Understanding Rain Gauge Undercatch Through Drop Size Distribution–Based Rainfall Classification and Artificial Rainfall Generation

Accurate rainfall measurement remains challenging, even for in-situ point observations commonly considered the “ground truth”, owing to precipitation undercatch primarily caused by wind effects and instrument design. These biases limit reliable rainfall estimation, especially at very high and low intensities, and hinder the robust characterisation of precipitation variability. This study first used disdrometer data from multiple sites across the UK to develop a new rainfall classification system based on observed drop size distributions rather than intensity thresholds alone. The proposed classification distinguished periods of rainfall with similar bulk intensities but different microphysical structures, providing a more physically meaningful framework for precipitation characterisation and supporting the development of more targeted undercatch correction strategies. Second, a custom-built rainfall simulator was developed to replicate the identified rainfall types under controlled laboratory conditions. The simulator enables independent control of rainfall rate and drop size distribution, allowing the reproduction of a wide range of precipitation regimes representative of natural UK rainfall. Controlled experiments were used to systematically quantify the response of rain gauges to different drop populations and intensities, providing new insights into the mechanisms driving undercatch and its dependence on rainfall microstructure. By explicitly linking drop-scale processes, controlled experimentation, and population-level rainfall classification, this work contributes to the improved accuracy of precipitation measurements and the representation of rainfall at hydrologically relevant scales, with direct implications for rainfall monitoring, model input uncertainty, and flood risk assessment.