Modeling stand-level mortality of loblolly pine (Pinus taeda L.) using stand, climate, and soil variables

Abstract

Accurate prediction of mortality is an important component of forest growth-and-yield systems, yet mortality remains one of the least understood components of the system. Remeasurement data collected from permanent sample plots established in 1980/1981 across the natural range of loblolly pine in the Coastal Plains and Piedmont were used. Biophysical data for plot locations for years 1980 –2003 were obtained from the Oak Ridge National Laboratory. The main objective of this study was to develop models for predicting stand-level mortality of loblolly pine plantations using stand characteristics and climate and soil variables. Models were constructed using two modeling approaches. In the first approach, mortality functions for directly predicting tree number reduction were developed using an algebraic difference equation method. In the second approach, a two-step modeling strategy was used. In the first step, a model predicting the probability of tree death occurring in a stand over a measurement period was developed; in the second step, a function that estimates the reduction in tree number was fitted. Performance of all the approaches was compared using leave-one-cluster-out cross-validation. When climate and soil variables were used in the model, the direct prediction approach performed the best.