An Interpolation Method For Weather Radar Data Based On Fourier Spectrum Analysis And Its Quantitative Comparison Analysis

An interpolation method based on Fourier spectrum analysis ("Fourier interpolation method" for short) has been developed for weather radar echo imagery data. This method is conducive to showing some severe convective centers that cannot be detected because of the lower spatial resolution at long distance in processing cumuliform cloud echoes with the characteristics that lead to convective precipitation. It may provide us the reference for further precise identification of position and intensity of strong echo centers.The algorithm of the interpolation method has been described in steps in processing weather radar echo images based on the theory of 1-D Fourier spectrum analysis. T principle for "Fourier interpolation method" is based on fitting discrete radar-sampling points with Fourier spectrum analysis and then re-sampling for the purpose of interpolation.We simulated radar echo images with different resolution at one time based on a squall line process detected by Jiangsu SA radar on April 28,2015.Fourier and bilinear interpolation methods have been carried out on the low resolution radar images and then compared with the high resolution radar images. The results show that Fourier interpolation method can show some strong echo centers that exist in real. Compared to bilinear interpolation method, the position and intensity of strong echo centers interpolated by Fourier method are more consistent with high resolution radar images, and error is smaller.Cases like typhoon and heavy rain radar images have been studied. The effect of Fourier interpolation and bilinear interpolation method in processing radar echo image at low angle has been assessed. Analysis shows that this method performs better in highlighting the structure characteristics of strong echoes in severe convective area than the commonly-used bilinear interpolation method and is conducive to showing some severe convective centers which cannot be detected because of the lower spatial resolution at long distance and the velocity of echo can be calculated more precisely by this method.Several interpolation approaches were investigated to turn a radar PPI reflectivity data onto rectangular coordinate grid. The results show that the strong echo centers which interpolated by Fourier spectrum analysis are actually exist compared to NUIST radar. The approach FVI (Fourier interpolation on range-azimuth planes combined with a linear interpolation in vertical direction) has been developed and compared to usual approach NVI in 3D Cartesian grid. It was found that the CAPPI image produced by FVI has better continuity in remote distance.