Land use/land cover change detection and prediction using the CA-Markov model: A case study of Quetta city, Pakistan
Keywords:
Change prediction; Markov Chain; Cellular Automata; GIS and Remote Sensing; Land use land cover change detection; QuettaAbstract
Background: Land use/land cover changes are the results of rapid urban growth and human activities. The anthropological activities, such as growth in population, rapid urbanization, and fast economic advancement, have modified the surface of the earth, which causes change at a local level and worldwide. Therefore, LULC monitoring and modelling are important for sustainable urban development.
Objectives: The current study aims to detect changes in land use/ land cover from 1998-2018 and to predict changes for the year 2028 using the integrated Cellular Automata-Markov model in Quetta city, Pakistan.
Methods: Three temporal satellite imageries were used for the detection of changes during 1998, 2008, and 2018. Maximum Likelihood Classification techniques were used for classification and change detection, whereas, Cellular Automata-Markov integrated model was used for the prediction of 2028. The standard kappa coefficient was used for assessing the validity of the model.
Results: The result shows an increase (27.1km2) in the built-up area, while a decrease (15.4 km2) in open spaces and 11.7 Km2 in green areas from 1998 to 2018. Moreover, the Prediction result shows that the green area and open spaces would likely be decreased (2.7 km2) and 3.86 km2 from 2018 to 2028 respectively, whereas a slight increase (6.56 km2) in built-up is expected from 2018 to 2028.
Conclusions: This study concluded that Quetta city had witnessed LULC changes over the last 20 years. The current study revealed that the condition of green areas and open spaces are much precarious due to the rapid urbanization in the city. Whereas, the alarming increase in the built-up area creates many complications for the current and future planning process. The use of GIS and RS can be used effectively for detecting and predicting LULC changes. The present study can be used as a direction for other studies using projected LULC models.
