Identifying the convective and stratiform rainfall regimes using stable isotopic measurement

It still remains an intriguing question in weather and climate research on identifying the convective rainfall from stratiform. In a future warmer climate, this will be important to know to predict the changing pattern of rain intensity and distribution and to plan an efficient usage of water resources. Although quite a few methods have been proposed to address this question, such as cloud top temperature value, height-integrated ice and cloud water paths, brightness temperature, drop size distribution etc., the fidelity and validity of those vary widely, and hence their applicability remains limited. In this work, we propose a method to identify these two rainfall regimes using a combination of surface and remote sensing measurements. We collected the rainwater samples daily at Port Blair, Sagar and Tezpur in India as part of a project CRP F31006, funded by the International Atomic Energy Agency (IAEA). We measured the ¹⁸O content in these collected rainwater samples by measuring its fractionation (δ¹⁸O) using a Triple Isotope Water Analyzer by Los Gatos Research, USA. Among these sites, Port Blair is an island on the Bay of Bengal near the tropics, Sagar is an inland location in central India's dry, arid climate and Tezpur is located in the wet and heavily forested northeast India. Whereas Port Blair is situated very close to the tropics, Sagar and Tezpur are closer to the subtropics. The dual-frequency precipitation radar in the global precipitation measurement (GPM) program provides the convective and stratiform rainfall records by looking at radar reflectivities. We utilize these records to estimate the area-averaged stratiform rainfall fraction over each of these locations. We find that the relation between rain intensity and stratiform rainfall fraction can be represented by a logarithmic regression, whereas, the relation between δ¹⁸O and rain intensity can be represented by linear regression. However, the logarithmic regression weakens with latitude, and, the slope of the linear relation changes from slightly negative to slightly positive. The three sites considered here are located under different environmental conditions (oceanic to continental, semi-arid to forest, southwest to northeast monsoon zones, etc.) and house different vegetation types. To better understand the underlying processes governing such relations, we also study the impact of different meteorological variables in regulating these relations. Based on our study, the δ¹⁸O can be used as a proxy to identify the relative contributions of convective and stratiform rain types in the total rainfall over a region.