Quantification of the environmental footprint of research software to help the UK research community achieve Net Zero

Project Description

The use of software in research is pervasive, however whilst the environmental impact of this is significant it is also poorly understood. This research will develop tools and techniques to help better quantify the environmental impact of research software and develop understanding around common software practices that do the most environmental harm.

Primary Supervisor: Dr Kirsty Pringle

Project Overview

Over 90% of researchers use software in their research, but the environmental impact of this research software use is significant but poorly understood. This PhD will focus on better quantifying the environmental impact of research software, understanding the common software practices that have the largest impact and identifying where reductions could be made. There is scope to examine tools, training or guidance that could be used to quantify environmental impact, and also the barriers to adoption. This work will inform the researchers, research software engineers and also policy makers, in order to help the community reduce the environmental impact of research software.

Although often keen to reduce their environmental impact, researchers are seldom trained in techniques to either quantify or minimise the environmental impact of their research. Furthermore, they often struggle to find resources and advice that is suitable for their particular research needs. Research software engineers (RSEs) work directly with researchers to develop research software. Although a relatively new role, the collaboration of researchers and RSEs has been shown to improve the standard of software used in research and facilitate the adoption of research software best practices. RSEs have great potential to help the research community learn how to understand, evaluate and reduce the environmental impact of their research. But while there have been some early efforts to build a community of practice around green software approaches, little research has been done to identify software approaches that are most appropriate to research, or to clarify the role that researcher software engineers could play in helping to achieve the changes required to achieve Net Zero.

Overview of the research area:

UK Research and Innovation (UKRI) is committed to becoming net zero by 2040, as part of this the UKRI recently published results from a scoping project which examined the challenge of achieving Net Zero in more detail. The report found that “there are very considerable savings to be made in many ways in the process of development and deployment of research software. Some of these gains are policy based, e.g. to avoid wasting computing time by occupying machines for access; others are concerned with reducing the need to re-run experiments (code review, testing); others with proper design of experiments

to minimise energy expenditure to obtain required results; and finally code efficiency, in terms of choice of programming language, program optimisation and compilation optimisation.” It also suggested that there is a role for research software engineers to help the development and adoption of good practices that will reduce the environmental footprint of research.

The research software engineering community has grown dramatically since the term was first coined 10 years ago, there are now 40 UK institutions that employ RSE and an estimated 10,000 RSEs worldwide. It is now widely recognised that working with RSEs can help researchers adopt software best practises, and thus improve the efficiency and impact of the research. The role of the RSE in aiding researchers to reduce the environmental impact of their research is, however, less clear, for example it has been suggested that it would be beneficial to encourage a new “Green RSE” role in which some RSEs would specialise in green software and work to inform researchers on best practises, and thus upskill the community.

The research landscape is constantly evolving, and present the environmental impact of the research is not currently evaluated at any stage of the research process, it is likely that in line with the UKRI’s commitment to becoming Net Zero by 2040 researchers will increasingly be asked to qualify and justify their environmental impact, significant training would be needed for the community to be able to do this accurately.

Student Requirements

Note that these are the minimum requirements to be considered for admission.

A UK 2:1 honours degree, or its international equivalent, in a relevant subject such as computer science and informatics, physics, mathematics, engineering, biology, chemistry and geosciences.

You must have significant programming experience in at least one major programming language (for example C, C++, Python, Fortran, or Java) and should be familiar with qualitative data analysis and statistics.

English Language requirements as set by University of Edinburgh

Recommended/Desirable Skills

Experience of qualitative research, conducting research interviews and designing surveys, and an understanding of the research software environment would be desirable.

How to apply

Applications should be made via the University application form, available via the degree finder. Please note the proposed supervisor and project title from this page and include this in your application. You may also find this page is an uneful starting point for a research proposal and we would strongly recommend discussing this further with the potential supervisor.