段金龙¹, 谭利华^,2, 杜世松², 陈长委², 伍永秋²1. 北京师范大学 政府管理学院,北京 100875
2. 北京师范大学 地理科学学部,北京 100875

Quantitative analysis of the source of aeolian sand on the east bank of the Cuona Lake in Tibet based on the composite fingerprinting identification technique

DUAN Jinlong¹, TAN Lihua^,2, DU Shisong², CHEN Changwei², WU Yongqiu²1. School of Government, Beijing Normal University, Beijing 100875, China
2. Faculty of Geographical Science , Beijing Normal University, Beijing 100875, China

通讯作者: 谭利华,男,湖南双峰人,讲师,研究方向为地貌与第四纪地质。E-mail: lihuatan@yeah.net

收稿日期:2019-06-14修回日期:2019-11-15网络出版日期:2019-12-25

基金资助:

第二次青藏高原综合科学考察研究专题.SQ2019QZKK1606 国家重点基础研究发展计划项目课题.2013CB956001

Received:2019-06-14Revised:2019-11-15Online:2019-12-25
作者简介 About authors
段金龙,男,河南息县人,博士生,研究方向为土地利用规划与管理E-mail:m15194450917@163.com。







摘要
本文通过对西藏错那湖地区基岩、第四纪沉积和风沙堆积的填图和采样,测量其化学元素成分,应用复合指纹识别技术定量化查明风沙沉积的物源组合及其贡献百分比,从而为错那湖地区更科学地进行沙漠化防治提供数据支撑和实践指导。分析结果表明：①使用复合指纹识别技术定量查明了错那湖东侧巴索曲、龙庆南沟和桑曲3个风沙分布区的物质来源,风沙主要来源于湖的东岸和北岸;②错那湖地区风沙的物质来源差异很大,以东岸贡献为主,对3个风沙区的贡献率分别为82.3%、60.4%和10.1%,呈现出自南向北逐渐减小的趋势;③结合青藏铁路分布可知,东岸物源对铁路的威胁最大,北岸物源仅对铁路以东到山坡的狭长区域有一些影响。夏季高温多雨期,巴索曲、龙庆南沟和桑曲3条河流将中上游的冻融、风化形成的细颗粒物质通过流水作用源源不断地输送到湖滨区域,来年冬春季干燥大风期时,这些暴露的细颗粒物质被大风输送到东侧形成风沙区,基于这一作用机理的认识,认为本地区风沙对铁路的威胁是持续的、长期存在的。因此,在铁路西侧防沙措施的基础上,还应加强铁路东侧的风沙的治理,并在铁路和山坡之间的狭长地带也设置一些防沙设施。
关键词： 风沙物源;复合指纹识别技术;多元混合模型;贡献率;青藏铁路;错那湖

Abstract
In this study, based on the mapping and sampling of bedrocks, quaternary sediments, and aeolian sand deposits in the Cuona Lake area as well as the analysis of their chemical composition, we used composite fingerprint identification technique to quantitatively investigate the source composition and contribution rate of each source to sand deposits, which can provide data support and practical guidance for the desertification control in the Cuona Lake area. The analysis results show that: (1) Using the composite fingerprinting identification technique, we have quantitatively identified the material sources in three aeolian sand distribution areas of Basoqu, Longqing South Gully and Sangqu on the east side of Cuona Lake, which indicates that the aeolian sand mainly drives from the east and north shore of the lake. (2) The material sources of aeolian sand vary greatly in the Cuona Lake area, most of which are mainly composed of the east bank source. In the aeolian sand distribution areas of Basuoqu, Longqing South Gully and Sangqu, the average contribution rates of east bank source is 82.3%, 60.4% and 10.1% , respectively, showing a decreasing trend from north to south. (3) Considering the route of the Qinghai-Tibet Railway, we conclude that the east bank source poses the greatest threat to the railway, while the north bank source has certain influence only in the narrow zone between the railway and the hillside. During the high temperature and rainy period in summer, the three rivers of Basoqu, Longqing South Gully and Sangqu continuously transport the fine particles, which were formed by freeze-thaw action and weathering in the middle-upper reaches of these rivers, to the lakeside area through flowing water. During the dry and windy period in winter and spring next year, these exposed fine particles are transported to the east side by strong wind to form the aeolian sand area. Based on the understanding of this mechanism, we believe that the threat of aeolian sand to the railway in this area is persistent and permanent. Therefore, on the basis of sand control measures on the west side of the railway, the control of aeolian sand on the east side of the railway should be strengthened, and some sand control facilities should also be set up in the narrow zone between the railway and the hillside.
Keywords：source of aeolian sand;composite fingerprinting identification technique;multivariate mixed model;contribution rate;Qinghai-Tibet Railway;Cuona Lake


PDF (10432KB)元数据多维度评价相关文章导出EndNote|Ris|Bibtex收藏本文
本文引用格式
段金龙, 谭利华, 杜世松, 陈长委, 伍永秋. 基于复合指纹识别技术的西藏错那湖东岸风沙来源的定量化分析. 资源科学[J], 2019, 41(12): 2316-2326 doi:10.18402/resci.2019.12.15
DUAN Jinlong. Quantitative analysis of the source of aeolian sand on the east bank of the Cuona Lake in Tibet based on the composite fingerprinting identification technique. RESOURCES SCIENCE[J], 2019, 41(12): 2316-2326 doi:10.18402/resci.2019.12.15

1 引言

风沙灾害是威胁沙区道路建设和运行的主要因素^[1,2],随着中国西部地区交通干线建设的加快,沿线风沙对铁路的危害问题显得愈发突出^[3,4,5]。青藏铁路是一条穿越多年冻土及沙漠的高原铁路,沿线风沙的形成过程与低海拔地区差异很大,风沙活动对铁路造成的危害也比低海拔地区更为突出^[3,6]。西藏错那湖地区位于藏北高原北端,生态环境极为脆弱,受季节性极端寒冷干旱大风气候、特殊的地质地貌组合以及复杂水文条件的交互作用,具备很强的风沙发育条件^[7,8,9],是青藏高原上极具代表性的小空间尺度的风沙分布区。青藏铁路从错那湖风沙分布区穿过,该地区的强烈风沙活动将严重威胁铁路的运行安全^[10,11]。

沙源是风沙产生的物质基础^[12],对风沙来源的研究是合理采取风沙防治措施的重要前提^[2,12]。目前,国内外相关研究一般采用地球化学元素法来分析风成沉积物的物质来源^[13,14,15],但普遍存在以下两个方面的问题：一是仅通过对潜在物源和风成沉积的示踪元素（常量元素、微量元素、REE和同位素等）指标特征来确定风沙的物质来源^[16,17,18],而很少进一步定量查明各类物源的贡献占比;二是仅在大空间尺度上确定风沙的物质来源,而极少在小空间尺度上识别风沙的物质来源。复合指纹识别技术则为定量化分析目标沉积物的各物质来源组合提供了非常有效的技术手段^[19],复合指纹识别技术结合严格的统计验证步骤和多元混合模型,仅需使用不同物源类型和目标沉积物的地球化学元素参数,即可计算出不同物源类型的相对贡献率^[20,21]。目前,复合指纹识别技术已经广泛使用在小流域范围内水成沉积物来源的研究上^[22,23],但在小空间尺度的风成沉积物来源研究上运用还非常少。

本文尝试将复合指纹识别技术应用于小空间尺度风沙来源的研究上,选取错那湖地区作为研究区,利用遥感影像和野外调查确定风沙及潜在物源的空间分布,采集风沙和潜在物源样品并测试其化学元素指标,构建最佳复合指纹因子,从而定量查明风沙沉积的物源组合及其贡献百分比,为错那湖地区更科学地进行沙漠化防治提供重要的数据支撑和实践指导,同时为更多小空间尺度风沙来源的定量化研究提供一定的方法和理论借鉴。

