Long-term precipitation prediction in different climate divisions of California using remotely sensed data and machine learning DOI

Shabnam Majnooni,

Mohammad Reza Nikoo, Banafsheh Nematollahi

et al.

Hydrological Sciences Journal, Journal Year: 2023, Volume and Issue: 68(14), P. 1984 - 2008

Published: Aug. 21, 2023

ABSTRACTThis study presented a novel paradigm for forecasting 12-step-ahead monthly precipitation at 126 California gauge stations. First, the satellite-based time series from Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS), TerraClimate, ECMWF Reanalysis V5 (ERA5), and Estimation Remotely Sensed Information using Artificial Neural Networks-Climate Data Record (PERSIANN-CDR) products were bias-corrected historical data. Four methods tested, quantile mapping (QM) was best. After pre-processing data, 19 machine-learning models developed. random forest, Extreme Gradient Boosting (XGBoost), extreme gradient boosting, support vector machine, multi-layer perceptron, K-nearest-neighbours chosen as best based on Complex Proportional Assessment (COPRAS) measurement. hyperparameter adjustment, Bayesian back-propagation regularization algorithm fused results. The superior models' predictions considered inputs, target's initial step labeled. next 11 steps each station followed this approach, fusion accurately predicted all steps. 12th step's average Nash-Sutcliffe efficiency (NSE), mean square error (MSE), coefficient of determination (R2), correlation (R) 0.937, 52.136, 0.880, 0.869, respectively, demonstrating framework's effectiveness high horizons to help policymakers manage water resources.KEYWORDS: bias correctionhyperparameterslong-term predictionmachine learning (ML)quantile (QM)satellite-based Editor A Castellarin; Associate F-J. ChangEditor ChangDisclosure statementNo potential conflict interest reported by authors.Supplementary materialSupplemental article can be accessed online https://doi.org/10.1080/02626667.2023.2248112.

Language: Английский

Developing a novel hybrid Auto Encoder Decoder Bidirectional Gated Recurrent Unit model enhanced with empirical wavelet transform and Boruta-Catboost to forecast significant wave height DOI
Masoud Karbasi, Mehdi Jamei, Mumtaz Ali

et al.

Journal of Cleaner Production, Journal Year: 2022, Volume and Issue: 379, P. 134820 - 134820

Published: Oct. 23, 2022

Language: Английский

Citations

27
Application of hybrid machine learning models and data pre-processing to predict water level of watersheds: Recent trends and future perspective DOI

Sarah Jasim Mohammed,

Salah L. Zubaidi, Sandra Ortega‐Martorell

et al.

Cogent Engineering, Journal Year: 2022, Volume and Issue: 9(1)

Published: Nov. 11, 2022

The community's well-being and economic livelihoods are heavily influenced by the water level of watersheds. changes in levels directly affect circulation processes lakes rivers that control mixing bottom sediment resuspension, further affecting quality aquatic ecosystems. Thus, these considerations have made monitoring process essential to save environment. Machine learning hybrid models emerging robust tools successfully applied for monitoring. Various been developed, selecting optimal model would be a lengthy procedure. A timely, detailed, instructive overview models' concepts historical uses beneficial preventing researchers from overlooking potential selection saving significant time on problem. recent research prediction using machines is reviewed this article present "state art" subject provide some suggestions methodologies models. This comprehensive study classifies into four types algorithm parameter optimisation-based (OBH), pre-processing-based (PBH), components combination-based (CBH), hybridisation with preprocessing-based (HOPH); furthermore, it explains pre-processing data detail. Finally, most popular optimisation methods future perspectives conclusions discussed.

Language: Английский

Citations

22
Characterizing drought prediction with deep learning: A literature review DOI Creative Commons
Aldo Márquez-Grajales,

Ramiro Villegas-Vega,

Fernando Salas-Martínez

et al.

MethodsX, Journal Year: 2024, Volume and Issue: 13, P. 102800 - 102800

Published: June 13, 2024

Drought prediction is a complex phenomenon that impacts human activities and the environment. For this reason, predicting its behavior crucial to mitigating such effects. Deep learning techniques are emerging as powerful tool for task. The main goal of work review state-of-the-art characterizing deep used in drought results suggest most widely climate indexes were Standardized Precipitation Index (SPI) Evapotranspiration (SPEI). Regarding multispectral index, Normalized Difference Vegetation (NDVI) indicator utilized. On other hand, countries with higher production scientific knowledge area located Asia Oceania; meanwhile, America Africa regions few publications. Concerning methods, Long-Short Term Memory network (LSTM) algorithm implemented task, either canonically or together (hybrid methods). In conclusion, reveals need more about using indices Africa; therefore, it an opportunity characterize developing countries.

