Enhancing Streamflow Forecasting in Glacierized Basins: A Hybrid Model Integrating Glacio-Hydrological Outputs, Deep Learning, and Wavelet Transformation DOI Creative Commons
Jamal Hassan Ougahi, John S. Rowan

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 15, 2024

Abstract Understanding snow and ice melt dynamics is vital for flood risk assessment effective water resource management in highly populated river basins rising inaccessible high-mountains. This study evaluated AI-enhanced hydrological modelling using a hybrid approach integrating glacio-hydrological model (GSM-SOCONT), with advanced machine learning deep techniques framed as alternative ‘scenarios’, leveraging both physical processes data-driven insights enhanced predictive capabilities. The standalone (CNN-LSTM), relying solely on meteorological data, outperformed the model. Additionally, series of models (CNN-LSTM1 to CNN-LSTM15) were trained data along three additional feature groups derived from outputs, providing detailed into streamflow simulation. (CNN-LSTM14), which relied glacier-derived features, demonstrated best performance high NSE (0.86), KGE (0.80), R (0.93) values during calibration, highest (0.83), (0.88), (0.91), lowest RMSE (892) MAE (544) validation. Furthermore, proposed hybridization framework involves applying permutation importance identify key wavelet transform decompose them multi-scale analysis, these (CNN-LSTM19), significantly enhances accuracy, particularly high-flow events, evidenced by improved (from 0.83 0.97) reduced 892 442) comparative analysis illustrates how improve accuracy runoff forecasting provide more reliable actionable managing resources mitigating risks - despite relative paucity direct measurements.

Language: Английский

Single-Objective and Multi-Objective Flood Interval Forecasting Considering Interval Fitting Coefficients DOI
Xinyu Chang, Jun Guo,

Hui Qin

et al.

Water Resources Management, Journal Year: 2024, Volume and Issue: 38(10), P. 3953 - 3972

Published: April 19, 2024

Language: Английский

Citations

52
Enhanced streamflow forecasting using hybrid modelling integrating glacio-hydrological outputs, deep learning and wavelet transformation DOI Creative Commons
Jamal Hassan Ougahi, John S. Rowan

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 22, 2025

Abstract Understanding snow and ice melt dynamics is vital for flood risk assessment effective water resource management in populated river basins sourced inaccessible high-mountains. This study provides an AI-enabled hybrid approach integrating glacio-hydrological model outputs (GSM-SOCONT), with different machine learning deep techniques framed as alternative ‘computational scenarios, leveraging both physical processes data-driven insights enhanced predictive capabilities. The standalone (CNN-LSTM), relying solely on meteorological data, outperformed its counterpart equivalents. Hybrid models (CNN-LSTM1 to CNN-LSTM15) were trained using data augmented representing snow-melt contributions streamflow. (CNN-LSTM14), only glacier-derived features, performed best high NSE (0.86), KGE (0.80), R (0.93) values during calibration, the highest (0.83), (0.88), (0.91), lowest RMSE (892) MAE (544) validation. Finally, a multi-scale analysis feature permutations was explored wavelet transformation theory, these into final (CNN-LSTM19), which significantly enhances accuracy, particularly high-flow events, evidenced by improved (from 0.83 0.97) reduced 892 442) comparative illustrates how AI-enhanced hydrological improve accuracy of runoff forecasting provide more reliable actionable managing resources mitigating risks - despite paucity direct measurements.

Language: Английский

Citations

3
Comparison of Machine Learning Models in Simulating Glacier Mass Balance: Insights from Maritime and Continental Glaciers in High Mountain Asia DOI Creative Commons
Weiwei Ren, Zhongzheng Zhu,

Yingzheng Wang

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(6), P. 956 - 956

Published: March 8, 2024

Accurately simulating glacier mass balance (GMB) data is crucial for assessing the impacts of climate change on dynamics. Since physical models often face challenges in comprehensively accounting factors influencing glacial melt and uncertainties inputs, machine learning (ML) offers a viable alternative due to its robust flexibility nonlinear fitting capability. However, effectiveness ML modeling GMB across diverse types within High Mountain Asia has not yet been thoroughly explored. This study addresses this research gap by evaluating used simulation annual glacier-wide data, with specific focus comparing maritime glaciers Niyang River basin continental Manas basin. For purpose, meteorological predictive derived from monthly ERA5-Land datasets, topographical obtained Randolph Glacier Inventory, along target rooted geodetic observations, were employed drive four selective models: random forest model, gradient boosting decision tree (GBDT) deep neural network ordinary least-square linear regression model. The results highlighted that generally exhibit superior performance compared ones. Moreover, among models, GBDT model was found consistently coefficient determination (R2) values 0.72 0.67 root mean squared error (RMSE) 0.21 m w.e. 0.30 river basins, respectively. Furthermore, reveals climatic differentially influence simulations glaciers, providing key insights into dynamics response change. In summary, ML, particularly demonstrates significant potential simulation. application can enhance accuracy modeling, promising approach assess

Language: Английский

Citations

8
Soil and Water Assessment Tool (SWAT)-Informed Deep Learning for Streamflow Forecasting with Remote Sensing and In Situ Precipitation and Discharge Observations DOI Creative Commons
Chunlin Huang, Ying Zhang,

