Much of our understanding of how land surface processes and terrestrial ecosystems affect weather, climate, and atmospheric composition comes from numerical models of surface energy fluxes, the hydrologic cycle, and biogeochemical cycles coupled to atmospheric models. Land surface models are coupled to atmospheric models to simulate the absorption of radiation at the land surface, the exchanges of sensible and latent heat between land and atmosphere, storage of heat in soil, and the frictional drag of vegetation and other surface elements on wind. These models were initially developed to provide the surface boundary conditions of radiative and turbulent fluxes required by atmospheric models. They have since evolved to simulate the hydrologic cycle, biogeochemical cycles, and vegetation dynamics so that the land and atmosphere are represented as a coupled system. This chapter reviews the historical development of land surface models. Model evaluation is discussed, as well as application of the models in climate model experiments.
Global climate models represent a set of numerical equations that describe the large-scale circulation of the atmosphere and ocean and their physical state, including interactions among oceans, atmosphere, land, and sea ice that affect climate. The land surface fluxes of energy, moisture, and momentum and the associated hydrologic cycle that regulates them have long been represented in global climate models. In these models, absorption of radiation at the surface, the reflection of solar radiation and emission of longwave radiation, sensible and latent heat fluxes, storage of heat in soil, and frictional drag of the surface on wind influence climate. The land surface models used with climate models provide these biogeophysical boundary conditions at the land–atmosphere interface. They partition net radiation at the surface into sensible and latent heat fluxes, soil heat storage, and snow melt. They also partition precipitation into runoff, evaporation, and water storage in snow or soil. The most recent versions of these models simulate biogeochemical cycles (e.g., carbon), wildfires, land use, and land-cover change. The models update the state variables (snow cover, soil moisture, soil temperature, vegetation cover, leaf area index, and carbon pools) that regulate surface fluxes with the atmosphere.
First Generation Models
The first generation of land surface models used aerodynamic bulk transfer equations and simple prescriptions of albedo, surface roughness, and soil water without explicitly representing vegetation or the hydrologic cycle (e.g., Manabe et al. 1965; Williamson et al. 1987).