About the data
Yuva compares every US town and city on the things that decide whether you’d like living there. Here is exactly where each number comes from.
Sources
Monthly average high / low temperature, precipitation, snowfall and threshold day-counts from weather stations.
License: US Government work — public domain · Pipeline:
scripts/data/ingest/noaa.ts (nearest station with an elevation penalty)Monthly sunshine (clear-sky index), dew point and wind speed from the MERRA-2 reanalysis, 1991–2020.
License: NASA open data · Pipeline:
scripts/data/ingest/power.ts (0.5° grid, cached)Days per year in each AQI category, median and 90th-percentile AQI, and dominant pollutants.
License: US Government work — public domain · Pipeline:
scripts/data/ingest/epa.ts (10 years; nearest monitored county for places without a monitor)Population, age, income, home value, rent, education, commuting and household characteristics for each place.
License: US Government work — public domain · Pipeline:
scripts/data/ingest/census.tsPlace names, coordinates, elevation (SRTM), time zones and county codes.
License: CC BY 4.0 · Pipeline:
bundled via the cities-1000-structured packageCounty-level risk scores (0–100) for 18 natural hazards plus a composite, combining expected loss, social vulnerability and community resilience.
License: US Government work — public domain · Pipeline:
scripts/data/ingest/fema.ts (FEMA’s ArcGIS feature service; each place inherits its county’s scores)Block-group walkability scores (1–20) from street connectivity, destination density and transit access.
License: US Government work — public domain · Pipeline:
scripts/data/ingest/sld.ts (population-weighted average of the census tracts around the place centre)Boundaries of national and state parks, national forest, BLM, wildlife areas and other protected public land, with who manages them and whether the public can get in. Used: fee and designation areas of 200+ acres, protection status 1–3, open or permit-restricted, public managers only.
License: US Government work — public domain · Pipeline:
scripts/data/ingest/outdoor.ts (USGS ArcGIS feature service, rasterised to a ~1 km grid, share of land within 10 and 30 miles)Worldwide elevation, blended from SRTM, USGS 3DEP and other public DEMs, served as map tiles. Used for the elevation spread around a place and to tell land from ocean.
License: Open data (SRTM, USGS 3DEP and other public-domain / open DEMs; see the registry page for each source) · Pipeline:
scripts/data/ingest/outdoor.ts (zoom-7 tiles, 5th–95th percentile elevation within 30 miles)Violent and property offences reported by each police department, per 100,000 residents.
License: US Government work — public domain · Pipeline:
scripts/data/ingest/fbi.ts (city police department matched to the place; needs a free api.data.gov key)8th-grade math and reading proficiency for each school district, joined to the place by county and district office location.
License: Public domain (US Department of Education) — redistributed by the Urban Institute · Pipeline:
scripts/data/ingest/nces.tsTop marginal state income-tax rate and combined state + average-local sales-tax rate for every state.
License: Facts, cited. Refresh yearly with `pnpm data:taxes` · Pipeline:
scripts/data/refresh-state-taxes.ts → src/data/state-taxes.ts- Computed by Yuva
Day counts from monthly normals, Köppen climate class, comfort score, daylight (astronomical), distances to coast and cities.
License: Methodology documented on the data page · Pipeline:
src/data/climate.ts
How we compute things
Weather is monthly normals. Every temperature, rain, snow, sunshine, dew-point and wind figure is a 30-year monthly average, so it shows the typical shape of the year rather than any one season’s weather. Filters can target the year, a season or any single month.
Sunshine hours are the hours between sunrise and sunset (computed from latitude, day by day) multiplied by the share of possible sunshine for that month — the same definition NOAA uses for “percent of possible sunshine.” Summing the twelve months gives hours per year. In live data the sunshine share is derived from a satellite/reanalysis clear-sky index rather than a sunshine recorder, so treat the hours as a good estimate, not a measurement.
Day counts (hot days, freezing nights, muggy days, comfortable days) come from NOAA’s threshold counts where a nearby station reports them, and otherwise are estimated from the monthly averages using a normal-distribution model of day-to-day swings.
Air quality is the EPA’s annual AQI summary for the county. Places without a monitor use the nearest monitored county and are flagged with the distance. Wildfire smoke can push a single day far above these annual figures.
