Non-native freshwater cladoceran Bosmina coregoni ( Baird , 1857 ) established on the Pacific coast of North America

The freshwater cladoceran Bosmina coregoni (Baird, 1857), native to Eurasia, has established and spread in the Great Lakes region of North America since the 1960s. Here we report the first detection of B. coregoni on the Pacific coast of North America, in three geographically distinct locations: the Lower Columbia River Estuary (LCRE), Lake Washington in western Washington state, and the Columbia River Basin in south eastern Washington state. Bosmina coregoni was detected on multiple sampling dates in Lake Washington and the LCRE between 2008 and 2012.


Introduction
Non-indigenous zooplankton have successfully invaded coastal and inland freshwater bodies throughout North America, sometimes with substantial effects on native zooplankton assemblages.For example, zooplankton communities declined in overall taxa richness and abundance after introduction of the predatory cladoceran Bythotrephes longimanus to the Laurentian Great Lakes and Canadian Shield lakes (Barbiero and Tuchman 1994;Strecker et al. 2006), and several introduced Asian copepods have impacted northeast Pacific estuaries (Bollens et al. 2002;Cordell et al. 2007;Cordell et al. 2008;Bollens et al. 2012).
The cladoceran Bosmina coregoni, common in its native Eurasia, has invaded and established in freshwater lakes throughout eastern North America since the 1960s (Deevey and Deevey 1971;Lieder 1991;Suchy et al. 2010).Genetic comparisons of endemic European and invasive North American populations reveal multiple founding events within the Great Lakes, likely via ship ballast water (Lieder 1991), and subsequent dispersal to surrounding inland lakes (Demelo and Hebert 1994).Until now this species has not been reported on the Pacific coast of North America, but has been detected as far west as Lake Winnipeg and Missouri (Suchy et al. 2010;Mabee 1998).
Bosmina coregoni's effects on native zooplankton and freshwater food webs have not been explicitly studied.Bosmina coregoni's life history traits and herbivorous feeding mode are likely to be similar to other bosminids, though species-level differences have not been examined.Here we report the presence of B. coregoni in three geographically distinct freshwater locations on the west coast: the Lower Columbia River Estuary (LCRE), Lake Washington in western Washington state, and the Columbia River Basin in south-eastern Washington state.

Methods
Bosmina coregoni specimens were collected through three separate monitoring and research projects (see Figure 1 for locations).The first was a multi-year field study of the planktonic community in the LCRE.Zooplankton were sampled monthly from a dock near Astoria, Oregon, USA, beginning in January 2005.The dock is located 20 km upstream of the river's mouth and extends 40 m from shore.Triplicate vertical tows with a 0.5-m mouth diameter, 73-µm mesh net equipped with a flow meter were used to collect zooplankton samples.Water column depth during sampling varied between 4.0 and 6.5 m, and tows were made from 0.5 m above the bottom to the water surface.Samples were preserved in 5% buffered formalin.Zooplankton were counted and identified using a dissecting microscope at 25 magnification.The LCRE zooplankton community is currently dominated by a mix of native and non-native copepods, including Eurytemora affinis, Diacyclops thomasi, Pseudodiaptomus forbesi, Limnoithona tetraspina, and Sinocalanus doerrii (Bollens et al. 2012).
Bosmina coregoni specimens were also collected as part of the long-term monitoring of Lake Washington that began in 1960 and continues to the present.Lake Washington is a large urban lake located at the eastern edge of Seattle, Washington, USA.The lake is 32.2 km long, an average of 2.5 km wide, and has a maximum depth of 64 m.It is used for various recreational purposes, and has two tributaries, the Cedar River at the southern end, and the smaller Sammamish River at the northern end.Lake Washington has been sampled bimonthly or monthly since 1960.The lake's zooplankton fauna is currently dominated by several Daphnia and Bosmina species, as well as the copepods Diacyclops thomasi (Forbes) and Skistodiaptomus reighardi (Litt, unpublished data).Zooplankton sampling and processing were done at three representative stations in the lake: Madison Park (MP), Arrowhead Point (AHP), and South of Southend (SSE), according to the methods of Edmondson and Litt (1982).Zooplankton were sampled every 1-3 weeks by taking oblique tows using a closing Clarke-Bumpus sampler equipped with a 73 and 130 µm nylon mesh net and flow meter.Zooplankton samples were preserved in 95% ethanol and counted (Edmondson and Litt 1982).The body lengths of three Bosmina and three Daphnia species from five 2011 samples (June-October) were measured, and differences in mean body lengths among taxa were tested for statistical significance using Welch's t-test for unequal variances.
Finally, B. coregoni was detected in samples collected in August 2012 as part of a limnological survey of 35 lakes in the Columbia River Basin in south-east Washington state.Bosmina coregoni was detected in Potholes Reservoir (maximum depth 9.7 m) and the adjacent Canal #1, and Moses Lake (10.2 m).Zooplankton sampling consisted of one open vertical net haul at the deepest attainable point in the water body using an 80 µm, 30 cm diameter zooplankton net.All zooplankton samples were preserved in 70% ethanol.Samples were examined under a dissecting microscope for species identification and counted for relative species abundances.Zooplankton were identified as Bosmina coregoni from morphological characters by Arni Litt.Distinguishing morphological features of B. coregoni include the complete lack of a mucron, extremely long antennules, the presence of only proximal pecten on the post abdominal claw, and the location of the sensory bristle near the tip of the rostrum (Kotov et al. 2009; see Figure 2).However, the carapace and rostrum may vary in form, therefore male morphology and genetic markers are required for final taxonomic confirmation (Kotov et al. 2009).Genetic analysis of males collected in Lake Washington was performed by Dr. Alexey A. Kotov of the A. N. Severtsov Institute of Ecology and Evolution, Moscow, Russia and by Derek Taylor of the State University of New York at Buffalo, to confirm species identification.

