A Hybrid Approach for the Prediction of Chlorophyll-a Concentration at the Non-monitoring Area in the Geum River, Korea

Recently, harmful algae bloom in the river or lake become serious water quality issues as it can affect public health. In this paper, we introduce a hybrid model to predict Chlorophyll-a concentration at the non-monitoring spot by combining the existing simulation- and deep learning-based prediction methods. Our long short-term memory (LSTM) based deep learning algorithm models the correlation among the water quality data and provides the estimated water quality data. Then, the environmental fluid dynamics code (EFDC) simulation performs a spatiotemporal simulation of the region of interest, with the generated water quality data. The experimental results show that our prediction method achieves reliable performance on the Chlorophyll-a prediction at non-monitoring spots with the estimated water quality data.