Functional significance of crayfish in stream food webs: roles of omnivory, substrate heterogeneity and sex

Heterogeneity of species interactions in food webs can result from characteristics of substrates as well as attributes of top consumers. We performed a streamside channel experiment to evaluate the impact of crayfish on lower trophic levels in detritus-based (leaf packs) and algal-based food webs (hard-bottoms). After 43 days, both male and female crayfish had dramatically promoted leaf decomposition, with males processin material at a faster rate. However, the difference in leaf processing rates was not related to a greater level of male activity. Despite the sex-related difference in residual leaf dry mass, densities of invertebrates in leaf packs were similarly low in the presence of crayfish of either sex, due to resource consumption, physical dislodgment (bioturbation) and/or predation. No trophic cascade was evident in the leaf pack assemblage. In the hard-bottom assemblage, the results confirmed circumstantial field evidence that crayfish reduce predatory Tanypodinae and indirectly increase collector-gatherer Chironominae following the prediction of a trophic cascade. However, no other taxa were indirectly facilitated, because of strong direct effects of crayfish on algal abundance (through direct consumption and bioturbation). Overall, impacts of crayfish on lower trophic levels were more pronounced in the structurally complex, detritus-based assemblages than in its hard-bottom, algal-based counterpart. This conflicts with the expectation that net predation effects should be weaker where structural complexity is greater but is mainly a consequence of the profound engineering effects of crayfish in reducing colonisable substrate when they shred and disturb detrital material. Effects of crayfish may therefore propagate differently and with varying strength depending on substrate. Moreover, engineering activities and predation by crayfish appear to have been of overwhelming significance with subtle sex differences in leaf processing rates failing to lead to differences in invertebrate densities.