Assessing in situ tolerances of eastern oysters (Crassostrea virginica) under moderate hypoxic regimes: implications for restoration.

ABSTRACT Increases in the frequency and duration of hypoxia and the loss of biogenic reefs are two of the most prominent environmental insults to estuaries. We investigated the interaction between moderate hypoxia and habitat restoration activities on estuarine ecosystems by measuring population growth and somatic growth for newly settled Eastern oysters (Crassostrea virginica). Experiments were conducted at a site that historically experiences moderate hypoxia (2.0 mg [l.sup.-1] < [[O.sub.2]] < 4.0 mg [l.sup.-1])(Whitehouse Reef) and at a site that experiences normoxia (Dauphin Island). Panels with known starting densities of oyster spat were deployed at the surface, 1.25 m above the bottom, and 0.5 m above the bottom at both sites. Population growth at the bottom and the 1.25 m panels at Whitehouse Reef was -1% to 0% individuals [day.sup.-1] caused by periods of moderate hypoxia; however, as oxygen conditions improved in September, population growth increased to approximately l to 3% individuals [day.sup.-1]. For these two panels, total population growth averaged 1% over the experiment duration. For the surface panel at Whitehouse Reef and all three depths at Dauphin Island, population growth remained positive with population growth of between 2% to 17% individuals [day.sup.-1]. Somatic growth indicated significant location and depth specific differences with somatic growth being negatively correlated with depth. Marginal water quality caused by moderate hypoxia may limit oyster population growth to a much greater extent than predicted by previous laboratory experiments. Additionally, we demonstrate that low-cost experiments prior to the initiation of restoration activities can help ensure success by providing critical in situ information on design and location.

KEY WORDS: Hypoxia, Crassostrea virginica, Eastern oyster, population dynamics, oxygen

INTRODUCTION

Success of habitat restoration in the marine environment has proven to be one of the most difficult tasks confronting habitat managers, despite the large investment of both time and money (Mann & Powell 2007). A reason for this lack of success may be that location and design of restoration projects are chosen without accounting for the complex interactions that are possible when working in situ. Historical data and surveys of the biological and physical conditions are often the only tools used in the decision making process (see Brumbaugh et al. 2006 and may fail to provide adequate data to ensure success. Integration of ecologically focused empirical testing into restoration design may help increase the odds of success by determining if an animal or habitat "can survive" rather than "should survive" (Hare et al. 2006, Kennedy et al. 2008). The advantage of integrating experimental ecology with restoration would be the successful identification of survival bottlenecks and the design criteria to mitigate such problems.

Among the myriad of environmental issues that may create such survival bottlenecks in estuarine ecosystems are increasing frequency and severity of low oxygen conditions and wide-scale loss of biogenic habitats (Diaz & Rosenberg 1995, Jackson et al. 2001, Orth et al. 2006). Increases in severity and frequency of low oxygen events in coastal marine systems have made it important to understand the role that hypoxia has in structuring coastal ecosystems (Diaz & Rosenberg 1995, Lenihan & Peterson 1998). For bivalves, a low oxygen event can be operationally classified into three discrete levels, moderate hypoxia (4 mg. [L.sup.-1] [greater than or equal to] [O.sub.2] [greater than or equal to] 2 mg [L.sup.-1]), severe hypoxia (2 mg. [L.sup.-1] [greater than or equal to] [O.sub.2] [greater than or equal to] 0.5 mg [L.sup.-1]), and anoxia ([O.sub.2] < 0.5 mg [L.sup.-1]) (Renaud 1986, Diaz & Rosenberg 1995, Turner et al. 2005). Because responses to severe hypoxia are often rapid and more easily detected, the focus of much benthic research has been on organism behavioral responses to severe hypoxia rather than the slower or cumulative responses expected during periods of moderate hypoxia (e.g., Diaz & Rosenberg 1995). Although moderate hypoxia can influence metabolism, growth, and movement patterns of organisms (Rosas et al. 1999, Mistri 2004, Montagna & Ritter 2006), the effects of moderate hypoxia are more subtle and rarely studied in situ. For example, severe hypoxia is known to be detrimental to bivalve populations (Saoud et al. 2000b, de Zwaan & Babarro 2001), but it is unknown how duration, intensity, and frequency of more moderate hypoxic events interact with respect to bivalve populations.

For benthic organisms, experimental research has concentrated on identifying and describing an organism's physical or physiological responses to hypoxia, then relating this back to that organism's ecology (e.g., de Zwaan 1977, Mistri 2004). Experiments are often conducted under highly controlled laboratory conditions (e.g., Seitz et al. 2003) and conducted at temperatures that often do not reflect conditions when hypoxic events are common, especially for locations in subtropical climates (NSTC 2003). Because of the lack of confounding factors and variability in oxygen concentrations for these laboratory experiments, such trials rarely replicate natural conditions (Diaz & Rosenberg 1995). The result is that few of the conclusions reached under laboratory conditions are robust enough to be applied to field conditions; however, the results from laboratory experiments are often accepted as prima facie for the observed patterns in the field.

Information obtained from the field can also result in data that is just as limiting as that obtained in the laboratory. Because the costs associated with restoring estuarine habitats are often high, field studies rather than habitat restoring pilot programs are often used to guide restoration. These studies are often limited to intermittent information about oxygen levels (e.g., Powers et al. 2005) and descriptions of organism settlement, recruitment, behavioral changes, and colonization during or after hypoxic events (Sagasti et al. 2000, Lenihan et al. 2001, Bell et al. 2003a, Bell et al. 2003b). The consequences are that projects are often designed without in situ data is that high-quality and biologically relevant for a specific location. To mitigate this, historical locations are often chosen with the assumption that simply returning the habitat to an earlier state will be enough to allow for the habitat to become stabile and thrive (Brumbaugh et al. 2006). One problem is that the conditions today (e.g., water quality, food supply, surrounding habitats, etc ...) are most likely different than they were before the habitat was damaged. As a result, only returning the benthic habitat to a previous site or condition may not be adequate to ensure success. Simple and inexpensive techniques are needed to assess the biological tolerances of benthic organisms to ensure that habitats are constructed in the most effective manner possible.

Interaction between the severity and duration of hypoxic conditions may affect mortality and growth of Eastern oysters (Crassostrea virginica); however, our ability to predict the fate of oysters is poor, especially as it relates to in situ conditions. In the laboratory, oysters are known to have the ability to successfully cope with hypoxia by utilizing anaerobic respiration (de Zwaan 1977, Widdows et al. 1989). However, hypoxia is considered to be one of the primary reasons for declining oyster abundances in field studies (Rothschild et al. 1994, Lenihan & Peterson 1998). Few studies have looked for a clear relationship between severity and duration of low oxygen conditions for shellfish because continuous oxygen data can be expensive to obtain. Thus, it is unknown for most survivorship studies if a single untimely (e.g., during settlement) hypoxic event impacted populations or if a more extended episode or cyclical events influenced the results (i.e., pycnocline tilting) (e.g., May 1973). Although oysters are sensitive to hypoxia at many life stages, there is a positive relationship between the ability to cope with hypoxia/anoxia and both development stage and body size (Widdows et al. 1989, Baker & Mann 1992). For example, in the laboratory there is significantly less settlement during hypoxic conditions (Baker & Mann 1992), but if juvenile oysters are exposed to anoxia, survival is upwards of 50% after 150 h at 22[degrees]C (Widdows et al. 1989). However, depending on temperature, juvenile oysters have an [LC.sub.50] in excess of 10 days at oxygen concentrations between 2.75 and 1.5 mg [l.sup.1] (Stickle et al. 1989, Widdows et al. 1989, Baker & Mann 1992). For larger, subadult oysters, they have the ability to manage oxygen uptake even at low salinity and oxygen levels or use anaerobic respiration (de Zwaan 1977, Shumway, 1982).

Our objective was to compare the instantaneous population growth and shell/somatic growth of oyster populations experiencing moderately hypoxic and nonnoxic conditions in Mobile Bay, AL. From these comparisons, we will determine if periodic exposure to oxygen concentrations within the range of tolerances demonstrated in the laboratory for oysters is transferable to field studies. Ultimately, this experimental design tests an efficient method to assay the fate of restored oyster reefs prior to a large monetary investment. We expect that the response to moderate hypoxia by newly settled oysters may result in reduction in both survival and growth caused by repeated exposure and an inadequate metabolic recovery period; however, we anticipate that this effect may be mitigated by increases in vertical relief.

METHODS

Between July and October 2007, we deployed three replicate moorings at two locations in Mobile Bay, AL. Moorings were placed at Whitehouse Reef in the south-central part of Mobile Bay (30.41[degrees]N, 88.07[degrees]W) and the eastern tip of Dauphin Island near the Dauphin Island Sea Laboratory (30.25[degrees]N, 88.07[degrees]W). Locations were identified a priori as sites where oysters were historically abundant (May 1971), having similar oyster larvae abundances (Hoese et al. 1972, Saoud et al. 2000a), and having the potential to experience different levels of hypoxia (May 1973, K. Park, unpublished data). We expected to observe moderate hypoxia at Whitehouse Reef and normoxia at Dauphin Island. Moorings were placed 20 m apart in depths of approximately 3.5 m during mean high tide and consisted of a concrete filled tire with a 30 cm post in the center. Extending from this post was a polypropylene line that reached the water's surface and was attached to a float to keep the line taught. A single 40 x 40 cm Vexar pouch (mesh size 1 [cm.sup.2]) was attached at 0.5 m above the bottom, 1.25 m above the bottom, and 0.25 m below the surface to hang down approximately 0.4 m. These bags were stitched together on three sides with a removable stitch on the fourth side allowing access to the bag's interior. Inside the pouch, we placed an additional piece of 30 x 30 cm mesh that had either 24 or 25 oyster shell halves attached. This pouch design prevented predation on oyster spat from common predators like blue crabs (Callinectes sapidus) and oyster drills (Stramonita haemastoma) (Smith 1983, Eggleston 1990). On each shell within the pouch, there were between 1 and 14 recently settled spat (mean abundance [+ or -] SE - 3.1 [+ or -] 0.17 [shell.sup.-1]; mean shell height [+ or -] SE 10 [+ or -] 2.10 mm). Juvenile oysters were settled during June 2007 at the Auburn Shellfish Laboratory, Dauphin Island, AL using clean articulated oyster shell. Each articulated shell had a small hole drilled through it where a cable tie would later be inserted to attach the shell to the mesh sheet inside the pouch. The articulated shells were arranged in a 5 x 5 or a 4 x 6 pattern based on the shape of the shells. Treatments had a mean of 79 [+ or -] 19 spat. After the oysters were placed into the pouch, it was sewn shut and attached to the mooring.

