Retrieving wind speed from rain-contaminated X-band nautical radar images

In this paper, an algorithm for retrieving wind speed from rain-contaminated X-band nautical radar images is presented. Two third-order polynomial functions for the relationship between overall average backscatter intensity and wind speed for rain-free and rain-contaminated images based on least-squares fitting with rain-free and rain-contaminated data, respectively, are derived. Radar images are then classified as rain-free and rain-contaminated according to their zero-pixel percentage (ZPP). For the rain-free data, wind speed is retrieved using the rain-free wind speed model. For the rain-contaminated data, wind speed is determined using the rain-contaminated wind speed model. Experimental results obtained from field data show that, by applying the new algorithm, the wind speed standard deviation and bias can be reduced by 1.2 m/s and 1.3 m/s, respectively.