Abstract
This study uses retrospective forecasts made using an APEC Climate Center seasonal forecast model to investigate the cause of errors in predicting the amplitude of El Niño Southern Oscillation (ENSO)-driven sea surface temperature variability. When utilizing Bjerknes coupled stability (BJ) index analysis, enhanced errors in ENSO amplitude with forecast lead times are found to be well represented by those in the growth rate estimated by the BJ index. ENSO amplitude forecast errors are most strongly associated with the errors in both the thermocline slope response and surface wind response to forcing over the tropical Pacific, leading to errors in thermocline feedback. This study concludes that upper ocean temperature bias in the equatorial Pacific, which becomes more intense with increasing lead times, is a possible cause of forecast errors in the thermocline feedback and thus in ENSO amplitude.
Highlights
ObjectivesUsing hindcast simulations of the APCC CCSM3, this study aims to understand how upper ocean mean temperature bias in the tropical Pacific, which intensifies with increasing forecast lead time, is associated with errors in ENSO-related feedbacks and errors in ENSO amplitude forecasts
All the monthly data at 1- to 6-month lead times from each of the ensemble member forecasts were reconstructed to obtain monthly datasets covering the period from 1982–2014 for each lead time
This study used hindcast simulations made using the APCC CCSM3 to investigate the possible cause of forecast errors in simulating ENSO amplitude
Summary
Using hindcast simulations of the APCC CCSM3, this study aims to understand how upper ocean mean temperature bias in the tropical Pacific, which intensifies with increasing forecast lead time, is associated with errors in ENSO-related feedbacks and errors in ENSO amplitude forecasts
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
