Speaker
Description
Keywords: urban climate shelters; policy translation; climate adaptation; planning under uncertainty
Climate shelters have gained attention as urban responses to extreme heat, particularly in European cities where schools, libraries, parks and civic facilities are being adapted to provide accessible spaces of thermal protection. Their emergence reflects a broader shift from temporary, event-based responses towards networks of everyday public spaces that can reduce vulnerability before, during and after climate extremes. Nevertheless, the concept remains closely associated with heat adaptation and with cities whose climatic, institutional and socio-spatial conditions differ substantially from those found in many contexts in the Global South. Its transfer to cities affected primarily by recurrent floods therefore raises a fundamental planning question: what can travel across hazards and territories, the physical model of a climate shelter, or the principles through which protective spaces become embedded in urban systems? This question is explored through Rio do Sul, a medium-sized city in the Itajaí Valley in southern Brazil. The city has a long history of river flooding, while intense rainfall also produces landslide risks in occupied hillsides. Recurrent disasters have generated considerable local experience in monitoring, warning, evacuation and emergency sheltering. At the same time, climate change, urban occupation of floodplains and slopes, population change and unequal adaptive capacities complicate decisions about how protection should be planned. These conditions make Rio do Sul a relevant case for examining whether the urban climate shelter concept can be translated beyond its dominant heat-focused framing and used to connect emergency response with longer-term, multi-hazard adaptation.
In this perspective, this study poses the question of under what conditions the urban climate shelter concept can be translated to support long-term adaptation in a flood-prone city. Rather than approaching transfer as the replication of a successful model, it understands it as a process of translation in which ideas are selected, interpreted and reassembled in relation to local risks, institutions, infrastructures and values (Dolowitz & Marsh, 2000; McCann & Ward, 2013). This distinction is particularly important for climate shelters. A schoolyard designed to provide shade during heatwaves, for example, cannot simply be reproduced as a solution for flood displacement. What may be transferable are broader principles: the use of familiar and accessible facilities; the multifunctionality of spaces before, during and after crises; the connection between physical protection and social infrastructure; and the integration of emergency management, spatial planning and community knowledge.
This perspective also reframes uncertainty. In Rio do Sul, uncertainty is not limited to forecasting rainfall or river levels. It concerns the location and changing composition of vulnerable populations, the willingness and ability of residents to evacuate, the availability and condition of shelter facilities, the distribution of responsibilities between public authorities and private or community organisations, and the resources required to maintain protective infrastructure between disasters. Climate adaptation measures are not inherently effective or equitable and may reproduce vulnerabilities when they ignore these institutional and social dimensions (IPCC, 2022; Schipper, 2020). Climate shelters must therefore be assessed not only as physical spaces, but as socio-spatial and governance arrangements whose effectiveness depends on accessibility, trust, coordination and the capacity to learn over time.
The paper develops an exploratory qualitative case study as part of ongoing doctoral research. It combines analysis of municipal planning and emergency-management arrangements with semi-structured interviews involving local actors engaged in municipal governance, civil protection, social assistance and the management of emergency shelters. The interviews examine perceptions of climatic risks and vulnerable groups, existing emergency and adaptation practices, the qualities and limitations of shelter spaces, institutional responsibilities, public participation and the possible local relevance of urban climate shelters. A visual elicitation exercise presents examples of climate shelters in other cities, particularly transformed schoolyards and public facilities, not as models to be copied but as prompts through which participants can identify possibilities, barriers and necessary adaptations. Interview material is being examined through thematic analysis, combining deductive categories derived from the climate-shelter framework with themes emerging from the Rio do Sul case.
Preliminary findings reveal a strong local capacity for emergency response alongside a less consolidated connection between disaster management and long-term spatial adaptation. Municipal actors describe a professionalised civil-protection system supported by hydrological and meteorological monitoring, real-time information and established procedures for opening shelters as river levels rise. Emergency operations mobilise several sectors: civil protection assesses conditions and activates shelters; public works supports evacuation and logistics; and social assistance manages registration, food, psychosocial needs and everyday life inside the facilities. This coordination represents an important institutional resource. However, it is principally activated in relation to an imminent or ongoing event, while the incorporation of sheltering into wider urban-planning and adaptation processes remains an evolving task.
Existing shelters expose the material and social limits of an emergency-oriented model. Many are gymnasia, community halls, religious facilities or other spaces that retain a different primary function and are frequently not owned by the municipality. They may provide immediate safety but were not designed for families to inhabit for days. Preliminary accounts emphasise insufficient privacy, sanitation and bathing facilities, limited possibilities for preparing food, inadequate electrical capacity, difficulties accommodating pets, and the emotional and interpersonal pressures created when people with different needs share large collective spaces. Internet access, safe electrical connections, spaces for children and shaded outdoor areas also become important during extended stays. These observations suggest that protection cannot be reduced to the number or geographic distribution of available facilities. The quality, usability and dignity of the sheltering experience are central adaptation concerns.
