Journal of Environmental Biology

 DOI : http://doi.org/10.22438jeb/40/3(SI)/Sp-11

Measuring surface water salinity of Pearl River Estuary by MODIS 250-m imageries

Paper received: 17. 04. 2018?????????????????????????????????????????????????????? Revised received: 14. 11.2018?????????????????????? ?????????????????????????????????????????????????????Accepted: 03.01. 2019

Authors Info

L. Fang^1,2,3*, Z. Liu¹, S. Chen² and W. Huang^3,4     

¹Jiangsu Province Support Software Engineering R&D Center for Modern Information Technology Application in Enterprise, Suzhou, 215104, China

²Guangdong Open Laboratory of Geospatial Information Technology and Application, Guangzhou Institute of Geography, Guangzhou, 510070, China

³Department of Civil Engineering, Florida State University, Tallahassee, FL 32310, USA

⁴Key Laboratory of Yangtze River Water Environment, Ministry of Education, Tongji University, Shanghai, 200092, China

*Corresponding Author Email :

fanglg@jssvc.edu.cn

Abstract

Aim: Moderate resolution imaging spectroradiometer (MODIS) 250-m imageries were used to evaluate the performance for developing the effective model of salinity using field spectra, the absorption coefficient of color dissolved organic matter (a_CDOM) collected on October 21and November 2, 2012 in the Pearl River Estuary (PRE), China.    

Methodology: Remote sensing reflectance of the four simulated MODIS bands (R_rs(Bi)) with a_CDOM (l) data were used for algorithms development, which include least squares regression of the single band, difference between bands, band ratios, sediment index and their varieties with a_CDOM (l) based on the dataset of November 2, 2012 (N=18). Additionally, the model of salinity with a_CDOM (l) data was also calibrated (N=18). ?

Results: Results illustrated the optimum performance of quadratic model of sediment index for estimation of a_CDOM (355) (R²= 0.67, N=18, P<0.001), and the reverse linear model of a_CDOM (355) is of the best accuracy for estimation of salinity (R²=0.83, N=18, P<0.001). An effective method to estimate surface water salinity from simulated MODIS 250-m bands was calibrated and validated using the independent dataset of October 21, 2012 (RMSE = 1.95?, MRE = 9.65%, N = 17) in the Lingding Bay of PRE. The preferred models were further applied to retrieve a_CDOM (355) and salinity data from the MODIS 250-m data in the coastal and inner Lingding Bay of PRE. The result presented rationality of salinity distribution of PRE. The MODIS 250-m salinity mapping was validated based on in-situ measurements, and presented a good salinity mapping accuracy (RMSE=1.72?, MRE=8.24%, N=10). The study proved the robustness of these algorithms for a_CDOM (355) and salinity estimation in the coastal and inner Lingding Bay of PRE by MODIS remote sensing. ?

Interpretation: Therefore, the effective method can be applicable in detecting variability of CDOM concentration and salinity during tidal cycle in a high frequency (two times during daytime), which proves the detection ability of MODIS 250-m imageries in estuarine and coastal waters. It provides the water supply and conservancy authorities spatially and temporally understand the marine intrusion in the Lingding Bay of PRE.

Key words: Color dissolved organic matter, MODIS 250-m, Pearl River Estuary, Salinity

Copyright ? 2019 Triveni Enterprises. All rights reserved. No part of the Journal can be reproduced in any form without prior permission. Responsibility regarding the authenticity of the data, and the acceptability of the conclusions enforced or derived, rest completely with the author(s).