2 研究区概况与研究方法

2.1 研究区概况

研究区位于唐古拉山南麓、安多县城附近的错那湖地区（91°24′E—91°45′E ,32°00′N—32°18′N）,地处藏北高原,平均海拔在4500 m以上,附近有青藏铁路、青藏公路等重大交通工程设施（图1）。本地区属于高原亚干旱气候^[8],据安多县气象站2006—2015年观测数据,年平均气温-1.5℃,最热月平均气温8.7℃,最冷月平均气温-12.3℃,气温日较差13.3℃,地表物理风化强烈;年平均降水量460.4 mm,年最大降水量543.9 mm,一年中的降水绝大部分集中在夏季,夏季（6—8月）平均降水量为456.5 mm,而春季、秋季和冬季加起来的平均降水量仅有3.9 mm,可见该地区除夏季之外的3个季节都极端干旱。本地区主导风向为W,风向为W的年平均日数为97天,次要风向为WSW和SW,对应的年平均日数分别为54.2天和33.4天;平均大风日数（≥8级）为93.5天,大风季节为每年10月至次年3月,定时、瞬时最大风速分别为20.9 m/s、32.2 m/s,风向都为W,可见风力非常强劲（图2）。此外,通过野外测量的5个流动沙丘倾向数据可反映出局部的风力合成状况（表1）,巴索曲和龙庆南沟流动沙丘的倾向都介于65°~100°之间,这与气象观测得到的主导风向并不完全一致,存在小范围的偏差,说明局部加入了地形风的影响。

图1

新窗口打开|下载原图ZIP|生成PPT
图1研究区地理位置及采样点分布

Figure 1Geographical location of the study area and sampling site distribution

图2

新窗口打开|下载原图ZIP|生成PPT
图2风向玫瑰图

Figure 2Wind rose map



Table 1
表1
表1错那湖地区流动沙丘倾向统计
Table 1Decline statistics of mobile sand dunes in the Cuona Lake area

沙丘样号	21e	22e	23e	27e	44e
倾向	70°	100°	65°	75°	75°

注：e代表风沙样。

新窗口打开|下载CSV

错那湖为高海拔构造湖,四周为构造抬升山体,有巴索曲、桑曲等河流呈向心状汇入湖泊,整体上属于一个封闭流域,受流域以外的影响较小。从图3看出,风沙主要分布在错那湖东岸,呈带状沿巴索曲、龙庆南沟和桑曲等河流谷地延伸分布,部分风沙甚至被搬运到高海拔山坡上,根据水系格局和地形状况等特征,可将东岸风沙由南向北进一步分为巴索曲、龙庆南沟和桑曲3个风沙区。通过野外调查和总结前人研究成果发现,错那湖东岸风沙总体上以细砂及以上粒级为主要组分^[8],再加上该地区风沙活动主要受偏西风控制,因此绝大部分沙源物质只能以跃移和蠕移方式沿地表自西向东进行近距离搬运,搬运距离很短,一般不超过十几到二十几千米^[7,24]。而错那湖湖面宽达十几到二十几千米,西岸的沙源物质很少能被风力搬运、越过湖泊上空到东岸堆积,可见西岸物源对风沙的贡献极小,故本文不考虑西岸物源对风沙的贡献。此外,南部湖岸非常狭窄,松散沉积物很少,起沙非常困难,加上南部湖岸山体高耸陡峻,阻挡了西风搬运的绝大分部南岸沙源物质的加入,可认为南岸物源对风沙的贡献非常小,故本文也不考虑南岸物源对风沙的贡献。由此可见,错那湖地区风沙基本是由东岸和北岸的沙源物质就地搬运形成的,所以本文只将东岸和北岸作为错那湖地区风沙的潜在物源区。又根据该区域地质状况^[25]可知,桑曲为断层所在,该断层将错那湖地区分割成2个截然不同的岩性区,断层南侧（即东岸）是由花岗片麻岩、花岗岩和花岗闪长岩等组成的花岗质岩性区,断层北侧（即北岸）是由泥岩、灰岩、砂岩等组成的沉积质岩性区。根据岩性的这种二元性特征,可将错那湖东岸风沙的物质来源以桑曲为界划分为东岸和北岸2个岩性截然不同的物源区。如图3和图4所示,桑曲以北为北岸物源区,包括北岸几个冲洪积扇以及桑曲河床上的冲洪积物,基本上来源于北岸的沉积质岩石;桑曲以南为东岸物源区,包括东岸湖积物,巴索曲、龙庆南沟和东南岸的冲洪积物,以及东岸山坡上的残坡积物（包括桑曲河谷南坡）,基本上来源于东岸的花岗质岩石。

图3

新窗口打开|下载原图ZIP|生成PPT
图3错那湖地区风沙及物源空间分布

Figure 3Spatial distribution of aeolian sand and sources in the Cuona Lake area

图4

新窗口打开|下载原图ZIP|生成PPT
图4错那湖地区地表物质照片

Figure 4Surface material in the Cuona Lake area

2.2 研究方法

2.2.1 样品采集

为了合理地进行物源样和风沙样的采集,需要充分考虑研究区的地质地貌状况^[25]。考虑到错那湖地区的风沙和潜在沙源物质主要分布在河谷、冲积扇和东岸山坡上,本文主要沿该地区的河流和冲积扇进行物源样和风沙样采集。为了确保采集的样品都是风沙或沙源物质,所有样品的采样深度都在地表至20 cm深度处^[26]。此外,为了保证有充足的样品用于各种指标的测试分析,样品采集的重量基本在100~200 g左右（基岩样除外）。根据采样点总体的空间分布特征,可将采样点分为巴索曲、龙庆村南沟、桑曲3条河流采样带以及北岸3条冲洪积扇采样带。共采集风沙样20个（其中流动沙丘样5个）,物源样41个（图3）。其中,东岸物源样23个,包括基岩样2个、残坡积样8个、冲洪积样10个和湖积样3个;北岸物源样18个,包括基岩样4个（其中,3个洪积扇扇顶的基岩为紫红色砂页岩,1个桑曲的基岩以河床砾石代替,为砂岩、灰岩、花岗片麻岩等的混合物）、残坡积样3个、冲洪积样10个和湖积样1个。

2.2.2 地球化学元素测定

指纹识别技术一般使用化学元素指标作为指纹因子,测试样品的多种化学元素指标是构建最佳复合指纹因子的首要环节^[19]。结合指纹识别技术中广泛使用的多种化学元素指标^[21,22,23],本文选择Al、Ca、Fe、K、Mg、Mn、Na、Ti、Ba和Pb 10个化学元素指标,采用北京师范大学分析测试中心的电感耦合等离子体发射光谱仪（ICP-AES）测试其化学元素浓度。在测试样品的化学元素浓度之前,必须对样品进行筛选和前处理。首先,虽然采集的样品中极少有>2 mm的颗粒,但是研磨机不能处理>2 mm的粒级,所以必须将>2 mm的粒级过筛剔除掉。然后,对所有<2 mm部分的样品进行前处理,前处理步骤主要参照饶文波等^[27,28]采用的处理办法,具体处理步骤如下：

第一步：研磨。首先用三头研磨机将样品研磨至74 μm以下,用醋酸在室温下处理超过8 h,以避免风化作用产生的次生碳酸盐对样品硅酸盐组分地球化学测量结果的影响,然后在烘箱中105℃烘干,并再次用三头研磨机研磨至74 μm以下。

第二步：加酸溶解。用万分之一电子天平称取0.1 g研磨样,误差范围控制在百分之一以内,放入溶解罐,加入5.0 mL HNO₃、2.0 mL HCl和1.0 mL HF在150℃高温下溶解6 h。