Lake Washington
Bosmina coregoni was not detected in monthly samples collected between 1960 and 2010 in Lake Washington.Bosmina coregoni was first detected in June 2011 and was present on each of 13 sampling dates from June 8, 2011 to December 2, 2011 (see Figure 3), occurring with a maximum density of 10.52 organisms L -1 at AHP on July 19, 2011 (0-10 meters depth).Males were present in samples from MP on two dates in October 2011, at densities between 0.0-0.039organisms L -1 , and at depths below 10 meters.Bosmina coregoni was not detected in samples collected in 2012 except for one female present in an April sample and a few individuals in November and December at densities of less than 0.03 organisms L -1 .In 2013, B. coregoni re-occurred with a maximum density of 2.091 organisms L -1 in July.Mean body size of mature female B. coregoni (n=85) was 0.587 mm, significantly larger than Bosmina liederi females (n=67; mean size 0.421 mm; p<0.001, df=111) and Bosmina longispina females (n=16; 0.553 mm; p=0.046, df=48), but significantly smaller than Daphnia galeata females (n=26; 1.830 mm; p<0.001, df=37), Daphnia pulicaria females (n=67; 2.171 mm; p<0.001, df=119), or Daphnia thorata females (n=39; 1.656 mm; p<0.001, df=84) (see Figure 4).All of these species co-occurred in 2011 samples.

Eastern Washington
Bosmina coregoni was detected in Potholes Reservoir and the adjacent Canal #1, and Moses Lake in eastern Washington state in August 2012.B. coregoni was the the most numerically abundant species in samples from all sites.Mature B. coregoni females made up 36% of the mesozooplankton in the sample from Moses Lake (excluding rotifers and immature stages), and the copepod Mesocyclops edax was the second most abundant species, making up 31% of the sample.Bosmina coregoni (88%) and M. edax (15%) were the most abundant species present in the sample from Potholes Reservoir.

Discussion
While we documented the presence of B. coregoni only in three freshwater locations along the west coast of North America, this species' prolific invasion history in the Great Lakes region suggests its current range may extend far beyond our sampling area.An exact invasion mechanism for this species in our study area may never be established, since cladocerans and their resting eggs can be transported in a variety of ways, including migratory birds, recreational boats, and direct hydrological connections (Figuerola and Green 2002;Havel and Medley 2006).The presence of this species in non-coastal eastern Washington suggests between-lake dispersal mechanisms such as recreational boating or migrating waterfowl, and suggests the possibility that B. coregoni's range has expanded across the continent in this fashion.
No studies from the Great Lakes region have directly addressed the ecological impacts of B. coregoni introduction to new aquatic habitats.In sporadically monitored lakes, realization of its presence can trail initial invasion by decades (Suchy and Hann 2007).Lake Washington's zooplankton community, monitored continuously since 1960 (Edmondson et al. 2003;Hampton and Schindler 2006), has undergone substantial shifts in species composition and abundance, often in clear response to changing environmental factors (Edmondson and Litt 1982).The addition of B. coregoni may shift population dynamics within the zooplankton community, since B. coregoni is larger than co-occurring bosminids B. liederi and B. longispina but likely too small to fall prey to juvenile sockeye salmon or other planktivorous fish (Scheuerell et al. 2005).
Though B. coregoni did not reoccur in significant abundance in Lake Washington samples during the years following its initial invasion, overall Bosmina density was also low, and its reappearance is likely.Sediment cores taken from Lake of the Woods, Ontario showed that B. coregoni can persist at low density relative to native Bosmina for decades after an initial invasion event (Suchy and Hann 2007).Previous literature indicates that B. coregoni mainly occurs in inland freshwater habitats, but this species' multi-year, low density persistence in the LCRE implies either its successful establishment in an estuarine environment, or repeated introductions.While the sampling site in the LCRE does periodically experience saline conditions (Bollens et al. 2012), this may not necessarily represent an expansion of B. coregoni's abiotic range, as zooplankton could have drifted from freshwater locations upstream.Unlike in Lake Washington or the LCRE, B. coregoni appears to be numerically dominant in the Eastern Washington lakes sampled; there is currently insufficient knowledge of the biotic and abiotic characteristics of these lakes to determine why B. coregoni has become so abundant there.Further sampling throughout the Pacific Northwest will likely expand its known range and elucidate its interactions with native freshwater communities.Future comparisons between Lake Washington and the LCRE will especially aid these efforts, as they are sampled regularly, overlap considerably in their zooplankton communities, but differ substantially in physical and chemical characteristics.

Figure 1 .
Figure 1.Recorded locations for Bosmina coregoni in the Pacific Northwest of the United States (2008-2012) (See Appendix 1 for details).
Bosmina coregoni was first detected in LCRE samples on November 22, 2008 and occurred regularly in samples through 2011.Maximum density occurred on July 21, 2009 at 0.123 organisms L -1 .Bosmina coregoni was present at low density relative to other cladocerans in LCRE samples (Bollens, unpublished data).

Figure 4 .
Figure 4. Mean body lengths (mm) of Bosmina and Daphnia spp.collected from Lake Washington between June and October 2011.Multiple bars per species correspond to samples collected on different dates.Error bars denote standard deviations.