Abstract Pocket parks have emerged as a practical form of small-scale urban green infrastructure for climate adaptation, yet thermal-comfort evidence at this scale remains methodologically fragmented. This review aimed to identify the dominant data sources, indicators, and methodological paradigms used in pocket-park thermal-comfort research, and to clarify how these choices vary across climates, seasons, spatial configurations, and user groups. Following PRISMA 2020, we searched Web of Science Core Collection, Scopus, EBSCOhost, and CNKI using concept-harmonised but database-adapted search strings. After deduplication and two-stage screening, 26 studies were included in the qualitative synthesis. A structured quality appraisal covering reporting transparency, data quality, validity, and reproducibility was then used to interpret evidence strength. The evidence base shows three co-existing research orientations: baseline cooling-effect verification, mechanism-oriented design optimisation, and people-centred multi-domain comfort assessment. PET and UTCI remain the dominant indices, frequently paired with subjective votes and, increasingly, behavioural evidence. Method choice varies systematically with study context: summer and heatwave studies are dominated by simulation and PET/UTCI-based analyses, whereas winter, user-group-specific, and multi-domain studies rely more strongly on field monitoring, questionnaires, and mixed-method designs. Pocket-park thermal-comfort research is maturing from single-site descriptive studies toward more integrated, design-oriented, and human-centred approaches, but cross-study comparability is still constrained by uneven climate coverage, inconsistent reporting, and variable evidence quality. This review provides a clearer method-indicator-context map for selecting appropriate approaches in pocket-park thermal-comfort studies and offers an explicit agenda for future design, monitoring, and simulation research.
Xin et al. (Sun,) studied this question.