Gaussian process machine learning and Kriging for groundwater salinity interpolation DOI Creative Commons
Tao Cui, Dan Pagendam, Mat Gilfedder

et al.

Environmental Modelling & Software, Journal Year: 2021, Volume and Issue: 144, P. 105170 - 105170

Published: Aug. 22, 2021

Gaussian processes (GPs) provide statistically optimal predictions in the sense of unbiasedness and maximal precision. Although modern implementation GPs as a machine learning technique is more capable flexible than Kriging, their employment environmental science less routine. Their flexibility capability spatial data interpolation are demonstrated by applying them to groundwater salinity prediction data-sparse region Australia. By from multiple sources, including AEM DEM data, have generated maps with rich local details quantified uncertainty support risk-based decision making. The results demonstrate great worth nonpoint regional coverage realistic heterogeneity aquifer properties that critical for many studies such contaminant transport. should be further encouraged prediction, especially when point measurements sparse predictors available.

Language: Английский

The role of deep learning in urban water management: A critical review DOI Creative Commons
Guangtao Fu, Yiwen Jin, Siao Sun

et al.

Water Research, Journal Year: 2022, Volume and Issue: 223, P. 118973 - 118973

Published: Aug. 11, 2022

Deep learning techniques and algorithms are emerging as a disruptive technology with the potential to transform global economies, environments societies. They have been applied planning management problems of urban water systems in general, however, there is lack systematic review current state deep applications an examination directions where can contribute solving challenges. Here we provide such review, covering demand forecasting, leakage contamination detection, sewer defect assessment, wastewater system prediction, asset monitoring flooding. We find that application still at early stage most studies used benchmark networks, synthetic data, laboratory or pilot test performance methods no practical adoption reported. Leakage detection perhaps forefront receiving implementation into day-to-day operation systems, compared other reviewed. Five research challenges, i.e., data privacy, algorithmic development, explainability trustworthiness, multi-agent digital twins, identified key areas advance management. Future expected drive towards high intelligence autonomy. hope this will inspire development harness power help achieve sustainable digitalise sector across world.

Language: Английский

Citations

213
Deep learning for water quality DOI
Wei Zhi, Alison P. Appling, Heather E. Golden

et al.

Nature Water, Journal Year: 2024, Volume and Issue: 2(3), P. 228 - 241

Published: March 12, 2024

Language: Английский

Citations

64
Prioritizing environmental determinants of urban heat islands: A machine learning study for major cities in China DOI Creative Commons
Haoran Hou, Qianqiu Longyang, Hongbo Su

et al.

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2023, Volume and Issue: 122, P. 103411 - 103411

Published: July 6, 2023

The exacerbated thermal environment in cities, with the urban heat island (UHI) effect as a prominent example, has been source of many adverse environmental issues, including increase health risks, degradation air quality and ecosystem services, reduced resiliency engineering infrastructure. Last decades have witnessed tremendous efforts resources being invested to find sustainable solutions for mitigation, whereas relative contributions different UHI attributes their patterns spatio-temporal variability remain obscure. In this study, we employed random forest (RF) method quantify importance four categories surface characteristics that regulate UHI, namely greenery fraction, land albedo, morphology, level human activities. We selected seventeen major cities from six megaregions China our study areas, RF training test sets obtained multi-sourced remote sensing observational data products. It is found coverage manifests most important determinants followed by albedo. results are informative planners, policymakers, practitioners design implement strategies mitigation.

Language: Английский

Citations

48
Conceptualizing future groundwater models through a ternary framework of multisource data, human expertise, and machine intelligence DOI
Chuanjun Zhan, Zhenxue Dai, Shangxian Yin

et al.

Water Research, Journal Year: 2024, Volume and Issue: 257, P. 121679 - 121679

Published: April 26, 2024

Language: Английский

Citations

34
Incorporating multiple grid-based data in CNN-LSTM hybrid model for daily runoff prediction in the source region of the Yellow River Basin DOI Creative Commons
F.X. Hu, Qinli Yang, J. Yang

et al.

