Body condition dependent dispersal in the semi-aquatic insect, Notonecta undulata

Background/Question/Methods

Dispersal is the movement of individuals across space, with the potential for gene flow among populations. Dispersers are often found to be a non-random sample of the population, differing from non-dispersers in a variety of traits. If dispersal causes the sorting of phenotypes across space, this may influence ecological and evolutionary processes.

Body condition is strongly tied to aspects of individual performance including competitive ability and fecundity and therefore connections between body condition and dispersal leading to sorting across the landscape may be especially important. However, empirical studies investigating the relationship between dispersal and body condition have produced mixed results. One hypothesis for this inconsistency is that body condition has interactive effects with responses to other factors.  This may make the relationship between body condition and dispersal context-dependent. Predation risk is one factor that may interact with body condition to shape dispersal responses. To assess this, we tested whether there is an interactive effect of body condition and predation risk on dispersal. We imposed diet treatments on backswimmers in the laboratory, and measured the effects of food availability on body condition. We then measured the effects of body condition and predation risk on dispersal in a field mesocosm experiment.

Results/Conclusions

We found that dispersal capacity was an increasing function of diet treatment, which was demonstrated by the fact that there was a positive relationship between fat content and diet treatment (p = 0.0004). Consistent with this, we found that dispersal probability was a positive function of body condition, although the effect of body condition decreased over time (diet × time: p = 0.004). Since body condition influences competitive ability and fecundity, dispersal that is non-random with respect to body condition may influence processes such as intra- and inter-specific interactions, and local adaptation. Dispersal probability increased with predation risk (p = 1.3 × 10^-6). However, the effects of body condition and predation risk were additive, not interactive (predator × diet: p = 0.266), indicating that an interaction between condition and predation risk is unlikely to be responsible for the inconsistency of the results of previous studies of condition-dependent dispersal. However, this does not rule out the possibility that interactions between body condition and other factors may alter the relationship between body condition and dispersal. Further investigation of these interactions in a greater diversity of systems may provide other insights. We will discuss additional factors that may be contributing to this inconsistency.