Atmospheric Radar Research Center

Vector wind is important for quantifying and forecasting weather. However, weather radars such as the WSR-88D can measure only the radial component of the wind. In contrast, the three-dimensional wind field (including wind profiles) can be measured either by Doppler beam swinging (DBS) or interferometric techniques [Doviak and Zrnic, 1993; Briggs et al., 1950].

For example, the Spaced Antenna Interferometer (SAI) technique has recently been developed allowing the measurement of the cross-beam wind components (i.e., the component parallel to the baseline connecting two receivers) as well as the along-beam wind component obtained from the mean Doppler shift [Doviak et al., 1996]. SAI can be understood as either a radar signal received at one receiver delayed with respect to the other, or as interference patterns formed on the antenna plane moving in correspondence to the scatterers’ motion.

Until now, wind measurements using spaced receivers have been limited to vertically pointing wind profilers. With the availability of NSSL’s NWRT, however, implementation of SAI to measure cross-beam wind, shear and turbulence now appears feasible for the first time with an S-band radar.

In close collaboration with NSSL scientists and engineers (Dr. Doviak, and others), we propose to perform research and development of SAI on NWRT. The project focuses on researching, developing and implementing the SAI and DBS techniques, and verifying results with winds obtained from other radars (e.g., NOAA’s wind profiler at the Kessler Farm Field Laboratory and the KTLX radar). We also plan to compare the relative performance of the DBS and SAI techniques.

Presently the difference channel is not enabled and needs to be activated. This important task is currently being accomplished by engineers from both NSSL and LCMO. Microwave receiver components to down-convert the 3200 MHz RF signal to an IF frequency of 57 MHz signals will be purchased, assembled, and tested. The sum and difference signals will be reconnected to the in-house fabricated dual-channel microwave receiver, and its output will be connected to an RVP-8 digital IF receiver. The two complex signals from both the sum and difference channels (i.e., the H and V channels on the RVP-8) will be recorded. Although this work will be primarily accomplished by NSSL and LCMO, we are willing to assist in any capacity needed.

A joint effort between NSSL and OU will be made to reconstruct the signals for the left and right halves of the PAR antenna from the sum and difference signals. We propose to study the attenuation ratio and phase difference for the two channels. As an initial step, we will investigate ground clutter signals to find the attenuation ratio, and the phase difference and then synthesize a pair of balanced signals, one from the left and the other from right half of the antenna array.

The cross-beam wind is estimated from the cross- correlation function estimates of signals from the left/right sides of the array by using the correlation method [Doviak et al., 1996; Zhang et al., 2004]. The accuracy of the resulting wind estimates will be quantified statistically, with an expected error of less than 1 m/s with a 10 s dwell time. Verification will be accomplished by comparing the wind measurements with profiler and/or other nearby Doppler radars.

In order to fully understand and optimize the performance of SAI for cross-beam wind measurement with the NWRT, theoretical and numerical studies will be performed in support of the experiment. SAI will be formulated based on wave scattering from randomly distributed scatterers in the presence of shear and turbulence. The performance and limitation of SAI in complex weather conditions such as shear will be studied through theoretical analysis and numerical simulations, possibly using the radar simulator described in Section 3.1. The auto and cross-correlation functions will be derived and calculated for various possible configurations/conditions with both NWRT and MPAR.