Evaluating the role of the Baiji Refinery in the development and dynamism of the city and surrounding settlements

Abstract:

The current research tries to understand the changes in urban land use in Biji city with regard to the three time periods of 2013, 2018 and 2023, which has been carried out using three different methods and algorithms. Urban land use classes have been evaluated in 6 different classes. According to the two objectives of this research, i.e. to investigate the effect of Biji Refinery on the development of settlements, especially urban ones, and to evaluate the changes in land use compared to urban land use, first we examined the land use changes in different periods of time, and in the next step, the land use changes were measured relative to each other. In order to evaluate its data, this research has used three non-parametric algorithms of support vector machine and random forest and decision tree in this research to extract the land use map from the OLI sensor image of Landsat 8 satellite and finally to choose the most suitable classification algorithm for Biji region in Iraq was conducted, the best algorithm that gave a favorable answer and had a lower error rate and was closer to reality is the random forest algorithm, which is very similar to the results of the research of Rinalz et al. (2013), which leads to the conclusion that The overall high accuracy of our classification map supports growing evidence that RF is a reliable classifier for heterogeneous landscapes. Land cover types other than cerrado were the most accurate with a combined accuracy of 91%. The results of the current research from the random forest show that the urban development around Beji Refinery was high, but an acute humanitarian crisis such as the ISIS war not only stopped this development but accelerated it. According to the random forest change of use from 2018 to 2023, it shows that after the war with ISIS, there is a change of use from desert areas to urban areas, and after that, these agricultural lands are added to urban areas. Shunde, also based on random forest method from 2013 to 2023, it is agricultural land use that has the most change towards urban areas, and agricultural and desert lands are in the next ranks of conversion, the continuation of this trend and indiscriminate urban development It leads to the destruction of the environment, especially useful agricultural lands, in fact, the result of the development of Beji Refinery leads to urban development in urban lands.