علوم زیستی دریا
Saeid Farhadi; Hossein Mohammad Asgari; Ali Dadolahi Sohrab; Seyed Mohammad Jafar Nazemosadat; Sayyed Hossein Khazaei
Abstract
Dust prediction such as prediction of wind and rain needs to synoptic information to the earth's surface, upper layers of the atmosphere, the prediction maps of land surface and upper levels as well as using radar and satellites. The purpose of this study, use of remote sensing technology and MODIS images ...
Read More
Dust prediction such as prediction of wind and rain needs to synoptic information to the earth's surface, upper layers of the atmosphere, the prediction maps of land surface and upper levels as well as using radar and satellites. The purpose of this study, use of remote sensing technology and MODIS images to estimate dust optical depth on the Persian Gulf surface and estimating the linear correlation relationship between the dust measurements in ground and atmospheric. The dust optical depth calculated using the code developed in MATLAB software. Evaluation of extracted data conducted using Pearson correlation coefficient, RMSE and RMSD index. In this study, optical depth obtained from image processing compared with the optical depths obtained from AERONET network. The evaluation results showed a high and significant correlation between the obtained optical depth and optical depths obtained from AERONET network (R2=0.99). The best and most suitable mode demonstrated for 1.243 and 1.643 bonds. At all stations, AOD value obtained from satellite image is bigger than AOD amount corresponding to the AERONET station and the algorithm used has overestimated. The cause of this more estimate can be use of limited particle's effective radius, because the scope of this effective radius is limited at the distribution of particle size in log-normal. Error resources at the retrieving particulate matter was defined such as sensor calibration error, pollution on the radiation angle, or poor predictor of water reflection.