Algorithm

Engineering

Ecology

Biology

Environmental economics

Markov chain

Constraint (computer-aided design)

Economics

Civil engineering

Computer science

Mathematical optimization

Subdivision

Urban expansion

Mathematics

Cellular automaton

Environmental resource management

Environmental science

Artificial intelligence

Geometry

Recreation

Land use

Ecosystem services

Machine learning

Ecosystem

Land Use Policy

Ecologically constrained urban expansion simulation (EC-UES) is an effective means to plan sustainable urban landscapes. Current studies typically set ecological constraints using Boolean logic while dismissing the spatially continuous and gradual features of ecological substrates. They also overlook the multiple ecosystem services (MESs) that an ecosystem provides and correlations among MESs. This study aggregated MESs (i.e., food productivity, water yield, carbon storage, biodiversity potential, erosion prevention, and outdoor recreation), based on an ordered weighted averaging (OWA) operator. An EC-UES was conducted for Wuhan, China by integrating the aggregation result as a constraint into a cellular automata (CA)-Markov chain model. By varying the β coefficient of the OWA, multiple scenarios of MESs constraint were designed and used to generate urban land patterns in different scenarios. We compared spatial patterns, quantities, and ecological effects of urban expansion in four typical scenarios. The results show that the incorporation of MESs constraint is beneficial for ecological conservation. However, the intensity of the constraint is not linearly proportional to benefit; very strong constraints from MESs may lead to the excessive loss of farmland and the irregularity and fragmentation of urban patterns. Compared to the conventional constraint strategy, a relatively soft constraint strategy was considered optimal. This study provides a reference for a win–win simulation between urban expansion and ecological conservation.

Urban expansion simulation under constraint of multiple ecosystem services (MESs) based on cellular automata (CA)-Markov model: Scenario analysis and policy implications