Buffer

A buffer produces a polygon around each input feature at a fixed distance. Points become circles. Lines become long thin polygons (think road right-of-way). Polygons grow outward (or shrink inward with a negative distance).

This page documents the Buffer step in isolation. It's a single derived-layer step and that's all it does on its own. Combine it with other steps via a derived layer's pipeline.

Inputs

• Source layer: any data layer or another derived layer.
• Distance: a number, positive (outward) or negative (inward on polygons).
• Units: meters, feet, kilometers, miles.

Output

A new layer whose geometry is ST_Buffer(source.geometry, distance) for each input row. Attributes are copied from the source row verbatim.

How to use it on its own

From Analyze (guided)

1. Open a data layer and switch to the Analyze tab.
2. Click Buffer this layer.
3. Enter the distance and unit (meters, feet, kilometers, …).
4. Click Create derived layer.

That posts a single-step buffer recipe to the portal. The job runs on PostGIS on the server — geodesic meters, not an in-browser degree buffer. The Analyze SQL builder deliberately does not offer meter buffers in DuckDB-WASM (no PROJ data in that build), so this guide is the honest path.

Share a direct link with ?tool=buffer (and &layer=<sublayerId> on multi-layer items) plus #data, for example /items/<id>?tool=buffer&layer=<key>#data. The Data tab opens Analyze with the Buffer form already expanded.

The Analyze Past runs shelf lists your earlier derived layers for this source. Re-run posts a new private copy of the same recipe; Edit & re-run opens this Buffer form again.

From a map, open the layer kebab on a data-layer source and choose Buffer… — that opens the same deep link in a new tab.

From the new-item wizard

1. Open the new-item wizard and pick Derived layer.
2. Pick a Source. The layer you want to buffer.
3. Add a Buffer step.
4. Enter the distance and pick the unit.
5. Click Save.

That's the whole single-step recipe. Nothing else is required. Derived-layer buffer is part of the always-on PostGIS analysis path; it is not gated on the optional server-heavy ANALYSIS_TIERS worker.

Example

You have a points layer of fire hydrants and want to know the area each one covers. Buffer the layer by 300 feet. The output is a polygon layer where each polygon is the 300-foot service area of one hydrant.

That's it. No dissolve. No centroid. No "first you join, then you...". Just buffer.

Notes

• Distance is geodesic. We project to a local equal-area CRS, buffer there, then reproject the result. This makes the 300-foot example actually 300 feet on the ground regardless of the layer's source CRS.
• Negative distances only do something useful on polygon inputs. A negative-buffered point becomes the empty geometry (and is dropped from the output).
• Buffered geometry follows the source on the next read. A Buffer Analyze shortcut creates a derived layer that stores the recipe and recomputes in PostGIS whenever the map or API reads it. There is no separate Refresh button on the derived-layer detail page.