← Back to Surroundings

References

Surroundings helps you compare property locations for early research.

Population and population density

Population data: © WorldPop, University of Southampton. The global map layers use Global2 R2025A v1 global population counts for 2016, 2021, and 2026, constrained and UN-adjusted, at 30 arc-second resolution (approximately 1 km at the equator).

Dataset credit: Bondarenko and colleagues, WorldPop Global Demographic Data Project, University of Southampton. Dataset DOI: 10.5258/SOTON/WP00845. Licensed under Creative Commons Attribution 4.0 International. See the release statement for the source methodology.

Surroundings modifies these count rasters by converting counts to people per square mile using geodesic cell areas and subtracting historical density from 2026 density on matching native grids. Low-zoom views aggregate differences; high-zoom views preserve native cells. Both dates must have valid data. Missing values remain transparent rather than becoming zero. Coverage follows the source extent, approximately 60°S–84°N.

The density layer shows the selected year: 2016, 2021, or 2026. Change layers compare 2016–2026 or 2021–2026 and saturate at ±2,000 people per square mile. Density colours use a logarithmic scale from 1 to 30,000 people per square mile. These are modeled estimates and projections, including the 2026 surface.

The selected-layer results sum native population counts inside each existing isochrone, weighted by fractional pixel coverage. Net change subtracts the starting population from 2026 population using only cells valid in both years. Positive values indicate growth; negative values indicate decline. Percent change is unavailable when the starting population is zero. These cumulative catchment totals overlap and should not be added together. The original saved-report statistics are retained separately.

New reports use 2026 population estimates. Historical reports omit current income enrichment. Any legacy 2001 reports use the Global1 series, whose methodology differs from Global2; it is excluded from the density-change layers.

Maps, addresses, and travel-time calculations

Mapbox supplies the interactive basemap, address lookup, and routing used by this app. Base-map data includes © OpenStreetMap contributors. The map retains the provider's required logo and attribution links.

Driving catchments use the Mapbox Isochrone API. Enter an address anywhere routing and data coverage are available to estimate the area reachable by car in 15, 30, 45, and 60 minutes. Historical transit examples use TravelTime. Accessible population is calculated by summing raster population within each catchment with fractional pixel coverage.

Travel times are routing estimates rather than guaranteed journeys. Compare reach scores using the same travel mode, timing, and time bands. The density-change layers compare population surfaces; they do not reconstruct historical networks, historical traffic, or historical timetables. Catchment geometry is held fixed when comparing historical population.

Income estimates

Area Income estimates total household income within a catchment. We use ACS aggregate household income (B19025), which equals mean household income times the number of households.

We distribute each tract’s total using WorldPop for the corresponding year, with covered area as the fallback when there is no positive population. The map shows income per square mile. Historical comparisons use the 2016, 2021, and 2024 ACS releases and convert amounts to constant 2024 USD using the annual US CPI-U.

Older reports may show a median-based Area Income proxy, calculated as tract median household income times estimated households.

Catchment results apply identical isochrone geometry and fractional cell coverage to both years. Only cells with data for both years contribute to the two totals and their difference. Results are cumulative for each travel time. The ACS windows overlap, so this modeled comparison is descriptive. It combines changes in income estimates, households, population allocation, and tract geography.

Housing units within reach

Housing map layers use ACS estimates from 2012–2016, 2017–2021, and 2020–2024. The 3- and 8-year change layers subtract the 2021 or 2016 estimate from the 2024 estimate over shared coverage. Housing map density is distributed by area on a common grid. The change scale reaches its strongest colors at a gain or loss of 1,000 housing units per square mile.

US housing-unit estimates come from ACS 2020–2024 table B25001. The total includes occupied and vacant units, including apartments and single-family homes. We allocate each tract's total by the share of its area inside the catchment, using 2024 Census cartographic tract boundaries and an equal-area projection. The estimates assume uniform distribution within each tract. They are cumulative by travel time and describe the covered US portion of a catchment.

Housing-market context

ZIP listing figures reflect the month covered by the listing data. US listing and inventory context uses Realtor.com research data. House-price trends use Federal Housing Finance Agency House Price Index data where available. Use these area-level estimates for comparison and early research. A property appraisal requires a separate valuation.

Saving and sharing reports

Premium reports are saved to your account. Reports created before signing in stay on the original device and are available when signed out. New analyses require an active subscription.

Analysis links share the location’s coordinates so Premium subscribers can run a fresh analysis. The saved report stays in your account or on your device.

References | Surroundings