Habitat‐dependent competition and the coexistence of Australian heathland rodents

When competing species depress one another's fitness in the habitats that they occupy, their competitive effects will emerge in each species’ pattern of density‐dependent habitat choice. Thus, a regression analysis of joint densities, corrected by the habitat effect, should reveal the magnitude of interspecific competition. We tested this idea by 1) demonstrating the connection between removal experiments and regression estimates of competition with those obtained from isodars (regressions that implicitly incorporate evolutionarily stable strategies of habitat selection) and 2) evaluating whether interspecific competition inferred from isodars corresponded with the inferences emerging from regression and field experiments. Previous removal experiments on two herbivorous rodents occupying coastal wet heathlands in eastern Australia documented that competition between Rattus lutreolus and Pseudomys gracilicaudatus is asymmetrically biased in favor of the much larger Rattus . The asymmetry in competition was also revealed by regression estimates of competition. Isodar analyses illustrate a habitat‐dependent mechanism for the asymmetry. Rattus compete effectively with Pseudomys in both ‘wetter’ and ‘drier’ patches of heath whereas Pseudomys appear to exert a competitive effect in only the drier sites. The magnitude of competition measured by a removal experiment in an area with more‐or‐less equal amounts of both habitats will be biased in favor of Rattus . More generally, one can use the isodar estimates to draw isolegs and isoclines of competitive coexistence. Isoclines for the two Australian rodents imply dynamic equilibria of stable competitive coexistence that vary with plant succession in fire‐dominated heathland ecosystems.