Inland harmful algal blooms (HABs) modeling using internet of things (IoT) system and deep learning

Harmful algal blooms (HABs) have been frequently occurred with releasing toxic substances, which typically lead to water quality degradation and health problems for humans aquatic animals. Hence, accurate quantitative analysis prediction of HABs should be implemented detect, monitor, manage severe blooms. However, the traditional monitoring required sufficient expense labor while numerical models were restricted in terms their ability simulate algae dynamic. To address challenging issue, this study evaluates applicability deep learning chlorophyll-a (Chl-a) phycocyanin (PC) internet things (IoT) system. Our research adopted LSTM simulating Chl-a PC. Among models, attention model achieved superior performance by showing 0.84 2.35 (μg/L) correlation coefficient root mean square error. preprocessing methods, z-score method was selected as optimal improve performance. The mechanism highlighted input data from July October, indicating that period most influential output. Therefore, demonstrated IoT system has potential detect quantify cyanobacteria, can eutrophication management schemes freshwater reservoirs.