基于PSR框架的内蒙古自治区土壤保持服务分区

土壤侵蚀是中国北方重要的生态问题,内蒙古自治区位于干旱半干旱地区,是中国北方典型的土壤侵蚀区。针对以往生态功能分区分析框架复杂多样、权重设定人为主观等问题,以内蒙古自治区为例,基于压力-状态-响应系统分析框架,从气候背景-土壤保持-植被条件三方面构建指标体系,运用自组织特征映射(SOFM)神经网络和GIS空间分析技术,以1280个小流域为基本单元进行土壤保持服务分区,按照分区单元的聚集度对多种方案进行优选,并依据区域共轭性原则最终将研究区划分为3个土壤保持服务区和11个土壤保持服务亚区。

Soil conservation service zoning in the Inner Mongolia Autonomous Region based on PSR framework

Soil erosion has been a critical ecological issue in northern China. Inner Mongolia Autonomous Region is located in the arid and semi-arid area, which is a typical soil erosion area in northern China. Based on the analytical framework of pressure-state-response (PSR), this study takes the Inner Mongolia Autonomous Region as the study area, and constructs the index system from three aspects: climatic background, soil conservation, and vegetation condition. It aims at addressing the remaining issue of complexity and subjectiveness in weigh specification towards ecological function zoning in the previous studies. In the study, self-organizing feature map (SOFM) neural network and GIS spatial analysis technique were applied to carry out the soil conservation service zoning with the 1280 small watersheds as the basic unit and the schemes were optimized according to the aggregation degree of the zoning units. The Inner Mongolia Autonomous Region was divided into three ecological functional zones and eleven sub-zones based on the principle of regional conjugation. The three ecological functional zones include Soil Conservation Services Protection Area, Soil Conservation Services Controlling Area, and Soil Conservation Services Rehabilitation Area. According to the characteristics of zoning, the results clearly indicate the zonal distribution of climatic background, soil conservation service and vegetation condition in Inner Mongolia, and quantify the characteristics and spatial differences of the ecological functions at watershed scale more comprehensively. The result of the optimal soil conservation service zoning was compared to the annual average temperature contours and annual precipitation contours in Inner Mongolia. It shows that the ecological function zoning results obtained by SOFM neural network model were in good agreement with the spatial patterns of basic ecological elements. The zoning scheme indicates a significant gradient and is consistent with changes in temperature and precipitation.