The development of cultivated land reserve is an important issue related to food security, and the evalu-ation of development potential is the foundation of cultivated land development. Currently, studies on cultivated land reserve in China largely concern with the land suitability evaluation which takes unused land as cultivated land reserve. However, there are few studies discussing the developmental potentiality of cultivated land reserve, which often paid much more attention to natural, social or economic conditions but focused less on ecological condition. Besides, the index weight determined by the traditional studies is always subjective, and the traditional evaluative methods are often incapable of expressing the fuzziness and randomness of the evaluative object precisely. The pur-pose of this study is to establish a feasible evaluating indication system and method for the developmental potentiali-ty of cultivated land reserve based on the statistical data in Jingtai. Methods of entropy weight and the Normal Cloud model are employed. Normal Cloud is a new model using language value to express the uncertain conversion be-tween qualitative description and quantitative description. It can overcome the ambiguity and uncertainty existing in the conversion from qualitative description to quantitative description when using traditional methods to make evalu-ations. Entropy can be used to express the confusion of a system. The weight of evaluation index is greater when the difference between evaluation index is bigger, and it is the basic idea that uses entropy to express the relative im-portance of evaluation index. Methods of entropy weight and Normal Cloud model can effectively reduce the errors caused by the traditional method. The results indicated that the cultivated land reserve developmental potentiality could be evaluated based on three aspects-ecological environment, natural conditions and social economic condi-tions. The evaluation score of each town in Jingtai ranged from 2. 2003 to 3. 9880, and the potential degree of each town could be categorized as high, middle and low potentiality according to the ecological environment, natural con-ditions and social conditions. What′s more, the towns as Shangshawo, Sitan and Luyang with high potentiality could be given priority to the development of cultivated land reserve. On the contrary, the Wufo town as low poten-tiality area could not develop as a cultivated land in the future. It concluded that the Normal Cloud model and en-tropy weight could reduce the error in the developmental potentiality evaluation of cultivated land reserve, and im-prove the accuracy of the evaluation effectively. This research could provide references for the developmental poten-tiality evaluation of regional cultivated land reserve.%耕地后备资源开发是关乎粮食安全的重要议题，耕地后备资源开发潜力评价是耕地后备资源开发的基础性工作。针对传统土地评价综合指数法、模糊综合法无法同时兼顾评价指标模糊性与随机性以及特尔菲法、层次分析法确定权重主观性过强的问题，研究以景泰县为例，将正态云模型与熵权法引入区域耕地后备资源开发潜力评价，建立综合考虑生态、自然、社会经济3方面的评价指标体系，对区域耕地后备资源开发潜力进行定量测度，并提出了针对性的政策建议。研究结果表明：景泰县各乡镇耕地后备资源开发潜力从2．2003到3．9880不等，潜力等别包括较低、中等和较高，对潜力等别为较高的上沙沃镇、寺滩乡、芦阳镇可作为耕地后备资源开发优先区域，对潜力等别为较低的五佛乡建议暂不考虑进行耕地后备资源开发。正态云模型与熵权法有效降低了偶然误差，提高了耕地后备资源开发潜力评价的准确性，该研究可为区域耕地后备资源开发潜力评价提供一定借鉴与参考。
展开▼
