Algorithms for wind parameter retrieval from rain-contaminated x-band marine radar images

Wang, Yali (2016) Algorithms for wind parameter retrieval from rain-contaminated x-band marine radar images. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (4MB)

Abstract

In this thesis, research for retrieving wind direction and speed from rain-contaminated X-band marine radar images is presented. Firstly, a method for retrieving wind direction from X-band marine radar data is proposed. The algorithm is used to investigate radar backscatter in the wavenumber domain and obtain wind direction from the wavenumber spectrum. For rain-contaminated images collected under low wind speeds (i.e. less than 8 m/s), wind directions are retrieved using spectral components with wavenumbers of [0.01, 0.2] rad/m. For rain-contaminated images obtained under high wind speeds and rain-free images, wind directions are retrieved using the spectral values at wavenumber zero. The algorithm was tested using X-band radar images and anemometer data collected on the east coast of Canada. Comparison with the anemometer data shows that the root mean square error (RMSE) of wind directions retrieved from low-wind-speed rain-contaminated images is reduced by 25.1 ◦ . Secondly, two methods for estimating wind speed from X-band nautical radar images are presented. One method is used to determine wind speeds by relating the spectral strengths of radar backscatter to the wind speeds using a logarithmic function. The other method is used to mitigate rain influence by applying gamma correction to rain-contaminated images, and then relate the average radar image intensities to measured wind speeds with a logarithmic function. Comparison with the anemometer data show that the two methods reduce the RMSEs of wind speeds estimated from rain-contaminated radar data by 5.9 m/s and 5.4 m/s, respectively. Unlike existing methods which require the exclusion of rain-contaminated data, the new wind parameter retrieval methods work well for both rain-contaminated and rain-free images.

Actions (login required)

View Item