Temporal Coupling of Urban Forest Phenology and Heating-Season Exposure from a Climate Adaptation Perspective: Implications for Air Quality

Urban forests are widely promoted for improving air quality, yet their effectiveness is typically assessed through static green-space indicators that ignore seasonal variation in vegetation activity. This limitation is especially consequential in cold-region cities, where winter heating-season pollution peaks coincide with the leaf-off period of deciduous trees. Using a monthly panel of 15 centrally heated cities in northern China (2015–2024; N = 1464), this study develops a phenology-aware framework integrating three indicators: effective forest capacity (EFC), which combines dynamic forest area with a sigmoid leaf-on share and city-specific evergreen fraction; heating-season exposure (HI); and a binary phenology–heating mismatch (PHM) flag. City–year–month fixed-effects models show that the EFC–PM 2.5 association is directionally negative but statistically inconclusive under conservative inference (city-clustered SE: p = 0.523 ; wild bootstrap: p = 0.541 ), whereas associations with SO 2 and O 3 are statistically robust. The central empirical contributions are the four-quadrant heterogeneity analysis and the topographic paired comparison: four-quadrant heterogeneity analysis reveals that forest capacity shows clearer negative associations in dry semi-humid cities, whereas HI dominates in heating-dominated plain cities. A paired topographic comparison between Urumqi and Xining illustrates how terrain-induced inversions can override forest signals. The results support differentiated urban greening strategies that coordinate forest expansion with heating-system transition, evergreen species planning, and ventilation-sensitive urban design.