The research, led by Hayley Leggett of Tohoku University with colleagues at the International Research Institute of Disaster Science, applied human-centred design, a methodology that places the needs, values and lived experiences of end users at the heart of the design process, to the problem of tsunami mitigation under rising seas. The team drew on earlier empirical work in Kesennuma showing that residents value visibility and connection to the sea, want defences that blend with the landscape, prefer alternatives to concrete, and place trust in infrastructure rooted in local identity as much as in perceived safety. Rather than treating these preferences as constraints on engineering, the researchers used them as generative design inputs, producing three conceptual multi-layered defence configurations and then testing each one rigorously through numerical simulation under combined tsunami and sea level rise scenarios.

The three designs represent distinct philosophies. Defence 1, an eco-hybrid system, integrates natural and low-impact solutions: offshore submerged concrete pile arrays with artificial reefs that dissipate wave energy through drag and turbulence while enhancing biodiversity, vegetated berms and bioswales along the coastline that absorb overflow and filter sediment, and elevated evacuation landscapes inland that double as public green space. Defence 2, a discreet integrated barrier system, prioritises openness: low-crest tetrapod fields arranged in staggered grids offshore, glass-composite seawalls with automated hydraulic floodgates that sit flush with the ground under normal conditions and rise automatically during tsunami alerts, and rooftop safe zones with elevated pedestrian pathways inland. Defence 3, a reinforced high-performance system, is the engineering-led benchmark, combining reinforced concrete pile fields at the harbour entrance, stepped seawalls of five to seven metres paired with buffer parks that temporarily store overtopping water, and floodable retention areas providing secondary containment.

To evaluate these configurations, the team employed TUNAMI-N2, a two-dimensional nonlinear shallow-water model that solves the depth-averaged nonlinear shallow-water equations for incompressible flow under the hydrostatic assumption. The model conserves mass and momentum, represents bottom friction through spatially varying Manning’s roughness coefficients, and was solved using a staggered leap-frog finite-difference scheme with second-order accuracy in space and time. A six-layer nested grid system resolved tsunami dynamics from regional propagation at 1,215-metre resolution down to 5-metre grids covering Central Kesennuma and Kesennuma Bay, capturing urban streets, river mouths and ports where run-up and inundation actually occur. A uniform one-second time step satisfied the Courant-Friedrichs-Lewy stability condition in the finest domain. TUNAMI-N2 has been extensively validated against observed data from the 2011 Tohoku tsunami along the Sanriku coast and the 2004 Indian Ocean tsunami, giving the comparative results a solid empirical foundation.

The simulation campaign was exhaustive: 150 runs covering all combinations of two study areas, three tsunami sources drawn from the Kuril Trench, the Japan Trench and the 2011 Tohoku event, five defence states including no defences and the current post-2011 infrastructure, and five sea level rise conditions. Those future scenarios came from the NASA Sea Level Projection Tool at the Ofunato II reference location, with increments of +0.40 metres by 2050 and +0.97 metres by 2100 under the intermediate SSP2-4.5 pathway, and +0.44 and +1.28 metres under the high-emissions SSP5-8.5 pathway. Each simulation produced inundation extent, fatality estimates based on depth-dependent empirical rates applied to census population data, and economic losses calculated per building using depth-damage functions and a conservative 20-million-yen replacement proxy.

The results revealed a consistent and, for advocates of human-centred design, remarkable hierarchy. Defence 3 achieved the strongest containment across every metric, as expected for the most structurally dominant configuration. But Defence 2, the transparent, open-feeling barrier system that most closely aligns with residents’ preferences for visibility and waterfront access, delivered reductions typically within 10 to 15 percent of Defence 3 across inundation extent, fatality rates and economic loss. Defence 1, the eco-hybrid, provided only limited hazard reduction, allowing broader and shallower flooding as its vegetated berms and bioswales were overtopped, and proved unsuitable as a standalone mitigation system. The ranking held across all three tsunami sources and all sea level scenarios, confirming that defence configuration and continuity, rather than sea level alone, are the principal determinants of protective performance.

Sea level rise acted primarily as an amplifier rather than a ranking-changer. Higher baseline water levels produced deeper and more extensive flooding, with isolated shallow zones merging into continuous inundation belts along the waterfront by 2100 under high emissions, particularly along inner bay margins, river corridors and reclaimed land. Yet the relative performance of the defences remained stable, and in fact Defence 2 and Defence 3 continued to improve relative to the no-defence baseline as sea levels rose, while Defence 1’s advantage fell sharply under late-century conditions. The authors stress that fatality figures are comparative indicators within a controlled framework, not predictions of real-world casualties, since evacuation behaviour, warning status, time of day and demographics all influence actual mortality but fall outside a deterministic depth-based model.

The findings carry weight because Kesennuma is not a blank slate. After 2011, the city invited public participation in reconstruction planning, unlike many neighbouring municipalities, and residents’ input produced real modifications: lowered seawall heights in certain areas, reinforced glass windows along the port seawall, and hydraulically operated barriers that rise with water ingress to preserve sightlines. Some regional media and neighbouring municipalities criticised this participatory stance as overly individualistic, arguing that negotiated reductions in defence height compromised protection. The new modelling offers a quantitative rebuttal to that criticism, demonstrating that openness and access need not come at a meaningful cost in safety when the design is engineered deliberately around those values from the start.

The study also highlights a sobering nuance: physical defences reduce economic loss far more directly than loss of life. The largest inundation reductions occurred in the port district and industrial areas where few residents live, so fatality reductions against the no-defence baseline were comparatively modest, consistent with evidence that hard structural measures are less effective at saving lives than at reducing physical damage. This suggests that closing the life-safety gap may depend as much on evacuation planning, early warning and education as on taller walls, and the authors note their framework cannot yet adjudicate between those routes.

The researchers are careful about limitations. The configurations are conceptual analytical prototypes, not construction plans; mechanical feasibility, constructability and long-term maintenance were not empirically tested, and vegetation effects were represented through surface roughness rather than stem-scale drag. The deterministic three-source approach reflects Kesennuma’s official hazard framework but does not capture the full variability a probabilistic assessment would provide, and larger or differently oriented megathrust sources could produce longer waves and longer overtopping durations that stress the defences in unmodelled ways. Future work should incorporate probabilistic modelling, cross-hazard analysis including storm surge, physical wave-tank validation, and presentation of the designs back to residents to test social acceptability. Still, the central message is striking: a transparent, socially integrated tsunami defence achieved nearly the same protection as towering concrete walls, offering coastal cities worldwide evidence that safety and human-centred design can genuinely coexist as sea levels climb.

Subject of Research: Human-centred design of multi-layered tsunami defences evaluated under sea level rise scenarios in Kesennuma, Japan

Article Title: Human-centred design approaches to tsunami mitigation under sea level rise in Kesennuma, Japan

Article References: Human-centred design approaches to tsunami mitigation under sea level rise in Kesennuma, Japan. (n.d.). https://doi.org/10.5194/nhess-26-4911-2026

Image Credits: AI Generated

DOI: 10.5194/nhess-26-4911-2026

Keywords: tsunami mitigation, sea level rise, human-centred design, Kesennuma, coastal engineering, TUNAMI-N2, multi-layered defence, Japan, disaster risk reduction, numerical modelling, seawalls, community resilience