Data_Sheet_1_Sub-seasonal to seasonal drivers of dry extreme rainfall events over Northeast Thailand.pdf

The interannual El Niño-Southern Oscillation (ENSO) and the Boreal Summer Intraseasonal Oscillation (BSISO) strongly modulate sub-seasonal to seasonal rainfall variability, leading to dry extreme rainfall events (DEREs) over Northeast (NE) Thailand. In this study, the ability of climate models to simulate the ENSO-BSISO-induced DEREs and associated synoptic features are evaluated using self-organizing maps. Observed DEREs occur most frequently during ENSO Neutral and La Niña conditions, when enhanced convection is located over central India and the Bay of Bengal. The intensity of observed DEREs are strengthened during El Niño when enhanced convection is observed over the western Pacific region. The climate models exhibit a diverse frequency of DEREs during ENSO phases, with some models showing better skill than others. On intraseasonal time scales, observed DEREs are favored when enhanced BSISO convection in phases 3–5 is located over the Bay of Bengal and the western Pacific region. Five models out of the 19 examined capture the observed pattern during BSISO phases 4 and 5, while only three models capture the BSISO phase 3 behavior. Composite maps of observed DEREs during the combined BSISO and ENSO conditions indicate that BSISO convection increases (decreases) DERE frequency during El Niño and Neutral (La Niña) phase(s). Climate models can simulate the occurrence of dry events associated with the combined BSISO and ENSO phases, but they do so in different BSISO phases.

厄尔尼诺-南方涛动（El Niño-Southern Oscillation, ENSO）与北半球夏季季节内振荡（Boreal Summer Intraseasonal Oscillation, BSISO）可显著调制次季节至季节尺度的降水变率，进而引发泰国东北部（NE）的极端少雨事件（dry extreme rainfall events, DEREs）。本研究采用自组织映射（self-organizing maps）方法，评估了气候模式模拟ENSO与BSISO共同诱导的DEREs及其相关天气尺度特征的性能。观测结果表明，DEREs在厄尔尼诺中性与拉尼娜状态下发生频次最高，此时对流异常活跃区位于印度中部与孟加拉湾上空；而在厄尔尼诺状态下，当西太平洋区域出现对流异常活跃时，观测到的DEREs强度会显著增强。不同气候模式在各ENSO位相下的DEREs发生频率存在显著差异，部分模式的模拟技巧优于其余模式。在季节内时间尺度上，当BSISO对流异常活跃区处于第3~5位相且覆盖孟加拉湾与西太平洋区域时，观测到的DEREs更易发生。在本次评估的19个模式中，有5个模式能够再现BSISO第4、5位相下的观测特征，仅3个模式可准确捕捉BSISO第3位相的相关行为。结合BSISO与ENSO位相的观测DEREs合成分析显示，在厄尔尼诺与中性状态下，BSISO对流活动增强会提升DEREs的发生频率，而在拉尼娜状态下则会降低该频率。气候模式虽能够模拟出BSISO与ENSO联合位相下的少雨事件发生特征，但对应的BSISO位相与观测结果并不一致。