Socioeconomic inequities within and between cities in objectively measured green space qualities at small geographical scales: Evidence from Australia
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Objective
To determine the extent of inequitable distributions in green space qualities in urban areas of Australia.
Method
Existing data from the cities of Sydney (n≈5M), Newcastle (n≈500k), and Wollongong (n≈300k) in Australia was used to define green space qualities relating to accessibility, amenities/activities, beaches/coastline, biodiversity, incivilities, landcover and land use. Green space qualities were measured within 100m, 200m, 300m, 400m, 800m and 1600m road network distance buffers for residential mesh block centroids and linked with the Australian Bureau of Statistics Index of Relative Socio-economic Disadvantage (IRSD). Correlations were analysed using Spearman’s rank correlation coefficient between IRSD score (reversed; higher scores are more disadvantaged) and green space qualities aggregated over mesh blocks. Single-level and multilevel models were used to examine the influence of IRSD, population density and random effects of population structures. Choropleth maps and hot spot (Getis-Ord Gi*) analyses were undertaken to depict spatial patterns and determine if qualities were spatially clustered.
Results
For green spaces at the 1600m scale, more disadvantaged areas tended to have lower percentages of nearby street trees to roads (Rho=-0.52, p≤0.001) and nearby street trees >4m to roads (Rho=-0.51, p≤0.001), lower percentages of slope >6° (Rho=-0.49, p≤0.001), lower likelihood of threatened mammal species/habitat occurrences (Rho=-0.47, p≤0.001), lower percentages of tree canopy (Rho=-0.46, p≤0.001) and percentage of surrounding tree canopy >4m nearby (Rho=-0.45, p≤0.001). More disadvantaged areas tended to have green spaces with higher percentages of open grass (Rho=0.38, p≤0.001) and bare earth (Rho=0.33, p≤0.001) and higher densities of robberies (Rho=0.34, p≤0.001) and malicious damage (Rho=0.33, p≤0.001). For selected qualities, multilevel models tended to support the relationships that were found using Spearman’s rank correlation.
Discussion
Socioeconomic inequities in tree canopy, biodiversity and incivilities are present within green spaces in large and mid-sized Australian cities.
