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), Год журнала: 2024, Номер unknown

Опубликована: Ноя. 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.

Язык: Английский

Application of HP-LSTM Models for Groundwater Level Prediction in Karst Regions: A Case Study in Qingzhen City DOI Open Access
Yan Bo, Chunlei Zhang, Xiaoyu Fang

и другие.

Water, Год журнала: 2025, Номер 17(3), С. 362 - 362

Опубликована: Янв. 27, 2025

Groundwater serves as an indispensable global resource, essential for agriculture, industry, and the urban water supply. Predicting groundwater level in karst regions presents notable challenges due to intricate geological structures fluctuating climatic conditions. This study examines Qingzhen City, China, introducing innovative hybrid model, Hodrick–Prescott (HP) filter–Long Short-Term Memory (LSTM) network (HP-LSTM), which integrates HP filter with LSTM enhance precision of forecasting. By attenuating short-term noise, HP-LSTM model improves long-term trend prediction accuracy. Findings reveal that significantly outperformed conventional LSTM, attaining R2 values 0.99, 0.96, 0.98 on training, validation, test datasets, respectively, contrast 0.92, 0.76, 0.95. The achieved RMSE 0.0276 a MAPE 2.92% set, outperforming (RMSE: 0.1149; MAPE: 9.14%) capturing patterns reducing fluctuations. While is effective at modeling dynamics, it more prone resulting greater errors. Overall, demonstrates superior robustness prediction, whereas may be better suited scenarios requiring rapid adaptation variations. Selecting appropriate tailored specific predictive needs can thus optimize management strategies.

Язык: Английский

Процитировано

1
Enhancing hydrological time series forecasting with a hybrid Bayesian-ConvLSTM model optimized by particle swarm optimization DOI Creative Commons
Hüseyin Çağan Kılınç,

Sina Apak,

Mahmut Esad Ergin

и другие.

Acta Geophysica, Год журнала: 2025, Номер unknown

Опубликована: Март 24, 2025

Язык: Английский

Процитировано

0
Incorporating hydrological constraints with deep learning for streamflow prediction DOI

Yi Zhou,

Yilin Duan, Hong Yao

и другие.

Expert Systems with Applications, Год журнала: 2024, Номер unknown, С. 125379 - 125379

Опубликована: Сен. 1, 2024

Язык: Английский

Процитировано

2
Review of Recent Advances in Remote Sensing and Machine Learning Methods for Lake Water Quality Management DOI Creative Commons
Ying Deng, Yue Zhang,

Daiwei Pan

и другие.

Remote Sensing, Год журнала: 2024, Номер 16(22), С. 4196 - 4196

Опубликована: Ноя. 11, 2024

This review examines the integration of remote sensing technologies and machine learning models for efficient monitoring management lake water quality. It critically evaluates performance various satellite platforms, including Landsat, Sentinel-2, MODIS, RapidEye, Hyperion, in assessing key quality parameters chlorophyll-a (Chl-a), turbidity, colored dissolved organic matter (CDOM). highlights specific advantages each platform, considering factors like spatial temporal resolution, spectral coverage, suitability these platforms different sizes characteristics. In addition to this paper explores application a wide range models, from traditional linear tree-based methods more advanced deep techniques convolutional neural networks (CNNs), recurrent (RNNs), generative adversarial (GANs). These are analyzed their ability handle complexities inherent data, high dimensionality, non-linear relationships, multispectral hyperspectral data. also discusses effectiveness predicting parameters, offering insights into most appropriate model–satellite combinations scenarios. Moreover, identifies challenges associated with data quality, model interpretability, integrating imagery models. emphasizes need advancements fusion techniques, improved generalizability, developing robust frameworks multi-source concludes by targeted recommendations future research, highlighting potential interdisciplinary collaborations enhance sustainable management.

Язык: Английский

Процитировано

1
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), Год журнала: 2024, Номер unknown

Опубликована: Ноя. 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.

Язык: Английский

Процитировано

0