Development of a TRMM-based Tropical Cyclone Precipitation Climatology

Principal Investigator: Frank Marks

Collaborating scientist(s): Shuyi Chen (UM/RSMAS)
Chris Kummerow (NASA/GSFC)
Joanne Simpson (NASA/GSFC)


Objective:
This proposal is a collaborative effort between NOAA/HRD, UM/RSMAS, and NASA to analyze the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) and Precipitation Radar (PR) surface rain estimates in all tropical cyclones (TC) encountered by the spacecraft to develop a climatology of TC precipitation fields.


Rationale:
One of the major shortfalls in prediction of TC precipitation is the lack of climatological distributions of rain in both space and time. Current forecast techniques rely on a simple rule of thumb linking the maximum storm total rain to a peak storm total (1310 mm) divided by the storm's motion (m s-1). While this rule of thumb provides a reasonable estimate of the peak storm total rain, it provides no information about the distribution of rain in space or time, the first step to producing a statistical TC rainfall model for validating quantitative precipitation forecasts. The global coverage of the TRMM rain estimates provide an excellent opportunity to develop a climatology of TC rain distributions from a large number of storms. In the first 13 months of operation TRMM sampled 84 TCs with 1050 orbits passing within 750 km of a TC center (16% of 6227 total orbits). This sample represents over an order of magnitude more data than we can obtain from any other platform.
Method:
Use the TRMM TMI and PR to estimate the spatial distribution of surface rain in TCs. The TRMM satellite was launched November 27, 1997 into a near circular orbit of approximately 350 km in altitude with an inclination of 35° to the equator and a period of 91.5 minutes (Figure1). The TMI is a 9 channel, 5 frequency, linearly polarized, passive microwave radiometric system. The instrument measures atmospheric and surface brightness temperatures at 10.7, 19.4, 21.3, 37.0, and 85.5 GHz. Each frequency has one vertically and one horizontally polarized channel, except for the 21.3 GHz frequency, which has only vertical polarization. TMI has a conical scanning geometry, rotating continuously about a vertical axis, receiving up welling radiation from 49° off nadir. Upwelling radiation is recorded over a 758.5 km swath, covered by 208 ~3.65 km resolution pixels (Figure 2). TMI is similar to the Special Sensor Microwave/Imager (SSM/I) instrument flown on the Defense Meteorological Satellite Program (DMSP) satellites with the addition of polarized 10 GHz channels, common scan geometry for every scan, and roughly twice as many pixels per scan. PR is an active 13.8 GHz radar, recording energy reflected from atmospheric and surface targets. The PR electronically scans every 0.6 s with a swath width of 215 km (Figure 3). Each scan contains 49 rays sampled at ~4.5 km resolution across track. For a given ray, the samples are recorded at 125 m intervals starting a fixed distance from the satellite to below the surface. PR products are referenced to the level of the earth ellipsoid at 250 m resolution.

Once the data is retrieved for the appropriate orbits, the TMI and/or PR rain estimates will be navigated to a storm-centered coordinate system, mapping the estimates into 10 km range intervals and quadrants from the storm center. The rain estimates (R) will be partitioned into 1 dBR (10log10R) intensity classes within each range interval and quadrant to produce a probability distribution function (PDF) of R. Characteristics of each PDF will be tabulated (e.g., mean, standard deviation, median, 90%, 99%, etc.). Stratifications by storm intensity and storm motion will be attempted to test the simple rule of thumb. We also plan to compare the TRMM-derived PDFs to those from SSM/I, WSR-88D, and gage observations. The PDF s of rain will be made available via the Internet for use by the TRMM PIs.


Accomplishment:

The TMI data was ordered (1050 orbits containing 9 Gbytes of data on 7 8-mm tapes). Software is being developed to scan the TMI orbits for each storm and to extract a 10° latitude X 10° longitude window of data around each TC as in Figure 2. During the scan we will also extract the total rain distribution from all 1050 orbits for comparison with the TC distributions. This subset of data will be stratified to produce the rain distributions by azimuthal quadrant and range from the storm center.


Key references:

Kummerow, C., W. Barnes, T. Kozu, J. Shiue, and J. Simpson, 1998: The Tropical Rainfall Measuring Mission (TRMM) Sensor Package. J. Atmos. Ocean. Tech., 15, 809–81

Simpson, J., C. Kummerow, W.-K. Tao and R. F. Adler, 1996: On the Tropical Rainfall Measuring Mission (TRMM). Meteor. Atmos. Phys., 60, 19-36.


Project page scientist index
Project page subject index

Main HRD home page

Last modified: 11/3/2000