The reliance on third-party facilities creates an additional governance challenge. Public investment in buildings that the municipality does not own is constrained by legal and administrative requirements, even when improvements are necessary for public protection. Facility owners may also be reluctant to continue serving as shelters when spaces are damaged during occupation. The transfer of the climate-shelter concept therefore depends on agreements concerning use, investment, maintenance, liability and post-event repair. These arrangements illustrate how apparently spatial interventions are shaped by institutional rules and relationships that become visible only when the concept is examined in practice.
At the same time, the interviews identify opportunities for reinterpretation. Participants responded positively to relatively modest adaptations such as access to drinking water, seating, shade, vegetation, improved sanitary facilities, connectivity and more welcoming outdoor areas. Existing municipal buildings already function informally as points of rest and assistance for some residents, suggesting that everyday practices can provide a foundation for a broader network of protective spaces. Schools and community facilities may be particularly important because they are distributed across neighbourhoods and familiar to local populations. Yet opening school grounds outside operating hours raises questions about staffing, security, maintenance and responsibility. The viability of these spaces consequently depends not only on design, but on operational arrangements and local trust.
The preliminary analysis also highlights unequal capacities to anticipate and respond to risk. Lower-income households often occupy more exposed locations and have fewer alternatives when flooding occurs. New residents and migrants may be unfamiliar with local river behaviour, warning systems and evacuation procedures, making risk communication and neighbourhood-level relationships especially important. Community organisations, neighbourhood associations and volunteers act as intermediaries between municipal institutions and affected populations, but participation appears stronger in communication, mobilisation and response than in the strategic decisions that shape adaptation. Translating the climate-shelter concept could therefore create an opportunity to involve residents not only as shelter users or volunteers, but as participants in determining where protective spaces are located, how they operate and which needs they address.
The findings suggest that transferring urban climate shelters to Rio do Sul requires a shift from object transfer to institutional and socio-spatial translation. The relevant outcome is not the importation of a European heat-shelter typology, but the development of a locally grounded network of spaces capable of performing different functions across time. Before an event, these spaces could support education, community activities, risk awareness and everyday well-being. During floods or other emergencies, they could provide safe and dignified temporary accommodation. After an event, they could support recovery, access to information and the restoration of community networks. Greening and heat protection may remain relevant within this model, particularly because flooding can occur in different seasons and shelter occupation may coincide with warm conditions, but these functions must be combined with the practical requirements of flood evacuation and displacement.
The Rio do Sul case contributes to planning debates by showing that transferability is itself a form of uncertainty. Concepts move through selective interpretation, encounter existing institutions and may acquire new meanings as they are implemented. This process can generate innovation, but it also risks stretching the climate-shelter concept until it becomes an imprecise label for any supportive public space. A meaningful transfer must therefore preserve the concept's core purpose - accessible protection from climate-related risks for vulnerable populations - while allowing its spatial forms, operational functions and governance arrangements to change. The case indicates that five interconnected conditions are particularly important: institutional integration, context-sensitive spatial design, equity and participation, stable financial and operational arrangements, and mechanisms for learning and adjustment.
The study concludes that climate shelters should not be treated as universal solutions or fixed models. Their value lies in providing a framework through which cities can reconsider the relationship between emergency infrastructure and everyday urban life. In Rio do Sul, the concept creates a lens for identifying how existing shelter practices could evolve from temporary crisis management towards a more anticipatory and socially embedded adaptation system. Uncertainty is not eliminated in this process. Instead, it is addressed through coordination, experimentation, community knowledge and iterative learning. Transferring climate shelters across hazards and contexts is therefore best understood not as copying an intervention, but as building the local capacity to translate principles of protection into institutions and spaces that respond to changing risks and unequal vulnerabilities.
References
Amorim-Maia, A. T., Anguelovski, I., Connolly, J., & Chu, E. (2023). Seeking refuge? The potential of urban climate shelters to address intersecting vulnerabilities. Landscape and Urban Planning, 238, 104836. https://doi.org/10.1016/j.landurbplan.2023.104836
Caldarice, O., Abastante, F., Mecca, B., Ozeren, Z., Pincegher, B., & Raico Torrel, E. P. (2026). Urban experimentation as a driver of climate adaptation: A European review of climate shelter in national adaptation policies and practices. Sustainability, 18(7), 3300.
Dolowitz, D. P., & Marsh, D. (2000). Learning from abroad: The role of policy transfer in contemporary policy-making. Governance, 13(1), 5–24. https://doi.org/10.1111/0952-1895.00121
Intergovernmental Panel on Climate Change. (2022). Climate change 2022: Impacts, adaptation and vulnerability. Cambridge University Press. https://doi.org/10.1017/9781009325844
McCann, E., & Ward, K. (2013). A multi-disciplinary approach to policy transfer research: Geographies, assemblages, mobilities and mutations. Policy Studies, 34(1), 2–18. https://doi.org/10.1080/01442872.2012.748563
Pincegher, B. (2026). From emergency shelters to urban climate shelters: Enabling conditions for long-term climate adaptation. Urban Research & Practice. Advance online publication. https://doi.org/10.1080/17535069.2026.2732162
Schipper, E. L. F. (2020). Maladaptation: When adaptation to climate change goes very wrong. One Earth, 3(4), 409–414. https://doi.org/10.1016/j.oneear.2020.09.014