Buildings are assessed using typical conditions, a questionable approach for a heating climate. This project will exploit novel datasets to develop a world atlas of weather events to stress-test buildings to enhance mitigation of and adaptation to climate change based on known and likely impacts.
A decarbonised built environment is key to climate change mitigation and adaptation, yet buildings still consume 29% of the global primary energy, mainly due to space conditioning demand . This is particularly worrying given the imminent expansion of the building stock associated with population growth to 9.7 billion by 2050, the service life of buildings and their costly retrofit . As the climate changes, there is a need to map closely and rigorously the boundary conditions of buildings through the so-called weather files, to facilitate learning from regular and atypical weather events and promoting robust planning. However, and due to historical limitations, current practice relies on: (1) the existence of weather stations with suitable records nearby the location of interest; and (2) carefully crafted single-year weather files that prompts the typical response of buildings according to a single key performance indicator like space heating demand or indoor overheating.
Thanks to satellite imaging and novel climate reanalyses datasets, a new generation of weather files is now possible to deliver unprecedented spatiotemporal resolution (the whole world across decades) to create a rigorous, global testbed for the built environment [4, 5]. Here, historical data is essential to assimilate past performance as it provides every relatable event while helping establish the methods to prepare and consume datasets for future weather (near future) and climate change, improving predictive control and resilience. This project will leverage new approaches to weather file creation that allow assimilating such data for the built environment with emphasis on robust decision-making, usability, and interpretability.
- What are the key mechanisms through which weather events affect building performance and indoor environments?
- What are the main weather event types that challenge mitigation of climate change?
- What are the main weather event types that challenge resilience to climate change?
- To what extent do extreme weather conditions influence decision-making by stakeholders when evaluating building performance?
The shift from appraisals under prevailing weather conditions to stress-testing buildings and their occupants with extreme events is a natural step to take but one that requires a different (a) approach to boundary conditions selection and (b) outcomes for the built environment. This project sets out to examine the hypothesis that it is possible to isolate meaningful events for building performance evaluation (1) that can be mass-produced from publicly available datasets on climate reanalyses and impacts in the built environment, (2) that capture the variability of observed weather-driven response of buildings, (3) that complement state-of-the-art single-year, single-parameter weather files, and (4) that facilitates decision-making by stakeholders.
- Review of the literature on weather file creation, available datasets, and extreme weather events. Work towards submission to specialist journal (months 1-9).
- Develop a codebase to establish a synthetic building stock database representative of different geographical contexts (months 7-12).
- Develop codebase to identify weather events from existing weather station and reanalysis datasets and testing on building physics engines (months 13-18).
- Appraise the representativeness of results based on impact on building occupants and outcomes for the built environment. Estimation of potential global coverage (months 18-24).
- Validate and evaluate performance of the novel weather files to current state of the art (months 24-30).
- Evaluate impact on decision-making by designers, facility managers or authorities, when appraising building performance in the presence of known, relatable weather events (months 30-36).
- Disseminate findings in academic journals, international conferences, and professional bodies (months 18-36).
- Thesis write up (months 36-42).
A comprehensive training programme will be provided comprising both specialist scientific training and generic transferable and professional skills. The PhD candidate will be introduced to comprehensive training options, within the University of Edinburgh and elsewhere. Depending on the experience of the candidate, options include (1) building physics simulation engines (commercial and/or research), (2) introduction to data science, (3) geospatial data analysis, (4) observation and analysis of surface solar radiation. The candidate will have the opportunity to become a teaching assistant following formal training, and there will also be opportunities to contribute to wider training and outreach activities. Further training in both academic and interdisciplinary skills will be available as part of Edinburgh’s Institute for Academic Development.
Candidates should have at least a 2:1 undergraduate degree in Engineering, Environmental Sciences, Physics or in another relevant programme. Experience or willingness to learn programming (Python, R, Julia, or similar) to manipulate large-scale datasets, manage simulations and analyse results will be essential. A background in building physics or meteorology would be an advantage but it is not essential.
The University of Edinburgh is committed to equality of opportunity for all its staff and students, and promotes a culture of inclusivity. Please see details here: https://www.ed.ac.uk/equality-diversity
This project is competing for funding under the E4 scheme at the University of Edinburgh. We also welcome candidates that would like to propose their own projects for the student-led category of the E4 funding scheme at the University of Edinburgh.