Beyond Static Cooling Accessibility: Lower-Tail Availability of Urban Park Cooling Services Across Expanding Pedestrian Catchments

Urban park cooling accessibility is often assessed using temporally aggregated thermal performance, which may obscure variation in accessible cooling potential. We combined Landsat-derived park cooling intensity with a fixed network of 77 park destination records and 63 subdistricts in Tianjin, China, across 14 screened summer observations in 2021–2025. Mean cooling accessibility (MeanCA), its 25th percentile (Q25CA), and absolute and relative lower-tail gaps (AG and TAG) were compared across 10, 15, and 20 min catchments. Positive-supply coverage increased from 23 to 41 and 51 subdistricts, while MeanCA and Q25CA retained similar spatial rankings. Among the 23 subdistricts already served at 10 min, TAG showed a small overall threshold difference (Kendall’s W = 0.178); its 10–20 min decline coexisted with an increase in AG. The TAG difference was not significant in the broader 41-subdistrict comparison from 15 to 20 min. Among the 27 eligible multi-source subdistricts, direct Q25CA modelling did not provide clear evidence of an additional asynchrony association after nonlinear MeanCA adjustment. Replacing TAG with AG changed 29 of 63 diagnostic labels. These findings distinguish expanded access from changes in relative and absolute lower-tail gaps. The framework offers a conditional planning diagnosis across observed dates, with category-level conclusions sensitive to indicator choice.