3 November 2010 Precision improvement of chlorophyll-a remote sensing inversion by data transformation in turbidity water under low concentration: a case of Taihu Lake, China
Author Affiliations +
Abstract
Estimation and monitoring Chlorophyll-a concentration (CHLA), especially low CHLA in lake using remote sensing data is very important for early warning of blue-green algal bloom. In spite of better overall goodness fit in three-band CHLA inversion model of turbidity water proposed by Gitelson, the estimation errors of samples with low CHLA are often higher, and this kind of error has great influence on the evaluation of lake nutritional status. In this paper, two methods of data transformation-logarithm of CHLA and continuum removal of spectrum-were used to decrease model error. Data set includes the routine monitoring sampling data collected from June to September, 2004 in Taihu Lake and field data in March, 2010 in Meiliangwan of Taihu Lake. Water surface spectrum data were measured in situ by ASD FieldPro. Comparative analysis showed that both logarithm transformation (LT) and continuum removal transformation (CRT) can increase model's accuracy. For all sample data, the average relative accuracy of model built by data after LT increased by 30%, and that of model built by data after LT and CRT increased by 35%. For the samples with CHLA lower than 50μg/L, the average relative error decreased from 76% of model built by data without transformation to 36% of LT and 27% of LT and CRT. The paper concluded that data transform is a simple and effective method to increase precision of CHLA remote sensing inversion.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yuchun Wei, Yuchun Wei, Chunmei Cheng, Chunmei Cheng, Lei Wang, Lei Wang, Jing Zhang, Jing Zhang, } "Precision improvement of chlorophyll-a remote sensing inversion by data transformation in turbidity water under low concentration: a case of Taihu Lake, China", Proc. SPIE 7858, Remote Sensing of the Coastal Ocean, Land, and Atmosphere Environment, 785805 (3 November 2010); doi: 10.1117/12.869421; https://doi.org/10.1117/12.869421
PROCEEDINGS
9 PAGES


SHARE
Back to Top