Characterizing the skill of CFSv2-based seasonal drought prediction at multiple spatiotemporal scales over China

Abstract:

A climate model’s predictive skill for seasonal temperature and precipitation generally varies with multiple factors, such as location, lead-time, season, and temporal and spatial scales. To fully understand the potential and limitation of the NCEP Climate Forecast System version 2 (CFSv2) in predicting seasonal drought in China, this study investigated how the seasonal drought predictive skill varies with such multiple factors in China. Six-month standardized precipitation index (SPI6) is used as the primary drought indicator to measure the medium-term meteorological drought. The predictive skill was then assessed by the correlation coefficient between observation-based SPI6 and CFSv2 forecast-based SPI6 at multiple spatial scales as well as multiple lead times during the period 1982-2008. Through this analysis, we will better characterize the capability of CFSv2 in seasonal climate forecast, which can help us to better utilize forecast information from such a system in drought prediction and water resource management.