Biodiversity wanes in New York

Throttling invasive species by TJ Sinnott

Bulwark for the Great Lakes and Hudson River by P Gerrity

Stopping ballast water "stowaways" by D Pughiuc

Biological pollutants in the Great Lakes by EL Mills, KT Holeck

Water quality signatures and the zebra mussel invasion by DA Matthews, SW Effler

Zebra mussel population dynamics: Implications for water quality modeling by CL Lange, DR Opdyke, JC Powers

Bad seeds: an introduction to invasive plants by AD Halpern, CA Boesse, AE Altor

You can help stop the plant invasion

President's message by D Ellis

Executive director's message by P Cerro-Reehil

People and places

NYWEA calendar

Sponsors at 73d Annual Meeting

Spring 2001 — Vol. 31, No. 1

Water quality signatures and the zebra mussel invasion

The zebra mussel (Dreissena polymorpha, Pallus) is a small bivalve mollusk that is native to southern Russia (Ludyanskiy and others 1993). This exotic invader was introduced to the Great Lakes in the mid-1980s, probably in water ballasts from a foreign ship (Hebert and others 1989, Mackie 1991). The spread of the zebra mussel has been rapid in North America in part because of the lack of natural ecological constraints (Ludyanskiy and others 1993) and its high reproductive rate (Ramcharan and others 1992a). By 1991 it had spread to all the Great Lakes and a number of major adjoining river systems. The zebra mussel is continuing to spread through much of the East, South, and Midwest of the United States and Canada (New York Sea Grant 2000).

Zebra mussels have been particularly successful in colonizing the hard waters of central New York (New York Sea Grant 2000). The Seneca River, both upstream and downstream of Cross Lake (Figure 1), was colonized by 1991. Particularly dense populations of zebra mussels have persisted downstream of Cross Lake (Effler and Siegfried 1994), and substantial population densities have been documented downstream to the Oswego River (Effler and Siegfried 1998). Zebra mussels have also colonized other adjoining systems, including the Finger Lakes, Oneida Lake, Oneida River, and Cross Lake. Densities in the colonized systems appear to have been limited only by the availability of solid substrate and suitable food.

Figure 1a. Seneca River system: Study reach of Seneca River including monitoring sites S1-S9.

Figure 1b. Three River System: Study reach within the larger river system and the Oswego River basin of New York, including Onondaga Lake.

Dense populations of the zebra mussel have been demonstrated to cause major changes in common measures of water quality associated with various aspects of their metabolism. For, example, decreases in concentrations of phytoplankton and tripton (nonliving particulate material), and attendant increases of light penetration, as a result of filter feeding of this exotic invader, have been reported for a number of invaded sites (for example, Reeders and Bij de Vaate 1990, Holland 1993, Leach 1993, Ludyanskiy and others 1993, Fahnenstiel and others 1995, Howell and others 1996, Caraco and others 1997, Effler and Siegfried 1998). Increases in ambient concentrations of ammonia and dissolved phosphorus have been attributed to excretion inputs from zebra mussels at several sites (for example, Holland and others 1995, Johengen and others 1995, Mellina and others 1995, Arnott and Vanni 1996, Effler and others 1996, Caraco and others 1997, Effler and others 1997, Effler and Siegfried 1998). Oxygen depletion associated with respiration of this bivalve has been documented in two large rivers (Effler and Siegfried 1994, Effler and others 1996, Caraco and others 2000).

Biology of the zebra mussel

A brief summary of the biology of the zebra mussel is presented in Table 1. The typical life span of the zebra mussel is 2 to 5 years. Most adults reach a shell length of 25 to 35 mm. The mussels attach to solid substrates with numerous byssal threads. Localized dispersal of the zebra mussel is achieved during the planktonic larval (veligers) stage. Wide-scale dispersal is enhanced by man's activities, including shipping and recreational boating (Ludyanskiy and others 1993).
 
Table 1. Zebra mussel (Dreissena polymorpha) biology fact sheet (modified from Effler and others 1996b).

Feature	Description/comments
Life span	2 to 5 years
Adult shell size	25 to 35 mm long
Attachment to substrate	sessile organism; byssal threads to solid substrate
Reproduction	2 to 5 spawnings/yr; fertilization at 12 to 24° C
Dispersal	swimming larvae-veligers; in plankton 1 to 5 wk
Feeding	filter feeder, particles > 1 µm diameter
Waste production	feces, psuedofeces [rejected (for example, inorganic) particles], and enriched excretions
Environmental requirements	salinity <=4 ppt; pH >=7.3; Ca2+ conc. >=20 mg/L; solid substrate; food particles (for example, phytoplankton)

Zebra mussels have certain environmental requirements that limit their occurrence (Ramcharan and others 1992b, Ludyanskiy and others 1993). They require alkaline (pH >=7.3) waters with adequate calcium concentrations (for example, [Ca²⁺] > 20 mg/L) to support shell formation and growth. Zebra mussels require solid substrates for colonization (Ramcharan and others 1992b, Ludyanskiy and others 1993); rock surfaces support particularly dense populations in infested systems (Mackie 1991, Effler and Siegfried 1994).

Filter feeding/material cycling

Zebra mussels are efficient filter feeders; particle removals of 100% have been reported for particles greater than 1 µm. It is this characteristic of the zebra mussel that is responsible for the improved clarity attributed to its invasion of particle-rich systems (Reeders and Bij de Vaate 1992, Leach 1993, Ludyanskiy and others 1993). Particles not selected for consumption are embedded in mucus in the mussel and discharged as pseudofeces (waste solids). Organic particles (for example, phytoplankton) taken in by mussels are rich in nitrogen and phosphorus as well as organic carbon (Hecky and others 1993). A potentially important net effect of zebra mussel metabolism is the conversion of particulate (unavailable) forms of nutrients to dissolved (available) forms. This conversion can have water-quality implications in infested and downstream systems.

Respiration/oxygen demand

The zebra mussel uses oxygen (respiration) to support its basic maintenance and active feeding needs (Schneider 1992) and releases carbon dioxide as a byproduct of this activity. Concern for the oxygen effect of the invasion focuses on streams and rivers (as opposed to lakes), because these environs are used to assimilate oxygen-demanding wastes and are generally more sensitive to additional oxygen demand (Thomann and Mueller 1987).

Seneca River: water quality effects of zebra mussel invasion

Seneca River, an alkaline, hard water system (Effler and others 1989), drains 9,000 km² of the 13,200-km² Oswego River basin of New York including the Finger Lakes region. The Seneca and Oneida Rivers combine to form the Oswego River, the second largest inflow into Lake Ontario. Bottom surveys of the Seneca River conducted in 1993 established that zebra mussels had colonized essentially all available solid (mostly rock) substrate > 3 cm in diameter along the bottom of the river from Cross Lake to Baldwinsville (Effler and Siegfried 1994). Particularly dense populations of zebra mussels (33,000 to 61,000 individuals/m²) were observed in the "State Ditch Cut" (Cut; Figure 1), a 1.7-km channel constructed to support navigation (Effler and Siegfried 1994). The size structure of this community indicated most of the population had been established in 1992 and 1993 (Effler and Siegfried 1994).

Major shifts in water quality reported for the Seneca River in 1993 were attributed to the effects of filter feeding, excretion, and respiration of this invader (Effler and others 1996). It was hypothesized that the conditions in this river, including anthropogenic effects and the presence of an intervening hypereutrophic lake, Cross Lake, promoted particularly severe water quality effects (for example, approaching a worst case or end-member) and long-term stability for this invader (Effler and Siegfried 1994, Effler and others 1996). We document here the major shifts in water quality over a ~20 km reach of the Seneca River brought about by the zebra mussel invasion, that have persisted 7 years.

Long-term water quality patterns

Long-term patterns are evaluated here within the context of annual mean values for various parameters at two sites bounding the Cut; temporal variability within years is represented by (minus) one standard deviation (Figure 2). Conditions for soluble reactive phosphorus (SRP; considered immediately available to support phytoplankton growth), ammonia (T-NH₃; total) and chlorophyll (Chl; the most commonly used surrogate measure of phytoplankton biomass) at site S3, just above the Cut, have reflected Cross Lake characteristics over the time of record. The similarities of the average conditions for all the parameters at S9, below Baldwinsville, with measurements at S3 before the invasion establish that these features remained largely uniform over this portion of the study reach before the zebra mussel invasion. The pre-invasion conditions were characterized by generally high concentrations of phytoplankton, low clarity, relatively low concentrations of available (for example, dissolved) nutrients, and nearly saturated dissolved oxygen (DO) concentrations.
 
Click here to see Figure 2 in a new window. Move the window to view text simultaneously.

Conspicuous changes occurred in these water quality characteristics over the ~16 km portion of the study reach that extends from the upstream boundary of the Cut (site S3) to S9 (Figure 1) following the invasion (Figure 2). On average, concentrations of SRP (Figure 2b) and total ammonia (Figure 2c) increased ten- and four-fold, respectively, downstream of Baldwinsville (S9) since the invasion. Concentrations of Chl at this location have decreased, on average, more than eight-fold (Figure 2f) and Secchi disk transparency (SD) increased more than two-fold (Figure 2h). Concentrations of DO have decreased, on average, about 20% (Figure 2j).

Post-invasion longitudinal patterns

Distinct longitudinal patterns have been manifested in affected parameters from the Cut (S3) to Baldwinsville (S9) following the zebra mussel invasion; DO and Chl concentrations decreased, and total ammonia and SRP concentrations and SD increased, with the approach to Baldwinsville (Figure 3). The progressive decrease in Chl from ~45 µg/L at S3 to < 5 µg/L about 4 km upstream of Baldwinsville (S7) on August 3, 1994 (Figure 3) reflects losses exceeding gains from phytoplankton growth.
 
Click here to see Figure 3 in a new window.

The observed decrease in DO along this reach (Figure 3c) reflects the operation of a sink process(es), that exceeded sources (for example, photosynthesis and reaeration). The oxygen concentration(s) in the downstream portion of the study reach on this day (Figure 3c) corresponded to ~40% saturation (T = 24°C), indicating the continued operation of a sink process of substantial magnitude (for example, compensating for reaeration inputs). Dissolved oxygen concentrations < 4 mg/L have been observed in this portion of the river on a number of occasions since the invasion.

Profiles of SRP (Figure 3d) and total ammonia (Figure 3e) depict sources of these constituents that far exceed potential sink processes (for example, plant uptake, and additionally nitrification for total ammonia) for a 7 km length of the river extending downstream from the upstream boundary of the Cut. The systematically higher concentrations in bottom samples during low flow intervals in 1994 (for example, Figures 3d and e), particularly at the downstream boundary of the Cut (S4), depict a source process operating at the river bottom.

Zebra mussels as the cause of observed changes

Changes in water quality characteristics of the Seneca River downstream of Cross Lake since the early 1990s and the spatial structure imparted (Figures 2 and 3) are likely a result of the zebra mussel invasion. First, the changes are temporally coincident with the invasion. Second, the features of the changes are consistent with the metabolism of the zebra mussel and the character of observations reported for other invaded systems with dense populations. These changes reflect at least three features of the metabolism of the invader:

• Respiration (decreases in DO)
• Filter feeding (decreases in Chl and increases in SD)
• Excretion (increases in SRP and total ammonia).

The persistence of high densities in the Cut supports the position that the conditions in this portion of the river promote high densities and population stability for zebra mussels (Effler and Siegfried 1994; Effler and others 1996). The European experience with the zebra mussel invasion indicates substantial variability in the long-term dynamics for individual systems and wide inter-system differences in temporal patterns (for example, Ramcharan and others 1992a). Within the context of those observations, the record for this part of the Seneca River depicts relatively stable conditions. This is probably attributable in part to the relatively stable upstream food source (phytoplankton) provided by hypereutrophic Cross Lake.

The zebra mussel invasion has converted the Seneca River downstream of Cross Lake from a low clarity, phytoplankton-rich, nutrient-depleted system, with nearly saturated oxygen concentrations, to a system with increased clarity, low phytoplankton levels, highly enriched in dissolved nutrients, with substantially undersaturated oxygen concentrations (Figure 2). These changes for this reach of the Seneca River are largely responsible for shifts in water quality of similar character reported (Effler and Siegfried 1998) for the mouth of the Oswego River and Oswego Harbor more than 50 km downstream. The water quality signature imparted by the invasion near Baldwinsville is the strongest encountered in our review of the literature. Further, there is no evidence of a waning in these effects over the post-invasion interval.

Onondaga Lake: water quality effects of zebra mussel invasion

Onondaga Lake and domestic waste inputs

Onondaga Lake is a hypereutrophic, alkaline, hard water, stratifying system located in Metropolitan Syracuse, New York (Figure 1). This lake has a volume of 131 x 10⁶ m³, a surface area of 12.0 km², and a maximum depth of 20 m. The lake discharges through a single outlet to the Seneca River.

Onondaga Lake was oligo-mesotrophic before European settlement in the late 1700s (Rowell 1996). Increasing inputs of domestic and industrial waste that accompanied urbanization in the watershed led to severe degradation and loss of uses of the lake (Effler 1996). The commercial cold-water fishery was eliminated by the late 1800s (Tango and Ringler 1996), and the lake was closed to ice harvesting in 1901, swimming in 1940, and fishing in 1970 (Effler and Harnett 1996). Onondaga Lake has been described as "the most polluted lake in the United States" (Hennigan 1990).

The Metropolitan Syracuse Wastewater Treatment Plant (Metro) discharges to the southern end of Onondaga Lake (Figure 1) via a surface outfall. It presently serves about 300,000 residents of Onondaga County, NY, as well as a number of local industries. This discharge [average flow ~3.0 m³/s (68 MGD)] represents nearly 20% of the annual inflow to the lake and often is the single largest inflow in late summer (Effler and others 1996a). Through 1997 Metro contributed approximately 90% of the total external load of total ammonia to the lake (Effler and others 1996a, Gelda and others 1999).

Although zebra mussels have been present in Onondaga Lake since at least 1992, their densities remained very low (~1 individual/m²) in the near-shore zone through 1998 (Spada 2000). These low densities appeared to be inconsistent with the available food (high concentrations of phytoplankton biomass; Effler 1996) and solid substrate (remained largely uncolonized; Spada 2000) in the lake. Veliger surveys and colonization experiments in 1997 and 1998 indicated the Seneca River was a source of zebra mussels for the lake but that survivorship in the lake was poor (Spada 2000). Spada evaluated several features of the lake's chemistry, commonly considered as influencing the distribution of this invader (Table 1), and concluded the pH, DO, salinity, and calcium conditions of Onondaga Lake were not limiting. Ammonia concentrations were identified as potentially limiting to zebra mussels in the lake (Spada 2000).
 
Click here to see Figure 4 in a new window.

Reductions in ammonia loading

Daily loads of total ammonia and total phosphorus (TP) from Metro are presented here for the April-October interval of 1999 (Figure 4a). These are compared to distributions of monthly average loads (represented by "box" plots) from Metro for the 1989-1997 interval to place the recent reductions in a historical context (Figure 4). With the exception of May, dramatic reductions in total ammonia loading from Metro were achieved (Figure 4a). The primary driver of these dramatic reductions in loading was improved treatment of total ammonia at Metro (Effler and others 2000a), largely associated with a recent upgrade in the aeration system.

Dramatic decreases in total ammonia concentrations in the lake's upper waters occurred in the summer of 1999 (Figure 4b) in response to the reductions in loading from Metro (Figure 4b). The concentrations in 1999 were the lowest observed in the lake over the record by a wide margin. Concentrations of total ammonia in 1999 remained below the New York State standard for non-salmonid systems from mid-July through the end of the study period (Figure 4b). This improvement is particularly noteworthy in light of the severe and extended violations of this standard that have prevailed annually in this system (Effler and others 1990 and Matthews and others 2000).

Zebra mussel invasion

Numerous small (< 1 year old) zebra mussels were found on a variety of artificial substrates (for example, buoys and fish nets) in the lake during the fall of 1999. Benthic sampling conducted in 2000 established that hard substrates in the littoral zone had experienced dense colonization (up to 12,000 individuals/m²; Table 2; Figure 5). The mussels collected in 2000 represented essentially two size classes, corresponding to colonization during the summers of 1999 and 2000. The timing of the sudden zebra mussel invasion coincides, and almost certainly is associated, with the dramatic decrease in the upper water concentrations of ammonia that occurred in 1999 (Figure 4b) as a result of reductions in loading from Metro (Figure 4a).
 
Table 2. Density estimates of zebra mussels in summer 1997, 1998, and 2000 at three locations in the near-shore zone of Onondaga Lake.

Site	1997/1998 (individuals /m²)	2000 (individuals /m²)
1	< 1	5,984
2	< 1	6,768
3	< 1	12,208

Figure 5. Zebra mussel encrusted rock from the shoreline of Onondaga Lake, fall 1999.

Zebra mussels reared under laboratory conditions are extremely sensitive to ammonia levels > 1 mgN/L (Nichols 1992). Nichols (1992) found that total ammonia concentrations of 2 mgN/L caused severe stress in lab-reared zebra mussels, and concentrations of 3 mgN/L caused 90-100% mortality. Coon and others (1993) reported that total ammonia concentrations exceeding 3 mgN/L cause 100% mortality in veligers. The high concentrations of T-NH3 maintained in the upper waters of Onondaga Lake (> 2 mgN/L) prior to 1999 apparently prevented the dense colonization of this system by zebra mussels.

Microcystis bloom

Onondaga Lake experienced a major, and largely unprecedented, cyanobacteria (previously bluegreen algae) bloom during July and August 1999 (Figure 4c). This bloom, dominated by Microcystis (~80% of the cyanobacteria biomass; Figure 4c), represents the single greatest change in the phytoplankton community, within the context of water quality management, since the late 1980s. The average Chl concentration for July and August was the second highest observed since the late 1980s and SD values dropped below the New York State standard for swimming safety of 1.2 m. Large accumulations of Microcystis developed as surface scums along the lake's shores that created noxious conditions (Figure 6). This shift in the phytoplankton community does not appear to have been driven by bottom-up processes, such as changes in nutrient concentrations or nutrient ratios.

Figure 6. Noxious surface scums that formed along the shoreline of Onondaga Lake during the late summer Microcystis bloom of 1999.

The severe late summer Microcystis bloom of 1999 was probably driven primarily by a top-down process, most likely associated with the coincident invasion of zebra mussels. The occurrence of Microcystis blooms has been reported elsewhere after the establishment of zebra mussels; for example, Saginaw Bay, Lake Michigan (Vanderploeg and others 2000), western basin of Lake Erie (Budd and others 1999), Gull Lake and Gun Lake in Michigan, and the Bay of Quinte, Lake Ontario (Vanderploeg and others 2000). Zebra mussel metabolism apparently selects for "unpalatable" strains of Microcystis that occur as large colonies, such as observed in Onondaga Lake in 1999. These colonies have been observed to be rejected by the mussels after filtration as loosely consolidated (viable) psuedofeces, while smaller more desirable algae are ingested (Vanderploeg and others 2000).

The expansion and growth of the lake's zebra mussel population could have potentially important implications for the lake's food web and measures of water quality. Direct or indirect interactions with the invader may influence the structure of various biological communities, including rooted plants (Skubinna and others 1995), zooplankton (Strayer and others 1999), fish and benthos (MacIssac 1996). Filter feeding by this invader may also influence water quality indicators of the hypolimnion, such as the rate of oxygen depletion, by routing much of the organic material that had previously reached the lower layers (via deposition) to the near-shore zone instead.
____________
David A. Matthews is a research scientist and Steven W. Effler is director of research for the Upstate Freshwater Institute. Corresponding author is David A. Matthews.

Click here for the references to this article.