HydroSight: Open-source data-driven hydrogeological insights

HydroSight is statistical toolbox for data-driven insights into groundwater dynamics and aquifer properties. Many hundreds of bores can be easily analysed, all without any programming, to quantify:

• drivers of groundwater trends, e.g. climate and pumping (Shapoori et al., 2015a) and landuse change (Peterson and Western, 2014).
• recharge over time Peterson et al., 2019).
• aquifer hydraulic properties (Shapoori et al., 2015c, Peterson et al., 2019)
• statistical identification of the major groundwater processes (Shapoori et al., 2015b).
• interpolate or extrapolate hydrographs to a regular time step (Peterson & Western, 2018.
• simulate groundwater level under different climate or, say, pumping scenarios.
• hydrograph monitoring errors and outliers (Peterson et al., 2018).

Installation Options

HydroSight is operating system independent and has been tested on Windows 10+ and Linux (Ubuntu 20.04 LTS). There are four installation options:

1. Stand-alone app within Windows. The latest .exe is available here.
2. Install Hydrosight Matlab source code by (i) downloading the source code, (ii) unzipping the downloaded file, (ii) setting the Matlab Current Folder to where the file was unzipped and (iv) entering HydroSight into the Matlab Command Window.
3. Install Hydrosight from within Matlab using the Add-Ons menu item and searching for HydroSight. From the Add button select Add to Matlab. Once installed, enter HydroSight into the Matlab Command Window.
4. Compile your own stand-alone app from within Matlab by (i) downloading the source code and (ii) running the command: makeStandaloneHydroSight()

For futher details see the installation wiki page.

Examples

Multiple examples are built into the HydroSight GUI, each highlighting aspects of the above papers. Soon, each example will be supported by online videos. In the meantime major aspects of the graphical interface and the algorithms are outlined on the wiki page.

HydroSight can also be run from the Matlab command window. For an example of this see here.

What does HydroSight look like?

The HydroSight graphical interface includes tabs for each step in the modelling of groundwater hydrographs:

1. Project documentation.
2. Hydrograph outlier detection.
3. Time-series model construction, specifically defining the data and the form of the model.
4. Model calibration and tools to examine the internal dynamics of the calibrated model, e.g. recharge. The screenshot below shows this tab and an estimate of the annual groundwater recharge.
5. Model simulations, allowing hydrograph decomposition, exploration of scenarios (e.g. different climate or pumping), hindcasting and interpolation.

Contributing

HydroSight is an ongoing research project and its funding depends upon evidence of impact. To help us demonstrate impact please ensure you cite the relevant papers (using the "Cite Project" option within the GUI).

And, if HydroSight doesn't do what you need then please consider (i) writing and sharing your own module by extending existing class definitions, (ii) or getting in touch to discuss collaborations.