Open GeoPDFs with your position on them
USGS quads, forest-service maps, alpine-club sheets, survey exports — the world's best maps ship as GeoPDF and GeoTIFF. NomadTracks opens them on iPhone and iPad with your live GPS position, automatically reading the embedded georeferencing. And when a file has none, you georeference it yourself in two minutes.
Automatic georeferencing
NomadTracks parses the geospatial metadata inside GeoPDF and GeoTIFF files and aligns them on import — no manual steps for properly tagged files. A great free source to try it with: the BLM’s georeferenced PDF library of US public-lands maps. The parser can be re-run on any map as it improves, and tells you why a file couldn't be auto-aligned.
Manual mode for everything else
Plain PDF? Scan? Photo of a paper map? Set ground control points by tapping matching spots on your map and the world map. Affine fits start from a single point pair; spread more points and NomadTracks upgrades to polynomial fits with a live ± meter accuracy readout.
Fine alignment for stubborn maps
Old maps with datum offsets can be consistently a few meters off. The fine-alignment tool nudges and rotates the whole overlay on top of the control-point fit — no re-placing every point.
If you're coming from Avenza Maps
Since March 2026, Avenza's free tier no longer allows importing your own maps at all, and Plus ($34.99/year) caps imports at 20. In NomadTracks the same workflow is free and unlimited: import your georeferenced PDFs and they behave as you'd expect — with full GPS track recording, POIs with photos, route planning and Apple Watch support around the map viewer.
Related
FAQ
Which formats are supported?
GeoPDF and GeoTIFF with automatic alignment; any PDF, TIFF or photo with manual control points. Custom MapLibre basemaps are supported for self-hosters.
Is there a limit on the number of maps?
No — import as many custom maps as your storage holds, and organize them in folders.
How do I know how accurate the alignment is?
NomadTracks shows the fit type and RMS error (e.g. 'polynomial · ± 2 m') for every georeferenced map — accuracy is always visible, never assumed.