Improving Process-Level Understanding of the Factors Underlying Long-Term Trends and Year to Year Variability in Carbon Sequestration of Northeastern Forests

The goal of the project is, using inverse data-model fusion techniques, to combine process-based modeling frameworks (a simple ecosystem C model, DALEC+, and a terrestrial biosphere model, ED2) with Monte Carlo inverse modeling techniques and rigorous model selection criteria based on Kullback-Leibler information theory to evaluate the degree to which each of the following competing hypotheses measurably contributes to explaining the observed interannual variability and multi-year trends in CO2 uptake and release: (i) gradual shifts in the environmental drivers of forest productivity are resulting in conditions that increase photosynthetic uptake and/or reduce respiratory losses; (ii) lengthening of the growing season (earlier spring onset or delayed autumn senescence) has increased C uptake and storage during the “shoulder” seasons; (iii) rising levels of atmospheric CO2 are resulting in “CO2 fertilization,” which is increasing instantaneous photosynthetic capacity and annual gross productivity; and (iv) successional dynamics and recovery from disturbance are resulting in demographic shifts in forest composition towards tree species with higher rates of growth and photosynthesis.