Limited monitoring. Monitors don’t always report all year. The EPA treats a year as complete when at least 75% of days (274) have readings, so when the latest year falls short we label the place’s air quality “Limited monitoring” and show how many days were measured. The numbers are still the real readings, but they may miss part of the year, so use them as a guide, not a full-year average.
Outdoor access asks how much open public land and how much mountain is within reach. For each place we measure, on a roughly 1 km grid, the share of land within 10 and 30 miles that is protected public land (USGS PAD-US: national and state parks, national forest, BLM, wildlife areas of 200+ acres, open or permit-restricted, never private or tribal land), and the spread of elevation within 30 miles (the 5th to 95th percentile, so one lone peak does not decide it). Those three measures are ranked against every place, blended (30-mile public land 35%, 10-mile 25%, relief 40%) and ranked again, so a score of 90 means more outdoor access than 90% of places. It does not yet include trail mileage, ski areas or water, so lake and river towns can score lower than they feel; lakes count as land and the ocean does not.
Comfort score is a transparent heuristic: highs of 65–85°F, dew points below 60°F and light rainfall score highest. It is a guide, not a scientific index.
Climate type uses the Köppen–Geiger rules of Peel et al. (2007) applied to each place’s own monthly normals. Distance to ocean is measured from the Natural Earth 1:50m coastline; the Great Lakes do not count as ocean.
Match scores from the quiz are weighted averages of how well each place fits your preferences, with a penalty when an important preference fails badly. Dealbreakers (budget, air quality, regions, tax) are hard filters.
Every property
All of these can be filtered, sorted and shown on the map.
Temperature
- Average daily highmonthly · averaged
- Mean of the daily high temperature (1991–2020 normals). (NOAA U.S. Climate Normals 1991–2020)
- Average daily lowmonthly · averaged
- Mean of the daily low temperature (1991–2020 normals). (NOAA U.S. Climate Normals 1991–2020)
- Average temperature
- Annual mean temperature. (Computed by Yuva)
- Seasonal swing
- Warmest month minus coldest month (mean temperature). Small = same weather all year; large = four distinct seasons. (Computed by Yuva)
- Days 90°F or hotter
- Average number of days per year where the high reaches 90°F. (NOAA U.S. Climate Normals 1991–2020)
- Freezing nights
- Average number of days per year where the low reaches 32°F. (NOAA U.S. Climate Normals 1991–2020)
- Days that never thaw
- Average number of days per year where even the high stays at or below 32°F. (NOAA U.S. Climate Normals 1991–2020)
- Comfortable days
- Days per year with a high between 65–85°F and a dew point under 62°F — no sweating, no shivering. (Computed by Yuva)
Rain & snow
- Annual precipitationmonthly · summed
- Total rain + melted snow per year. (NOAA U.S. Climate Normals 1991–2020)
- Annual snowfallmonthly · summed
- Total snowfall per year. (NOAA U.S. Climate Normals 1991–2020)
- Wet days
- Days per year with at least 0.1 inch of precipitation. (NOAA U.S. Climate Normals 1991–2020)
Sun & humidity
- Sunshinemonthly · averaged
- Share of possible sunshine that actually reaches the ground (clear-sky index). (NASA POWER climatology)
- Sunshine hours per yearmonthly · summed
- Hours of actual sunshine in a typical year: the hours between sunrise and sunset, times the share of that daylight the sun is actually out. (Computed by Yuva)
- Average dew pointmonthly · averaged
- The best single measure of how humid it feels: under 55°F is dry and crisp, above 65°F is muggy. (NASA POWER climatology)
- Muggy days
- Days per year with a dew point of 65°F or higher. (Computed by Yuva)
- Average wind speedmonthly · averaged
- Mean wind speed at 10 m. (NASA POWER climatology)
Air quality
- Median AQI
- Median daily Air Quality Index for the most recent year. 0–50 is good, 51–100 moderate. (EPA Air Quality System — annual AQI by county)
- Good air days
- Share of days with an AQI of 50 or less. (EPA Air Quality System — annual AQI by county)
- Unhealthy-air days
- Days per year with AQI above 100 (unhealthy for sensitive groups or worse). (EPA Air Quality System — annual AQI by county)
- Worst day AQI
- Highest daily AQI of the year — usually a wildfire-smoke or ozone episode. (EPA Air Quality System — annual AQI by county)
- 90th-percentile AQI
- AQI on a typical bad day (only 10% of days are worse). (EPA Air Quality System — annual AQI by county)
