EGU22-5839, updated on 28 Mar 2022
EGU General Assembly 2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Exploring Behavioral Determinants of Flood Insurance Adoption with Explainable Machine Learning in the Continental US

Nadja Veigel1,2,3, Heidi Kreibich2, and Andrea Cominola1,3
Nadja Veigel et al.
  • 1Chair of Smart Water Networks, Technische Universität Berlin, Straße des 17. Juni 135, 10623 Berlin, Germany
  • 2GFZ German Research Centre for Geosciences, Section Hydrology, Potsdam, Germany
  • 3Einstein Center Digital Future, Wilhelmstraße 67, 10117 Berlin, Germany
Flood insurance is a straightforward way to provide resources for ex-post recovery from the damages caused by floods and, therefore, strengthen household resilience against this type of natural hazards. The US National Flood Insurance Program is the centralized source of flood insurance in the US providing more than 5 million policies in force today. However, only less than 5% of all US households are currently insured against flood damage. Understanding the determinants of flood insurance purchase is key to support the development of future resilience strategies. Yet, the question of which household characteristics and motivations lead to flood insurance purchase is still not answered. In this work we consider flood insurance adoption at the spatial scale of census tract (unit of ~ 3000 inhabitants) as an indicator for flood resilience. We test 397 candidate features to identify relevant determinants of flood resilience in the continentall US. Our feature set predominantly includes socio-economic variables from the American Community Survey, along with the flood history, rate discounts, and home ownership. We apply an explainable Machine Learning approach based on Light Gradient Boosting Machine (LightGBM) to predict insurance coverage and estimated the SHAP values (SHapley Additive exPlanations) for each feature. SHAP values indicate the marginal contribution of each feature to the model output for every census tract. This enables us to understand how our data-driven model deducted the predictions and to reduce the initial set of candidate features to a subset of representative features that explain flood insurance adoptionWe found that insurance coverage at the whole US scale is driven by home ownership, previous flood severity and frequency, as well as financial incentives. Conversely, the impact of socio-economic background is marginal at this scale of aggregation. In other words, if a census tract experienced a very severe flood in the past, more inhabitants are insured, compared to inhabitants in census tracts with no direct experience of severe floods. The same counts for regular flooding, yet to a smaller degree. Also, people in census tracts which do not profit from their communities voluntarily implementing floodplain management strategies to acquire subsidized insurance rates are less willing to purchase private insurance. Our results overall suggest that households will get insured irrespective of their social background, if the community provides financial incentives by participating in the community rating system or has experienced severe flooding. Finallly, we identify areas prone to fluvial flooding (e.g., Lower Mississippi) with potential to improve flood resilience by community subsidization. Targeted risk communication should be aimed at urban areas with high fluctuation of inhabitants that are unaware of the flooding history.

How to cite: Veigel, N., Kreibich, H., and Cominola, A.: Exploring Behavioral Determinants of Flood Insurance Adoption with Explainable Machine Learning in the Continental US, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-5839,, 2022.


Display file

Comments on the display

to access the discussion