Application of Cellular Automata and Markov Chain model for urban green infrastructure in Kuala Lumpur, Malaysia

Abstract

Kuala Lumpur of Malaysia, as a tropical city, has experienced a notable decline in its critical urban green infrastructure (UGI) due to rapid urbanization and haphazard development. The decrease of UGI, especially natural forest and artificial forest, may reduce the diversity of ecosystem services and the ability of Kuala Lumpur to build resilience in the future. This study analyzed land use and land cover (LULC) and UGI changes in Kuala Lumpur based on Landsat satellite images in 1990, 2005, and 2021and employed the overall accuracy and Kappa coefficient to assess classification accuracy. LULC was categorized into six main types: natural forest, artificial forest, grassland, water body, bare ground, and built-up area. Satellite images in 1990, 2005, and 2021 showed the remarkable overall accuracy values of 91.06%, 96.67%, and 98.28%, respectively, along with the significant Kappa coefficient values of 0.8997, 0.9626, and 0.9512, respectively. Then, this study utilized Cellular Automata and Markov Chain model to analyze the transition of different LULC types during 1990–2005 and 1990–2021 and predict LULC types in 2050. The results showed that natural forest decreased from 15.22% to 8.20% and artificial forest reduced from 18.51% to 15.16% during 1990–2021. Reductions in natural forest and artificial forest led to alterations in urban surface water dynamics, increasing the risk of urban floods. However, grassland showed a significant increase from 7.80% to 24.30% during 1990–2021. Meanwhile, bare ground increased from 27.16% to 31.56% and built-up area increased from 30.45% to 39.90% during 1990–2005. In 2021, built-up area decreased to 35.10% and bare ground decreased to 13.08%, indicating a consistent dominance of built-up area in the central parts of Kuala Lumpur. This study highlights the importance of integrating past, current, and future LULC changes to improve urban ecosystem services in the city.