Add X, Y, and Z samples. The tool interpolates the field, traces isolines, and builds a readable color surface automatically.
The contour surface has not been calculated yet.
| Samples | X minimum | X maximum | Y minimum | Y maximum | Z minimum | Z maximum | Z mean |
|---|
A contour plot turns a measured field into lines of equal Z value. It is useful for elevation, temperature, pressure, response surfaces, concentration maps, and any quantity that varies across two dimensions.
Use one column each for X position, Y position, and the measured Z value. CSV, TSV, TXT, JSON, XLSX, XLS, and pasted tables are processed locally in your browser.
The tool uses inverse-distance weighting to estimate a regular surface between samples. Grid resolution controls visual smoothness, interpolation power controls how local each sample feels, and the nearest-samples setting limits distant influence.
Contour lines mark exact thresholds while the color surface makes broad patterns easy to scan. Turn on sample points to compare the interpolated field with the original observations.
Adjust titles, axes, palette, opacity, line weight, level count, labels, and legend before downloading a high-resolution PNG or native SVG.
Imagine sensors placed at irregular X and Y positions, each recording temperature as Z. The contour surface estimates values between those sensors, while isolines connect locations with the same estimated temperature. Display the original sample points to see which regions are supported by nearby measurements and which are mostly interpolation.
Inverse-distance weighting gives closer samples more influence, but it does not know about walls, boundaries, wind, terrain, or other physical processes. Sparse edges and large unsampled gaps are especially uncertain. Test multiple powers, neighbor limits, and grid resolutions; a feature that disappears under modest settings may not be robust.
Use evenly spaced levels when equal numeric changes should receive equal emphasis. A sequential palette suits low-to-high measurements, while a diverging palette is appropriate only when a meaningful midpoint exists. Include units in the legend, avoid rainbow palettes that create false boundaries, and do not imply more precision than the source measurements provide.
A scatter plot can reveal the sampling layout, while a hexbin chart can summarize dense coordinate coverage. Comparing these views helps separate a real spatial pattern from one produced by uneven sampling.
Do not use interpolation for unrelated category pairs or coordinates that have no meaningful distance. A contour map can imply a continuous field even when movement between samples is impossible. If the goal is simply to compare measured X, Y, and Z records without estimating the spaces between them, use points or bubbles and keep the observations discrete.