Dataset extent

Map data © OpenStreetMap contributors

Deep recurrent neural networks for daily streamflow prediction

This dataset contains daily streamflow and weather data and the python code that was used to developed deep learning models for streamflow prediction. The data and code can be used to replicate the experiment on Google Drive using Google Colaboratory.

Streamflow information is crucial for effectively managing water resources. The declining number of active gauging stations in many rivers is a global concern, necessitating the need for reliable streamflow estimates. Deep learning (DL) techniques offer potential solutions, but their application in southern Africa remains largely underexplored. To fill this gap, this study evaluated the predictive performance of Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) networks using two headwater catchments of the Steelpoort River, South Africa, as case studies. The model inputs included rainfall, maximum and minimum temperature, as well as past streamflow, which was utilized in an autoregressive sense. The inclusion of streamflow in this way allowed for the incorporation of simulated streamflow values into the look-back window (LBW) for predicting the streamflow of the testing set. Two modifications were required to the GRU and LSTM architectures to ensure physically consistent predictions. Including a change in activation function of the GRU/LSTM cells in the final hidden layer, and a non-negative constraint that was used in the dense layer. Models trained using commercial weather station data produced reliable streamflow estimates, while moderately accurate predictions were obtained using freely available gridded weather data.

Data and Resources

Additional Info

Field Value
Author 1
Author first name
Author surname
Author organization
University of Pretoria
Plant and Soil Sciences
Is this author a contact person for the dataset?
Contact person
Contact 1
Contact name
Contact organization
Contact 2
Contact name
Contact organization
Recommended citation Schutte, C., Van der Laan, M., & Van der Merwe, B. (2024). Leveraging historic streamflow and weather data with deep learning for enhanced streamflow predictions. Journal of Hydroinformatics.
Did the author / contact organization collect the data? true
Name of organization that collected the data
Dataset language
Publisher Water Research Commission
Publication date 2024-04-30
Project number C2020/2021-00440
License Open (Creative commons)
License URL
Keywords GRU, LSTM, Rainfall-runoff modelling
Geographic location or bounding box coordinates [-22.1265, 16.4699, -34.8212, 32.8931]
Topic category Hydrological data and modelling
Data structure category Structured (clearly labelled and in a standardised format)
Uploader estimation of extent to which data have been processed Access
Is the data time series or static Time series
Data reference date
Data reference date 1
Data reference date (from)
Data reference date (to)
Alternate identifier
Vertical extent datum 1800
Vertical minimum-maximum extent
Vertical minimum-maximum extent 1
Minimum vertical extent
Maximum vertical extent
I agree to the data management plan and terms and conditions of the WRO true