Asia, one of the most densely populated regions in the world, receives 50%-80% of annual rainfall during the summer monsoon season. While agricultural yield and water resource over this region benefit greatly from the summer rainfall, human lives and infrastructures are, at the same time, threatened by the frequently occurring heavy downpour. Although large efforts have been devoted to delineate the characteristics and variations of Asian monsoon extreme rainfall, its dynamical triggers and the physical mechanisms underlying the past and future changes remain poorly understood. To address the knowledge gap, this thesis aims to provide a process-oriented perspective on monsoon rainfall extremes with special attention given to the heavy-rain producing weather systems, namely the monsoon low-pressure systems (LPSs).
In Chapter 1, an objective feature-tracking algorithm is adopted to compile the observed trajectories of monsoon LPSs over the East Asia monsoon region during the post-1979 satellite era. Two types of LPS are identified. One forms near the downwind side of the Tibetan Plateau (i.e., southwestern China) and travels northeastward toward north-central China. The other forms over the western North Pacific Ocean and migrates along the southern and western peripheries of the subtropical high. The two types of LPS together account for approximately half of the rainfall extremes. The terrestrial LPSs are responsible for a great majority of extreme rainfall over inland areas, whereas the influences of marine LPSs are primarily confined to the coastal regions where they frequently make landfall. The observed long-term change in extreme rainfall, featured by a “south flood-north drought” pattern, aligns well with the change in LPS activity.
The decreasing number of northeastward-moving terrestrial LPSs leads to an extreme rainfall dipole with negative trends in north-central China and positive trends in southern China, while the increasing number of northward-recurving marine LPSs enhances the extreme rainfall along the southeastern China coast. These trends are driven by the weakening of the monsoonal southwesterlies and the eastward retreat of the subtropical high. Despite the great importance of terrestrial LPSs in modulating extreme rainfall over East Asia, these storms have so far received limited attention in research community because of the lack of a track archive. Chapter 2 further investigates the dynamical processes fueling the different evolution regimes of individual terrestrial LPSs and explores the environmental factors controlling their evolution.
Chapters 3 and 4 concentrate on the South Asian monsoon region, where the long-term trend of LPS activity remains debatable owing to the potential errors arising from the manual and subjective identification of LPSs from weather charts. Using two different tracking algorithms, in Chapter 3 we find that the trends of extreme rainfall and LPS activity indeed exhibit a strong coherence. Over time, the LPSs propagate preferentially through south-central India rather than north-central India, imparting a corresponding dipole footprint in rainfall extremes. In agreement with previous studies that the LPS propagation is a combined effect of the northwestward-propagating component due to horizontal nonlinear adiabatic advection and the southwestward-propagating component due to diabatic heating, the LPSs traveling through south-central India have stronger updrafts on their west-southwestern flank than those passing through north-central India. The increased frequency of LPSs propagating through south-central India is likely due to a strengthened cross-equatorial moisture transport over the Arabian Sea, which favors more vigorous storm convection through the conditional instability of second kind mechanism.
Chapter 4 then focuses specifically on the role of LPS in triggering the record-breaking Pakistan flood during summer 2022, when most of the South Asian LPSs were able to propagate into Pakistan with intensity and longevity far exceeding historical records. The abnormal LPS activity was fueled by a historically-high cross-equatorial moisture transport, which is in agreement with the fingerprint of anthropogenic warming in the Coupled Model Intercomparison Project - Phase 6 (CMIP6) models.
The last chapter of this thesis proceeds to evaluate the performance of CMIP6 models in simulating the monsoon rainfall extremes and to explore whether the performance is affected by the degree to which the models could realistically capture the LPS activity. The modelled precipitation often occurs more frequently and the extreme events are commonly less intense than in observations. A robust improvement of model performance in simulating monsoon rainfall extremes as resolution increases is seen across most models, both in terms of spatial distribution and intensity. The dry biases get improved in the regions with high exposure to monsoon LPSs, such as central India, southern China, and western North Pacific. The improvement is associated with a better representation of LPSs, which become more frequent and stronger at finer resolution.
Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/kgbp-9261 |
Date | January 2023 |
Creators | You, Yujia |
Source Sets | Columbia University |
Language | English |
Detected Language | English |
Type | Theses |
Page generated in 0.0024 seconds