Evaluating the performance of the Community Land Model (CLM4.5) for a western US coniferous forest under annual drought stress

Tuesday, 16 December 2014
Henrique Duarte1, John C Lin1 and James R Ehleringer2, (1)University of Utah, Department of Atmospheric Sciences, Salt Lake City, UT, United States, (2)University of Utah, Department of Biology, Salt Lake City, UT, United States
The Community Land Model (CLM) is the land model of NCAR's Community Earth System Model (CESM), encompassing land biogeophysics, biogeochemistry, hydrology, and ecosystem dynamics components. Several modifications were implemented in its most recent release (CLM4.5), including a revised photosynthesis scheme and improved hydrology, among an extensive list of updates. Since version 4.0, CLM also includes parameterizations related to photosynthetic carbon isotope discrimination. In this study we evaluate the performance of CLM4.5 at the Wind River Field Station AmeriFlux site (US-Wrc), with particular attention to its parameterization of ecosystem drought response. US-Wrc is located near the WA/OR border in a coniferous forest (Douglas-fir/western hemlock), in a region characterized by strongly seasonal climate and summer drought. Long-term meteorological/biological data are available through the AmeriFlux repository (almost a decade of L4 (gap-filled) data available, starting in 1998). Another factor that makes the site so unique is the availability of a decade-long record of carbon isotope ratios (δ13C). Here we run CLM in offline mode, forced by the observed meteorological data, and then compare modeled surface fluxes (CO2, sensible heat, and latent heat) against observed eddy-covariance fluxes. We also use the observed δ13C values to assess the parameterizations of carbon isotope discrimination in the model. We will present the result of the analysis and discuss possible improvements in the model.