Ranking system for the viable options that arise from different initial landscapes.

(A) Mosaic richness indices along the paths shown in Fig 6 that originated from the initial landscape shown in Fig 4A. The mosaic richness indices of the intermediate landscapes that occur along the three different paths are very similar, so much so that it is difficult to differentiate between the dashed line (corresponding to Fig 6C) and the solid line (corresponding to Fig 6A). Thus, for this initial condition, it is clear that several good options exist that do not correspond to dominant minimal paths (e.g., the dashed and dotted lines), and highlights the reason one may want to consider minimal deviants. (B) Mosaic richness indices along the paths shown in Fig 6 that originated from the initial landscape shown in Fig 5A. For this initial condition, the mosaic richness indices of the intermediate landscapes vary considerably along the paths. It is clear that the path shown in Fig 7B (dotted line) is not the best choice for this initial landscape. Alternatively, the paths shown in Fig 7A and 7C are comparable, but the graph supports the argument that, in this situation, maybe the path shown in Fig 7C (dashed) that leads to a global optimum is not as good as the locally optimal path shown in Fig 7A (solid).