| dc.description.abstract | An accurate seasonal prediction of Indian summer monsoon rainfall (ISMR) holds immense importance for the socio-economic well-being of South-Asian countries, as it directly impacts agricultural output and overall economies. The interannual variability of ISMR is strongly associated with climatic events occurring in various ocean basins, particularly the eastern and central Pacific oceans. However, despite this association, accurately predicting ISMR with a lead time of one season remains a challenging task. While there exist several theories on ISMR teleconnections, they differ significantly in their approaches. Hence, gaining insights into the role of global climatic patterns through a unified physical mechanism is crucial to improve ISMR prediction. Therefore, in this study, we investigate the physical processes that connect major climatic events with the Indian summer monsoon through a single physical mechanism by utilizing both observations and climate model simulations. By employing moisture budget theory, we highlight the crucial role of surface pressure modulation in controlling the ISMR variability, which is difficult to simulate accurately using numerical prediction models.
In the first part, we proposed a unified teleconnection framework that links tropical teleconnections to ISMR through the net moisture convergence driven by surface pressure (𝑃𝑠) gradients surrounding the Indian region. The positive and negative phases of major tropical climate patterns modulate these pressure gradients asymmetrically in the zonal and/or meridional directions leading to asymmetric changes in moisture convergence and ISM rainfall. Stronger El Nino droughts than La Nina floods are due to greater decreased eastward moisture flux over the Arabian Sea during El Nino than the corresponding increase during La Nina driven by proportionate meridional 𝑃𝑠 gradients. While the equatorial Atlantic Ocean’s sea surface temperature in boreal summer and El Nino Southern Oscillation (ENSO) in the preceding winter changes ISMR significantly, moisture convergence anomalies driven by the Indian Ocean Dipole were insignificant. Moreover, while ISMR extremes during ENSO are due to asymmetric changes in zonal and meridional gradients in 𝑃𝑠, non-ENSO ISMR extremes arise due to the zonal gradient in zonally symmetric 𝑃𝑠 anomalies.
In the second part, we employ the previously proposed theory to examine the underlying physical processes that contribute to the occurrence of ISMR seasonal droughts in both observation and Climate Forecast System version 2 (CFSv2) model. In observations, a reduction in incoming zonal moisture flux over the Arabian Sea (𝐹𝑊 ) is essential for droughts to occur. In addition, an increase
in outgoing flux over the Bay of Bengal (𝐹𝐸) results in a severe drought. On the contrary, droughts in CFSv2 primarily occur due to an enhancement in 𝐹𝐸, seldom accompanied by a decrease in 𝐹𝑊 . This hypersensitivity of the CFSv2 ISMR to 𝐹𝐸 is further explained using the Matsuno-Gill response to moist convection. During El Nino droughts, precipitation decreases over the equatorial western
Pacific and eastern Indian Oceans. The resulting anomalous diabatic cooling increases local surface pressure (𝑃𝑠), intensifying meridional 𝑃𝑠 gradient, and thus, 𝐹𝐸. The reduction in 𝐹𝑊 , however, is associated with a cooling of the central north Pacific Ocean in tandem with El Nino. During non-El Nino droughts, frequent occurrences of cold sea surface temperatures over the western north
Pacific Ocean are noticed. This cooling decreases 𝑃𝑠 over east Asia, increasing 𝐹𝐸. To summarize, droughts in CFSv2 are controlled by the pan-Pacific climate but weakly decreasing 𝐹𝑊 . But in observations, a strong decrease in 𝐹𝑊 and a moderate increase in 𝐹𝐸 together lead to droughts.
Numerous previous studies have consistently highlighted that most climate models tend to overestimate the relationship between ENSO and ISMR. However, in the final part of this thesis, we argue that this strong relationship is not an inherent characteristic of the CFSv2 model but rather stems from the methodology employed to calculate the ensemble mean. The dominance of the ENSO-ISMR relationship in the ensemble mean results from a higher occurrence of ensemble members having the same sign of an anomaly. However, for non-ENSO forcing, the limited number of ensemble members with coherent anomalies leads to a diminished impact and inadequate ISMR response in the ensemble mean. This study highlights the limitations of relying solely on the ensemble mean for analyzing model characteristics and making forecasts. By exclusively focusing on the ensemble mean, there is a risk of making incorrect assessments of the model’s teleconnection patterns.
This thesis presents a comprehensive physical theory that connects tropical teleconnections and the Indian summer monsoon through the net moisture convergence driven by 𝑃𝑠 gradients around the Indian region. We have demonstrated the utility of this theory in analyzing the characteristics of climate models, which serves as a yardstick for enhancing the performance of dynamic models used for seasonal prediction. | en_US |
