Session 5R.4 Possible algorithms for the Dual-frequency Precipitation Radar (DPR) on the GPM core satellite

Tuesday, 25 October 2005: 4:15 PM
Alvarado ABC (Hotel Albuquerque at Old Town)
Toshio Iguchi, NICT, Koganei, Tokyo, Japan

Presentation PDF (202.7 kB)

This paper concerns precipitation retrieval algorithms for a spaceborne or airborne dual-frequency radar, in particular for the Dual-frequency Precipitation Radar (DPR) onboard the Global Precipitation Measurement (GPM) core satellite. From the dual-frequency radar data we can in principle estimate two parameters of the drop size distribution (DSD) at each range bin. In practice, however, estimates from such algorithms are not reliable in many cases unless we are given other information such as the attenuation to the first gate or to the farthest range gate that mitigates the effect of uncertainties in radar calibration and attenuation due to unobservable gases and particles, i.e., water vapor and cloud water.

This paper first reviews the general principles of dual-frequency algorithm and then examines some possible alternative algorithms that do not require additional information but can estimate reliable rain parameters by abandoning the possibility of estimating two independent DSD parameters at each range bin. Some possible methods are derived by assuming a certain plausible relationship between DSD parameters or between some rain parameters. The simplest one is to estimate a single parameter, for example, the path-averaged rainfall rate over an interval by taking the difference of the attenuation differences between the two end points of the interval at two frequency channels. This simple method does not require the radar calibration or surface reference, and it is a rather robust method. Between this simplest method on one hand and the algorithms that estimate two rain parameters at each range on the other hand, there are a few different types of methods that can estimate at least one parameter at each range. Such intermediate methods use an assumption or assumptions among some rain parameters or DSD parameters. This paper compares the performance of these methods and suggests a possible algorithm for the GPM's DPR data processing.

Some of the methods suggested are also applicable to a ground-based dual-frequency radar system.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner