Physiologically Active Vegetation Reverses Its Cooling Effect in Humid Urban Climates

Abstract

Efforts to green cities for cooling are succeeding unevenly because the same vegetation that cools surfaces can also intensify how hot the air feels. Previous studies have identified humid heat as a growing urban hazard, yet how physiologically active vegetation governs this trade-off between cooling and moisture accumulation remains poorly understood, leaving mitigation policy and design largely unguided. Here we quantify how vegetation structure and function influence the Heat Index (HI), a combined measure of temperature and humidity in 138 Indian cities spanning tropical savanna, semi-arid steppe, and humid subtropical climates, and across dense urban cores and semi-urban rings. Using an extreme-aware, one kilometre reconstruction of HI and an interpretable machine-learning framework that integrates SHapley Additive Explanations (SHAP) and Accumulated Local Effects (ALE), we isolate vegetation-climate interactions. Cooling generally strengthens for EVI >= 0.4 and LAI >= 0.05, but joint-high regimes begin to reverse toward warming when EVI >= 0.5, LAI >= 0.2, and fPAR >= 0.5,with an earlier onset for fPAR >= 0.25 in humid, dense cores. In such environments, highly physiologically active vegetation elevates near-surface humidity faster than it removes heat, reversing its cooling effect and amplifying perceived heat stress. These findings establish the climatic limits of vegetation-driven cooling and provide quantitative thresholds for climate-specific greening strategies that promote equitable and heat-resilient cities.

Figures (5)

• Figure 1: Conceptual framework linking canopy structure, greening features, and urban form to thermal discomfort. The upper panel illustrates how canopy density regulates heat stress mechanisms: dense canopies enhance shading and evapotranspiration through higher Leaf Area Index (LAI) and Fraction of Absorbed Photosynthetically Active Radiation (fPAR), leading to latent-dominated heat stress; whereas sparse canopies promote ventilation but increase sensible heat flux due to lower LAI and fPAR, resulting in sensible-dominated heat stress. The lower panel conceptualizes how these processes interact with urban morphology within a microclimatic boundary. Solar radiation, air temperature, and humidity define the ambient thermal environment, while greening features such as, Enhanced Vegetation Index (EVI; vegetation area), LAI (canopy depth), and fPAR (vegetation health) and urban features such as night-time lights (NTL; built intensity) and Local Climate Zones (LCZ; urban form), jointly regulate local heat exchange. Their combined influence determines the degree of thermal discomfort experienced within urban neighborhoods.
• Figure 2: Spatial and temporal patterns of Heat Index (HI). (a) City-level linear trends in HI (°C yr$^{-1}$; 2003–2020) across 138 Indian cities, grouped by Köppen–Geiger class and population density (high/low). Stippling marks $p<0.05$. (b,c) Interannual variability of mean HI for high- and low-density cohorts, respectively; vertical bars denote $\pm1$ SD across cities in each class.
• Figure 3: Relative importance and direction of greening and urban-form features on HI. (a) Feature importance derived from SHapley Additive exPlanations (SHAP) for EVI, fPAR, LAI, and NTL across Aw, BSh, and Cwa classes and urbanization strata (Urban–High/Low, Semi–High/Low) for 2003–2010 and 2011–2020. Darker shading indicates stronger explanatory power (rank = 1). (b) Decadal shifts in the direction and strength of associations shown as paired donut charts; inner rings represent urban cores and outer rings semi-urban zones.
• Figure 4: Joint effects of vegetation attributes on Heat Index (HI). Filled markers show $\mu_{\mathrm{all}}$ (domain mean) and hollow markers $\mu_{\mathrm{HH}}$ (joint-high mean) for Urban-core (top) and Semi-urban (bottom) zones during 2003–2010 (left) and 2011–2020 (right). Circles = EVI–LAI, squares = EVI–fPAR, triangles = LAI–fPAR. Horizontal bars show $\pm1$ SD. Color intensity indicates cooling coverage (% of pixels with net SHAP $<0$); negative values denote associations with lower HI, positive with higher HI.
• Figure 5: Greening–productivity interactions and their effects on Heat Index (HI, °C) across climate types and urbanization levels. (a) Partial dependence plots showing the marginal effects of Enhanced Vegetation Index (EVI), Leaf Area Index (LAI), and Fraction of Photosynthetically Active Radiation (fPAR) on HI across urban (high, low) and semi-urban (high, low) classes for tropical (Aw), arid steppe (BSh), and humid-subtropical (Cwa) climates. Shaded bands represent 95% confidence intervals. (b) Two-dimensional Accumulated Local Effects (2D-ALE) maps showing combined interactions among EVI, LAI, and fPAR pairs, with contours indicating zero-crossing isolines ($\Delta \mathrm{HI} = 0$). Red regions denote warming (positive $\Delta \mathrm{HI}$) and blue regions denote cooling (negative $\Delta \mathrm{HI}$), illustrating nonlinear and synergistic effects of greening and productivity on thermal regulation across urban gradients. All ALE results pool the full 2003–2020 period by design to illustrate “what happens if” we increase or decrease greening features, singly or in pairs, rather than to compare decades.