Jim R. Muirhead

Identifying the source of species invasions: sampling intensity vs. genetic diversity

Population geneticists and community ecologists have long recognized the importance of sampling design for uncovering patterns of diversity within and among populations and in communities. Invasion ecologists increasingly have utilized phylogeographical patterns of mitochondrial or chloroplast DNA sequence variation to link introduced populations with putative source populations. However, many studies have ignored lessons from population genetics and community ecology and are vulnerable to sampling errors owing to insufficient field collections. A review of published invasion studies that utilized mitochondrial or chloroplast DNA markers reveals that insufficient sampling could strongly influence results and interpretations. Sixty per cent of studies sampled an average of less than six individuals per source population, vs. only 45% for introduced populations. Typically, far fewer introduced than source populations were surveyed, although they were sampled more intensively. Simulations based on published data forming a comprehensive mtDNA haplotype data set highlight and quantify the impact of the number of individuals surveyed per source population and number of putative source populations surveyed for accurate assignment of introduced individuals. Errors associated with sampling a low number of individuals are most acute when rare source haplotypes are dominant or fixed in the introduced population. Accuracy of assignment of introduced individuals is also directly related to the number of source populations surveyed and to the degree of genetic differentiation among them (FST). Incorrect interpretations resulting from sampling errors can be avoided if sampling design is considered before field collections are made.

Contrasting patterns in genetic diversity following multiple invasions of fresh and brackish waters.

Biological invasions may combine the genetic effects of population bottlenecks and selection and thus provide valuable insight into the role of such processes during novel environmental colonizations. However, these processes are also influenced by multiple invasions, the number of individuals introduced and the degree of similarity between source and receiving habitats. The amphipod Gammarus tigrinus provides a useful model to assess these factors, as its invasion history has involved major environmental transitions. This species is native to the northwest Atlantic Ocean, although it invaded both brackish and freshwater habitats in the British Isles after introduction more than 65 years ago. It has also spread to similar habitats in Western Europe and, most recently, to Eastern Europe, the Baltic Sea, and the Laurentian Great Lakes. To examine sources of invasion and patterns of genetic change, we sampled populations from 13 native estuaries and 19 invaded sites and sequenced 542 bp of the mitochondrial COI gene. Strong native phylogeographical structure allowed us to unambiguously identify three allopatrically evolved clades (2.3–3.1% divergent) in invading populations, indicative of multiple introductions. The most divergent clades occurred in the British Isles and mainland Europe and were sourced from the St Lawrence and Chesapeake/Delaware Bay estuaries. A third clade was found in the Great Lakes and sourced to the Hudson River estuary. Despite extensive sampling, G. tigrinus did not occur in freshwater at putative source sites. Some European populations showed reduced genetic diversity consistent with bottlenecks, although selection effects cannot be excluded. The habitat distribution of clades in Europe was congruent with the known invasion history of secondary spread from the British Isles. Differences in salinity tolerance among lineages were suggested by patterns of habitat colonization by different native COI clades. Populations consisting of admixtures of the two invading clades were found principally at recently invaded fresh and brackish water sites in Eastern Europe, and were characterized by higher genetic diversity than putative source populations. Further studies are required to determine if these represent novel genotypes. Our results confirm that biological invasions need not result in diminished genetic diversity, particularly if multiple source populations, each with distinctive genetic composition, contribute to the founding populations.

Range expansion of quagga mussels Dreissena rostriformis bugensis in the Volga River and Caspian Sea basin

