Streamflow Forecasting
This is a public entry point for a suite of hydrologic ensemble forecasting modelling tools developed in Australia over the past decade. Most scientific modelling features were authored by the CSIRO. While designed with unique features for ensemble streamflow forecasting, these tools can equally be used in non-ensemble simulation mode for other, “traditional” semi-distributed hydrologic modelling.
Applications
Some of the modelling tools are used by the Australian Bureau of Meteorology (the Bureau) to provide the 7-day Ensemble Streamflow Forecasts service. An overview of the research and outcomes in streamflow forecasting of the research alliance between the Bureau and CSIRO is available at this page.
More recent versions of the tools are also used for further research, development and applications. The core modelling features are usually native (C++) libraries, accessed seamlessly by user from R, Python and Matlab.
The documentation page has links to sample modelling workflows in R and Python that hydrologic modellers can browse through to get an idea of the features.
Installation
Installation packages are available for Windows, Debian/Ubuntu Linux, and MacOSX. They currently cannot all be made publicly downloadable, but access for evaluation or research purposes can be arranged. Contact David Robertson at david.robertson@csiro.au and/or jean-michel.perraud@csiro.au. Then follow the instructions at the Installation page.
Architecture
The core of the toolset is written in C++ for execution speed, but in practice users mostly access the features via packages in R, python, or Matlab.
Besides building upon established third party open source software, some components of our stack are also open source. See csiro-hydroinformatics on GitHub, for instance the uchronia time series handling module.
The developer page also contains more detailed technical information on the software stack.
Acknowledgements
Most of the tools covered by this documentation is the output of research supported by the Water Information Research and Development Alliance between CSIRO and the Australian Bureau of Meteorology.
Publications
SWIFT2: Advanced software for continuous ensemble short-term streamflow forecasting