The summer of 2022 saw unprecedented heavy rainfall in Pakistan and northwestern India, an event that surpassed even the 2010 event and was not predicted by most seasonal forecasting systems. This study investigates the reasons for this predictive failure, finding that positive sea surface temperature anomalies (SSTAs) in the northern coastal Arabian Sea (NAS) played a key role. While the ensemble mean prediction captured only 7% of the observed precipitation anomalies, sensitivity experiments nudging the model's NAS SST to observations captured about 25%. Improving NAS SST prediction is crucial for better forecasting extreme rainfall events in the region. Further multi-model analysis is needed.

Takeshi Doi, Swadhin K. Behera, Toshio Yamagata