Coupled Modes over Indian Ocean at Sub-seasonal time Scales and its Prediction

Abstract:

Sub-seasonal variability over the Indian Ocean, such as Madden-Julian Oscillation impacts weather and climate globally. However, the prediction of tropical sub-seasonal variability (TSV) remains a challenge, and understanding air-sea interactions on TSV time-scales is likely to be an important part of the prediction problem. The purpose of this paper is to examine the predictability of sub-seasonal variability in the tropical Indo-Pacific region. The analysis emphasizes on variability associated with coupled air-sea interactions in observational estimates, and how well these coupled modes are simulated and predicted within the context of a 30-year retrospective forecast experiment with a state-of-the-art atmosphere-ocean coupled model. The analysis shows that Sea Surface Temperature anomalies (SSTA) over the Indian Ocean tend to precede precipitation anomalies by 7-11 days with maximum amplitude over the Arabian Sea and the Bay of Bengal for summer and along the Seychelles-Chagos Thermocline Ridge (SCTR) region for winter. Though these coupled modes are captured by the models, the forecasts fail to predict its evolution. Based on the diagnosis of these coupled modes, we introduce a SCTR-SST index and an index that measures the modulation of the low-frequency amplitude (LFAM) of sub-seasonal SSTA variability over SCTR as a way to predict the coupled modes. Based on correlation with the observed variability, SCTR-SST has forecast skill of about 45 days over the Indian Ocean. However the sub-seasonal SSTAs in the predictions and the observational estimates do not have any direct ENSO tele-connection. In contrast, the LFAM of the sub-seasonal SSTA variance over SCTR is strongly correlated with ENSO, suggesting enhanced sub-seasonal variance on seasonal time-scales is potentially predictable.