Towards Operational Use of the SOPHy X-Band Radar in Peru: A Bayesian Fuzzy Logic Approach for Ground Clutter Filtering in Complex Orography

Abstract

Ground clutter (GC) echoes can negatively impact the quality of radar measurements and derived products such as quantitative precipitation estimation. Therefore, effective removal of GC while preserving hydrometeor echoes is essential. The most advanced GC filtering techniques rely on polarimetric variables; however, radars without polarimetric capabilities are limited to spatial analysis of reflectivity, which reduces filtering performance. In this context, we propose a GC filtering algorithm based on fuzzy logic adapted to non-polarimetric variables for X-band radars. For demonstration, observations collected between January and March 2025 by an X-band radar located in the Mantaro Valley (Peru, 3300 m.a.s.l.) were analyzed. A representative dataset that included dry and rainy conditions was selected to derive the membership functions required for the fuzzy logic framework. Statistical results show that the bias in fractional area (dimensionless) caused by GC was reduced from 0.187 and 0.073 to -0.017 and -0.007 at elevation angles of 1.0° and 2.6°, respectively. Furthermore, the spatial analysis was performed separately in stratiform and convective precipitation. The stratiform case proved the most challenging due to the similarity in signal intensity and velocity with GC, leading to classification errors and loss of meteorological signal. These results show that the proposed method can be easily implemented using a small dataset in both dry and rainy conditions.We also demonstrate that GC can be successfully removed using only Doppler velocity and reflectivity measurements.