Riparian dunes in deserts exhibit unique geographic features due to aeolian-fluvial interactions.In this study,we collected 510 surface sediment samples from eight drainage basins and conducted a systematic analysis t...Riparian dunes in deserts exhibit unique geographic features due to aeolian-fluvial interactions.In this study,we collected 510 surface sediment samples from eight drainage basins and conducted a systematic analysis to examine the grain size characteristics of major riparian dunes in the typical cold and arid deserts of China.The results indicate that major riparian dunes of deserts in study area can be classified into three types based on their grain size characteristics.The Bartlett test of sphericity and the Kaiser-Meyer-Olkin(KMO)test were also performed,and their significance values were found to be 0.000 and 0.584,respectively.The results of the principal component analysis revealed that the cumulative contribution rate of the total variance reached 85.9%for the two principal components with characteristic roots greater than 1.0.The primary principal component included medium sand,whereas the secondary principal component included fine sand.We conducted a cluster analysis and classified the samples into three major types.Type I rivers include the Keriya River,Langqu River,Tora River and Heihe River,which are characterized by by fine particle size,and well-sorted.Type II includes Mu Bulag River,Kuye River,and the Xar Moron River,Compared with type I,it has a relatively coarser mean grain size and relatively poor sorting for this type.Type III includes the Maquan River,which is characterized mainly by fine sand and medium sand,accounting for more than 90%,and the sorting coefficient(0.52)suggests relatively well sorting in this pattern.Moreover,principal component analysis was applied to determine the particle sizes of samples from different watersheds.Moreover,these sediments exhibit both hydromorphic and aeolian features.At the drainage basin scale,the mode and intensity of aeolian-fluvial interactions depend on climatic conditions.In arid and semi-arid climate regions,wind is the dominant force,and the grain size exhibits significant aeolian features.Conversely,in the semi-humid region,flowing water is the dominant force,and riparian dunes in this region are formed by aeolian-fluvial interaction.The angle between the wind direction and flow direction in different reaches influences both the supply of sediment sources and the development of riparian dunes.This study will provide a new perspective for evaluating aeolian-fluvial interactions on riparian dunes in the deserts of China’s cold and arid regions.展开更多
当前,土地沙漠化已成为全球十大生态环境问题之一,严重威胁着众多国家和地区。为有效监测沙漠化动态,本研究基于Landsat遥感影像,通过辐射定标、大气校正等预处理算法获取沙漠区域影像。本研究首先采用人工目视解译方法对典型沙漠边界...当前,土地沙漠化已成为全球十大生态环境问题之一,严重威胁着众多国家和地区。为有效监测沙漠化动态,本研究基于Landsat遥感影像,通过辐射定标、大气校正等预处理算法获取沙漠区域影像。本研究首先采用人工目视解译方法对典型沙漠边界区域进行标注,构建训练样本;随后基于结构状态空间对偶的双分支编码沙漠分割模型提取大范围沙漠边界信息;最终形成2000–2020年每5年一次的中国沙漠边界数据集。本数据集涉及中国八大沙漠,即腾格里沙漠、塔克拉玛干沙漠、巴丹吉林沙漠、库布齐沙漠、乌兰布和沙漠、库姆塔格沙漠、古尔班通古特沙漠、柴达木盆地沙漠。在质量控制方面,通过引入Overall Accuracy(OA)、Mean Intersection over Union(MioU)、Mean Pixel Accuracy(MPA)、Mean Recall(MRecall)和Mean F1 Score(MF1)等指标进行精度评估,最终OA达98.78%,MIou达97.58%,MPA达98.77%,MRecall达98.77%,MF1达98.77%。本数据集在沙漠化监测、环境保护、资源管理等领域具有重要的应用价值,为沙漠化研究提供了有力的数据支撑。展开更多
基金Under the auspices of the General Project of Science and Technology Department of Shaanxi Province(No.2023-JCYB-264)General Program of National Natural Science Foundation of China(No.41801004,42371008,42471012)。
文摘Riparian dunes in deserts exhibit unique geographic features due to aeolian-fluvial interactions.In this study,we collected 510 surface sediment samples from eight drainage basins and conducted a systematic analysis to examine the grain size characteristics of major riparian dunes in the typical cold and arid deserts of China.The results indicate that major riparian dunes of deserts in study area can be classified into three types based on their grain size characteristics.The Bartlett test of sphericity and the Kaiser-Meyer-Olkin(KMO)test were also performed,and their significance values were found to be 0.000 and 0.584,respectively.The results of the principal component analysis revealed that the cumulative contribution rate of the total variance reached 85.9%for the two principal components with characteristic roots greater than 1.0.The primary principal component included medium sand,whereas the secondary principal component included fine sand.We conducted a cluster analysis and classified the samples into three major types.Type I rivers include the Keriya River,Langqu River,Tora River and Heihe River,which are characterized by by fine particle size,and well-sorted.Type II includes Mu Bulag River,Kuye River,and the Xar Moron River,Compared with type I,it has a relatively coarser mean grain size and relatively poor sorting for this type.Type III includes the Maquan River,which is characterized mainly by fine sand and medium sand,accounting for more than 90%,and the sorting coefficient(0.52)suggests relatively well sorting in this pattern.Moreover,principal component analysis was applied to determine the particle sizes of samples from different watersheds.Moreover,these sediments exhibit both hydromorphic and aeolian features.At the drainage basin scale,the mode and intensity of aeolian-fluvial interactions depend on climatic conditions.In arid and semi-arid climate regions,wind is the dominant force,and the grain size exhibits significant aeolian features.Conversely,in the semi-humid region,flowing water is the dominant force,and riparian dunes in this region are formed by aeolian-fluvial interaction.The angle between the wind direction and flow direction in different reaches influences both the supply of sediment sources and the development of riparian dunes.This study will provide a new perspective for evaluating aeolian-fluvial interactions on riparian dunes in the deserts of China’s cold and arid regions.
文摘当前,土地沙漠化已成为全球十大生态环境问题之一,严重威胁着众多国家和地区。为有效监测沙漠化动态,本研究基于Landsat遥感影像,通过辐射定标、大气校正等预处理算法获取沙漠区域影像。本研究首先采用人工目视解译方法对典型沙漠边界区域进行标注,构建训练样本;随后基于结构状态空间对偶的双分支编码沙漠分割模型提取大范围沙漠边界信息;最终形成2000–2020年每5年一次的中国沙漠边界数据集。本数据集涉及中国八大沙漠,即腾格里沙漠、塔克拉玛干沙漠、巴丹吉林沙漠、库布齐沙漠、乌兰布和沙漠、库姆塔格沙漠、古尔班通古特沙漠、柴达木盆地沙漠。在质量控制方面,通过引入Overall Accuracy(OA)、Mean Intersection over Union(MioU)、Mean Pixel Accuracy(MPA)、Mean Recall(MRecall)和Mean F1 Score(MF1)等指标进行精度评估,最终OA达98.78%,MIou达97.58%,MPA达98.77%,MRecall达98.77%,MF1达98.77%。本数据集在沙漠化监测、环境保护、资源管理等领域具有重要的应用价值,为沙漠化研究提供了有力的数据支撑。
