Seasonal variations in sediment and bottom water chemistry of western Long Island Sound: implications for lobster mortality.

ABSTRACT In the fall of 1999 the lobster population of western Long Island Sound (WLIS) experienced a massive die-off (LoBue et al. 2003). In fact, the lobster populations in WLIS still have not recovered as of this writing (Benway et al. 2004). Factors suspected of playing roles in this mortality event include water temperature, hypoxia and a neoparameoba. The overall objective of this study is to obtain an understanding of bottom water chemical conditions in WLIS, and their potential influence on benthic community structure and lobster health. Beginning in May 2002, water samples were collected within 5.0 cm and at 1.0 m above the bottom at 12 selected stations in WLIS and analyzed for dissolved NH4, H2S and O2. Grab samples for benthic community determination and sediment-profile images were simultaneously obtained with the water samples at 6 of the 12 stations. The data collected provide an accurate field record of the apparent levels of dissolved oxygen in the sediments of western Long Island Sound, as reflected in apparent redox depths (as recorded in sediment-profile images) and the amount of ammonia, hydrogen sulfide and oxygen present in the bottom waters, over a period critical to the LIS lobster fishery. Levels of dissolved oxygen in actual (within 5 cm) bottom waters of WLIS remain low throughout the summer and fall, despite the increase in dissolved oxygen recorded in the fall in waters 1 m above the bottom. Dissolved sulfide and ammonia concentrations in bottom waters of WLIS increased during spring and fall. These chemicals have been shown by other researchers to be physiologic stressors on lobsters and other marine life that live within benthic environments (Knezovich et al. 1996, Lianso 1991, McMahon & Wilkens 1975, Shumway & Scott 1983, Theede et al. 1969, Vargo & Sastry 1978, Vismann 1990, Wang & Widdows 1991). Given that the lobster die-off of 1999 began in late summer to early fall, it is hypothesized that ammonia and sulfide released from the sediments during that time may have played a role in weakening the lobsters. A comparison of water quality and temperature data from this study with that collected by the CT DEP at certain stations in WLIS during 1999 revealed some similarities and further strengthens the argument that environmental stressors played a role in the lobster mortality event of 1999.

KEY WORDS: lobster, mortality, stressors, sulfide, ammonia, hypoxia, imaging, Homarus americanus

INTRODUCTION

From historical times, Long Island Sound fisheries have played an important role in the economies of New York and Connecticut. The Long Island Sound lobster fishery alone was worth over $30 million in 1998. The lobster die-off that occurred during the late summer to fall of 1999 effectively shut down this important fishery, especially in western Long Island Sound (WLIS).

Lobster die-offs of varying degrees have occurred previously in Long Island Sound, usually during the fall (CT DEP 2000a) and have been attributed various pathogens and/or stressors. Although all dead lobsters necropsied from the 1999 die-off contained evidence of the paramoeba, Neoparamoeba pemaquidensis, the relative role this organism played in the actual demise of the lobster remains uncertain as of this writing (Robohm et al. 2004). Most experts agree that adverse environmental conditions, including the presence of bottom water hypoxia (Valente & Cuomo 2001, Cuomo et al. 2004, Valente & Cuomo in press), elevated water temperatures (LoBue et al. 2003), and unusual storm events played a significant role in weakening the LIS lobsters in 1999, making them more susceptible to infestation by biologic pathogens, including the parasitic paramoeba. This study investigates seasonal changes in bottom water sediment and chemical conditions in WLIS and the potential role they may have played in the Long Island Sound lobster mortality event of 1999.

BACKGROUND

Water column hypoxia, as defined by the CT DEP, begins when the oxygen content of waters in WLIS drops to [less than or equal to] 3.5 mg/L. The first reported occurrence of hypoxia in WLIS waters was made in 1909 (NYCDEP 1998). By the late 20th Century, however, summertime hypoxia had become a recurring phenomenon in WLIS (CT DEP 2000b). One of the main driving forces behind water column hypoxia in WLIS appears to be nitrogen enrichment from both point and nonpoint sources of pollution (CT DEP 2000a, 2000b, Welsh & Eller 1991, Welsh et al. 1994). This enrichment results in excess phytoplankton productivity that, in turn, creates an oxygen draw down as the phytoplankton dies and sinks through the water column. Additional driving forces on summertime hypoxic conditions in WLIS include seasonal stratification of the water column and warm water temperatures. Interestingly, hypoxic water conditions during the summer of 1999 were less severe and affected a smaller area than they had in previous and subsequent years (CT DEP 2000a) despite the presence of a strongly stratified water column that resulted from a lack of rainfall. Bottom and surface water temperature measurements taken during the summer of 1999 revealed water temperatures up to 3[degrees]C higher than those present in previous years. Yet, the lobster mortality event did not occur during the summer months when these conditions were present, but rather began in September and continued unabated through the fall of 1999. Were there other environmental factors present in the fall that may have contributed to the lobster mortality?

Sediment profile images (SPI) taken immediately after the lobster die-off showed the presence of an extremely shallow (<1 cm deep) Redox Potential Discontinuity (RPD) (1) in the sediments underlying most lobster fishing grounds in the western Sound and the general absence of benthos. Such an RPD is also often indicative of hypoxic-anoxic conditions present at the sediment-water interface. Yet, the CT DEP data (CT DEP 2000a) clearly demonstrated that the water column was not hypoxic during this period. This condition raises the following questions--what factors might cause a fall bottom water hypoxic/anoxic event in WLIS separate from the summertime water column hypoxic events and might conditions associated with such an event be enough to cause lobster mortality?

Work conducted by the authors (Cuomo et al. 2004, Valente & Cuomo in press) during the late summer and fall of 2000 revealed the presence of bottom water (within 5 cm of the sediment-water-interface) hypoxic conditions present within WLIS at a time when the overlying waters were fully oxygenated. Their work also revealed the presence, in the bottom waters, of minor amounts of ammonia and sulfide--both chemicals associated with the anaerobic decomposition of organic matter. Their study showed the development of bottom water hypoxic conditions began in late August to early September, peaked in October, and began to dissipate by November. The timing of the study was such, however, that it missed the critical summer months when the overlying water column was experiencing hypoxia.

The work presented here further investigated bottom water and sediment chemistry dynamics in WLIS, and the potential influence of near-bottom chemistry on benthic community structure and lobster health. It is hypothesized that exposures to hypoxia/anoxia and sub-lethal levels of reduced metabolites associated with anaerobic organic matter decomposition may have induced a physiologically-stressed state in the lobsters living in WLIS during the tall of 1999, weakening their immune system and increasing their susceptibility for disease.

METHODS

This study consisted of a series of field surveys designed to obtain sediment profile photographs (SPI) and bottom water data from 12 sampling stations in WLIS between the months of April and December 2002. The objective of this field survey was to examine overall benthic habitat quality in WLIS, as revealed by SPI, bottom water chemical conditions, and benthic organisms. The data obtained from this study were compared with that collected from CTDEP water quality monitoring surveys and other sources for the same period to look for points of agreement and divergence. Additionally, the data were compared with the conditions known to be in existence during the summer and fall of 1999 to further understand the role that hypoxia and reduced metabolites associated with anaerobic organic matter decomposition, may have played in the massive lobster mortality of 1999.

The 12 "critical sampling sites" were culled from 36 total stations previously sampled by the authors (Valente & Cuomo 2001, Cuomo et al. 2004, Valente & Cuomo in press) after the lobster die-off in 1999, and again in 2000. The stations generally followed a north to south transect across the Eastern and Western Narrows of Long Island Sound (Fig. 1). The sites represented a mixture of bottom water conditions (e.g., severely hypoxic, ammonia and hydrogen sulfide present and a shallow RPD to oxygenated with a deep RPD). All 12 stations were occupied during the three SPI surveys undertaken and during the seven water-sampling cruises taken. Because of mechanical difficulties, a set of water samples for May was unable to be retrieved. Benthic samples were taken concurrent with the deployment of the sediment-profile camera. The ship's GPS was used in positioning each station and the ship's engines were idled during each sampling event.

[FIGURE 1 OMITTED]

Once the ship reached station, the bottom water sampler was deployed. Discrete samples of seawater (6 total) were obtained from 1 m (hereafter referred to as near-bottom water or NBW) and 5 cm (hereafter referred to as bottom water or BW) above the seafloor, using the [Cuomo-Dobkowski.sup.patent pending] bottom water sampler, a weighted sampler designed specifically for this purpose. The sampler's weight is distributed in such a way as to ensure that the sampler maintains contact with the seafloor during deployment. In addition, the line attached to the sampler remains slack throughout the sampler's deployment. At each station, the sampler was deployed to retrieve the near-bottom water sample, first, followed by deployment and retrieval of the bottom water sample. During collection of the bottom water, the sampler was allowed to sit on the bottom for approximately 1 rain prior to sample collection to assure that any sulfides or ammonia accidentally released from the bottom upon contact of the sampler with bottom sediments would be cleared away by bottom currents. During SPI cruises, all water samples were collected prior to the deployment of the sediment-profile camera at every station to ensure clean, uncontaminated water samples.

Upon retrieval, all water samples were immediately fixed and/ or frozen for later analysis. Samples were analyzed for dissolved oxygen (Parsons et al. 1984), ammonia (Parsons et al. 1984), and hydrogen sulfide (Cline 1969, Fonselius et al. 2002). A minimum of three replicates were analyzed per sample and blanks were run after every sample to insure consistency. Standardization of each method was carried out immediately before the analyses using blanks and fresh reagents each time.

One benthic grab sample was taken at five selected stations in May and December and at six stations in September to groundtruth the benthic communities identified in the sediment-profile images. The sites chosen represented areas that previous SPI studies had identified as ranging from anoxic-azoic (Stage 0, sensu Rhoads & Germano 1982) to deeply bioturbated, oxygen-rich sediments (Stage III, sensu Rhoads & Germano 1982). Each sample was sieved on board the research vessel; all material retained on sieves 250-m mesh was preserved in 10% formalin and stained with rose bengal. Samples were returned to the laboratory where they were picked, identified, enumerated and recorded.

A Benthos[R] Model 3731 SPI camera was used to obtain three sediment-profile images at each of the 12 stations sampled during each survey. The replicates were spaced anywhere from 3-10 m apart from each other, the estimated distance between successive drops of the camera, whereas the ship retained its on-station position. For a complete description of the design and operation of the sediment-profile camera and the physical and biologic parameters measured and analyzed from the collected images, please see Rhoads and Germano (1982, 1986). Measurements obtained from all 3 replicates at each station were averaged and mapped, using the ArcView Geographic Information System. The aRPD (apparent RPD) refers to the SPI's ability to accurately record the redox potential discontinuity transition as a measure of photoreflectance. Because no direct measurement of pore water chemistry is made with the sediment-profile camera, the vertical transition based on photoreflectance is referred to as the apparent RPD.

RESULTS

Water Chemistry

At all sites, average NBW dissolved oxygen increased from August to September, decreased in October, and then rose in December (Fig. 2A). In general, average BW dissolved oxygen levels were elevated in May, decreased to their lowest levels in August, increased in September, declined again in October, and rose again in December, paralleling the near-bottom water conditions (Fig. 2B).

[FIGURE 2 OMITTED]

Average NBW dissolved sulfides (Fig. 3A) were low throughout the study period, although they increased in September and October and decreased in December. Average BW dissolved sulfides followed a similar trend (Fig. 3B). Average NBW dissolved ammonia (Fig. 4A) increased steadily from August to October at most stations, then declined in December; average BW ammonia levels (Fig. 4B) were elevated in May, varied in August and September, increased significantly in October at the westernmost stations and declined in December at all stations.

[FIGURES 3-4 OMITTED]

SPI

SPI collected from stations in WLIS during 2002 were consistent with those from previous surveys conducted in 1999 and 2000. Surface sediments at almost all of the 2002 stations were fine-grained, consisting of either silt-clay or silt, with a small proportion of fine sand. Very soft mud and unconsolidated silty sediments were the primary types of benthic habitats observed within the deeper stations of the surveyed area.

The overall average aRPD (2) for the May survey was 2.1 cm; the aRPD depths were highly variable among stations in the May survey, with roughly half the stations having thin aRPD depths of less than 2.0 cm and the other half having values above this value. Most of the shallowest aRPD depths (<2.0 cm) were observed in the Western Narrows. Extremely black sediment (indicating sediment anoxia) was present at 12 of the 17 stations sampled (Fig. 5A to B). A cluster of stations in the Western Narrows had anoxic sediment present at depth and at the sediment surface. In the September survey, the mean aRPD depths ranged from 0-5.2 cm, with an overall average of 2.1 cm. The overall range of values in this survey was wider than it was in May or December. Of particular note, however, were the extremely shallow or nonexistent aRPDs observed at Stations 27, 29 and 31 in the Western Narrows. A methane gas bubble (evidence of anaerobic bacterial decomposition of organic matter) was visible within the sediment column at Station 29 and white sulfur bacteria were present at the sediment surface. Images from Station 31 and 32 also showed the presence of white sulfur bacteria at the sediment surface. Overall, extremely black, anoxic sediment was observed at almost all stations sampled in the September survey, either present at depth or exposed at the sediment surface.

[FIGURE 5 OMITTED]

The December images revealed aRPDs that were uniformly deeper than the values observed in the May or September surveys, with an overall average of 2.7 cm. The range of values for December aRPDs was narrower than the range for either May or September, with a minimum of 1.7 and a maximum of 4.0 cm. Despite the presence of an aRPD at all of the stations, the underlying sediments retained their extremely black appearance.

Using the criteria set forth by Rhoads & Germano (1982, 1986) for assigning benthic community status to SPI, the following succession stages were assigned. Stage I (conditions associated with small, surficial tube-dwelling polychaetes) was assigned to 32% of the replicate images collected in May, 77% of the replicate images from September and 36% of the replicate images from December. Evidence of Stage III (conditions associated with head-down, deposit-feeding infauna) was detected in 45% of the May images, 14% of the September images, and 61% of the December images. When present, Stage III conditions were generally accompanied by either Stage I or Stage II conditions at the sediment-water interface. It should be noted that Stage I conditions were widespread in all three of the 2002 surveys, although some of the samples stationed in the Western Narrows seemed to be completely azoic (Stage 0).

Benthos

The results of the benthic sampling are shown in Figures 6 A to C. There is little variation in species composition across all stations and across seasons. In addition, the density of organisms per square meter was low at all stations. Most species present were small polychaete worms. From a functional perspective, the organisms that dominated the benthos in WLIS during this study are part of the typical late Stage I to very early Stage II community of Rhoads & Germano (1982, 1986).

[FIGURE 6 OMITTED]

DISCUSSION

The CT DEP (http://dep.state.ct.us/wtr/lis/monitoring/summer 2002.htm) reports that hypoxia in the waters of WLIS was particularly severe during the summer of 2002. Hypoxid conditions persisted for 64 days, beginning June 26th and ending August 28th. The maximum area covered by hypoxic water conditions (<3.0 mg/L DO) reached 130 square miles. The monitoring data from mid-July showed an area of hypoxia larger than the historical average for July, and the early August survey found an area <1 mg/L (42 square miles) that was the largest ever observed by the CT DEP.

This study indicates that BW and NBW hypoxia was moderately severe (<2.0 mg/L) at all stations in August 2002. Such data are consistent with the data collected by the CT DEP for the same time period. NBW dissolved oxygen measured in this study at all stations in September 2002 was also consistent with CT DEP data and fall within the moderate to excellent scale used by CT DEP; measured BW oxygen levels, however, fall within the moderately severe to moderate hypoxia scale of the CT DEP at all stations except for the two easternmost stations that had readings above 4.0 mg/L. This pattern was similar to that observed during the 2000 study (Valente & Cuomo in press). Near-bottom and bottom waters returned to hypoxic conditions in October despite the breakdown of stratification in the western Sound. NBW measurements were <2.0 mg/L at all stations, except for Station 14, which had measured dissolved oxygen concentrations slightly >2.0 mg/L. Similar conditions were recorded by the authors in the fall of 2000 (Cuomo et al. 2004). All NBW and BW measurements for December were near or >3.0 mg/L, placing all of the stations within the marginal-excellent water quality framework of the CT DEP monitoring program.

NBW and BW dissolved oxygen levels in WLIS seem to follow a seasonal pattern, with levels rising during the winter and decreasing in the summer and/or fall. Increasing water temperatures, coupled with the aerobic water column during the summer months, are known to contribute to the onset of hypoxic water column conditions in WLIS during July and August. These conditions, when severe, may, as in the present study, reach the very bottom waters of WLIS. The breakdown of stratification in the late summer to early fall (September for the 2002 study), however, results in a return of oxygen to bottom and near-bottom waters in WLIS. The data from this study demonstrate a return to hypoxic bottom water conditions in the early to mid fall (October for the 2002 study). Such conditions are hypothesized to be caused by the anaerobic decomposition of the fall plankton bloom on and within the sediments of WLIS and diffusive processes occurring at the sediment-water-interface. It is further hypothesized that because water and sediment temperatures continue to drop during the late fall and early winter, bacterial decomposition in the sediments likely slows and the bottom waters return to an oxygenated state. Thus, the overall pattern of dissolved oxygen in the bottom waters in WLIS seems to be bimodal--high DO in the winter and early spring, low DO in the summer, a brief return to higher levels in the late summer to early fall as stratification breaks down and a return to low levels in the fall as benthic processes dominate.

Measurements of dissolved sulfide taken at stations with organic-rich sediments in the westernmost Sound in May ranged from 1-4.5 mmol/1; those taken at the more easterly or sandier stations (10, 14, 19, 24, 32, A5 and B3) were all <1 mmol/l. Although hypoxia was well developed in August, dissolved sulfide values for near-bottom and bottom waters at all stations remained <1 mmol/1, strongly supporting the idea that hypoxic conditions during the summer months in WLIS are not related to anaerobic decomposition processes occurring in the sediments. As the water column destratified in the fall, sulfide values increased at all stations, although not as dramatically as they had in the 2000 study. The data from these two sampling years suggest that sulfide inputs to bottom waters cycle over the course of a year in WLIS. Elevated levels of bottom water sulfide in the spring are hypothesized to come from the bacterial decomposition of both freshly deposited plankton and stored sedimentary organic matter. These levels decrease during the summer as the water column becomes stratified and plankton production decreases. The return to elevated sulfide levels in the fall is hypothesized to result from the arrival of dead plankton from the fall bloom, the warm temperature of the sediments and their associated increased bacterial production rates. It is further hypothesized that as sediment temperatures drop over the remainder of the fall and into the winter, bacterial decomposition rates slow down, no new plankton is added to the sediment, dissolved oxygen levels rise as water temperatures decrease, and oxygen chemically diffuses from the water into the uppermost cm of sediments, effectively cutting off sulfide release.

Yearly ammonia trends in WLIS bottom waters are similar to those of sulfides. Once again, the levels measured in 2000 were higher than those measured in 2002, but the overall trend is the same for both years. Ammonia levels begin to increase in the spring, after the plankton bloom. Ammonia levels decrease during the summer and increase again in the fall. As sediment and water temperatures cool down in the late fall, ammonia production in the sediments falls off and ammonia release into the bottom waters declines.

Overall, the results of the three SPI surveys conducted in western LIS during 2002 are consistent with those of previous surveys conducted in 1999 and 2000 and indicate that the habitat quality of the seafloor in western LIS remains somewhat compromised, mainly because of the widespread distribution of anoxic sediments at or near the sediment surface. Surface sediments that appeared to be anoxic were visible in the profile images in all three of the 2002 surveys. This anoxic sediment was either exposed as the sediment-water interface or present just below a thin surface veneer of oxidized sediment. The chronic, widespread presence of relatively thin aRPD depths and highly anoxic sediment in western LIS throughout the months of May, September and December of 2002 is attributed to a well-documented combination of factors, including nutrient over-enrichment (eutrophication), resultant high inputs of labile organic matter to bottom waters and sediments and seasonal thermal stratification of the water column leading to near-bottom hypoxia (Parker & O'Reilly 1991, Welsh & Eller 1991, CTDEP 2000b). Similar to previous years, the stations in the Western Narrows generally exhibited more severe degradation in benthic habitat quality than those in the Eastern Narrows. There was also a more noticeable temporal change in benthic habitat quality across the three surveys at the Western Narrows stations compared with the Eastern Narrows stations.

Surface-dwelling, opportunistic benthic organisms (Stage I) were widespread in all three of the 2002 surveys, although some of the Western Narrows stations sampled in September appeared to have azoic sediments (i.e., no visible macrofauna). Stage I taxa have high population turnover rates and are often able to survive in hypoxic and/or sulfidic sediments. In particular, the tube-building, surface-dwelling polychaetes that typically comprise Stage I in Long Island Sound are among the most sulfide-resistant marine invertebrates (Wang & Chapman 1999, Knezovich et al. 1996, Cuomo 1984, Cuomo 1985).

Conditions representative of more-sensitive Stage III taxa also seemed to be present, albeit at low apparent densities, in all three surveys. Yet, data from the benthic surveys do not support this interpretation. This result illustrates one of the challenges of SPI interpretation: namely, that it sometimes can be difficult to distinguish an active feeding void from an inactive one (i.e., a relic structure); this is because the small subsurface excavations created by head-down deposit-feeding infauna can persist for some time (weeks to months) after the organisms themselves have either died or moved away. Therefore, it is possible that some of the subsurface features identified as active feeding voids in the SPI might be artifacts of the camera's penetration or inactive/relic structures. This is particularly likely during the September survey, after the hypoxia event of July and August had been severe and prolonged enough to kill many larger-bodied, Stage III infauna (Diaz & Rosenberg 1995). Therefore, some of the SPI interpretations of "Stage III," based on the observation of apparent feeding voids in the images, likely overestimated the actual presence of these organisms. The annual cycle of benthic recruitment and die-off in response to hypoxia in western LIS is a subject worthy of additional, more detailed study.

To see what insights these data may provide into the lobster mortality event of 1999, it is necessary to compare conditions measured in 1999 with the data collected in this and other studies. It is known that hypoxic conditions were present in western Long Island Sound during the summer of 1999 and it is also known that September temperatures exceeded 20.8[degrees]C--a temperature considered critical to lobster survival--at almost all stations sampled. Temperature, which exerts strong influences on water column dissolved oxygen levels, bacterial degradation of organic matter in and on sediments, water column mixing and lobster health, undoubtedly played a role in the 1999 mortality and in the 2002 mortality. The data from this study reveal a decoupling of processes controlling bottom water chemical conditions in WLIS over the course of a year. It seems that in the summer, when water temperatures are at their warmest and stratification is present, hypoxia is a process driven from the "top-down." That is, hypoxia in WLIS in the summer is primarily driven by density stratification, aerobic decomposition of plankton in the water column and lack of bottom water oxygen renewal owing to this stratification. Such information is not new and has formed the underlying basis for the ongoing successful water quality-monitoring program established by EPA and carried out by CT DEP. This gradual depletion of oxygen from the bottom waters of WLIS, however, explains the lack of lobsters and other mobile bottom-dwelling organisms reported by lobstermen during the summer months preceding the die-off in WLIS in 1999. Organisms capable of migrating out of areas of low oxygen, such as flounder and lobsters, undoubtedly do. It is likely that the lobsters and other mobile organisms migrated to more oxygen-rich areas of LIS.

The more interesting question regarding the lobster die-off of 1999 concerns the timing of the event--it occurred not during the peak of the hypoxia--but in the fall of 1999, after the water column had become remixed. Data from this and the 2000 study strongly suggest that the fall is the time when sediment processes determine bottom water chemistry, as opposed to surface processes, although the two remain connected. In other words, in the fall, bottom water hypoxia and anoxia result from sediment-driven processes, a "bottom-up" effect. This effect is not captured by routine water column monitoring measurements as it is generally confined to within 0.25 m or less of the sediment-water interface. More important, however, is the fact that lobsters and other benthic organisms live within the zone where this effect is occurring. The release of reduced end products of anaerobic organic matter decomposition into an otherwise oxygenated water column, such as appears to occur in WLIS in the fall, affects both chemical and biologic processes. First, the reduced end products undergo oxidation reactions in the presence of oxygenated bottom waters, resulting in a reduction in the local dissolved oxygen levels. Second, organisms present in the area are going to encounter both sulfides and ammonia, along with oxygen, in the bottom waters. Numerous studies have revealed the negative effects of hydrogen sulfide, ammonia and/or hypoxia on benthic organisms (Theede et al. 1969, Voyer & Hennekey 1972, Vargo & Sastry 1978, Shumway & Scott 1983, Stickle et al. 1989, Vismann 1990, Lianso 1991, Wang & Widdows 1991). Studies on lobsters indicate that adult lobsters vary in their tolerance to low oxygen conditions (McLeese 1956, McLeese & Watson 1968, Young 1973, McMahon & Wilkens 1975, Cooper & Uzmann 1980, Penkoff & Thurberg 1982); recent work by Robohm et al. (2005) has demonstrated that adult lobsters exposed to a combination of low oxygen, high temperatures, and concentrations of sulfide and ammonia equal to those measured in this study, undergo high mortalities.

The presence of ammonia and sulfide in near-bottom and bottom waters suggests that anaerobic decomposition processes are dominant over aerobic processes in near-surface sediments of WLIS. This is supported by the chemical data, the SPI and the benthic data collected during this study and the study conducted by Valente & Cuomo (in press). The SPI from the present study's three surveys revealed anoxic sediments either exposed at the sediment-water interface or covered by a thin veneer of oxidized sediment. The benthic data reveal the absence of an abundance of deep deposit-feeders and the dominant presence of small, surface-dwelling organisms--the organisms that make up the Stage I communities of LIS. Many Stage I organisms (e.g., capitellid polychaetes) have been shown to be tolerant of reduced conditions, such that they might be able to survive in areas where reduced metabolites are defusing out of the sediments even as overlying waters are oxygenated (Vismann et al. 1998, Cuomo 1984, 1985). This, combined with the chemical data, strongly suggests that the organic-rich bottom sediments of WLIS are potentially compromised and remain in a functionally hypoxic-anoxic state for most of the year.

CONCLUSIONS

The CT DEP hypoxia-mapping project (http://dep.state.ct.us/ wtr/lis/monitoring/summer 2002.htm) reported that the summer of 2002 was one of the worst recorded hypoxic events to ever occur in LIS. DEP hypoxia maps reveal significant regions of LIS with water column dissolved oxygen levels <1.0 mg/L. This is a functionally hypoxic-borderline anoxic state. Water temperature trends for the same time reveal higher than normal conditions at the Battery in WLIS and the Race in ELIS commencing in July and lasting until mid September (Fig. 7). The results of the study reported on here were consistent with these observations. Nearbottom and bottom dissolved oxygen levels remained low throughout the duration of this study and paralleled the DEP values. Nearbottom levels of DO recorded in this study were consistent with those measured by the DEP at approximately the same depth in the water column. Bottom water DO measurements were consistently lower than near-bottom waters and represent the influence of sediment oxygen demand, as well as lack of bottom water oxygen renewal, in the system. Highest levels for bottom water and nearbottom water DO were recorded in May and December, when colder water was present and WLIS was well mixed.

[FIGURE 7 OMITTED]

The measured levels of sulfide and ammonia do not exhibit a simple correlation to low oxygen: however, they do exhibit somewhat of a reverse correspondence. Ammonia and sulfide releases from the sediment occurred in the late spring and fall in WLIS. Both of these events were accompanied by local bottom water decreases in dissolved oxygen levels. The low levels of bottom DO measured in October apparently result from the consumption of oxygen by the reoxidation of anaerobic organic matter degradation products actively diffusing into the water column from the surface sediments because the upper water column was fully oxygenated during this period. Whereas the trends are the same as 2000, sulfide and ammonia levels were detected in concentrations well over 1.0 retool/1 in the fall of that year, compared with the lower levels detected this year. Nevertheless, because there is no sudden return to stratified conditions recorded for this period, it becomes necessary to invoke some other explanation on this fall hypoxic occurrence. The data strongly support the idea that fall hypoxia is driven by processes occurring within the sediments and at the sediment-water interface.

Given the information presented here, it is likely that exposure to sulfides and ammonia during the fall of 1999, in conjunction with an extended period of summertime hypoxia, and warmer than average bottom water temperatures was enough to exert a physiologic stress on the lobsters in WLIS, leaving them extremely susceptible to the parasite, Neoparameoba pemaquidensis, which was present in the waters of WLIS during the summer and fall of 1999.