- Days led by fine particles
- Share of days where PM2.5 (smoke, soot) was the main pollutant. (EPA Air Quality System — annual AQI by county)
- Days led by ozone
- Share of days where ozone (smog) was the main pollutant. (EPA Air Quality System — annual AQI by county)
- Air-quality trend
- Change in median AQI between the first and last three years on record. Negative = getting cleaner. (EPA Air Quality System — annual AQI by county)
- Distance to an air monitor
- How far the air-quality data was measured from the place. Bigger = less certain. (EPA Air Quality System — annual AQI by county)
Housing
- Median home value
- Median value of owner-occupied homes. (U.S. Census Bureau — American Community Survey 5-year)
- Median rent
- Median gross rent (including utilities) per month. (U.S. Census Bureau — American Community Survey 5-year)
- Home price ÷ income
- Median home value divided by median household income. Under 3 is very affordable; over 5 is stretched. (Computed by Yuva)
- Renter households
- Share of households that rent. (U.S. Census Bureau — American Community Survey 5-year)
- Vacant homes
- Share of housing units that are vacant (includes vacation homes). (U.S. Census Bureau — American Community Survey 5-year)
- Median year homes were built
- Older = historic housing stock; newer = suburban growth. (U.S. Census Bureau — American Community Survey 5-year)
Economy & education
- Median household income
- Median annual household income. (U.S. Census Bureau — American Community Survey 5-year)
- Poverty rate
- Share of residents below the federal poverty line. (U.S. Census Bureau — American Community Survey 5-year)
- Unemployment rate
- Share of the labour force that is unemployed. (U.S. Census Bureau — American Community Survey 5-year)
- College-educated adults
- Share of adults 25+ with a bachelor’s degree or higher. (U.S. Census Bureau — American Community Survey 5-year)
- Employment rate
- Share of everyone 16 and over who has a job. Low values often mean a retirement community or a college town. (U.S. Census Bureau — American Community Survey 5-year)
- Median worker earnings
- Median annual earnings of people who work (individuals, not households). (U.S. Census Bureau — American Community Survey 5-year)
- Population growth (5 yr)
- Change in population between the two most recent non-overlapping ACS 5-year windows. Growing places usually have a growing job market — and rising housing costs. (U.S. Census Bureau — American Community Survey 5-year)
Schools
- Math proficiency (8th grade)
- Share of 8th graders scoring proficient on the state math test in the school district serving the place. States set their own tests and bars, so compare places within a state rather than across states. (EDFacts proficiency via the Urban Institute Education Data Portal)
- Reading proficiency (8th grade)
- Share of 8th graders scoring proficient on the state reading / language-arts test in the school district serving the place. Compare places within a state rather than across states. (EDFacts proficiency via the Urban Institute Education Data Portal)
Safety
- Violent crime rate
- Murder, rape, robbery and aggravated assault reported to the FBI per 100,000 residents in a year, for the local police department. Reporting practices differ between agencies, so treat small gaps as noise. (FBI Crime Data Explorer — summary reporting)
- Property crime rate
- Burglary, larceny and motor-vehicle theft reported to the FBI per 100,000 residents in a year, for the local police department. (FBI Crime Data Explorer — summary reporting)
Natural hazards
- Overall natural-hazard risk
- FEMA National Risk Index composite, 0–100 against every US county (higher = more expected loss from natural hazards, adjusted for how vulnerable and resilient the community is). County-level. (FEMA National Risk Index)
- Flood risk
- FEMA National Risk Index score (0–100) for river and coastal flooding. County-level. (FEMA National Risk Index)
- Wildfire risk
- FEMA National Risk Index score (0–100) for wildfire. County-level. (FEMA National Risk Index)
- Hurricane risk
- FEMA National Risk Index score (0–100) for hurricanes. County-level. (FEMA National Risk Index)
- Tornado risk
- FEMA National Risk Index score (0–100) for tornadoes. County-level. (FEMA National Risk Index)
- Earthquake risk
- FEMA National Risk Index score (0–100) for earthquakes. County-level. (FEMA National Risk Index)
- Heat-wave risk
- FEMA National Risk Index score (0–100) for heat waves. County-level. (FEMA National Risk Index)
Taxes
- Property tax rate
