Toward more robust projections of forest landscape dynamics under novel environmental conditions: Embedding PnET within LANDIS-II

Ecological models built on phenomenological relationships and behavior of the past may not be robust under novel conditions of the future because global changes are producing environmental conditions that forests have not experienced historically. We developed a new succession extension for the LANDIS-II forest landscape model, PnET-Succession, to simulate forest growth and succession using physiological first principles. PnET-Succession integrates the tree physiology model PnET-II with the existing LANDIS-II Biomass Succession extension. PnET-Succession simulates the competition of tree species cohorts for water and light as a function of photosynthetic processes driven by foliar nitrogen. Competition for water is simulated on each grid cell through a dynamic soil-water balance that receives precipitation and loses water through runoff, consumption in photosynthesis, and evapotranspiration. Competition for light is modeled by tracking solar radiation through canopy layers according to a standard Beer-Lambert formula. PnET-Succession requires average monthly photosynthetically active radiation, atmospheric CO2 concentration, temperature and precipitation as inputs. The new extension also dynamically calculates species establishment probabilities in each time step as a function of water and radiation stress. We calibrated PnET-Succession to biomass and LAI measurements from the Duke Experimental Forest in North Carolina (USA) and tested the calibrated model against data from the Green Ridge State Forest in Maryland. The new extension shows considerable promise for studying forest response to climate change, including changes in carbon stocks. (C) 2014 Elsevier B.V. All rights reserved.