第三步：蒸发与定容。打开溶解罐盖子,将样品溶液放在通风罩下烘干至1滴,再加入1.0 mL HClO₃,转移到试管定容至100 mL,然后用ICP-AES测试其化学元素浓度。

原始化学元素浓度结果以μg/0.1g形式呈现,最后需换算成Al₂O₃、CaO、Fe₂O₃、K₂O、MgO、MnO和Na₂O常量元素稳定氧化物以及Ba、Pb微量元素的重量百分含量形式参与后续的计算^[29,30]。

2.2.3 最佳复合指纹因子构建

对指纹因子进行初步筛选前,必须将异常数据进行剔除。首先,对所有物源样的各类化学元素数据进行描述统计探索分析,再将所得25分位（Q1）和75分位（Q3）数据代入以下公式^[31]：

（1）最高临界值=Q3+2.2×(Q3-Q1)最低临界值=Q1-2.2×(Q3-Q1)
如果一个物源样的某个化学元素浓度大于最高临界值,或小于最低临界值^[31],则整个物源样被视为异常值,经以上分析,7个物源样（6个基岩样,1个残坡积样）数据异常,直接予以剔除。此外,对于数据异常的风沙样5e也直接剔除。上述数据异常的样品不参与最佳复合指纹因子构建和混合模型计算^[32]。

最佳复合指纹因子是对物源样识别程度最高的多个指纹因子的组合,只有构建出最佳指纹因子才能利用多元混合模型精确地计算出各个物源的贡献率百分比^[19,22]。最佳复合指纹因子构建^[33]步骤如下：第一步,利用Kruskal-Wallis H检验,初步筛选可区分不同风沙来源的指纹因子,当H值大于卡方检验结果（即当P≤0.05）时,表明该指纹因子可区别不同风沙来源;第二步,在上一步检验结果的基础上,利用多元逐步判别分析,每步都是Wilks’ Lambda概计量最小的指纹因子进入判别函数,从而可找到最佳复合指纹因子。

2.2.4 风沙来源分析

将最终筛选出的物源样和风沙样的最佳指纹因子数据代入多元混合模型进行计算,即可定量计算出2个物源区对各风沙样的贡献率百分比。鉴于已经通过前处理剔除了有机质和粒度的影响,因此本文采用Walling等^[20]简化后的多元混合模型：

（2）Res=∑i=1nCssi-∑s=1mCsiPsCssi2
式中：R_es为残差平方和;C_ssi为风沙样中指纹因子i的浓度;P_s为物源区S的风沙贡献率;C_si为物源区S中指纹因子i的平均浓度;m为物源区数量;n为最佳指纹因子的数量。其中,风沙样和物源样的指纹因子浓度需要标准化为无量纲数据^[19],标准化便于直观比较和数据计算。在此函数中,物源区个数m=2,最佳指纹因子个数n=4。

在使用混合模型函数计算贡献率时,贡献率P_s必须满足以下条件^[34]：

（3）0≤Ps≤1;∑s=1nPs=1
在满足以上条件的基础上,当函数式 Res取最小值时,即可得到各风沙物源区的贡献百分比。此外,为了防止计算结果误差过大,所有混合模型的计算结果数据还必须通过拟合优度检验（GOF）,GOF值越接近1,说明拟合程度越好,而GOF值越接近0,说明拟合程度越差^[35]。常维娜等^[32]认为当GOF>0.8时,通过混合模型计算得到的贡献率 Ps才能被接受。Motha等^[36]给出单个指纹因子的拟合优度函数,在此基础上,常维娜等^[32]给出了最佳复合指纹因子的拟合优度函数（GOF）如下：

（4）GOF=1-1n∑i=1n|Cssi-∑s=1mCsiPsCssi

3 结果与分析

3.1 错那湖风沙及物源空间分布特征

如图3所示,湖积物、冲洪积物、风沙、残破积物和基岩呈现出从低到高的空间分布格局（图3）,裸露湖积物主要分布在湖泊东北岸,冲洪积物主要分布在冲洪积扇,风沙则主要分布在湖岸到东岸山坡之间,残破积物和基岩主要分布在高海拔山地,其中,冲洪积物、残坡积物和基岩的分布范围相对较大。

如表2和图3所示,东岸的巴索曲、龙庆南沟和桑曲3个风沙区的分布面积分别为60.0 km²、19.2 km²和7.7 km²,占研究区面积的比例依次为5.6%、1.8%和0.7%。冲洪积物的分布面积达306.9 km²,占比为28.4%,尤其在巴索曲、龙庆南沟宽谷段以及桑曲入湖冲积扇上有大量冲洪积物。裸露湖积物的分布面积仅有27.8 km²,占比为2.6%,其中西岸的湖积物与东岸风沙距离很远,基本与风沙形成无关。残坡积物的分布面积最大,达351.1 km²,占比为32.5%,东岸山体坡度大且基本无植被,表面零星出露有严重破碎的花岗质基岩,周边散布着大量片状分布的残坡积物。基岩分布面积达306.6 km²,占比为28.4%,北岸山地主体为泥岩、砂岩和灰岩等沉积质岩石,东岸山顶上则零星有花岗质岩石出露。

Table 2
表2
表2错那湖地区各物质类型空间分布统计
Table 2Statistics of the spatial distribution of all material types in the Cuona Lake area

统计指标	风沙			冲洪积	湖积	残坡积	基岩
	巴索曲	龙庆南沟	桑曲
面积占比/%	5.6	1.8	0.7	28.4	2.6	32.5	28.4
面积大小/km2	60.0	19.2	7.7	306.9	27.8	351.1	306.6

新窗口打开|下载CSV

从空间演化来看,来自错那湖东岸和北岸的各类物源碎屑物,在偏西风的吹刮下自西向东、从湖岸向山坡搬运,在湖岸和山坡之间堆积形成风沙区。错那湖风沙区是典型的风水两相作用地区,夏季降水和冰雪融水量丰富,河流迎来汛期,径流将中上游的细颗粒风化碎屑物搬运到湖滨堆积形成新的沙源物质,至来年冬春季节,强劲的西风将湖滨的细颗粒物质分选出来,通过蠕移和跃移带至东侧区域,形成风沙区,错那湖东侧和北侧的山地为本地区的风沙提供了源源不断的物源供给。

3.2 错那湖风沙物质来源分析

3.2.1 潜在物源区化学性质及指纹因子筛选

从各化学性质的平均值可看出（表3）,Fe₂O₃、MgO、MnO、Na₂O和TiO₂等化学指标在不同物源区差异较为明显。其中,北岸物源区Fe₂O₃含量是东岸物源区的1.5倍,MgO含量是东岸物源区的1.6倍,MnO含量是东岸物源区的1.5倍,TiO₂含量是东岸物源区的1.4倍,而东岸物源区Na₂O含量是北岸物源区的1.3倍,这种差异是由花岗质火成岩、变质岩和沉积碎屑岩的差异特征决定的,表明这几个化学指标能够对风沙沉积的不同物质来源进行有效辨别。将物源样10个待筛选的指纹因子用Kruskal-Wallis H检验,结果表明有6个化学指标Fe₂O₃、MgO、MnO、Na₂O、Ti₂O、Ba通过检验,作为初步筛选出的指纹因子组合,能明显判别出不同的物质来源。

Table 3
表3
表3潜在物源区化学性质及其Kruskal-Wallis H检验结果
Table 3Chemical properties of potential sources and their Kruskal-Wallis H test results

