Affiliation:
1. Nakdong River Support Team, Presidential Water Commission, Changwon-si 51439, Republic of Korea
2. Department of Civil & Environmental Engineering, Myongji University, Yongin 17058, Republic of Korea
Abstract
Several changes have occurred in the river environment of Nakdong river due to the construction of multifunctional weirs as part of the Four Major Rivers Project. This river currently exhibits characteristics that are similar to those of a stagnant water area in which the river depth increases and the flow velocity decreases. Consequently, blue-green algae are frequently observed. Toxic substances secreted by blue-green algae are harmful to aquatic ecosystems and the human body; therefore, ensuring the stability of the water quality of Nakdong river is of utmost importance. Various factors are associated with the occurrence of blue-green algae. Therefore, the causal relationship between these causative factors must be identified. In this study, we investigated factors influencing algal growth, such as water quality, hydraulics, and weather, and algal occurrence patterns by site were analyzed. Recent studies have used data mining and machine-learning techniques in algal management to quantitatively identify the characteristics of blue-green algae. In machine learning, the prediction results differ depending on the selection of parameters, which are an important aspect in the management of blue-green algae with complex causes. In this study, we quantitatively analyzed the conditions for the occurrence of cyanobacteria according to the influencing factors using decision trees and random forests, which are machine-learning techniques, along with an analysis of the major complex factors influencing the occurrence of blue-green algae in the Nakdong river weirs. Considering the water quality and hydraulic factors, we analyzed the characteristics of algal generation in each weir at different hydraulic volume times. In addition, we investigated the possibility of improving the accuracy of cyanobacterial prediction according to the learning factors. Through these analyses, we attempted to study the characteristics of blue-green algae in stagnant rivers.
Funder
National Research Foundation of Korea
Subject
Fluid Flow and Transfer Processes,Computer Science Applications,Process Chemistry and Technology,General Engineering,Instrumentation,General Materials Science
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献