Integrating ANNs and Cellular Automata–Markov Chain to Simulate Urban Expansion with Annual Land Use Data

Abstract

Accurately simulating and predicting the urban expansion process, especially in expeditious urbanization areas, is an important aspect of managing limited land resources and adjusting flawed land use policies. This research was conducted on the basis of a high-temporal-resolution land use dataset to precisely model urban expansion in a rapidly developing zone by integrating the Artificial Neural Network (ANN), cellular automata (CA), and Markov Chain (MC). An urban suitability index (USI) map was created using ANN and fed to CA–MC to identify possible changed-to-urban cells. Two ANNs, multiple-layer perceptron (MLP) and long short-term memory network (LSTM), were implemented as simulation models for comparison. Due to its ability to capture more temporal information, LSTM outperformed MLP in modeling urban expansion dynamics over a short temporal interval. The simulated results were validated by (fuzzy) kappa simulation and the results revealed that the combination of ANN and CA–MC can precisely model the urban development locations due to its strength in revealing the nonlinear relationship between the expansion process and its drivers. The same model was applied to southern Auckland, and the compared results show that the most simulated variance is caused by the land use policies implemented by different types of governments.

Modelling Land Use/Land Cover Change of River Rwizi Catchment, South-Western Uganda Using GIS and Markov Chain Model

“…For example, a study on the upper Rwizi catchment [64], reported that cropland covered majority of the land cover and increased between 2000-2014, while grazing land reduced; all due to conversions from and to other land cover classes respectively. This is similar to findings by [65], on the Lake Mburo pastoral area, where farmland increased by conversion from wetland area during the 1987-2020 period [26], in the Mt. Elgon region (1978-2020), reported a similar conversion trend.…”

Multi-Scenario Simulation to Predict Ecological Risk Posed by Urban Sprawl with Spontaneous Growth: A Case Study of Quanzhou

“…Simulating fine-scale changes in urban land use is an important way to explore the ecological risks manifested by the loss of the ecosystem services provided by an ecological space or by an increased sensitivity of the ecological environment [ 22 , 23 ]. However, simulations of urban expansion that consider “urban–non-urban” or land use changes involving transitions between multiple land types generally combine impervious surfaces into a single type of urban land [ 24 , 25 ]. Research remains sparse on the simulation of the spatiotemporal evolution of various types of construction land such as industrial land, mining land, urban construction land, and rural settlements.…”

Multi-scenario simulation of land use dynamics and ecological risk: a case study of the liaohe estuary national wetland reserve using PLUS-Markov and PSR models

“…However, prior researches on land use change modeling have mostly concentrated on transitioning between common land types, such as impermeable surfaces, cultivated land, etc. (Hu et al, 2022;Xu et al, 2022), while there is still little knowledge on nature reserves, which are composed mainly of wetlands, such as freshwater swamps, saline swamps, Phragmites australis fields, and Suaeda salsa area, et al (Bao et al, 2019;Athapaththu et al, 2020). Moreover, by taking into account the interactions within different patches, the CA-based model has been widely used to simulate the complicated self-organization behavior of land use (Guan et al, 2011).…”