pourpoint

pourpoint is the Python package for the pourpoint watershed delineation engine. Give it a point on a river and it returns the whole upstream watershed as a polygon. It reads HFX datasets, which are folders of pre-built river-network files. Only HFX v0.3.0 datasets load; older versions report a clear unsupported format error.

The wheel bundles the full native stack, including GDAL, PROJ, GEOS, libtiff, SQLite, and more, so no system GDAL install is needed.

Install

uv add pourpoint

(or pip install pourpoint)

Prebuilt wheels are published for:

• macOS (Apple Silicon + Intel)
• Linux (x86_64 + aarch64)
• Windows (x86_64)

as macosx_11_0_arm64, macosx_11_0_x86_64, manylinux_2_28_x86_64, manylinux_2_28_aarch64, win_amd64.

Zero-download quickstart

Use the hosted public GRIT (Global River Topology) dataset without downloading it first:

import pourpoint

# No local dataset: this reads the hosted GRIT dataset over the network.
engine = pourpoint.Engine("https://basin-delineations-public.upstream.tech/grit/hfx-v0.3.0/")
result = engine.delineate(lat=47.3769, lon=8.5417)
print(result.area_km2)

The engine fetches only the pieces of the dataset it needs, so the full dataset never lands on your machine. The one public dataset hosted today is GRIT 2.0.0 (the source river network), compiled to HFX v0.3.0 (the format version). To use a different HFX dataset, change the URL and nothing else in your code changes.

Local quickstart

import pourpoint

engine = pourpoint.Engine("/path/to/hfx/dataset")
result = engine.delineate(lat=47.3769, lon=8.5417)
print(result.area_km2)

Engine accepts local paths and remote dataset URLs such as s3:// and https://. For constructor options such as snap search radius and geometry repair, see the Tuning Knobs section of API.md.

Reuse the Engine

The first open fetches dataset metadata over the network and is slower, so keep the engine around and reuse it. Repeated delineations in the same session reuse data already fetched, so overlapping watersheds are faster.

Going further

Logging and verbose output, batch delineation with a progress callback, the staged step-by-step API, and GeoParquet export are documented in API.md and on the docs site at https://cooperbigfoot.github.io/pourpoint/.

What it does

• Finds the catchment your point sits in.
• Gathers every catchment upstream.
• Merges them into one watershed polygon.
• Returns the geometry plus geodesic area in km².

API Reference

For the full developer-oriented API surface, including argument types, return types, and the exception hierarchy, see API.md.

Links

• Source & issues: https://github.com/CooperBigFoot/pourpoint
• HFX dataset spec: https://github.com/CooperBigFoot/hfx
• License: MIT for pourpoint; bundled native libraries retain their own licenses. See LICENSES/.