7B.3
Comparing Detection Methods of Deep Convective Clouds in China with AMSU-B and FY-2C
Zhoujie Cheng, Institute of Aviation Meteorology, Beijing, , China; and Y. Zhu, J. Liu, J. Bai, and W. Li
Detection and identification of deep convective clouds play a key role in the studies of mesoscale weather phenomena, especially in the tracking and forecasting convective storms. The satellite-based techniques including passive measurements of visible, infrared, and microwave radiances have been studied in the analyses on the deep convective clouds, some optical and microwave criteria have been presented over the years. The purpose of this paper is to compare these algorithms and to investigate the microwave and optical characteristics differences between the techniques in determining the deep convective clouds in China. The optical and microwave algorithms are presented, which are based on FY-2C and AMSU-B image data. By detecting and analyzing deep convective clouds in some cases, the techniques are investigated, and by matching surface conventional data the results of various methods are validated and compared.
The microwave frequencies have the ability to penetrating clouds and the effects of clouds and precipitation on microwave radiances at the AMSU-B channels have been examined through simulations and observations. Some criteria are suggest to discriminate deep convective clouds, such as the difference between brightness temperatures at 183.3±3 and 183.3±7GHz, 89 and 220GHz, or differences between three water vapor channels. The present study derives methods for detecting deep convective clouds in two ways. The first one is the brightness temperature thresholds from two window channels at 89 and 150GHz, the criteria of TB89<240K and TB150<220K are tested to detect deep convective clouds, the thresholds are selected by statistical brightness temperature of three years from 2004 to 2006 in summer from June to August. The second one is the differences between the three water vapor channels. For three water vapor channels near to the water vapor absorption line centered at 183.3GHz(183.3±1, ±3, ±7GHz ), the criterion of three water vapor channels is ΔT17>ΔT37>0, ΔT17>ΔT13>0, ΔT37>ΔT13>0, being used to detect deep convective clouds.
The optical remote sensing signals (including VIS, NIR, IR, and WV) based on FY-2C are came from the cloud top information and the optical techniques include infrared brightness thresholds detection of cloud top temperatures, the water vapor and infrared window temperature differences determination, and the classification of cumulonimbus clouds correlating with deep convective clouds applying with infrared/water vapor spectral features. In order to compare with the microwave techniques and investigate the advantages and limitations of these methods, different thresholds from IR1 are selected in this study, which include -65°C, -58°C, -43°C, -40°C. The next method of positive differences between WV (6.5µm) and IR(10.7µm) channels is tested to identify deep convective TB6.7-10.7>0. The third technique based on the stepwise classifier using IR and WV spectral features are applied to discriminate cumulonimbus clouds, which has been studied and tested to be a stable approach to classify the cumulonimbus clouds from other cloud types, while the aim of this study is to test the identification areas and the efficiency of the classifier in monitoring deep convective clouds.
The results show that microwave brightness temperatures from window channels can discriminate deep convective clouds efficiently, while the brightness temperatures of 89GHz are affected by surface features and the cold water surface are mistaken to convective clouds, the brightness temperatures of 150GHz are weakly influenced by surface characteristics, the detection areas are coincident with those from water vapor channels microwave brightness differences identification, which can identify the deep convective clouds well and depend on the thresholds less. The FY-2C different infrared brightness thresholds are given the detection regions are more or less, the single threshold are applicable to the local areas. The water vapor and infrared window temperature differences determination areas are smaller. The stepwise cluster can identify cumulonimbus clouds correlating with deep convective clouds applying with infrared/water vapor spectral features, the detection areas is coincident with AMSU-B detection areas, and the surface conventional data can validated the results, which include hazards weather and cumulonimbus clouds.
Session 7B, Developments in Use of Satellite and Radar Data
Tuesday, 28 October 2008, 10:45 AM-12:00 PM, South Ballroom
Previous paper Next paper