Language: Английский

Citations

5
A novel stabilized artificial neural network model enhanced by variational mode decomposing DOI Creative Commons
Ali Danandeh Mehr, Sadra Shadkani, Laith Abualigah

et al.

Heliyon, Journal Year: 2024, Volume and Issue: 10(13), P. e34142 - e34142

Published: July 1, 2024

Language: Английский

Citations

5
Review of Recent Trends in the Hybridisation of Preprocessing-Based and Parameter Optimisation-Based Hybrid Models to Forecast Univariate Streamflow DOI Open Access

Baydaa Abdul Kareem,

Salah L. Zubaidi, Nadhir Al‐Ansari

et al.

Computer Modeling in Engineering & Sciences, Journal Year: 2023, Volume and Issue: 138(1), P. 1 - 41

Published: Sept. 19, 2023

Forecasting river flow is crucial for optimal planning, management, and sustainability using freshwater resources. Many machine learning (ML) approaches have been enhanced to improve streamflow prediction. Hybrid techniques viewed as a viable method enhancing the accuracy of univariate estimation when compared standalone approaches. Current researchers also emphasised hybrid models forecast accuracy. Accordingly, this paper conducts an updated literature review applications in estimating over last five years, summarising data preprocessing, modelling strategy, advantages disadvantages ML techniques, models, performance metrics. This study focuses on two types models: parameter optimisation-based (OBH) hybridisation preprocessing-based (HOPH). Overall, research supports idea that meta-heuristic precisely techniques. It's one first efforts comprehensively examine efficiency various (classified into four primary classes) hybridised with revealed previous applied swarm, evolutionary, physics, metaheuristics 77%, 61%, 12%, respectively. Finally, there still room improving OBH HOPH by examining different pre-processing metaheuristic algorithms.

Language: Английский

Citations

11
Utilizing Hybrid Machine Learning Techniques and Gridded Precipitation Data for Advanced Discharge Simulation in Under-Monitored River Basins DOI Creative Commons
Reza Morovati, Özgür Kişi

Hydrology, Journal Year: 2024, Volume and Issue: 11(4), P. 48 - 48

Published: April 4, 2024

This study addresses the challenge of utilizing incomplete long-term discharge data when using gridded precipitation datasets and data-driven modeling in Iran’s Karkheh basin. The Multilayer Perceptron Neural Network (MLPNN), a rainfall-runoff (R-R) model, was applied, leveraging from Asian Precipitation—Highly Resolved Observational Data Integration Toward Evaluation (APHRODITE), Global Precipitation Climatology Center (GPCC), Climatic Research Unit (CRU). MLPNN trained Levenberg–Marquardt algorithm optimized with Non-dominated Sorting Genetic Algorithm-II (NSGA-II). Input were pre-processed through principal component analysis (PCA) singular value decomposition (SVD). explored two scenarios: Scenario 1 (S1) used situ for calibration dataset testing, while 2 (S2) involved separate calibrations tests each dataset. findings reveal that APHRODITE outperformed S1, all showing improved results S2. best achieved hybrid applications S2-PCA-NSGA-II S2-SVD-NSGA-II GPCC CRU. concludes datasets, properly calibrated, significantly enhance runoff simulation accuracy, highlighting importance bias correction modeling. It is important to emphasize this approach may not be suitable situations where catchment undergoing significant changes, whether due development interventions or impacts anthropogenic climate change. limitation highlights need dynamic approaches can adapt changing conditions.

Language: Английский

Citations

4
A post-processing machine learning framework for bias-correcting National Water Model outputs by accounting for dominant streamflow drivers DOI

Savalan Naser Neisary,

Ryan Johnson, Muddasser Alam

et al.

Environmental Modelling & Software, Journal Year: 2025, Volume and Issue: unknown, P. 106459 - 106459

Published: April 1, 2025

Language: Английский

Citations

0
A physically-informed long short-term memory-based tool for predicting extensive droughts in the distant future DOI Creative Commons
A Ghaffari, Shrouq Abuismail, Yi‐Chen E. Yang

et al.

Journal of Hydrology, Journal Year: 2025, Volume and Issue: unknown, P. 133316 - 133316

Published: April 1, 2025

Language: Английский

Citations

0
Adaptive ensemble weighting for GCMs to enhance future drought characterization under various climate change scenarios DOI
Muhammad Taimoor Shakeel,

H. Abbas,

Ayesha Waseem

et al.

Theoretical and Applied Climatology, Journal Year: 2025, Volume and Issue: 156(5)

Published: April 30, 2025

Language: Английский

Citations

0
Prediction of Hydrological Drought in Semi-arid Regions Using a Novel Hybrid Model DOI
Anas Mahmood Al-Juboori

Water Resources Management, Journal Year: 2023, Volume and Issue: 37(9), P. 3657 - 3669

Published: April 18, 2023

Language: Английский

Citations

9