Urban heat islands (UHI) amplify the burdens of a warming climate across cities globally. UHIs have disproportionately severe consequences for people in poorly insulated housing, those on low incomes, older adults, and other population groups with limited adaptive capacity. Beyond acute heat-health issues, the long-term impacts include elevated morbidity and mortality, reduced labour productivity, exacerbation of energy poverty through rising cooling demand, and stress on urban ecosystems and infrastructure. As warming intensifies, cities will require adaptation strategies that can both reduce thermal exposure and distribute benefits fairly, so that today’s investments protect future generations, and prevent entrenching inequalities further. Nature-based solutions (NbS) are a tractable intervention with potential to cool surfaces, attenuate stormwater, and enhance biodiversity; however, without explicit attention to equity, deployment risks clustering benefits in already advantaged areas.
This project will develop an AI-based, multi-objective optimisation framework that integrates environmental effectiveness and distributive justice in the design of urban heat adaptation strategies. The UK has a built environment which is poorest in quality across the European continent and urgently needs attention in protecting people from the impacts of rising heat. Using UK-wide analyses, the research will combine gridded temperature climatologies with small-area indicators of social vulnerability to identify where interventions in the form of various NbS (green roofs, trees, planted beds, stormwater capture, etc.) can generate the greatest aggregate heat reduction while also reaching populations with compounded disadvantages. The optimisation process using evolutionary algorithms will formalise these twin aims as simultaneous objectives — maximising cooling and maximising coverage of vulnerable populations — subject to realistic cost and feasibility constraints such as placement on facades, rooftop structure, street network design. Stakeholder engagement will be designed during the project with the MET and Charnwood Borough Council (Loughborough) which will inform objectives, constraints, and evaluation criteria, and the resulting evidence will be translated into web-based interactive maps for investment allocation, open-source modelling frameworks, and policy briefs.
Methodologically, the work advances the coupled use of AI multi-objective optimisation with urban-climate and social data, providing a pragmatic basis for cities to plan just, future-oriented heat adaptation strategies.
This project is not suitable for CASE funding
Each host has a slightly different application process.
Find out how to apply for this studentship.
All applications must include the CENTA application form.
Choose your application route
The study will construct an AI-based multi-objective optimisation workflow using UK gridded temperature observations. Baseline thermal fields will be derived from HadUK-Grid (daily Tmax/Tmin at ~1 km), aggregated to summer climatologies and transformed into UHI indicators via urban–rural masking. A seven-dimension social vulnerability index (economic, cultural, community, gender, built environment, institutional, health/wellbeing) will be built at LSOA/MSOA scales from the UK IMD; indicators will be normalised, weighted, and stress-tested with bootstrap resampling to capture uncertainty. A parametric cooling model will then estimate expected temperature reductions from candidate NbS deployments, and parameters (e.g., cooling intensity, diminishing returns) sampled to reflect physical variability. These components will be coupled through evolutionary optimisers (e.g., NSGA-II) to allocate NbS across ground, façade, or roof subject to budget, jointly maximising heat reduction and social coverage. Outputs will include maps of efficiency–equity trade-offs, hosted in an open-source adaptation web interface.
DRs will be awarded CENTA Training Credits (CTCs) for participation in CENTA-provided and ‘free choice’ external training. One CTC can be earned per 3 hours training, and DRs must accrue 100 CTCs across the three and a half years of their PhD.
The doctoral researcher will develop expertise in urban-climate analysis and AI-based multi-objective optimisation, including formulation of equity and efficiency objectives, evolutionary algorithms, and Pareto-front interpretation. They will work with UK gridded observations (HadUK-Grid), small-area socio-environmental indicators, and uncertainty quantification (bootstrapping, Monte-Carlo sensitivity). Technical skills will include Python/R for spatial data science, data engineering, and reproducible pipelines (version control, testing). They will also learn to appraise the use of various NbS technologies. Professional skills will include stakeholder engagement with public agencies, translating results into policy-ready maps/briefs, scientific writing and presentation, and research ethics and responsible data management.
Dr Nazli Aydin will support the candidate on urban-adaptation strategy, governance levers and implementation pathways to ensure scenarios reflect feasibility of governance mechanisms. Dr Juliana Gonçalves will support on building-physics aspects — roof typologies, cooling kernels and structural feasibility of different NbS — informing optimisation constraints. These partners will attend quarterly meetings held via hybrid modes, co-author papers and co-design workshops with local stakeholders in the two case-study cities. Recently dated commitment emails from Aydin and Gonçalves are attached.
Year 1: Spatial data science and comparative urban analysis.
