Monday, 11 August 2003
Cloning radar cells at outer ranges: a test with the Bauru radar
Since the early days of operations of the radar facility at Bauru the detectability of (the upper part of) thunderstorms at the outermost ranges, i.e., around 400 km, was evident. Because of specific logistic and economical reasons, this has drawn much attention, eventually leading to successfull attempts of far range correction of radar rainfall estimations with the Bauru radar. Notwithstanding, due to the low MDZ and to observations restricted to the lower reflectivities of the top of storms, only the tip of the core is detected impairing the operational utilization of the radar in the outer range ring.
One approach to mitigate this limitation, is the use of satellite observations to restore to it fullest, the far radar cells.
This paper explore such an approach, as applied to storms surveyed by both the Bauru and Presidente Prudente radars, in Central S. Paulo.
The basics of the procedure, essentially involving a kind of "cloning" of a radar cell, based in a reference at near range, to the furthest distance range, is presented.
One event of the wet-to-dry transition period, e.g. October, for wich the needed satellite data was available, was selected for a test run, involving both the Bauru radar - used as a control - and the Presidente Prudente (PP) radar.
A control cell at about 80 km from the Bauru radar provided the functional correspondences between radar reflectivities and satellite data, for VIS and IR channels of the GOES-8, and the HSB (Humidity Sounder of Brazil) 183.33 +/-7 GHz channel of the Aqua spacecrafts, respectively.
The test cell was approximately 360 km far from the PP radar, its detected core reflectivity was range corrected and "surrounded" by the satellite image envelopes,which were apropriatelly tagged with radar rainfall values.
In this phase of the project the test cell as enveloped by the HSB (microvawe) observations was quantitatively verified against the radar truth taken as the test cell as observed by the Bauru radar (approximatelly 220 km away). The first results suggest tha the achievement of a retrieval accuracy acceptable for numerical forecasting is within reach. The rain boundaries of about 240 K found are compatible with those obtained in recent works carried out using the AMSU-B instrument on board NOAA satellites. While VIS and IR retrieved test cells have shown to the quantitatively usefull for operational practices, work is proceeding for a quantitative verification, as well.
Most immediate operational applications of the procedure are planned for the Bauru radar observations as an auxiliary for local radar observations at the very sensitive region of the Paraíba Valley ( in the São Paulo-Rio axis), and of frontal activity observations over water with a coastal radar, far off the shore line.
Supplementary URL: