Conductivity Limits Survival and Growth
of the New Zealand Mud Snail
from the Upper Owens River, California
David B. Herbst1
Michael T. Bogan
Robert A. Lusardi
Sierra Nevada Aquatic Research Laboratory
University of California
HCR 79, Box 198
Mammoth Lakes, CA 93546
1. Contact: herbst@lifesci.ucsb.edu
760.935.4536
December 21, 2006
Final Report to California Trout
Summary (Abstract)
The New Zealand Mud Snail (NZMS) is an invasive exotic species of aquatic snail that is becoming widespread in inland and coastal waters of the western United States. Where populations have become abundant, numbering in the hundreds of thousands per square meter, they may consume a large fraction of available algae food resources, and potentially compete with and displace native invertebrates and thereby disrupt aquatic food webs that sustain recreational fisheries. This snail likely spreads through both transport by humans on fishing gear or boats, and movement of snails within habitats where they have been introduced, and may colonize as single individuals that reproduce through asexual clones. Tracking of the distribution and abundance of the NZMS is revealing a rapid expansion but little is understood about the types of habitats that are most vulnerable and the conditions that permit snails to thrive. Observational studies of the NZMS invasion on the Upper Owens River (east of the Sierra Nevada in California) suggested that the snail may be limited to certain water chemistry conditions, so experimental studies of growth and survival were undertaken to evaluate survival and growth as a function of mineral content of the water. Snails were collected from the Upper Owens River, targeting an early life stage cohort of uniform size-class, and reared in dilutions of natural river water adjusted to 10, 50, 100, 200 and 300 µS conductivity. In addition, calcium-free artificial river water was prepared at 200 µS to test for the independent effect of limitation of this mineral ion required for shell-building. Experiments were also conducted with newborn clones raised in river water dilutions ranging from 25 to 200 µS to examine mortality and growth at this sensitive stage of development. Low survival and no growth was found at and below 25 µS and significant graded reductions in growth occurred as river water was diluted from 300 to 50 µS conductivity. Growth was also inhibited in Ca+2-free artificial water compared to the same conductivity of natural river water, indicating that lack of this mineral impedes development. Even though growth was reduced in Ca+2-free media, it was lower still in 50 µS river water, suggesting that dilute water even with Ca+2 may inhibit growth through osmotic stress and/or imbalance in other ions. These results show that the distribution and productivity of the NZMS may be restricted by ecophysiological constraints. The chemistry of water may confine invasion of the NZMS to conductivities above 25 µS, and produce physiological debilitation over the range 25 to 200 µS that would place snails at an energetic disadvantage in competition for food or space with native benthic organisms. Productive populations are likely those that inhabit water of high alkalinity (CaCO3 content), where habitat disturbance has also undermined the viability of native benthic invertebrates. This paradigm provides a framework for predictive modeling of the potential vulnerability of streams and rivers to invasion, a mechanism for understanding where colonization and expansion has occurred, and potential controls on NZMS infestations.
Introduction
Since the initial introduction of Potamopyrgus antipodarum (NZMS) in the mid-1980s into streams of the western interior United States, there has been growing concern over how this exotic species might alter freshwater ecosystems. As the range and incidence of the snail invasion has expanded, studies have been undertaken on how to decontaminate fishing gear thought to be a primary vector for dissemination (Richards et al. 2004, Hosea and Finlayson 2005), conduct public education campaigns (www.protectyourwaters.net), and examine how the snails affect aquatic ecosystems (Hall et al. 2006). Unfortunately there are few studies outside of New Zealand (Winterbourn 1970, Broekhuizen et al. 2001, Schreiber et al. 2003, Suren 2005) that examine basic habitat requirements that might enable an understanding of where and how invasion has or will occur. Mapping of the spread of the NZMS has enabled tracking of new records, but the absence of systematic physicochemical data collection at these sites severely hinders the utility of this information for revealing shared patterns of habitat type underlying the distribution.
Is this an invasive exotic on the verge of ecological explosion? The problem is widely perceived in the potential for snails to displace native species of stream invertebrates, altering ecosystem structure and function, and threatening recreational fisheries and indigenous fish through loss of food resources. The impression given in many information sources is that the snail can occur in almost any kind of habitat from inland freshwater streams to coastal estuaries (internet link for basic information may be found at www.esg.montana.edu/aim/mollusca/nzms), but the associated environments for distribution and abundance of the mud snail are known only in general outline. Areas of concentration in the western US include the Snake River system (ID), Greater Yellowstone area (MT, WY), Bonneville basin (UT), coastal estuaries (CA, OR), lower Colorado River (AZ), and Owens River (CA). Among the traits that make the NZMS especially problematic as an invasive species is that it typically reproduces through asexual cloning, gives birth to live young from an internal brood chamber, can resist desiccation through closure of the shell opening by an operculum (a hardened “door” that closes behind it when the snail retracts into its shell), survives a wide range of temperatures, thrives in sediment-laden habitats (the mud in mud-snail), and is capable of living in environments ranging from mountain lakes to brackish seawater conditions. The study presented here provides an explicit test of how snails perform with regard to this last aspect, over a conductivity range in river water.
Introduction of the NZMS into the Upper Owens River likely occurred in 1998. The snail has since expanded downstream and upstream to apparent boundaries at Crowley Reservoir below, and Big Springs above, an area of groundwater-dominated inflow where conductivity increases 4-fold (from 60 to 230 µS, Figure 1). Though habitat disturbance from livestock grazing, bank erosion, sedimentation, and geomorphic modification associated with augmented flows and diversions likely plays a role in NZMS occurrence in the Upper Owens (Herbst, unpublished), the continued absence of snails in the more dilute tributary waters of Deadman Creek (above Big Springs) suggest snails might not survive in waters of more dilute mineral content than the river. The objective of this research was to test the hypothesis that physiological capacity for survival and growth is controlled by ionic and osmotic requirements for shell building and metabolic homeostasis. If confirmed, this hypothesis would provide a simple mechanistic explanation of the apparent limitations on NZMS distribution: NZMS require higher conductivity water to facilitate survival and growth. We also outline how these results may be used to interpret known distribution and anticipate future invasions.
Methods
Conductivity Range-Finding Experiment
NZMS were collected from the Upper Owens River at Benton Crossing Road using a D-frame sampling net of 500-micron mesh in late March 2006. Sediment and macrophytes were removed in the field and mud snails were then sorted in the laboratory to obtain a size cohort of approximately 1 mm spire height (range: 0.75-1.5mm). Fifty mud snails were randomly ed from the 1 mm size cohort and sacrificed for measurement of initial length and weight distributions. All length measurements were made with an eyepiece micrometer under a stereoscope at 10X to ± 0.01 mm. Dry weights of individual snails were obtained using a Cahn Electrobalance after drying at 70C for thirty-six hours.
Water collected from the Upper Owens River at Benton Crossing (conductivity = 365S) was used to prepare all experimental treatment solutions. After filtering water through 0.45 micron GF/B filters, the river water was diluted with de-ionized water to create treatment solutions of five conductivity levels: 300, 200, 100, 50 and 10 S. An artificial 200 S calcium-free solution was also prepared using a combination of the following salts: 65.9 mg/L NaHCO3, 15.5 mg/L KCl, 38.6 mg/L MgCl2-7H2O and 46.5 mg/L MgSO4-7 H2O (final concentrations after adjusting to 200 S. This artificial media was prepared to test the role of Ca+2 independent of conductivity in altering growth and survival of snails. Conductivity was measured with a portable Oakton 35630-00 pH/conductivity meter throughout the entire experiment (temperature compensated readings at 25 °C reference).
Benthic algae were collected from shallow river margin substrates at Benton Crossing on the Upper Owens River to serve as a food source during the experiment. Collected periphyton were sieved through a 2 mm mesh screen. Algae and fines passing the sieve were placed in a volume of 2 L natural river water and 10ml/L of NaNO3 added to stimulate algal growth. Eight clear plastic culture containers were then filled with 250ml of this algae and nutrient solution and placed in direct sunlight. Any snails found were picked out of the culture containers until none remained. After approximately 10 days, at the beginning of experimental treatments, algae cultures were combined, allowed to settle, supernatant poured off, and remaining algae centrifuged for 5 minutes. Pelleted algae were combined and re-suspended in de-ionized water to achieve an algal ‘paste’ for use as an initial food source in each treatment dish.
Eight replicate plastic Petri dishes were prepared for each of the six conductivity treatments. Each replicate dish was prepared by combining 30 ml of treatment solution with 1 ml of algal paste. Twenty-five randomly ed mud snails (of the 1 mm cohort) were then added to each replicate dish. All replicate dishes were examined for mud snail mortality, approximately every four days, and mortalities were removed. Mortality was judged as an absence of response to light tactile stimulation (touching soft tissue with a probe). One replicate dish from each conductivity treatment was sacrificed at about weekly intervals to create length and weight growth curves over the course of the experiment. To ensure that water quality did not degrade over the course of the experiment, 15 ml of water was removed from each dish and replaced with fresh treatment solution every three to four days. Air temperature in the culture room was recorded daily. The experiment continued as described for 7 weeks. The mud snails in the eighth and final dish for each treatment were allowed to grow for 2 additional weeks before sacrifice.
Newborn Clone-Initiated Experiment
After establishing a response range in the first experiment, a narrower range of conductivities were tested using clonal newborn mud snails which are presumably more sensitive to varied environmental conditions and so should provide a more realistic assessment of ecological limits in nature. Mature adult NZMS were collected from the Owens River at Benton Crossing in October 2006 as a brood-source for newborn clones. These snails were monitored regularly and new clonal progeny were collected within a 24-hour period for use in experimental treatments.
Six treatment solutions were prepared for the juvenile experiment. Five were created by diluting filtered Owens River water as previously described to obtain a conductivity series (200, 100, 75, 50, and 25 S), while the sixth was the 200 S calcium-free solution described above. One Petri dish was prepared for each of the six treatments with 30 ml of treatment solution and 1 ml of algae paste (prepared from fresh algae cultures as described above). Twenty-five new (24-hour cohort) clonal progeny were then added to each Petri dish (using light-weight broad-tip forceps) and checked twice to ensure all juvenile mud snails survived the transfer. In each Petri dish, 15 ml of treatment solution was replaced regularly, as described above. Solution removal was monitored under a microscope to ensure that juveniles were not accidentally removed from the dishes during solution replacement. Air temperature of the culture room was again monitored daily. All treatments were monitored daily for mortality, and dead snails removed. After 1 month, all snails were harvested and lengths were measured as described above.
Statistical Analyses
Differences in mortality were evaluated with cumulative mortality curves over the duration of the experiment to estimate relative mortality between treatment groups. This did not account for the removal of one dish from each treatment for growth harvests, but these were equal across treatments. Growth in terms of dry weight was compared among treatments in the range-finding experiments, and as shell spire length in the newborn clone experiment. Overall differences in growth rate regressions among treatments were tested using analysis of covariance, followed by Tukey HSD multiple comparisons to test for differences between treatments. Shell spire lengths of newborn snails after one month growth were compared by a Kruskal-Wallis test followed by paired treatment comparisons using Dunn’s procedure.
Results
In the initial range-finding experiment of varied conductivity effects of Upper Owens River water on survival and growth, the snail cohorts did not survive in the most dilute water of 10 µS (or were moribund and without any sign of growth), and showed mortalities at 50 and 100 µS but high rates of survival at 200 and 300 µS (Figure 2). Death appeared to be related to shell-thinning (fragile, whitened shell at growth margin), invasive growth of fungus on soft tissue, and general morbidity (slow or inactive, with no feeding). NZMS also exhibited significant and graded declines in growth rates with a decrease in treatment conductivity from 300 down to 50 µS over the 2-month course of the experiment (Figure 3). In addition to low conductivity limits on survival and growth, Ca+2-free culture media also reduced growth relative to natural Ca+2-containing water of equal conductivity (Figure 4). Analysis of covariance demonstrated significant between-treatment effects on growth (F = 56.205, p<0.0001), and using Tukey HSD multiple comparisons test, growth rates were ordered as follows: 300 = 200 > 100 = 200 Ca-free > 50 µS (p<0.0001). Though the decreasing series of conductivities were not all separated as significant, continuation of the experiment beyond 2 months would likely have produced differences, evident in the greater slopes of the regressions over the series (300>200≥100>200 Ca-free>50). These contrasts also show that while Ca+2-free 200 µS inhibited growth relative to natural 200 µS river water, this media still supported higher growth than river water at 50 µS that contained calcium ion [Upper Owens River water from the Benton Crossing collection site (365 µS) had an alkalinity of about 180 mg/L (as CaCO3), so the molar concentration of Ca+2 at 50 µS would be about 0.24 mM, but have a much lower osmotic concentration than the artificial 200 µS river water].
In the experiment using newborn clone cohorts, only a few snails survived the treatment at 25 µS, and those snails were moribund and showed no growth. Growth measured as shell spire length also was curtailed as conditions became more dilute (Kruskal-Wallis: K = 51.38, p <0.0001), with paired comparisons showing growth in the following order: 200=100 > 75=50=200 Ca-free (p<0.0001), but again these were in graded series indicating that had the experiment been extended, significant differences between treatments would probably have developed (Figure 5). As in the first experiment, Ca+2-free 200 µS media resulted in reduction in growth relative to the 200 µS river water treatment.
Discussion and Conclusions
There is very little information available from either field studies or experimental work to evaluate the relationship of dissolved solute content to NZMS distribution or growth and survival. In certain streams of the Greater Yellowstone area, where several published studies of abundant NZMS populations have been conducted (Hall et al. 2003, Hall et al. 2006, Kerans et al. 2005), conductivities are high – in the range of conductivities shown in this study to support high growth rates (Table 1). These populations can reach densities of hundreds of thousands of indivduals per square meter in local patches, similar to those in the Upper Owens River. Under these conditions of high conductivity and alkalinity the NZMS can also consume large fractions of algal production (75% in Polecat Creek WY, Hall et al. 2003) and dominate the benthic invertebrate community. In a detailed study of distribution along a gradient of stream sites in Yellowstone National Park (Kerans et al. 2005), NZMS were present at all locations below inflows from thermal groundwater sources, but were completely absent above these inflows where conductivities were much lower – in a range of growth inhibition (60-120 µS in the Firehole River and 50-200 µS in the Gibbon River). In the single experimental study where dissolved solute concentration was manipulated (Jacobsen and Forbes 1997), only very high conductivities of diluted seawater (5 to 15 ppt, or about 10,000 to 25,000 µS) were tested in relation to an artificial freshwater solution (est. 500-1,000 µS). While survival and growth to maturity were found in all these treatments, even this fairly high conductivity “zero salinity” freshwater treatment showed lower growth and viability relative to the dilute seawater treatments.
The range of conductivities tested in the present study is representative of a wide variety of streams found in mountainous areas with snow-melt-dominated hydrology, and many other streams where groundwater sources can dominate flows at least during parts of the annual cycle. The results indicate that no survival would be expected at conductivities less than 25 µS (common in many mid-to-high elevation Sierra Nevada streams), growth is inhibited over the range of 25 to 200 µS, and above 200-300 µS there is little or no restriction of conductivity on growth or survival. Reduced growth rates are likely to lead to later age at maturity, delay in reproduction, and consequent reduction in the intrinsic rate of population growth. Even if reproduction is not delayed, maturation would likely be at a smaller size with reduced fecundity and/or lower viability of newborn clones. Shell thinning at low conductivity river water and reduced growth in Ca+2-free water relative to the same conductivity of river water indicates that limitations on shell-building explain part of the growth inhibition of snails. Slower growth in 50 µS river water compared with 200 µS Ca+2-free water suggests that, even with calcium present, dilute conditions carry costs for maintaining internal osmotic and ionic balance against a high electrochemical gradient. Below about 100 µS (<0.5 mM) there may often be a small gradient for Ca concentration from external to internal fluids that requires active transport, as shown in the freshwater snail Limnaea stagnalis (Greenaway 1971). The gradient for acquiring ions such as sodium and potassium will require even greater energy costs since these are essential internal solutes but occur only in low concentration in river water. That alkalinity (CaCO3 content) can play an important ecological role has been shown in studies of freshwater gastropod distribution in Tennessee, one of the richest regions in the world for molluscan diversity. Below an alkalinity of 20 mg/L only 10% of streams held any snails, while above this level at least 40 to 75% of streams held a varied snail fauna (Shoup 1943).
We propose that the distribution and productivity of the NZMS is constrained by ecophysiological limitations and metabolic costs (Figure 6). The chemistry of water may confine invasion of the NZMS to conductivities above 25 µS, and produce physiological debilitation over the range 25 to 200 µS that would place snails at an energetic disadvantage in competition for food or space with native benthic organisms. Above 300 µS there may be no growth limitation with respect to calcium requirements for shell growth and minimal cost related to osmotic and ionic regulation. Beyond these physiological boundaries, ecological factors may be more important in NZMS population dynamics. Productive populations are likely those that inhabit water of high alkalinity (CaCO3 content), where habitat disturbance has also undermined the viability of native benthic invertebrates. Substrate quality, flow regime, sedimentation and algal production or type may also influence snail abundance.
This paradigm provides a framework for predictive modeling of the potential vulnerability of streams and rivers to NZMS invasion, identifies a mechanism for the colonization and expansion of NZMS populations, and suggests potential controls on NZMS infestations. Given the detailed data coverages of water quality measurements, including conductivity, that are available for western North America (USGS, various regional government water quality agencies), it should be possible to develop GIS-based predictive models and maps showing where new invasions of snails are most likely to occur and proliferate. Population fluctuations and presence/absence of the NZMS (and native benthic invertebrates affected directly or indirectly by snail competition), may also be interpreted in new light as related to spatial and temporal variations in conductivity. While control of conductivity in natural waters is seldom feasible, dissolved salts from stormwater runoff in urban areas and agricultural return flows in rural areas represent potential causes of increased conductivity. This type of mineral solute-polluted runoff could be regulated as a means of reducing the favorability of habitat conditions for NZMS infestation.
Acknowledgements
We are grateful to California Trout for funding support and recognition of the importance of the NZMS problem as a conservation issue. Thanks to Jeff Kane and Sandi Roll for assistance in setting up experiments, and to Sean Eagan and Bob Hall for data on river conductivities in the Greater Yellowstone area.
References
Broekhuizen, N., S. Parkyn, and D. Miller. 2001. Fine sediment effects on feeding and growth in the invertebrate grazers Potamopyrgus antipodarum (Gastropoda, Hydrobiidae) and Deleatidium sp (Ephemeroptera, Leptophlebiidae). Hydrobiologia 457:125-132.
Greenaway, P. 1971. Calcium regulation in the freshwater mollusc, Limnaea stagnalis (L.) (Gastropoda: Pulmonata). Journal of Experimental Biology 54:199-214.
Hall, R., J. L. Tank, and M. F. Dybdahl. 2003. Exotic snails dominate nitrogen and carbon cycling in a highly productive stream. Frontiers in Ecology and the Environment 1:407-411.
Hall, R.O., M.F. Dybdahl and M.C. Vanderloop. 2006. Extremely high secondary production of introduced snails in rivers. Ecological Applications 16:1121-1131.
Hosea, R.C. and B. Finlayson. 2005. Controlling the spread of New Zealand Mud Snails on wading gear. Unpublished report of the California Department of Fish and Game.
Jacobsen, R., and V. E. Forbes. 1997. Clonal variation in life-history traits and feeding rates in the gastropod, Potamopyrgus antipodarum: Performance across a salinity gradient. Functional Ecology 11:260-267.
Kerans, B. L., M. F. Dybdahl, M. M. Gangloff, and J. E. Jannot. 2005. Potamopyrgus antipodarum: distribution, density, and effects on native macroinvertebrate assemblages in the Greater Yellowstone Ecosystem. Journal of the North American Benthological Society 24:123-138.
Richards, D. C., P. O'Connell, and D. C. Shinn. 2004. Simple control method to limit the spread of the New Zealand mudsnail Potamopyrgus antipodarum. North American Journal of Fisheries Management 24:114-117.
Schreiber, E. S. G., G. P. Quinn, and P. S. Lake. 2003. Distribution of an alien aquatic snail in relation to flow variability, human activities and water quality. Freshwater Biology 48:951-961.
Shoup, C.S. 1943. Distribution of fresh-water gastropods in relation to total alkalinity of streams. The Nautilus 56:130-134.
Suren, A.M. 2005. Effects of deposited sediment on patch ion by two grazing stream invertebrates. Hydrobiologia 549:205-218.
Winterbourn, M. 1970. Population studies on the New Zealand freshwater gastropod, Potamopyrgus antipodarum (Gray). Proceedings of the Malacological Society of London 39:139-149.
Table 1. Conductivity of rivers in the Greater Yellowstone area where the NZMS has been found in abundance.
Name of River Conductivity Range µS (seasonal)
Firehole 300-600
Gibbon 150-550
Madison 200-550
Polecat Creek 155 (no seasonal data available)
Where NZMS is absent:
Upper Firehole (above geyser basin inflow) 60-120
Upper Gibbon (above geyser basin inflow) 50-200
Figure 1. Aerial photograph of Upper Owens River showing the varied conductivities in the mainstem of the river and tributaries. Note that the NZMS occurs in the mainstem river (in light blue), and in the Hot Creek tributary, but are not found in Deadman Creek or Mammoth Creek. Conductivities are shown from summer collections of 2004, but vary seasonally and annually within about ±10-20% of these values.
Figure 2. Relative mortality of NZMS among different conductivity treatments of diluted water from the Upper Owens River, except 200A which refers to calcium-free artificial river water adjusted to 200 µS.
Figure 3. Growth regressions of NZMS dry weight (µg) over 60+ days at varied conductivities (µS) of Upper Owens River water.
Figure 4. Growth regressions of NZMS dry weight (µg) over 60+ days at 200 µS of Upper Owens River water in comparison
to 200 µS of calcium-free artificial river water.
Figure 5. Growth of clonal newborn NZMS at varied conductivities of Upper Owens River water after 1 month.
Y-axis scaled based on an initial average size of 0.40 mm. Sample size in order = 16, 20, 21, 21, 19 snails.
Figure 6. Model of ecophysiological constraints on distribution, abundance and productivity of NZMS. Ionic and osmotic limits on metabolic costs to function and viability. Physiological debilitation over the range of growth inhibition reduces the competitive dominance that the NZMS can exhibit under physiologically optimal conditions.
