Single-Doppler velocity retrievals with rapid-scan radar data

The past fifteen years has seen the development of a variety of new techniques to estimate the three-dimensional vector wind field from single-Doppler radar data (e.g., Tuttle and Foote 1990; Sun et al. 1991; Sun and Crook 1994; Qiu and Xu 1992; Shapiro et al 1995; Laroche and Zawadzki 1994, 1995; Xu et al 1994a,b, 1995; Zhang and Gal-Chen 1996, Gao et al. 2001, Lazarus et al. 2001). These Single-Doppler Velocity Retrievals (SDVRs) can be used as stand-alone algorithms for hazard-warning or nowcasting, or as a means for numerical model initialization or data assimilation (e.g., Shapiro et al. 1996; Wilson et al. 1998; Weygandt et al. 1999). Although the various SDVRs have successfully retrieved cross-beam winds in a number of simulated and real radar datasets, solution non-uniqueness and other difficulties are persistent problems in some circumstances. Our presentation describes the use of rapid-scan radar data to alleviate this problem.

A series of retrieval experiments were performed with rapid-scan (~ 1 min scan rate) data gathered by two Doppler-on-Wheels (DOW) research radars (Wurman et al. 1997). We consider three datasets: a Florida thunderstorm complex (summer 1997), a slow-moving non-convective Oklahoma cold front (spring 2000) and an intense landfalling convective cold-front from the CALJET field experiment (winter 1998). For each case, data from one radar were supplied to the retrieval, while data from both radars were used to construct a dual-Doppler wind analysis to verify the retrieved azimuthal wind component. The retrieval algorithm used in these experiments is a simple 4DVAR method that combines the conceptually simple Lagrangian framework of Laroche and Zawadzki (1994) with the method of Xu et al. (1995) in which bulk source terms in the equations of motion are retrieved as part of the problem. The most recent version of this retrieval also uses a local least squares approach to retrieve the winds on a small sector or "patch" (as in Laroche and Zawadzki 1995), and incorporates a mass conservation constraint. Our key experiments focussed on determining the optimum time and space resolutions for the retrieval and the optimum length of the assimilation window.

Our results suggest that a dramatic improvement in retrieval error statistics could be achieved as the volume scan times decreased from 5 minutes (characterizing the current WSR-88D scan rates) down to 1 minute, the fastest scan rates available from the Doppler-on-Wheels at the time of the respective field deployments. The trend suggests that even greater improvements can be attained with the faster scan rates available from phased array weather radars.