Evaluation of the Subseasonal Forecast Performance of Marine Heatwaves in the Tropical Indian Ocean

In recent years, intensified global warming has led to increasingly frequent marine heatwaves (MHWs) in the tropical Indian Ocean, exerting severe impacts on marine ecosystems and coastal socio-economic systems. Using the ECMWF Subseasonal to Seasonal (S2S) reforecast data and NOAA OISST observations, this study systematically evaluates the subseasonal forecast performance of MHWs in the tropical Indian Ocean. For deterministic forecasts, the days and cumulative intensity of MHWs are overestimated near the equator (by up to 40% and 25%, respectively, primarily dominated by false positives), whereas they are underestimated in off-equatorial regions (by up to 50% and 45%, respectively, mainly due to false negatives). Both overestimation and underestimation become more pronounced with increasing forecast lead time. The mean intensity of MHWs is underestimated across the entire region, with the underestimation increasing from about 9% to 14% with lead time. Moreover, forecast biases are more pronounced for strong MHW events than for weak ones. For probabilistic forecasts, MHW forecasting exhibit relatively high skill at lead times of 1–7 days, with predominantly positive Brier Skill Scores and AUC values exceeding 0.80. Although Brier Skill Scores are generally lower near the equator than in off-equatorial regions, AUC values are comparable between the two regions. Overall, the ECMWF S2S system shows promise for subseasonal forecast of MHWs in the tropical Indian Ocean, but notable deficiencies remain for intense events and in specific regions.