The spiny water flea Daphnia lumholtzi (Sars, 1885) (Order Cladocera, family Daphniidae) is a potentially invasive species native to Eastern Africa, Australia and the subcontinent of India (Benzie 1988). Though its presence was first recorded in an eastern Texas reservoir in 1991 (Sorensen and Sterner 1992), researchers believe D. lumholtzi was accidentally introduced to the United States in a shipment of Nile perch (Lates niloticus; Linnaeus, 1758) in 1983 (Havel and Hebert 1993). Since its introduction, D. lumholtzi has spread to over 170 lakes and reservoirs in the U.S. ranging from Florida to Arizona, and as far north as the Laurentian Great Lakes (Lennon et al. 2001; Havel et al. 2005; Havel and Graham 2006). The ephippia of this species resist desiccation, making rapid migration between water bodies common. While the complete distribution of D. lumholtzi in North Carolina is unknown, its presence has been documented in three reservoirs in western NC: Lake Norman, Cedar Creek Lake, and Greenwood Lake (Havel and Shurin 2004; Fig. 1). This is the first report of D. lumholtzi in B. Everett Jordan Lake, a reservoir in central NC.
Though D. lumholtzi can survive in a variety of aquatic environments ranging from freshwater lakes to rivers and streams, the full effects of D. lumholtzi establishment are still unknown. Some studies suggest that D. lumholtzi is a weak competitor, particularly in the presence of Chydorus sphaericus, another cladoceran species (Lennon et al. 2003); others suggest its size, reproductive ability, and high temperature tolerance give it a competitive advantage over native species (Engel and Tollrian 2009). Additionally, D. lumholtzi undergoes cyclomorphosis, where it grows long head and tail spines, making the species less vulnerable to predation by young fishes and predatory invertebrates (Frisch and Weider 2010). Other studies have shown, however, that D. lumholtzi may become an important food source for juvenile fish when native zooplankton populations are low during the summer months (Lienesch and Gophen 2005). The ecological effects of D. lumholtzi in Jordan Lake cannot be determined without data on its abundance over time. Our objective is to report the population dynamics of D. lumholtzi among native zooplankton in B. Everett Jordan Lake, an impoundment of the Haw River located in the upper Cape Fear River drainage basin in Chatham and Durham counties in central NC.
Macroscopic zooplankton samples were collected from the Farrington Point boat ramp near the Bush Creek inlet at the northern end of Jordan Lake (Fig. 1) monthly between 19 July 2010 and 10 August 2011 using a plankton net (30.5 cm diameter and 243 µm mesh size). Samples were collected near the middle of each month, except in August 2010 when samples were collected on the 5th and 23rd. All sampling was done between 930 AM and 1100 AM when migrating zooplankton may have been at deeper depths, so the net was lowered vertically to the lake bottom to sample the entire water column (approximately 2.5 m deep—total sample volume of 160 L). Four replicate plankton tows were collected and preserved in a 90% ethanol solution for analysis (but only two replicates were collected on 19 July 2010). Additionally, subsurface water temperature and dissolved oxygen (DO) were measured using a YSI 550 dissolved oxygen meter (although some DO data were discarded due to instrument malfunction), and water clarity was measured using a Secchi disk 18 September 2010 through 10 August 2011. Copepod zooplankton were identified to order (calanoids or cyclopoids), and cladoceran zooplankton were identified to genus using Pennak (1989). D. lumholtzi was identified to species level using Benzie (1988). The zooplankton were then counted under a dissecting microscope. For samples with total densities >100 L−1, three subsamples comprising ∼5–10% of the total were counted to estimate the total sample. Using these counts, the average density of each species group was calculated.
Zooplankton samples were also collected from the Vista Point boat ramp and New Hope Overlook boat ramp (Fig. 1) on 28 July 2010 and 3 August 2010, respectively, using the same 30.5 cm diameter 243 µm mesh net and were preserved in a 90% ethanol solution. The zooplankton were examined under a dissecting microscope and D. lumholtzi was identified using Benzie (1988). The collection depths, however, were not measured, so the densities for these samples could not be calculated.
RESULTS AND DISCUSSION
The population of D. lumholtzi remained higher than other zooplankton populations at Farrington Point from July 2010 to November 2010, with more than half of the zooplankton community consisting of D. lumholtzi (Fig. 2). Total zooplankton densities during this time ranged from <1–2 L−1 (Fig. 2b inset). The D. lumholtzi population then declined in December and was not present in samples collected between January and May, although the native zooplankton populations thrived (Fig. 2b). Native cladocerans were dominated by Diaphanosoma spp., Daphnia spp., Bosmina spp., and Ceriodaphnia spp. D. lumholtzi reappeared in early June 2011, but failed to re-establish itself and was not observed in any midsummer samples.
In additional to Farrington Point, D. lumholtzi was positively identified at both Vista Point and New Hope Overlook, two locations near the southern end of Jordan Lake (Fig. 1) during July and August 2010, respectively. Although densities cannot be calculated from these qualitative samples, the presence of D. lumholtzi at both the northern and southern regions of Jordan Lake suggests that the population may have been established throughout the lake during the late summer of 2010.
The fluctuations in the D. lumholtzi population may result from seasonal temperature changes. The surface water temperature dropped to 5°C by 13 December 2010 (Fig. 3), coinciding with the disappearance of D. lumholtzi, which performs poorly below 10°C in lab experiments (Lennon et al. 2001). Record high water temperatures between July and August 2011 are believed to have caused striped bass (Morone saxatilis) kills (Sorg 2011), which could have influenced populations of planktivorous fishes. Also, Leptodora spp., a large (ca. 1 cm diameter) predatory cladoceran which fed selectively on D. lumholtzi in lab experiments (Effert and Pederson 2006), was observed in July and August 2011 samples, but not on other dates. Increased predation by Leptodora could have contributed to the absence of D. lumholtzi in samples collected after June 2011.
Future research directions should include continued monitoring of the D. lumholtzi population in Jordan Lake, especially with respect to native zooplankton populations, planktivorous fishes and environmental variables. We recommend sampling of additional sites throughout the lake, including pelagic habitats, as our observations at three shoreline sites may not be representative of lake-wide patterns. This research could lead to an enhanced understanding of the conditions under which this non-native species might become invasive within Jordan Lake and other eastern freshwater ecosystems.
We thank undergraduate students in Dr. Sandra Cooke's Writing 20 class for their contribution to data collection and analysis, Terry Corliss of the Biology Department at Duke University for providing microscopes and lab space, Dr. Christine Erlien for assistance in creating the maps, and an anonymous reviewer for helpful feedback on an earlier version of this manuscript. This research was supported by a Trinity College of Arts and Sciences course enhancement grant from The Duke Endowment.