A Data-Driven Framework for Studying Urban Rainfall Modification

In this study, we present a comprehensive framework for assessing the impact of urbanization on storm characteristics. Our approach involves analyzing diverse precipitation datasets and employing statistical techniques to investigate historical modifications in storm intensity and frequency over urban regions. Initially, we analyzed precipitation data from urban and adjacent non-urban stations to detect disparities in average daily rainfall intensity, and storm frequency between urban and non-urban regions. We further examined changes in the statistics of storm events occurring in two consecutive time periods (i.e., 1980-2000 and 2000-2020) for both the urban and its surrounding rural regions. Moreover, we assess annual maxima time series trends across all stations (urban and non-urban) to evaluate the hypothesis that urbanization leads to more rapid increases in annual maxima statistics, particularly over downwind of the urban areas. Also, we apply multiple causal discovery algorithms to a dataset comprising annual maxima rainfall, regional average temperature, Oceanic Nino Index (ONI), and urban built-up area time series to discover causal relationships with the aim of unveiling unconditional dependencies between changes in urban built-up area and annual maxima rainfall intensity. Lastly, we employ Stage IV radar and Integrated Multi-satellite Retrievals for GPM (IMERG) datasets to track historical storms over the urban region of interest. We apply this framework over Austin, Texas urban region. We scrutinize potential signs of urban-induced storm modifications, considering factors like urban heat island effects, aerosol emissions, and increased surface roughness. Despite existing studies on urban storm intensification, limitations in dataset selection and methodology have introduced biases. Our approach addresses these limitations by embracing a holistic view and incorporating diverse datasets and analyses. Through this comprehensive methodology, we aim to provide a more accurate understanding of the complex interactions between urbanization and storm dynamics.