The ability of a fuzzy logic classifier to dynamically\nidentify non-meteorological radar echoes is demonstrated\nusing data from the National Centre for Atmospheric\nScience dual polarisation, Doppler, X-band mobile radar.\nDynamic filtering of radar echoes is required due to the variable\npresence of spurious targets, which can include insects,\nground clutter and background noise. The fuzzy logic classifier\ndescribed here uses novel multi-vertex membership functions\nwhich allow a range of distributions to be incorporated\ninto the final decision. These membership functions are derived\nusing empirical observations, from a subset of the available\nradar data. The classifier incorporates a threshold of certainty\n(25% of the total possible membership score) into the\nfinal fractional defuzzification to improve the reliability of\nthe results. It is shown that the addition of linear texture\nfields, specifically the texture of the cross-correlation coefficient,\ndifferential phase shift and differential reflectivity, to\nthe classifier along with standard dual polarisation radar moments\nenhances the ability of the fuzzy classifier to identify\nmultiple features. Examples from the Convective Precipitation\nExperiment (COPE) show the ability of the filter to identify\ninsects (18 August 2013) and ground clutter in the presence\nof precipitation (17 August 2013). Medium-duration\nrainfall accumulations across the whole of the COPE campaign\nshow the benefit of applying the filter prior to making\nquantitative precipitation estimates. A second deployment at\na second field site (Burn Airfield, 6 October 2014) shows the\napplicability of the method to multiple locations, with small\necho features, including power lines and cooling towers, being\nsuccessfully identified by the classifier without modification\nof the membership functions from the previous deployment.\nThe fuzzy logic filter described can also be run in near\nreal time, with a delay of less than 1 min, allowing its use on\nfuture field campaigns.
Loading....