Investigation of Land Surface Temperature Fluctuations Using Time Series of MODIS Sensor Images from Terra and Aqua Satellites Under Climatic Conditions(Study Area: Tehran Province)

Analysis of Land Surface Temperature (LST) time series data significantly improves our understanding of long-term climate change. Since about two-thirds of Iran is covered by arid and semi-arid climates; Therefore, it is important to study temperature fluctuations in the country in order to determine and predict the resulting crises. The purpose of this study was to investigate the trend of temperature changes in Tehran province because in recent years we are facing with overpopulation, which is one of the issues affecting climate change in Tehran province because overpopulation increases consumption of fossil fuels, greenhouse gases emissions and As a result, the temperature rises. In this essay, the trend of temperature changes was investigated using the time series of MODIS sensor images of Terra satellite (since 2002) and the time series of MODIS sensor images of Aqua satellite (since 2003) until the end of 2018. The trend of temperature changes of the received images was determined using TerrSET software ETM analyzer and their significance was determined by parametric methods of correlation coefficient and non-parametric Mann-Kendall significance test (at 1% level). These methods were used to identify the trend of changes in maximum, medium and minimum temperature variables on a monthly and annual basis. The trend estimate of these changes was converted to degrees Celsius by ordinary least squares analysis (OLS). In addition, the trend of minimum temperature data changes of synoptic stations in Tehran province was obtained using regression methods and was compared and validated with trend maps of minimum temperature changes obtained from satellite images. In order to compare the performance of MODIS sensor in Terra and Aqua satellites, the time series of MODIS sensor images of Terra satellite were compared with the time series of MODIS sensor images of Aqua satellite by Linear Modeling analysis and R correlation coefficient. Also, the classified areas of significant increase, decrease and no significant trend obtained from the MODIS satellite sensor were compared with the classified areas obtained from the Aqua satellite sensor by the Kappa index. In addition, in order to identify the ecological effects of the trend of surface temperature changes, the shape file of protected areas and national parks of Tehran province was added to the maps of the trend of temperature changes in Arc Gis software. According to the results, the maximum, average and minimum monthly temperatures during the years under study were almost without a significant trend However, a significant increase in surface temperature was observed in the study of annual variables in most parts of Tehran province, especially in Tehran city, desert and plain areas. Also, the irrational or sudden increase in temperature in an area has reasons such as uncontrolled construction, proximity to industrial towns and drying of rivers, lakes, wetlands, etc. Comparison of the minimum temperature trend of satellite images with the minimum temperature trend of synoptic stations datas in Tehran province showed 98.3% similarity. Based on this, it can be said that there is an acceptable agreement between the trend of changes in satellite images and the trend of changes in terrestrial data, and because the degree of concordance of the results obtained from the Aqua satellite is more similar to the data of the synoptic stations than the results obtained from the Terra satellite, it is therefore recommended to study the temperature changes of the Aqua satellite. Comparison of MODIS sensor images time series of Terra satellite with Aqua MODIS sensor time series in most parts of Tehran province showed a high correlation coefficient (99% similarity). Except Tehran city and the east of Tehran province (Firoozkooh city) which can be the difference due to housing, human interference, pollution due to higher population density and agricultural use of land and harvest. Also, kappa index is below 0.7, which indicates the lack of similarity in the classification of significant areas in the whole province by the MODIS sensor between the two satellites. Based on this, it can be said that although MODIS sensor is recommended for long-term changes and better understanding of the change process but there seems to be uncertainty in the interpretation of the data obtained from the classification of MODIS sensor images. Therefore, care must be taken in choosing the type of satellite for the MODIS sensor. Based on the results obtained from identifying the ecological effects of temperature changes, Jajroud Protected Area has a trend of more temperature changes than other protected areas in Tehran province. Considering the trend of rising land surface temperature and its role in increasing evapotranspiration, we should look for solutions to better manage water resources and improve its exploitation methods, especially in agriculture and industry in Tehran province.