Urban Water Absorbance to Predict Chlorophyll a and Turbidity
- https://doi.org/10.2991/edep-18.2018.26How to use a DOI?
- Chlorophyll a, LS-SVM, Spectral absorbance, SVM, Turbidity
This study aimed to use water absorbance in the range of 200-900 nm to predict the turbidity and chlorophyll a concentration (Chl-a) in urban water. Six kinds of water samples were artificially prepared in this study: spirulina samples (S), chlorella vulgaris samples (C), turbidity samples (T), mixed water samples of S-T, C-T, and S-C. The correlation analysis results showed that the turbidity had a strong correlations with the absorbance at most of the wavelengths, and linear models were built to predict turbidity for each type of water samples, the Rv2 of each specific model was higher than 0.981, and the overall Rv2 reached to 0.955. For Chl-a prediction, SVM method had better accuracies (Rv2>0.986) for the algae water samples than those (Rv2<0.826) for the turbidity mixed samples. In order to improve Chl-a prediction accuracy for turbidity mixed samples, LS-SVM method was used to estimate Chl-a (Rv2>0.987), which was increased by 31.1% comparing to the corresponding SVM method. Furthermore, the Rv2 of the overall Chl-a decoupled model for both S-T and C-T also reached to 0.915. The results showed that the absorbance of unprocessed water samples had great potential to predict Chl-a fast and accurately.
- © 2018, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Ya-Nan CAO AU - Xiu-Hua LI AU - Shao-Dui MA AU - Jiao-Yan AI AU - Hong-Xiang ZHU PY - 2018/10 DA - 2018/10 TI - Urban Water Absorbance to Predict Chlorophyll a and Turbidity BT - Proceedings of the 2018 International Conference on Energy Development and Environmental Protection (EDEP 2018) PB - Atlantis Press SP - 165 EP - 172 SN - 2352-5401 UR - https://doi.org/10.2991/edep-18.2018.26 DO - https://doi.org/10.2991/edep-18.2018.26 ID - CAO2018/10 ER -