山区农村居民点时空格局演变特征及蚁群仿真优化——以重庆澄江镇为例

山地农村居民点系统因受自然因素及人类活动等多重扰动而处于深刻转型中,探索山区农村居民点时空格局特征及其优化策略,对于优化山区国土空间格局、推进山区国土空间综合治理与乡村振兴具有重要意义。基于构建理论框架,集成运用景观格局指数法、栅格权重叠加评价法、改进蚁群算法（ACO）等,以重庆澄江镇为例,探索山区农村居民点时空格局演变特征及优化过程,提出优化调控策略。结果表明：（1）研究区农村居民点空间格局演变特征呈现集中、靠拢的演化发展趋势,总体呈现匀质、线状、向心、院落四种分布形态;同时高程、坡度、水域、城镇、道路、基础设施等自然—经济社会因素均对山区农村居民点时空格局演化产生重要影响。（2）改进蚁群算法优化后,研究区农村居民点斑块面积（CA）减少10.07%,居民点斑块数量（NP）和密度（PD）分别减少83.12%和81.16%,平均斑块面积（MPS）增加了4.3倍,平均最近邻体距离（MENN）增加了42.82%。优化后69.1%的农村居民点斑块都集中在适宜区和较适宜区内,且重点向交通便利、公共设施完善、生产便利和中心村庄集中。优化后的农村居民点空间格局更加紧凑有序,土地集约节约效率更高。（3）基于仿真优化结果,综合考虑村庄振兴“产业—人口—土地”互动耦合机制,提出城郊融合、产业集聚、景村融合、搬迁撤并四类农村居民点格局优化方案,助推山区乡村振兴。（4）优化结果同时表明改进的蚁群算法非常适用于复杂环境条件下的较小尺度要素空间布局优化,能够为山区国土空间规划编制提供技术支撑。

Spatio-temporal patterns of rural settlements in mountainous areas and optimization with ant colony algorithm: Evidence from Chengjiang town in Chongqing

Due to the dual disturbance of natural conditions and human activities, rural settlements systems in mountainous areas are experiencing a profound transformation. Exploring the spatial and temporal pattern characteristics of rural settlements in mountainous areas and their optimization strategies is conducive to promoting the comprehensive management of land space and rural revitalization in such regions. By employing landscape pattern index, grid weight overlay evaluation, improved ant colony algorithm, etc., this study, with Chengjiang town of Chongqing as an example, constructs a theoretical analysis framework to explore the spatio-temporal evolution characteristics of rural settlements under rapid urbanization, so as formulate the optimization and control strategy of rural settlements pattern in mountainous areas. The results show that: (1) The spatial pattern evolution of rural settlements in the study area showed a trend of concentrated and close evolution, and generally showed four distribution patterns of homogeneity, linearity, centripetal and courtyard, meanwhile, the elevation, slope, water area, town, infrastructure and other natural, economic and social factors have an significant impact on the spatial and temporal pattern evolution of rural settlements in mountainous areas. (2) The results of improved ant colony algorithm optimization shows that the patch area (CA) of rural settlements in the study area decreased by 10.07%. Meanwhile, the patch number (NP) and density (PD) decreased by 83.12% and 81.16%, respectively. The mean patch size (MPS), and the mean nearest neighbor distance (MENN) increased by 4.3 times and 42.82%, respectively. The results of optimization show that 69.1% of the rural residential patches are concentrated in suitable areas and relative suitable areas, and the concentration occurs close to convenient transportation, perfect public facilities, convenient production and central villages. The layout of rural settlements is more compact and orderly, and the land is more intensive and economical. (3) The optimization results also show that the improved ant colony algorithm is very suitable for the spatial layout optimization of town scale elements under complex environmental conditions, and can provide a scientific basis for the optimization of land use and living space pattern in the compilation of land spatial planning in mountainous areas.