Contour
Generate iso-lines (or iso-polygons) from a value column on a point or cell layer.
advancedA contour step takes a layer of points (or fishnet cells) with a numeric value and produces iso-lines (or iso-polygons) at specified value intervals. The output represents lines of equal value, like contour lines on a topographic map.
Use for elevation contours from a point cloud, signal-strength isobars from a measurement grid, "minutes from the nearest hospital" service zones.
Inputs
- Source layer of points or fishnet cells.
- Value field: the numeric attribute to contour on.
- Mode: - Iso-lines. Output is line geometry at each level. - Iso-polygons. Output is filled polygon bands between levels.
- Levels: a list of values (
100, 200, 300, ...) or an interval (every 50). - Smoothing: how much to smooth the result. 0 = strict mathematical contours from the input grid; higher values produce visually smoother but slightly less precise lines.
Output
A new layer of line (or polygon) geometry with attributes:
level: the value of the contour line / lower edge of the polygon band.level_top: the upper edge (iso-polygons only).- Aggregates from the original layer (mean, count) per band.
How to use it on its own
- Build a value layer first: typically points-with-a-value or a fishnet-binned summary.
- New-item wizard → Derived layer.
- Pick the value layer as Source.
- Add a Contour step.
- Pick the value field, mode, levels, and smoothing.
- Save.
Example: drive-time service zones
- Start with a point grid (a fishnet's cell centroids).
- For each cell, calculate drive time to the nearest hospital (an external routing tool, or a built-in approximation).
- Contour on
drive_time_minutes, iso-polygons, levels at5, 10, 15, 20, 30. Output is a polygon layer with bands you can render as a heatmap-style service-area map.
Notes
- Input density matters. Coarse input grids produce jagged contours; fine grids produce smooth contours at higher cost. For elevation data, expect to use a point cloud or DEM-derived fishnet at meter-level cell size.
- Edges are interpolated linearly between adjacent input values. The contours are continuous within the input's coverage and stop at the input's bounding box.
- Run cost. Contour is expensive on large inputs. Run as a derived layer with scheduled refresh (nightly), not as an on-the-fly step.