Regional validation and improved parameterization of the 3-PG model for Pinus taeda stands

The forest simulation model, 3-PG, has the capability to estimate the effects of climate, site and management practices on many stand attributes using easily available data. The model, once calibrated, has been widely applied as a useful tool for estimating growth of forest species in many countries. Currently, there is an increasing interest in estimating biomass and assessing the potential impact of climate change on loblolly pine (Pinus taeda L.), the most important commercial tree species in the southeastern U.S. This paper reports a new set of 3-PG parameter estimates for loblolly pine, and describe new methodologies to determine important estimates. Using data from the literature and long-term productivity studies, we parameterized 3-PG for loblolly pine stands, and developed new functions for estimating NPP allocation dynamics, biomass pools at variable starting ages, canopy cover dynamics, effects of frost on production, density-independent and density-dependent tree mortality and the fertility rating. The model was tested against data from replicated experimental measurement plots covering a wide range of stand characteristics, distributed across the southeastern U.S. and also beyond the natural range of the species, using stands in Uruguay, South America. We used the largest validation dataset for 3-PG, and the most geographically extensive within and beyond a species' native range. Comparison of modeled to measured data showed robust agreement across the natural range in the U.S., as well as in South America, where the species is grown as an exotic. Across all tested sites, estimations of survival, basal area, height, quadratic mean diameter, bole volume and above-ground biomass agreed well with measured values, with R2 values ranging between 0.71 for bole volume, and 0.95 for survival. The levels of bias were small and generally less than 13%. LAI estimations performed well, predicting monthly values within the range of observed LAI. The results provided strong evidence that 3-PG could be applied over a wide geographical range using one set of parameters for loblolly pine. The model can also be applied to estimate the impact of climate change on stands growing across a wide range of ages and stand characteristics.