Remote sensing of absorption and scattering coefficient using neural network model: Development, validation, and application

doi:10.1016/j.rse.2014.04.013

CORC > 自然资源部第一海洋研究所 > 自然资源部第一海洋研究所 > 业务部门 > 海洋物理与遥感研究室

	Remote sensing of absorption and scattering coefficient using neural network model: Development, validation, and application
	Chen, Jun 1,2; Quan, Wenting 3; Cui, Tingwei 4; Song, Qingjun 5; Lin, Changsong 1
刊名	REMOTE SENSING OF ENVIRONMENT
	2014-06
卷号	149 页码:213-226
关键词	Global oceanic and coastal waters Remote sensing Neural network Inherent optical property
ISSN号	0034-4257
DOI	10.1016/j.rse.2014.04.013
英文摘要	The total absorption (a(lambda)) and backscattering (b(b)(lambda)) coefficients of natural waters are the most significant factors affecting light propagation within water columns, and thus play indispensable roles in the estimation of aquatic biomass, primary production, and carbon pools. Despite its importance, no accurate retrieval model has been specifically developed for both oceanic and coastal waters, but significant efforts have been made in regard to oceanic inversion models. The objectives of the present study are to evaluate the applicability of the quasi-analytical algorithm (QAA) in deriving a(lambda) and b(b)(lambda) from oceanic and coastal waters, and to improve it using a neural network-based semi-analytical algorithm (NNSAA). Based on a comparison of the a(X) and bb(X) predicted by these models with field measurements taken from the national aeronautics and space administration bio-optical marine algorithm dataset (NOMAD), the Yellow Sea and China East Sea, it is shown that the NNSAA model (R-2 > 0.82 and mean relative error, MRE = 20.6-35.5%) provides a stronger performance than the QAA model (R-2 < 0.73 and MRE = 32.2-69.6%). The model was also applied to MODIS data after atmospheric correction using a near-infrared-based and shortwave infrared-based combined model. Through validation by field measurements, it was shown that the NNSAA model can predict a(lambda) and b(b)(lambda) with high accuracy (R-2 >0.77 and MRE < 39.9%). Finally, the NNSAA model was used to map the global climatological seasonal mean a(443) and b(b)(555) for the time range of July 2002 to September 2013. Except the coastal zones, it was shown that the a(443) and b(b)(532) in some high-latitude areas are much higher than in the mid- and low-latitude regions, due to the effects of spurious signals from neighboring sea-ice. In the equatorial oceans, the a(443) value in the surface water is considerably higher in the equatorial Pacific than in the equatorial Atlantic in the upwelling region, while the integrate a(443) is much higher in the Atlantic than in the throughout the entire tropical gyre areas. The difference between a(443) and b(b)(532) in the subsurface water is due to a pronounced deep biomass maximum existing in the equatorial Atlantic, which is associated with the higher nitrate in the lower euphotic zone. (C) 2014 Elsevier Inc. All rights reserved.
资助项目	Serial Maps of Geology and Geophysics on China Seas and Land on the Scale of 1:1000000[200311000001]
WOS研究方向	Environmental Sciences & Ecology ; Remote Sensing ; Imaging Science & Photographic Technology
语种	英语
出版者	ELSEVIER SCIENCE INC
WOS记录号	WOS:000337651300017
内容类型	期刊论文
源URL	[http://ir.fio.com.cn/handle/2SI8HI0U/3828]
专题	业务部门_海洋物理与遥感研究室
作者单位	1.China Univ Geosci, Sch Ocean Sci, Beijing 100083, Peoples R China; 2.Qingdao Inst Marine Geol, Key Lab Marine Hydrocarbon Resources & Environm G, Qingdao 266071, Peoples R China; 3.Shaanxi Agr Remote Sensing Informat Ctr, Xian 71000, Peoples R China; 4.State Ocean Adm, Inst Oceanog 1, Qingdao 266071, Peoples R China; 5.Natl Satellite Ocean Applicat Serv, Beijing 100081, Peoples R China
推荐引用方式 GB/T 7714	Chen, Jun,Quan, Wenting,Cui, Tingwei,et al. Remote sensing of absorption and scattering coefficient using neural network model: Development, validation, and application[J]. REMOTE SENSING OF ENVIRONMENT,2014,149:213-226.
APA	Chen, Jun,Quan, Wenting,Cui, Tingwei,Song, Qingjun,&Lin, Changsong.(2014).Remote sensing of absorption and scattering coefficient using neural network model: Development, validation, and application.REMOTE SENSING OF ENVIRONMENT,149,213-226.
MLA	Chen, Jun,et al."Remote sensing of absorption and scattering coefficient using neural network model: Development, validation, and application".REMOTE SENSING OF ENVIRONMENT 149(2014):213-226.