|Published in||Journal of Hydrometeorology, v. 18(8): 2297-2311|
Spennemann, P.C., Rivera, J.A., Osman, M., Saulo, A.C., Penalba, O.C.
Centro de Investigaciones del Mar y la Atmósfera, Consejo Nacional de Investigaciones Científicas y Técnicas&ndashUniversidad de Buenos Aires, UMI&ndashInstituto Franco-Argentino sobre Estudios de Clima y sus Impactos/CNRS, Buenos Aires, Argentina
|Assessment of Seasonal Soil Moisture Forecasts over Southern South America with Emphasis on Dry and Wet Events.pdf|
The importance of forecasting extreme wet and dry conditions from weeks to months in advance relies on the need to prevent considerable socioeconomic losses, mainly in regions of large populations and where agriculture is a key value for the economies, such as southern South America (SSA). To improve the understanding of the performance and uncertainties of seasonal soil moisture and precipitation forecasts over SSA, this study aims to 1) perform a general assessment of the Climate Forecast System, version 2 (CFSv2), soil moisture and precipitation forecasts against observations and soil moisture simulations based on GLDAS, version 2.0 2) evaluate the ability of CFSv2 to represent wet and dry events through the forecasted standardized precipitation index (SPI) and standardized soil moisture anomalies (SSMA) and 3) analyze the capability of a statistical methodology (merging observations and forecasts) in representing a severe drought event. Results show that both SPI and SSMA forecast skill are regionally and seasonally dependent. In general, a fast degradation of the forecasts skill is observed as the lead time increases, resulting in almost no added value with regard to climatology at lead times longer than 3 months. Additionally, a better performance of the SSMA forecasts is observed compared to SPI calculated using three months of precipitation (SPI3), with a higher skill for dry events against wet events. The CFSv2 forecasts are able to represent the spatial patterns of the 2008/09 severe drought event, although it shows crucial limitations regarding the identification of drought onset, duration, severity, and demise, considering both meteorological (SPI) and agricultural (SSMA) drought conditions.