Estimating GATE Rainfall with Geosynchronous Satellite Images

John S. Stout Space Science and Engineering Center, Madison, WI 53706

Search for other papers by John S. Stout in
Current site
Google Scholar
PubMed
Close
,
David W. Martin Space Science and Engineering Center, Madison, WI 53706

Search for other papers by David W. Martin in
Current site
Google Scholar
PubMed
Close
, and
Dhirendra N. Sikdar University of Wisconsin, Milwaukee, WI 53201

Search for other papers by Dhirendra N. Sikdar in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

A method of estimating GATE rainfall from either visible or infrared images of geosynchronous satellites is described. Rain is estimated from cumulonimbus cloud area by the equation R = a0A + a1dA/dt, where R is volumetric rainfall (m3 s−1), A cloud area (m2), t time (s), and a0 and a1, are constants. Rainfall, calculated from 5.3 cm ship radar, and cloud area are measured from clouds in the tropical North Atlantic. The constants a0 and a1 are fit to these measurements by the least-squares method. Hourly estimates by the infrared version of this technique correlate well (correlation coefficient of 0.84) with rain totals derived from composited radar for an area of 105 km2. The accuracy of this method is described and compared to that of another technique using geosynchronous satellite images. We conclude that this technique provides useful estimates of tropical oceanic rainfall on a convective scale.

Abstract

A method of estimating GATE rainfall from either visible or infrared images of geosynchronous satellites is described. Rain is estimated from cumulonimbus cloud area by the equation R = a0A + a1dA/dt, where R is volumetric rainfall (m3 s−1), A cloud area (m2), t time (s), and a0 and a1, are constants. Rainfall, calculated from 5.3 cm ship radar, and cloud area are measured from clouds in the tropical North Atlantic. The constants a0 and a1 are fit to these measurements by the least-squares method. Hourly estimates by the infrared version of this technique correlate well (correlation coefficient of 0.84) with rain totals derived from composited radar for an area of 105 km2. The accuracy of this method is described and compared to that of another technique using geosynchronous satellite images. We conclude that this technique provides useful estimates of tropical oceanic rainfall on a convective scale.

Save