Addressing Ground Clutter Corruption of Sodar Measurements

Abstract

Echoes from ground clutter around a SODAR may result in a low bias in the wind speed measured by a SODAR. Typically, some, but not all, of the data collected by a SODAR at a site with ground clutter may be corrupted in this manner. In this study, methods to identify acceptable data points in already-collected data are explored, as well as the use of those remaining data points for the estimation of wind speeds. The paper also explores two approaches to filtering that can remove the effects of ground clutter from the raw data, before averaging. One filters out individual wind speed measurements based on signal amplitude. The other approach attempts to filter the raw acoustic data, using wavelets, to eliminate the effect of ground clutter before any further data processing. The filtering of data based on signal amplitude works well, but may result in few data samples with which to determine 10-minute averages. The wavelet approach did not successfully filter out the effects of ground clutter.