Spatiotemporal dynamics of urban expansion and its simulation using CA-ANN model in Ulaanbaatar, Mongolia

Understanding, analysis, monitoring and modelling of urban expansions are most important for urban planners in the process of sustainable development. This article investigated spatio-temporal process of urban expansion using cellular automata (CA) model integrated with artificial neural network (ANN) method. This integrated model was calibrated with data from 1974 to 2018, and used to predict expansions for 2030 and 2040 with two datasets of explanatory variables including slope, forest, river, specially protected areas, road, railroad and urban centres in an urban and non-urban area. The urban suitable area was estimated by criteria evaluation weights and analytic hierarchy process (AHP). The results show that the urban growth of Ulaanbaatar was equal to +36,201 ha (84.8%) for the period of 1974-2018 and this city will be projected to reach approximately +43,111 ha by 2030 (+50.3% from 2018) and +111,441 ha by 2040 (+44.3% from 2030 and totally +72.3% from 2018).

Automated summary

This article investigates the spatio-temporal process of urban expansion using cellular automata and artificial neural network models. The results predict that Ulaanbaatar city will continue to expand in the coming decades.