Modeling the city development of Bojnourd with Logistic regression

Describing and quantifying the spatiotemporal dynamics of urban expansion in developing countries play a crucial role in determining the mechanisms of urban growth and decision making processes. Urban dissonant development is one of the most important issues in land using, and predicting urban development is not simply possible, but it needs a model to consider the complex nature of urban processes. Urban land uses constitute a complex system which is controlled both by human activities and spatial-temporal dynamics. The rapid population growth and urbanization in developing countries provoke uncontrolled urban sprawl and correspondingly pose a threat to the environment. This urbanization process in territorial structures led to a higher complexity and pressure on the natural environment Logistic regression is one of the experimental predicting model. Determination of weight of mobile factors based on empirical rather than expert knowledge is one of the benefits of this model. Another superiority of this model is the ability to import more variables. Considering the importance of this issue, this study is working around the modeling the development of Bojnourd, using logistic regression in 28 years. This method is a descriptive and analytic model that uses 9 independent variables for modelling. The Pseudo-R2 and ROC, respectively were 0.2835 and 0.9335 that are in an acceptable range. Therefore, this modeling is confirmed. Evaluating the sensitivity of the model, using the method of eliminating independent variables, represents high impacts of tilt and its distance of water surfaces in modelling. Southern urban areas and rural areas are the most suitable areas for the development in the future.