Comparing TRMM rainfall retrieval with NOAA buoy rain gauge data

Abstract

This study compares rain rate measurements from the Tropical Rainfall Measuring Mission (TRMM) satellite to rain rate measurements from rain gauges on open-ocean buoys. The rain gauges are part of the instrument package on the Next Generation Autonomous Temperature Line Acquisition System (ATLAS) buoys in the Tropical Atmosphere-Ocean/Triangle Trans-Ocean Buoy Network (TAO/TRITON) array in the tropical Pacific. The rain rate data from TRMM and 25 buoys are collected from January of 1998 to December of 2001. TRMM's 3G68 product provides instantaneous rain rate data averaged over 0.5⁰ x 0.5⁰ latitude-longitude grid boxes for the TRMM Microwave Imager (TMI), Precipitation Radar (PR), and a combined algorithm (COMB). The buoy's rain rate data are averaged over and reported in 10-minute intervals. Buoy data are compared to 1.0⁰ x 1.0⁰ TRMM area-averaged values centered on each of the 25 buoy locations. The 1.0⁰ x 1.0⁰ boxes are composed of four 0.5⁰ x 0.5⁰ 3G68 grid boxes. The OVERLAP subset consists of all TRMM (TMI, PR and COMB) and buoy rain rate data from the observation period of January 1998 to December 2001. The MATCH dataset consists of only those 10-minute periods where both TRMM and the buoy data are available. The rain rates from both subsets are averaged and compared to each other. Scatterplots are drawn, the best-fit line is determined, and the 95% confidence interval on the slope of the best-fit line is calculated. If the confidence interval contains 1.0, the two methods of measuring rain rates are not biased. This study shows that the confidence intervals for the OVERLAP and MATCH TMI versus buoy cases do contain 1.0. The confidence intervals for the OVERLAP and MATCH PR versus buoy and COMB versus buoy cases, however, do not. This study concludes that TMI and the buoy rain rate measurements are unbiased and that the PR and COMB product underestimate rain rate as compared to the buoy.