Scaling Forest Management Practices in Earth System Models: Case Study of Southeast and Pacific Northwest Forests

Abstract:

A wide array of human-induced disturbances can alter the structure and function of forests, including climate change, disturbance and management. While there have been numerous studies on climate change impacts on forests, interactions of management with changing climate and natural disturbance are poorly studied. Forecasts of the range of plausible responses of forests to climate change and management are need for informed decision making on new management approaches under changing climate, as well as adaptation strategies for coming decades. Terrestrial biosphere models (TBMs) provide an excellent opportunity to investigate and assess simultaneous responses of terrestrial ecosystems to climatic perturbations and management across multiple spatio-temporal scales, but currently do not represent a wide array of management activities known to impact carbon, water, surface energy fluxes, and biodiversity.

The Ecosystem Demography model 2 (ED2) incorporates non-linear impacts of fine-scale (~10^-1 km) heterogeneity in ecosystem structure both horizontally and vertically at a plant level. Therefore it is an ideal candidate to incorporate different forest management practices and test various hypotheses under changing climate and across various spatial scales. The management practices that we implemented were: clear-cut, conversion, planting, partial harvest, low intensity fire, restoration, salvage, and herbicide. The results were validated against observed data across 8 different sites in the U.S. Southeast (Duke Forest, Joseph Jones Ecological Research Center, North Carolina Loblolly Pine, and Ordway-Swisher Biological Station) and Pacific Northwest (Metolius Research Natural Area, H.J. Andrews Experimental Forest, Wind River Field Station, and Mount Rainier National Park). These sites differ in regards to climate, vegetation, soil, and historical land disturbance as well as management approaches. Results showed that different management practices could successfully and realistically be implemented in the ED2 model at a site level. Moreover, sensitivity analyses determined the most important processes at different spatial scales, and also those which could be ignored while minimizing overall error.