Characteristics of convective cloud cluster formationover Thailand through satellite image analysis

Sammanfattning: Weather forecasting relies on the availability of observational data as input parameters. However,such data are not readily available, because of difficulties to collect weather data due toinaccessibility to many places in the world, such as oceans or mountain regions. For this reason,satellite surveillance is a suitable tool to observe the atmosphere in regions where it is notpossible by other means. This master thesis is a study of convective cloud cluster formation over Thailand, conductedthrough satellite image analysis. Characteristics of cloud cluster formations are investigatedthrough an implementation of the Maximum Spatial Correlation Technique (MASCOTTE),described by Carvalho and Jones (2001). This method allows tracking of convective cloud systemsthrough region based analysis of satellite images. The aim of this study is to investigate whether satellite image analysis, through the implementationof the MASCOTTE methodology, can provide characteristics of convective cloud systems,in order to discern convective systems by intensity, accurately enough to be able to discernsevere thunderstorms from ordinary thunderstorms. The annual distribution of the occurrenceof life cycles detected through the analysis is studied, as well as their monthly distribution ofmean and maximum life times. Moreover, the yearly distribution of life cycle mean and minimumbrightness temperatures are analysed, as well as the number of detected split and mergeevents. This is followed by a comparison of life cycle structural properties to investigate thepossibility to use individual parameters, alone or in combination with each other, as indicatorsof the degree of convective activity within life cycles. Yearly distributions were studied in order to verify if this method could reveal seasonal variations,such as the onset period of the wet season, in terms of the occurrence of life cycles andtheir life time. The findings of this study verified that the most convectively intense life cycles exist under theinfluence of the Inter Tropical Convergence Zone (ITCZ), during the onset and beginning ofthe monsoon season. Analysis of life cycle structural properties, showed that properties likemean and minimum brightness temperature as well as fractional convective area, could be usedas indicators to discern between life cycles with different level of convective activity. However,it is concluded that studies, including ground-based remote sensing technologies such asRADAR/LIDAR, as well as data from rawinsondes, needs to be conducted in order to clarifyif it is possible to use this methodology to successfully discern severe thunderstorms fromordinary thunderstorms.