Detection of spatio-temporal changes in the vegetation of Tehran and satellite cities in association with land surface temperature

Change detection is a process to find the paradoxical regions in different temporal imageries of a similar area. Vegetation is very effective at the absorption of grime and lead, prevention of the spread of contaminations in urban environments, clearing the air, and reduction of heat islands. The need to investigate the decrease or increase in vegetation is extremely important in Tehran as a metropolis, as well as its satellite counties because of the increase in population and construction. The purpose of this research was to investigate the spatiotemporal changes in the vegetation of Tehran and its satellite cities in association with temperature during different temporal periods. The results of this research can be useful in studies concerning urban viability, reducing the effects of urban heat islands, and environmental sciences.

Initially, the extraction and preparation of data were carried out using the ETM+ sensor of Landsat 7 satellite from 2001 to 2015, with June being selected as the hottest month of the study area. Then, the selection of days to be studied and correcting imagery, preparing LULC maps and plotting the area percentage was done. The computation of vegetation indices and built-up areas and the calculation of land surface temperature along with the assessment of the accuracy of surface temperature data were other stages of the research methodology. Finally, the area percentage of each index, as well as the scatter plot and Gaussian function chart were produced and the spatial variation of vegetation was studied.

According to the land use and land cover map (LULC) in 2001, 2005, 2010, and 2015, the vegetation significantly increased in 2015 compared to previous courses. The development of the residential area in the west region was higher than in other regions. In the charts of the area percentage for each land use class and its change over the four selected years, the vegetation percentage has been decreased since 2001, which continued in 2010 and 2015. In this study, the relationships between surface temperature and NDVI and SAVI indices were not linear, which showed that there is another controlling factor. In the normal density function chart, which is usually described by mean and standard deviation, variations of NDVI and SAVI indices were similar in the second and third periods, and the mean increased in these two periods compared to the first period, and the height of curve increased due to the reduction of variance. In this study, the results showed a decrease in the value of LST in the second and third periods from 2006 to 2015. On the other hand, the vegetation area was increasing in the region. From a spatial point of view, Tehran has the highest percentage of class one of NDVI that have no vegetation surfaces.

Regarding the study of NDVI, SAVI and NDBI indices, the overall trend of vegetation cover in the study area was increasing. Considering the values of 15R2"> , the vegetation in Ray County was defunct. In the cities of Robat Karim and Tehran, vegetation cover was increasing. The high growth of satellite towns in the surroundings of Tehran has led to the allocation of land and fields and vegetation to residential areas, which exacerbate the heat islands and the unfavorable conditions of life.