文献类型：期刊文献

英文题名：An approach for developing Landsat-5 TM-based retrieval models of suspended particulate matter concentration with the assistance of MODIS

作者：Wu, Guofeng[1,2] Cui, Lijuan[3] Duan, Hongtao[4] Fei, Teng[2] Liu, Yaolin[2]

第一作者：Wu, Guofeng

通信作者：Wu, GF[1]

机构：[1]Shenzhen Univ, Shenzhen Key Lab Spatial Informat Smart Sensing &, Coll Life Sci, Shenzhen 518060, Peoples R China;[2]Wuhan Univ, Sch Resource & Environm Sci, Key Lab Geog Informat Syst, Minist Educ, Wuhan 430079, Peoples R China;[3]Chinese Acad Forestry, Inst Wetland Res, Beijing 100091, Peoples R China;[4]Chinese Acad Sci, Nanjing Inst Geog & Limnol, State Key Lab Lake Sci & Environm, Nanjing 210008, Peoples R China

年份：2013

卷号：85

起止页码：84-92

外文期刊名：ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING

收录：;EI(收录号：20133916792579);Scopus(收录号：2-s2.0-84884525789);WOS:【SCI-EXPANDED(收录号:WOS:000326558200009)】；

基金：This study was supported by the National Natural Science Foundation of China (Grant Nos. 41171290, 40971191 and 41171271) and the Special Foundation of the Ministry of Finance of China for Nonprofit Research of Forestry Industry (Grant No. 200904001).

语种：英文

外文关键词：Water quality; Remote sensing; Model development; Exponential model

摘要：It is challenging to develop Landsat-5 TM (TM5) image-based retrieval models for estimating the suspended particulate matter concentration (C-SPM) in water when missing coincident ground C-SPM measurements. This study, with the Poyang Lake in China as a case study, proposed an approach for developing TM5-based C-SPM retrieval models with the assistance of moderate resolution imaging spectroradiometer (MODIS) images. After validation with an independent dataset, a cubic C-SPM retrieval model of 250 m MODIS red band was used to estimate the C-SPM values at 100 sampling points from the MODIS images (MODIS-based C-SPM) captured at three time periods. The MODIS-based C-SPM values at the time period with the largest C-SPM variation were combined with their coincident TM5 image reflectance for TM5-based model calibrations. The linear, quadratic, cubic, power and exponential models of MODIS-based C-SPM against TM5 single bands and their combinations were calibrated, respectively. Four best-fitting TM5-based C-SPM models were selected to retrieve the C-SPM values at 100 sampling points from the TM5 images (TM5-based C-SPM) at the other two time periods, and the coincident MODIS- and TM5-based C-SPM values were compared to assess TM5-based model performances. Model calibration results showed that the cubic and exponential models of TM5 red band (band 3) and red subtracting mid-infrared band (band 5) obtained the best fitting for estimating C-SPM from the TM5 image on 12 August 2005, and they explained 94-97% of the variation of MODIS-based C-SPM values with an estimated standard error of 6.617-8.457 mg/l. Model validations indicated that the exponential model of TM5 red band got the best result for estimating C-SPM from TM5 images when the MODIS-based C-SPM values were assumed as ground truths (correlation coefficient between MODIS- and TM5-based C-SPM values = 0.96, root mean square error = 4.60 mg/l). We concluded that the TM5-based C-SPM retrieval models could be developed with the assistance of MODIS, and the approach proposed in this study will be helpful for other researchers who also want to retrieve C-SPM from TM5 image archive but without coincident ground C-SPM measurements. (C) 2013 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS) Published by Elsevier B.V. All rights reserved.