Effects of Spatial and Temporal Land Use Changes and Urban Development on the Increase of Land Surface Temperature Using Landsat Multi-Temporal Images (Case study: Gorgan City)

Introduction More than 50% of the world population lives in urban areas. It is predicted the value will be increased to 69.6% by 2050. In recent decades, urban population is rapidly increasing due to the natural growth of cities, the migration from villages to cities, climate change, reduction of water resources, loss of agricultural lands, animal husbandry and other factors. These factors have led to physical expansion of the cities and the subsequent destruction and reduction of green spaces and forests, and increase of streets, buildings and asphalt roads. These changes in land use and land cover in urban areas cause many environmental problems and warming of the temperature of the city and its surroundings. Gorgan city as one of the northern cities of Iran is noticeable in urban physical expansion and land use change mainly due to conversion of agricultural landuse and green space into built-up areas. These reasons have created a special climatic condition in terms of air temperature, humidity and precipitation. The purpose of this study is to investigate the increase in temperature as a result of changes in the various landuse and the impact of each landuse on the increase in surface temperature and identifying effective landuse to better management. Methodology In the present study, we have used Landsat satellite images in 1992, 2001, 2009 and 2015 in the sensors of TM5, ETM+, and OLI/TIRS. In order to complete the input parameters for mapping the surface temperature using satellite images, we have used MODIS water vapor product. To provide control points, we have used field views, Google Earth images, and topographic maps prepared by “Iran National Cartographic Center”.   After  providing Landsat time series, we applied preprocessing steps including atmospheric and geometrice correction. Then, the images were classified by Support Vector Machine method. They were classified into four classes including built up, fallow, agriculture and green space. After classifying the control points, the accuracy of the images was calculated. In the next step, we have used the Mono Window algorithm to obtain surface temperature for each image. At the end, we investigated the changes between different images and their relationship with the Earth's surface changes.   Results and discussion The results of landuse changes in Gorgan indicated that during the first period (1992-2001), the extent of fallow and green space increased 48.55% and 31.95%, respectively. The agricultural and green spaces decreased 68.68% and 5.9%, respectively. This is the most important cause of this decline in agricultural landuse during this period in the fallow landuse. In the second time period (2001-2009), the area of ​​green, agricultural and built up landuse increased by 17.1%, 86.59% and 14.51%. The fallow landuse because of cultivation decreased about 18.68%. Also, in the third time period (2009-2015), the extent of the built up and fallow landuse was increased by 12.24% and 7.84%, respectively. The area of the green spaces and agriculture landuse is decreased by 0.72% and 29.49%, respectively. The use of green space due to its particular geographic location, including special topographic conditions, has not changed during the study period.   The highest temperature related to the fallow landuse, because of this increase in temperature for the fallow is the thermal capacity and low heat transfer capacity of the dry soil. Also, the highest temperature is related to green space landuse. This is resulted from evapotranspiration for reducing the temperature for the green space landuse. The variation in temperature classes is different. The very cold temperate class has a faster rate of reduction, so that the area of ​​5875.51 hectares in 1992 changed into an area of 1260.1 hectares in 2015. Also the normal and hot temperatures class in these years had the growing trend. The area of the warm class was zero in  1992. It increased by 319.73, 1226.91, and 1686.13 hectare in the years 2001, 2009 and 2015, respectively.   Conclusion   The results of the image classification in the research indicate a positive effect of the NDVI index and the LST map to increase the accuracy of image classification. Landuse changes indicated that the most changes were observed between the agriculture and fallow landuse. If this trend continues, other landuse will undergo fundamental changes. The trend of temperature changes in the earth surface is an increasing and the highest temperature is related to the fallow and built up landuse. Also, the highest increase in temperature is related to the changes in the green space to fallow landuse. Investigating the relationship between the characteristics of vegetation density and the earth surface temperature indicates that different classes of land use/cover, the presence of vegetation could decrease the surface temperature during study period. It was found that surface temperature in dense urban areas were higher than those in other areas. Hence, it can be noted that the role of vegetation in reducing the surface temperature of the city was important. With studying the temperature classes in the study area, it showed that cold temperatures classes have decreasing trend and warm temperature classes have increasing trend because of the changes occurred in landuse.