To monitor water quality at Whitehouse Reefa YSI 6600 V2 sonde was deployed at the surface, 1.6 m off the bottom, and 0.3 m off the bottom. The data for temperature, salinity, dissolved oxygen, and depth were obtained using the burst setting in twenty minute intervals (mean values recorded after 5 min, followed by 15 min of inactivity). At Dauphin Island, physical measurements were taken from the permanent monitoring station maintained by the Dauphin Island Sea Laboratory. This station consists of a YSI 6600 V2 sonde located approximately 0.5 m above the bottom and approximately 100 m from the oyster moorings. Data collection interval was set at 30 min. Because this station had been in operation for multiple years and revealed little evidence of hypoxia near the bottom, we did not feel it was necessary to deploy additional oxygen sensors at intermediate depths.

Oysters were deployed on July 11th and sampled approximately every two weeks until September 9 and again on October 6th. At each collection, the moorings were hauled aboard a vessel, oysters were then removed from the bag, brushed clean of any biofouling, and counted. Shells where oyster spat mortality was 100% were replaced with a new shell containing spat from the same cohort. These new oysters were held in bags near the surface of the water at the Dauphin Island Sea Laboratory's docking facilities to ensure that they experience similar conditions (minus oxygen concentrations) as those deployed in the experiment. The oysters were then returned to the bags, sewn shut, and the moorings redeployed. During the final sampling, oysters were returned to the laboratory and enumerated. If present, up to 50 spat on shells that remained throughout the duration of the experiment without being replaced were measured (length and height) for each bag. Throughout the experiment, oyster drills were present on or around the moorings; however, there was no evidence of predation on oysters within the bags.

Statistically, we tested population growth by measuring instantaneous population growth (individuals day-l). This metric was used because new oyster settlement was expected during the experiment and there was no efficient technique to tag the approximately 2,500 oysters deployed to differentiate them from settlers. Percentages were calculated between each sample date. Population growth was tested using a 2-way repeated measures analysis of variance (rm-ANOVA) where location and depth were our independent variables and time was our repeated measure 'Davis 2002). Because rm-ANOVA requires values at each time (Davis 2002), we used average values from all the other sampling times at the corresponding location and depth for any bags that were damaged or lost (10% total) to maintain three replicates. For measurements of growth, we used a two-way ANOVA where location and depth were our independent variables. When significant interaction terms were present, we used a one way ANOVA on Ranks to make comparisons between locations and among depths. For depth, Dunn's posthoc tests were performed when significant differences existed. All values were considered significant at the P [less than or equal to] 0.05 level.

[FIGURE 1 OMITTED]

RESULTS

Physical Parameters

Dissolved oxygen levels at both sites resulted in values that met our a priori assumptions about oxygen gradients based on location and depth (Fig. 1). At Whitehouse Reef, bottom waters exhibited a pattern that consisted of 6% of the dissolved oxygen measurements being severe and 28% of the measurements being moderately hypoxic. Dissolved oxygen concentrations at the bottom had a mean of 4.65 [+ or -] 0.02 mg [L.sup.-1] and ranged between 0.57 and 7.94 mg [L.sup.-1] Mean bottom temperature was 29.6 [+ or -] 0.03[degrees]C and ranged between 25.7 and 32.6[degrees]C. Mean salinity on the bottom was 21.9 [+ or -] 0.03 psu with a range of 15.3-27.6 psu. Prior to September 1st 44% of the measurements taken at Whitehouse Reef had oxygen concentrations of 4.0 mg [L.sup.-1] or less, but after this date only 15% of the values were hypoxic. This date was chosen because of the increase in population growth exhibited between 8/27 and 9/06 (Fig. 2). Measurements taken at 1.6 m above the bottom resulted in 7% of the oxygen concentrations being moderately hypoxic and less that 1% being severely hypoxic/anoxic. Dissolved oxygen concentrations had a mean of 6.0 [+ or -] 0.02 mg [L.sup.-1] and ranged between 1.38[degrees]C and 9.00 mg [L.sup.-1]. Temperature had a mean of 29.3[degrees]C [+ or -] 0.03[degrees]C and ranged between 25.7[degrees]C and 32.8[degrees]C. Mean salinity was 20.5 [+ or -] 0.03 psu with a range of 15.3-27.3 psu. In the surface waters, approximately 1% of dissolved oxygen concentrations measured less than 4 mg [L.sup.-1]. Dissolved oxygen concentrations had a mean of 6.68 [+ or -] 0.01 mg [L.sup.-1] and ranged between 3.16 and 8.91 mg [L.sup.-1] Temperature had a mean of 29.5 [+ or -] 0.03[degrees]C and ranged between 25.4 and 33.3[degrees]C. Mean salinity was 19.9 [+ or -] 0.03 psu with a range of 15.2-23.6 psu.