- Median real-estate tax paid by owners ÷ median home value: the share of a home’s value you pay in property tax every year. (U.S. Census Bureau — American Community Survey 5-year)
- Top state income-tax rate
- Highest marginal state tax rate on wages (0 = none). Most households pay less than this because lower brackets apply to the first dollars. Local income taxes are not included. (Tax Foundation — state tax tables)
- Sales tax (state + local)
- State sales-tax rate plus the average local rate in that state. Your exact rate depends on the city and county. (Tax Foundation — state tax tables)
Outdoors
- Outdoor access
- How much open public land and mountain terrain is within reach, ranked 0–100 against every place here: 90 means more outdoor access than 90% of them. It blends public land within 10 and 30 miles with the elevation spread around the place; 90+ is an outdoor hub (think Bend, Boulder, Missoula), under 30 means little nearby to explore. Trail mileage and water are not included yet. (Computed by Yuva)
- Public land within 10 miles
- Share of the land within 10 miles that is open public land: national and state parks, national forest, BLM, wildlife areas and other protected public land (permit or fee areas included, closed and private land not). Lakes count as land; ocean does not. (USGS Protected Areas Database of the United States (PAD-US))
- Public land within 30 miles
- Share of the land within 30 miles (about a 45-minute drive) that is open public land: parks, national forest, BLM, wildlife areas and other protected public land. (USGS Protected Areas Database of the United States (PAD-US))
- Mountain relief within 30 miles
- Elevation spread of the land within 30 miles, from the low ground to the high ground (5th to 95th percentile, so one lone peak does not skew it). Under 500 ft is flat; 3,000+ ft means real mountains on the doorstep. (AWS Terrain Tiles (Terrarium))
People
- Median age
- Median age of residents. (U.S. Census Bureau — American Community Survey 5-year)
- Residents 65+
- Share of residents aged 65 or older. (U.S. Census Bureau — American Community Survey 5-year)
- Residents under 18
- Share of residents under 18 — a proxy for how family-oriented a place is. (U.S. Census Bureau — American Community Survey 5-year)
- Young adults (18–24)
- Share of residents aged 18–24. Very high values usually mean a college town. (U.S. Census Bureau — American Community Survey 5-year)
- Diversity index
- Chance (0–1) that two random residents are of different race / ethnicity. (U.S. Census Bureau — American Community Survey 5-year)
- Foreign-born residents
- Share of residents born outside the US. (U.S. Census Bureau — American Community Survey 5-year)
Getting around
- Average commute
- Mean one-way travel time to work. (U.S. Census Bureau — American Community Survey 5-year)
- Work from home
- Share of workers who work from home. (U.S. Census Bureau — American Community Survey 5-year)
- Take public transit
- Share of workers commuting by public transit. (U.S. Census Bureau — American Community Survey 5-year)
- Walk or bike to work
- Share of workers who walk or bike to work. (U.S. Census Bureau — American Community Survey 5-year)
- Walkability index
- EPA National Walkability Index, 1–20: how many destinations are close, how connected the streets are and how much transit is nearby. 15+ is very walkable, under 6 is car-dependent. (EPA Smart Location Database — National Walkability Index)
- Homes with broadband
- Share of households with a broadband internet subscription (cable, fibre, DSL, satellite or fixed wireless). (U.S. Census Bureau — American Community Survey 5-year)
Location & size
- Population
- Residents within the city / town limits. (U.S. Census Bureau — American Community Survey 5-year)
- Population density
- People per square mile of land within the place limits. (U.S. Census Bureau — American Community Survey 5-year)
- Elevation
- Height above sea level. (GeoNames)
- Distance to the ocean
- Straight-line distance to the nearest Atlantic, Pacific or Gulf shoreline. (Computed by Yuva)
- Distance to a big city (metro of 900k+)
- Straight-line distance to the principal city of the nearest metro area of about a million people (900,000 or more; 0 if you live in one). Metro size, not city-limits size, so Salt Lake City counts. (Computed by Yuva)
- Distance to a mid-size city (50k+)
- Straight-line distance to the nearest place with at least 50,000 residents (0 if you live in one). (Computed by Yuva)
- Longest day
- Hours of daylight at the summer solstice. (Computed by Yuva)
- Shortest day
- Hours of daylight at the winter solstice. (Computed by Yuva)