Jinliang Hou

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(21), P. 3999 - 3999

Published: Oct. 28, 2024

In order to anticipate residual errors and improve accuracy while reducing uncertainties, this work integrates the long short-term memory (LSTM) with Soil Water Assessment Tool (SWAT) create a deep learning (DL) model that is guided by physics. By forecasting of SWAT model, SWAT-informed LSTM (LSTM-SWAT) differs from typical approaches predict streamflow directly. Through numerical tests, performance LSTM-SWAT was evaluated both LSTM-only SWAT-only models in Upper Heihe River Basin. The outcomes showed performed better than other models, showing higher lower mean absolute error (MAE = 3.13 m3/s). Sensitivity experiments further how quality training dataset affects LSTM-SWAT. results study demonstrate may prediction greatly remote sensing situ observations. Additionally, emphasizes need for detailed consideration specific sources uncertainty predictive capabilities hybrid model.

Language: Английский

Citations

5
Spatial and temporal runoff variability in response to climate change in alpine mountains DOI
Bing He, Jianxia Chang, Aijun Guo

et al.

Journal of Hydrology, Journal Year: 2025, Volume and Issue: unknown, P. 132779 - 132779

Published: Feb. 1, 2025

Language: Английский

Citations

0
Probabilistic daily runoff forecasting in high-altitude cold regions using a hybrid model combining DBO and transformer variants DOI
Qiying Yu, Wenzhong Li, Yungang Bai

et al.

Journal of Hydrology Regional Studies, Journal Year: 2025, Volume and Issue: 59, P. 102311 - 102311

Published: March 17, 2025

Language: Английский

Citations

0
Spatially non-stationarity relationships between high-density built environment and waterlogging disaster: Insights from xiamen island, china DOI Creative Commons
Qianwen Wang, Runze Zhao, Ning Wang

et al.

Ecological Indicators, Journal Year: 2024, Volume and Issue: 162, P. 112021 - 112021

Published: April 18, 2024

A profound comprehension of the relationship between built environment and waterlogging disasters in high-density urban areas is essential to improve resilience achieve safe sustainable living environments. Based on multi-source data, this study employs a spatial autocorrelation model, geodetector multi-scale geographically weighted regression model construct comprehensive method. This has led formulation more effective integration method for investigating non-stationary relationships disasters. The results show that: (1) distribution events displays strong dependence. (2) Floor area ratio, building structure index, slope, water surface rate, etc., exhibit limited independent explanatory capability concerning density events. However, their nonlinear augmentation effect becomes noteworthy upon amalgamation with other factors, especially combination characteristics number rain-sewage mixing nodes. (3) It reliable efficient reduce risk by optimizing rainwater sewage diversion system reducing nodes instead modifying underlying controlling morphological impervious ground. (4) Where demonstrates intricate diverse attributes, emergence heightened sensitivity changes environment. provides novel perspective understanding formation mechanisms heterogeneity

Language: Английский

Citations

3
Predicting Gross Primary Productivity under Future Climate Change for the Tibetan Plateau Based on Convolutional Neural Networks DOI Creative Commons
Meimei Li, Zhongzheng Zhu, Weiwei Ren

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(19), P. 3723 - 3723

Published: Oct. 7, 2024

Gross primary productivity (GPP) is vital for ecosystems and the global carbon cycle, serving as a sensitive indicator of ecosystems’ responses to climate change. However, impact future changes on GPP in Tibetan Plateau, an ecologically important climatically region, remains underexplored. This study aimed develop data-driven approach predict seasonal annual variations Plateau up year 2100 under changing climatic conditions. A convolutional neural network (CNN) was employed investigate relationships between various environmental factors, including variables, CO2 concentrations, terrain attributes. analyzed projected from Coupled Model Intercomparison Project Phase 6 (CMIP6) four scenarios: SSP1–2.6, SSP2–4.5, SSP3–7.0, SSP5–8.5. The results suggest that expected significantly increase throughout 21st century all scenarios. By 2100, reach 1011.98 Tg C, 1032.67 1044.35 1055.50 C scenarios, representing 0.36%, 4.02%, 5.55%, 5.67% relative 2021. analysis indicates spring autumn shows more pronounced growth SSP3–7.0 SSP5–8.5 scenarios due extended growing season. Furthermore, identified elevation band 3000 4500 m particularly change terms response. Significant increases would occur east Qilian Mountains upper reaches Yellow Yangtze Rivers. These findings highlight pivotal role driving dynamics this region. insights not only bridge existing knowledge gaps regarding over coming decades but also provide valuable guidance formulation adaptation strategies at ecological conservation management.

Language: Английский

Citations

2
Glacier mass change and evolution of Petrov Lake in the Ak-Shyirak massif, central Tien Shan, from 1973 to 2023 using multisource satellite data DOI

Yingzheng Wang,

Donghai Zheng, Yushan Zhou

et al.

Remote Sensing of Environment, Journal Year: 2024, Volume and Issue: 315, P. 114437 - 114437

Published: Oct. 17, 2024

Language: Английский

Citations

2
Enhancing flood monitoring and prevention using machine learning and IoT integration DOI

Syed Asad Shabbir Bukhari,

Imran Shafi,

Jamil Ahmad

et al.

Natural Hazards, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 4, 2024

Language: Английский

Citations

1