Mismatch between household flood preparedness and objective flood risk in the Netherlands

Climate change is projected to increase flood frequency and severity, with disproportionate impacts on vulnerable populations. Adequate preparedness – understood both as being aware of the risks and as taking proactive actions to reduce them – can make the difference between a disruptive event and a catastrophic one, lessening the economic and social impacts, reducing loss of life, and preserving critical infrastructure.

As knowledge has grown regarding the identification of flood zones and the estimation of flood damages, the dissemination of risk information through publicly accessible flood maps, community outreach, and targeted communication strategies has increasingly become a core component of flood risk management in many countries. Governments and agencies aim to make risk information available to the public not only to improve awareness but also to encourage private preparedness and facilitate informed decision-making.

From an economic policy perspective, risk-based preparedness is desirable as it aligns individual behavior with efficient risk allocation. However, if risk perceptions and private adaptation fail to correlate with objective flood risk, this may compromise crucial instruments for managing flood risk, including investment in protection infrastructure and the viability of insurance schemes.

Despite advances in flood mapping and household adaptation research, the relationship between expected damages and adaptation decisions remains poorly understood. Most existing studies examine willingness to adapt in response to perceived flood risk, which is often shaped by psychological factors, personal experience, and socioeconomic characteristics rather than objective risk metrics. Understanding whether adaptation aligns with objective risk is essential but technically challenging, as household-level data on exposure, perceptions, and adaptation actions are rarely observed together.

Here, we provide novel insights into spatial patterns of household flood adaptation by combining: (i) objective household-level flood risk from publicly available street-level flood maps, (ii) household flood damages simulated using a national hydraulic model, and (iii) a large-scale survey (n > 1000) of household adaptation measures and flood risk perceptions, geolocated at the address level. We focus on the South of the Netherlands, a "best case scenario" given its accurate flood risk information and recent flood experiences.

We find a substantial mismatch between private adaptation measures and objective flood risks, as well as significant heterogeneity in risk perceptions. Simulations show that expected damages could be reduced substantially if high-risk households invested more in adaptation relative to low-risk households: Although expected damages vary by orders of magnitude, high-risk households take only slightly more protective measures than those facing little risk. Adaptation is also poorly aligned with households' flood risk perceptions, indicating that perceived danger does not reliably translate into action.

These findings reveal important limits to the effectiveness of private adaptation when left to individual decision-making and underscore the need for policies that enhance the accessibility, relevance, and actionability of flood risk information to support climate resilience.