Calibration of X-band radar for extreme events in a spatially complex precipitation region in north peru: machine learning vs. empirical approach

Abstract

Cost-efficient single-polarized X-band radars are a feasible alternative due to their highsensitivity and resolution, which makes them well suited for complex precipitation patterns. Thefirst horizontal scanning weather radar in Peru was installed in Piura in 2019, after the devastatingimpact of the 2017 coastal El Niño. To obtain a calibrated rain rate from radar reflectivity, we employa modified empirical approach and draw a direct comparison to a well-established machine learningtechnique used for radar QPE. For both methods, preprocessing steps are required, such as clutterand noise elimination, atmospheric, geometric, and precipitation-induced attenuation correction,and hardware variations. For the new empirical approach, the corrected reflectivity is related to raingauge observations, and a spatially and temporally variable parameter set is iteratively determined.The machine learning approach uses a set of features mainly derived from the radar data. Therandom forest (RF) algorithm employed here learns from the features and builds decision trees toobtain quantitative precipitation estimates for each bin of detected reflectivity. Both methods capturethe spatial variability of rainfall quite well. Validating the empirical approach, it performed betterwith an overall linear regression slope of 0.65 and r of 0.82. The RF approach had limitations with thequantitative representation (slope = 0.44 and r = 0.65), but it more closely matches the reflectivitydistribution, and it is independent of real-time rain-gauge data. Possibly, a weighted mean of bothapproaches can be used operationally on a daily basis