This blog is part of the Research Software Camp: Careers and Skills in Research Software series.

Digital skills are becoming increasingly important for research, unlocking powerful tools that accelerate discovery and innovation - but these advances can come with an environmental cost. Every dataset stored, every AI model trained, and every simulation run consumes energy and contributes to emissions. And as the scale of digital research grows, so too does its environmental footprint.

The UK Government's commitment to Net Zero by 2050 will require a reduction in emissions from all sectors, including research. Digital research should be at the forefront of this transition, helping to unlock new ways of working that reduce environmental impact while maintaining quality, efficiency, and innovation. At the same time, it is increasingly important to ensure that the environmental footprint of digital research is minimised, without compromising the efficiency or quality of the work itself.

The good news is that many sustainable digital research practices are simply good digital research practices. Efficient code runs faster, well-managed data is easier to reuse, and optimised workflows save time and resources. So time spent transitioning to greener practices can also be mutually beneficial for research.

Why Training Matters

The challenge is that research is complex - there’s no one-size-fits-all route to more sustainable science. What’s essential for one field might be irrelevant to another, and that’s okay. The key is to identify what changes can make the biggest difference in your work.

The Greening Digital Research project (led by Weronika Filinger with Jeremy Cohen, Martin Jukes and Kirsty Pringle) is a collaboration between the CHARTED, NetDRIVE, DisCouRSE and SCALE-UP NetworkPlus projects. It has been working with training experts, including the team behind the DIRECT Framework, as well as researchers and digital research technical professionals (people who support research in a wide range of roles), to better understand where training efforts should be focused.

To explore this, we ran a workshop at a recent NetDRIVE meeting aimed at identifying how digital research training needs align with emerging green skills and sustainability goals. It’s early days, so the list is likely to change, but take a read through our initial points and think about what might apply to your work.

Ten Training Priorities for Net Zero Research

Here, we highlight ten areas where workshop participants identified training as important. Different priorities will matter for different research areas and styles - start wherever makes sense for your work.

Active Data Management & Reuse

Be deliberate about what data you keep, how long you keep it, and how you describe it.

Learn how to apply FAIR principles (Findable, Accessible, Interoperable, Reusable), create clear metadata, and avoid the trap of “store everything forever.”

Why it matters: Every unnecessary terabyte stored increases the need for additional storage hardware, which uses additional energy in both its operation and construction. Good metadata and reuse reduce duplication across the community.

Efficient & Responsible AI

AI and machine learning are powerful tools, but are also power-hungry. Training in efficient model design, experiment planning, and impact assessment helps you make sure AI is used wisely, not just because it’s trendy.​

Why it matters: Unnecessary use of AI models or poorly designed AI models can increase energy use and associated emissions.​

Code Profiling & Optimisation

Learn how to write lean, efficient code and use profiling tools (which can help you to see where your programs waste time and energy). Understanding compiler options, algorithms, and libraries can make your software (and your research!) run faster and greener.

Why it matters: Small improvements in code performance can be important when scaled to lots of runs or users.

Robust Software Practices

Test early, test often, and fix bugs before scaling up. Training in debugging, version control, and code review saves resources, time, and frustration.​

Why it matters: Every failed or repeated run is wasted energy. Plus it will save users time!

Sustainable High Performance Computing & Workflows

High-performance computing (HPC) is a key part of modern research, but it’s also a major energy consumer. Learn to optimise job submissions, right-size resources, and (if appropriate) use carbon-aware scheduling.  If you are unsure then speak to the HPC provider - they will have the expertise to help.  ​

Why it matters: A well-optimised job can do the same science with a fraction of the energy.

Smarter Cloud & Container Use

Cloud platforms are convenient, but easy to use incorrectly. Understanding when cloud solutions are appropriate and how they are deployed helps avoid hidden carbon costs

Why it matters: “Set and forget” cloud jobs often run long after they’re needed.​

Embedding Sustainability into Project Management

Make environmental impact part of your planning, not an afterthought. Learn to include carbon costing, risk assessment, and sustainability checkpoints in your project lifecycle.  Good project management can also reduce the risk of failures, which both speeds up research and wastes fewer resources.

Why it matters: What you measure, you can manage - and improve.

Leading for Change

You don’t have to manage a team to show leadership; you can advocate for sustainability wherever you work. For example, running training for your colleagues or peers, sharing best practices, and promoting opportunities can all have a positive impact.

Why it matters: Change happens when people lead by example.

Ethics, Adaptability & Collaboration

Being an ethical researcher means considering not only what you do, but how you do it. Developing skills in ethical reasoning, adaptability, and teamwork helps you make balanced decisions, work effectively with others, and avoid unnecessary waste or duplication. These skills apply across the whole research lifecycle, not just digital aspects.

Why it matters: Sustainable research is about people as much as technology.

Communication & Community

Good communication spreads good practice. Training in clear, evidence-based communication and community leadership helps to engage others and build collective change. It also raises awareness of the importance of sustainable research practices and encourages wider participation.

Why it matters: Sustainability is a shared journey, not a solo effort.

What do you think?

These topics came out as important from the workshop, but what do you think? We would love to hear any comments or suggestions: Greening Digital Research Training - Suggestion Form​

Where to Start

If you’re new to sustainability in research, here are a few great places to begin:

Green Software Foundation:  Free course on reducing the environmental impact of software, and lots of relevant blog articles.

Carbon Literacy Project:  Provides training to help individuals and organisations understand and act on carbon emissions, paid but often available through research institutions.

Digital Humanities Climate Coalition Toolkit: This toolkit is a guide to making your research practices more environmentally responsible. It is geared towards digital practices, but also touches on general areas such as travel and advocacy.

Final Thought

Achieving Net Zero in research won’t happen overnight, but every small change helps. Whether you’re optimising your code, rethinking your data storage, or mentoring others to do the same, you’re contributing to a more sustainable research ecosystem. You don’t need to do everything - just start somewhere.