41 / 2026-01-25 14:59:31

Urban heat island dynamics under dynamically evolving urban boundaries: Spatiotemporal patterns and coupling effects in Chengdu (2000–2022)

Urban heat island; Urban expansion; Greenhouse gas emissions; Vegetation–climate coupling; Thermal–ecological imbalance

主题2 大气物理、化学、环境 > 专题2.16 大气环境多尺度监测与智慧决策

Between 2000 and 2022, Chengdu experienced rapid urbanization, with the built-up area expanding from approximately 600 km² to over 1,600 km², accompanied by a near twofold increase in greenhouse gas (GHG) emissions. These pronounced changes provide a representative background for examining long-term urban thermal dynamics. However, most previous studies of the Urban Heat Island (UHI) have relied on static urban boundaries, which are unable to capture the continuously evolving spatial extent of cities and may therefore underestimate cumulative thermal impacts associated with urban expansion.

In this study, we investigate the spatiotemporal evolution of the Urban Heat Island in Chengdu, China, from 2000 to 2022 using dynamically evolving urban boundaries. Urban heat island intensity (UHII) is quantified using three complementary indices—annual mean, maximum, and minimum UHII—to jointly represent background thermal conditions, extreme heat responses, and baseline warming. By integrating station-based air temperature observations with data on urban expansion, GHG emissions, and vegetation dynamics derived from MODIS NDVI, this study examines the coupled interactions among urban growth, emissions, vegetation change, and the urban thermal environment.

The results indicate a persistent intensification of UHII over the study period. All three UHII indices exhibit clear upward trends, reflecting a sustained strengthening of the urban thermal environment under rapid urban expansion. Spatial analyses reveal a distinct core–periphery structure of UHI intensity, with higher UHII values concentrated in the urban core and lower values prevailing in peripheral areas, highlighting the strong influence of urban form and expansion patterns on thermal heterogeneity.

GHG emissions show strong coupling with UHII across multiple temporal scales. Correlation analyses consistently indicate positive relationships between emissions and all UHII indices. Lead–lag analysis further reveals that changes in GHG emissions precede UHII variations on multi-year timescales, with the strongest coupling occurring at a lag of approximately seven years. This temporal structure suggests that urban thermal responses are governed by cumulative and delayed forcing effects, rather than immediate or short-term emission changes.

Vegetation dynamics exhibit pronounced spatial contrasts and stage-dependent interactions with UHII. Comparisons between urban expansion and non-expansion areas show that NDVI is dominated by degradation in newly urbanized regions, whereas non-expansion areas are characterized by widespread vegetation improvement. Temporal analyses indicate that the vegetation–UHII relationship varies across different stages of urbanization, reflecting context-dependent regulation rather than a fixed or uniform cooling effect.

Coupling coordination analysis shows that the overall system maintains a relatively high level of coordination throughout the study period, with the coupling coordination degree (CCD) generally remaining in the range of approximately 0.7–0.8 after 2012. Nevertheless, increasing divergence among the urban development, emission, and eco-thermal subsystems indicates increasing divergence between the rapidly expanding urban system and the lagging eco-thermal regulation, despite the overall coordination degree remaining stable. Overall, this study demonstrates that static boundary-based assessments may underestimate long-term urban thermal risks, and highlights the importance of dynamic boundary frameworks for understanding coupled urban–environmental processes and supporting sustainable urban climate management.