矿区废弃地复垦为农用地潜力评价方法的比较

Comparison for evaluate methods of potentiality for agricultural land reclamation from waste land in mining area

矿区废弃地复垦潜力是编制土地整治规划的重要依据,复垦潜力的大小主要依靠评价来实现,而评价方法的选取对评价结果具有直接影响。该文以山西省朔州市平鲁区为研究区域,采用指数和法、模糊综合评价法和人工神经网络模型对矿区废弃地进行复垦潜力评价,探讨这3种方法所得结果的一致性与差异性及其产生的原因;并在结果分析的基础上,运用层次分析法对各方法所得结果进行综合评价。结果表明:在同一指标体系和同一权重下,由于各方法在计算过程中对于评价单元的处理侧重点不同,致使3种方法所得各级别中相同图斑数、同一潜力级别下的面积及其空间位置分布具有一定差异性;理论上讲,人工神经网络方法较其他2种方法所得结果更具客观性;平鲁区矿区废弃地分布与综合评价结果对比分析显示,历史遗留矿区废弃地和因矿业整合产生的废弃地复垦潜力集中在Ⅰ、Ⅱ级,潜力级别总体较高,矿区正在作业区复垦潜力多为Ⅲ、Ⅳ级,级别相对较低;研究表明,综合评价的结果是更科学、更客观的理论值,对于矿区废弃地系统总体而言,其复垦潜力的实际值还需考虑矿区废弃地复垦实践中复垦年限、经费、人员组织、产权关系等影响因素进行修正。

Waste land reclamation potential in mine areas is an important basis for a Land Consolidation and Rehabilitation Plan. The size of potential value calls for the evaluation, so the selection of evaluation methods has a direct impact on the result of the evaluation. In general, the steps of reclamation evaluation are choosing the evaluation objects, screening evaluation indices, grading the evaluation indices, weighted the indices, and evaluating the potential. In the current study, most research uses just one method to evaluate the reclamation potential, or improve the method during the process of reclamation evaluation, like improving the method for obtaining weight, and taking this as the final evaluation result may introduce a kind of randomness. Based on the Shuozhou city in Pinglu area in Shanxi Province, this article will use the Index method, the Fuzzy comprehensive evaluation method, and the Artificial neural network model to evaluate the reclamation potential, and discuss the similarities, differences, and the lead reasons among the results provided by these three methods. Then, the Analytic Hierarchy Process is used to combine the results from each method, and the comprehensive evaluation result is obtained based on that. The result indicates that under the same index system and weight, because of the different evaluation methods and the different dimensionless methods used during these processes, each potential level is different among the numbers of figure spot, area, and spatial location under the same potential level. Under the theory that both the Index method and Fuzzy comprehensive evaluation method are affected by the indicator system and weight, the difference is that before using the Index method, the experimentalist should standardize parameter values, which will be solved during the Fuzzy comprehensive evaluation process. The＆nbsp;similarity is that both methods will provide weighted sums of the indicator system to obtain the final value, which will be a numerical value whose size represents the potential level by the Index method, while the result by the Fuzzy comprehensive evaluation method is a row vector. This paper depends on the weighted average principle to get the potential level. As for the artificial neural network model, after normalizing all parameters, because the value for each node in the hidden layer and output layer, instead of the simple weighted sum, is calculated, the whole process is dependent on its self-learning capability. So, whereas the artificial neural network model is little influenced by subjectivity, the result by the artificial neural network model is more objective. The comparison between the waste land distribution and the comprehensive evaluation result indicates that where the abandoned mining areas and mining consolidation areas are concentrated in the first or second level, the potential is overall high. Where the mined areas stay in the third or fourth level, the potential is relatively low. Research shows that the comprehensive evaluation result is more scientific and has a more objective theoretical value; and for the waste land system, the actual value calculated still needs to include values for the reclamation years, funds, personnel organization, property right relations, and other factors.