Do dams and road crossings limit round goby distribution in great lakes tributaries?

Background/Question/Methods

Access to spawning grounds in tributaries of the Great Lakes is drastically reduced for dozens of migratory fishes by >275,000 dams and road crossings, motivating efforts to restore habitat connectivity by removing or modifying these barriers.  Unfortunately, these efforts may also increase available habitat for invasive species like round gobies (Neogobius melanostomus).  Round gobies, which compete with native fishes for food and habitat, rapidly invaded all five Great Lakes after their discovery in 1990, and their distribution continues to expand within the Lakes and their watersheds.  We investigated whether dams and road crossings currently limit the distribution of round gobies in Great Lakes tributaries.  If so, management decisions should account for the likelihood of facilitating the spread of this invasive species. We used two existing spatial datasets, one of round goby occurrences and one of dams and road crossings, which included estimates of dam and road culvert passability for fishes with three different swimming abilities. Using the available passability estimates for slower swimming, small bodied fishes, we estimated the proportion of tributaries currently accessible to gobies, as well as the tributaries that would become accessible under several proposed barrier removal scenarios.

Results/Conclusions

Of the 980 documented instances of round goby invasion in the Great Lakes basin, 23% are in tributaries upstream of at least one dam or road crossing.  Round gobies found upstream of impassable barriers (e.g., dams; n=75) may be attributable to ballast water or bait bucket transfer. The minimum passability for all other barriers encountered by round gobies moving upstream (n=141) is 0.52 (where the passability of a fully passable road culvert or bridge is 1 and of a dam is 0), indicating that the current distribution of round gobies is likely to be limited by dams and road crossings and that a passability threshold may exist for round gobies. Combining this inferred passability threshold with other environmental variables should strengthen existing distribution models for round gobies. For example, only 6.1% of the total length of Great Lakes tributaries is currently accessible to round gobies based on the observed passability threshold. However, the amount of habitat accessible to round gobies could rise rapidly as barrier removal projects are completed. We conclude that barrier removal strategies must account for trade-offs between benefits for native species and costs associated with enabling the spread of invasive species.