Multistorm Analysis of Satellite Precipitation Informed Surface Hydrology for landfalling Tropical Cyclones

After the historically active 2020 hurricane season, the last three hurricane seasons have been climatologically average (1991 to 2020 climatology) in the Atlantic basin, but have still managed to set records for landfalling tropical cyclones (TC). These storms have inflicted substantial damage, affecting coastal and inland communities with deaths and destruction. TCs are a global phenomenon and many regions lack real-time precipitation estimates and flood prediction systems.

This study focuses on the feedback between floods and localized rainfall from TCs, crucial for developing regional infrastructure and vulnerability assessments. Using globally available data sets, we analyze eight tropical cyclones to address key questions: (i) Are satellite quantitative precipitation estimates (QPEs) reliable during TC scenarios? (ii) How do gridded precipitation data products (ground-based, radar QPEs, satellite QPEs) vary from each other? and (iii) What are the implications of these differences for surface hydrology?

Using storms such as Hurricane Charley (2004) and Hurricane Michael (2018), we evaluated these impacts through three approaches: (i) within the HUC03 region, (ii) along the TC track at varying distances, and (iii) in different landuse regions such as urban, forested, and agricultural. Our findings highlight that the IMERG satellite-derived product often underpredicts precipitation at higher quantiles, but aligns closely with the ground-based GCOOP and radar-based STAGEIV products at lower quantiles. IMERG generally shows higher discrepancies in total runoff for urban areas compared to agricultural and forested regions. Along the TC track regions themselves, discrepancies in total runoff informed by IMERG range from 0 to 18 mm, indicating the relatively stable performance of IMERG in these regions. Overall, IMERG was found to reliably estimate hydrological variables in 90\% of TC scenarios along the track and in agricultural and forested regions, with errors ranging from 0 to 10 mm. These insights highlight the utility of IMERG and similar datasets in assessing short-term hydrological impacts of TCs globally, despite variations in precipitation intensity and location ac