Modeling urban land use dynamics using Markov-chain and cellular automata in Gondar City, Northwest Ethiopia

Abstract

Modeling urban land-use dynamics is critical for urban experts’ and infrastructure managers’ planning. This study attempts to explore the land-use/land-cover (LULC) dynamics of Gondar using satellite images from 1984 to 2020. Markov-Chain and Cellular Automata (MC-CA) models have been recognized as performing well in predicting urban land-use change. However, only a few models work in Ethiopia in general, and no study in Gondar has applied this approach to study urban land-use patterns. Therefore, Gondar land-use/land-cover changes of Gondar were predicted using the MC-CA model in IDRISI. The built-up area in Gondar city covered 1 413 ha (3% of the total area) in 1984 and increased to 2 380 ha (5%) in 1994; 21 153 ha (45.5%) in 2004; 22 622 ha (48.7%) in 2014; and 23 427 ha (50.5%) in 2020. The area has been predicted to reach 57.5% in the 2050s, showing a faster increase that will cause a very vast loss of farmland. This will increase urban sprawl challenges as well as overall environmental disequilibrium in the preceding decade. Thus, innovative and careful structures and systems in urban planning are required to secure a sustainable urban future and to make our cities livable and competitive in the paradigm of sustainable cities.

Keywords

• Modeling urban growth
• Markov chain
• Cellular automata
• Remote sensing
• IDRISI