Atmospheric Radar Research Center

WTC is a challenging problem due to the non-stationarity of the signal. Unlike ground clutter, it is observed that WTC will vary with wind speed, direction, blade pitch, etc. Although we do not control these parameters, knowledge of their relationship to the WTC signal can be exploited to mitigate this significant problem.

Although experience provides some insight into the expected signal characteristics of WTC, it is extremely important to verify these assumptions by a thorough signal analysis. Level-I data from the Dodge City, Kansas and Great Falls, Montana radars were gathered during the Spring and Winter of 2006. A complete analysis of the collected data was performed and the following conclusions were reached:

• The tower structure creates a strong zero-Doppler return signal and can be removed using traditional stationary clutter filtering techniques.
• The turbine blade interference is only significant as each blade passes a vertical possition. This occurs six times per rotation for a three blade turbine. This confirms an earlier study done in the United Kingdom [Poupart, 2003].
• The stationary tower is the primary source of any WTC detected in the sidelobes; the blade interference is minimal in the sidelobes. Again, standard clutter filtering techniques can be applied.
• Wind direction plays an important role in the degree of spectral contamination. The orientation of the rotor axis with respect to the radar beam is a direct function of the wind direction and determines the range of radial velocities detected by the radar.

An example of the temporal evolution of the WTC Doppler spectrum and a PPI plot showing the WTC reflectivity are shown in Figure 1.

With the upgrade of the WSR-88D network to digital IF receivers (RVP8), it will be possible to readily implement sophisticated signal processing algorithms. For example, the new Gaussian Model Adaptive Processing (GMAP) algorithm [Siggia and Passarelli, 2005], used for clutter rejection and moment estimation, would not have been possible with the legacy system.
After the development of the adaptive WTC filtering scheme, the algorithm will be tested on the KDDC Level-I data in an off-line mode. Actual data under different meteorological conditions will be important in order to test the robustness of the algorithm.

Given the mission of the ROC, real-time implementation of the developed WTC filtering algorithms is the ultimate goal. In order to accomplish this goal, the investigators will work closely with the engineers, programmers, and scientists at the ROC in order to gain a better understanding of the capabilities and limitations of the RVP8 receiver. This knowledge will assist in refining the code for possible future real-time implementation.

Due to the complex nature of WTC, conventional spectral filtering techniques are not effective in recovering the true atmospheric signal. As a result, alternative techniques were explored that would reduce the impact of the WTC on the weather signal. Interpolation is a technique that was deemed appropriate due to the high spatial correlation of the atmosphere over relatively short distances. An example of the interpolation technique is shown in Figure 2.