Navigating uncertainty: timberland management with variable fertilization responses

Abstract Forest productivity is a key determinant of timberland financial returns. To increase timber production, landowners and forest managers often have to decide on fertilization levels early in a plantation’s establishment. However, the biological response to fertilization is uncertain and does not guarantee increased productivity. In this study, we developed a stochastic harvest scheduling model at the forest level to account for these uncertainties. Using inventory plots and fertilization trial data across the Southeastern US, we constructed a two-stage stochastic programming model using timber production as the terminal stage. Our findings highlight the need to consider evolving market dynamics, including land conversion, in forest management decisions. Future research should explore multi-stage stochastic programming, site-specific silviculture, and financial factors like timber price volatility to refine growth and yield predictions. By extending the application of stochastic optimization, this study provides forest managers with a more nuanced understanding of silvicultural risks and uncertainties, helping them make informed decisions that balance economic returns with ecological sustainability.