Asymmetric Impacts of the Indian Ocean Dipole on Rainfall: Evidence from the Tropical Indian Ocean
摘要
The Indian Ocean Dipole (IOD) is a dominant mode of climate variability in the Indian Ocean (IO), characterized by oscillating sea surface temperature (SST) anomalies between the western and southeastern equatorial regions. The two distinct phases of the IOD—positive (pIOD) and negative (nIOD)—exert significant influence on climate dynamics across the tropical IO. While the spatial response of rainfall to pIOD and nIOD phases is well documented, the asymmetry in rainfall intensity between the positive and negative phases of the IOD remains relatively unexplored in IO region, leaving an important research gap. This study addresses this gap by analyzing observational data and historical simulations from the Institute Pierre-Simon Laplace Climate Model version 6 A–Low Resolution (IPSL-CM6A-LR). Results during the IOD peak season (September - November; SON) reveal that pIOD events increase rainfall over the Western Indian Ocean (WIO) and decrease it in the Southeastern Indian Ocean (SEIO), off Sumatra. In contrast, nIOD events produce an opposite, though generally weaker, pattern. Composite residual highlights the stronger intensity of pIOD-related rainfall anomalies in the IO region. Rainfall anomalies are found to be driven by zonal SST gradients in the IO, which modulate moisture convergence and vertical convection. The asymmetry in rainfall intensity is attributed to the positive skewness of the Dipole Mode Index, which results in stronger SST gradients, moisture convergence, and more vigorous convection during pIOD events. These results highlight the asymmetric nature of IOD-induced rainfall variability in the tropical IO, offering improved understanding of regional climate dynamics and their implications for tropical islands such as Sri Lanka, where pIOD events exert a stronger influence on seasonal rainfall.
Graphical AbstractThe graphical abstract presents an overview of the influence of the Indian Ocean Dipole (IOD) on rainfall variability across the Indian Ocean (IO). The analysis used Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) and ERA5 rainfall, 850 hPa wind, sea surface temperature (SST), moisture divergence, outgoing longwave radiation (OLR), in-situ rainfall data from Sri Lanka, and historical simulations from the IPSL-CM6A-LR model. Anomalies were derived by removing the seasonal cycle and long-term linear trend, and the IOD-induced effect was isolated using multiple linear regression and data reconstruction. Positive and negative IOD (pIOD and nIOD) events were defined using the Dipole Mode Index (DMI) standard deviation. Composite maps of the IOD phases highlight spatial and temporal rainfall impacts. The composite residual shows that pIOD events exert stronger intensity than nIOD events, including over Sri Lanka, which lies in the central IO. The results are consistent with the positive skewness of DMI, indicating the dominance of pIOD. During pIOD, warmer SSTs in the Western Indian Ocean (WIO) enhance convection, while cooler SSTs in the Southeastern Indian Ocean (SEIO) suppress convection. This modifies the pressure gradient across the IO, which in turn reinforces westward advection of warm waters and amplifies the SST gradient. The resulting ocean–atmosphere feedback intensifies moisture transport and convergence over the WIO, with reduced OLR confirming enhanced cloud cover. Historical simulations further support the observational evidence, demonstrating the model’s ability to capture IOD-induced rainfall asymmetry.