Who has access to green relief? A cost-distance equity index for green infrastructure planning in Niš

Abstract

The ecosystem services of Urban green infrastructure (UGS) are well-documented, however conventional accessibility measures like Euclidean buffers and arbitrarily weighted composite indices poorly capture real pedestrian access or defensible priority-setting. This study develops a network-based cost-distance equity index for green infrastructure in the urban core of Niš, Serbia, using a 100 m analysis grid. Walking-time costs to local, district, and city parks were derived via GRASS r.cost, transformed through negative exponential decay functions calibrated with planning standards, and combined into a non-compensatory composite deficit score. Bivariate Local Indicators of Spatial Association (LISA), tested across three spatial weight specifications, identified statistically robust priority clusters overlapping deficit with four age-vulnerable population subgroups. In total, 6,134 grid cells are examined. Of these, 917 cells, or 15.0%, fall into robust Tier 1 priority zones. These zones hold about 57 to 59% of the vulnerable population for each subgroup. That is a large share, even though the zones cover only 15% of the study area. The resulting three-tier classification, together with descriptive attributes for each cluster, gives a ranked set of planning outputs that can be used for decision making process. This approach avoids arbitrary weights at each step, and keeps the same logic from the decay calibration, through the deficit index build, cluster robustness test. Thus it offers a transparent, reproducible framework that other mid-sized cities with comparable open data availability could adapt to their own green infrastructure planning processes.