Randal D. Koster, S. P. P. Mahanama, T. J. Yamada
GMAO, NASA/Goddard Space Flight Center
Greenbelt, MD, USA
Gianpaolo Balsamo
ECMWF
Shinfield Park
RG2 9AX Reading, UK
A. A. Berg
Department of Geography, University of Guelph
Guelph, Ontario, Canada
M. Boisserie
Center for Ocean-Atmospheric Prediction Studies, FSU
Tallahassee, Florida, USA
P. A. Dirmeyer
Center for Ocean-Land-Atmosphere Studies
Calverton, Maryland, USA
Francisco J. Doblas-Reyes
ICREA and IC3
Doctor Trueta 203
08005, Barcelona, Spain
The second phase of the Global Land-Atmosphere Coupling Experiment (GLACE-2) is a multi-institutional numerical modeling experiment focused on quantifying, for boreal summer, the subseasonal (out to two months) forecast skill for precipitation and air temperature that can be derived from the realistic initialization of land surface states, notably soil moisture. An overview of the experiment and model behavior at the global scale is described here, along with a determination and characterization of multi-model “consensus” skill. The models show modest but significant skill in predicting air temperatures, especially where the rain gauge network is dense. Given that precipitation is the chief driver of soil moisture, and thereby assuming that rain gauge density is a reasonable proxy for the adequacy of the observational network contributing to soil moisture initialization, this result indeed highlights the potential contribution of enhanced observations to prediction. Land-derived precipitation forecast skill is much weaker than that for air temperature. The skill for predicting air temperature, and to some extent precipitation, increases with the magnitude of the initial soil moisture anomaly. GLACE-2 results are examined further to provide insight into the asymmetric impacts of wet and dry soil moisture initialization on skill.