Comparison of three recurrent neural networks for rainfall-runoff modelling at a snow-dominated watershed

Yokoo, Kazuki; Ishida, Kei; Nagasato, Takeyoshi; Ercan, Ali; Tu, Tongbi

doi:10.1088/1755-1315/851/1/012012

Cited by 2 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…for the Kowmung River at Cedar Ford in Australia;Luu et al (2021aLuu et al ( , 2021b for the Thu Bon-Vu Gia catchment, Vietnam;Yokoo et al (2021Yokoo et al ( , 2022 for the Ishikari River watershed in Japan;Li et al (2021e) for the middle Yangtze River, China;Hashemi et al (2021) for 740 gauges in France;Kwak et al (2021) for the Hyeongsan River basin, South Korea;Zakhrouf et al (2021) for two stations in Algeria;Rahimzad et al (2021) for a station in Kentucky, US; Chen and Xu for the Dadu River Basin, China;Chen and Qiao (2021) for the Nanjing River, China;Kim and Kim (2021) for the Yeongsan River basin, South Korea;Wang et al (2021b) for the Tunxi Basin, China utilized LSTMs.In one of the ensemble model studies,Mirzaei et al (2021) coupled LSTMs with Principal Component Analysis (PCA) for the Samarahan and Trusan river basins, Malaysia. Using PCA as well,Lian et al (2021) utilized Bayesian optimization (BO) and LSTMs for the Yellow River, China.…”

mentioning

confidence: 99%

A Systematic Review of Deep Learning Applications in Streamflow Data Augmentation and Forecasting

Sit¹,

Demiray²,

Demir³

2022

Preprint

View full text Add to dashboard Cite

The volume and variety of Earth data have increased as a result of growing attention to climate change and, subsequently, the availability of large-scale sensor networks and remote sensing instruments. This data has been an important resource for data-driven studies to generate practical knowledge and services, support environmental modeling and forecasting needs, and transform climate and earth science research thanks to the increased availability of computational resources and the popularity of novel computational techniques like deep learning. Timely and accurate simulation and modeling of extreme events are critical for planning and mitigation in hydrology and water resources. There is a strong need for short-term and long-term forecasts of streamflow, benefiting from recent developments in data availability and novel deep learning methods. In this study, we review the literature for studies that employ deep learning in tackling tasks that are either to improve the quality of the streamflow data or to forecast streamflow. The study aims to serve as a starting point by covering the latest developments of deep learning approaches in those topics as well as highlighting problems, limitations, and open questions with insights for future directions.

show abstract

mentioning

confidence: 99%

A Systematic Review of Deep Learning Applications in Streamflow Data Augmentation and Forecasting

Sit¹,

Demiray²,

Demir³

2022

Preprint

View full text Add to dashboard Cite

show abstract

Enhancing Rainfall-Runoff Simulation via Meteorological Variables and a Deep-Conceptual Learning-Based Framework

et al. 2022

View full text Add to dashboard Cite

Accurate streamflow simulation is crucial for many applications, such as optimal reservoir operation and irrigation. Conceptual techniques employ physical ideas and are suitable for representing the physics of the hydrologic model, but they might fail in competition with their more advanced counterparts. In contrast, deep learning (DL) approaches provide a great computational capability for streamflow simulation, but they rely on data characteristics and the physics of the issue cannot be fully understood. To overcome these limitations, the current study provided a novel framework based on a combination of conceptual and DL techniques for enhancing the accuracy of streamflow simulation in a snow-covered basin. In this regard, the current study simulated daily streamflow in the Kalixälven river basin in northern Sweden by integrating a snow-based conceptual hydrological model (MISD) with a DL model. Daily precipitation, air temperature (average, minimum, and maximum), dew point temperature, evapotranspiration, relative humidity, sunshine duration, global solar radiation, and atmospheric pressure data were used as inputs for the DL model to examine the effect of each meteorological variable on the streamflow simulation. Results proved that adding meteorological variables to the conceptual hydrological model underframe of parallel settings can improve the accuracy of streamflow simulating by the DL model. The MISD model simulated streamflow had an MAE = 8.33 (cms), r = 0.88, and NSE = 0.77 for the validation phase. The proposed deep-conceptual learning-based framework also performed better than the standalone MISD model; the DL method had an MAE = 7.89 (cms), r = 0.90, and NSE = 0.80 for the validation phase when meteorological variables and MISD results were combined as inputs for the DL model. The integrated rainfall-runoff model proposed in this research is a new concept in rainfall-runoff modeling which can be used for accurate streamflow simulations.

show abstract

Comparison of three recurrent neural networks for rainfall-runoff modelling at a snow-dominated watershed

Cited by 2 publications

References 13 publications

A Systematic Review of Deep Learning Applications in Streamflow Data Augmentation and Forecasting

A Systematic Review of Deep Learning Applications in Streamflow Data Augmentation and Forecasting

Enhancing Rainfall-Runoff Simulation via Meteorological Variables and a Deep-Conceptual Learning-Based Framework

Contact Info

Product

Resources

About