Tuesday, 22 January 2008: 11:00 AM
Intercomparisons of AIRS, AMSR-E, and CloudSat for improving GPCP high-latitude precipitation estimates
217-218 (Ernest N. Morial Convention Center)
Eric J. Nelkin, SSAI and NASA/GSFC, Greenbelt, MD; and G. J. Huffman, R. F. Adler, D. T. Bolvin, J. Susskind, and J. M. Haynes
The Global Precipitation Climatology Project (GPCP) provides global estimates of monthly precipitation from 1979-present, and of daily precipitation from 1997-present, using a combination of satellite and gauge data. At high latitudes, the estimates are less reliable due to the limited number of gauges and to difficulties in retrieving precipitation over cold/frozen surfaces. Historically, the technique has relied upon the Susskind et al. algorithm applied to TOVS data outside of seasonally-varying boundaries at approximately 45°N-45°S. With the demise of NOAA-14, the final satellite in the TOVS series, during the first half of 2005, the TOVS scheme was adapted and applied to Version 5 Aqua Atmospheric Infrared Sounder (AIRS) precipitation data, available at the Goddard Earth Sciences Data and Information Services Center (http://disc.gsfc.nasa.gov). To maintain long-term consistency in the GPCP record, a seasonal histogram-matching relationship was developed using data from the AIRS/TOVS overlap era, then applied to AIRS data going forward.
As GPCP moves towards a modernized “Version 3” product, one goal is to improve precipitation estimates at high latitudes. The CloudSat Precipitation Radar (CPR), launched in mid-2006, is crucial to this effort, despite viewing only along a nadir track, since all sensors considered are part of the A-train satellite constellation. A two-step procedure is presented. First, CPR estimates are compared against coincident estimates from the high-quality AMSR-E. Because AMSR-E is susceptible to the familiar problems that passive microwave sensors experience in retrieving precipitation over cold surface, this comparison is critical to understanding how best to jointly use AMSR-E and AIRS estimates at higher latitudes. Next, focusing on these latitudes, the calibrated CPR is compared against coincident AIRS data to develop a relationship that will be applied to all AIRS data. The goal is a product that takes advantage of the strengths of multiple satellite observing systems to enhance our understanding of precipitation at higher latitudes, including transferring the AIRS calibration to the prior TOVS data set.
Supplementary URL: