This brief looks at three estimates of the characteristics of public school students who are likely to be at risk of experiencing flooding at their school or in their neighborhoods. The three estimates are (1) zip code–level estimates of flood risk from First Street’s 2.1 model, (2) census tract–level estimates of flood risk from the Federal Emergency Management Agency’s (FEMA’s) flood risk indicators, and (3) school site–specific estimates of flood risk from our analysis of the FEMA National Flood Hazard Layer database.

Why This Matters

Government estimates suggest that the number of properties at risk of flooding will increase by about 10 percent over the next 30 years. Floods in and around schools can prove particularly disruptive to education. More than 50 percent of all school closure days from 2011 to 2019 were attributable to natural disasters, including flooding. River and creek flooding caused 111 prolonged school closures during this period (prolonged closures are five days or longer), and rain and hurricanes caused 10,501 prolonged closures.

Key Takeaways

• By any measure, a small but substantial share of American public school students are predicted to have elevated flood risk. Our measures estimate that 2.3 to 10.2 percent of public school students are at high risk or relatively high risk of flooding. This translates to between 1.1 million and 5.1 million students at risk, depending on measure type.
• Flood risk varies by student demographics within each measure, but students from low-income families are always more at risk. Across all three measures, students from households below the federal poverty level are substantially more at risk of experiencing flooding at their school or in their neighborhood. American Indian and Alaska Native students and Native Hawaiian and Pacific Islander students also stand out as having more risk across the three measures.
• School building–specific flood risks can be different than their corresponding neighborhood flood risks. At the state level, our estimate of school building site risk is positively but not completely correlated with estimates at the census tract or zip code level (r = 0.657 and r = 0.585, respectively). Some states have higher or lower aggregate estimates of school building risk relative to aggregate school neighborhood risk, though it is difficult to assess how much of this difference is attributable to differences in the available measures. Ideally, school building sites should be located out of flood risk areas whenever possible, as schools often serve as shelter sites during natural disasters. State policy could be strengthened to better support school building resilience by supporting school siting decisions and by prioritizing state capital revenue for projects that support hardening educational infrastructure against climate change.

How We Did It

We use data on public school location and student enrollment, linked to three types of geographic flood data, to understand which students are more at risk of flooding exposure. In lieu of using outdated enrollment zones, we opt for measures of risk in the zip code where a public school is located or in the census tract where a public school is located.