全球变化知识资源中心

Old-growth coniferous forests of the Pacific Northwest are among the most productive temperate ecosystems and have the capacity to store large amounts of carbon for multiple centuries. To date, there are considerable gaps in modeling ecosystem fluxes and their responses to physiological constraints in these old-growth forests. These model shortcomings limit our ability to understand and project how the old-growth forests of the Pacific Northwest will respond to global climate change. This study applies the cohort-based Ecosystem Demography Model 2 (ED2) to the Wind River Experimental Forest (Washington, USA), a well-studied old-growth Douglas-fir-western hemlock ecosystem. ED2 is calibrated and validated using an extensive suite of forest inventory, eddy covariance, and biophysical observations. ED2 is able to reproduce observed forest composition and canopy structure, and carbon, water, and energy fluxes at the site. In the simulations, the effect of limited water supply on ecosystem carbon fluxes is mediated primarily by the forest's gross primary productivity (GPP) response, rather than its heterotrophic respiration response. The simulation indicates that stomata! conductance is mainly determined by soil moisture during periods of low vapor pressure deficit (VPD). However, when VPD is high, stomatal conductance is greatly reduced regardless of soil moisture status. During summer droughts, reduced soil moisture and increased VPD result in considerable stomatal closure and GPP reduction, which in turn decreases net carbon uptake. Cohort-based scheme integrates all canopy layers (species) that have distinct sensitivity to microclimate and respond distinctly to drought. This study is an initial first step to explore the potential importance of cohort-based model in simulating forest with complex structure, and to lay the foundation for applying cohort-based model at regional scales across the Pacific Northwest.