Concave Hull Generator
Upload any GeoJSON, KML, or GPX file and compute a concave hull (alpha shape) — a tight-fitting polygon around your features. Adjust the max edge length to control how tightly the hull wraps.
Drop file here or click to browse
GeoJSON, KML, GPX — any geometry type
Smaller values = tighter hull. Larger values = looser, approaching convex hull. If generation fails, try increasing this value.
Powered by Turf.js and GDAL WebAssembly. All processing happens in your browser.
What is a concave hull?
A concave hull (also called an alpha shape) is a polygon that closely follows the outline of a set of features, allowing the boundary to curve inward where the data has indentations or gaps. Unlike a convex hull — which is always the tightest convex shape — a concave hull can represent non-convex distributions much more accurately.
The tightness of the concave hull is controlled by a max edge length parameter: smaller values produce tighter outlines that follow the data closely; larger values produce looser outlines that approach the convex hull. The algorithm is based on Delaunay triangulation — edges longer than the threshold are removed, and the remaining boundary forms the concave hull.
Common use cases
Urban extent mapping
Outline the built-up extent of a city from building footprint data, following street corridors and excluding undeveloped land.
Species distribution boundaries
Approximate the range boundary of a species from occurrence point data, producing a tighter outline than a convex hull.
Cluster delineation
Draw boundaries around identified point clusters that follow the actual cluster shape, including concave regions.
Survey area outlining
Automatically generate the boundary polygon of a field survey from GPS trackpoints, excluding gaps where no data was collected.
Frequently asked questions
What is the max edge length parameter?
The max edge length controls how tightly the hull wraps around your features. A smaller value produces a tighter, more detailed outline but may create holes or disconnected shapes if the data is sparse. A larger value produces a looser hull closer to the convex hull. Start with a value roughly equal to the typical gap between your input points.
How is a concave hull different from a convex hull?
A convex hull cannot curve inward — it is always the tightest convex shape possible. A concave hull (alpha shape) allows indentations, following the outline of the dataset more closely. This makes it more useful when the data has an irregular or non-convex distribution.
What happens if my max edge length is too small?
If the max edge length is smaller than the typical spacing between points, the algorithm may produce multiple disconnected polygons or leave gaps in the hull. Increase the max edge length if you see unexpected holes or fragmented output.
Can the concave hull produce holes?
Yes — if the data has interior gaps larger than the max edge length, the algorithm will leave holes in the output polygon. This can be useful (e.g., representing a donut-shaped dataset) or undesirable depending on your use case.