At Dauphin Island, conditions near the bottom experienced minimal hypoxia. Ninety-six percent of the values collected were considered to be normoxic and the remaining 4% were moderately hypoxic. Dissolved oxygen concentrations had a mean of 5.75 [+ or -] 0.01 mg [L.sup.-1] and ranged between 2.52 and 10.98 mg [L.sup.-1]. Temperature had a mean of 29.8 [+ or -] 0.03 [degrees] C and ranged between 25.7 and 33.7[degrees]C. Mean salinity was 26.9 4 [+ or -] 0.03 psu with a range of 21.5-33.3 psu.

For Whitehouse Reef, population growth generally increased with time for the 0.5 m and 1.25 m depths, whereas growth near the surface averaged approximately 2.0% for the duration of the experiment (Fig. 2). During the first month of the experiment, there was a net loss of oysters on the panels at the 0.5 and 1.25 m depths followed by a steady increase for the 0.5 m depth. The 1.25 m panel increased until 9/9/2007, but decreased by October. Mean cumulative population growth was 0.007 [+ or -] 0.003 individuals [day.sup.-1] at 0.5 m, 0.008 [+ or -] 0.003 individuals [day.sup.-1] at 1.25 m, and 0.019[+ or -] 0.002 individuals day-1 at the surface (Fig. 3). For oysters located at Dauphin Island, population growth remained positive throughout the experiment except for one measurement at the surface. At the 0.5 and 1.25 m depths, a large population spike was evident during the initial period. Panels near the surface experienced a similar spike but it did not occur until the third sample period (Fig. 2). Overall, mean instantaneous population growth for all depths was approximately 5%. Mean cumulative population growth was 0.055 [+ or -] 0.009 individuals [day .sup.-1] at 0.5 m, 0.055 [+ or -] 0.008 individuals [day.sup.-1] at 1.25 m, and 0.042 [+ or -] 0.004 individuals [day.sup.-1] at the surface (Fig. 3). Rm-ANOVA results showed that there was a significant difference in instantaneous oyster population growth between the two locations ([F.sub.1, 32] = 319.4, P < 0.001 ), but not among the three depths or any of the interaction terms.

[FIGURE 2 OMITTED]

Somatic growth of oysters on our panels showed that oysters residing near the surface had the greatest increase in shell height and length and that growth decreased for both parameters with depth in both locations (Fig. 4). Two-way ANOVAs resulted in significant interactions between location and depth for both shell height ([F.sub.2, 683] = 12.79, P < 0.001) and length ([F.sub.2, 683] = 13.23, P < 0.001) indicating that depth specific growth was dependent on location. For shell height, ANOVA on ranks resulted in a significant differences based on location ([H.sub.1] = 29.06, P < 0.001) and depth ([H.sub.1] = 99.05, P < 0.001) (Fig. 4). Posthoc analysis showed that each of the three depths were significantly different from the other (surface v. 0.5 m - [Q.sub.2] = 9.73, P < 0.05; surface v. 1.25 m - [Q.sub.2] = 2.88, P < 0.05; 1.25 m v. 0.5 m - [Q.sub.2] = 6.50, P < 0.05). For shell length, ANOVA on ranks resulted in a significant differences based on location ([H.sub.1] = 17.91, P < 0.001) and depth ([H.sub.1] = 124.2, P < 0.001). Posthoc analysis showed that each of the three locations were significantly different from the other (surface v. 0.5 m - [Q.sub.2] = 10.95, P< 0.05; surface v. 1.25 m - [Q.sub.2] = 63.49, P< 0.05; 1.25 m v. 0.5m - [Q.sub.2] = 7.08, P < 0.05).

[FIGURE 3 OMITTED]

DISCUSSION

Our findings explain why restored oyster reefs located in areas that experience only moderate levels of hypoxia can fail and become demographic sinks. Our results suggest that the more amount of time that an oyster larva experiences moderate hypoxia, the less likely that it will successfully recruit to a suitable habitat. Overall our conclusion is not unique, Baker and Mann (1992) demonstrated that hypoxia could inhibit recruitment; however, we found that the magnitude of the response as described by population growth in situ was greater than expected by previous laboratory studies like Baker and Mann (1992). Previous laboratory experiment designs did not incorporate variations in duration and severity that would be expected in natural systems. We were able to demonstrate this concept in situ using an inexpensive methodology that could be used to guide restoration efforts.

