HomeSources
Sources and methods
Every number on this site comes from one of six public datasets. Scripts pull them, check them against values confirmed by hand, and write the pages. Nothing is estimated by a person or a model.
Soil: USDA NRCS SSURGO
The Soil Survey Geographic Database, queried through USDA's Soil Data Access service. Survey versions on this build run to August 5, 2026 (3,379 soil survey areas). Public domain.
How the two main shares are counted
- Expansive clay share: the share of mapped land where USDA's national interpretation "Dwellings without basements" lists shrink-swell as a limiting feature with a rating of 0.5 or more on its 0-to-1 scale. USDA's rule for that feature scores the weighted-average linear extensibility between 25 and 100 cm deep. The interpretation assumes a house of three stories or less on shallow spread footings, which covers slabs and shallow foundations.
- High water table share: the share of mapped land where USDA's "Dwellings with basements" interpretation lists depth to a saturated zone as a limiting feature with a rating of 0.5 or more. USDA's rule for that feature is named for ground water between 75 and 180 cm (about 2.5 to 6 feet) below the surface.
- Weighting: each map unit's acres times each soil component's percentage of that map unit. Every component counts in the denominator, including urban land and water, which USDA doesn't rate, so a heavily built county can show lower shares than its soils alone would.
- Counties: acres come from SSURGO's own county overlap tables, which split survey areas that cross county lines. For 33 counties those tables can't be used (Connecticut's planning regions, split Alaska boroughs, survey areas without overlap acres), so we clip the Census county boundary against the soil map instead.
- Cities: each Texas and Ohio city's Census boundary (TIGERweb, simplified to within a few tens of meters) is clipped against the soil map with Soil Data Access spatial queries, and the same weights apply inside city limits.
- Also shown: the share rated "very limited" for septic absorption fields; linear extensibility (LEP), the percent change in a soil's length between moist and oven-dry, taking each component's highest value in the top 150 cm (USDA classes: under 3% low, 3-6% moderate, 6-9% high, 9% and up very high); and the soil series covering the most land.
USDA's own words
- The houses the ratings describe: "Dwellings are single-family houses of three stories or less. For dwellings without basements, the foundation is assumed to consist of spread footings of reinforced concrete built on undisturbed soil at a depth of 2 feet or at the depth of maximum frost penetration, whichever is deeper." With basements, the footings are assumed "at a depth of about 7 feet."
- Shrink-swell as a limit for houses without basements: "Excessive shrink-swell in the thickest layer between 25 and 100cm deep can cause building foundations to shift, flooring slabs to crack and roadways to buckle."
- A shallow water table as a limit: "Soils with shallow depth to a water table may become waterlogged and boggy during periods of heavy precipitation and are slow to drain."
- Linear extensibility: "the change in length of an unconfined clod as moisture content is decreased from a moist to a dry state." The National Soil Survey Handbook's classes: low under 3.0%, moderate 3.0 to 5.9%, high 6.0 to 8.9%, very high 9.0% and up.
Quoted from the interpretation descriptions in USDA NRCS Soil Data Access and the National Soil Survey Handbook, part 618.
A soil rating is a risk measure for the ground, not a record of damage. No public dataset records actual foundation failures by place. USDA's Soil Survey Manual notes that reinforced slab-on-ground construction "has markedly reduced the limitation of shrink-swell for small building construction", so a high share describes the ground, not the condition of any particular house.
