Simulation modeling to test different harvest regimes for stand conversion and old growth restoration in southern pine

Background/Question/Methods   Managing forests with multiple ages and size classes can promote both economic and ecological services. Interest among landowners in managing for multiple uses under structurally diverse uneven-aged management has increased recently in Florida (USA), partly due to a trend toward public landownership. Increasingly, forest managers and land owners are interested in converting structurally homogenous plantations into more diverse conditions commonly found in uneven-aged stands. The objective of the present study is to assist in this process by developing growth, mortality and recruitment models for slash pine (Pinus elliotti. elliottii) in Florida, and also to use these models to simulate harvesting regimes such as thinnings, single tree selection and group selection. The regimes were evaluated based on their utility in achieving a set of desirable structural characteristics. Using 70 permanent plots in natural and plantation forests, we developed diameter and height growth, ingrowth and mortality models. We used a mixed modeling approach, comparing linear and non-linear mixed models. Models were evaluated with information criteria, bias, and precision measures. Due to the unavailability of an independent dataset, we validated the models using a leave one out cross validation.

Results/Conclusions   Our results indicated that the mixed models produced better estimates of growth. Mortality and ingrowth were best estimated with logistic regression, with number recruited modeled linearly. Results suggested that depending on objectives, single tree and group selection should be used for achieving the desired structural characteristics.