Intercomparison of Surface Precipitation in Seven SCMs Against the ABRFC Observation over the ARM SGP Site
Hua Song | Brookhaven National Laboratory |
Wuyin Lin | Brookhaven National Laboratory |
Yanluan Lin | Geophysical Fluid Dynamics Laboratory |
Leo Donner | Geophysical Fluid Dynamics Laboratory |
Roel Neggers | Royal Netherlands Meteorological Institute |
Audrey Wolf | NASA - Goddard Institute for Space Studies |
Anthony Del Genio | NASA |
Yangang Liu | Brookhaven National Laboratory |
Category: Precipitation
This study evaluates the overall performances of seven SCMs (NCAR CAM 3, 4, 5; GFDL AM2, 3; GISS, and ECMWF) by comparing simulated hourly surface precipitation with observations at the ARM SGP site from 01 January 1999 to 31 December 2001. Results show that most SCMs can reasonably well reproduce the observed total precipitation. The model performance is better in the nighttime than in the daytime. The frequency of precipitation, especially light precipitation, is higher in most SCMs than in the observation. The frequency of very strong precipitation (> 50 mm/day) is underestimated in all SCMs. For precipitation stronger than 10 mm/day in general, the frequency tends to be overestimated in the daytime but underestimated in the nighttime in most SCMs. The disparities in the convective/stratiform partitioning among the SCMs are further examined. It is also shown that most SCMs produce a spurious precipitation peak in the weak vertical motion regimes, while the observed precipitation is more tightly coupled to the large-scale vertical motions. Several extreme cases will be analyzed to further reveal the meteorological backgrounds against which the model underestimation/overestimation events may occur.
This poster will be displayed at ASR Science Team Meeting.