كليدواژه :
گردوغبار , بازتابندگي سطح , عمق اپتيكي , دماي سطح , سنجش از دور
چكيده فارسي :
در اين مقاله هدف بهبود بخشيدن به مدل OMI در استخراج عمق اپتيكي براي آشكارسازي مناطق متأثر از گردوغبار با استفاده از تغييرات بازتابندگي سطح، دما و خصوصيات طيفي گردوغبار مي‌باشد. داده‌هاي مورداستفاده شامل بازتابندگي سطح OMI، عمق اپتيكي OMI، كسر ابر OMI، دماي سطح MODIS به همراه عمق اپتيكي AERONET مي‌باشد. تصاوير مورد استفاده همزمان و در باند‌هاي يكسان از دو منطقه در الجزاير و نيجريه واقع در قاره آفريقا مي‌باشد. براي آشكارسازي ميزان گردوغبار پيكسل‌هاي خاصي تعريف گرديدند كه در منطقه الجزاير پس از حذف اثر ابر با مقايسه مقادير بازتابندگي سطح و عمق اپتيكي به رابطه‌اي ميان اين دو بدست آمد كه با مقادير به‌دست‌آمده از AERONET، همبستگي 0.788 و 0.148rmse= براي كانال nm500 به دست آمد. كه در مقايسه با مقادير اوليه با اندازه‌گيري‌هاي زميني،0.13 همبستگي بالاتر و دقت 0.24 بالاتر را گزارش مي‌كند. اما براي نيجريه كه جنس گردوغبار از نوع سطح است به دليل عدم كارايي بازتابندگي سطح، از رابطه بين گردوغبار و دما استفاده گرديد. با اين فرض كه با ارتفاع گرفتن گردوغبار، دما كاهش يافته و با نبود گردوغبار، دما افزايش مي‌يابد. براي اعتبار سنجي داده‌هاي حاصل از مدل از مقادير AERONET استفاده شد. در اين مقايسه، ضريب همبستگي 0.93 و 0.166rmse= براي كانال nm500 نانومتر به دست آمد. درنهايت براي آشكارسازي ميزان گردوغبار در هر نقطه جهان بايد نوع پوشش منطقه، مقدار بازتابندگي سطح و دماي سطح همچنين اطلاع از شرايط ابر منطقه در اختيار باشد. نتايج به‌دست‌آمده از تحقيق نشان مي‌دهد كه روش اول را براي پوشش گياهي و روش دوم براي نقاط بياباني مناسب خواهد بود.
چكيده لاتين :
Dust properties can be sensed remotely from space and the ground. The main problem of remote sensing is that it cannot measure dust directly and collect data about dust by optical behavior of the Earth-Atmosphere System. Nevertheless, the advantage of remote sensing techniques is that they measure dust properties in ambient environment, and they can have access to the mass of data over a large geographical area via satellite. With regard to sampling, field measurements are made at regional and pin point scale regarding both the surface and volume which can be problematic. Although the optical depth of dust derived from satellites are less accurate than ground-based observation, satellite-based remote sensing can supplement ground-based remote sensing in their methods of dust observation since satellite-based remote sensing covers large areas of research. Therefore, this combination can be beneficial and cost-effective in calculating the Air-Quality Index. Aura is the chemistry mission of NASA with the overall objective to study the chemistry and dynamics of Earthchr('39')s atmosphere from the three observation platforms on ground through the mesosphere. The observation of the complex interactions of atmospheric constituents contributes to global environmental health. This thesis aims at modifying OMI model in deriving the optical depth to detect dust affected region using changes in surface reflectance, temperature, and dust spectral properties. The remote sensing data sets included OMI surface reflectance, OMI optical depth, OMI cloud fraction, MODIS surface temperature, and AERONET optical depth. The images included Al Jazeera and Nigeria in Africa, which were close enough in time and band. To detect the dust, particular pixels were defined. Regarding Al Jazeera, after removing the cloud effect, the correlation coefficient of surface reflectance and optical depth was 0.788 and rms=0.148 for channel 500nm. The findings revealed 0.13 higher correlation coefficient and 0.24 higher accuracy in comparison with ground-based observations. Considering Nigeria, the correlation coefficient of dust and temperature was calculated since dust is of surface type and surface reflectance could not be taken into account. The presupposition was that altitude and temperature are inversely related. This means that the higher the altitude, the lower the temperature. To assess the validity of study, the comparison of computations based on AERONET measurement was made. The results showed correlation coefficient= 0.93 and rms=0.166 for channel 500nm. Finally, to detect the amount dust in different regions of world there is a need of information on the vegetation type, the amount of surface reflectance, temperature, as well as clouds of that region. The findings of this study suggested that the first method seemed appropriate for vegetation and the second method for desert areas.
