Enabling Organisational Change Through Ground-Up Initiatives: A Case Study from the STFC Scientific Computing Department

In the modern world, scientific discovery relies heavily on vast networks of computers that crunch numbers, model complex systems, and store immense amounts of data. These machines, often housed in large data centers, are the engines of research, but they come with a heavy price tag: they consume enormous amounts of electricity. As the global community strives to reach "Net Zero" targets—a goal where the amount of greenhouse gases released into the atmosphere is balanced by the amount removed—the energy used by these digital tools has become a critical concern. The electricity demand from data centers is rising rapidly, driven by the growing use of artificial intelligence and the sheer scale of modern computing. If the scientific community is to meet its environmental goals, it must find a way to make its digital infrastructure cleaner and more efficient without slowing down the pace of discovery.

This challenge is at the heart of a recent case study from the Scientific Computing Department at the Science and Technology Facilities Council in the United Kingdom. This department supports tens of thousands of researchers with a massive array of computing power and software. Faced with the need to reduce its carbon footprint, the department did not wait for a top-down mandate to dictate every move. Instead, a small group of staff members formed a team to build a sustainability strategy from the ground up. They recognized that while high-level goals existed, the real work of changing how thousands of people use computers required a different approach. Their story offers a practical blueprint for how a large organization can transform its daily operations to align with environmental targets, turning abstract policies into concrete actions.

The journey began with a group of volunteers who met informally to discuss sustainable computing. They quickly realized that without dedicated time and resources, their ideas would remain just that: ideas. To move forward, they developed a structured strategy that focused on five key areas where they could make a direct difference. They decided against trying to solve every problem at once, such as fixing global supply chains, which were outside their control. Instead, they concentrated on what they could influence: monitoring their own emissions, teaching users how to be more efficient, setting standards for best practices, and providing long-term support to keep these efforts alive.

A central part of their work involved understanding exactly how much carbon their computers were producing. In the past, this data was often missing or too difficult to gather across their diverse systems. The team tackled this by training early-career staff and apprentices to build tools that could track energy use. They created software that could monitor the emissions from specific computing jobs, allowing users to see the environmental cost of their work. For example, they developed a system that could tell a researcher the best time to run a calculation based on how clean the electricity grid was at that moment. They also built dashboards that showed users how much carbon their virtual machines were generating, turning invisible energy use into visible data that people could act upon.

With better data in hand, the team turned their attention to the people using the systems. They understood that simply telling researchers to "be greener" was not enough; they needed to show them how. The team organized workshops and created training modules to help developers and users understand the environmental impact of their choices. They worked on features that would display real-time usage graphs and forecasts, helping scientists decide when to schedule their tasks or when to shut down unused workspaces to save energy. By embedding these sustainability considerations directly into the tools researchers used every day, they made it easier for individuals to make informed decisions without needing to become experts in environmental science themselves.

To ensure these changes stuck, the team also focused on building a culture of certification and shared knowledge. They encouraged different groups within the department to aim for a recognized digital sustainability certification, which provided a clear roadmap for improvement. To help these groups succeed, the team created templates and guides, reducing the effort required to get started. They established a network of representatives from each research theme to share updates and keep the conversation going. While maintaining this momentum proved difficult, as people are often busy with their primary research, the team found that regular meetings and a clear sense of community helped keep the initiative alive. They also made a point of documenting their work and sharing it with others, creating a record of what worked and what did not.

The results of this approach have been promising. The department has successfully moved from a state of scattered, voluntary efforts to a coordinated strategy with clear goals and measurable progress. They have trained a new generation of staff in sustainable computing skills, developed tools that give users visibility into their carbon footprint, and created a framework that other departments can follow. The team acknowledges that they still face challenges, particularly in securing enough funding and staff time to sustain these efforts long-term. They have not yet solved every problem, such as the environmental cost of the hardware itself or the storage of massive data sets, but they have laid a solid foundation.

Ultimately, this case study demonstrates that significant organizational change does not always require a massive overhaul from the top. By empowering a small group to coordinate efforts, provide tools, and foster a culture of shared responsibility, a large scientific department can begin to align its operations with global environmental goals. The work shows that when researchers are given the right information and the right support, they can make the changes necessary to reduce the environmental impact of the digital tools that drive modern science. This approach offers a practical path forward for any organization looking to turn the broad ambition of Net Zero into the daily reality of its operations.