فهرست مطالب مجید محمدیوسفی

The prediction and modeling of land use changes is important for understanding the quantity and quality of possible future changes. The purpose of this research is to monitor land use changes in the past and to investigate the possibility of predicting them in the future using the LCM model in the Beheshabad watershed of Chaharmahal va Bakhtiari province. In this study, Landsat 5 TM sensors images of 1991 and 2008 and Landsat 8 OLI sensors images in 2016 were used and analyzed. Images of all three periods were classified into five categories of range lands, urban-construction areas, agricultural lands, garden lands and Bare lands. The prediction of land use status for 2016 was carried out using user maps of 1370 and 1387 using the LCM model based on artificial neural networks and Markov chain analysis. For this purpose, the spatial variables of the distance from the pastures, the distance from the residential and urban areas, the distance from the agricultural land, the distance from the garden, the distance from the Bare Lands, distance from the river, distance from the road, elevation or digital elevation model, slope and direction of the slope, are used as factors affecting changes in artificial neural network. The results of modeling the transmission force using artificial neural network in most. of the sub-models showed high accuracy (62 to 94 percent). The total error in modeling for the year 2016 obtained approximetly 23%, which reflects the large mismatch of the predicted image of the model with the image of the Earth reality and the acceptability of the model. The prediction results for the years of 1420 and 1429 showed that the area of range land and bare lands would be reduced and the area of urbanconstruction areas, garden lands and agricultural lands would becincrease.

In order to plan and manage the land and its changes, it is necessary to identify and evaluate the factors affecting it. Land use / cover changes are one of the main factors in global environmental change that is defined as a change in the type of land use; it is one of the major factors changing hydrologic flow, land erosion and destruction of biodiversity. The main purpose of this study was to assess the trends of land use changes in Beheshtabad Watershed of Chaharmahal and Bakhtiari Province with an area about 3847 square kilometers by using remote sensing and GIS during a 25-year period. In this research, first, analyzing and pre-processing the satellite images of Landsat 5 TM sensors from 1991 and 2008 were done, and Landsat 8 of OLI sensor of 2016 was applied. Then, by using the hybrid classification method, 5 land use classes including pasture lands, urban-building lands, agricultural lands, garden lands and bare lands, land use maps for the three time periods were prepared. The overall accuracy of the obtained land use maps for 1991, 2008 and 2016, was 92.17%, 94.29% and 93.41%, respectively, indicating the acceptable accuracy of the maps. Then, the process of land use change and the contribution of each land use classes and the percentage of changes in each land use class were determined in two study periods. The results of this study showed some changes occurred in the studied watershed. The total area of pasture lands during two periods indicated the decreasing trend, but urban-building and garden lands during two periods represented the increasing one. Agricultural lands during the first period indicated the decreasing trend and during the second period showed the increasing trend, while bare lands during the first period showed the increasing trend and during the second period, reflected the decreasing trend. In general, it should be noted that in the Beheshtabad watershed, we could see an increase in the replacement of pastures by urban-building class, rainfed agriculture, gardens, and bare lands, the incidence of destruction in the region.