沟道是黄土高原重要的地形特征和水文特征,汇流量阈值是采用水文分析方法提取沟道的关键,目前已有汇流量阈值的确定方法具有较强的主观性。为解决汇流量阈值不确定性的问题,采用30 m分辨率DEM为基础数据,以延安市宝塔区碾庄沟流域为研究区,提出了一种基于受试者工作特征曲线(ROC曲线)确定汇流量阈值的方法。首先采集69个实际沟道起点,分析沟道起点的汇流特征,并确定了汇流量集聚区间;其次提取了500个一级沟道点作为样本点,采用ROC曲线分析了汇流量对沟道的诊断价值,得到了汇流量最优阈值。结果表明:汇流量对沟道具有较高的诊断价值,存在某一汇流量阈值能够较好地判定样本点是否为沟道;采用ROC曲线法确定的汇流量最优阈值提取的沟道均方根误差<2.5个像元,结果较为理想。
Abstract
Gully is the essential geographical and hydrological feature of loess plateau. Runoff accumulation threshold is the key to extracting loess gully by hydrological analysis. Present method for determining the accumulation threshold is highly subjective. To address the uncertainty problem of the accumulation threshold, we present a method of determining the runoff accumulation threshold based on the receiver operating characteristic (ROC) curve. We select the 30 m resolution DEM of Nianzhuanggou catchment in Baota District of Yan’an City as the basic data. Firstly, we collect 69 starting points of actual gullies and analyze the convergence characteristics of the starting points, and determine the interval of confluence gathering. Secondly, we extracted 500 first-order gully points as sample points and obtain the optimal threshold of runoff accumulation by analyzing the diagnostic value of runoff accumulation to the gully based on ROC curve. Research results demonstrate that the runoff accumulation is of high diagnostic value for gully. An accumulation threshold could well determine whether the sample point is a gully; the root mean square error of the gully extracted under the optimal accumulation threshold determined by the ROC curve analysis method is less than 2.5 pixels, and the result is ideal.
关键词
黄土沟道 /
沟道提取 /
水文分析 /
汇流量阈值 /
ROC曲线
Key words
loess channel /
gully extraction /
hydrological analysis /
runoff accumulation threshold /
ROC curve
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 刘彦随,李裕瑞.黄土丘陵沟壑区沟道土地整治工程原理与设计技术[J].农业工程学报,2017, 33(10):1-9.
[2] 魏宏安,王介勇.延安市黄土丘陵沟壑区沟道土地整治适宜性评价研究[J].地域研究与开发, 2013,32(3):129-132.
[3] 崔海兴,郑风田,王立群.退耕还林工程对耕地利用影响的实证分析:以河北省沽源县为例[J]. 农村经济, 2009(3):28-31.
[4] 肖 潇,铁 鑫.浅谈可行性研究报告中的风险分析[J].中国工程咨询,2013(7):23-24.
[5] 李 喆,伍 文.成都市土地开发整理工程选址决策规则集适宜性评价[J].中国土地科学, 2013,27(10):57-61.
[6] 闾国年,钱亚东,陈钟明,等.黄土丘陵沟壑区沟谷网络自动制图技术研究[J].测绘学报, 1998,27(2):131-137.
[7] BAND L E. Topographic Partition of Watersheds with Digital Elevation Models[J]. Water Resources Research, 1986, 22(1):15-24.
[8] TARBOTON D G,BRAS R L,RODRIGUEZITURBE I. On the Extraction of Channel Networks from Digital Elevation Data[J]. Hydrological Processes,1991,5(1):81-100.
[9] 陈 涛 ,朱金兆 ,马 浩,等.黄土高原沟壑区沟道提取时阈值范围的研究[J].长江科学院院报,2008,25(3):28-30.
[10]GYASIAGYEI Y, WILLGOOSE G, DE TROCH F. Effects of Vertical Resolution and Map Scale of Digital Elevation Models on Geomorphological Parameters Used in Hydrology[J]. Hydrological Processes, 2010, 9(3/4): 363-382.
[11]LUO Ming-liang, TANG Guo-an, DONG You-fu. Uncertainty of Flow Accumulation Threshold Influence in Hydrology Modeling—A Case Study in Qinling Mountain SRTM3 DEM based[C]∥Proceedings of 2008 International Workshop on Education Technology and Training & 2008 International Workshop on Geoscience and Remote Sensing. DOI: 10.1109/ETTandGRS.2008.280.
[12]熊立华,郭生练.基于DEM的数字河网生成方法的探讨[J].长江科学院院报,2003, 20(4):14-17.
[13]李 俊,汤国安,张 婷,等.利用DEM提取陕北黄土高原沟谷网络的汇流阈值研究[J].水土保持通报,2007,27(2):75-78.
[14]孔凡哲,李莉莉.利用DEM提取河网时集水面积阈值的确定[J].水电能源科学,2005,23(4):65-67.
[15]苑露莎,郑新奇,郭继发.DEM提取沟谷的尺度效应分析[C]∥2013全国土地资源开发利用与生态文明建设学术研讨会论文集.西宁:中国自然学会,中国地理学会,2013:403-409.
[16]沈晶玉,史明昌,田玉柱,等.DEM网格尺寸与沟谷提取精度研究[J].中国水土保持,2007, 2007(2):56-59.
[17]吴秉校,侯 雷,宋敏敏,等.基于汇流累积计算的沟壑密度分析方法[J].水土保持研究,2017, 24(3):39-44.
[18]巫晓玲,熊礼阳,胡二乐,等.基于DEM的黄土高原沟谷提取汇流阈值自适应研究[J].地理与地理信息科学,2017,33(4):93-98.
[19]卞鸿雁.黄土高原南部地区土地利用变化与生态效应时空动态测评[D].西安:陕西师范大学,2011.
[20]GROHMANNC H. Evaluation of TanDEM-X DEMs on Selected Brazilian Sites: Comparison with SRTM, ASTER GDEM and ALOS AW3D30[J]. Remote Sensing of Environment, 2018, 212:121-133.
[21]JAIN A O, THAKER T, CHAURASIA A, et al. Vertical Accuracy Evaluation of SRTM-GL1, GDEM-V2, AW3D30 and CartoDEM-V3.1 of 30-m Resolution with Dual Frequency GNSS for Lower Tapi Basin India[J]. Geocarto International, 2017: 1-20. DOI: 10.1080/10106049.2017.1343392.
[22]赵明伟,汤国安,李发源.黄土塬区DEM水文分析中消除地面伪沟谷的方法[J].中国水土保持, 2011(8):36-38.
[23]GREENLEED D. Raster and Vector Processing for Scanned Linework[J]. Photogrammetric Engineering & Remote Sensing, 1987, 53(10): 1383-1387.
[24]ZHANG T, HAN L, CHENW, et al. Hybrid Integration Approach of Entropy with Logistic Regression and Support Vector Machine for Landslide Susceptibility Modeling[J]. Entropy, 2018, 20(11):884. DOI: 10.3390/e20110884.
[25]HUGHESG. Youden’s Index and the Weight of Evidence[J]. Methods of Information in Medicine, 2015, 54(2):198-199.
[26]沈 涛,苏山舞.基于DEM的河流多尺度显示研究[J].测绘科学,2003,28(2):20-22.
基金
国家重点研发计划项目 (2017YFC0504700)
PDF(4386 KB)
