| A machine learning approach for early warning of cyanobacterial bloom outbreaks in a freshwater reservoir |
|---|
|
학술지명 Journal of Environmental Management
저자 신재기,김성환,김진휘,박용은,백상수,이한규,전강민,조경화
발표일 2021-03-16
|
|
Understanding the dynamics of harmful algal blooms is important to protect the aquatic ecosystem in regulatedrivers and secure human health. In this study, artificial neural network (ANN) and support vector machine (SVM)models were used to predict algae alert levels for the early warning of blooms in a freshwater reservoir. Intensivewater-quality, hydrodynamic, and meteorological data were used to train and validate both ANN and SVMmodels. The Latin-hypercube one-factor-at-a-time (LH-OAT) method and a pattern search algorithm were appliedto perform sensitivity analyses for the input variables and to optimize the parameters of the models, respectively.The results indicated that the two models well reproduced the algae alert level based on the time-lag input andoutput data. In particular, the ANN model showed a better performance than the SVM model, displaying a higherperformance value in both training and validation steps. Furthermore, a sampling frequency of 6- and 7-day weredetermined as efficient early-warning intervals for the freshwater reservoir. Therefore, this study presents aneffective early-warning prediction method for algae alert level, which can improve the eutrophication managementschemes for freshwater reservoirs. |