Journal of Tropical Oceanography ›› 2009, Vol. 28 ›› Issue (5): 35-42.doi: 10.11978/j.issn.1009-5470.2009.05.035cstr: 32234.14.j.issn.1009-5470.2009.05.035

• Marine physics • Previous Articles     Next Articles

An evaluation of two semi-analytical ocean color algorithms for waters off South China

WANG Wen-qi1, DONG Qiang1, SHANG Shao-ling1, WU Jing-yu1, LEE Zhong-ping2   

  1. 1. Joint Key Laboratory of Coastal Study (Xiamen University,Fujian Institute of Oceanography), Xiamen 361005, China; 2. Northern Gulf Institute, Mississippi State University, MS 39529
  • Received:2009-01-12 Revised:2009-04-07 Online:2009-11-10 Published:2009-10-10
  • Contact: 商少凌 E-mail:slshang@xmu.edu.cn
  • Supported by:

    863计划项目(2006AA09A302,2008AA09Z108);国家自然科学基金项目(40521003)

Abstract:

With 42 in situ measurements of remote sensing reflectance and component absorption coefficients taken in the South China Sea and coastal waters off Fujian (China) in different seasons, the performance of the Quasi-Analytical Algorithm (QAA) and the Garver-Siegel-Maritorena (GSM) algorithm for water’s absorption coefficients are evaluated. It is found that the retrieval performance of the two algorithms is similar as those by other researchers performed in different regions. To waters in this study, QAA performed better in the South China Sea than for waters off Fujian coast. For total absorption coefficient at 443 nm (a443), the root-mean-square-error (RMSE) is 0.046 in the South China Sea, with an averaged percentage error of 7.9%, and averaged error in log scale (δ) close to 0. For coastal waters off Fujian, RMSE, an averaged percentage error and δ are 0.194, 30.6%, and -0.167, respectively. The performance of GSM is similar for the two regions of waters. For a443, RMSE and an averaged percentage error are 0.161 and 27.7% in the South China Sea, respectively; and are 0.149 and 32.1% for waters off Fujian coast. Their δ values, however, are negative (-0.142, indicating underestimation) in the South China Sea, and positive (0.016, indicating slight overestimation) in coastal waters off Fujian. Further analysis indicate that the differences between the empirical parameters employed in the algorithms and actual values of the waters under study are the main reasons causing errors in remote sensing retrievals, and it is necessary to regionally refine those parameters in order to improve the algorithm performances.wei

Key words: semi-analytical algorithm, absorption coefficient, South China Sea, coastal Fujian

CLC Number: 

  • TP79