Hydrologic model calibration requires reliable precipitation data. Most US urban collection system modeling studies rely on project-specific gages deployed coincident with intensive flow metering programs or use long-term rainfall measurements obtained by collection system owners or other agencies. When spatial variability is a concern, such as for monitoring conducted during summer, ground gage data may be replaced with gage-adjusted radar rainfall data. Radar-derived dataset development usually entails considerable effort to perform quality control on ground measurements and is burdensome to produce for extended periods. NOAA’s Multi-Radar Multi-Sensor (MRMS) quantitative precipitation estimation (QPE) dataset can potentially transform calibration and application of urban hydrologic models. The MRMS QPE offers a high quality, nationwide hourly dataset on a 1-km grid that can be readily applied anywhere in the US with modest effort, making it potentially very valuable for model calibration and hindcasting in municipal engineering studies. The principal limitation of the dataset is that portions of most urban collection systems have short times of concentration and are thus typically calibrated to 15-minute or finer data. This study examined application of a stochastic method for synthetic disaggregation of hourly precipitation records. MRMS QPE data was used to recalibrate the Hartford (Connecticut) Metropolitan District’s integrated collection sewer model to summer monitoring data. The recalibration improved model skill. The study presents a straightforward methodology for application of MRMS QPE data for urban hydrologic modeling, particularly in citywide models where ground rainfall gages inadequately represent the true rainfall distribution across the study area.
