Glasgow's hidden energy patterns: Energy clustering and social geospatial analysis for a just transition

Worldwide, governments have recognized that a place-based approach is required to decarbonize heat and achieve net zero targets. Previous research has identified groups of buildings in need of energy efficiency upgrades and areas most suitable for heat networks. However, a deeper understanding of how potential interventions may impact different sociodemographic groups is imperative to ensuring a just energy transition. This paper uses clustering techniques and Geographic Information Systems (GIS) to addresses this gap in the literature. The framework proposed can be applied to cities worldwide to analyse the varying relationships between domestic energy consumption, housing, and sociodemographic attributes at sub-city level, visualise how current decarbonization plans cut across different citizen-groups, and identify gaps and opportunities for all stakeholders. Using Glasgow City as a case study, application of the proposed framework reveals that high levels of social deprivation (prevalent in Glasgow) act as a barrier to energy consumption, while poor energy efficiency - common in buildings owned by the city's more affluent residents - is associated with high energy consumption. Spatial comparison to existing decarbonization plans highlights several groups of citizens that may benefit or lose out under existing plans, providing useful insight to policy makers, and validating the benefits of this novel approach. A clear strategy is needed to enable low-income citizens in their transition to net zero and encourage the most affluent citizens to take measures to improve energy efficiency in their homes. • Spatial maps of citizen-personas based on energy consumption • Assessment of social implications of existing decarbonization plans • Identification of significant decarbonization opportunities and risks • Consideration of social factors and implications for a just transition • Easy replication for other cities with same (or similar) data sources