Machine learning-based estimation of riverine nutrient concentrations and associated uncertainties caused by sampling frequencies

Accurate and sufficient water quality data is essential for watershed management and sustainability. Machine learning models have shown great potentials for estimating water quality with the development of online sensors. However, accurate estimation is challenging because of uncertainties related t...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	PloS one Ročník 17; číslo 7; s. e0271458
Hlavní autoři:	Chen, Shengyue, Zhang, Zhenyu, Lin, Juanjuan, Huang, Jinliang
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	San Francisco Public Library of Science 13.07.2022 Public Library of Science (PLoS)
Témata:	Accuracy Algorithms Ammonia Analysis Artificial neural networks Back propagation networks Biology and Life Sciences Chemical oxygen demand Coastal waters Computer and Information Sciences Datasets Dissolved oxygen Earth Sciences Ecology and Environmental Sciences Electrical conductivity Electrical resistivity Environmental aspects Estimation Evaluation Hydrogen Hydrogen ion concentration Hydrogen ions Indicators Ion concentration Laboratories Learning algorithms Machine learning Management Modelling Monte Carlo simulation Neural networks Nitrogen Nutrient concentrations Nutrients Oxygen Phosphorus Physical Sciences Research and Analysis Methods Research methodology Sampling Security Sensors Support vector machines Sustainability Sustainable development Turbidity Uncertainty Water Water quality Water quality management Water security Water temperature Watershed management Watersheds China
ISSN:	1932-6203, 1932-6203
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Accurate and sufficient water quality data is essential for watershed management and sustainability. Machine learning models have shown great potentials for estimating water quality with the development of online sensors. However, accurate estimation is challenging because of uncertainties related to models used and data input. In this study, random forest (RF), support vector machine (SVM), and back-propagation neural network (BPNN) models are developed with three sampling frequency datasets (i.e., 4-hourly, daily, and weekly) and five conventional indicators (i.e., water temperature (WT), hydrogen ion concentration (pH), electrical conductivity (EC), dissolved oxygen (DO), and turbidity (TUR)) as surrogates to individually estimate riverine total phosphorus (TP), total nitrogen (TN), and ammonia nitrogen (NH 4 + -N) in a small-scale coastal watershed. The results show that the RF model outperforms the SVM and BPNN machine learning models in terms of estimative performance, which explains much of the variation in TP (79 ± 1.3%), TN (84 ± 0.9%), and NH 4 + -N (75 ± 1.3%), when using the 4-hourly sampling frequency dataset. The higher sampling frequency would help the RF obtain a significantly better performance for the three nutrient estimation measures (4-hourly > daily > weekly) for R 2 and NSE values. WT, EC, and TUR were the three key input indicators for nutrient estimations in RF. Our study highlights the importance of high-frequency data as input to machine learning model development. The RF model is shown to be viable for riverine nutrient estimation in small-scale watersheds of important local water security.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 Competing Interests: The authors have declared that no competing interests exist.
ISSN:	1932-6203 1932-6203
DOI:	10.1371/journal.pone.0271458