Journal of Hydrology Regional Studies, Journal Year: 2024, Volume and Issue: 51, P. 101652 - 101652

Published: Jan. 9, 2024

The Source Region of the Yellow River Basin (SRYRB), China To improve daily runoff prediction accuracy in data-scarce areas, this study focuses on incorporating multiple grid-based data (precipitation, EVI, soil moisture (SM)) to drive CNN-LSTM hybrid model. spatial features precipitation and underlying surface basin can be extracted by CNN, while temporal input series captured LSTM. model is compared with single models (CNN, LSTM), performances under different driven are also investigated. Driven in-situ precipitation, (GPM) SM data, achieved best result NSE 0.834, outperforming LSTM (NSE=0.510) CNN (NSE=0.612). It indicates that captures spatiotemporal change basin. When using only GPM as input, comparable 0.827. implies could serve a good alternative provide additional value prediction. This highlights model, which provides new insights into regions.

Language: Английский

Citations

26
A critical review of machine learning algorithms in maritime, offshore, and oil & gas corrosion research: A comprehensive analysis of ANN and RF models DOI
Md Mahadi Hasan Imran, Shahrizan Jamaludin, Ahmad Faisal Mohamad Ayob

et al.

Ocean Engineering, Journal Year: 2024, Volume and Issue: 295, P. 116796 - 116796

Published: Jan. 30, 2024

Language: Английский

Citations

21
Distributed Hydrological Modeling With Physics‐Encoded Deep Learning: A General Framework and Its Application in the Amazon DOI Creative Commons
Chao Wang, Shijie Jiang, Yi Zheng

et al.

Water Resources Research, Journal Year: 2024, Volume and Issue: 60(4)

Published: April 1, 2024

Abstract While deep learning (DL) models exhibit superior simulation accuracy over traditional distributed hydrological (DHMs), their main limitations lie in opacity and the absence of underlying physical mechanisms. The pursuit synergies between DL DHMs is an engaging research domain, yet a definitive roadmap remains elusive. In this study, novel framework that seamlessly integrates process‐based model encoded as neural network (NN), additional NN for mapping spatially physically meaningful parameters from watershed attributes, NN‐based replacement representing inadequately understood processes developed. Multi‐source observations are used training data, fully differentiable, enabling fast parameter tuning by backpropagation. A hybrid Amazon Basin (∼6 × 10 6 km 2 ) was established based on framework, HydroPy, global‐scale DHM, its backbone. Trained simultaneously with streamflow Gravity Recovery Climate Experiment satellite yielded median Nash‐Sutcliffe efficiencies 0.83 0.77 dynamic simulations total water storage, respectively, 41% 35% higher than those original HydroPy model. Replacing Penman‒Monteith formulation produces more plausible potential evapotranspiration (PET) estimates, unravels spatial pattern PET giant basin. parameterization interpreted to identify factors controlling variability key parameters. Overall, study lays out feasible technical modeling big data era.

Language: Английский

Citations

19
Use of one-dimensional CNN for input data size reduction in LSTM for improved computational efficiency and accuracy in hourly rainfall-runoff modeling DOI
Kei Ishida, Ali Ercan, Takeyoshi Nagasato

et al.

Journal of Environmental Management, Journal Year: 2024, Volume and Issue: 359, P. 120931 - 120931

Published: April 27, 2024

Language: Английский

Citations

17
Enhancing Water Level Prediction Using Ensemble Machine Learning Models: A Comparative Analysis DOI
Saleh Alsulamy, Vijendra Kumar, Özgür Kişi

et al.

Water Resources Management, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 11, 2025

Language: Английский

Citations

3
A novel ensemble-based conceptual-data-driven approach for improved streamflow simulations DOI Creative Commons
Anna E. Sikorska‐Senoner, John Quilty

Environmental Modelling & Software, Journal Year: 2021, Volume and Issue: 143, P. 105094 - 105094

Published: June 2, 2021

A novel ensemble-based conceptual-data-driven approach (CDDA) is developed where a data-driven model (DDM) used to "correct" the residuals from an ensemble of hydrological (HM) simulations. The CDDA respects processes via HM and it benefits DDM's ability simulate complex relationship between input variables. can accomodate any DDM, allowing for different configurations be tested. tested streamflow simulation in three Swiss catchments HM, HBV (Hydrologiska Byråns Vattenbalansavdelning), coupled with eight DDMs: Multiple Linear Regression, k Nearest Neighbours Second-Order Volterra Series Model, Artificial Neural Networks, two variants eXtreme Gradient Boosting (XGB) Random Forests (RF). proposed was able improve mean continuous ranked probability score by 16–29% over standalone HM. Since XGB RF demonstrated best performance, they are recommended simulating residuals.

Language: Английский

Citations

66