Constructing ecological network based on multi-objective genetic algorithms: a case study of Changsha City, China

Regional ecological security faces serious threats in a changing world. Ecological network (EN) provides decision-makers with spatial strategies for maintaining ecological security and landscape sustainability via alleviating the contradiction between ecological conservation and economic growth. Despite years of intense and fruitful studies, accurately identifying ecological source patches when facing multiple conflicting objectives still remains a challenge.

This study aimed to propose an advanced framework for recognizing ecological source patches with consideration of multiple objectives and further constructing EN, which would promote a more profound understanding of local ecological condition and provide spatial guidance for ecological conservation planning.

Taking Changsha City as the study area, we evaluated the ecological condition by considering three key ecosystem services, i.e., habitat maintenance, carbon sequestration and water yield using the InVEST model. Ecological source patches were identified using multi-objective genetic algorithms (MOGA) in view of ecosystem services, landscape connectivity and the total area of ecological source patches. Ecological corridors were extracted by applying Minimum Cumulative Resistance (MCR) model based on modified ecological resistance surface. The EN was established by combining these ecological source patches with ecological corridors.

The EN in Changsha City was comprised of 51 ecological source patches and 50 ecological corridors. The ecological source patches were primarily distributed across the eastern and western mountainous areas with the total area of 2842 km², occupying 24.05% of the study area. There was a clear lack of ecological source patches along the Xiangjiang River owing to the high level of urbanization, which deserved particular attention for ecological restoration. Overall, the identified ecological source patches provided 87.31% of ecosystem service supply and 82.49% of the whole landscape connectivity by occupying 67.09% of the dominant patch area. The depicted ecological corridors formed two clusters in the central and northeastern parts of the study area.

This study offered new insights into accurately identifying ecological source patches by coordinating various conservation objectives. With the application of MOGA, the proposed framework consolidated ecosystem services, landscape connectivity and patch area to effectively delineate core ecological patches.