Understanding the Authagraph Projection
The Authagraph projection is a cylindrical map projection developed by Japanese astronomer Hajime Narukawa and officially unveiled in 2014. It gained significant attention after the National Geographic Society included it in their list of top ten new maps in 2016. The core idea is straightforward: instead of projecting the globe directly onto a cylinder like Mercator, the Authagraph wraps the sphere inside an octagonal prism, maps the surface onto that, and then unfolds it into a rectangle. This geometric approach is what gives it its particular distortion profile. What you end up with is a world map where landmass areas are far more accurate than Mercator, and coastline shapes are reasonably preserved too. Greenland, for example, appears roughly its true size relative to Africa rather than the exaggerated version most people grew up with. The polar regions are still stretched, but not nearly as badly as in Mercator. That said, the map does have a distinctive jagged edge along the top and bottom where the octagonal folding creates discontinuities. You'll see thin strips of ocean separating the main rectangle from small satellite panels that represent the poles. It looks unusual at first, and some people find it visually jarring even though the geometry is sound.
Downloading the Authagraph Map Of The World
The official source for high-resolution Authagraph files is the Authagraph website maintained by the project team. They offer several formats: vector SVG files, PNG raster images, and tiled web maps. The vector files are typically around 50 to 100 megabytes depending on resolution, while the PNG variants can range from 5 megabytes for standard screens up to 50 or 60 megabytes for print-quality exports. There is no cost involved for personal or educational use, but commercial licensing requires contacting the project directly. I've downloaded the SVG version multiple times over the years and it has consistently been reliable. The geometry behind the Authagraph is more involved than a standard cylindrical or conic projection, but the practical result is easier to grasp. Start with a sphere. Wrap it in a rectangular sheet of material. Now pinch that rectangle along four lines so it forms an octagon when viewed from above. Project the globe's surface onto that octagonal shape. Then cut and unfold it back into a rectangle. The result is a mapping where the equatorial regions are represented with minimal distortion, and the polar areas get redistributed across those smaller satellite rectangles at the top and bottom. In practice, this means that any coordinate transformation from lat-long to Authagraph space requires a piecewise calculation. The main rectangle uses one set of equations, and each polar panel uses a different set. If you're working with GIS software that doesn't natively support this projection, you'll need to implement the transform yourself or find a plugin. PROJ.4 and GDAL do include Authagraph support in recent versions, which makes things significantly easier than it was a few years ago when you had to write custom code for almost everything.
I ran into a specific issue last year when trying to overlay satellite imagery onto an Authagraph base map for a presentation. The imagery was in Web Mercator (EPSG:3857), and I needed it reprojected cleanly onto the Authagraph grid. Standard QGIS tools handled the central rectangle fine, but the polar regions showed severe stretching and misalignment because the reprojection engine was treating the entire output space as a single cylindrical projection. The workaround was to split the polar panels into separate layers, reproject each individually using their own coordinate definitions, and then merge them back together. It added about twenty minutes to the workflow, but it produced a clean result without visible seams at normal viewing distances.
Get the Full Details

Common Pitfalls and What People Miss
Most users approach the Authagraph assuming it solves the distortion problem entirely. It doesn't. No flat map projection does. The Authagraph minimizes combined area and shape distortion better than Mercator, but it introduces its own trade-offs. Distance measurement along certain paths, particularly those crossing the octagonal fold boundaries, can be misleading if you're not accounting for the projection's piecewise nature. Route planning applications that assume a uniform projection will give you incorrect bearings near the polar panels. Another thing that catches people off guard is the map's behavior at extreme zoom levels. When you zoom into a small region in the main rectangle, the distortion is minimal and the map behaves almost like a simple cylindrical projection. Zoom into a polar panel and the scale factor changes abruptly at the panel boundary. This isn't a bug in your software, it's a fundamental property of the projection. If you're building an interactive web map with it, you'll need to handle these transitions carefully or your users will notice scale jumps that look like rendering errors. The Authagraph also doesn't solve the fundamental problem that any world map has to represent a sphere on a flat surface. Certain spatial relationships that are intuitive on a globe become awkward on this projection. For instance, the relative positions of countries near the fold lines can shift in ways that feel wrong to someone used to Mercator. Japan and Australia, for example, appear much closer together than they do on Mercator, which actually is more geographically accurate, but viewers familiar with the standard map often question whether the rendering is correct. You should expect this kind of pushback and have the geometric explanation ready if you're presenting the map publicly.
When to Use It and When Not To
The Authagraph is well-suited for educational contexts, museum displays, and general reference where accurate area representation matters more than directional accuracy or seamless tiling. It's also useful for thematic maps showing global distributions of population, climate zones, or resource reserves where Mercator's polar inflation would skew the visual message. If you need a clean rectangular output for embedding in documents or slides, it works reasonably well. It is not suitable for navigation, web mapping at global scale, or any application requiring consistent distance or bearing measurements across the full extent. For those use cases, equal-area projections like Eckert IV or compromise projections like Robinson remain more practical. If you're building a web map that needs to tile efficiently across zoom levels, consider using a standard projection for the base layer and overlaying Authagraph-based visualizations only where the specific accuracy benefits outweigh the implementation complexity. The projection is also limited by its visual complexity. The discontinuous layout with separate polar panels means you can't create a seamless panorama or a continuous wrap-around scroll effect without additional processing. Some publishers have experimented with creating a continuous variant by rearranging the panels, but that introduces new distortions that defeat much of the original design's purpose. The official version remains the most mathematically coherent representation available.