Applicability of Rain Gauge Blended Radar Precipitation for The Hydrologic Modeling of Adayar River Basin

Abstract Submission: For retrospective Hydroclimatic modeling and flood forecasting, a precise estimate of precipitation is crucial. The spatial and temporal variability of rainfall significantly influences catchment response in hydrologic modeling. While rain gauges cannot fully capture this variability at a finer scale, weather radar offers a more detailed representation of rainfall patterns. The study uses the radar rain rates: derived from reflectivity and specific differential phase values as the input for the Hydraulic Engineering Center’s -Hydrologic Modeling System (HEC-HMS)-based distributed hydrologic model of the Adayar basin, Tamil Nadu, India. The runoff estimates during the northeast monsoon flood events of 2015 and 2023-Michaung cyclonic storms were simulated and compared. To improve the model performance, the radar precipitation was blended with rain gauge measurements. Rain gauge observations were modeled as a function of radar estimate from the nearest grid point and variables like elevation, latitude, longitude, and coastal proximity by using a random forest model, followed by spatial kriging. This model effectively captured the rain gauge precipitation and showed the robustness of the model. In comparison with a hydrologic model with rain gauges as precipitation input, the model with blended radar rainfall estimate outperformed the original radar-derived data in both flood events. Comparing the hydrologic models provides valuable insights, allowing more accurate predictions and improved flood management strategies.

Learning Objectives/Expected Outcome (Optional) : Applicability of radar- rain gauge blended products in hydrologic modeling. and which can be utilized for planning statergies
validating the robustness of radar-gauge blended random forest model in comparison with point rain gauge data