Dylan Young, Gaby Lopez-Gonzalez, Pia Benaud and Richard Grayson, University of Leeds.
Large scale data syntheses are critical for policy and practice decisions about peatlands. We can start by making it easier to access and exchange our data
Imagine you have spent several years collecting peatland monitoring data - water table depths, vegetation surveys, carbon flux measurements that have been carefully recorded in the format that worked best for your project and organisation. Then a colleague from a partner organisation, working on a national synthesis of these data, asks if they can use the data that you have collected. You agree, but several weeks later, you are still reformatting spreadsheets, trying to reconcile your column headers with theirs, and wondering how a task that sounded so simple became so time consuming.
Anyone who has worked to combine peatland data from multiple sources will recognise this story. It's a story of a sector that has grown organically, with data collected by many different organisations using workflows that make sense for their own purposes. However, they were never designed to be exchanged with one another.
The problem is a lack of common standards for data definitions and exchange
Ask almost anyone working in peatland science or restoration whether they think data sharing matters, and the answer is an immediate yes. The value of being able to pool data across sites, organisations and countries is to be able to explore change at scale and to identify what restoration approaches work under what conditions.
Yet data often remains stubbornly siloed. Sharing data can be a time-consuming task because it may need to be cleaned, reformatted and explained. And sometimes a dataset is simply not available. Long-established data collection workflows, developed over years and embedded in team culture and institutional systems, are difficult to change even when there is appetite to do so. And without a common structure for how peatland data is described, every act of sharing becomes a translation exercise.
This was the problem a community of practice set out to address in 2023 with workshops following in May 2025 and November 2025.
Building a standard — the first steps
A community of practice of practitioners, researchers and data specialists working on peatlands first met to develop a draft peatland metadata standard. The standard is a shared set of terms and definitions for describing peatland datasets, aligned with existing data structures and vocabulary used more broadly in environmental science. A common metadata standard provides the vocabulary that makes data discoverable and comparable. But it quickly became clear that a standard alone was not sufficient.
However, even with an agreed standard, organisations still record their data in many ways: different column names, different units, different structures are used. Asking everyone to change how they collect and store data to conform to the new standard is unrealistic.
We needed a way to connect the standard to the data as it exists, without demanding that organisations make significant changes to their existing practices.
An unexpected solution from an unexpected place
When organisations respond to humanitarian crises they coordinate food distribution, medical supplies, population movements and face a version of the same problem. Data arrives from dozens of sources, in dozens of formats, and needs to be combined and acted upon immediately.
Their solution was the Humanitarian Exchange Language, or HXL, which is a set of simple hashtags and attributes that are added to existing datasets as a single tagged row. These tags identify what each column of data represents, using a standardised vocabulary, without altering anything about the underlying structure of the file. The data stays exactly as it was collected. The tags make the data easily exchanged and aggregated.
Because HXL is published under an open Creative Commons licence (CC BY 4.0), we were able to adapt this approach for peatland data. We call it Peatland Exchange Tags, or PXT. PXT adds a layer of shared meaning to peatland datasets without touching the underlying data.
Testing it in practice
In November 2025, the community of practice met again to review and test PXT. The workshop focused on whether the tags were fit for purpose. We looked at whether they captured the right concepts, in the right level of detail, to fit the range of uses people have for peatland data.
Alongside the tagging approach, we have developed a browser-based application, which works entirely offline and allows users to import a dataset, apply the relevant PXT tags to their columns, and save the result. The original data structure is left entirely intact; all that is added is a single row of tags at the top of the file.
During the meeting we identified gaps in the tag vocabulary, highlighted areas where the guidance needed to be clearer, and revealed practical questions that were not initially obvious.
