부산대학교 도서관

Although advanced land surface schemes have been developed in the past decade, many biosphere models still use the simple bucket model, partly due to its efficiency when it is coupled with an CGCM model. In this paper, we use a sophisticated land surface model, the Simulator for Hydrology and Energy Exchange at the Land Surface (SHEELS), including an explicit vegetation canopy and its physiological control on evapotranspiration and multiple, interactive subsurface soil layers. It is found that this model has potential for improving the carbon cycling description of a widely used biosphere model, the Carnegie-Ames-Stanford approach (CASA), especially for multiple seasonal integrations. Verifying with observations from Oklahoma Atmospheric Surface-layer Instrumentation System (OASIS) stations, we show that a bucket model as implemented in the CASA produces good simulations of the seasonal cycle of soil moisture content, but only for the upper 15-cm soil depth, no matter how it is initialized. This is partly due to its inability to include vegetation characteristics other than a fixed wilting point. Although only approximate, the soil depth to which CASA simulates storage of below-ground carbon is assumed to be about 30cm depth, significantly deeper than the 15cm depth. The bucket model cannot utilize the soil profile measurements that have recently been made widely available. A major finding of this study is that carbon fluxes are sensitive to the soil moisture simulations, especially the soil water content of the upper 30cm. The SHEELS exhibits potential for simulating soil moisture, and hence the total soil water amount, accurately at every level. For the Net Primary Production (NPP) parameter, the differences between two hydrological schemes occur primarily during the growing seasons, when the land surface processes are more important for climate. However, soil microbial respiration is found to be sensitive all year round to soil moisture simulations at our 7 selected Oklahoma Mesonet stations. These suggest that for future implementing of interactive representation of soil carbon in CGCMs, the accompanying hydrological scheme should not be over-simplified.