When people speak about disaster resilience, they often focus on infrastructure, warning systems, or policy frameworks. But in places like the Indian Sundarbans Delta, resilience is also shaped by everyday negotiations over water, land, livelihoods, and voice. That is what led us to this study. We wanted to understand not only whether co-design “works,” but how it works in practice when communities facing floods, salinity intrusion, waterlogging, and cyclones are brought into decision-making processes that are usually dominated by technical institutions. Our paper explores this through three embedded cases in the Sundarbans: climate-resilient betel vine cultivation, integrated water management for salinity control, and community-led mangrove afforestation.

One of the most important things to clarify is that our research was an ex-post evaluation. We did not arrive as outside experts to design the interventions ourselves. Instead, we examined how co-design unfolded within ongoing public programmes and mandated forums such as Gram Sabhas, Gram Panchayat Development Plan convergence platforms, and Water Users’ Association processes. That distinction mattered to us because we wanted to study how local actors, government staff, facilitators, and community members negotiated design choices in real institutional settings rather than in an idealized experimental environment.

What made the research especially compelling was that the most meaningful design changes were often not the most technically dramatic ones. Sometimes they were procedural. In one case, a young landless agricultural worker explained that workshops were initially held in the morning, exactly when wage workers were out in the fields. After repeated requests, organisers shifted some sessions to the evening. That seemingly simple adjustment changed who could participate, and attendance reportedly rose from 25% to 85%. In another case, a smallholder farmer insisted that a hand-dug canal used locally be added to the official map, even though it did not exist on the technical base map. In yet another setting, dissatisfaction with bureaucratic language led facilitators to rely more on local dialect and pictorial representation. These moments reminded us that inclusion is not achieved by merely opening the door. It has to be designed into the process.

That insight shaped one of our core arguments. Co-design in high-risk environments is not simply collaborative problem-solving. It is a process of epistemic negotiation. Farmers, fishers, women’s groups, technical experts, and local institutions do not all enter the room with the same authority, vocabulary, or forms of evidence. Yet what we saw in the Sundarbans was that participatory mapping, charrette-style design sessions, and validation circles created spaces where local memory and scientific reasoning could actively reshape one another. Historical flood routes, knowledge of micro-topography, seasonal “water-flow memory,” and fishing access routes did not remain anecdotal. They became design-relevant data.

The betel vine case captured this especially well. Farmers dealing with salinity stress, waterlogging, and cyclone-related damage worked through design modifications such as raised beds, trenches, improved boroj layouts, organic mulch, drainage adjustments, standby pumps, and access to simple irrigation hardware. What stood out was that the turning point was not just the adoption of a technical package. It was the shift from externally prescribed trials to farmer-led experimentation. The paper reports that use of new methods increased from 30% to 80% after the intervention, alongside improved productivity and renewed confidence among participants that they could adapt under changing environmental conditions.

The mangrove afforestation case added another layer. Here, ecological restoration was inseparable from livelihood politics. Community members linked mangrove restoration to fish habitats and local economic survival, while others raised concerns about whether nurseries would remain viable once initial funding ended. A Dalit cooperative member described struggles over keeping the nursery in a location compatible with fishing livelihoods. These were not side issues. They revealed that long-term sustainability depends on whether ecological interventions are socially negotiable and institutionally durable.

Perhaps the biggest surprise in this project was that co-design did not erase power relations. Instead, it made them visible and contestable. That is a more realistic and, in many ways, more useful finding. Rather than romanticising participation, the paper shows that co-design can move processes from token presence toward meaningful influence when it is culturally grounded, iterative, and attentive to timing, facilitation, and representation. Our broader contribution is a process model built around epistemic negotiation, engineered inclusivity, and temporal alignment. We hope this helps move debates on participation beyond general endorsement and toward more precise questions about what kinds of design moves actually redistribute decision-making power.

For us, the paper ultimately became about something larger than three case studies in one delta. It became about how public-sector decision-making can be changed when narratives, lived experience, and community knowledge are treated as legitimate inputs rather than afterthoughts. As climate risks intensify, the challenge is not only to design better interventions, but to design better relationships among institutions, expertise, and affected communities. That is the deeper story behind this paper.https://doi.org/10.1080/15710882.2026.2632233