UrbanStudyUrban Research Radar
Urban Ecology & Green Space
Journal articleOpen access

Safety Perceptions in Urban Green Spaces: A Global Systematic Review

Published 13 Sep 2026 · Added to radar 5 Nov 2026

Source abstract

Urban green spaces (UGS) provide critical benefits for urban quality of life, yet safety perceptions significantly influence their utilisation and effectiveness. Despite the growing literature on UGS and safety perceptions, two critical gaps persist: there is no synthesis that has examined how physical design priorities, socio-economic factors, and usage patterns vary across global regions; and existing research remains biased towards the Global North. This article addresses these gaps through a systematic review of 135 articles, using machine learning-assisted screening and applying PRISMA guidelines to enable a comprehensive cross-regional analysis of safety perceptions in UGS. The results reveal critical distinctions between universal and context-dependent safety factors. Physical characteristics, particularly vegetation structure, maintenance, and lighting, are determinants of safety perceptions across all regions. However, regional studies identified distinct safety concerns, including physical incivility in Sub-Saharan and North Africa, social incivility and police presence in Latin America and the Caribbean, and infrastructure-related factors in the Middle East and North America. The analysis reveals that socio-economic factors show context-dependent influences, underscoring the risk of standardised design approaches. Findings also show that safety perceptions influence UGS use, with physical activity and recreation, stress relief, and contact with nature as the primary drivers across all regions. Overall, the study advances theoretical understanding of the extent to which safety perception frameworks are universally valid or require contextual adaptation. This cross-regional synthesis demonstrates that effective UGS planning must account for both universal design principles and context-specific safety priorities.

Abstract metadata supplied by the research index; wording belongs to the original source.

Structured record

Source-reported findings
  • The results reveal critical distinctions between universal and context-dependent safety factors.
Methods
systematic review, machine learning

Read the original ↗ DOI ↗