Value Skyline
See what land is worth, in three dimensions
Overview
Value Skyline draws every parcel in the current view as a 3D prism, streamed live from the landrecords.us parcel vector tile layer. How tall it stands is its total assessed value; what colour it takes is its value per square foot of land. Reading the two together separates a big cheap tract from a small expensive one — a warehouse on twenty acres and a corner lot downtown can carry the same total value while sitting at opposite ends of the ramp.
It works anywhere the nationwide parcel layer has assessed values, not just in cities. Both scales rescale to whatever is on screen, so a rural township reads with the same contrast as a downtown core.
How to use
Search an address, or zoom in
Type an address or place into the search box and pick a suggestion — the map flies there, drops a pin, and opens the lot underneath it. Pressing Enter takes the first suggestion. Otherwise just pan and zoom: parcels render at zoom 14 and above — the tile layer serves nothing below that — and the map asks you to zoom in. The chip at the top names the county under the centre of the view and links to its coverage page.
Tilt and orbit
Right-drag (or two-finger drag on a trackpad, two-finger twist on touch) to change the pitch and bearing. The skyline only reads from an angle. The compass in the top-right resets to north and flat.
Read the two encodings
Height is total assessed value, log-scaled from the cheapest parcels in view up to the single most valuable one — only that parcel touches full height. Colour is value per square foot on the magma ramp. Both rescale to what is on screen, so a rural township reads with the same contrast as a downtown core. Very tall or very large parcels draw slightly translucent, so the short parcels behind them stay visible.
Drag the legend to filter
Drag across the colour bar to keep only the lots inside that band of dollars per square foot; everything else disappears from the map. It is the fastest way to pull one value tier out of a dense field — the cheapest land near an expensive core, say. Click the bar once, or use the reset link, to clear it.
Click a lot
The popup gives the address, city, land use where the county files a readable one, total assessed value, lot size, and the resulting dollars per square foot.
How the numbers are built
One prism is one parcel record.The hosted tile layer carries each county record as its own feature, so a condominium — filed as one record per unit, every one on the building's shared footprint — draws once per unit. The statistics in the panel are computed the same way, per record, so the legend always describes the map that is actually on screen.
Lot area comes from the geometry. Value per square foot divides total value by calcarea, the computed area of the parcel polygon, rather than the filed taxacres, which carries systematic outliers.
The scales are derived from the viewport. Colour snaps its domain to the 5th–95th percentile of dollars per square foot in view, rounded to 1-2-5 steps so the legend holds still as you pan. Height runs from the 5th percentile of total value up to the most valuable parcel in view — the exact maximum, so a landmark is never clipped to look like its neighbours.
Where it comes up empty
Assessed value is filed by the county, and not every county files it. Maricopa County, Arizona is the largest example — a downtown Phoenix viewport has around 18,000 mapped parcels and not one carries a value. Rather than showing an empty map, the panel says how many parcels are mapped there and names the county whose assessor data is missing.
Data layers
Parcel boundaries and 90 attribute columns — assessed value, computed lot area, address, land use — streamed as Mapbox vector tiles from the hosted web API. The View API button in the top bar shows the exact integration.
The viewport statistics — counts, medians, and the colour and height scale anchors — are computed live against the same nationwide parcel table the tiles are rendered from.
Address and place search, proxied server-side and biased toward the current map centre. The same geocoder behind the Turtle Release Locator and Terra Firma.
Sharing
The map URL encodes the camera, not just the location, so a shared link reproduces the shot you were looking at. The parameters are lon, lat, z, pitch and bearing.