The phrase “urban green space” becomes insufficient when environmental performance depends on relations that vegetation alone cannot describe. Lagos, Cape Town, Addis Ababa and Tunis expose four different limits of the green-space category. Lagos demonstrates that heat risk depends on the intersection of climatic hazard, built morphology, vegetation, population exposure and socioeconomic vulnerability. Cape Town shows that exceptional biodiversity can coexist with profound barriers to access. Addis Ababa reveals that planned green areas do not persist without governance and maintenance structures. Tunis shows that land represented as available for urbanisation may already function as hydrological and ecological infrastructure. Metropolitan nature emerges from these studies as relational capacity rather than as a quantity of green surface. This distinction changes the analytical object. In Lagos, the relevant object is not simply urban heat but the configuration through which heat becomes dangerous to specific populations. A Local Climate Zone is not intrinsically risky; risk intensifies where dense morphology, limited vegetation, poverty, vulnerable age groups and weak infrastructure coincide. The contribution of Obe, Morakinyo and Mills is therefore not only a more detailed map. Their model demonstrates that environmental exposure is socially and morphologically organised. The map becomes an instrument for locating where metropolitan form and social vulnerability reinforce one another.

Cape Town reveals a parallel structure through access. Protected nature can be spatially close and practically remote. Milliken’s evidence shows how private-car dependence, inadequate public transport, safety concerns, information deficits and metropolitan segregation shape who reaches conservation landscapes. Access is thus a relational property between environmental space and urban systems, not an intrinsic attribute of the reserve. A hectare of protected biodiversity may function as a major ecological asset while remaining weakly integrated into everyday metropolitan life. This difference is essential for any project interested in lived spatial culture rather than environmental provision alone. Addis Ababa relocates the same problem inside institutions. Green space survives through governance arrangements that connect planning, land, maintenance, public participation and management. The formal existence of a park or community green space does not guarantee its persistence. Land allocated for greenery may be converted to other uses, agencies may fail to cooperate, and management can fragment across administrative levels. Yet community-managed spaces also demonstrate that governance capacity does not reside exclusively in municipal hierarchies. Environmental infrastructure is maintained through distributed actor networks whose legitimacy and effectiveness can differ from formal planning structures. Tunis adds a more fundamental challenge: ecological capacity may predate the city project itself. The sebkha is already performing storage, filtration, erosion protection and habitat functions before architecture arrives. Treating such land as undeveloped territory is therefore an epistemic error with spatial consequences. Mili’s design-research approach makes this visible by asking architecture to begin with geomorphology and hydrology rather than with a formal programme. Her insistence that an appropriate conclusion may be to build less or not build at all transforms environmental knowledge from a constraint into a project-generating intelligence. Together, these cases support a comparative method based on relations rather than universal indicators. Lagos contributes hazard–exposure–vulnerability; Cape Town contributes provision–access–use; Addis Ababa contributes designation–governance–maintenance; Tunis contributes site process–ecological function–project decision. These relational structures can travel between cities as questions without pretending that their answers are transferable. A heat-risk protocol entering Tunis would encounter different climatic and hydrological assumptions. A Cape Town access framework applied in Addis Ababa would need to adapt from conservation reserves to community-managed neighbourhood spaces. A sebkha-based hydrological reading transferred to Lagos would require reinterpretation within a humid coastal megacity shaped by different settlement and water systems. The comparative value therefore lies in identifying what survives contextual change and what does not. A shared framework can ask how morphology, vegetation, water, accessibility, maintenance and governance interact, but each city must retain authority over the meaning of those relations. This is the operational strength of Metropolitan Natures: common variables establish a field of comparison while city-specific evidence prevents comparison from becoming homogenisation. Environmental knowledge becomes useful when it can move without losing the conditions that produced it. For Socioplastics, the resulting archive should preserve not only topic and city but method, spatial scale, environmental process, governance regime, affected population, evidence limitation and project implication. A heat-risk layer, an access survey, a governance model and a wetland-design studio cannot be treated as equivalent evidence objects, but they can be relationally connected. The graph becomes stronger when methodological difference remains visible. Metropolitan nature then appears not as a universal category waiting to be measured, but as a family of situated relations through which cities distribute thermal comfort, ecological access, institutional responsibility and future environmental capacity.