Development of a Regional Gridded Runoff Dataset Using Long Short-Term Memory (LSTM) Networks

Ayzel, Georgy; Kurochkina, L. Ya.; Abramov, Dmitriy; Zhuravlev, Sergei

doi:10.3390/hydrology8010006

Cited by 18 publications

(9 citation statements)

References 81 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the framework of the R5 project, extensive datasets of gridded runoff products have been developed for the studied region (for methodology, see [27,29]). These products include:…”

Section: Runoff Datamentioning

confidence: 99%

Machine Learning Reveals a Significant Shift in Water Regime Types Due to Projected Climate Change

Ayzel

2021

IJGI

Self Cite

View full text Add to dashboard Cite

A water regime type is a cumulative representation of seasonal runoff variability in a textual, qualitative, or quantitative form developed for a particular period. The assessment of the respective water regime type changes is of high importance for local communities and water management authorities, increasing their awareness and opening strategies for adaptation. In the presented study, we trained a machine learning model—the Random Forest classifier—to predict water regime types in northwest Russia based on monthly climatological hydrographs derived for a historical period (1979–1991). Evaluation results show the high efficiency of the trained model with an accuracy of 91.6%. Then, the Random Forest model was used to predict water regime types based on runoff projections for the end of the 21st century (2087–2099) forced by four different General Circulation Models (GCM) and three Representative Concentration Pathway scenarios (RCP). Results indicate that climate is expected to modify water regime types remarkably. There are two primary directions of projected changes. First, we detect the tendency towards less stable summer and winter flows. The second direction is towards a shift in spring flood characteristics. While spring flooding is expected to remain the dominant phase of the water regime, the flood peak is expected to shift towards earlier occurrence and lower magnitude. We identified that the projected changes in water regime types are more pronounced in more aggressive RCP scenarios.

show abstract

“…In the framework of the R5 project, extensive datasets of gridded runoff products have been developed for the studied region (for methodology, see [27,29]). These products include:…”

Section: Runoff Datamentioning

confidence: 99%

Machine Learning Reveals a Significant Shift in Water Regime Types Due to Projected Climate Change

Ayzel

2021

IJGI

Self Cite

View full text Add to dashboard Cite

show abstract

“…Forecast is established on the probability of the river flow and its strong historical data or records. As a result of highly efficient decision-making capability, both the short-term and longterm (e.g., hourly, and daily) forecasting models have a significant interest into the research and scientist community [12][13][14][15]. Due to the complexity, climate changeability and the effects on anthropology, hydrological data holds a strong constraint to the advancement of short-term forecasting models [16][17][18][19].…”

Section: Introductionmentioning

confidence: 99%

Exploring Temporal Dynamics of River Discharge using Univariate Long Short-Term Memory (LSTM) Recurrent Neural Network at East Branch of Delaware River

Mehedi¹,

Khosravi²,

Yazdan³

et al. 2022

Preprint

View full text Add to dashboard Cite

River flow prediction is a pivotal task in the field of water resource management during the era of rapid climate change. The highly dynamic and evolving nature of the climatic variables e.g., precipitation has a significant impact on the temporal distribution of the river discharge in recent days making the discharge forecasting even more complicated for diversified water-related issues e.g., flood prediction and irrigation planning. To predict the discharge, various physics-based numerical models are used using numerous hydrologic parameters. Extensive lab-based investigation and calibration are required to reduce the uncertainty involved in those parameters. However, in the age of data-driven predictions, several deep learning algorithms showed satisfactory performance in dealing with sequential data. In this research, Long Short-term Memory (LSTM) neural network regression model is trained using over 80 years of daily data to forecast the discharge time series up to 3 days ahead of time. The performance of the model is found satisfactory through the comparison of the predicted data with the observed data, visualization of the distribution of the errors and Root Mean Squared Error (RMSE) value of 0.09. Higher performance is achieved through the increase in the number of epochs and hyper parameter tuning. This model can be transferred to other locations with proper feature engineering and optimization to perform univariate predictive analysis and potentially be used to perform real-time river discharge prediction.

show abstract

“…Postprocessor approaches typically use forms of regression with measured data, including frequency or distribution matching, multivariate statistical analysis, and machine learning regression, to map biased model outputs to bias-corrected values [29][30][31][32]. Machine learning-centered approaches for bias correction postprocessors and related analysis are particularly common in recent publications [33][34][35][36].…”

Section: Introductionmentioning

confidence: 99%

SABER: A Model-Agnostic Postprocessor for Bias Correcting Discharge from Large Hydrologic Models

et al. 2022

View full text Add to dashboard Cite

Hydrologic modeling is trending toward larger spatial and temporal domains, higher resolutions, and less extensive local calibration and validation. Thorough calibration and validation are difficult because the quantity of observations needed for such scales do not exist or is inaccessible to modelers. We present the Stream Analysis for Bias Estimation and Reduction (SABER) method for bias correction targeting large models. SABER is intended for model consumers to apply to a subset of a larger domain at gauged and ungauged locations and address issues with data size and availability. SABER extends frequency-matching postprocessing techniques using flow duration curves (FDC) at gauged subbasins to be applied at ungauged subbasins using clustering and spatial analysis. SABER uses a “scalar” FDC (SFDC), a ratio of simulated to observed FDC, to characterize biases spatially, temporally, and for varying exceedance probabilities to make corrections at ungauged subbasins. Biased flows at ungauged locations are corrected with the scalar values from the SFDC. Corrected flows are refined to fit a Gumbel Type 1 distribution. We present the theory, procedure, and validation study in Colombia. SABER reduces biases and improves composite metrics, including Nash Sutcliffe and Kling Gupta Efficiency. Recommendations for future work and a discussion of limitations are provided.

show abstract

Development of a Regional Gridded Runoff Dataset Using Long Short-Term Memory (LSTM) Networks

Cited by 18 publications

References 81 publications

Machine Learning Reveals a Significant Shift in Water Regime Types Due to Projected Climate Change

Machine Learning Reveals a Significant Shift in Water Regime Types Due to Projected Climate Change

Exploring Temporal Dynamics of River Discharge using Univariate Long Short-Term Memory (LSTM) Recurrent Neural Network at East Branch of Delaware River

SABER: A Model-Agnostic Postprocessor for Bias Correcting Discharge from Large Hydrologic Models

Contact Info

Product

Resources

About