Improved performance over time of integration in momentum flux estimation using Postset Beam Steering technique

Abstract

Using time series of vertical data collected from middle and upper atmospheric (MU) radar, the adaptive Capon beamforming technique is used to synthesize beams in the desired pointing directions within the radar beamwidth. Beam synthesis has been performed at the tilt angle of 1.5° with different beam configurations (4 beams, 8 beams, 16 beams, 32 beams, 48 beams and 64 beams), which are equally separated azimuth plane. The Power Spectral Density (PSD) is obtained from the synthesized beam using the eigenvector (EV) sub-space based spectral estimation method. The first moment is derived from the EV produced spectrum using the adaptive moment estimation method. From the first order moments derived along equally spaced pointing directions, radial wind velocities are readily obtained in the corresponding directions. From radial velocity obtained in various pointing directions, momentum flux of short duration (<2h) is estimated using the symmetric beam method (SBM) which requires 4 symmetric beams separated by 90° in the azimuth plane. A comparative study has been performed to study the optimum beam configuration, required for the flux estimation, which shows that 32 beam configuration is sufficient to estimate the flux with least error. Study with 32 beam configuration, gives 8 sets of symmetric beams (8×4). Using 8 set of beams, the flux is estimated for each set of beams and averaged. The averaged flux is further integrated over different lengths of time up to 14h. This systematic method for the estimation of momentum flux reveals that spatial averaging of beams in azimuth and integration over different lengths of time have reduced the time of integration from 15 to 16h for the conventional approach to 8–9h for the new approach using Postset Beam Steering (PBS) technique. The flux estimated with spatial-averaging of beam using PBS technique has been compared with other standard methods.