待选指纹 因子/%	物源分类							风沙沉积物
	东岸			北岸		H值	P值	东岸风沙区
	平均值	变异系数		平均值	变异系数			平均值	变异系数
Al2O3/%	6.70	0.29		7.67	0.26	1.96	0.162	4.93	0.14
CaO/%	7.16	0.36		9.61	0.56	1.68	0.195	8.03	0.33
Fe2O3/%	1.85	0.40		2.77	0.34	7.09	0.008*	1.25	0.18
K2O/%	1.76	0.25		1.88	0.22	0.62	0.432	1.38	0.14
MgO/%	0.61	0.48		0.96	0.46	7.27	0.007*	0.40	0.22
MnO/%	0.039	0.42		0.060	0.28	10.52	0.001*	0.027	0.22
Na2O/%	1.10	0.28		0.82	0.15	6.73	0.009*	0.85	0.24
TiO2/%	0.23	0.32		0.33	0.33	7.84	0.005*	0.17	0.14
Ba/%	0.032	0.25		0.039	0.25	5.08	0.024*	0.026	0.11
Pb/%	0.0015	0.31		0.0018	0.19	3.79	0.052	0.0011	0.12

注：*表示P≤0.05时差异显著。

新窗口打开|下载CSV

3.2.2 最佳复合指纹因子分析

6个初步筛选出的指纹因子再经逐步判别分析筛选后,有4个指纹因子MnO、Na₂O、Ba、Ti₂O入选最佳复合指纹因子组合,前三步的累积正确判别率依次为77.1%、91.4%和94.3%,最终的累积正确判别率可达94.4%（表4）。在2个物源区的基础上,如果将每个物源区的物质类型分为3类,那么一共可分为6种物源类型,依次为东岸残破积物、东岸冲洪积物、东岸湖积物、北岸残破积物、北岸冲洪积物和北岸湖积物。但是经Kruskal-Wallis H检验和多元判别分析后,其判别正确率仅为57.1%,表明在此种分类下并不能有效区分这6种不同物源类型,也即只能区分出东岸和北岸2个岩性不同的物源区,而不能再进一步区分每个具体的物质类型,这是因为错那湖地区的空间范围较小,尚不足以形成明显的元素迁移序列,导致不同沉积物类型的元素迁移差异很小。

Table 4
表4
表4最佳复合指纹因子
Table 4Best composite fingerprinting factors

步骤	指纹因子	Wilk统计量λ	累积正确判别率/%
1	MnO	0.724	77.1
2	Na2O	0.448	91.4
3	Ba	0.288	94.3
4	Ti2O	0.284	94.4

新窗口打开|下载CSV

3.2.3 风沙来源贡献率分析

在满足多元混合模型函数式（R_es）取最小值条件下,风沙样对应的2个不同物源区的贡献率,只有当GOF>0.8时其贡献率计算结果才能被接受^[32];当GOF<0.8时表明其未通过拟合优度检验,即其贡献率计算结果误差较大,故将GOF<0.8的风沙样44e和52e剔除。通过GOF检验的风沙样,其不同物质来源的相对贡献率及GOF检验结果如表5所示。

Table 5
表5
表5风沙来源贡献率分析结果
Table 5Contribution rates of the sources of aeolian sand

区域	样品编号	不同物源区相对贡献率/%		GOF
		东岸	北岸
巴索曲	17e	81.2	18.8	0.97
	21e	70.6	29.4	0.98
	23e	70.8	29.2	0.96
	27e	74.8	25.2	0.92
	31e	96.4	3.6	0.94
	67e	100	0	0.86
	平均值	82.3	17.7	0.94
龙庆村	49e	35.4	64.6	0.93
	37e	46.6	53.4	0.89
	38e	99.3	0.7	0.85
	平均值	60.4	39.6	0.89
桑曲	54e	9.1	90.9	0.88
	56e	6.1	93.9	0.90
	63e	15.0	85.0	0.96
	平均值	10.1	89.9	0.91

新窗口打开|下载CSV

总体上来看,东岸风沙区由南向北分为巴索曲、龙庆南沟和桑曲3个风沙分布区,风沙主要来自于湖的东岸和北岸,两个物源对不同风沙区的贡献存在较大差异,东岸是风沙的主要物源区,贡献率呈现出从巴索曲到桑曲自南向北逐渐减小的趋势,北岸的贡献较少,贡献率自北向南呈减少趋势。因此,在青藏铁路错那湖段的大部分路段,东岸物源对铁路沙漠化的威胁最大,而北岸物源对铁路沙漠化的威胁则较小,仅影响铁路以东至东岸山地西侧坡脚的狭长区域,且影响程度自南向北逐渐增大。不同风沙分布区的物源组合状况也差异很大,具体特征如下：

（1）巴索曲风沙区。巴索曲的风沙来自东岸和北岸物源区的相对贡献率范围分别为70.6%~100%以及0%~29.4%,其平均值分别为82.3%和17.7%,可见巴索曲的风沙主要来源于东岸。在主导风向西风的作用下,巴索曲的风沙从湖岸向东岸山坡搬运,受青藏铁路和层层沙障阻挡,来自湖滩和巴索曲冲积扇的沉积物组分相对减少;而次要风向西北风到达山坡后,在铁路路基东侧至山坡西侧之间近地面形成NNW向风力,使得部分风沙能往南到达较远距离,导致远程搬运来的北岸物源组分相对增加。在巴索曲西北支沟段,随着坡度的急剧增大和山谷的突然收窄,阻挡了西北风远程搬运来的部分北岸物源组分,且在地形风的加持下,次要风向西南风的风力增强,使得就地风化的残坡积物和冲洪积物的影响逐渐加大,故此风沙序列物源组成变化特征为东岸物源组分逐渐增加、北岸物源组分逐渐减少。特别是爬坡沙样31e,东岸物源区的贡献率高达96.4%,因其所处山坡非常陡且为东南倾向,基本隔绝了西北风远程搬运来的北岸物源组分。此外,在错那湖东岸风沙带的最东端,青藏公路附近的风沙样67e完全来源于东岸物源区,东岸物源贡献率达100%,说明往东远程搬运而来的北岸物源完全消失,基本上都来源于就地风化残坡积物起沙形成的东岸物源,沙物质来源趋于单纯。由此可见,东岸物源对巴索曲段铁路的威胁最大,加上山地的碎屑物质不断被水流搬运到湖滨地带形成新的沙源物质,在强劲西风的吹刮下很容易起沙,进一步增大了东岸物源对该段铁路的影响。因此,在巴索曲段铁路西侧设置防沙工程措施的基础上,还应加强铁路东侧山坡上的工程防治措施,以阻止东侧山坡上的风沙和风化碎屑物质转化成新的沙源物质。

（2）龙庆南沟风沙区。龙庆南沟的风沙来自东岸和北岸物源区的相对贡献率范围分别为35.4%~99.3%以及0.7%~64.6%,其平均值分别为60.4%和39.6%。从贡献率变化来看,龙庆南沟大致可分为3段：进沟之前、进沟后到河谷急转弯处和急转弯为近南北向后。①进沟之前,风沙主要来源于桑曲冲积物及其多条支沟的洪积扇物质,北岸物源组分占优势,此段风沙样49e的北岸物源贡献率可达64.6%。②进沟之后至急转弯处,东岸物源组分增加,北岸物源组分减少但仍有相当比重,此段风沙样37e的北岸物源贡献率降至53.4%,北岸物源组分可能更多与龙庆南沟北侧发育的多条支沟有关,这几条支沟溯源侵蚀导致其与桑曲的分水岭较低,年次要的西北向风力将桑曲起沙的物质搬运至龙庆南沟中段,再由风力就地起沙搬运沉积而成。③河谷急转弯后的沙丘样38e,东岸物源相对贡献率高达99.3%,其物质主要来源于南侧和西南侧缓坡上的风化残坡积物和东南侧的河流冲洪积物,此处的残坡积物和风沙被流水再次搬运至河谷,这是导致其北岸物源贡献率趋近于零的重要原因。由于铁路与东岸山坡之间的狭长地带有部分北岸物源加入,导致铁路东侧的北岸物源组分高于其西侧。因此,在龙庆南沟段铁路两侧设置防沙工程措施的基础上,还亟需在铁路与东岸山坡之间的狭长地带也设置一些东西向延伸的防沙工程措施,以阻断北岸物源继续向南搬运,而目前这一狭长地带尚未建造东西向的防沙工程体系。

