St.Petersburg Regional Center for Hydrometeorology and Environmental Monitoring,
ROSHYDROMET - Russia
Squall lines, downpours, thunderstorms - basic severe meteorological phenomena connected with convection in Northwest region of Russia in summer time. For their detection and analysis the special satellite data processing system is used. The initial data are the AVHRR/ NOAA multispectral information. The process of detection and analysis of severe convective phenomena is divided into 2 main stages and based on the identification of mass of severe convective clouds.
At the first stage the identification of cloudiness and earth surface is carried out. Thus the original clustering technique of multidimensional brightness histogram of the initial image in a combination with a threshold methods used for identification of cloud and earth surface clusters. For identification of cumuliform, cumulus congestus and cumulonimbus cloud clusters the mean characteristic sizes of cloudy elements for examined region are involved and also the threshold meanings in channels 1, 3, 4 of radiometer AVHRR are used.
At the second stage of satellite data processing on the basis of special models of satellite measurements and cloud optical models the inverse problem of retrieving of important for operative practice cloud characteristics such as cloud top temperature, cloud top height, cloud thickness and liquid water path is solved for each cloud cluster. Thus for each cloud type the certain microphysical model is used. In operative practice the rather simplified cloudy model used for solution of incorrect inverse problem for retrieving cloud parameters allowing receiving estimations of this parameters quickly. Thus the mean error in estimation of cloud top height are within the limits of 1.0-1.5 km.
Severe convective phenomena associate with a mass of severe convective clouds, which operatively retrieved with the abovementioned information satellite processing system.