Contour lines from a grid of elevations
Lines of equal elevation at a chosen interval from a grid of elevations, as vector lines with their lengths, by marching squares.
3 contour lines at 3 levels, 182.599 m in all.
| Elevation | Lines | Closed loops | Length |
|---|---|---|---|
| 105.000 m | 1 | 1 | 105.558 m |
| 110.000 m | 1 | 1 | 56.030 m |
| 115.000 m | 1 | 1 | 21.011 m |
- Levels
- Total length
- Lowest elevation
- Highest elevation
Provenance
- Computed by
- raster.terrain.contours 1.0.0, core 0.1.0
- Model
- Marching squares (the two-dimensional case of Lorensen and Cline 1987): in each grid square, a level crosses an edge whose corners lie on either side of it (a corner exactly at the level counts as below, as in contourpy), at the linearly interpolated point; the crossings are paired, and a square crossed on all four edges joins its high corners through the middle when the mean of its corners is above the level. Levels are the base plus whole intervals within the grid's range. Squares with a no-data corner are skipped. A geographic grid uses WGS 84 east-west and north-south degree lengths at its center latitude, and its vertices retain the grid's affine latitude-longitude placement
- Accuracy
- Contour placement is exact for the surface made of linear pieces along the grid lines. Geographic lengths use the WGS 84 degree lengths at the grid's center latitude; a real hillside between grid points is only as well known as the grid
- Notes
- None
- Cites
- Lorensen, W. E., and Cline, H. E., ACM SIGGRAPH Computer Graphics, Marching cubes: A high resolution 3D surface construction algorithm
Something look off?
How we got thisFormula, worked example, sources, and proof
Model: Marching squares (the two-dimensional case of Lorensen and Cline 1987): in each grid square, a level crosses an edge whose corners lie on either side of it (a corner exactly at the level counts as below, as in contourpy), at the linearly interpolated point; the crossings are paired, and a square crossed on all four edges joins its high corners through the middle when the mean of its corners is above the level. Levels are the base plus whole intervals within the grid's range. Squares with a no-data corner are skipped. A geographic grid uses WGS 84 east-west and north-south degree lengths at its center latitude, and its vertices retain the grid's affine latitude-longitude placement
Accuracy: Contour placement is exact for the surface made of linear pieces along the grid lines. Geographic lengths use the WGS 84 degree lengths at the grid's center latitude; a real hillside between grid points is only as well known as the grid
When to use this: Use this to turn a small grid of elevations into contour lines: a survey of spot heights on a regular grid, a DEM clip exported as numbers, or a site model, when you want lines to draw, measure, or hand to CAD rather than a picture. It gives each line's points in order with higher ground on the left, the lines and closed loops per elevation, and their lengths, so a closed loop can be read as a hill or a hollow at a glance.
Limitations: The grid must be regular, north-up, and unrotated, with elevations in meters. A meter grid is square-celled and stays in local east-south coordinates; a geographic grid uses equal latitude and longitude steps, whose ground widths differ by latitude, and also returns map coordinates. Between grid points the ground is assumed to change linearly along each grid line, so a contour cannot show a feature smaller than the grid spacing, and where a square is crossed on all four sides the choice of which way to join is a convention (the mean of the corners, as contourpy and Matplotlib use) rather than something the grid knows. Squares that touch a no-data value are left out, so lines stop at gaps. At most 300 by 300 points, 500 levels, and 100,000 contour vertices.
Worked example: 5 m contours on a geographic hill. Source: contourpy 1.3.0 gives the same three closed loops at 105, 110, and 115 m; the geographic placement follows the grid coordinates and the WGS 84 degree lengths.
You enter
- Cell size in degrees
- 0.0001
- Elevation grid
- 100, 102, 104, 102, 100 102, 106, 110, 106, 102 104, 110, 118, 110, 104 102, 106, 110, 106, 102 100, 102, 104, 102, 100
- Contour interval
- 5 m
- North-west latitude
- 40 deg
- North-west longitude
- -105 deg
You get
- Lines
- 3
- Levels
- 3
- Total length
- 182.599 m
- Lowest elevation
- 100 m
- Highest elevation
- 118 m
Review: Not yet independently reviewed by a GIS professional.
Last verified: 2026-09-24, when a maintainer last confirmed this tool's sources at the issuer. See the sources ledger.
Status: version 1.0.0, core 0.1.0. See this tool in the verification report.
Changes
- 2026-10-12, changed: The contour tool now accepts a north-up geographic grid from its north-west coordinate and angular cell size. It uses latitude-aware WGS 84 east-west and north-south dimensions, returns latitude and longitude on every contour vertex, and draws separate contour polylines on the map. Existing meter-grid results and contour topology are unchanged. Changelog
Checked against: 25 golden test vectors (download the test vectors, each with its source and tolerance). See how results are checked and every source.
Constants this tool assumes
Values you do not enter, and where each comes from.
- WGS 84 semi-major axis, for geographic grid spacing
6378137 m— Department of Defense World Geodetic System 1984- WGS 84 inverse flattening, for geographic grid spacing
298.257223563— Department of Defense World Geodetic System 1984
Sources
- Marching cubes: A high resolution 3D surface construction algorithm, Lorensen, W. E., and Cline, H. E., ACM SIGGRAPH Computer Graphics, Volume 21, issue 4, pages 163-169. The cell-by-cell case table with linear interpolation along edges, of which marching squares is the two-dimensional case.