WWETAC Projects

Project Title: Probabilistic risk models for multiple disturbances: an example of forest insects and wildfires

Principal Investigators: Haiganoush K. Preisler, USDA Forest Service, Pacific SW Research Station, Albany, CA; Alan A. Ager, WWETAC, Pacific NW Research Station, Prineville, OR; Jane L. Hayes, USDA Forest Service, Pacific NW Research Station, La Grande, OR; Jeffrey Hicke, WWETAC, USDA Forest Service, Univ. of Idaho.


Key Issues/Problems Addressed:

A better understanding of multiple forest disturbances and their interactions (i.e., fire-insect) is required to aid in planning for climate change impacts. A framework to estimate forest disturbance probabilities in the presence of multiple stressors is needed.

Setting and Approach:

This study reviewed historical data from 1980 to 2004 on fire occurrence and forest insect outbreaks collected in Oregon and Washington. Spatial data on insect activity was obtained from aerial sketch maps created by the Forest Service Forest Health Protection program. Federal wildfire data, including information on the date, location, and size of the fire, was retrieved from the Desert Research Institute (DRI). Average monthly temperature, precipitation, and other climate variables were derived from extrapolating station data and results from climate suitability models that incorporate winter and year-round temperature effects as well as drought stress on host trees.

Key Findings:

  • The developed framework is especially amenable to quantifying uncertainties from multiple sources, including stochastic disturbance, sampling and measurement errors, approximate models, and missing predictor variables.
  • Stands (one km2) with previous year fire sizes of 750 ha or less exhibited an increase in the odds of bark beetle attacks with increasing fire size while stands with previous year fires sizes greater than 1000 ha exhibited a decrease in the odds with increasing fire size.  (Stands with weakened but not dead trees were more prone to bark beetle attacks than stands with mostly dead trees).
  • Locations with 3- to 6-year-old spruce budworm damage greater than 100 ha seem to decrease risk of large crown fires.


The developed framework provides an objective tool and methods for modeling complex processes and estimating associated probability maps for competing risks (multiple disturbances).


Preisler, H.K., A.A. Ager and J.L. Hayes. 2010. Probabilistic risk models for multiple disturbances: an example of forest insects and wildfires. Pages 371-379 in Pye, J. M.; H.M. Rauscher, Y. Sands, D.C. Lee, J.S. Beatty technical editors, Advances in threat assessment and their application to forest and rangeland management. USDA Forest Service General Technical Report PNW-GTR-802. Portland, OR. (PDF, 387 KB)

McMahan, A., A.A. Ager, H. Maffei, J.L. Hayes and E.L. Smith. 2008. Modeling bark beetles and fuels on landscapes: A dem of ArcFuels and a disc of possible model enhancements. Pages 40-52 in R.N. Havis, and N.I. Crookston compilers. 2008. Third Forest Vegetation Simulator Conference; 2007 February 13–15; Fort Collins, CO. Pro­ceedings USDA Forest Service RMRS-P-54. Fort Collins, CO.

WWETAC Project ID:  FY07AA34