3-PG simulations of young ponderosa pine plantations under varied management intensity: Why do they grow so differently?

Models can be powerful tools for estimating forest productivity and guiding forest management, but their credibility and complexity are often at issue for forest managers. We parameterized a process-based forest growth model, 3-PG (Physiological Principles Predicting Growth), to simulate growth of ponderosa pine (Pinus ponderosa) plantations in Northern California. We used data collected from the “Garden of Eden” study, which was established in the 1980s to determine the effect of silvicultural treatments on plantation growth. We picked three sites representing a gradient of water availability and site productivity to run 3-PG. We modified the original linear canopy closure function to a power curve to capture observed stand dynamics in situ. We also added new functions to estimate the leaf area index and transpiration of the trees’ understory competitors. These new functions restricted shrub growth with light intensity and assumed a fix ratio of shrub/tree transpiration per leaf area index. A δ13C submodel, which estimated the ratio of stable carbon isotopes (δ13C) in plant tissue, played a key role in assigning values to gas-exchange parameters in the model. The resulting parameter values were similar to those fitted using sap flux. We replaced the original age modifier with tree-height based functions to reflect the decreased forest productivity as trees grew taller; tree height drove the change of maximum canopy conductance and its responsiveness to water vapor pressure deficit in the new functions. Some key parameters differed among sites, including quantum yield, maximum canopy conductance, and leaf allocation. The model successfully simulated the tree growth responses to fertilization and vegetation control at all three sites. The temporal variation of simulated shrub leaf area index was similar to the observed variation in shrub cover. These results help us to understand forest-growth responses to fertilizer and vegetation control, identify key tree and site parameters, and provide tuned model parameterizations that can predict the results of management alternatives in a changing climate.