摘要
在资料稀缺的背景下,遥感数据是提供湿地系统长时间序列的理想方案。然而面向"一带一路"沿线地区复杂下垫面,国家级湿地缺乏系统的长时序梳理。利用Landsat系列数据,基于随机森林分类模型,研究近30年中国典型尾闾湖湿地的时空分布模式、空间转换规律和景观连通性。结果表明:随机森林算法在艾比湖湿地分类应用中取得较高精度(Kappa系数大于0.9)。1991—2017年艾比湖湿地总面积增加425.06 km^(2),河流增加47.97 km^(2),湖泊增加233.95 km^(2),人工湿地增加48.74 km^(2),盐沼增加109.41 km^(2),沼泽减少15.01 km^(2)。艾比湖湿地年内时空变化显著,年内季节间盐沼转化率最大,湖泊年内逐渐缩小,主要转化为沼泽。此外,艾比湖湿地空间连通性理想度排序为:春季>夏季>秋季,湿地景观连通性取决于较大面积的湿地斑块,连通效率东移。相关结果可弥补稀缺资料区基础湿地资料,为"一带一路"地区生态补水长效机制提供典型示范。
Remote sensing data can provide long-term sequences for replenishing wetlands in changing environments under the context of scarce data.The accurate identification of wetland systems in arid regions is important to the comprehensive regulation and feedback mechanism of water and soil resources.However,we lack long-term data accumulation in the national wetlands in the context of the complex underlying surface of the Belt and Road region.In this study,random forest(RF)classification model was used to map the spatial distribution pattern of the typical terminal lake wetland in the arid regions over the past 30 years.Through the accurate wetland mapping,the spatial conversion rule of the wetland landscape is calculated,and the spatial connectivity of the wetland landscape is determined.We present the results of a high-precision classification study conducted in the Ebinur Lake National Wetland Reserve(Kappa coefficient is greater than 0.9).The spatial and temporal changes in the Ebinur Lake wetland during 1991-2017 was extremely significant,especially in different seasons of the year.From 1991 to 2017,the total area of wetlands,rivers,lakes,constructed wetlands,and salt marshes increased by 425.06 km^(2),47.97 km^(2),233.95 km^(2),48.74 km^(2),and 109.41 km^(2),respecctively,while the total area of marshes decreased by 15.01 km^(2).The annual change in salt marsh conversion is the largest,and the lake wetlands were shrinking gradually from spring to autumn,mainly into swampy wetlands.In addition,the ideal spatial connectivity of Ebinur wetland is listed in the order of spring>summer>autumn.The connectivity between lake wetlands,marsh wetlands and non-wetlands is better than that of other types of wetlands.The connectivity of wetland landscapes depends on the wetland patches with larger areas.To a certain extent,this research compensates the basic wetland data in the scarce data area,and provides a typical demonstration for the long-term mechanism of ecological water replenishment in the Belt and Road region.
作者
丁建丽
葛翔宇
王敬哲
DING Jian-li;GE Xiang-yu;WANG Jing-zhe(Key Laboratory of Oasis Ecology,Xinjiang University,Urumqi 830046,China;Key Laboratory of Smart City and Environment Modelling of Higher Education Institute,Xinjiang University,Urumqi 800046,China;Key Laboratory for Geo-Environmental Monitoring of Coastal Zone of the Ministry of Natural Resources,Shenzhen 518060,Guangdong,China)
出处
《自然资源学报》
CSSCI
CSCD
北大核心
2021年第8期1949-1963,共15页
Journal of Natural Resources
基金
新疆水专项(2020.B-001)
国家自然科学基金项目(41961059,41771470)。
关键词
湿地提取
随机森林
时间序列
生态效应
wetland data extraction
random forest
time series
ecological effect