Simulating Future Patterns of Land Use Changes in Sanandaj Urban Area Using the CLUE-S and MOLAND Models

Abstract
Today, understanding the importance of land-use changes and its affecting factors in urban planning is undeniable. Knowing the ratio of land uses and how it changes over time is one of the critical issues in management planning. Modeling and predicting urban land-use change can help to understand and recognize urban dynamics and it can be considered as an effective and essential tool for planners. The goal of this study at first is to investigate the extent and manner of conversion of existing agricultural and bare lands to built-up areas in the urban area of Sanandaj, and secondly to model the future pattern of these changes up to the horizon of 2032 using the CLUE-S and MOLAND models. The research method used in this study is a combined method based on historical, descriptive, correlation, and comparative methods. This article is an applied research and quantitative and comparative methods have been used to analyze the subject. First, effective factors in the land-use change in Sanandaj urban area were identified based on literature review, observations, and field studies, as well as interviews with local officials and planners, and then, logistic regression as an analytical method is used in SPSS 16.0 software environment to analyze the relationship between land-use variables and independent factors. The results indicate that land-use changes often take place in lands having the highest degree of spatial desirability for that type of land use. The results of both models show a decrease in agricultural and bare lands and an increase in constructed areas. Therefore, in this regard, the eastern and northwestern regions of the study area are facing more land-use changes, and more actions need to be taken to prevent it in the future.
Introduction
Urban development has been very rapid, especially in recent decades which the conversion of agricultural and barren lands into urban uses within the area and city exclusion are its main consequences. Population growth and urban development are very important issues that increase the acceleration of land use changes. Developing countries such as Iran, has experienced a similar trend over the past few decades. So that with the formation of evolution in the socio-economic and political system of Iran, especially since the 1960s and following the land reform law, migration trends from rural to urban areas increased and became the most important factor and source of land use changes.
Land use change models play an important role in understanding the causes, mechanisms and consequences of land use dynamics. At present, various models and methods of remote sensing and GIS are used to study the trend of land use change and their prediction. But in this study, in order to increase the accuracy of modeling and predicting land use changes, a comparative study of CLUE-S and MOLAND models are used. Compared to the other land use change models, the CLUE-S model also includes natural, socio-economic, spatial and non-spatial factors which makes it a more comprehensive, open and comprehensive model. The model of monitoring land use dynamics and land cover patterns (MOLAND) also uses socio-economic information such as population and employment to calculate the amount of land demand and allocates this demand to the space using a sub-model based on cellular automata and considering the natural features and management factors in the area. Combining the two models of CLUE-S and MOLAND can be an effective method for modeling spatial-temporal changes in land use and they are more accurate than other older models such as SLEUTH. Therefore, combining these two models can provide better results than using them separately, which is the innovation of the present study. Sanandaj urban area, especially in recent decades, has witnessed extensive land use changes and now, it has become one of the major challenges. Therefore, the overall purpose of this study is to model land use changes using two models of CLUE-S and MOLAND in the next years.
Methodology
In the present study, historical, descriptive and comparative methods have been used. Data collection was done through review of policy documents, interviews with local authorities, residents, and field observations. To determine the factors affecting land use change, review of documents and interviews with experts and trustees have been used. Also, logistic regression method in SPSS 16.0 software environment has been used to determine the relationships between land use variables and factors affecting it. In this research, to simulate the future pattern of land use changes in Sanandaj urban area in the horizon of 2030, the results of CLUE-S (Dyna-CLUE) model, GIS and MOLAND model (GEONAMICA) have been used. All files are first saved as ASCII files to access Dyna-CLUE and GEONAMICA software.
The present study has been done in several stages. First, the issue and causes of land use changes occur, are discussed. In the next step, the CLUE-S modeling framework is examined. Then, the current land use status of Sanandaj urban area is examined. Factors affecting land use change are identified and data and information related to them are collected. Then, in order to increase the accuracy of modeling and precise the results, the MOLAND model was used to simulate the future pattern of land use changes in the study area at horizon 2030. All information and maps to prepare for the model, first turned into a raster file and then ASCII ones. Finally, according to the results of the model, the future pattern of land use changes in the study area is simulated.
Conclusion
Awareness of land use status in the next years and assessing the damage caused by changes, has a great role in land use management. Controlling and reducing damage requires a lot of costs when the land use changes occur and the damage appears. But it would be easier to control the situation if the simulation conditions and possible damages are assessed in advance. Undoubtedly, the destruction of natural resources and their conversion into residential areas and built-up surfaces in the not-too-distant future has exposed the Sanandaj metropolitan area to crisis, with consequences such as increased flooding, erosion and sedimentation, agricultural pollution and earth mass movements. Studies show that we will see an increasing trend in urban uses and built-up areas and a decreasing trend in agricultural and barren land uses, up to the modeling horizon (2030), which can indicate urban development in the region.
According to the results of the CLUE-S model, between 2015 and 2030, about 2484 hectares of agricultural, barren and undeveloped lands (including arboriculture and forestry, green and middle green areas of roads, gardens and agricultural lands, hills Forest parks) have been reduced in the study area while according to the scenario of fast changes, this amount has even reached about six thousand hectares. The most important reason for this can be considered unplanned development, especially around the city of Sanandaj. According to the results of modeling the future pattern of land use changes in Sanandaj urban area on the horizon of 2030 (Figures 8 and 9), more land use changes will occur in the eastern and northern parts of the study area because in these parts, there are more possibilities for land conversion. Settlements around the city of Sanandaj, especially "Naysar", "Hassanabad", "Dushan", "Qar", "Asavleh" and "Nanaleh", have experienced significant growth in recent decades and are more exposed to land use changes. The existence of agricultural lands around these settlements has been further caused and due to its easy conversion to other land-uses, there is a great potential for further growth and expansion of these settlements. The prediction results using the MOLAND model showed that between 1991 and 2030, about 20% of the area of barren and undeveloped lands (including agriculture and orchards, natural zone, green space and barren) decreased and in contrast , urban surfaces have increased by 3.8 times. This means that the demand for residential, service and generally built-up land uses, up to the modeling horizon (2030) continues to increase and this demand has led to a decrease in agricultural levels and their land-use change. In general, using the results of the two models studied, managers and planners can be informed about the future conditions of the region to include the problem of land use change in their planning.
Keywords
Land use changes, Simulation, physical development, CLUE-S model, MOLAND model, Sanandaj urban area