城市周边及交通沿线基本农田保护风险预警及应用——以广州增城区为例  

Early warning for monitoring of basic farmland surrounding cities and traffic lines——A case study about Zengcheng, Guangzhou

作　　者：张婷[1] 马世发[1] 黄韬[1] 

机构地区：[1]广东省国土资源技术中心,广东广州510075

出　　处：《国土与自然资源研究》2017年第6期55-58,共4页Territory & Natural Resources Study

基　　金：广东省国土资源科研项目(GDGTKJ2017001)资助

摘　　要：本文的研究目的是建立空间预警模型,分析基本农田流失风险,提高土地利用"卫片执法"遥感检测的针对性。研究方法为基于空间区位论,利用人工神经网络建立数据挖掘模型,分析土地城镇化概率,预警耕地流失风险。研究结果显示城市周边及交通沿线优质耕地分布较多,质量也较好,但流失风险也较大,耕地保护矛盾十分突出。本文的研究结论为基于数据挖掘的空间预警模型能有效预测城市周边及交通沿线耕地流失风险,可为土地利用"卫片执法"提供监测指引。The purpose of this paper was to propose an early wanting model based on spatial data mining method to analysis the risk for protection of basic farmland, which will help for monitoring illegal land use with t^mote sensing. Methods include the urban expansion location theory and artificial neural network. Results demonstrate that high quality arable land usually located around cities and traffic lines, but these areas are also with a high probability for urban growth. The basic farmland which located around metropolitan areas are easier to be lost. The paper concluded that artificial neural network could be used to discover the land use change probability for early warning the occupation of arable land, which could help discover the hot areas for illegal land use monitoring.

关 键 词：基本农田 风险预警 人工神经网络 卫片执法 

分 类 号：F321.1[经济管理—产业经济]