Urban greening may help to reduce the population health impacts of loneliness and its concomitants, such as hopelessness and despair. However, the literature lacks both a critical appraisal of extant evidence and a conceptual model to explain how green space would work as a structural intervention. Both are needed to guide decision making and further research. We conducted a systematic review of quantitative studies testing associations between green space and loneliness, searching seven databases. Twenty two studies were identified by 25/01/2022. Most of the studies were conducted in high-income countries and fifteen (68%) had cross-sectional designs. Green space was measured inconsistently using either objective or subjective indicators. Few studies examined specific green space types or qualities. The majority of studies measured general loneliness (e.g. using the UCLA loneliness scale). Different types of loneliness (social, emotional, existential) were not analysed. Of 132 associations, 88 (66.6%) indicated potential protection from green space against loneliness, with 44 (33.3%) reaching statistical significance (p<0.05). We integrated these findings with evidence from qualitative studies to elaborate and extend the existing pathway domain model linking green space and health. These elaborations and extensions acknowledge the following: (a) different types of green space have implications for different types of loneliness; (b) multilevel circumstances influence the likelihood a person will benefit or suffer harm from green space; (c) personal, relational, and collective processes operate within different domains of pathways linking green space with loneliness and its concomitants; (d) loneliness and its concomitants are explicitly positioned as mediators within the broader causal system that links green space with health and wellbeing. This review and model provide guidance for decision making and further epidemiological research on green space and loneliness.