ACKNOWLEDGMENTS

The authors thank the CTDEP, the National Marine Fisheries Service in Milford, CT and the Marine Sciences Research Center at Stony Brook University, for the use of their vessels, they especially thank Dr. Anthony Calabrese and Ms. Nancy Balcom for all of their assistance with this manuscript. This work was funded by Connecticut Sea Grant, New York Sea Grant, the Connecticut Department of Environmental Protection (CTDEP), and the United States Environmental Protection Agency.

LITERATURE CITED

Benway, J., R. Burgess, C. Giannini, P. Howell, J. Hayden & K. McKown. 2004. Long-term population trends in American lobster in Long Island Sound. Abstract. Fourth LIS Lobster Health Symposium, Stony Brook University, October 4.

Cline, J. D. 1969. Spectrophotometric determination of hydrogen sulfide in natural waters. Limnol Ocean 14(3):454-458.

CT DEP. 2000a. Information regarding the impact of 1999 lobster mortalities in Long Island Sound. Bureau of Natural Resources, Marine Fisheries Office, February 4, 2000. 45 pp.

CT DEP. 2000b. Long Island Sound ambient water quality monitoring program: summer hypoxia monitoring survey 1991 to 1998 Data review. Hartford, CT: Bureau of Water Management. 46 pp.

Cooper, R. A. & J. R. Uzmann. 1980. Ecology of juvenile and adult Homarus americanus. In: J. S. Cobb & B. F. Phillips, editors. The biology and management of lobsters. Vol. II. Ecology and management, New York: Academic Press. pp. 97-142.

Cuomo, M. C. 1984. The ecological and paleoecological significance of sulfides in marine sediments. Yale University Ph.D. thesis. Department of Geology & Genphysics. 196 pp.

Cuomo, M. C. 1985. Sulfide as a larval settlement cue for Capitella sp I. Biogeochem. 1:169-181.

Cuomo, C., R. Valente & D. Dogru. 2004. Monitoring of bottom water and sediment conditions at critical stations in western Lung Island Sound. In: P. Van Patten, editor. Proceedings of the 6th biennial Long Island Sound research conference, October 24-26, 2002. Groton, CT: Connecticut Sea Grant. pp. 13-21.

Diaz, R. J. & R. Rosenberg. 1995. Marine benthic hypoxia: a review of its ecological effects and the behavioural responses of benthic macrofauna. Oceanogr. Man Biol. Ann. Rev. 33:245-303.

Fonselius, S., D. Dyrssen & B. Yhlen. 2002. Determination of hydrogen sulfide. In: K. Grasshoff, K. Kremling & M. Ehrhardt, editors. Methods of seawater analysis, 3rd ed. Germany: Wiley-VCH. pp. 91

Knezovich, J. P., D. J. Steichen, J. A. Jelinski & S. L. Anderson. 1996. Sulfide tolerance of four marine species used to evaluate sediment and porewater toxicity. Bull. Environ. Contain. Toxicol. 57:450-457.

Lianso, R. J. 1991. Tolerance of low dissolved oxygen and hydrogen sulfide by the polychaete Streblospio benedicti (Webster). J. Exp. Mar. Biol. Ecol. 153:165-178.

LoBue, C., P. LoCicero & J. Hayden. 2003. The American lobster fishery in New York waters: a summary of fishery dependent sampling information. 1998 to 2002. NYS DEC, Div. of Fish Wildlife and Marine Resources. 43 pp.

McLeese, D. W. 1956. Effects of temperature, salinity, and oxygen on the survival of the American lobster. J. Fish. Res. Board Can. 13(2):247272.

McLeese, D. W. & J. Watson. 1968. Oxygen consumption of the spider crab (Chionoecetes opilio) and the American lobster (Homarus americanus) at a low temperature. J. Fish. Res. Board. Can. 25(8):17291732.

McMahon, B. R. & J. L. Wilkens. 1975. Respiratory and circulatory responses to hypoxia in the lobster, Homarus americanus. J. Exp. Mar. Biol. 62:637-655.

NYCDEP. 1998. New York harbur water quality survey, executive summary. New York City DEP Bureau of Wastewater Pollution Control. 16 pp.

Parker, C. A. & J. E. O'Reilly. 1991. Oxygen depletion in Long Island Sound: a historical perspective. Estuaries 14:248-264.

Parsons, T. R., Y. Maita & C. M. Lalli. 1984. A manual of chemical and biological methods fur seawater analysis. Pergamon Press. 173 pp. pp. 14-17.

Penkoff, S. J. & F. P. Thurberg. 1982. Changes in oxygen consumption of the American lobster, Homarus americanus, during the molt cycle. Comp. Biochem. Physiol. 72A(4):621-622.

Rhoads, D. C. & J. D. Germano. 1982. Characterization of organism-sediment relations using sediment profile imaging: An efficient method of Remote Ecological Monitoring of the Seafloor (REMOTS[TM] System). Mar. Ecol. Prog. Ser. 8:115-128.

