Threat Assessment Using an Integrative Approach of Probabilistic Risk Modeling and Stochastic Programming with an Application to Southern Pine Beetle Outbreaks

Darmawan Prasodjo, Jianbang Gan, and Bruce A. McCarl

Texas A&M University Department of Agriculture Economics (1,3) and Department of Forest Science (2)

One of the major challenges in assessing the impact of natural threats such as weather extremes, wildfire, and pest/disease outbreaks is how to take into account the stochastic nature of these threats in terms of both time and spatial dimensions and to incorporate both direct and indirect impacts.  Probabilistic risk modeling is an effective tool for risk characterization, and a multiregional and multisectoral stochastic programming model is capable of incorporating spatial uncertainty as well as interregional and intersectoral linkages.  Combining the strengths of these two modeling methods appears to be a promising tool for threat assessment.

This paper will first offer an extensive review of probabilistic risk modeling and stochastic programming and their applications in risk analysis and decision-making under uncertainty.  Their advantages and disadvantages as well as suitability for threat assessment will be discussed.  Drawing on this synthesis, we will then attempt to integrate these two modeling approaches and demonstrate its applicability using the case of the southern pine beetle (SPB), the most detrimental insect of southern pines.  Using the probabilistic modeling approach, the risk of SPB outbreaks will be characterized.  This will provide input (the stochastic component) to the Forest and Agricultural Sector Optimization Model--Green-House Gas version (FASOMGH) to simulate the impacts.  FASOMGH is a multiregional, dynamic, price-endogenous, nonlinear programming model of the forestry (including forest and wood processing) and agricultural sectors in the U.S.  Because of its unique features, this model is able to take into account interregional and intersectoral linkages as well as variations in the SPB outbreak risk across time and locations.  The simulation results will in turn offer strategies for mitigating SPB infestations.

This paper will not only describe the modeling approach and demonstrate its applicability in threat assessment, but also offer empirical results on the characterization of SPB outbreak risk, economic impacts of SPB outbreaks under various scenarios (e.g. representing the historical trends and global climate change), and mitigation strategies.  The impact will be measured in terms of output, price, and welfare.  The mitigation options will include species mix, land-use change, salvage harvest, and improved management practices, among others.  The model developed can be applied to assessing the threats of other natural and anthropogenic disturbances, and the empirical findings will provide guidelines for mitigating future SPB damage and supporting healthy forest management.

Native Pests Session - Wednesday Afternoon

corresponding author:

Darmawan Prasodjo
Agriculture Economics
Texas A&M University
113 Jack K. Williams Administration Bldg.
College Station, TX 77843-2142
979-845-8224
dprasodjo@ag.tamu.edu