（3）桑曲风沙区。桑曲的风沙来自东岸和北岸物源区的相对贡献率分别为6.1%~15.0%和85.0%~93.9%,其平均值分别为10.1%和89.9%。由此可见,桑曲的风沙主要来源于北岸,次要风向西南风携带来的沙源物质极少。其中,位于桑曲南侧陡峻山坡上的爬坡沙样63e,其东岸物源的相对贡献率增加到15%,比另2个位于桑曲入湖冲积扇上的风沙样的东岸物源组分高,显然是因为63e处地势陡峭,导致在搬运过程中北岸冲洪积物源减少,就地风化的东岸残坡积物增加造成的。由于桑曲风沙区的北侧为河谷宽广且水量丰富的桑曲,南侧为NE-SW走向的陡坡,限制了风沙的南北向扩散。由于桑曲冲积扇上的沙源物质非常丰富,再加上河谷的风力强劲,因此桑曲段铁路西侧还需要进一步加强防沙工程措施。

3.2.4 风沙物源贡献差异的成因分析

巴索曲、龙庆南沟、桑曲3个风沙带的风沙物质来源有明显差异,巴索曲主要来源于东岸物源,桑曲主要来源于北岸物源,龙庆南沟则呈前两者的过渡状态。在巴索曲和龙庆南沟的风沙序列中存在一些波动性变化,这些变化可能与以下影响因素有关：①搬运距离。在桑曲、巴索曲、龙庆南沟河谷以及湖东近岸地区,以湖积物和冲洪积物为主导物源,随着离干流和湖泊距离增大,就近的冲洪积物和残坡积物影响越来越大,因此越往东去东岸物源贡献率越大;②风向。主要风向西风与河谷走向相适应,是造成风沙带状分布的主要原因,也是东岸物源总体占优势的原因,次要风向西北风沿山地西侧和支沟河谷向南和东南侧搬运,形成北岸物源组分的补充;③地形。桑曲南侧的山岭一定程度上阻挡了北岸物源组分的进入,仅局部低分水岭构成北岸物源通道,山坡高处风沙主要来源于就近物质补充;④交通和防沙工程措施。次要风向西北风受西侧山坡和工程措施尤其是铁路路基的影响转为NNW向,形成狭长的风沙搬运带,是造成铁路东侧北岸物源组分高于其西侧的主要原因。

4 结论与讨论

本文通过遥感解译、野外调查采样和化学元素测试,初步查明了错那湖地区风沙及物源的空间分布,并定量化分析了风沙沉积的物源组合及其贡献百分比,得到结论如下：

（1）错那湖、青藏铁路和东岸风沙区总体上呈自西向东的空间演替,来自错那湖东岸和北岸的物源碎屑物,在偏西风的吹刮下自西向东搬运,越过青藏铁路,在东岸山坡上堆积形成东岸风沙区。受风水两相作用,错那湖东侧和北侧的山地为湖滨的沙源提供了充足的物质供给。

（2）总体上来看,错那湖地区的3个风沙区（巴索曲、龙庆南沟和桑曲）的风沙物源组成有很大差异。3个风沙区东岸物源的平均贡献率依次为89.9%、39.6%和17.7%,呈现出自北向南逐渐减小的趋势。此外,每个风沙分布区内部风沙的物源组成也有一定差异。具体来看,巴索曲的风沙主要来源于东岸物源区,其范围为70.6%~100%;龙庆南沟风沙的东岸物源贡献率自西向东从35.4%增加到99.3%;桑曲的风沙主要来源于北岸冲洪积物,北岸物源贡献率范围为85.0%~93.9%。影响风沙物源组合的主要影响因素包括搬运距离、地形、风向和交通与防沙工程措施等。

（3）结合青藏铁路的路线分布可知,在青藏铁路错那湖段的大部分路段,东岸物源演化形成的风沙对铁路的威胁最大,北岸物源演化形成的风沙对铁路的影响则较小,仅影响铁路以东至山坡的狭长区域,且影响程度自南向北逐渐增大。因此,在铁路西侧防沙措施的基础上,还应加强铁路东侧风沙和风化碎屑物的防治,并在铁路东侧到山坡之间的狭长地带也增加一些防沙设施。

（4）复合指纹识别技术构建的最佳复合指纹因子可以精确计算出东岸和北岸物源对风沙的贡献率百分比,说明使用复合指纹识别技术定量化查明小空间范围内风沙的物质来源组合是可行的。

需要指出,复合指纹识别技术应用的前提是能够有效识别出潜在物源类型,本文的研究区是一个封闭湖盆,很少受外界物质的干扰,物源也相对简单,因此使用复合指纹识别技术能够较为精确地查明风沙的物质来源。但是,对于受外部影响较大的开放区域,潜在物源存在很大的不确定性,这种情况下复合指纹识别技术不一定适用。此外,为进一步提高风沙物源贡献率的计算精度,建议综合考虑水系、风向、地形和交通等因素,更科学地布置采样点,并且增加多种测试指纹因子,构建更合理的最佳复合指纹因子。

参考文献 View Option 原文顺序
文献年度倒序
文中引用次数倒序
被引期刊影响因子

[1] Bergstrom D J, Boucher K M, Derksen D . Wind flow over an elevated roadway
[J]. Journal of Wind Engineering & Industrial Aerodynamics, 1992,44(1) : 2697-2698.
DOI:10.1038/s41598-019-56514-8URLPMID:31882768 [本文引用: 1]
Climate variability and climate change in Eastern Boundary Upwelling Systems (EBUS) affect global marine ecosystems services. We use passive tracers in a global ocean model hindcast at eddy-permitting resolution to diagnose EBUS low-frequency variability over 1958-2015 period. The results highlight the uniqueness of each EBUS in terms of drivers and climate variability. The wind forcing and the thermocline depth, which are potentially competitive or complementary upwelling drivers under climate change, control EBUS low-frequency variability with different contributions. Moreover, Atlantic and Pacific upwelling systems are independent. In the Pacific, the only coherent variability between California and Humboldt Systems is associated with El Ni?o Southern Oscillation. The remaining low-frequency variance is partially explained by the North and South Pacific expressions of the Meridional Modes. In the Atlantic, coherent variability between Canary and Benguela Systems is associated with upwelling trends, which are not dynamically linked and represent different processes. In the Canary, a negative upwelling trend is connected to the Atlantic Multi-decadal Oscillation, while in the Benguela, a positive upwelling trend is forced by a global sea level pressure trend, which is consistent with the climate response to anthropogenic forcing. The residual variability is forced by localized offshore high sea level pressure variability.

