Land Surface Modeling: A Mini-Workshop
1. Executive Summary
Background. A mini-workshop on land surface representation in global climate models was held at the NASA Goddard Institute for Space Studies on 14-15 May, 1998. The workshop was organized by Elaine Matthews, James Hansen and Roni Avissar. The objective was to obtain improved land surface modeling in the GISS model through increased collaboration with the outside research community, but also to contribute generally to the land surface/climate research discussion.
Workshop topic. The specific topic was the land surface modeling required to study interannual to multi-decadal climate variations. It was assumed that the principal model use will be simulation of forced and unforced climate variability of the past few decades, which are data rich, and climate projections for a few decades. Applications of the land surface models include study of: (1) vegetation response to climate change, and thus also the vegetation feedback on climate; (2) seasonal to interannual climate prediction, including the role of soil moisture and evapotranspiration; (3) land-use as a climate forcing, i.e., the transient climate effects of land-use changes.
Climate applications place high expectations on the global models. The climate models therefore need to have realistic unforced climate variability and realistic climate sensitivity. This implies that the land surface representation must provide realistic fluxes to the atmosphere and accurate runoff of fresh water to the ocean, and these quantities must respond properly to changes of climate.
Workshop aims. The first aim was fundamental discussion leading to mutual guidance regarding land surface representation in GCMs. A second aim was to identify the most appropriate data for judging model capabilities, and discuss how the data can best be obtained and used. A third aim was to improve coordination among different groups to maximize effectiveness of limited modeling resources.
Outcome. The outcome of the workshop is summarized in a table (Table 1) of priority research foci for land surface models in GCMs, including prospective gains and the current limitations and strengths. We identify groups pursuing these topics and links among the groups that are being or can be pursued to maximize progress. There was general agreement to improve modularization of land surface models and components, to promote exchange of these submodels, and to define and exchange sets of standard model diagnostics thus facilitating model comparisons. The research foci also suggest potential scientific collaborations, including interactions that can be carried out via students and post-doctoral scientists.