Population growth on the surface panel at Whitehouse Reef showed that there was an adequate supply of larvae to this area; however, population growth at the deeper panels did not reflect this until later in the summer, even when oxygen levels were well within the tolerances of exposure duration and severity reported in laboratory studies for juvenile oysters (Widdows et al. 1989, Baker & Mann 1992, Baker & Mann 1994). Because the vertical swimming ability of presettlement oyster larvae decreases at the pediveliger stage (Galtsoff 1964, Baker 1991), there should be a greater abundance of settlement ready larvae near the bottom panels compared with the surface panels in a highly stratified system (e.g., Mobile Bay) (Dekshenieks et al. 1996, Noble et al. 1996). According to Dekshenieks et al. (1996), their model suggests that the abundance of ready to settle larvae near the surface would be 5 times less than abundances near the bottom. What is unknown is what happens when these oyster larvae settle during periods of normoxia between periods of hypoxia. Although many more larvae may settle, few may survive the next hypoxic period because they may not have the ability to recover from the energetic demands of settlement before the onset of another period of moderate hypoxia, resulting in very high mortality under seemingly benign conditions.

[FIGURE 4 OMITTED]

If population growth is correlated with environmental data, then as the pattern in oxygen concentrations became increasingly normoxic starting in late August or early September, oyster populations responded as evidenced by positive growth on the two deepest panels. The longer periods of normoxia allow for oysters to settle and recover before experiencing another period of moderate hypoxia. Thus, an older oyster may be more capable of utilizing an alternative source of energy production (anaerobic) during this period (de Zwaan 1977). The similarity in patterns in population growth between the 1.25 and 0.5 m panels at Whitehouse Reef and the disparity with the patterns in oxygen concentrations was unexpected. Although the oxygen levels indicated at 1.6 m that little hypoxia occurs, growth at 1 m reflected a population growth pattern similar to the bottom panels. This may be the result of moderate hypoxia generally extending into the water column to a height that that the oysters experienced (~1 m), but less than the depth of the oxygen meter (1.6 m above the bottom). The pycnocline in Mobile Bay is often present at approximately 1.0-1.5 m above the bottom and rapid declines in oxygen concentrations are correlated with this gradient (Park et al. 2007). As such, it would be expected that the response between the two panel depths would be similar. Pycnocline tilting may be responsible for the extended period of moderate hypoxia during mid August at the 1.25 m panel and has been historically responsible for declines in oyster spat settlement (May 1973).

Somatic growth showed a similar pattern between the two locations with a negative relationship between growth and depth, but because of its location near the mouth of the bay, flow may be greater at the Dauphin Island site as exhibited by the increased growth in the surface and 1.25 m panels. Because oysters are active suspension-feeders, the vertical pattern in somatic growth can be explained by the reduction in water flow with an increase in depth (Lenihan et al. 1996). The lack of differences in oyster growth at the deepest depths between the two locations may indicate that flow may have been similar along the bottom regardless of location or that a benthic boundary layer exists at both locations that can limit the amount of food available for growth (Frechette et al. 1989, McKee et al. 2004).

By controlling the larval supply because of location selection and eliminating the effects of predation caused by caging, we were able to minimize two of the most important population controlling variables between Whitehouse Reef and Dauphin Island. Additionally, because growth at the deepest moorings was similar between the locations, we can assume that the quality or quantity of food was similar for oysters residing along the bottom. This leaves few explanations for the decreased population growth except that the larvae that settle at Whitehouse reef must travel a farther distance prior to settlement than those at Dauphin Island. At Whitehouse Reef, the closest source of larvae is 5.2 km south of the reef. At Dauphin Island, the closest larval source is 2.8 km to the north. For oyster larvae, transportation upstream is most likely a combination of selective tidal stream transport (Wood & Hargis 1971) and passive transport (Mann 1986). Larval oysters moving north would have to experience many more periods of moderate hypoxia as they move along with the bottom waters or remain near the bottom to escape prevailing water movement patterns, whereas those moving south could rapidly follow the net seaward flow. Although larval oyster abundance may be high, when a larval oyster encounters a suitable habitat at Whitehouse Reef they may not be healthy enough to successfully settle. If an oyster does survive settlement, it may not be able to recover quick enough to survive a postsettlement moderate hypoxic event. Survival would rely on normoxic conditions remaining stable long enough for them to be capable of anaerobic respiration or rebuild energetic reserves (de Zwaan 1977). If this process is influencing the settlement patterns in each of these locations, then this may explain the failure of some of the restoration efforts in Mobile Bay (M. Johnson, unpublished data) and may help further quantify why reefs in certain locations may be acting as larval sinks rather than sources (North et al. 2008).