Checks
The county shares reproduce the values checked by hand when this site was planned (September 28, 2026):
| Check | Ours | Checked | Result |
|---|---|---|---|
| soil shrink-swell Bexar County, TX | 51 | 51 | match |
| soil wet-basement Bexar County, TX | 0 | 0 | match |
| soil shrink-swell Dallas County, TX | 51 | 51 | match |
| soil wet-basement Dallas County, TX | 6 | 6 | match |
| soil shrink-swell Harris County, TX | 25 | 25 | match |
| soil wet-basement Harris County, TX | 53 | 53 | match |
| soil shrink-swell Cook County, IL | 25 | 25 | match |
| soil wet-basement Cook County, IL | 59 | 59 | match |
| soil shrink-swell Franklin County, OH | 12 | 12 | match |
| soil wet-basement Franklin County, OH | 71 | 71 | match |
| soil shrink-swell Kent County, MI | 0 | 0 | match |
| soil wet-basement Kent County, MI | 38 | 38 | match |
| soil shrink-swell Fulton County, GA | 0 | 0 | match |
| soil wet-basement Fulton County, GA | 11 | 11 | match |
| drought D1+ 48029 2026-09-22 | 60.7 | 60.7 | match |
| drought D1+ 48029 2026-09-15 | 32.1 | 32.1 | match |
| drought D1+ 39049 2026-09-22 | 0 | 0 | match |
| flood records 2016-2025 48201 (v3 vs v2) | 66080 | 66075 | match |
| flood records 2016-2025 17031 (v3 vs v2) | 966 | 966 | match |
| flood records 2016-2025 48029 (v3 vs v2) | 264 | 264 | match |
| flood records 2016-2025 39049 (v3 vs v2) | 245 | 244 | match |
| rain y2025_anom_in 48201 | -13.7 | -13.7 | match |
| rain y2025_anom_in 39049 | -1.69 | -1.69 | match |
| rain y2025_anom_in 17031 | -8.14 | -8.14 | match |
| homes county:48029 | 1990, 262200, 0.588 | 1990, 262200, 0.588 | match |
| homes county:39049 | 1979, 288400, 0.528 | 1979, 288400, 0.528 | match |
| homes county:17031 | 1961, 324500, 0.577 | 1961, 324500, 0.577 | match |
| homes place:3918000 | 1980, 252900, 0.441 | 1980, 252900, 0.441 | match |
| homes place:4865000 | 1985, 235700, 0.522 | 1985, 235700, 0.522 | match |
Drought: US Drought Monitor
County statistics from the Drought Monitor's data services: the percent of each county's area in each drought category, every week. We show the share in D1 (moderate drought) or worse, and the last 52 weekly maps. Newest map on this build: September 22, 2026. Maps are valid Tuesday and published Thursday. Credit: the U.S. Drought Monitor is jointly produced by the National Drought Mitigation Center at the University of Nebraska-Lincoln, the United States Department of Agriculture, and the National Oceanic and Atmospheric Administration.
Flood losses: FEMA OpenFEMA
FEMA's NFIP Redacted Claims v3 dataset (API endpoint https://www.fema.gov/api/open/v3/NfipClaims), refreshed by FEMA on September 9, 2026 (newest loss September 7, 2026) and retrieved by us on September 28, 2026 at 21:10 UTC. We count loss records with a date of loss from January 1, 2016, by county and by ZIP code, and the amount paid for building, contents and compliance costs. Foundation types come from the dataset's foundation field where it is filled in. Water depth is the exterior depth in inches. City pages count the ZIP codes whose area overlaps the city by at least 10% of the city or 25% of the ZIP, so they reach past city limits.
This product uses the Federal Emergency Management Agency's OpenFEMA API, but is not endorsed by FEMA. The Federal Government or FEMA cannot vouch for the data or analyses derived from these data after the data have been retrieved from the Agency's website(s).
Rain: NOAA
NOAA NCEI's county precipitation series (nClimDiv), monthly since 1895, through August 2026. The average year is the mean of the 1991-2020 calendar-year totals. Days with an inch or more of rain come from NOAA's 1991-2020 Climate Normals at the nearest station with 24 or more years of record within 50 km of the county's center (a shorter-record station only when none qualifies). Mountain snow sites are excluded.
Homes: Census Bureau ACS
American Community Survey ACS 2020-2024 5-year, released January 29, 2026: housing units, year built, tenure, value and units in structure. Pre-1980 means built in 1979 or earlier. City figures are not shown for places with fewer than 100 housing units.
Boundaries and places
Counties and incorporated places: Census Bureau 2026 Gazetteer files and TIGERweb boundaries. Texas and Ohio pages cover every active incorporated place. ZIP overlaps: Census 2020 ZCTA-to-place relationship file. Soil series descriptions: USDA NRCS series descriptions (OSD).
Repair costs
Cost ranges are copied from published cost guides with the page link and the page's own update date. They are ranges, not quotes.