Evidence for micropatch partitioning and effects of boundaries on patch use in two species of gerbils

1. Spatial heterogeneity in resource abundance creates opportunities to exploit patchiness at different scales. The ability to subdivide patches into smaller micropatches is termed micropatch partitioning. At various scales, patches can be recognized by distinct boundaries or by vague boundaries delineated by changes in resource abundances. The ability to micropatch partition was tested for and the effect on foraging of vague vs distinct boundaries was assessed for two species of coexisting desert rodents, the gerbils Gerbillus allenbyi and G. pyramidum. To do so, the distribution of seeds in experimental resource patches in a large enclosure were manipulated. 2. Both G. allenbyi and G. pyramidum are imperfect micropatch assessors. When seeds were concentrated into micropatches, there were lower patch-wide giving-up densities (GUDs) than if seeds were randomly distributed throughout the patch. Both species could micropatch partition at scales of 0.14 m². 3. GUDs in experimental micropatches were greater than expected for perfect assessors. 4. With G. allenbyi, a micropatch with fewer vague boundaries was foraged to a lower GUD. In contrast, GUDs for G. pyramidum were similar regardless of the types of patch boundaries. 5. Gerbillus allenbyi showed greater error in the assessment of micropatch size. 6. These results have implications for diet selection and coexistence. These species coexist via temporal partitioning, with G. pyramidum biasing its activity to early in the night when resource patches are richer and G. allenbyi biasing its activity to later in the night. Gerbillus pyramidum encounters a richer and more variable environment where errors in patch assessment are costlier, and micropatch partitioning is more valuable. Gerbillus allenbyi encounters a poorer environment where resources vary at larger spatial scales.