talk · Tuesday 22 September · 08:59–⁠09:11 · ODIN

Adopting Environmental Sustainability as a Strategy to Build Resilient Data Infrastructures in Biodiversity Science

Giuditta Parolini · Building resilient data infrastructures for biodiversity science

Recording time 35:19–44:47Open on Vimeo ↗

Summaries and key moments are machine-produced. Check the recording for the speaker’s exact words.

The short versionBuilding frugal, environmentally sustainable data infrastructures is a practical risk-management strategy for resilience.

Overview

What this was about

Giuditta Parolini argues that environmental sustainability, rarely discussed in research data management, should be treated as a risk-management strategy rather than a nice-to-have. Drawing on the International Energy Agency (growing demand, geopolitical fragmentation, grid limits strained by AI data centres, pressure on critical materials such as rare earths and lithium, slow climate mitigation) and a 2023 ESFRI report on energy and supply challenges, she defines resilience as continued availability of core services under political, legal, financial and technological change. Frugal infrastructures that need fewer resources are better placed to face rising hardware and energy costs and growing EU environmental reporting duties, and biodiversity data are especially at risk because they are politically contentious.

Why it matters. It reframes green IT from a values question into a resilience argument that infrastructure managers and funders can act on as energy, hardware and compliance risks grow.

Key ideas

In the room

  • Sustainability in RDM is usually discussed in financial or social terms, rarely environmental, even though IT infrastructure use of electricity, hardware and e-waste keeps growing.
  • The IEA identifies growing energy demand against geopolitical fragmentation and grids not keeping pace, partly due to generative-AI data centres.
  • Supply chains for critical materials are under pressure (e.g. the 2025 Chinese export restrictions on rare earths; lithium provisioning).
  • Resilience = continued availability of core services under political, legal, financial and technological change; build it by defining core services, identifying risks and adopting a risk-management strategy.
  • ESFRI (2023, after COVID and the Ukraine invasion) recommended greening operations, including materials, reuse and recycling, as risk management.
  • Biodiversity data are politically contentious and face energy/hardware cost risks plus EU environmental regulation and reporting duties.
Jump in

Notable moments

In their words

Transcript

Automatically generated captions can contain mistakes, especially in names and technical terms. Times are relative to the room recording.

Read the transcript ↓
Loading transcript…
Programme abstract

Adopting Environmental Sustainability as a Strategy to Build Resilient Data Infrastructures in Biodiversity Science

Giuditta Parolini

In the aftermath of the COVID pandemic and the Russian invasion of Ukraine, ESFRI (European Strategy Forum for Research Infrastructures) reported on the challenges posed to research by high energy costs and shortages of critical resources required by scientific computing and large experimental facilities. As a response to the difficult situation, ESFRI recommended research infrastructures to increase the environmental sustainability of their operations by decreasing energy use, promoting eco-friendly processes, and increasing reuse and recycling, especially for critical resources in short supply. Since the report was published, new conflicts and geopolitical tensions have further destabilised energy markets and jeopardised the supply chain of key elements for computer hardware, such as rare earth elements. ESFRI's recommendation is now more relevant than ever.

The paper considers how data infrastructures in biodiversity science can become more resilient by adopting environmental sustainability as a guiding principle. Considering the growing amount of research data in biodiversity science and the increasing use of resource-intensive technologies like generative AI as part of research, biodiversity science can greatly benefit from understanding the environmental footprint of its data infrastructures and implement strategies to lower it. This is not only a required contribution to address global environmental challenges, such as climate change, but also a risk management strategy to reduce operational and strategic risks due to scarcity and/or high costs of energy and computing hardware, and to be prepared to comply with more stringent environmental reporting obligations that are likely to involve research infrastructures over time, at least in the European Union.

Submitted to the conference programme before the talk, so the talk itself may differ.