1. Title, Journal and Authors
Title : Development of early-warning protocol for predicting chlorophyll-a concentration using machine learning models in freshwater and estuarine reservoirs, Korea
Journal : science of the Total Environment 502 (2015) 31-41
Authors : Yongeun Parka, Kyung Hwa Chob, Jihwan Parka, Sung Min Chac, Joon Ha Kima,*
a School of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 261 Cheomdan-gwagiro, Buk-gu
b School of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Eonyang-eup, Ulju-gun, Ulsan 689-798, Republic of Kore
c Jellanam-do Environmental Industries Promotion Institute, 650-94songgye-ro, Seongjeon-myeon, Gangjin-gun, Jellanam-do, 527-811, Republic of Korea
This study provides a rational model for the early-warning prediction of Chl-a concentration using a comprehensive evaluation of two machine learning models, Artificial Neural Networks and Support Vector Machine, in a reservoir system. Two machine learning models were used to predict the Chl-a concentration using weekly water quality data and meteorological data over 7 years’ period. And SVM showed a higher prediction accuracy than ANN. The results of sensitivity analysis using LH-OAT method showed that the causal relation between Chl-a concentration environmental variables. Especially, PO4-P was the most sensitive input variable. Overall, SVM showed good performance for early-warning prediction of Chl-a for different sampling intervals. This study suggested that 7-day interval is a reasonable interval for early warning.
3. Originality & Creativity
This study showed goodness of SVM performance in Chl-a concentration prediction and the reasonable interval for early warning.
Sora Shin / Ph.D. program
Environmental Systems Engineering Lab.
School of Earth Sciences & Environmental Engineering
Gwangju Institute of Science and Technology
123 Cheomdangwagi-ro, Buk-gu Gwangju, 61005, Korea
Phone : +82-10-8796-0728
E-mail : email@example.com