Skip to content
This repository has been archived by the owner. It is now read-only.
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

ARCHIVED - U.S. National Hydrography Dataset Network & Barrier Analysis Tools

Build Status Coverage Status


This library has been archived as a stand-alone project and is no longer being actively developed.

All NHD-related processing for the Southeast Aquatic Barrier Prioritization Tool has been subsumed into


This library was used to help perform network connectivity analysis for the Southeast Aquatic Barrier Prioritization Tool. See for more information about specific processing for that project.

This library is intended to provide more general processing utilities to assist with analyzing connectivity using the National Hydrography Dataset (NHD) - High Resolution Plus version.

We do not currently intend to add support for the NHD - Medium Resolution dataset. Pull requests are welcome to add this functionality.

Due to the large size of NHD data, it may be possible to only process a single region at a time, or a group of regions. The key limits are based on the amount of available memory (RAM) and the file sizes of the outputs (shapefiles are limited to 2 GB in size).

Key features:

  • preprocessing utilities to prepare NHD data for analysis within this library
  • merging of NHD flowlines between adjacent basins or regions
  • automatic snapping barriers to nearest flowlines, including heuristics to aid with manual QA/QC
  • cutting of NHD flowlines at barriers
  • construction of functional upstream networks from a barrier to the next upstream barriers or origins of a stream network
  • network statistics
  • optimized data I/O using the feather file format for intermediate data products and customized serialization / deserialization of spatial data


  • reading / writing shapefiles using Geopandas can be very slow. We preferred to use the feather format for intermediate files (geofeather provides spatial compatibility).
  • data from NHD are downloaded as ArcGIS File Geodatabases. While these formats can be read (usually) using Geopandas, it is not possible to write this format, so shapefile outputs are generally the only option for use in GIS.


pip install nhdnet

This project uses GeoPandas, Pandas, rtree, and shapely in Python 3.6+.

We do not intend to support Python < 3.6.

Due to the complexity of these libraries, installation instructions for your platform may vary from the following.

rtree first requires the separate installation of libspatialindex. On MacOS:

brew install spatialindex

Python dependencies and virtual environment are managed using pipenv.

pipenv install

If you do not wish to use pipenv, see the Pipfile for the list of dependencies.

Source Data

NHD High Resolution data are downloaded by HUC4 from NHD Data Distribution Site.


This project uses black for autoformatting and pylint for linting.


Known issues:


This project was made possible in partnership with the Southeast Aquatic Resources Partnership as part of a larger project to develop a comprehensive inventory of aquatic barriers in the Southeastern US and assess impacts of these barriers on aquatic systems.

The results of this project are available in the Southeast Aquatic Barrier Prioritization Tool.


US National Hydrography Dataset Barrier Analysis Tools







No releases published


No packages published