张家口坝上地区湿地退化特征及风险评价

Wetland Degradation Characteristics and Risk Assessment in Bashang Region of Zhangjiakou

区域内湿地生态系统变化可反映出区域生态环境质量状况,精准、客观地做好湿地退化风险评价是湿地管理及可持续发展的关键。该研究以河北省张家口坝上4县为研究区,利用1990-2020年多期土地利用和景观指数数据,分析湿地退化特征。通过选取8个指标构建湿地退化风险评价体系,并基于随机森林(RF)和支持向量机(SVM)2种机器学习算法进行湿地退化风险评价。结果表明:1990-2020年张家口坝上地区湿地面积呈明显的退化状态,总面积减少290.79 km^(2),湿地盐碱化较为严重;模型评价精度结果显示,RF和SVM两种算法的AUC值分别为0.549和0.613,表明SVM模型更适用于该区域的湿地退化风险评价;基于SVM模型湿地退化风险评价结果分析可知,重度退化风险区占比为23.94%,主要集中于研究区西部边缘及中部安固里淖流域;轻度退化风险区占比为22.54%,呈散落分布,多为面积小且破碎度高的季节性湖淖,湖泊间存在河道、沟渠等连通性较好的区域退化风险相对较低;湿地退化影响因素权重分析表明,夜间灯光数据的权重占比最大,达到17.74%,表明人类活动是主要影响因素。研究结果有助于当地相关部门识别张家口坝上湿地退化风险区域,为维护区域湿地可持续发展提供科学决策指导。

The change of wetland ecosystem in the region can reflect the quality of regional ecological environment.Accurate and objective assessment of wetland degradation risk is the key to wetland management and sustainable development.In this study,four counties in Zhangjiakou,Hebei Province are selected as the research area,and the characteristics of wetland degradation are analysed by using the multi-period land use and landscape index data from 1990 to 2020.Eight indicators are selected to construct a wetland degradation risk evaluation index system.Based on two machine learning algorithms,random forest(RF)and support vector machine(SVM),the risk of wetland degradation in the Bashang Region is evaluated and partitioned.The results show that the wetland area in the Bashang Region is obviously degraded from 1990 to 2020,and the total area is reduced by 290.79 km^(2),and the salinization of the wetland is more serious.The results of model evaluation accuracy show that the AUC values of RF and SVM algorithms are 0.549 and 0.613,respectively,indicating that the SVM model is more suitable for wetland degradation risk assessment in this area.The analysis of wetland degradation risk assessment results based on SVM model shows that the proportion of severe degradation risk area is 23.94%,which is mainly concentrated in the western edge of the study area and the central Angulinao watershed.The proportion of mild degradation risk area is 22.54%,which is scattered,mostly seasonal lakes with small area and high fragmentation.The degradation risk of areas with good connectivity such as rivers and ditches between lakes is relatively low.The weight analysis of the influencing factors of wetland degradation shows that the weight of nighttime light accounted for the largest proportion,reaching 17.74%,indicating that human activities are the main influencing factors.The results of the study are helpful for local relevant departments to identify the risk area of wetland degradation in Zhangjiakou Bashang,and provide scientific decision-making guidance for maintaining the sustainable development of regional wetlands.