Abstract/Summary

The Indian summer monsoon can be described as a convectively coupled phenomenon – a factor that complicates modelling and forecasting. Interactions between the large-scale monsoon circulation and local convective systems are two-way and non-linear, but these relationships are not well documented. Based on data collected during the 2016 INCOMPASS field campaign, this thesis enhances our ability to objectively analyse monsoon storms using operational weather radars and provides insight into the complex relationships between these storms and the large-scale. The impressive potential for radar-based model evaluation for the region is also demonstrated. Firstly, an existing calibration technique is applied to the operational radar data for the first time and shown to lead to improved rainfall estimates. Next, the calibrated radars are used to gain further understanding of how monsoon onset and other aspects of variability during the 2016 monsoon season regulate convection. Storms most frequently reach heights of around 6–8 km for all locations. Intraseasonal variability during the monsoon affects the regional patterns of storm heights, especially for western and central India, and also modulates the diurnal cycle of storm occurrence for all regions. Over land, the horizontal area of storms peaks in the afternoon and evening, and typically in the early morning over the surrounding ocean, consistent with the diurnal cycle of precipitation. These results confirm that Indian monsoon convective regimes are partly regulated by the large-scale environment within which they are embedded. Finally, radar-derived storms are compared to those simulated by convection-permitting models over India for the first time. The model is found to simulate too much shallow convection; modelled storms are also often too wide and intense, and are largest around the freezing level, higher than observed. Some of these biases have been found before for other regions, and potential reasons for the major discrepancies are discussed.