An algorithm to detect tropical deep convective clouds through AMSU-B water vapor channels

Abstract

An algorithm to detect tropical deep convective clouds and deep convective overshootings based on the measurements from the three water vapor channels (183.3 GHz±1 GHz, 183.3 GHz±3 GHz and 183.3 GHz±7 GHz) of the Advanced Microwave Sounding Unit-B (AMSU-B) is presented. This algorithm is an improved version of the method of Hong et al. (2005). The proposed procedure is based on the statistical analysis of seven years’ (2001–2007) measurements from AMSU-B on NOAA-16. From the 1-d histograms of the brightness temperature of the three water vapor channels and the 2-d histograms of the brightness temperature difference between these channels, new thresholds for brightness temperature differences and the brightness temperature of channel 18 (183.3 GHz±1 GHz) are suggested. The new algorithm is employed to investigate the mean distribution of tropical deep convective clouds and convective overshootings from 30°S to 30°N for the years 2001 to 2007. The major concentration of deep convective clouds and convective overshootings is found over the Intertropical Convergence Zone (ITCZ), the South Pacific Convergence Zone (SPCZ), tropical Africa, South America, the Indian Ocean and Indonesia with an average fraction of 0.4%. In terms of these clouds we identify the secondary Intertropical Convergence Zone located in the eastern South Pacific and parallel to the main ITCZ in the North Pacific. The convective overshooting is more frequently observed over land than over the ocean.