Forecasted water demand using Extended Cellular Automata Markov Chain Model: case of Saida and Jezzine regions in Lebanon

摘要

Abstract Forecasting water demand in urban areas is crucial for local officials and water network planners in allocating investments in water infrastructure. Estimates of future climate variation, urbanization, population growth, socioeconomic development, water pricing and water management programs are considered the baseline data to forecast future water demand. However, there is a continuous change in the urban fabric and dynamic transitions in land use. The land use/land cover evolution and the changes of the urban agglomerations and population distribution could additionally affect the water demand. This paper discusses the applicability of the Extended Cellular Automata Markov Chain Model to forecast future water demands in urban areas based on forecasting future land use dynamics. Results generated from simulating the future land use pattern reveal that new urban settlements will take place in areas with very low coverage of water supply service. Furthermore, differences in urban densities provide an overview of the water demand relatively to the population density. It is worth noting that 57.6 of new urban development, taking place between the years 2018 and 2054, is expected to have a population density less than 500 per km2. Further studies are needed to investigate water demand in low density areas since the corresponding housing types may increase the water usage.