基本农田规划决策支持:模型构建与系统研发  

The Decision Support for Basic Farmlands Planning:Model Application and System Development

作　　者：汪西林[1] 胡娈运[2] 

机构地区：[1]北京林业大学水土保持学院重点实验室,北京100083 [2]湖南省第三测绘院,湖南长沙410007

出　　处：《生态环境学报》2017年第10期1689-1695,共7页Ecology and Environmental Sciences

基　　金：国家重点研发项目(2017YFC0505500);林业科技创新平台运行补助项目(2016-LYPT-SY-013);国家科技支撑项目(2006BAB15B05)

摘　　要：基本农田的保护是确保粮食安全的"底线"。随着中国1.03亿公顷永久基本农田的划定工作进入尾声,基本农田划定中的高效操作经验有待系统梳理和总结。文章依据沈阳苏家屯区土地利用调查分等定级成果,构建基本农田识别划定模型,辅以决策支持系统的开发,探索科学识别、快速划定基本农田,减轻人为干扰,提高基本农田的识别划定效率的方法。该研究提取9项评价指标——自然质量分、自然质量等指数、土地利用等指数、农用地等别指数、耕作距离、中心城市影响度、道路通达度、对外交通便利度、农贸市场影响度,通过主成分分析"降维"为3个主成分因子,特征值分别为5.084、1.342、1.011,累积贡献率达到82.632%。其中,土地利用等指数、农用地等别指数、自然质量等指数、自然质量分、农贸市场影响度在第一主成分上有较高载荷,反映自然条件和社会投入-产出值对耕地综合质量的影响;对外交通便利度、中心城市影响度在第二主成分上有较高的载荷,反映区位条件对耕地综合质量的影响;耕作距离在第三成分上有较高的载荷,反映经济条件对耕地综合质量的影响。利用主成分分析计算耕地综合质量指数,再预测基本农田规划面积约束,建立0-1整数规划模型,识别筛选综合质量指数较高耕地,通过决策系统快速划定入选基本农田的耕地。该模型全面反映了自然因素、经济因素、区位因素对耕地综合质量的影响,其决策系统的模型构建与系统应用为区县尺度的基本农田划定提供了科学识别、基本农田面积确定和基本农田的划定三大功能,使基本农田规划决策工作得以有效提升。To ensure its national food security, the Chinese government plans to delineate and protect 103 million hectares permanent basic cropland. However, the area needs to be selected and analyzed based on a combined land use survey and a decision support system. Using Sujiatun District, Shenyang City as the case study area, nine indices, including natural quality score, natural quality grading index, land use grading index, agricultural land grading index, farming distance, influence degree of central city, road accessibility, external traffic convenience degree, and influence degree of market, were applied for the analysis. Through a Principle Component Analysis （PCA）, the listed indices were reduced to three major component factors which explained 82.63% of the accumulated variations in the metrics （eigenvalues were 5.084, 1.322, and 1.011 for the three components, respectively）. The first principal components showed that the quality of cultivated land was strongly associated with land use grading index, agricultural land grading index, natural quality grading index, natural quality score and market influence degree, which indicated the influence of natural and socio-economic conditions. The second component was associated to the external traffic convenience degree, and the influence degree of central city, which reflected the importance of farmland location. While farming distance indicating the impacts from economic conditions, was reflected in the third component. Based on these results, the quality of the cultivated land could be predicted using the impacts of natural, economic and location factors. In addition, the “0-1” integer model was established, and cultivated land with a relatively higher quality score would be assigned as protected basic farmland. Supported by scientificapproach, this proposed method and model system could support county-level basic farmland designation to effectively promote farmland planning and decision-making activities.

关 键 词：:基本农田 分等定级 主成分分析 0-1整数规划模型 决策支持系统 

分 类 号：X144[环境科学与工程—环境科学]