Modeling slash pine mortality rates incorporating responses to silvicultural treatments and fusiform rust infection rates

Abstract

The evaluation of stand survival at any age is an essential task that allows foresters to estimate stand dynamics and ultimately, the value of a forest stand. Therefore, any growth and yield system developed with the aim of predicting or projecting standing value requires precise estimates of the number of trees surviving at any point in time. When describing survival through modeling the mortality rate, differential equations using height increments rather than time increments have been shown to improve the overall fit of survival models. Using a long-term data set of slash pine plantations (Pinus elliottii Engelm.) in which silvicultural treatments (i.e., bedding and complete vegetation control) were applied during the establishment phase, and fusiform rust (Cronartium quercuum Berk.) infection rates were recorded, a survival/mortality model of this nature was constructed. Height increments were taken from a proposed dominant height model that included explicit treatment effects. In addition, the proportion of trees infected with fusiform rust at an age of 5 years was added as a predictor describing the mortality rate. Our results show that stand survival is better described by a model in which time increments are used rather than height increments, and although silvicultural treatments were essential for modeling dominant height, mortality was not greatly affected by these treatments and therefore, no additional parameter modifiers associated to these treatments were needed. On the other hand, the inclusion of average fusiform rust infections was essential to describe mortality rates in these stands, with higher infection rates associated to higher mortality rates.