Does a skillful prediction of ENSO help in forecasting the 2012 Midwestern summer drought?
Speaker: Jonghun Kam, Graduate Student
Series: EEWR Brown Bag Seminars
Location: Engineering Quad E225
Date/Time: Friday, March 15, 2013, 12:00 p.m. - 1:00 p.m.
The 2012 Midwestern (MW) U.S. summer drought was characterized by the lowest summer precipitation in the last three decades with total crop losses of over US$20 billion. To mitigate the impacts of future MW droughts, skillful seasonal forecasting is necessary. One of the largest sources of predictability for MW precipitation is the El Niño-Southern Oscillation (ENSO) due to its persistence and relatively high predictability. To diagnose how well climate forecast models represent El Niño-Southern Oscillation (ENSO) teleconnections with MW summer droughts in the nature, we use the singular value decomposition (SVD) analysis for global sea surface temperatures (SSTs) and conterminous US (CONUS) precipitation during summers from observational data (1915-2012) and climate forecast models (1982-2012). From observations, the first SVD SST mode relates to Pacific Decadal Oscillation (1922-1951; PDO) and ENSO (1952-1981 and 1982-2012). During 1982-2012, the ENSO-like SVD modes from climate forecast models explain the squared covariance by from 15% up to 44% whereas observational data have the squared covariance fractions (SCFs) of the three leading SVD modes, 10%, 9.8%, and 7%, respectively. Due to their strong coupling, the climate forecast models are still limited to forecast MW summer precipitation anomalies even at zero month lead time forecasting regardless of our skillful prediction for ENSO. We carefully speculate that ENSO teleconnections with MW summer precipitation are overconfident in the current climate forecast models.