The first year culminates in a data-led paper that maps and explains the geography of urban heat and social vulnerability across contrasting UK urban forms. Using HadUK-Grid to derive summer thermal climatologies and small-area indicators to characterise sensitivity and adaptive capacity, the study will build UK-wide UHI baselines, develop urban-form typologies, and quantify how exposure covaries with housing stock, morphology, and socio-demographic composition, social factors that may be correlated with the impact of climate heating. This will help identify the forms of urban adaptation necessary for protecting the most vulnerable from climate impacts.
Year 2: AI framework and equity–efficiency trade-offs.
The second year produces a methodological paper that constructs the AI multi-objective optimisation framework and evaluates trade-offs between aggregate cooling and coverage of vulnerable populations. The work will formalise objectives, constraints, and feasibility elements; implement evolutionary optimisers to generate Pareto frontiers; and apply Monte-Carlo and scenario analyses to test robustness to budgets, vulnerability definitions, and modelling assumptions.
Year 3 — Policy pathways and transformational change.
The final year delivers a synthesis paper that translates optimisation results into actionable policy pathways capable of driving transformational change across UK cities. It will integrate institutional constraints, explore multi-year investment, and compare alternative governance levers — standard policies, incentives, and place-based mechanisms — under uncertainty, to analyse the determinants of effective investments by government.
All years — Validation will use Year-1 analyses, Year-2 modelling framework, and workshops with partners to assess plausibility and generalisability to other contexts (for example, The Netherlands where other external partners are based) that are different to the UK in policy implementation. This will lead to testing of governance levers and its impact on trade-offs between efficiency and equity.
Cutter, S.L., Boruff, B.J. & Shirley, W.L. (2003) ‘Social vulnerability to environmental hazards’, Social Science Quarterly, 84(2), 242–261. https://doi.org/10.1111/1540-6237.8402002
Frantzeskaki, N., McPhearson, T., Collier, M.J., Kendal, D., Bulkeley, H., Dumitru, A. et al. (2019) ‘Nature-based solutions for urban climate change adaptation: Linking science, policy and practice communities for evidence-based decision-making’, BioScience, 69(6), 455–466. https://doi.org/10.1093/biosci/biz042
Santamouris, M. (2020) ‘Recent progress on urban overheating and heat island research: Integrated assessment of the energy, environmental, vulnerability and health impact—Synergies with global climate change’, Energy and Buildings, 207, 109482. https://doi.org/10.1016/j.enbuild.2019.109482
Susca, T., Gaffin, S.R. & Dell’Osso, G.R. (2011) ‘Positive effects of vegetation: Urban heat island and green roofs’, Environmental Pollution, 159(8–9), 2119–2126. https://doi.org/10.1016/j.envpol.2011.03.007
Tapia, C., Abajo, B., Feliu, E., Mendizabal, M., Martínez, J.A., Fernández, J.G., Laburu, T. & Lejarazu, A. (2017) ‘Profiling urban vulnerabilities to climate change: An indicator-based vulnerability assessment for European cities’, Ecological Indicators, 78, 142–155. https://doi.org/10.1016/j.ecolind.2017.02.040
Weber, S., Sadoff, N., Zell, E. & de Sherbinin, A. (2015) ‘Policy-relevant indicators for mapping the vulnerability of urban populations to extreme heat events: A case study of Philadelphia’, Applied Geography, 63, 231–243. https://doi.org/10.1016/j.apgeog.2015.07.006
Vanclay, F. (2002) ‘Conceptualising social impacts’, Environmental Impact Assessment Review, 22(3), 183–211. https://doi.org/10.1016/S0195-9255(01)00105-6
Lin, M., Dong, J., Jones, L., Liu, J., Lin, T., Zuo, J., Ye, H., Zhang, G. & Zhou, T. (2021) ‘Modeling green roofs’ cooling effect in high-density urban areas based on law of diminishing marginal utility of the cooling efficiency: A case study of Xiamen Island, China’, Journal of Cleaner Production, 316, 128277. https://doi.org/10.1016/j.jclepro.2021.128277
Hill, B., Marjoribanks, T., Moore, H., Bosher, L., & Gussy, M. (2025) ‘Market-based instruments to fund nature-based solutions for flood risk management can disproportionately benefit affluent areas’, Communications Earth & Environment, 6(1), 714. https://doi.org/10.1038/s43247-025-02706-2
For any enquiries related to this project please contact Prof Trivik Verma, [email protected].
To apply to this project:
Applications must be submitted by 23:59 GMT on Wednesday 7th January 2026.