A recent paper rethinks how spatial uncertainty is shown in maps by treating it as a layer in the same grammar that builds all other plot elements. The authors show how uncertainty‑specific symbols can be composed, customised and extended alongside standard visual components.

They built a new R package, <code>ggincerta</code>, that fits into the standard ggplot2 workflow. The package adds specialised layers and scales—for example, a <code>geom_sf()</code> layer with a <code>duio()</code> aesthetic lets users create bivariate choropleth maps that encode a value and its uncertainty in one image.

Compared with Vizumap, which requires a cascade of dedicated functions for reading data, building palettes, mapping and adding keys, ggincerta lets users finish the same map with a single layer specification. This smoother workflow runs directly inside the layered grammar of graphics and avoids manual assembly of map and key.

The package also introduces new visual techniques such as the dual map, where the polygon’s fill shows uncertainty while a glyph inside the polygon shows the central estimate. Dual maps make patterns of uncertainty obvious at a glance, but the authors warn that colour perception can be affected when background lightness varies, so they recommend a neutral greyscale background to minimise contrast interference.

The paper demonstrates ggincerta on three real‑world data sets—child undernutrition in Bangladesh, brain‑fMRI measures and a crop‑improvement experiment—highlighting its flexibility and the richer visual stories it can support for spatial data.

Overall, ggincerta brings uncertainty visualisation into the layered grammar of graphics, offering a unified, flexible workflow for creating maps that convey both estimates and their uncertainties in a single, composable graphic.