Using the four periods of land use data and socio-economic statistics of Lanzhou City in 1995, 2005, 2015 and 2018, we examined the evolution characteristics and driving forces of urban land use in Lanzhou, with expansion intensity index, Logistic regression, and principal component analysis method. Based on the multi-agent model, the expansion of urban land use in Lanzhou under multiple scenarios in 2025 was simulated using Netlogo software. The results showed that the intensity and scale of urban land expansion in Lanzhou City from 1995 to 2018 showed an increasing trend, with a total of 307 km2 of cultivated land and water area. The expansion of urban land in Lanzhou City was significantly related to regional GDP, social fixed asset investment, population density, population, policy, and terrain. The multi-agent model on the Netlogo software could simulate the expansion trend of urban land in Lanzhou, which provided a reliable simulation method for the expansion of valley-type cities. From 2020 to 2025, urban land use under the comprehensive development scenario, natural expansion development scenario, ecological and farmland protection development scenario all showed an increasing trend. The scale of expansion would range between 42.6-134.3 km2. The expansion mode would mainly be infilled expansion. Our results could provide decision support for the land space planning and ecological environment layout optimization for Lanzhou City.
利用1995、2005、2015、2018年4期兰州市土地利用数据和社会经济统计数据,运用扩展强度指数、Logistic回归、主成分分析等方法揭示兰州市城镇用地演变特征及驱动力,基于多智能体模型用Netlogo软件实现了2025年多情景下兰州市城镇用地扩展模拟。结果表明: 1995—2018年,兰州市城镇用地扩展强度和规模均呈增长趋势,侵占耕地和水域面积共307 km2;兰州市城镇用地扩展与地区生产总值、全社会固定资产投资、人口密度、城镇人口、政策和地形显著相关;从模拟结果来看,多智能体模型在Netlogo软件上能够较好地模拟兰州市城镇用地扩展趋势,在一定程度上为河谷型城市扩展模拟提供了可靠的仿真手段;2020—2025年,综合发展情景、自然扩张发展情景、生态及农田保护发展情景下城镇用地模拟结果均呈增长趋势,扩展规模预计为42.6～134.3 km2,扩展模式将以填充式和飞地式扩展为主。本研究结果可为兰州市的国土空间规划和生态环境布局优化提供决策支持。.
Keywords: Lanzhou City; multi-agent model; scenario simulation; temporal and spatial characteristics; urban land expansion.