Turning publications into data – imagining a world of linked hydrologic knowledge

As a science, hydrology faces diverse sets of interacting processes combined with a vast heterogeneity of our environment. Ideally, one would be broadly knowledgeable in all processes of the water cycle, including their variations across the planet, but taking such a holistic approach to our science has become problematic due to the vast number of hydrologic case numbers published. More than 25 000 articles were published on the topic of water resources in 2022 alone. Such publication numbers make it impossible to keep up with the current literature, not to mention the knowledge acquired over time.

But at the same time, these publications comprise a vast source of information and data that is not being utilised at the moment. For example, we are currently unable to connect our highly localised process knowledge for a broader understanding.

One solution that has been discussed in the past is to extend our article metadata to relevant hydrologic information to support search and synthesis of hydrologic knowledge. And there is a wide range of potential metadata that can be relevant: research topic, study location, models used, time period covered, data availability…. In regard to data services, it could be used to link data collections or networks with the models that use that data, the researchers who employ the models, and the publications that summarise the knowledge gained.

For this data to be useful and used by the community, it will need a collaborative platform to host this information. We discuss the use of Wikidata, a free, accessible, machine-readable, and editable by all, database, for this task. An added benefit is that Wikidata can easily link to existing data repositories and identification numbers, thus making the concept not only of interest for article metadata but also a potential interoperable approach for data services.