[2] 张克存, 屈建军, 鱼燕萍 , 等. 中国铁路风沙防治的研究进展
[J]. 地球科学进展, 2019,34(6):573-583.
[本文引用: 2]

[ Zhang K C, Qu J J, Yu Y P , et al. Progress of research on wind-blown sand prevention and control of railways in China
[J]. Advances in Earth Science, 2019,34(6):573-583.]
[本文引用: 2]

[3] 冯连昌, 卢继清, 邸耀全 . 中国沙区铁路沙害防治综述
[J]. 中国沙漠, 1994,14(3):47-53.
[本文引用: 2]

[ Feng L C, Lu J Q, Di Y Q . Review on the prevention of sand damages to railway line desert areas of China
[J]. Journal of Desert Research, 1994,14(3):47-53.]
[本文引用: 2]

[4] 屈建军, 凌裕泉, 刘宝军 , 等. 我国风沙防治工程研究现状及发展趋势
[J]. 地球科学进展, 2019,34(3):225-231.
[本文引用: 1]

[ Qu J J, Ling Y Q, Liu B J , et al. The research status and development trends of wind-sand engineering in China
[J]. Advances in Earth Science, 2019,34(3):225-231.]
[本文引用: 1]

[5] 鲍锋, 董治宝 . 青藏铁路察尔汗盐湖段风沙活动特征
[J]. 中国沙漠, 2017,37(4):621-625.
[本文引用: 1]

[ Bao F, Dong Z B . Characteristic of sand-drift activities along the Qarhan Salt Lake section of Qinghai-Tibet Railway
[J]. Journal of Desert Research, 2017,37(4):621-625.]
[本文引用: 1]

[6] 肖建华 . 青藏铁路(格拉段)风沙危害及其防治研究
[D]. 北京: 中国科学院大学, 2014.
[本文引用: 1]

[ Xiao J H . Study on Blown Sand Disaster and Its Prevention and Control of Qinghai-Tibet Railway (Golmud-Lhasa Section)
[D]. Beijing: University of Chinese Academy of Sciences, 2014.]
[本文引用: 1]

[7] Jiang Y S, Gao Y H, Dong Z B , et al. Simulations of wind erosion along the Qinghai-Tibet Railway in north-central Tibet
[J]. Aeolian Research, 2018,32:192-201.
DOI:10.1016/j.aeolia.2018.03.006URL [本文引用: 2]

[8] 姚正毅, 屈建军 . 青藏铁路格尔木-拉萨段风成沙物源及其粒度特征
[J]. 中国沙漠, 2012,32(2):300-307.
URL [本文引用: 3]
青藏铁路格尔木-拉萨段沿线地表松散沉积物广泛分布,为风沙活动的产生提供了丰富的物质来源。地表松散沉积物按成因分为现代风成沙,河流冲积物,古风成沙和洪积、湖积沉积物。粒度分析结果显示,格-拉段沿线风沙物质（沙丘沙、防沙体系积沙）样品（56个）平均粒径变化于1.29～3.25 &Phi;之间,均值2.36 &Phi;（0.19 mm）。平均粒径在2～3 &Phi;（0.25～0.125 mm）之间的样品占总数的78.57%。粒度组成以细沙为主（65.20%）,其次是中沙（20.53%）,0.5～0.1 mm范围重量百分比在49.44%～99.67%之间,平均为85.98%。在所有样品中,细沙和中沙合计在90%以上者占样品总数的55.36%。极细沙平均含量为7.99%,粗沙平均含量为5.50%。样品整体分选程度较好,标准偏差&sigma;1在0.3～1.26之间,平均为0.58。偏度变化在-0.41到 0.36之间,样品中以正态分布为主,占53.57%;正偏者占12.50%;负偏者占19.64%;极负偏和极正偏各占7.14%。KG值变化于0.56～1.24,平均为1.00,为中等峰态。KG值在0.90～1.11之间者占71.43%。粒度分析表明,不同来源沙物质粒度特征有明显的差别,能很好地反映其形成过程和环境特征。
[ Yao Z Y, Qu J J . Source and grain size of aeolian sands along Golmud-Lhasa section of Qinghai-Tibet Railway
[J]. Journal of Desert Research, 2012,32(2):300-307.]
URL [本文引用: 3]
青藏铁路格尔木-拉萨段沿线地表松散沉积物广泛分布,为风沙活动的产生提供了丰富的物质来源。地表松散沉积物按成因分为现代风成沙,河流冲积物,古风成沙和洪积、湖积沉积物。粒度分析结果显示,格-拉段沿线风沙物质（沙丘沙、防沙体系积沙）样品（56个）平均粒径变化于1.29～3.25 &Phi;之间,均值2.36 &Phi;（0.19 mm）。平均粒径在2～3 &Phi;（0.25～0.125 mm）之间的样品占总数的78.57%。粒度组成以细沙为主（65.20%）,其次是中沙（20.53%）,0.5～0.1 mm范围重量百分比在49.44%～99.67%之间,平均为85.98%。在所有样品中,细沙和中沙合计在90%以上者占样品总数的55.36%。极细沙平均含量为7.99%,粗沙平均含量为5.50%。样品整体分选程度较好,标准偏差&sigma;1在0.3～1.26之间,平均为0.58。偏度变化在-0.41到 0.36之间,样品中以正态分布为主,占53.57%;正偏者占12.50%;负偏者占19.64%;极负偏和极正偏各占7.14%。KG值变化于0.56～1.24,平均为1.00,为中等峰态。KG值在0.90～1.11之间者占71.43%。粒度分析表明,不同来源沙物质粒度特征有明显的差别,能很好地反映其形成过程和环境特征。

[9] 段青龙 . 青藏铁路错那湖活动沙丘的形成机制及治理措施
[J]. 岩土工程技术, 2002, ( 6):311-314.
URL [本文引用: 1]
通过对青藏铁路错那湖活动沙丘分布特征及形成机制的剖析,提出了合理的工程地质选线原则及工程治理措施,以使风沙活动对铁路工程及铁路运营的危害程度尽可能地减少.
[ Duan Q L . Formation mechanism and treatment measure of the motive sand-flood of Cuona Lake on Qinghai-Tibet Railway
[J]. Geotechnical Engineering Technique, 2002, ( 6):311-314.]
URL [本文引用: 1]
通过对青藏铁路错那湖活动沙丘分布特征及形成机制的剖析,提出了合理的工程地质选线原则及工程治理措施,以使风沙活动对铁路工程及铁路运营的危害程度尽可能地减少.

[10] 蔡迪文, 张克存, 安志山 , 等. 青藏铁路格拉段风动力环境及其对铁路沙害的影响
[J]. 中国沙漠, 2017,37(1):40-47.
[本文引用: 1]

[ Cai D W, Zhang K C, An Z S , et al. Wind energy environments and its impacts on railway sand hazards along Gerlha section of the Qinghai-Tibet Railway
[J]. Journal of Desert Research, 2017,37(1):40-47.]
[本文引用: 1]

[11] 陈长委, 伍永秋, 谭利华 , 等. 青藏铁路错那湖段沙漠化土地变化及成因分析
[J]. 干旱区地理, 2019,42(4):885-892.
[本文引用: 1]

[ Chen C W, Wu Y Q, Tan L H , et al. Analysis on the desertified land change and its causes in Co Nag Lake region along Qinghai-Tibet Railway
[J]. Arid Land Geography, 2019,42(4):885-892.]
[本文引用: 1]

[12] 王涛 . 中国沙漠与沙漠化[M]. 石家庄: 河北科学技术出版社, 2003.
[本文引用: 2]

[ Wang T. Desert and Aeolian Desertification in China[M]. Shijiazhuang: Hebei Science & Technology Press, 2003.]
[本文引用: 2]

[13] Aarons S M, Blakowski M A, Aciego S M , et al. Geochemical characterization of critical dust source regions in the American West
[J]. Geochimica et Cosmochimica Acta, 2017,215:141-161.
DOI:10.1016/j.gca.2017.07.024URL [本文引用: 1]

[14] Muhs D R . Evaluation of simple geochemical indicators of aeolian sand provenance: Late Quaternary dune fields of North America revisited
[J]. Quaternary Science Reviews, 2017,171:260-296.
DOI:10.1016/j.quascirev.2017.07.007URL [本文引用: 1]

[15] Budahn J R, Muhs D R . New geochemical evidence for the origin of North America’s largest dune field, the Nebraska Sand Hills, central Great Plains, USA
[J]. Geomorphology, 2019,332:188-212.
DOI:10.1016/j.geomorph.2019.02.023URL [本文引用: 1]

[16] Jiang Q D, Yang X P . Sedimentological and geochemical composition of aeolian sediments in the Taklamakan Desert: Implications for provenance and sediment supply mechanisms
[J]. Journal of Geophysical Research, 2019,124(5):1217-1237.
[本文引用: 1]

