School of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology, 50 UNIST-gil, Eonyang-eup, Ulju-gun, Ulsan 689-798, Republic of Korea.
Water Quality Assessment Research Division, National Institute of Environmental Research, Hwangyeong-ro 42, Seogu, Incheon 22689, Republic of Korea.
Water Res. 2020 Nov 1;186:116349. doi: 10.1016/j.watres.2020.116349. Epub 2020 Aug 26.
Machine learning modeling techniques have emerged as a potential means for predicting algal blooms. In this study, synthetic spatio-temporal water quality data for a river section were generated with a 3D water quality model and used to investigate the capability of a convolutional neural network (CNN) for predicting harmful cyanobacterial blooms. The CNN model displayed a reasonable capacity for short-term predictions of cyanobacteria (Microcystis) biomass. In the nowcasting of Microcystis, the CNN performance had a Nash-Sutcliffe Efficiency (NSE) of 0.87. An increase in the forecast lead time resulted in a decrease in the prediction accuracy, reducing the NSE from 0.87 to 0.58. As the spatial observation density increased from 20% to 100% of the input image grids, the CNN prediction NSE had improved from 0.70 to 0.84. Adding noise to the data resulted in accuracy deterioration, but even at the noise amplitude of 10%, the accuracy was acceptable for some applications, with NSE = 0.76. Visualization of the CNN results characterized its performance variations across the studied river reach. Overall, this study successfully demonstrated the capability of the CNN model for cyanobacterial bloom prediction using high temporal frequency images.
机器学习建模技术已成为预测藻类水华的一种潜在手段。本研究利用三维水质模型生成了河段的合成时空水质数据，并利用该数据调查了卷积神经网络（CNN）预测有害蓝藻水华的能力。CNN 模型对蓝藻（微囊藻）生物量的短期预测具有合理的能力。在 Microcystis 的临近预报中，CNN 的纳什效率（NSE）为 0.87。随着预测提前期的增加，预测精度降低，NSE 从 0.87 降低到 0.58。随着空间观测密度从输入图像网格的 20%增加到 100%，CNN 预测 NSE 从 0.70 提高到 0.84。向数据中添加噪声会导致精度下降，但即使在噪声幅度为 10%的情况下，对于某些应用，NSE=0.76 的精度也是可以接受的。CNN 结果的可视化描绘了其在研究河段的性能变化。总的来说，本研究成功地证明了 CNN 模型在使用高时间频率图像进行蓝藻水华预测方面的能力。
