Many species have life cycles where a life history stage (e.g. winged insects) undertakes dispersal. Recruitment effected by successful dispersers from source communities can have profound effects on the species diversity of receiving communities. This hypothesis is largely unexplored for riverine fauna.
Previously, we showed that many caddisfly species are able to fly upstream to catchment boundaries and link rivers on either side (boundary species), whereas others do not (non-boundary species). Hypothetically, insect populations in catchments that share boundaries with many other catchments (i.e. total area of surrounding catchments [ASC] is large) should have frequent successful dispersal; this should result in higher species diversity than in catchments that are relatively isolated from others (i.e. have a small ASC). We have termed this the Insect Superhighway Hypothesis (ISH). A related question is: how do boundary species navigate successfully up to catchment boundaries and then fly down-slope to reach rivers?
First, we tested the ISH and found support. Species diversity of boundary species was related to ASC (explaining 50% of variance) but not home catchment area. In contrast, diversity of non-boundary species was weakly associated with home catchment area but, as expected, was unrelated to ASC. Second, we explored whether caddisflies use thermally driven slope winds to reach, and then fly over, catchment boundaries. We found evidence that boundary species ride katabatic winds to move from ridgelines down into river valleys after sunset.
These findings are challenging current models. The Isolated Headwaters model proposed that species at the tops of catchments are isolated from those in areas downstream. Far from being isolated, our results suggest that headwater areas act as conduits for dispersal. Given often extensive clearing of native vegetation from catchment ridge-lines, our results suggest a new explanation for the troubling rate of species extinctions among aquatic insects.