High-resolution observations of dynamic structures in the boundary layer using a range imaging wind profiler and an FMCW radar

Wind profiling radars must operate within stringent frequency management constraints, which limit their range resolution. Therefore, Range Imaging (RIM) has been developed as a means of improving range resolution through frequency diversity. In this technique, several closely spaced carrier frequencies are transmitted and received, and a constrained optimization method is used to generate high-resolution maps of the reflectivity field in range. Basically, a weighting is applied to the received signals at the different frequencies in order to create constructive interference between the different signals at a particular range within a nominal range gate. A RIM validation experiment was conducted during June of 2002 at the Boulder Atmospheric Observatory in Colorado. For the experiment, comparative measurements were collected using two vertically pointing, collocated radars. One was a UHF boundary layer radar (BLR) that had been configured to operated in a RIM mode. The other was an S-Band frequency-modulated continuous wave (FMCW) radar. The four frequencies chosen for the BLR were 914.0, 914.33, 915.33, and 916.0 MHz. The transmitted pulse width was 1.4 microseconds, which corresponds to a nominal range resolution of 210 m. The returned signal was sampled at 700 ns (105 m). The FMCW radar has a center frequency of 2.9 GHz and a bandwidth up to 200 MHz. The range resolution selected during the experiment was 2 m. Range-time-intensity maps of the radar reflectivity are presented from the two radars. Data from the BLR are shown both with and without RIM processing. Not surprisingly, the RIM processed data exhibit much more fine-scale vertical structure than can be seen from the conventionally processed data. Furthermore, the atmospheric structures revealed using the RIM technique agree remarkably well with those obtained from the FMCW radar. High-resolution range-time maps of the vertical velocity field obtained from the RIM processed BLR data are also presented.