Evaluating zebra mussel spread pathways and mechanisms in order to prevent further spread
Activity 1: Identification of the sources of inland invasions throughout Minnesota, and the pathways through which zebra mussels have spread throughout the state
We have completed genotyping and analyzing our first data set, which includes 43 geographic sites, 16 lakes and 3 river systems, and 1,281 mussels at 9 microsatellite marker loci (samples gathered in the fall of 2016). Using genetic methods to study invasion sources and pathways at the scale of a U.S. state is novel. We then expanded our genotyping further to investigate regional spread and the lack of contribution from perceived “super-spreader” lakes. In total, we have now genotyped 2,050 mussels from 40 lakes and 4 rivers at 9 microsatellite markers with a total of 69 geographic sites included. We found that:
- Genetic diversity in lakes is high. That means that lakes are colonized by large numbers of mussels or larvae.
- Several lakes are genetically distinct which allows for the study of invasion sources.
- Genetic clusters identify regional patterns of spread. Minnesota contains three large regions of many lakes – Detroit, Alexandria, and Brainerd Lakes regions – in which infested lakes are clustered. Genetically, lakes within each reach fall into one or more localized genetic clusters found nowhere else in the state. Since mussels in lakes in each region are closely related and genetically unique, we conclude that vectors active within the region have spread genetically similar “colonizing” mussels, from the first lake(s) infested, then from lake to lake. This calls for more efforts to identify and block regional spread vectors.
- “Super-spreaders” Mille Lacs and Prior Lake have, surprisingly, not infested other lakes in Minnesota. Their super-spreader status is inferred due to high boater traffic. Mille Lacs watercraft inspection data shows that departing boaters next visit a large number of both recently infested and uninfested lakes. We evaluated 35 lakes that were infested after 2005, testing an invasion model in which Mille Lacs served as the source for the new lake, and have not yet detected a single case in which Mille Lacs was selected by the model to be the source.
This extreme discord between boater movements and genetics must mean that watercraft inspection and decontamination has been effective on Mille Lacs and should be continued — on this and other high traffic lakes.
In light of Sequence-Based Genotyping (SBG) becoming available in 2016, our lab worked with the University of Minnesota Genomics Center to develop a strategy to use SBG for zebra mussel genotyping. This new strategy uses a different enzyme to generate a higher-quality genetic sequence. We are now in the process of analyzing SNP (single nucleotide polymorphism) data from 13 lakes and four rivers.
Data analysis for sources and pathways of spread is ongoing, particularly with the clustered lakes invasion model for more complex regions. For example, in the Brainerd Lakes region, we are interested in determining whether “stepping stone” spread occurred from Ossawinamakee (the first infested natural inland lake in Minnesota) to other Crow Wing County lakes that were infested later. These complex analyses are being conducted at the Minnesota Supercomputing Institute. Finally, adequate analysis of SNP data and subsequent model testing is just beginning and will extend into Phase II.
Activity 2: Clarifying downstream drift as a mechanism of spread of zebra mussels between lakes
In order to address the risk of spread through downstream drift (the phenomenon of veligers naturally moving in lakes and streams, settling, and founding new populations), samples have been gathered, counted, and photographed using our CPLM-image analysis system. MAISRC team member Maxwell Kleinhans wrote a program that interfaces with Image J software for rapid capture of these images and various analyses of size and shape of larvae, which is now semi-automated. Samples were gathered from the Pelican River, Pine River, Gull River, and Minnehaha Creek. We found that:
- Settlement occurs immediately downstream of the headwater lake and declines to zero, typically within the first kilometer downstream.
- Small streams (< 10 m wide) are not good habitat for zebra mussels, so the management issue is not colonization of the stream bottom for most of its length, rather the concern is dispersal of larvae.
- The number of larvae leaving infested lakes can range from 10 million to more than a billion per day.
- Downstream larval dispersal sharply declined with distance, and this decline becomes steeper as summer progresses.
- Early in the season, it may be possible for veligers to disperse downstream as far as 40 miles in the Pelican system. This could explain the infestation of Dayton Hollow and Orwell Reservoirs below Fergus Falls.
Population genomics of zebra mussel spread pathways
Phase II of this effort focuses on preventing zebra mussel invasions by developing genetic evidence of spread sources and pathways so that they may be interrupted. It also lays the groundwork for potential biocontrol through genetic modification technologies.
The prevention research will result in direct evidence of sources and pathways for zebra mussel invasions in Minnesota and will provide accompanying prevention management recommendations based on these findings. We will use highly variable population genetic markers called microsatellite DNAs, and variable DNA positions in the zebra mussel genome—Single Nucleotide Polymorphisms, or SNPs—to genetically type zebra mussel populations, and assign these populations to the source waters from which they were carried to infest new waters. We will complete this work for approximately 75 waterbodies, while also creating a database that will enable a more powerful analysis of additional waterbodies that may be studied in the future (e.g. new infestations). Learn more about Phase II of this project here.