Fig. 3: Size-frequency distributions of craters formed on layered ejecta deposits on Mars.

a–h show counting areas (yellow polygons in a–d) and size-frequency distributions (e–h) of crater populations formed in smooth deposits of single- or multiple-layered ejecta (SLE or MLE). Gratteri has SLE deposits (a, e), and Zunil (b, f), Tooting (c, g), and Mojave (d, h) have MLE deposits. i Counting areas (red and blue polygons) and craters (yellow circles) collected on the inner ejecta layer of Steinheim, which is a Type 1 double-layered ejecta (DLE) crater. The red polygon is for the western half of the inner ejecta layer, and it overlaps the blue polygon that is for the entire inner ejecta layer. j Cluster analysis for the crater population formed in the inner layer revealing a clustered distribution for craters with D = 95–120 m in the eastern half. Dots represent the locations of craters collected in the inner ejecta layer. Colors saturation is based on the strength of membership, with dark being strong and light being weak (see Methods). Craters in the same cluster have the same color, and unclustered craters are gray. Analyses of spatial distribution of the entire crater population are shown in Supplementary Fig. 6. k Size-frequency distributions of the D = 95–120 m craters in the entire (blue) and western half (red) of the inner ejecta layer. The approximate population of the extra self-secondaries (black curve) is derived by subtracting the western crater population from the entire crater population in the inner ejecta layer. The equivalent model age for the population of extra self-secondaries is derived using the Hartmann crater chronology². The 95% confidence interval is represented by color shades in (e–h) and (k) (see Methods). IDs and available addresses of base images used are in the Supplementary Table 2.