The simulation of energy, water vapor and carbon dioxide fluxes over common crops by the Canadian Land Surface Scheme (CLASS)

The performance of the Canadian Land Surface Scheme (CLASS) was evaluated against observed eddy covariance flux data from four commonly cultivated crops growing in different climatic conditions. Of the four crops examined, wheat and soybean are C-3 plants, whereas maize and millet are C-4 plants. Physiological modules to simulate photosynthesis, respiration and growth for C-4 plants were added to CLASS. Apart from maximum photosynthesis capacity, and root and soil respiration rates at reference temperatures, most model parameters were kept constant for each crop. The diumal and seasonal cycles of simulated energy, water vapor and carbon dioxide fluxes were similar to the measurements made during the growing seasons for each of the four crops. Simulated values of canopy conductance were in agreement with observations. The model was able to explain 84, 76, 84 and 58% of the variance in the half-hourly net ecosystem productivity values for the maize, soybean, wheat and millet crops, respectively. Similarly, 77, 76, 77 and 78% of the variance in observed half-hourly evaporative fluxes was accounted for by the model for the maize, soybean, wheat and millet crops, respectively. We conclude that despite the process-based physics of the model, it is still critical to correctly specify detailed model parameters such as maximum carboxylation rate V-cmax, plant and soil carbon pools and temperature sensitivity coefficients to obtain close agreement between simulated and observed fluxes. Increasing the complexity of the model physics further necessitates the accurate specification of additional model parameters depending on the intended use of the model. The enhanced version of CLASS when coupled to Canadian global and regional climate models or when used as a stand-alone component can potentially be a valuable tool to evaluate the impact of climate change on the growth and health of terrestrial ecosystems. (c) 2005 Elsevier B.V.. All rights reserved.