Premium
Improving the dynamical seasonal prediction of western Pacific warm pool sea surface temperatures using a physical–empirical model
Author(s) -
Chen Ping,
Sun Bo
Publication year - 2020
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.6481
Subject(s) - western hemisphere warm pool , hindcast , sea surface temperature , climatology , environmental science , predictability , mathematics , geology , statistics
The western Pacific warm pool (WPWP) has a profound impact on the global climate. In this study, the forecast skill of ENSEMBLES model for predicting the WPWP sea surface temperature (SST) for the period 1960–2006 is evaluated, where a WPWP index (WPWPI) is defined to represent the interannual variability of WPWP SST. The result indicates that the ENSEMBLES exhibit a poor skill in predicting the WPWPI during January–April (2‐ to 5‐month forecasts starting on November 1). To improve the ENSEMBLES‐predicted WPWP SSTs during January–April, a physical–empirical (PE) model is developed based on two predictors, using the year‐to‐year increment method and the linear regression method. The two predictors include the ENSEMBLES‐predicted sea level pressure during January and the observed northern tropical Atlantic SSTs during the preceding August. The mechanisms associated with the two predictors are illuminated. The 1‐year‐out cross‐validation and the independent hindcast indicate that this PE model may notably improve the WPWPI prediction of ENSEMBLES, with a correlation coefficient (CC) above 0.6 between the PE‐model‐predicted WPWPI and the observed WPWPI during January–April. The physical mechanisms expounded in this study and the PE model utilized in this study can be considered to improve the prediction of WPWP SST of numerical models in the future.