New paradigm for watershed model development by coupling machine learning algorithm and mechanistic model

•New paradigm was used for development of watershed model.•Machine learning algorithm and mechanistic model are effectively coupled.•Equations for best management practices evaluation were established.•The paradigm shows higher interpretability based on mechanistic model. The development of watershe...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of hydrology (Amsterdam) Ročník 636; s. 131264
Hlavní autoři: Liu, Guowangchen, Chen, Lei, Wang, Wenzhuo, Wang, Shuai, Zhu, Kaihang, Shen, Zhenyao
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 01.06.2024
Témata:
algorithms
Best management practices (BMPs)
biomass
case studies
China
data collection
hydrologic models
lakes
Machine learning (ML)
mechanistic models
Multilayer perceptron regression (MLPR)
neural networks
Nonpoint source (NPS) pollution
pollution
regression analysis
soil types
vegetation
Watershed model
watersheds
China
Best management practices (BMPs)
Machine learning (ML)
Multilayer perceptron regression (MLPR)
Watershed model
Nonpoint source (NPS) pollution
ISSN:0022-1694, 1879-2707
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•New paradigm was used for development of watershed model.•Machine learning algorithm and mechanistic model are effectively coupled.•Equations for best management practices evaluation were established.•The paradigm shows higher interpretability based on mechanistic model. The development of watershed models faces double pressure of data requirements and physical interpretation. The simulation of nonpoint source (NPS) pollution is an important application of watershed model, and the best management practices (BMPs) have attracted wide attention as the main control approach of NPS pollution. In this study, a new paradigm was proposed based on the integration of data-driven and mechanistic methods, taking BMP evaluation as an example. Specifically, comprehensive databases were constructed for filter and retention BMPs by collecting, classifying and analyzing published data. Twelve machine learning algorithms were employed for regression analysis between BMPs efficiency and their influencing factors, while the constructed equations were coupled with physical-based models. A case study was performed in a typical catchment of Chaohu Lake Watershed, China. The results demonstrated total interception area, soil type, and vegetation biomass had significant impacts on BMPs performances, while the multilayer perceptron regression (MLPR), K nearest neighbor regression (KNRD), and extremely randomized tree (ETR) methods had the best performances of nutrients removal. The study generated over ten thousand datasets using mechanistic processes, resulting in more efficient and interpretable BMPs evaluator. Compare to traditional methods, this new paradigm has shown advantages in the model application at the watershed scale, enhancing the BMPs evaluator with a higher level of interpretability by coupling the approach with mechanistic models.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0022-1694
1879-2707
DOI:10.1016/j.jhydrol.2024.131264