Premium
Dynamical prediction of Indian monsoon: Past and present skill
Author(s) -
KöhnReich Lisei,
Bürger Gerd
Publication year - 2019
Publication title -
international journal of climatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.58
H-Index - 166
eISSN - 1097-0088
pISSN - 0899-8418
DOI - 10.1002/joc.6039
Subject(s) - climatology , monsoon , environmental science , indian ocean , forecast skill , atmosphere (unit) , convergence (economics) , period (music) , meteorology , atmospheric sciences , geology , geography , oceanography , physics , economics , macroeconomics , acoustics
Ongoing development of dynamical atmosphere–ocean general circulation models keep expectations high regarding seasonal predictions of Indian monsoon rainfall. This study compares past and present skill of four currently operating forecasting systems, CFSv2 from NCEP, ENSEMBLES, System 4 and the newest SEAS5 from ECMWF, by analysing correlations of respective hindcasts with observed all‐India summer rainfall. For the common time period 1982–2005, only ENSEMBLES and CFSv2 give significantly skilful forecasts. It is shown that skill is highly dependent on the chosen time period. Especially the intense El Niño of 1997 seems to degrade the predictions, most notably for SEAS4 and SEAS5 which seem to be linked to El Niño too strongly. We show that by discarding that year, a regime shift in the 1990s is no longer visible. Overall, we observe a convergence of skill towards the present with correlations of about 0.4 for CFSv2 and of 0.6 for System 4 and SEAS5.