Land use Changes Modelling Using Logistic Regression and Markov Chain in the Haraz Watershed

Assessment of land use spatio-temporal changes presents the perfect data for managers to elaborate plans. Land use change modeling is one of the methods used by planers to manage land use changes. The present study aims to predict land use changes using logistic regression and Markov chain model in Haraz watershed. Land use maps of the study area were prepared from Landsat images (L5-TM-1988, L7-ETM -2000 and -L8-OLI-2013). For this purpose, image classification method and maximum likelihood algorithm was used in ENVI 4.8 software. Then transition potential modeling was performed using Land Change Modeler (LCM) and Logistic Regression (LR). In order to prediction of land use for 2025, maps of the calibration periods of 1988-2000, 2000-2013 and 1988-2013 using a Markov chain model and hard prediction were used. The results of the calibration periods using the GEOMOD method and its parameters (N(n), N(m), H(m), M(m), K(m), P(m) and P(p)) and kappa coefficients showed that period of 1988-2013 with highest accuracy was selected to predict 2025 land use map. The results of the land use changes showed that over the period 1988-2013, the rate of decreasing in forest, grassland and irrigated land was 4.20, 5.09 and 0.63 percent, respectively. Also during the period residential areas, orchard and bare land, increased 1.28, 2.20 and 6.62 percent, respectively. Most of the changes in this period was transition of forest and rangeland to orchards, residential area and bare land with 8836.4, 5165.1 and 426598.4 ha, respectively. Also the results of land use modeling for 2025 revealed that the area of forest and rangelands will be decrease 2978.18 and 6367.41 ha, respectively and irrigated land, residential area, orchards and bare land will be increase 391.86, 29.38, 1453.42 and 7214.94 ha, respectively.