Modelling carbon‐coupled energy and water dynamics of a boreal aspen forest in a general circulation model land surface scheme

Abstract The ecosystem carbon (C) and nitrogen (N) processes and the C‐coupled energy and water dynamics were developed in the Canadian Land Surface Scheme (CLASS). The ecosystem C and N simulations include plant photosynthesis and respiration, plant tissue growth, senescence, root N uptake, soil heterotrophic respiration, and N mineralization and immobilization. These simulations are driven by variables (i.e. leaf temperature and water potential, and soil temperature and moisture) obtained from the C‐coupled energy and water balance simulations and feed back the model‐determined vegetation parameters (i.e. leaf area index, stomatal resistance, root length and distribution), which in turn control the land surface energy and water processes. In this paper, we introduce the C‐coupled energy and water balance scheme. The water flow process developed for the soil–plant–atmosphere system includes leaf and canopy stomatal resistance driven by leaf net CO 2 fixation, plant water capacitance, and soil rhizosphere and plant root resistances. This water flow scheme is dynamically coupled with the canopy energy balance so that the full water balance and energy balance equations can be solved simultaneously. The model was run at a time step of 30 min and tested in a stand‐alone mode driven by meteorological observations obtained at the old aspen ( Populus tremuloides ) site in the southern study area of the Boreal Ecosystem–Atmosphere Study (BOREAS). Results show that the model reproduced the observed diurnal and seasonal patterns of energy fluxes fairly well. Canopy conductance in the mid‐growing season was simulated to reach above 0.5 mol m −2 s −1 . Plant water capacitance was found to buffer the plant water flow process significantly and affect the canopy latent heat exchange under dry soil conditions. Comparisons of modelled and measured daily evapotranspiration in the two years of 1994 and 1996 gave the root‐mean‐square error of 0.71 mm day −1 and correlation coefficient of 0.75. Copyright © 2002 Royal Meteorological Society.