Modelling the urban growth of Adana using remote sensing and geographical information systems

The objective of this study was to suggest a predictive modelling system, capable of depicting the impacts of different land use or land management policies for the future urban growth of Adana City considering the three different development policies including; (1) current trends, (2) managed trends, (3) ecologically sustainable growth.Cellular automata (CA) has gained attention from researchers attempting to simulate and predict spatial patterns of urban development. These models require that space should be represented as a grid of cells that can change state as the model iterates. Given its success with regional scale modelling, CA based modelling approaches such as, SLEUTH (slope, land use, exclusion, urban extent, transportation, hillshade), multiple regression, decision tree (regression and classification tree), artificial neural network and Markov chain models were adopted to achieve the modelling process for the year 2023 of Adana. 1967-1977 CORONA airphotos, 1987-1998 SPOT and 2007 ALOS AVNIR-2 satellite images were used for this study. The model was calibrated using historic time series of remotely sensed data and future growth was projected out to 2023 assuming three different policy scenarios.The SLEUTH model and Markov Chain were resulted in the largest overall accuracy of 75 % and 72 % respectively. LR and YSA yielded the least accurate results with an overall accuracy of 66 %. Different modeling approaches have their own merits and advantages. However, the SLEUTH model was the most accurate for handling the variability present in urban development in Adana City