For restoration activities at Whitehouse Reef, these data combined with previous studies suggest that any reef building activity should result in reefs that are at least 1.5 m in height with a base of at least 25 x 25 m. In Mobile Bay, continued monitoring of 25 x 25 m experimental reefs deployed in 2004 (see Gregalis et al. 2008 for description) noted that the reef shape changed over a period of four years from trapezoidal with a large flat crest to mound shaped with a small crest that remains at the original height of l m (M. Johnson, unpublished data). The additional vertical relief combined with the larger base should help ensure that the reef will retain its height long enough to ensure newly settled larvae will experience minimal fluctuations in oxygen concentrations. It will also allow the core section of the reef remains above the critical vertical threshold required for oyster survival as the reef matrix naturally settles prior to any cementing by newly settled oysters. At the Dauphin Island site, a reef built to a height of 0.1 m would be sufficient based on oxygen data. Unfortunately, based on a cost of $70.00 per [m.sup.3] the cost of reef restoration at Whitehouse Reef will cost 15 times that of a restored reef at Dauphin Island. Conversely, for the same cost, a reef 15 times larger could be built at Dauphin Island for the same amount. If the potential for predation is factored into the equation, the probability for success would be in favor of Whitehouse Reef over Dauphin Island because the middle part of the bay experiences the variations in salinity because of location and vertical height that are required to reduce the impact of oyster drills (Noble et al. 1996). Regardless, the low cost of this experimental design was able to provide the critical information that is often missing during the design process, ultimately leading to a more effective reef building activities in Mobile Bay.

Currently, there are sufficient data that quantifies the tolerances of larval and juvenile oysters under brief or continuous hypoxic conditions in the laboratory (Widdows et al. 1989; Baker and Mann 1992, 1994, Diaz & Rosenberg 1995); however, further research is needed to examine the cumulative effects of repeated exposure to even moderate hypoxia. The cumulative effects of repeated exposure to hypoxia has been shown to have population level effects by resulting in differential genetic responses between grass shrimp (Palaemonetes pugio) exposed to chronic hypoxia (1.5-1.8 mg [l.sup.-1]) versus cyclical hypoxia (Brown-Peterson et al. 2008). The consequence is a decrease in the number of broods and eggs produced, resulting in a slower estimated rate of population growth. This experiment has demonstrated that experimental ecology can have a low-cost role in directing reef restoration efforts to help ensure success. In this instance, our results help explain why previous restoration efforts in the area have failed (M. Johnson, unpublished data). Subsequent reef building activities should incorporate adequate reef relief into the building process, even in a location like the Whitehouse reef area that does not experience hypoxia at levels typically associated with the failure of shellfish populations (May 1973).

ACKNOWLEDGMENTS

The authors acknowledge C. Hightower, M. Kenworthy, S. Bosarge, J. Herrmann, and several unnamed interns for their tireless work drilling oyster shells, sewing Vexar mesh, and pulling moorings. Additionally, the authors thank S. Rikard and the staff at the Auburn Shellfish Lab for providing spat and allowing us to use the wet lab facilities prior to the start of this experiment. Funding for this experiment was provided by the National Marine Fisheries Service through the University of South Alabama's Oyster Reef Restoration Program. This is publication number 397 of the Dauphin Island Sea Lab.

LITERATURE CITED

Baker, P. 1991. Effect of neutral red stain on settlement ability of oyster pediveligers, Crassostrea virginica. J. Shellfish Res. 10:455-456.

Baker, S. M. & R. Mann. 1992. Effects of hypoxia and anoxia on larval settlement, juvenile growth, and juvenile survival of the oyster Crassostrea virginica. Biol. Bull. 182:263-269.

Baker, S. M. & R. Mann. 1994. Description of metamorphic phases in the oyster Crassostrea virginica and effects of hypoxia on metamorphosis. Mar. Ecol. Prog. Ser. 104:91-99.

Bell, G. W., D. B. Eggleston & T. G. Wolcott. 2003a. Behavioral responses of free-ranging blue crabs to episodic hypoxia. 1. Movement. Mar. Ecol. Prog. Ser. 259:215-25.

Bell, G. W., D. B. Eggleston & T. G. Wolcott. 2003b. Behavioral responses of free-ranging blue crabs to episodic hypoxia. 2. Feeding. Mar. Ecol. Prog. Ser. 259:227-235.

Brown-Peterson, N. J., C. Manning, V. Patel, N. D. Denslow & M. Brouwer. 2008. Effects of cyclic hypoxia on gene expression and reproduction in a grass shrimp, Palaemonetes pugio. Biol. Bull. 214:6-16.

Brumbaugh, R. D., M. W. Beck, L. D. Coen, L. Craig & P. Hicks. 2006. A practitioners" guide to the design and monitoring of shellfish restoration projects: An ecosystem services approach. The Nature Conservancy, Arlington, Virginia, 33 pp.