[17] Zhao W C, Liu L W, Chen J , et al. Geochemical characterization of major elements in desert sediments and implications for the Chinese loess source
[J]. Science China (Earth Sciences), 2019,62(9):1428-1440.
DOI:10.1007/s11430-018-9354-yURL [本文引用: 1]

[18] Wang X M, Sun J M, Qiang M R , et al. Geochemical characteristics of dust aerosol availability in northwestern China
[J]. Arabian Journal of Geosciences, 2019, DOI: http://www.resci.cn/article/2019/1007-7588/10.1007/s12517-019-4580-0.
[本文引用: 1]

[19] Collins A L, Walling D E, Leeks G J L . Source type ascription for fluvial suspended sediment based on a quantitative composite fingerprinting technique
[J]. Catena, 1997,29(1):1-27.
DOI:10.1016/S0341-8162(96)00064-1URL [本文引用: 4]

[20] Walling D E . Tracing suspended sediment sources in catchments and river systems
[J]. Science of the Total Environment, 2005,344(1-3):159-184.
DOI:10.1016/j.scitotenv.2005.02.011URLPMID:15907516 [本文引用: 2]
Recent years have seen a growing awareness of the wider environmental significance of the suspended sediment loads transported by rivers and streams. This includes the importance of fine sediment in the transport of nutrients and contaminants, including phosphorus (P). Sediment source exerts a key control on the physical and geochemical properties of suspended sediment, including its P content, and will also influence the potential for implementing effective sediment and diffuse source pollution control strategies. Information on suspended sediment source, defined in terms of both source type and spatial origin, is therefore increasingly needed. Such information is difficult to obtain using traditional monitoring techniques, but source tracing or fingerprinting techniques afford a valuable and effective alternative approach to establish the relative importance of potential sediment sources. This contribution reviews the development of source fingerprinting techniques, presents several examples of their application in UK catchments and discusses the need for future development of the approach and the potential for extending its application.

[21] Gholami H, Telfer M T, Blake W H , et al. Aeolian sediment fingerprinting using a Bayesian mixing model
[J]. Earth Surface Processes and Landforms, 2017,42(14):2365-2376.
DOI:10.1002/esp.v42.14URL [本文引用: 2]

[22] Niu B C, Qu J J, Zhang X C , et al. Quantifying provenance of reservoir sediment using multiple composite fingerprints in an arid region experiencing both wind and water erosion
[J]. Geomorphology, 2019,332:112-121.
DOI:10.1016/j.geomorph.2019.02.011URL [本文引用: 3]

[23] Patault E, Alary C, Franke C , et al. Quantification of tributaries contributions using a confluence-based sediment fingerprinting approach in the Canche River Watershed (France)
[J]. Science of the Total Environment, 2019,668:457-469.
DOI:10.1016/j.scitotenv.2019.02.458URLPMID:30852221 [本文引用: 2]
Since a few years, land use management aims to reduce and control water erosion processes in watersheds but there is a lack of quantitative information on the contribution of the sources of transported sediment. This is most important in agricultural areas where soils are sensitive to erosion. The geology of these areas is often characterized by large expanses of relatively homogeneous quaternary silts. The possibility of distinguishing the sources of erosion according to their geological substratum is thus very delicate. This information is important because its lack can lead to the mis-implementation of erosion control measures. To address this request, a confluence-based sediment fingerprinting approach was developed on the Canche river watershed (1274?km²; northern France), located in the European loess belt, an area that is affected by diffuse and concentrate erosion processes. Suspended particulate matter was collected during five seasonal sampling campaigns using sediment traps at the outlet of each tributary and confluence with the main stream of the Canche river. The final composite fingerprint was defined using physico-chemical and statistical analyses. The best tracer parameters for each tributary were selected using stepwise discriminant function analyses. These parameters were introduced into a mass balance mixing model incorporating Monte-Carlo simulations to represent the uncertainty. Estimates of the overall mean contributions from each tributary were quantified at different temporal scales. The annual sediment flux tributaries contributions range from 3 to 22% at the outlet of the Canche river, and annual sediment flux range from 0.87 to 40.7?kt?yr^-1. The Planquette and the Créquoise tributaries appear to be those producing the largest sediment flux. In contrast, tributaries with the highest number of erosion control on their area exhibit the lowest values of sediment flux. Our results indicate a positive impact of recent land management policies in the Canche river watershed.

[24] 马淑红, 戈峰, 陈晓光 , 等. 古尔班通古特沙漠瞬间最大风速的时空分布特征
[J]. 资源科学, 2007,29(4):46-53.
URL [本文引用: 1]
利用古尔班通古特沙漠地区以及周围20个气象站45年（1961年～2005年）历史资料，结合古尔班通古特沙漠3个野外风沙监测站近5年（2001年～2005年）的5层梯度风（0.5m、2.0m、4.0m、6.0m、10.0m）野外监测资料，对古尔班通古特沙漠瞬间最大风速时空分布与活化沙丘治理对策进行分析。在此基础上，选取活化沙丘顶部为床面，分析了古尔班通古特沙漠活化沙丘顶部边界层内10分钟平均最大风速与瞬间最大风速空间相关性，发现风沙边界层内任意两层的瞬间最大风速相关显著，相关系数在0.9654～0.9922之间，且随高度增加相关系数呈现增加，而影响这一变化特征的主要因素是下垫面状况。本文通过大量分析，建立了活化沙丘顶部风沙边界层内1分钟平均风速的预测模式，其公式为u=（Z/Z₁)^0.1026，为今后古尔班通古特沙漠活化沙丘的治理提供科学依据。
[ Ma S H, Ge F, Chen X G , et al. The spatial-temporal distribution of maximum instantaneous wind speed and the control of active dunes in Gurbantunggut Desert
[J]. Resources Science, 2007,29(4):46-53.]
URL [本文引用: 1]
利用古尔班通古特沙漠地区以及周围20个气象站45年（1961年～2005年）历史资料，结合古尔班通古特沙漠3个野外风沙监测站近5年（2001年～2005年）的5层梯度风（0.5m、2.0m、4.0m、6.0m、10.0m）野外监测资料，对古尔班通古特沙漠瞬间最大风速时空分布与活化沙丘治理对策进行分析。在此基础上，选取活化沙丘顶部为床面，分析了古尔班通古特沙漠活化沙丘顶部边界层内10分钟平均最大风速与瞬间最大风速空间相关性，发现风沙边界层内任意两层的瞬间最大风速相关显著，相关系数在0.9654～0.9922之间，且随高度增加相关系数呈现增加，而影响这一变化特征的主要因素是下垫面状况。本文通过大量分析，建立了活化沙丘顶部风沙边界层内1分钟平均风速的预测模式，其公式为u=（Z/Z₁)^0.1026，为今后古尔班通古特沙漠活化沙丘的治理提供科学依据。

[25] 潘桂棠, 丁俊 . 青藏高原及邻区地质图[M]. 成都: 成都地图出版社, 2004.
[本文引用: 2]

[ Pan G T, Ding J. Geological Map of the Qinghai-Tibet Plateau and Adjacent Areas[M]. Chengdu: Chengdu Map Press, 2004.]
[本文引用: 2]

[26] Du S S, Wu Y Q, Tan L H . Geochemical evidence for the provenance of aeolian deposits in the Qaidam Basin, Tibetan Plateau
[J]. Aeolian Research, 2018,32:60-70.
DOI:10.1016/j.aeolia.2018.01.005URL [本文引用: 1]

[27] Rao W B, Chen J, Tan H B , et al. Sr-Nd isotopic and REE geochemical constraints on the provenance of fine-grained sands in the Ordos Deserts, north-central China
[J]. Geomorphology, 2011,132(3-4):123-138.
DOI:10.1016/j.geomorph.2011.05.003URL [本文引用: 1]