Rhoads, D. C. & J. D. Germano. 1986. Interpreting long-term change in benthic community structure: a new protocol. Hydrobiol. 142:291-308.

Robohm, R. A., A. F. J. Draxler, D. Wieczorek, D. Kapareiko & S. Pitchford. 2005. Effects of environmental stressors on disease susceptibility in American lobsters: a controlled laboratory study. J. Shellfish Res. 24(3):773-779.

Robohm, R., S. Pitchford, D. Kapareiko & G. Messick. 2004. Infectivity of parasitic amoebae in lobsters: Testing of Koch's Postulates. In: The briefing book of the Long Island Sound lobster research initiative working meeting. May 2-3, 2004, Groton, CT: UCONN Avery Point.

Shumway, S. E. & T. M. Scott. 1983. The effects of anoxia and hydrogen sulfide on survival, activity, and metabolic rate in the coot clam, Mulinia lateralis (Say). J. Exp. Mar. Biol. Ecol. 71:135-146.

Stickle, W. B., M. A. Kapper, L. Liu, E. Gnaiger & S. Y. Wang. 1989. Metabolic adaptations of several species of crustaceans and molluscs to hypoxia: tolerance and microcalorimetric studies. Bio. Bull. 177:303-312.

Theede, H., A. Ponat, K. Hiroki & C. Schleiper. 1969. Studies on the resistance of marine bottom invertebrates to oxygen-deficiency and hydrogen sulfide. Mat. Biol. 2:325-337.

Valente, R. & C. Cuomo. A sediment-profile imaging (SPI) study of sediment anoxia and bottom-water chemistry in relation to the 1999 lobster mass mortality event in western Long Island Sound, USA. Estuaries. (In press).

Valente. R. & C. Cuomo. 2001. REMOTS[TM] imaging and bottom water chemistry of Western Long Island Sound a preliminary exploration of the contribution of hypoxia to the lobster mortality event of 1999. EPA Special Report, Office of Long Island Sound Programs.

Vargo, S. L. & A. N. Sastry. 1978. Interspecific differences in tolerance of Eurytemora affinis and Acartia tonsa from an estuarine anoxic basin to low dissolved oxygen and hydrogen sulfide. In: D. S. McClusky & A. J. Berry, editors. Physiology and behaviour of marine organisms. Proceedings of the 12th European Marine Biology Symposium. Stirling, Scotland: Pergamon Press.

Vismann, B. 1990. Sulfide detoxification and tolerance in Nereis (Hediste) diversicolor and Nereis (Neanthes) virens (Annelida:Polychaeta). Mar. Ecol. Prog. Ser. 59:229-238.

Vismann, B., M. K. Grieshaber & O. Giere. 1998. Ecophysiological differentiation of Capitella capitata (Polychaeta) sibling species from different sulfidic habitats. Mar. Ecol. Prog. Ser. 175:155-166.

Voyer, R. A. & R. J. Hennekey. 1972. Effects of dissolved oxygen on two life stages of the mummichog. Prog. Fish. Culture 34:222-225.

Wang, F. & P. M. Chapman. 1999. Biological implications of sulfide in sediment--a review focusing on sediment toxicity. Environ. Toxicol. Chem. 18(11):2526-2532.

Wang, W. X. & J. Widdows. 1991. Physiological responses of mussel larvae Mytilus edulis to environmental hypoxia and anoxia. Mar. Ecol. Prog. Ser. 70:223-236.

Welsh, B. L. & F. C. Eller. 1991. Mechanisms controlling summertime oxygen depletion in western Long Island Sound. Estuaries 14:265-278.

Welsh, B. L., R. I. Welsh & M. L. DiGiacomo-Cohen. 1994. Quantifying hypoxia and anoxia in Long Island Sound. In: K. R. Dyer & R. J. Orth, editors. Changes in fluxes in estuaries: implications for science and management. Olsen & Olsen. pp. 131-137.

Young, J. S. 1973. A marine kill in New Jersey coastal waters. Mar. Pollut. Bull. 4(5):70.

(1) Redox Potential Discontinuity refers to the location in a given sediment column where free oxygen disappears and [E.sub.h] goes to zero.

(2) The depth of the aRPD* measured in REMOTS images provides an indication of the degree of oxygen penetration into the sediment surface resulting from both passive diffusion and active mixing by benthic organisms.

CARMELA CUOMO, * RAYMOND VALENTE AND DEREN DOGRU

Department of Geology & Geophysics, Yale University, New Haven, Connecticut 06520 and SAIC, Marine Environmental Science & Information Management Division, Newport, Rhode Island 02840

* Corresponding author. Present address: Department of Biology & Environmental Sciences, University of New Haven, 300 Orange Avenue, West Haven, Connecticut 06516; E-mail: [e-mail]