by 27.8 Tg C (relative to a control run) on 185,272 km2 over 26 years (1980 to 2005) in Washington and Oregon. Simulations for the entire western US will be presented; including maps and statistics of insect mortality from 1997 to 2008 and associated impacts on carbon cycling.