GLACE: The Global Land-Atmosphere Coupling Experiment. Part I: Overview

Randal D. Koster*, Zhichang Guo, Paul A. Dirmeyer, Gordon Bonan, Edmond Chan, Peter Cox, Harvey Davies, C. T. Gordon, Shinjiro Kanae, Eva Kowalczyk, David Lawrence, Ping Liu, Cheng Hsuan Lu, Sergey Malyshev, Bryant McAvaney, Ken Mitchell, David Mocko, Taikan Oki, Keith W. Oleson, Andrew PitmanY. C. Sud, Christopher M. Taylor, Diana Verseghy, Ratko Vasic, Yongkang Xue, Tomohito Yamada

*Corresponding author for this work

Research output: Contribution to journalReview articlepeer-review

624 Citations (Scopus)


The Global Land-Atmosphere Coupling Experiment (GLACE) is a model intercomparison study focusing on a typically neglected yet critical element of numerical weather and climate modeling: land-atmosphere coupling strength, or the degree to which anomalies in land surface state (e.g., soil moisture) can affect rainfall generation and other atmospheric processes. The 12 AGCM groups participating in GLACE performed a series of simple numerical experiments that allow the objective quantification of this element for boreal summer. The derived coupling strengths vary widely. Some similarity, however, is found in the spatial patterns generated by the models, with enough similarity to pinpoint multimodel "hot spots" of land-atmosphere coupling. For boreal summer, such hot spots for precipitation and temperature are found over large regions of Africa, central North America, and India; a hot spot for temperature is also found over eastern China. The design of the GLACE simulations are described in full detail so that any interested modeling group can repeat them easily and thereby place their model's coupling strength within the broad range of those documented here.

Original languageEnglish
Pages (from-to)590-610
Number of pages21
JournalJournal of Hydrometeorology
Issue number4
Publication statusPublished - Aug 2006


Dive into the research topics of 'GLACE: The Global Land-Atmosphere Coupling Experiment. Part I: Overview'. Together they form a unique fingerprint.

Cite this