Simulating Scenarios of Future Intra-Urban Land-Use Expansion Based on the Neural Network–Markov Model: A Case Study of Lusaka, Zambia

Abstract

Forecasting scenarios of future intra-urban land-use (intra-urban-LU) expansion can help to curb the historically unplanned urbanization in cities in sub-Saharan Africa (SSA) and promote urban sustainability. In this study, we applied the neural network–Markov model to simulate scenarios of future intra-urban-LU expansion in Lusaka city, Zambia. Data derived from remote sensing (RS) and geographic information system (GIS) techniques including urban-LU maps (from 2000, 2005, 2010, and 2015) and selected driver variables, were used to calibrate and validate the model. We then simulated urban-LU expansion for three scenarios (business as usual/status quo, environmental conservation and protection, and strategic urban planning) to explore alternatives for attaining urban sustainability by 2030. The results revealed that Lusaka had experienced rapid urban expansion dominated by informal settlements. Scenario analysis results suggest that a business-as-usual setup is perilous, as it signals an escalating problem of unplanned settlements. The environmental conservation and protection scenario is insufficient, as most of the green spaces and forests have been depleted. The strategic urban planning scenario has the potential for attaining urban sustainability, as it predicts sufficient control of unplanned settlement expansion and protection of green spaces and forests. The study proffers guidance for strategic policy directions and creating a planning vision.

Keywords

• intra-urban land use
• artificial neural network
• Markov model
• modeling and simulation
• scenario analysis
• Lusaka