Regional land-use allocation with a spatially explicit genetic algorithm

Abstract

Land-use allocation is an important way to promote the intensive and economic use of land resources and achieve the goal of sustainable development. It is a complex spatial optimization problem, and heuristic algorithms have been one of the most effective ways to solve it in past studies. However, heuristic algorithms lack the guidance of planning knowledge, which makes land-use patterns usually unreasonable in practice. This research proposes a spatially explicit genetic algorithm (SEGA) that integrates land-use planning knowledge with the genetic algorithm (GA). The SEGA transforms the spatially implicit computation mode of the GA into a spatially explicit optimization style, which helps to promote the effectiveness of regional land-use allocation. Gaoqiao Town, China, was selected as the study area to test the SEGA. Results show that: (1) land-use conversions are reasonable in accordance with planning knowledge, and they improve overall land-use suitability and spatial compactness; (2) compared with the GA, the SEGA is superior in achieving global objectives and simulating local dynamics. We demonstrated that planning knowledge is essential to heuristic algorithms for land-use allocation.