[28] Rao W B, Chen J, Tan H B , et al. Nd-Sr isotopic and REE geochemical compositions of Late Quaternary deposits in the desert-loess transition, north-central China: Implications for their provenance and past wind systems
[J]. Quaternary International, 2014, 334-335:197-212.
DOI:10.1016/j.quaint.2013.06.009URL [本文引用: 1]

[29] Liu Q Q, Yang X P . Geochemical composition and provenance of aeolian sands in the Ordos Deserts, northern China
[J]. Geomorphology, 2018,318:354-374.
DOI:10.1016/j.geomorph.2018.06.017URL [本文引用: 1]

[30] Ding Z Y, Lu R J, Lv Z Q , et al. Geochemical characteristics of Holocene aeolian deposits east of Qinghai Lake, China, and their paleoclimatic implications
[J]. Science of the Total Environment, 2019,692:917-929.
DOI:10.1016/j.scitotenv.2019.07.099URLPMID:31539996 [本文引用: 1]
The paleoclimate evolution of the northeastern Tibetan Plateau (TP), especially in the Qinghai Lake Basin (QLB), has long been a subject of interest to scholars due to the particularity of the geographical location. However, because of the uncertainties of the chronologies and the interpretations of the proxies used, climate change in this region remains controversial during the Holocene. The Hudong dunefield is located to the east of Qinghai Lake and is the largest sand accumulation area in the QLB. In this study, deposits in the Holocene aeolian sand-paleosol sequences Chengou East (CGE) and Qinghaihu Country (QHH) in the Hudong dunefield were analyzed to determine their elemental geochemical characteristics and paleoclimatic implications. Combining the grain size, total organic carbon (TOC) and redness, we investigate the paleoclimate changes in this region and its response to the East Asian Summer Monsoon (EASM) during the Holocene. The high Na₂O/Al₂O₃ ratios and low chemical index of alteration (CIA) values suggest that most of the sediments are unweathered to weakly weathered, although some mid-Holocene samples are moderately weathered. The multiproxy analysis indicates that the local climate was broadly coincident with that of the northeastern TP and most regions of northern China, implying that the paleoclimate of the QLB was closely related with the evolution of the EASM during the Holocene. Additionally, after the 9.2?ka BP cold event, the chemical weathering increased gradually. The higher CIA and TOC contents and lower redness and mean grain size from 8.7 to 4.0?ka BP are possibly associated with the mid-Holocene optimum period and indicate intensified chemical weathering, denser vegetation cover and weakened aeolian activity in the QLB in response to a warmer and more humid climate. After 4.0?ka BP, the lower degrees of chemical weathering indicate that the study area was dominated by a relatively cold and dry climate, and several alternating warm-wet and cold-dry intervals occurred from 3.1 to 0.6?ka BP.

[31] 王旭 . SPSS数据处理与分析[M]. 北京: 人民邮电出版社, 2016.
[本文引用: 2]

[ Wang X. Data Processing and Analysis Based on SPSS[M]. Beijing: People Post Press, 2016.]
[本文引用: 2]

[32] 常维娜, 周慧平, 高燕 . 基于复合指纹法的九乡河小流域泥沙来源解析
[J]. 水土保持学报, 2014,28(6):106-110.
[本文引用: 4]

[ Chang W N, Zhou H P, Gao Y . Sediment sources apportionment in a small catchment of the upper Jiuxiang River using composite fingerprinting
[J]. Journal of Soil and Water Conservation, 2014,28(6):106-110.]
[本文引用: 4]

[33] Poleto C, Merten G H, Minella J P . The identification of sediment source in a small urban watershed in southern Brazil: An application of sediment fingerprinting
[J]. Environmental Technology, 2009,30(11):1145-1153.
DOI:10.1080/09593330903112154URLPMID:19947145 [本文引用: 1]
Soil particles eroded from the land surface and transported into rivers by runoff are considered one of the main components of non-point source pollution in urban watersheds. These particles also serve as a vector for a wide variety of both organic and inorganic constituents. As a result, the identification of sediment sources in an urban watershed is necessary not only to understand erosion dynamics, but also to help implement more effective measures to control and/or remediate non-point source pollution. The present study employs sediment 'fingerprinting' to determine the main sediment sources in a small residential urban watershed (0.83 km2) on the outskirts of Porto Alegre in southern Brazil. Based on an evaluation spanning 12 rainfall events, the results show that paved and unpaved roads and the stream channel itself contribute, on average, 46%, 23%, and 31%, respectively, to the suspended sediment flux in the watershed. Furthermore, the source contributions varied both between events and over the course of a single event. This appears to imply that source contributions, at least to some extent, depend on local precipitation patterns. The results from this study indicate that the level of uncertainty in source ascription tends to decline with increasing numbers of tracers; hence, successful sediment fingerprinting and source ascription in complex hydrologic environments, such as urban watersheds, may require the use of a large number of chemical and/or physical tracers.

[34] Franz C, Makeschin F, Wei B H , et al. Sediment in urban river basins: Identification of sediment sources within the Lago Paranoa Catchment, Brasilia D F, Brazil-using the fingerprint approach
[J]. Science of the Total Environment, 2014, 466-467:513-523.
DOI:10.1016/j.scitotenv.2013.07.056URLPMID:23933453 [本文引用: 1]
The development of effective sediment management strategies is a key requirement in tropical areas with fast urban development, like Brasilia DF, Brazil, because of the limited resources available. Accurate identification and management of sediment sources areas, however, is hampered by the dearth of reliable information on the primary sources of sediment. Few studies have attempted to quantify the source of sediment within fast urbanizing, mixed used, tropical catchments. In this study, statistically verified composite fingerprints and a multivariate mixing model have been used to identify the main land use specific sources of sediment deposited in the artificial Lago Paranoá, Central Brazil. Because of the variability of urban land use types within the Lago Paranoá sub-catchments, the fingerprinting approach was additionally undertaking for the Riacho Fundo sub-catchment. The main contributions from individual source types (i.e. surface materials from residential areas, constructions sites, road deposited sediment, cultivated areas, pasture, farm tracks, woodland and natural gullies) varied between the whole catchment and the Riacho Fundo sub-catchment, reflecting the different proportions of land uses. The sediments deposited in the silting zones of the Lago Paranoá originate largely from urban sources (85 ± 4%). Areas with (semi-) natural vegetation and natural gullies contribute 10 ± 2% of the sediment yield. Agricultural sites have only a minor sediment contribution of about 5 ± 4% within the whole catchment. Within the Riacho Fundo sub-catchment there is a significant contribution from urban (53 ± 4%) source, such as residential areas with semi-detached housings (42 ± 3%) with unpaved roads (12 ± 3%) and construction sites (20 ± 3%) and agricultural areas (31 ± 2%). The relative contribution from land use specific sources to the sediment deposition in the silting zone of the Lago Paranoá demonstrated that most of the sediment is derived from sites with high anthropogenic impact.

[35] Reghenzani F, Massari G, Santinelli L , et al. Statistical power estimation dataset for external validation GoF tests on EVT distribution
[J]. Data in Brief, 2019,25:1-10.
DOI:10.1016/j.dib.2019.104071URLPMID:31211211 [本文引用: 1]
This paper presents the statistical power estimation of goodness-of-fit tests for Extreme Value Theory (EVT) distributions. The presented dataset provides quantitative information on the statistical power, in order to enable the sample size selection in external validation scenario. In particular, high precision estimations of the statistical power of KS, AD, and MAD goodness-of-fit tests have been computed using a Monte Carlo approach. The full raw dataset resulting from this analysis has been published as reference for future studies: https://doi.org/10.17632/hh2byrbbmf.1.

[36] Motha J A, Wallbrink P J, Hairsine P B , et al. Determining the sources of suspended sediment in a forested catchment in southeastern Australia
[J]. Water Resources Research, 2003,39(3):1056-1069.
[本文引用: 1]