Seagrass presence structures densities and distributions of two estuarine gobies: laboratory and field evidence

Pamela J. Schofield, University of Southern Mississippi
& U. S. Geological Survey, Gainesville, FL

Presented at the Estuarine Research Foundation biennial meeting in Seattle, WA September 2003.

click to enlarge

Densities of the code goby Gobiosoma robustum Ginsburg and clown goby Microgobius gulosus (Girard) vary across regions of Florida Bay.  Although their distributions overlap to a limited degree, G. robustum is found in great abundance in the seagrass beds of the western and southern portion of the bay, where a strong Gulf of Mexico influence maintains salinities at a relatively constant level (35 – 40 ppt). Conversely, M. gulosus is most common in the north-eastern section of the bay that is influenced by freshwater inflow and generally is characterised by lower salinities and sparse submerged vegetation.  In this study, the effects of salinity, habitat complexity, competition, and susceptibility to predation were examined as possible factors underlying the distribution of these two species.

Effects of salinity and competition on growth:

The relative influences of salinity and competition on growth of both goby species was evaluated in a laboratory experiment.  Both species were grown in the lab over 27 d at two salinity levels (5 and 35 ppt), two food levels (low and high) and both with and without the presence of the other species. Both species exhibited greatest growth at the high food level and the low (5 ppt) salinity.  Neither species was affected by the presence of the other species, and there were no overall differences in growth between the two species.  Thus, the competitive superiority of G. robustum over M. gulosus does not seem to confer an advantage relative to feeding success. Furthermore, as growth of G. robustum was greater at the lower salinity, it is clear that some factor other than salinity is restricting the bay-wide distribution of this species from north-eastern Florida Bay.

Mean residuals (+ 1 SE) from analysis of variance test showing growth of M. gulosus
 and G. robustum at salinities (5 and 35 ppt). Residuals were used rather than
 raw means to account for differences attributable to initial mass.
 

Acute salinity tolerance:

The premise of this study was to determine whether M. gulosus and G. robustum exhibited differential tolerance to acute salinity shifts, which may be related to their distributions within Florida Bay. Given that M. gulosus inhabits the north-eastern region of Florida Bay that is more likely to be exposed to rapid shifts in salinity than the less variable habitat over which G. robustum is distributed, it was hypothesized that M. gulosus would be more tolerant to salinity shifts. To test this hypothesis, acute (e.g., "plunge-type") salinity tolerance tests were performed with both species. Both species were remarkably tolerant to rapid shifts in salinity, and there was no compelling evidence for differences in acute salinity tolerance.

Survival (%) for two gobies transferred from 30 ppt to 5, 10, 15 and 30 ppt (control).

Gulf toadfish, Opsanus beta,  a goby predator

Habitat effects on susceptibility to predation:

To determine whether the presence of a structurally complex habitat mitigates predation for either species, a laboratory experiment was conducted in which gobies (1 per trial) were placed in aquaria containing either a) bare mud or b) bare mud + artificial seagrass substrate. The predator (Opsanus beta) was more successful preying upon M. gulosus than G. robustum, and there was no habitat (e.g., seagrass) effect. This was surprising, given that M. gulosus is a burrowing species that was expected to fare better than the non-burrowing G. robustum.
 

Gobiosoma robustum

Density: 0.1 - 9,     > 10 Click graphics and images to enlarge

Microgobius gulosus

Competition and habitat selection:

In Florida Bay, G. robustum is generally found in structurally complex habitats (e.g., seagrass beds), while M. gulosus is more often found in bare mud areas. To determine whether this habitat partitioning was an effect of interspecific competition, I conducted a series of laboratory experiments, in which each species was presented with structured (artificial seagrass) versus non-structured (bare sand) habitats and their frequency of choosing either habitat type was measured. Use of structured versus non-structured habitats were then examined when the two species were placed together in a mixed group. Both goby species selected grass over sand in allopatry; however, in sympatry, M. gulosus occupied sand more often when paired with G. robustum than when alone.  Logistic regression was used to analyse the data, with habitat type (grass or sand) designated as the dichotomous dependent variable and species (M. gulosus or G. robustum) and social grouping (allopatric and sympatric) as the main effects. The interaction effect (species X group) was significant (X²=11.1, df=1, P=0.0009), reflecting the shift in habitat selection of M. gulosus when G. robustum is present.  Gobiosoma robustum appears to directly influence the habitat choice of M. gulosus:  It seems that M. gulosus is pushed out of the structured habitat that is the preferred habitat of G. robustum. Thus, competition appears to modify habitat selection of these species when they occur in sympatry.
 

click to enlarge

Field microhabitat distributions:

I documented patterns of microhabitat use in south-eastern Florida Bay, where both G. robustum and M. gulosus are present.  The substrate at Crab Key, the site of this study component, consisted of dense beds of turtlegrass and drift algae interspersed with small bare patches (see photo, below). Ten pairs of plots measuring 1 m² each were quantitatively sampled with a throw trap in bare unvegetated (i.e., propeller scar) and adjacent seagrass microhabitats. At Crab Key, G. robustum was more abundant in seagrass than unvegetated microhabitats (t = 4.832, df = 9, P = 0.001, two-tailed test). Conversely, M. gulosus was more abundant in unvegetated (bare mud) microhabitats than seagrass (t = 2.449, df = 9, P = 0.037, two tailed test). These results corroborate the conclusion of the laboratory experiment on habitat selection:  It seems that M. gulosus is pushed out of the structured habitat that is the preferred habitat of G. robustum.

Conclusions & implications for water management:

The microhabitat distribution of G. robustum and M . gulosus in the Gulf-influenced (e.g., southern and western) regions of Florida Bay can be explained by the selection of G. robustum for structurally-complex seagrass habitats and its competitive superiority over M. gulosus for seagrass-dominated microhabitats. The low salinities of north-eastern Florida Bay do not limit expansion of G. robustum into this area; however, G. robustum may be rare in north-eastern Florida Bay due to the lack of seagrass in this region. Future manipulative field studies are needed to more fully test this hypothesis. Alternately, it is possible that prey species (especially those that co-occur with seagrass) utilised by G. robustum are not as abundant in the less-vegetated north-eastern Florida Bay as opposed to southern and western regions of Florida Bay.

Salinity variation in NE Florida Bay has increased dramatically since the placement of the C-111 canal, which shunts water away NE Florida Bay toward the Atlantic Ocean.  There is some evidence that this increase in variability has resulted in a lower standing crop of seagrass.  Results from this study indicate that the abundance of G. robustum is not affected by salinity variability directly, but may be indirectly affected by its affinity for seagrass.  Both seagrass and seagrass-associated fishes (such as G. robustum) may increase in abundance if Everglades restoration restores historic freshwater flow to NE Florida Bay.
 

Acknowledgments

This poster summarizes a multi-year research program that would not have been possible without the support of numerous individuals, including: USM: Stephen T. Ross, Mark Peterson; NOAA: Gordon Thayer, Allyn Powell; USGS: George D. Dennis, Nicholas Funicelli, Mike Robblee, Ken Sulak, Susan Walls, Stephen J. Walsh, George Yeargin and numerous others.  Buck Albert took the lab photographs. Allyn Powell (NOAA) provided the maps of Florida Bay. Clinton Hittle (USGS) and Kevin Kotun (NPS) provided salinity data.