The forest vegetation simulator: A review of its structure, content, and applications

Abstract

The Forest Vegetation Simulator (FVS) is a distance-independent, individual-tree forest growth model widely used in the United States to support management decisionmaking. Stands are the basic projection unit, but the spatial scope can be many thousands of stands. The temporal scope is several hundred years at a resolution of 5–10 years. Projections start with a summary of current conditions evident in the input inventory data. FVS contains a self-calibration feature that uses measured growth rates to modify predictions for local conditions. Component models predict the growth and mortality of individual trees, and extensions to the base model represent disturbance agents including insects, pathogens, and fire. The component models differ depending on the geographic region represented by regionally specific model variants. The differences are due to data availability and the applicability of existing models. The model supports specification of management rules in the input, such as thinning if density is too high. The rules can be extended to represent other factors. For example, the effect of climate change on stand development by entering rules that specify how growth and mortality will change in response to changing climate. Applications range from development of silvicultural prescription for single stands to landscape and large regional assessments. Key issues addressed with FVS include forest development, wildlife habitat, pest outbreaks, and fuels management. The predictions are used to gain insights into how forested environments will respond to alternative management actions. Broad-scale forest management policies have been studied with FVS. For the 30 years since the model was initially introduced, the development team has anticipated and provided needed enhancements and maintained a commitment to working with and training users. The existence of an adequate user interface and the continued use of the original programming language are often overlooked factors for the success of this model. Future work will focus on improving FVS by adopting recent biometric techniques and including new information linking geomorphology to mortality and growth. Extending the model to more closely represent biophysical processes and adapting the model so that it is more relevant to management questions related to predicted climate change are also foci. Providing ways to dynamically link FVS to other models is our current strategy for providing major new capabilities.