Detection and monitoring of two dust storm events by multispectral modis images.

During a dust storm event, loose sand/dust from a dry area is picked up and tossed into the air. These particles get together and form a layer of dust. These layers carry large amount of dust from one place to another and affect global meteorological conditions. The present study addresses detection, migration, altitude measurements and spread of two intense dust storm outbreaks events over the Arabian Sea region using multi-satellite observations. The first event occurred on 15 th October 2008 when intense plumes of dust blew over Arabian Sea from the south coast of Pakistan. The second event of the dust storm originated in Arabian desert on March 25 th . The plume extended hundreds of kilometers across southern shore of Oman, over Arabian Sea to the coast of Pakistan. The dust storm lasted over the Arabian Sea till 30 th March. MODIS sensors on both Terra and Aqua Satellites captured images of both events. From the difference in emissive/transmissive nature of the mineral aerosol, suspended dust is distinguished from the ground scene. The algorithms based on split window brightness temperature difference, relations among reflectance at different wavelengths and their ratios and normalized difference indices like Normalized Difference Dust Index (NDDI), Normalized Difference Vegetation Index (NDVI) and Modified Normalized Difference Vegetation Index (MNDVI) were utilized to map the dust events. Selection of an appropriate algorithm is based on nature of the underlying surface, density of dust plumes, presence of clouds, geographic region of the event and time of image acquisition. CALIPSO LiDAR Data is analyzed over the footprint of CALIPSO (Cloud –Aerosol, LiDAR and Infrared Pathfinder Satellite Observation) on the MODIS (Moderate Resolution Imaging Spectroradiometer) scene and altitude of the dust storm is estimated from a profile of total attenuated backscatter. CALIPSO LiDAR measurements are also used to validate the observations. In this study, two dust storm events with dissimilar origins have been studied to evaluate different techniques of dust detection.