Using the T-EFA method in a cellular automata-based urban growth simulation's calibration step

Abstract

Changes in land cover driven by urban sprawl increase the threat of urbanization of forests and agricultural lands. Therefore, monitoring urban sprawl by creating simulation models is frequently carried out to understand sustainable city management. Cellular automata-based models are mostly preferred to reduce the damage led by urban sprawl, and the SLEUTH model is the most well known. Several methods have been developed for the SLEUTH model calibration step, such as optimum SLEUTH metrics and total exploratory factor analysis (T-EFA), to improve the model accuracy. This study aims to create a high-accuracy urban growth simulation model using low-resolution data, investigate the T-EFA method's success in the calibration step, and find the urban sprawl effects on land cover change. Istanbul was selected as our study area due to witnessing its tremendous urban sprawl since the 1950s. According to our results, the urban growth that occurred between 2000 and 2018 could be defined more closely to reality using the T-EFA method, and Istanbul will continue to grow until 2040, with approximately 428.7 km2 of agricultural lands, 553.4 km2 of forests, and 0.1 km2 of wetlands being transformed to urban. In addition, the geologically risky areas under threat of urbanization will increase by 60% between 2018 and 2040.