Accelerating hydrological research by embracing FAIR

Accelerating hydrological research by embracing FAIR Bart Schilperoort, Peter Kalverla, Stefan Verhoeven, Mark Melotto, Rolf Hut, and many more A “simple” experiment Meteo Hydrological models How much water is in the river? Predict using meteorological data, i.e., precipitation ? IFS CMIP ERA5 A “complex” landscape The “eWaterCycle” dream Work with written in for with focusing on not any hydrological model any programming language any catchment any data source, hydrology computer science ewatercycle.org Step 1: input data •ESGF provides access to climate model data through a common interface •ESMValTool can access this data and provides it according to standard formats (names, units, coordinates…) •Used to generate figures for IPCC reports Standardized analysis of weather + climate data Step 2: the models •To run and interact with the models we use the Basic Model Interface •Standard functions to control and interact with a numerical model •With software containers we can reliably run models on any platform •We put container images on the GitHub Container Registry; linked to the code repository Basic Model Interface Step 3: the user interface •Users interact through Jupyter notebooks •Python is appropriate for entry level hydrologists •Mixes text and code •Allows for self-documented workflows •We host eWaterCycle on Surf Research Cloud CMIP ERA5 Adopt existing standards! Standardized analysis of weather + climate data Basic Model Interface Containers Standardized output in a JupyterHub (python) interface Overview Project showcase •http://www.ewatercycle.org/projects/ •Showcases BSc and MSc thesis projects •Making students upload their work to Github allows others to continue reuse/built upon their work