Coupling an Advanced Land Surface–Hydrology Model with the Penn State–NCAR MM5 Modeling System. Part I: Model Implementation and Sensitivity

This paper addresses and documents a number of issues related to the implementation of an advanced land surface–hydrology model in the Penn State–NCAR fifth-generation Mesoscale Model (MM5). The concept adopted here is that the land surface model should be able to provide not only reasonable diurnal variations of surface heat fluxes as surface boundary conditions for coupled models, but also correct seasonal evolutions of soil moisture in the context of a long-term data assimilation system. In a similar way to that in which the modified Oregon State University land surface model (LSM) has been used in the NCEP global and regional forecast models, it is implemented in MM5 to facilitate the initialization of soil moisture. Also, 1-km resolution vegetation and soil texture maps are introduced in the coupled MM5–LSM system to help identify vegetation/water/soil characteristics at fine scales and capture the feedback of these land surface forcings. A monthly varying climatological 0.15° × 0.15° green vegetation fraction is utilized to represent the annual control of vegetation on the surface evaporation. Specification of various vegetation and soil parameters is discussed, and the available water capacity in the LSM is extended to account for subgrid-scale heterogeneity. The coupling of the LSM to MM5 is also sensitive to the treatment of the surface layer, especially the calculation of the roughness length for heat/moisture. Including the effect of the molecular sublayer can improve the simulation of surface heat flux. It is shown that the soil thermal and hydraulic conductivities and the surface energy balance are very sensitive to soil moisture changes. Hence, it is necessary to establish an appropriate soil moisture data assimilation system to improve the soil moisture initialization at fine scales.