Prediction of aquatic vegetation growth under ecological recharge based on machine learning and remote sensing.
In: Journal of Cleaner Production, Jg. 452 (2024-05-01), S. N.PAG
academicJournal
Zugriff:
Ecological water recharge could be an important approach to alleviating water scarcity and improving water quality deterioration in water-scarce areas. However, the water quality status of the water source for ecological recharge was a key factor that may have an impact on the recharging water body. This study applied remote sensing and machine learning (ML) algorithms to predict the growth of the Huangtai algae in the Ulansuhai Lake under high nitrogen ecological recharge and the factors that influence the growth of the Huangtai algae. The results indicated that the ecological recharge could help to control the bloom of Huangtai algae and avoid the swamping of the water body caused by the massive growth of emergent vegetation. The machine learning model predicted the changes in Huangtai algae in the case of ecological recharge (R2 train = 0.9999, RMSE train = 0.0004, MAE train = 0.0003). Meanwhile, it was demonstrated that the water depth of the receiving lake was an important factor for Huangtai algae (Value of importance = 0.37). Apart from that nitrogen, water quantity, and phosphorus of ecological recharge were key factors (0.25, 0.21, and 0.09). This study provides preliminary evidence that ecological water could be a source of recharge for water-scarce areas, and that machine learning techniques could be applied in the future to make high-precision predictions of rivers, providing further insight into the drivers and facilitating water resource management. • Ecological recharge facilitates the controlled outbreak of Huangtai algae in lakes. • Predicting variation in the area of Huangtai algae using machine learning based on ecological recharge. • Water depth may be an important factor influencing the growth of Huangtai algae. [ABSTRACT FROM AUTHOR]
Titel: |
Prediction of aquatic vegetation growth under ecological recharge based on machine learning and remote sensing.
|
---|---|
Autor/in / Beteiligte Person: | Du, Caili ; Cui, Jianglong ; Wang, Dianpeng ; Li, Guowen ; Lu, Haoran ; Tian, Zhenjun ; Zhao, Chen ; Li, Maotong ; Zhang, Lieyu |
Zeitschrift: | Journal of Cleaner Production, Jg. 452 (2024-05-01), S. N.PAG |
Veröffentlichung: | 2024 |
Medientyp: | academicJournal |
ISSN: | 0959-6526 (print) |
DOI: | 10.1016/j.jclepro.2024.142054 |
Schlagwort: |
|
Sonstiges: |
|