Davis, C. S. 2002. Statistical methods for the analysis of repeated measurements. Heidelberg, Germany: Springer Verlag. 415 pp.

de Zwaan, A. 1977. Anaerobic energy metabolism in bivalve molluscs. Oceanogr. Mar. Biol. Annu. Rev. 5:103-187.

de Zwaan, A. & J. M. Babarro. 2001. Studies on the causes of mortality of the estuarine bivalve Macoma balthica under conditions of (near) anoxia. Mar. Biol. 138:1021-1028.

Dekshenieks, M. M., E. E. Hofmann, J. M. Klinck & E. N. Powell. 1996. Modeling the vertical distribution of oyster larvae in response to environmental conditions. Mar. Ecol. Prog. Ser. 136:97-110.

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

Eggleston, D. B. 1990. Foraging behavior of the blue crab, Callinectes sapidus, on juvenile oysters, Crassostrea virginica: Effects of prey density and size. Bull. Mar. Sci. 46:62-82.

Frechette, M., C. A. Butman & W. R. Geyer. 1989. The importance of boundary-layer flows in supplying phytoplankton to the benthic suspension feeder, Mytilus edulis L. Linmol. Oceanogr. 34:19-36.

Galtsoff, P. S. 1964. The American oyster Crassostrea virginica (Gmelin). U.S. Fish and Wildlife Service Bulletin 64:1-480.

Gregalis, K. C., S. P. Powers & K. L. Heck, Jr. 2008. Restoration of oyster reefs along a bio-physical gradient in Mobile Bay, Alabama. J. Shellfish Res. 27:1163-1170.

Hare, J. A., H. J. Walsh & M. J. Wuenschel. 2006. Sinking rates of late-stage fish larvae: Implications for larval ingress into estuarine nursery habitats. J. Evp. Mar. Biol. Ecol. 330:493-504.

Hoese, H. D., W. R. Nelson & H. Beckert. 1972. Seasonal and spatial setting of fouling organisms in Mobile Bay and Eastern Mississippi Sound, Alabama. Alabama Marine Resources Bulletin, Montgomery, Alabama. pp. 9-17.

Jackson, J. B. C., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W. Botsford, B. J. Bourque, R. H. Bradbury, R. Cooke, J. Erlandson, J. A. Estes, T. P. Hughes, S. Kidwell, C. B. Lange, H. S. Lenihan, J. M. Pandolphi, C. H. Peterson, R. S. Steneck, M. J. Tegner & R. R. Warner. 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science 293:629-638.

Kennedy, B. P., K. H. Nislow & C. L. Folt. 2008. Habitat-Mediated foraging limitations drive survival bottlenecks for juvenile salmon. Ecology 89:2529-2541.

Lenihan, H. S. & C. H. Peterson. 1998. How habitat degradation through fishery disturbance enhances impacts of hypoxia on oyster reefs. Ecol. Appl. 8:128-140.

Lenihan, H. S., C. H. Peterson & J. M. Allen. 1996. Does flow speed also have a direct effect on growth of active suspension-feeders: An experimental test on oysters. Linmol. Oceanogr. 41:1359-1366.

Lenihan, H. S., C. H. Peterson, J. E. Byers, J. H. Grabowski, G. W. Thayer & D. R. Colby. 2001. Cascading of habitat degradation: Oyster reefs invaded by refugee fishes escaping stress. Ecol. Appl. 11:764-782.

Mann, R. 1986. Sampling of bivalve larvae. In: Jamieson, G., Bourne, N. (Eds.), North Pacific workshop on stock assessment and management of invertebrates. Canadian Special Publication of Fisheries and Aquatic Sciences, pp. 107-116.

Mann, R. & E. N. Powell. 2007. Why oyster restoration goals in the Chesapeake Bay are not and probably cannot be achieved. J. Shellfish Res. 26:905-917.

May, E. B. 1971. A survey of the oyster and oyster shell resources of Alabama. Alabama Department of Conservation, M.R., Bulletin 4, Montgomery, AL. pp. 1-52.

May, E. B. 1973. Extensive oxygen depletion in Mobile Bay, Alabama. Limnol. Oceanogr. 18:353-366.

McKee, B. A., R. C. Aller, M. A. Allison, T. S. Bianchi & G. C. Kineke. 2004. Transport and transformation of dissolved and particulate materials on continental margins influenced by major rivers: benthic boundary layer and seabed processes. Cont. Shelf Res. 24:899-926.

Mistri, M. 2004. Effects of hypoxia on predator-prey interactions between juvenile Carcinus aestuarii and Musculista senhousia. Mar. Ecol. Prog. Ser. 275:211-217.

Montagna, P. A. & C. Ritter. 2006. Direct and indirect effects of hypoxia on benthos in Corpus Christi Bay, Texas, U.S.A. J. Exp. Mar. Biol. Ecol. 330:119-131.

Noble, M. A., W. W. Schroeder, W. J. Wiseman, Jr., H. F. Ryan & G. Gelfenbaum. 1996. Subtidal circulation patterns in a shallow, highly stratified estuary: Mobile Bay, Alabama. J. Geophys. Res. 101:25689-25703.

