Radar inter - calibration analysis: potential use for weather radar networks

Weather radar networks, originally centralized and based on single vendor solutions, are evolving towards increasing heterogeneity. This is due to both the adoption of upgrades of existing installations at different time and to include, in radar networks, also systems having different features (e.g.: working at different bands). A paradigmatic case is the I talian weather radar network, built up according to a federated paradigm upon existing and heterogeneous radars independently build and operated by different organizations, typically regional weather services. As known, any fruitful usage of network radar data either for quantitative precipitation estimation or just for operational monitoring, must deal with a careful check of data quality. This aspect is particularly critical for networks with heterogeneity radar having different features (products, scans trategy, operative algorithms and so on). Complex orography can further leads to miscalibration problems during mosaicing and make difficult the matching of observations of the same phenomenon form different radars. In addition, scheduled radar maintenance procedures can be insufficient to determine promptly whether a radar is performing correctly or not. With the advent of mosaic radar data and products, the need for techniques to monitor and diagnose the performance of radar networks is increasing. In this work a preliminary analysis on the inter-comparison of reflectivity data, collected by two C-band radars both installed in central Italy, but having different characteristics is shown. The radars used for the analysis are those of the Consiglio Nazionale delle Ricerche (CNR) in Rome and that of Region Abruzzo, Italy. The first is located at an elevation of 162 m, whereas the second is at 1660m. asl. Therefore, different observations can result both from miscalibration of sensors, and also by other factors related to influence of attenuation viewing geometry and different influence of vertical reflectivity profile. In the light of what mentioned above, the proposed work goes in the direction to set up an automatic routine to remotely detect miscalibration of radars with overlapping coverage within a network and possibly compensate it. The practical implications of this work are the optimization of the radar maintenance and the increasing of the quality of network radar products.