Cited by A probabilistic machine learning framework for daily extreme events forecasting

DeepBase: A Deep Learning-based Daily Baseflow Dataset across the United States DOI

Parnian Ghaneei, Hamid Moradkhani

Scientific Data, Journal Year: 2025, Volume and Issue: 12(1)

Published: Jan. 7, 2025

High quality baseflow data is important for advancing water resources modeling and management, as it captures the critical role of groundwater delayed sources in contributing to streamflow. Baseflow main recharge source runoff during dry period, particularly understanding interaction between surface systems. This study focuses on estimating using deep learning algorithms that enhance estimation capabilities both gauged ungauged basins. Recognizing shortage accessible high daily data, our objective generate a dataset across contiguous United States (CONUS) 1661 basins from 1981 2022. provides valuable information earth environmental scientists, resource managers, enhancing cycle. It also an foundation contributions extreme events such droughts floods. The can be used new benchmark future studies aimed at improving hydrological predictions managing more effectively.

Language: Английский

Citations

Predicting the Evolution of Extreme Water Levels With Long Short‐Term Memory Station‐Based Approximated Models and Transfer Learning Techniques DOI

Samuel Daramola, David F. Muñoz, Paúl Muñoz

et al.

Water Resources Research, Journal Year: 2025, Volume and Issue: 61(3)

Published: March 1, 2025

Abstract Extreme water levels (EWLs) resulting from cyclones pose significant flood hazards and risks to coastal communities interconnected ecosystems. To date, physically based models have enabled accurate prediction of EWLs despite their inherent high computational cost. However, the applicability these is limited data‐rich sites with diverse characteristics. The dependence on quality spatiotemporal data, which often computationally expensive, hinders regions either or data‐scarce conditions. address this challenge, we present a Long Short‐Term Memory (LSTM) network framework predict evolution beyond site‐specific training stations. framework, named LSTM‐Station Approximated Models (LSTM‐SAM), consists collection bidirectional LSTM enhanced custom attention mechanism layer embedded in architecture. LSTM‐SAM incorporates transfer learning approach applicable target (tide‐gage) stations along U.S. Atlantic Coast. Importantly, helps analyze: (a) underlying limitations associated learning, (b) evaluate EWL predictions domains, (c) capture caused by tropical extratropical cyclones. demonstrates satisfactory performance “transferable” achieving Kling‐Gupta Efficiency (KGE), Nash‐Sutcliffe (NSE), Root‐Mean Square Error (RMSE) ranging 0.78 0.92, 0.90 0.97, 0.09–0.18 m at stations, respectively. We show that can accurately not only but also over time, is, onset, peak, dissipation, could assist operational forecasting resources set up models.

Language: Английский

Citations

Evidential uncertainty quantification with multiple deep learning architectures for spatiotemporal drought forecasting DOI

Ahlem Ferchichi, Mejda Chihaoui,

Radhia Toujani

et al.

Neural Computing and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 15, 2025

Language: Английский

Citations

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

A Cluster-based Temporal Attention Approach for Predicting Cyclone-induced Compound Flood Dynamics DOI

Samuel Daramola, David F. Muñoz, Hamed Moftakhari

et al.

Environmental Modelling & Software, Journal Year: 2025, Volume and Issue: unknown, P. 106499 - 106499

Published: April 1, 2025

Language: Английский

Citations

Case Studies on Generative Adversarial Networks in Environmental Changes DOI

K. Udayakumar,

M. Revathi,

K. Janani

et al.

Advances in geospatial technologies book series, Journal Year: 2025, Volume and Issue: unknown, P. 199 - 226

Published: April 30, 2025

Environmental changes have been rooted in both natural and man-made. It affect greatly the ecosystems of Earth, human health, economic systems, societal structures. Understanding these impacts is therefore vital formulation policies that could promote sustainability diminish adverse effects. Deep generative models categories currently attracting a lot attention include Generative Adversarial Networks (GANs). These networks can model complex distributions are based on variety data. GANs become very powerful tool most branches research areas, especially with much stress environmental research. Their capabilities giving trustworthy data development robust dataset presented, which proven useful monitoring responding to environment. Applications observing air quality, tracking process deforestation, climate change modeling, water resource management, others. GAN applications pave way for obtaining more accurate detailed information conditions. This will make efforts minimize combat stronger. Based studies, possible use mitigating various fields discussed further. A couple case studies reflect extensive and, therefore, great potential addressing earth's surface. Steadily progressing computational resources technologies predict an expansion important role within scope science. open new solutions points insight into severe problems sphere.

Language: Английский

Citations

Enhancing Streamflow Prediction in Ungauged Basins Using a Nonlinear Knowledge‐Based Framework and Deep Learning DOI

Parnian Ghaneei, Ehsan Foroumandi, Hamid Moradkhani

et al.

Water Resources Research, Journal Year: 2024, Volume and Issue: 60(11)

Published: Oct. 30, 2024

Abstract In hydrology, a fundamental task involves enhancing the predictive power of model in ungagged basins by transferring information on physical attributes and hydroclimate dynamics from gauged basins. Introducing an integrated nonlinear clustering framework, this study aims to develop comprehensive framework that augments performance where direct measurements are sparse or absent. uniform manifold approximation projection (UMAP) is used as method extract essential features embedded hydro‐climatological properties. Then, Growing Neural Gas (GNG) find potentially share similar behaviors. Besides UMAP‐GNG, integration Principal Component Analysis (PCA) linear reduce dimensionality with common methods also assessed serve benchmarks. The results reveal combination algorithms PCA may lead loss while can more informative features. efficacy proposed across Contiguous United States (CONUS) training single Base Model using long short‐term memory (LSTM) for centroids all clusters then, fine‐tuning each cluster separately create regional model. indicate extracted UMAP‐GNG guide significantly improve accuracy most enhance median prediction within different 0.04 0.37 KGE ungauged

Language: Английский

Citations

A probabilistic machine learning framework for daily extreme events forecasting DOI

A. Sattari,

Ehsan Foroumandi, Keyhan Gavahi

et al.

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: unknown, P. 126004 - 126004

Published: Dec. 1, 2024

Language: Английский

Citations