| dc.description.abstract | Most nature and health research use the normalized difference vegetation index (NDVI) for measuring
greenness exposure. However, little is known about what NDVI measures in terms of vegetation types
(e.g., grass, canopy coverage) within certain analysis zones (e.g., 500m buffer). Additionally, more
exploration is needed to understand how to interpret changes in NDVI (e.g., per 0.1 increments) for
policy intervention in urban greening. This study aims to address such gaps in the literature. We,
therefore, measured mean values of NDVI and other vegetation metrics for different buffer zones
(100, 300, and 500 m) at sample locations within the Greater Manchester study area by applying focal
statistics. For each scale, we fitted linear and nonlinear (i.e., GAM) regression models to explore (1)
what vegetation types and amounts best predict the NDVI values in a multivariate model and (2) how
changes in NDVI explain changes in different vegetation coverages in univariate models. We found
that NDVI is most sensitive to tree canopy at 300 meters scale and that changes at different levels of
vegetation values and types predict dissimilar changes in NDVI. The models with three vegetation
types can explain approximately 80% variance in NDVI values (i.e., Pseudo R²) for buffer zones of
300 and 500 meters. We observed that forb and shrub density is most sensitive (12.21% change) to
mid-range increments in mean NDVI (0.45 to 0.55) within 300 meters. These sensitivities usually
follow nonlinear patterns for all the vegetation types in multivariate and univariate models. Our
results indicate that NDVI values are more sensitive to certain types and amounts of vegetation within
various buffer zones. Overall, interpreting changes in NDVI values for urban greening interventions
would require careful evaluation of the relative changes in types and quantities of vegetation for
different buffer zones. | |
