Preliminary Results of the AMIP II Diagnostic Subproject 12: Implications for Modeling Coupled Atmosphere/Land Interactions

Some 30 global modeling centers and 20 diagnostic subprojects are participating in the second phase of the Atmospheric Model Intercomparison Project (AMIP II), a major initiative of the World Climate Research Programme (WCRP). The intercomparison involves the diagnosis of different facets of simulations of the observed climate for the period 1979-1996,subject to commonly specified values of solar constant, carbon dioxide concentration, and ocean/sea ice boundary conditions. Among the many differences in AMIP II model features are the respective land-surface schemes (LSS), which range in complexity from classic "bucket" models to detailed representations of vegetation and soil hydrology. Because of the rich collection of LSS represented, the WCRP's Project for Intercomparison of Landsurface Parameterization Schemes (PILPS) has organized Phase 3 of its work as AMIP II Diagnostic Subproject 12. The Subproject has begun to analyze data from a subset of the AMIP II models that recently have become available.

One of the goals of AMIP II Diagnostic Subproject 12 is to identify possible relationships between LSS type and the resulting simulation of continental climate. However, if such relationships exist, they will be obscured by the coupling of each LSS to its atmospheric host model, in that the simulation of land-surface climate will be a product of two-way (possibly compensatory) interactions. To reveal putative LSS "signatures" in these coupled simulations, diagnostics that highlight "intrinsic" LSS characteristics within the context of host model differences (e.g. in surface radiative fluxes and/or precipitation) need to be devised.

Diagnostic Subproject 12 will explore two approaches: the first will investigate whether various global similarity measures of land-surface climate cluster according to LSS type; the second approach will assess whether the LSS type has a discernable impact on land-surface energy and moisture partitionings (e.g. dimensionless Bowen and runoff ratios).Preliminary results from applying these diagnostics to available AMIP II simulations will be presented.

This work was performed under the auspices of the U.S. Department of Energy, Environmental Sciences Division, by the Lawrence Livermore National Laboratory under Contract W-7405-ENG-48.