Afrouz Bagheri; Bahram Malekmohammadi; Banafsheh Zahraei; Amirhesam Hasani; Farzam Babaei
Abstract
1. Introduction Nowadays, changes in environmentaldynamics including changes in land cover, land use, water supplies and climate are considered to be challenging issues of human communities. Thus, it is especially important to investigate different aspects of land use change and their effects on the ...
Read More
1. Introduction Nowadays, changes in environmentaldynamics including changes in land cover, land use, water supplies and climate are considered to be challenging issues of human communities. Thus, it is especially important to investigate different aspects of land use change and their effects on the past and future trends ofdifferent plains. Identification of previous changes and prediction of future trends help planners and managers to compensate for losses and avoid similar mistakes in future.Therefore, the present study is divided into two parts. In the first part, land usechanges of Lenjanat Plain in the 1990-2015 period are analyzed. In the second part, future land use changes (2015-2035) of the area are investigated.Adjustment coefficient is calculated to show the effect of land use changes on runoff coefficient. Materials and methods 2.1 Study area and its characteristics Lenjanat Plain is a sub-basin of Gavkhuni Wetland located in the central part of the Iranian plateau with the longitude of 51˚ 8' to 51˚45' E and latitude of 32˚2' to 32˚24' N. 2.2 Land use change In the first step, preprocessing of satellite data and preparation of the information were carried out through geometric and atmospheric correction of the digital imageswith the purpose of correcting errors, removing defects in the images and omitting system errors. Landsat-4 and 5, TM sensor, Landsat-7, ETM+ sensor and Landsat-8 OLI sensor were used to evaluate and predict land use changes in the study area. Image selection was performed based on the availability criteria, and Landsat satellite images were thus obtainedfor 1990, 2005, and 2015. In the next step, unsupervised classification was used to create a general understanding of land use classes in the study area as a useful tool for determining training samples. ENVI software was used to identify suitable training samples for classification. To realize the second goal of the study, Marco integration model and a cellular automaton model were used and future land use changes in Lenjanstudy area were predicted for 2035 based on the base map and the assumption that the current trend in land use changes will continue. For this purpose, the Marco and CA-MARKOV modules were utilizedin IDRISI SELVA. CA-Markov model was used to predict land use changes with spatial contiguity and spatial transitions over time. 3. Results and discussion 3.1 Measuringland usechanges Finall and use maps represent the percentage and spatial distribution of each landuse type in the study area in the past, and at present. These maps area also used to evaluate the effects of management on the intensity of land use changes in the study area. Man-made surfaces have almost doubled in the region and reached from 3922 to 7202 hectares. In the past, 3922, 22516, 81613 and 367 hectaresof man-made areas (such as residential and industrial), agricultural lands, barren lands, and riverbeds were located in the study area which have reached 7202, 17943, 82793 and 229 hectares, respectively. 3.2 Prediction of future land usechanges Land use types in 2035 were predicted using CA-Markov chain model. Results indicate that manmade surfaces will exhibit a rising trend and increase from 7,202 to 9,122 hectaresduring 2015-2035 period.To determine the compatibility or incompatibility of actual maps and modelingresults, model validation was performed. In this regard, land usesof the study area was predicted for 2015 through the aforementionedmodel and the predicted map was compared with the actual land use map in 2015. In this method, the Kappa index of 0-1 was used to interpret the results. 3.3 Adjustment Factor Before anything else, the present study have determinedland usechangespercentage. Then, runoff coefficient of the forecast period was divided by runoff coefficient of land use changes in the pastto calculate the adjustment factor.Based on the findings of this study and the land use changesforecasted forLenjanat plain in 2035, the adjustment coefficient for the region equals 1.051. 4. Conclusions The present study aimed to evaluate various criteria affecting the quantity of water resources. Moreover, it has evaluated and determinedadjustment factor. For this purpose, Lenjan plain was used as a representative of the plainsin the country. Five land use types, including man-made areas (such as residential and industrial), agricultural lands, Barren lands, riverbeds and rock beds were identified for 1990, 2005 and 2015. CA-Markov was applied to predict land use changes for 2035. Adjustment coefficient is also calculated to show the effect of land use changes on runoff coefficient
