thedavidweng/china-village-boundaries — explained in plain English
Analysis updated 2026-05-18
Load village-level boundaries for a Chinese province into QGIS for mapping.
Read the boundary data in Python using GeoPandas for spatial analysis.
Study administrative geography patterns across China's 33 provincial divisions.
Use as a base layer for academic or personal GIS research.
| thedavidweng/china-village-boundaries | 920linjerry-stack/capital-studio | aahonarmand/neticu | |
|---|---|---|---|
| Stars | 16 | 16 | 16 |
| Language | — | Python | Swift |
| Setup difficulty | moderate | easy | easy |
| Complexity | 2/5 | 3/5 | 2/5 |
| Audience | researcher | researcher | general |
Figures from each repo's GitHub metadata at analysis time.
Some provinces use GBK text encoding instead of UTF-8, which can cause garbled village names.
This repository is a geospatial dataset containing the boundaries of villages across all of China, compiled into Shapefile format for use in mapping software. It covers all 33 provincial-level administrative divisions and contains roughly 875,000 individual boundary records at the village level, which is the smallest unit of administrative geography in China. The data was originally shared by another user online and has been reorganized and re-uploaded here. The files are organized by province, with one or more Shapefile sets per region. All files have been standardized to the WGS84 coordinate system, which is the same standard used by GPS and most web mapping tools. The data is estimated to come from roughly 2010 to 2022, with most records reflecting the situation around 2016. In addition to Shapefile, some provinces are also available in GeoJSON, KML, GML, and a format called FileGDB. The data can be loaded directly into QGIS by dragging in the file, or read in Python using a library called GeoPandas. The README documents several known issues. The field names and structure differ from province to province, so a mapping table is provided to help align them across regions. Some provinces store village names using an older Chinese text encoding called GBK rather than the standard UTF-8, which can cause garbled text in some software. A few provinces have field names that were cut off because the Shapefile format imposes an 11-byte limit on field names. The README explains how to work around these issues. The license is CC0, meaning the compilers have waived their own rights to the reorganization work. However, the README is explicit that the copyright status of the underlying data is unknown, the boundaries have not been officially verified, and the dataset is recommended only for personal research, academic analysis, and learning GIS. It is not recommended for commercial use or as a legal or administrative reference. Anyone who believes they hold rights to the original data can request removal.
A geospatial dataset of roughly 875,000 village-level boundaries across all of China, packaged as Shapefiles for mapping software.
The compilers waived their own rights to the reorganization work, but the copyright status of the underlying data is unknown.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Verify against the repo before relying on details.