Published · May 2026
Flood Map Reality Gap
79.4M daily precipitation observations · 1,099 FEMA county maps · Measuring where official flood risk has fallen behind reality
79.4M
Daily observations
5,545
NOAA stations
1,099
FEMA county maps
539
Counties scored
35%
Getting worse since map date
110
Significant trends (p<0.05)
Key Findings
🗺️
35% of counties have more extreme rain days after their map was issued
For 187 of 535 scored counties, average annual extreme precipitation days (>25mm) are higher after the FEMA map effective date than before — meaning the map was already falling behind at publication, or conditions have worsened since.
📅
The oldest maps are 27+ years out of date
Harrison County, TX and Mississippi County, AR both have FEMA maps from the late 1990s — nearly three decades old. FEMA recommends updates every 5 years. Over 60% of scored counties have maps older than 10 years.
📈
110 counties show statistically significant worsening trends
Using OLS regression on annual extreme-day counts since each county's map effective date, 110 counties (p<0.05) show a measurable upward trend in extreme precipitation — meaning the gap between map and reality is growing.
🌧️
Extreme rain days are rising nationally
The 10-year rolling average of extreme precipitation days per station has trended upward since the 1980s. The 2020s are tracking above every prior decade in the dataset.
👥
2.5M people live in medium-risk lag counties
The 39 counties in the 'medium' lag tier collectively house 2.5 million people. No counties have yet reached 'high' or 'critical' — a reflection of partial FEMA coverage (~35% of US counties), not absence of risk.
💰
Lower-income counties may face disproportionate exposure
The income vs. lag score scatter shows a weak negative correlation — counties with lower median household incomes tend to cluster at higher lag scores, suggesting an environmental justice dimension to map staleness.
Top Counties by Lag Score
| # | County | Map Age | Trend (days/yr) | Lag Score | Tier |
|---|---|---|---|---|---|
| 1 | Harrison County, TX | 27.4 yr | +0.786 | 0.493 | medium |
| 2 | Mississippi County, AR | 27.4 yr | +0.481 | 0.456 | medium |
| 3 | Anderson County, TX | 5.0 yr | +3.800 | 0.437 | medium |
| 4 | Benton County, MN | 14.6 yr | +2.264 | 0.431 | medium |
| 5 | Montgomery County, AL | 7.0 yr | +3.000 | 0.412 | medium |
| 6 | Prince Edward County, VA | 15.6 yr | +1.951 | 0.411 | medium |
| 7 | Angelina County, TX | 4.4 yr | +3.600 | 0.410 | medium |
| 8 | Rockbridge County, VA | 19.7 yr | +1.152 | 0.397 | medium |
| 9 | Swift County, MN | 19.9 yr | +1.049 | 0.387 | medium |
| 10 | Miller County, AR | 19.0 yr | +1.134 | 0.386 | medium |
Data Sources
NOAA GHCN-D
79.4M daily rows · 1840–2026
Daily precipitation per station
FEMA NFHL REST API
1,099 counties · 1998–2026
FIRM effective dates
Census ACS 5-year
3,222 counties · 2022
Population, housing, income
Census TIGER 2022
County shapefile · 2022
Spatial join for station→county
Coverage note: FEMA data covers ~1,099 of ~3,100 US counties (~35%). The lag index is computed only for counties with all three data sources. Expanding FEMA coverage would score more counties and likely push some into the high/critical tier.
Flood Map Reality Gap · Sources: NOAA GHCN-D · FEMA NFHL REST API · Census ACS 5-year 2022 · Published May 2026