Spatial Dynamics of Environmental Health: The Impact of Vegetation Greenness and Heat Exposure on Mental Health Outcomes Across California Census Tracts

Abstract

This study examines spatial relationships between environmental factors, socioeconomicconditions, and mental health across California census tracts using a Spatial Durbin Error Model. Analysis of 7,963 tracts revealed significant spatial autocorrelation in mental health distress (Moran's I = 0.4713, p < 0.001). Median household income was the strongest predictor of mental health distress (direct effect: β = -0.0000489, p < 0.001; indirect effect: β = -0.0000193, p < 0.001). Vegetation greenness showed a significant protective direct effect (β = -3.8818, p < 0.001) without significant spillover effects, indicating localized benefits. Conversely, maximum temperature demonstrated no significant direct effect but had significant positive indirect effects (β = 0.1022, p = 0.0016), suggesting regional rather than local influence. The substantial spatial error parameter (λ = 0.73511) and strong spatial autocorrelation in both vegetation (r = 0.820) and temperature (r = 0.992) validate the spatial modeling approach. These findings enhance understanding of how environmental factors influence mental health through different spatial mechanisms and inform targeted intervention strategies addressing both socioeconomic and environmental determinants of health.