North, E. W., Z. Schlag, R. R. Hood, M. Li, L. Zhong, T. Gross & V. S. Kennedy. 2008. Vertical swimming behavior influences the dispersal of simulated oyster larvae in a coupled particle-tracking and hydrodynamic model of Chesapeake Bay. Mar. Ecol. Prog. Ser. 359:99-115.

NSTC. 2003. An assessment of coastal hypoxia and eutrophication in U.S. waters. National Science and Technology Council, Committee on Environmental and Natural Resources, Washington, DC. pp. 82.

Orth, R. J., T. J. B. Carruthers, W. C. Dennison, C. M. Duarte, J. W. Fourqurean, K. L. Heck, A. R. Hughes, G. A. Kendrick, W. J. Kenworthy, S. Olyarnik, F. T. Short, M. Waycott & S. L. Williams. 2006. A global crisis for seagrass ecosystems. Bioscience 56:987-996.

Park, K., C.-K. Kim & W. W. Schroeder. 2007. Temporal variability in summertime bottom hypoxia in shallow areas of Mobile Bay, Alabama. Estuaries Coasts 30:54-65.

Powers, S. P., C. H. Peterson, R. R. Christian, E. Sullivan, M. J. Powers, M. J. Bishop & C. P. Buzzelli. 2005. Effects of eutrophication on bottom habitat and prey resources of demersal fishes. Mar. Ecol. Prog. Ser. 302:233-243.

Renaud, M. L. 1986. Hypoxia in Louisiana coastal waters during 1983: Implications for fisheries. Fish. Bull. (Wash. DC) 84:19-26.

Rosas, C., E. Martinez, G. Gaxiola, R. Brito, A. Sanchez & L. A. Soto. 1999. The effect of dissolved oxygen and salinity on oxygen consumption, ammonia excretion and osmotic pressure of Penaeus setiferus (Linnaeus) juveniles. J. Exp. Mar. Biol. Ecol. 234:41-57.

Rothschild, B. J., J. S. Ault, P. Goulletquer & M. Heral. 1994. Decline of the Chesapeake Bay oyster population: A century of habitat destruction and overfishing. Mar. Ecol. Prog. Ser. 111:29-39.

Sagasti, A., L. C. Schaffner & J. E. Duffy. 2000. Epifaunal communities thrive in an estuary with hypoxic episodes. Estuaries 23:474-487.

Saoud, I. G., D. B. Rouse, R. K. Wallace, J. Howe & B. Page. 2000a. Oyster Crassostrea virginica spat settlement as it relates to the restoration of Fish River reef in Mobile Bay, Alabama. J. World Aqua. Soc. 31:640-650.

Saoud, I. G., D. B. Rouse, R. K. Wallace, J. E. Supan & S. Rikard. 2000b. An in situ study on the survival and growth of Crassostrea virginica juveniles in Bon Secour Bay, Alabama. J. Shellfish Res. 19:809-814.

Seitz, R. D., L. S. Marshall, Jr., A. H. Hines & K. L. Clark. 2003. Effects of hypoxia on predator-prey dynamics of the blue crab Callinectes sapidus and the Baltic clam Macoma balthica in Chesapeake Bay. Mar. Ecol. Prog. Ser. 257:179-188.

Shumway, S. 1982. Oxygen consumption in the American oyster Crassostrea virginica. Mar. Ecol. Prog. Ser. 9:59-68.

Smith, D. J. 1983. Chemoattraction of the southern oyster drill Thais haemastoma towards the oyster Crassostrea virginica, as a function of temperature and salinity. Louisiana State University, Baton Rouge, Louisiana.

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. Biol. Bull. 177:303-312.

Turner, R. E., N. N. Rabalais, E. M. Swenson, M. Kasprzak & T. Romaire. 2005. Summer hypoxia in the northern Gulf of Mexico and its prediction from 1978 to 1995. Mar. Environ. Res. 59:65-77.

Widdows, J., R. I. E. Newell & R. Mann. 1989. Effects of hypoxia and anoxia on survival, energy metabolism, and feeding of oyster larvae (Crassostrea virginica, Gmelin). Biol. Bull. 177:154-166.

Wood, L & W. J. Hargis. 1971. Transport of bivalve larvae in a tidal estuary. In: D. Crisp, editor. Proceedings of the fourth European Marine Biological Symposium. Cambridge, UK: University Press, pp. 29-44.

MATTHEW W. JOHNSON, (1) * SEAN P. POWERS, (2,1) JOSEPH SENNE (2,1) AND KYEONG PARK (2,1)

(1) Dauphin Island Sea Lab, Fisheries Ecology Lab, 101 Bienville Blvd., Dauphin Island, Alabama 36528;

(2) University of South Alabama, Department of Marine Sciences LCSB 025 Mobile, Alabama 36688

* Corresponding author. E-mail: [e-mail]