NERC Environmental Data Services : API 4 AI project kickoff workshop report

Booth, Jonathan; Bell, Patrick; Kingdon, Andrew ORCID: https://orcid.org/0000-0003-4979-588X; Heaven, Rachel; Card, Chris; Sauze, Colin; Tso, Michael ORCID: https://orcid.org/0000-0002-2415-0826. 2025 NERC Environmental Data Services : API 4 AI project kickoff workshop report. NERC, 38pp. (OR/25/019) (Unpublished)

Before downloading, please read NORA policies.

Preview

Text OR25019.pdf Download (4MB) | Preview

Abstract/Summary

The NERC Environmental Data Service (EDS) provides a focal point for scientific data and information spanning environmental science domains: atmosphere and climate; earth observation, polar and cryosphere; marine, terrestrial and freshwater; geoscience, and solar and space physics. Improving access to quality-assured, high-resolution environmental information and associated software tools will enable new users and communities to access environmental research by facilitating integrated analyses of environmental processes in response to societal challenges and environmental change. The wholesale availability of Machine Learning (ML) and other Artificial Intelligence (AI) technologies is changing the way that environmental data is being used. They enable a new scale of processing and allow the identification of trends and signals in data streams that were previously impossible to identify or practically impossible to deliver due to a lack of computing power. Application Programming Interfaces (API) are a core enabling technology for AI as they provide a machine readable connection between programmes / algorithms to connect directly to other programmes and crucially data automatically without human interference. Widening the availability of API-supplied data will deliver new functionality for AI capabilities. This project aims to share experience in creating and applying standardised APIs for utilisation in AI workflows. This will widen data access to environmental researchers and enable systematic AI analysis of multiple environmental data types to underpin the development of predictive environmental modelling and digital twins. The project received UKRI Digital Research Infrastructure Programme funding through the opportunity entitled ‘Enhancing digital research infrastructures by trialling approaches to skills and software’. This funding provided resources across the British Oceanographic Data Centre (BODC) hosted at National Oceanographic Centre (NOC), the Environmental Information Data Centre (EIDC) hosted at UK Centre for Ecology & Hydrology (UKCEH) and the National Geoscience Data Centre (NGDC) hosted at British Geological Survey (BGS) with a deadline of 31st March 2025. This workshop brought together staff from those organisations as well as representatives from the Polar Data Centre hosted at British Antarctic Survey (BAS) and the Centre for Environmental Data Analysis (CEDA), part of the National Centre for Atmospheric Science. The purposes of the workshop were many fold: • To introduce staff working on the project to generate a team spirit to be carried through the project and identify common objectives and priorities • To establish a common baseline agreement and understanding of the aims and objectives of the project • To demonstrate previous work that was relevant and could be utilised within the project • To begin establishing science use cases that would be used as exemplars to structure the provision of data APIs for utilisation in AI workflows • To begin establishing the AI workflow technologies that would be utilised in the project • To begin establishing the data APIs that would be required to address the selected science use cases • To collaboratively co-design the project and identify its deliverables, a project timeline and participant responsibilities • To start generating ideas for further work to build on the outcomes of this project that can be put forward in response to future funding opportunity calls The full agenda for the workshop can be viewed in Appendix A – Agenda. The day was structured into three main sections, opening with introductions and a series of knowledge-sharing presentations across the organisations. These introduced current AI workflow technologies being utilised; the NOC Data Science Platform (DSP) and the UKCEH Datalabs platform. In addition, BGS provided an overview of their standards-based approach to API development. The second part of the day focused on a workshop activity to identify science use cases and the AI workflows and data APIs needed to address them. This resulted in 25 use cases with ideas from across all organisations. The results of this exercise can be viewed in Appendix B – Use cases The final part of the day encompassed a second workshop activity that generated a project timeline to address the identified project deliverables that built towards successfully developing and integrating the required data APIs and AI workflows to meet the chosen science use cases. This can be seen in Appendix C – Timeline

Item Type:	Publication - Report
UKCEH and CEH Sections/Science Areas:	National Capability and Digital Research (2025-)
Funders/Sponsors:	NERC, British Geological Survey, National Oceanography Centre, UK Centre for Ecology & Hydrology, British Antarctic Survey, Centre for Environmental Data Analysis
NORA Subject Terms:	Computer Science Data and Information
Date made live:	19 Mar 2025 13:29 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/539107

Actions (login required)

View Item

Document Downloads

More statistics for this item...