Image analysis of geostationary meteorological satellite for monitoring movement of mesoscale convective systems over Tibetan Plateau

Disaster weather forecasting is becoming increasingly important. In this paper, the trajectories of Mesoscale Convective Systems (MCSs) were automatically tracked over the Chinese Tibetan Plateau using Geostationary Meteorological Satellite (GMS) brightness temperature (Tbb) from June to August 1998, and the MCSs are classified according to their movement direction. Based on these, spatial data mining methods are used to study the relationships between MCSs trajectories and their environmental physical field values. Results indicate that at 400hPa level, the trajectories of MCSs moving across the 105°E boundary are less influenced by water vapor flux divergence, vertical wind velocity, relative humidity and K index. In addition, if the gravity central longitude locations of MCSs are between 104°E and 105°E, then geopotential height and wind divergence are two main factors in movement causation. On the other hand, at 500hPa level, the trajectories of MCSs in a north-east direction are mainly influenced by K index and water vapor flux divergence when their central locations are less than 104°E. However, the MCSs moving in an east and south-east direction are influenced by a few correlation factors at this level.