In 1992, we discovered populations of the nonindigenous quagga mussel Dreissena rostriformis bugensis in the middle reaches of the Volga River. The same species was found in samples collected between 1994 and 1997 in the Volga delta and in shallow areas of the Northern Caspian Sea. D. r. bugensis always co-occurred with its more widespread congener, the zebra mussel D. polymorpha (Pallas 1771). The quagga mussels contribution to total Dreissena abundance increased over time in the middle Volga reservoirs and Volga River delta. D. r. bugensis was common in the Volga portion of Rybinsk Reservoir during 1997 and, by 2000, it was in Uglich, Rybinsk and Gorky Reservoirs on the Upper Volga River. D. r. bugensis was neither found in Ivankov Reservoir, nor in terminal sections of the Volga-Baltic corridor including the eastern Gulf of Finland. Presently, all but the northern-most regions of the Volga River have been colonized by D. r. bugensis. We hypothesize that its introduction into the Volga River and Caspian basin occurred no later than the late 1980s via commercial shipping that utilized the Volga-Don waterway to navigate between the source Black-Azov Sea region and recipient areas on the Volga River. Larval drift likely contributed to establishment of populations at downstream sites, while human-mediated vectors may be responsible for introductions to upstream locations on the Volga River. We anticipate continued northward dispersal in conjunction with shipping activities.

Realized vs apparent reduction in enemies of the European starling

Release from parasites, pathogens or predators (i.e. enemies) is a widely cited ‘rule of thumb’ to explain the proliferation of nonindigenous species in their introduced regions (i.e. the ‘enemy release hypothesis’, or ERH). Indeed, profound effects of some parasites and predators on host populations are well documented. However, some support for the ERH comes from studies that find a reduction in the species richness of enemies in the introduced range, relative to the native range, of particular hosts. For example, data on helminth parasites of the European starling in both its native Eurasia and in North America support a reduction of parasites in the latter. However, North American ‘founder’ starlings were likely not chosen randomly from across Eurasia. This could result in an overestimation of enemy release since enemies affect their hosts on a population level. We control for the effects of subsampling colonists and find, contrary to previous reports, no evidence that introduced populations of starlings experienced a reduction in the species richness of helminth parasites after colonization of North America. These results highlight the importance of choosing appropriate contrast groups in biogeographical analyses of biological invasions to minimize the confounding effects of ‘propagule biases’.

Development and experimental assessment of an underwater video technique for assessing fish-habitat relationships

Developing tools that aquatic managers can use to understand the impact of human development on fish habitat is important in an era where our aquatic resources are under increasing pressure. To this end, we examined whether an underwater video camera was useful for quantifying fish habitat use in inland lakes by 1) examining patterns in fish habitat use, residency time and feeding behaviour among habitat types, 2) determining the precision and statistical power of the aforementioned estimates, and 3) assessing whether our habitat-specific camera estimates were reflective of whole-system estimates. Lastly, we used our protocol in an experimental situation to test site-specific fish habitat use at sites where habitats were manipulated (removed or added). We demonstrated that our underwater video protocol could successfully capture site-level habitat use that corresponded with whole-system abundance estimates, addressing the concern that habitat-based surrogates of fish productivity be validated on a whole-system scale. Unfortunately, our underwater video technique was unable to discriminate fish habitat use patterns among simple habitat types and unable to consistently separate differences in among-habitat fish feeding and residency behaviours. Our ability to detect a difference was low in all among-habitat comparisons. In the aquatic systems where habitat was added, we documented a significant shift in fish habitat use towards the addition sites and away from control sites, but no corresponding increase in system-wide fish biomass or production; no changes were apparent in the habitat removal lakes. A combination of longer filming duration, more filming sites or changing to a mobile transect method would likely address the data deficiencies that limited our ability to make site-level inferences about fish habitat use.

Consistent, long-term change in rotifer community composition across four Polish lakes.

Few studies exist documenting changes in rotifer communities over long time intervals. Here, we explore seasonal and long-term variation in rotifer communities in four Polish lakes sampled in 1976 and again in 1997. Rarefied, asymptotic species richness did not differ significantly across study years, although values in 1997 tended to be higher. Simpson’s and Shannon–Wiener diversity measures provided inconsistent temporal results, with only the former indicating significantly higher richness in 1997. Sorensen’s coefficient of community similarity was as high among lakes in 1976 (0.81) and in 1997 (0.76) as within lakes across the 21-year span (0.77). Nonlinear redundancy analysis of species’ abundances revealed large, consistent seasonal changes across lakes, smaller consistent shifts between sampling periods, and small differences between lakes. Collectively, these metrics indicate that species composition was relatively stable among lakes within years and within lakes between years, while species’ abundance patterns were far more variable and most affected by season.