A New Method to Calibrate Cardinal Temperatures for Eucalyptus Plantation †

Abstract

Developing a good understanding of the interactions between forest plantation growth and climate is essential for predicting the impact of climate change on terrestrial ecosystems and for assessing the adaptation and vulnerability of tree species. One such interaction, the response in growth rate of a forest stand to changes in temperature, may be described mathematically. Some models that run on monthly time steps assume a yearly optimum, minimum, and maximum temperature for simplicity, which may not represent well to actual forest growth. Here, we developed a finer-resolution methodology that encompasses monthly growth rates and temperature limits to calibrate the parameters for an envelope curve in Eucalyptus plantations in South America. Several polynomial curves were tested to determine temperature patterns, and their yearly tree growth patterns demonstrated that responses to temperature differed by as much as 10 °C among seasons. The best curve was a second-degree polynomial curve, whose extreme values indicated the optimum temperature and whose real roots limited the minimum and maximum temperatures for growth. This polynomial was fitted every month to describe yearly changes in optimum, maximum, and minimum temperatures. When fitted to annual data, it determined 7 °C, 19 °C, and 31 °C as the minimum, optimum, and maximum temperatures for tree growth, respectively. The monthly model predictions indicated that the minimum, optimum, and maximum temperatures lay between 8 °C and 16 °C, 18 °C and 22 °C, and 27 °C and 30 °C, respectively. These monthly temperature ranges can improve the estimation of productivity in process-based models. Our results contribute to the understanding of tree growth dynamics and its relationship to changes in temperature. Accurate ranges of temperature can be used to improve productivity predictions in new expanding planting regions with no previous information or to suggest a regionalization for potential species.