Production-grade Python & GIS automation for precision agriculture.
Drone imagery, vegetation indices, yield mapping, variable-rate prescriptions, and ISOBUS-ready compliance.
ag-gis.com is a free, engineering-grade reference for building precision-agriculture and crop-GIS pipelines in Python. Every page is written for people who ship real systems — agtech engineers, farm data analysts, Python GIS developers, and drone/imagery teams — and treats spatial accuracy, radiometric calibration, and equipment compatibility as first-class engineering concerns.
Guides are not blog posts. Each one delivers a working implementation path: concept framing → environment and prerequisites → runnable Python with explicit EPSG codes and verification asserts → validation gates → failure modes and gotchas. Code blocks are syntactically complete and ready to drop into batch workers, Dask clusters, or ISOBUS export pipelines. 49 guides, ~95,000 words, and growing.
The site is organized around three end-to-end playbooks:
Coordinate reference systems, vector and raster ingestion, projection discipline, geospatial file-format choices (Shapefile, GeoPackage, GeoParquet, COG), and a systematic reference for debugging CRS and projection errors — the load-bearing infrastructure every prescription and index depends on.
End-to-end UAV pipelines: radiometric correction, cloud and shadow masking, band math (NDVI / NDRE / SAVI / GNDVI / CIre), index selection by crop stage, temporal aggregation, threshold mapping, and batch/async orchestration of multi-flight jobs with Celery, asyncio, and Dask.
Combine telemetry cleaning, spatial interpolation (IDW vs Kriging), management-zone classification, ISOXML / shapefile export, controller-format conversion across John Deere, Trimble, Raven, and CNH terminals, and prescription-export debugging against the ISO 11783-10 schema.
rasterio · geopandas · pyproj · shapely · xarray · dask · numpy · scipy · scikit-learn · pykrige · lxml · Celery — with real sensor and hardware context (MicaSense RedEdge-MX, DJI P4 Multispectral, Parrot Sequoia, RTK GPS) throughout.
A static site built with Eleventy (11ty) and deployed on Cloudflare Pages. Content is authored in Markdown with hand-authored, theme-aware inline SVG diagrams, JSON-LD structured data on every page, and a fully cross-linked internal graph.
npm install # install dependencies
npm run build # build the static site into _site/
npm run serve # local dev server with live reload
npm run deploy # build and deploy to Cloudflare PagesSpotted an error in a code sample, or want a topic covered? Open an issue. Suggestions grounded in real production ag-GIS workflows are especially welcome.
Commits to this repository are authored solely by the maintainer account — no co-authors are used. All content is provided for educational use. © ag-gis.