Adjoint-Method Retrievals of Low-Altitude Wind Fields from Single-Doppler Reflectivity Measured during Phoenix II

View More View Less
  • 1 Cooperative Institute for Mesoscale Meteorological Studies, University of Oklahoma/NOAA, Norman, Oklahoma
© Get Permissions
Restricted access

Abstract

The simple adjoint method of Qiu and Xu is upgraded and tested with the Phoenix II data for retrieving the low-altitude winds from the movements of reflectivity patterns measured by a single-Doppler radar. The upgraded method uses an improved reflectivity advection equation that contains not only the advection terms but also the eddy diffusion terms. The test results show that (i) utilizing multiple-time-level data provides more information and, thus, increases the accuracy of the retrieval; (ii) the adjoint method can retrieve not only the time-mean (or running mean) velocity field but also the coefficients of horizontal and vertical eddy diffusion (retrieving the eddy coefficients improves the velocity retrieval); (iii) the retrieval is improved by judiciously setting the weights in the cost function and by spatial noise-filtering and temporal interpolation of the data; (iv) the retrieval is further improved when the observed radial wind (along the radar beam) is used.

Abstract

The simple adjoint method of Qiu and Xu is upgraded and tested with the Phoenix II data for retrieving the low-altitude winds from the movements of reflectivity patterns measured by a single-Doppler radar. The upgraded method uses an improved reflectivity advection equation that contains not only the advection terms but also the eddy diffusion terms. The test results show that (i) utilizing multiple-time-level data provides more information and, thus, increases the accuracy of the retrieval; (ii) the adjoint method can retrieve not only the time-mean (or running mean) velocity field but also the coefficients of horizontal and vertical eddy diffusion (retrieving the eddy coefficients improves the velocity retrieval); (iii) the retrieval is improved by judiciously setting the weights in the cost function and by spatial noise-filtering and temporal interpolation of the data; (iv) the retrieval is further improved when the observed radial wind (along the radar beam) is used.

Save