Green infrastructure (GI), such as urban forest and parks, can promote clean air and water, and an array of other ecosystem benefits. Equitable access to the benefits of GI in cities is one tenet of environmental justice. Identifying areas that could most benefit from increased GI investment can assist planners in promoting environmental justice in urban areas. We compared plant greenness in different income-level neighborhoods of Portland, Oregon using remotely-sensed spectral data and analysis techniques. Using Landsat 8 Operational Land Imager (OLI)/Thermal Infrared Sensor (TIRS) data, a standard index for measuring vegetation greenness, the Normalized Difference Vegetation Index (NDVI) can be calculated. Plants absorb solar radiation in the photosynthetically active spectral region, and then cells of leaves re-emit radiation in the near-infrared region (0.7 to 1.1 µm). A higher NDVI indicates a higher density of green leaves, while lower NDVI indicates a lower density of green leaves. We asked, “Are differences in plant greenness of neighborhoods in Portland related to income level?” We hypothesized that high-income neighborhoods will have higher NDVI, while those with low-income will have the lowest NDVI, for multiple reasons, including lot size, percent impervious surface, or site management.
Results/Conclusions
Using Landsat 8 OLI/TIRS data for 2016, NDVI was calculated for Portland, Oregon. The mean NDVI for all of Portland was 0.393 with a median of 0.396. Tests for statistical significance (Kruskall-Wallis) indicate that plant greenness is not equally distributed among the different median household income (MHI) level neighborhoods (p < 0.05). A post-hoc test (Benjamini-Hochberg FDR method) revealed significant differences in mean NDVI between the two highest median income household groups as compared to the three lower median income groups (p < 0.05). These patterns may potentially be explained by: 1) structural difference in lot sizes, proportion of pavement, and ways that built infrastructure in different neighborhoods constrain the potential amount and size of vegetation; 2) differential effort by residents and/or managers to promote and maintain healthy green infrastructure in lower MHI areas, and/or 3) residents with higher incomes have the ability, preference, and financial resources to move to areas with greater amounts of healthy, green plants. Our results can be used to inform managers about where more vegetation or attention to vegetation health is needed to increase equitable access to GI within the various MHI neighborhoods of Portland.