首页 | 本学科首页   官方微博 | 高级检索  
     检索      

黄河口海域悬浮物浓度Landsat8 OLI分段线性反演
引用本文:刘振宇,崔廷伟,张胜花,赵文静.黄河口海域悬浮物浓度Landsat8 OLI分段线性反演[J].光谱学与光谱分析,2018,38(8):2536-2541.
作者姓名:刘振宇  崔廷伟  张胜花  赵文静
作者单位:1. 中南民族大学资源与环境学院,湖北 武汉 430074
2. 国家海洋局第一海洋研究所,山东 青岛 266061
3. 中南民族大学人地关系研究中心,湖北 武汉 430074
4. 环保部华南环境科学研究所,广东 广州 510655
基金项目:国家自然科学基金项目(41506202),湖北省自然科学基金项目(2015CFC874),广东省自然科学基金项目(2014A030310287),中韩海洋科学共同研究中心项目(PI-2017-3)资助
摘    要:黄河每年输送大量泥沙进入渤海。研究黄河口海域悬浮物浓度,对于黄河输沙以及周边海域的环境监测具有重要意义。利用2011年夏、冬两季实测遥感反射率以及同步测量悬浮物浓度数据,开展了黄河口海域悬浮物浓度分段线性反演研究。结果表明,不同浓度范围下,悬浮物浓度反演的敏感波段不同;浓度小于等于50 mg·L-1(≤50 mg·L-1),敏感比值波段为(600~700 nm)/(400~600 nm),浓度高于50 mg·L-1(>50 mg·L-1),敏感比值波段为(750~900 nm)/(420~720 nm),Landsat8 OLI的对应组合方式分别为B4/B2和B5/B3;根据上述浓度分段范围分别建立线性模型,其精度R2,RMSE和APD分别为0.873 5,4.08 mg·L-1和22.81%(≤50 mg·L-1),以及0.969 3,102.96 mg·L-1和17.51%(>50 mg·L-1),整个浓度下三个精度参数分别为0.975 3,67.03 mg·L-1和20.45%,均优于常用单一模型在分段和整体浓度下的相应参数,且具有良好的稳定性。分段线性模型,更适合浓度变化大的黄河口海域悬浮物浓度反演。

关 键 词:黄河口  悬浮物浓度  Landsat8  OLI  分段线性  
收稿时间:2017-08-25

Piecewise Linear Retrieval Suspended Particulate Matter for the Yellow River Estuary Based on Landsat8 OLI
LIU Zhen-yu,CUI Ting-wei,ZHANG Sheng-hua,ZHAO Wen-jing.Piecewise Linear Retrieval Suspended Particulate Matter for the Yellow River Estuary Based on Landsat8 OLI[J].Spectroscopy and Spectral Analysis,2018,38(8):2536-2541.
Authors:LIU Zhen-yu  CUI Ting-wei  ZHANG Sheng-hua  ZHAO Wen-jing
Institution:1. College of Resource and Environment Science, South-Central University for Nationalities, Wuhan 430074, China 2. The First Institute of Oceangraphy, State Oceanic Administration, Qingdao 266061, China 3. Research Center of Human-Environment Relations, South-Central University for Nationalities, Wuhan 430074, China 4. South China Institute of Environmental Sciences. MEP, Guanzhou 510655, China
Abstract:Much sediment is transported into Bohai by Yellow River every year. Therefore the study on suspended matter concentration(SPM) in Yellow River estuary is significant to the environmental monitoring and sediment transport of surrounding ocean. The piecewise linear retrieval model was proposed based on the remote sensing reflectance and suspended matter concentration synchronous sampled in summer and winter in 2011. The results showed: the sensitive bands combination for suspended matter concentration inversion was different under different concentration ranges, (600~700 nm) /(400~600 nm) and (750~900 nm)/(420~720 nm) respectively to the concentration range below and above 50 mg·L-1, and the corresponding bands for Landsat 8 OLI were B4/B2 and B5/B3 respectively. The R2, RMSE and APD of the Piecewise model were 0.873 5,4.08 mg·L-1 and 22.81%(SPM≤50 mg·L-1),0.969 3, 102.96 mg·L-1, and 17.51%(SPM>50 mg·L-1) respectively, and 0.975 3,67.03 mg·L-1and 20.45% for the overall concentration, which were better than parameters of common single model under entire concentration range. In summary, the piecewise linear retrieval model is more suitable for suspended matter concentration inversion of the Yellow River estuary with large variation of concentration.
Keywords:Yellow River Estuary  Suspended particulate matter  Landsat8 OLI  Piecewise linear retrieval model  
本文献已被 CNKI 等数据库收录!
点击此处可从《光谱学与光谱分析》浏览原始摘要信息
点击此处可从《光谱学与光谱分析》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号