the effects of sea urchin grazing and drift algal blooms on a subtropical seagrass bed community

LJournal of Experimental Marine Biology and Ecology246 (2000) 53–67

www.elsevier.nl / locate / jembe

The effects of sea urchin grazing and drift algal blooms on asubtropical seagrass bed community

*´Silvia MaciaCenter for Marine and Environmental Analyses, University of Miami, Rosenstiel School of Marine and

Atmospheric Science, 4600 Rickenbacker Causeway, Miami, FL 33149, USA

Received 27 January 1999; received in revised form 18 November 1999; accepted 25 November 1999

Abstract

Subtropical seagrass beds can be subject to relatively high levels of direct herbivory and largeblooms of drift algae, both of which can have important effects on the floral and faunal componentsof the community. Caging experiments were used to investigate these factors in a Thalassia tes-tudinum bed in Biscayne Bay, Florida. Abundance of sea urchins, Lytechinus variegatus, and driftalgae was manipulated within the cages. Naturally occurring levels of urchin grazing do not appearto affect the T. testudinum population. With experimentally increased urchin densities in the winter,seagrass shoot density and aboveground biomass decreased significantly. Similar effects were notdetected in the summer, indicating that the impact of grazing on T. testudinum is lessened during thistime of year. Shoot density was more vulnerable to grazing than aboveground biomass. This may bea result of grazing-induced increases in seagrass productivity, in which the remaining shootsproduce more or longer leaves. In the winter, drift algal blooms form large mats that cover the sea-grass canopy. Under the normal grazing regime these algal blooms do not have significant negativeeffects on the seagrass. With increased grazing pressure, however, there is a synergistic effect of

22grazing and drift algae on seagrass shoot density. At intermediate urchin density (10 per m ), cageswithout algae did not undergo significant decreases in shoot density, while those with algae did. Atthe high density of urchins, the number of seagrass shoots in cages both with and without algae de-creased, but the effect was more pronounced for cages with algae. Invertebrate abundance at thefield site was low relative to other seagrass beds. There were no discernible effects, either positive ornegative, of urchin and algae manipulations on the sampled invertebrate community. 2000Elsevier Science B.V. All rights reserved.

Keywords: Drift algae; Grazing; Seagrass community; Sea urchin; Thalassia

*Tel.: 1 1-305-361-4833; fax: 1 1-305-361-4077.´E-mail address: [email protected] (S. Macia)

0022-0981/00/$ – see front matter 2000 Elsevier Science B.V. All rights reserved.PI I : S0022-0981( 99 )00172-0

´54 S. Macia / J. Exp. Mar. Biol. Ecol. 246 (2000) 53 –67

1. Introduction

Seagrass beds are complex ecosystems composed of many interacting species ofplants, invertebrates, and fishes. Tropical and subtropical seagrass beds in particular havean added level of complexity resulting from the direct grazing of seagrasses, which ismore prevalent than in temperate areas (Ogden, 1976, 1980; Zieman, 1982). Theseasonal occurrence of large blooms of drift algae also contributes to the complexity ofsubtropical seagrass meadows (Holmquist, 1994, 1997; Bell and Hall, 1997). These twofactors can have major effects on the structure and function of warm-water seagrassbeds, yet they have rarely been studied in concert.

Several studies have documented blooms of drift algae in subtropical seagrass beds(Gore et al., 1981; Virnstein and Carbonara, 1985; Holmquist, 1994, 1997; Bell and Hall,1997). These blooms can result in very large mats of algae that compete with theseagrass for light and form bare patches void of seagrass (Hull, 1987; Short et al., 1995;Holmquist, 1997). Dense accumulations of algae can also decrease the flux of oxygen tothe sediment, thereby lowering the amount of oxygen available to animals living in orbeneath the algae (Hull, 1987; Hansen and Kristensen, 1997). Conversely, drift algalmats greatly increase the structural complexity of the seagrass community and canprovide superior habitat for invertebrates (Gore et al., 1981; Holmquist, 1994, 1997).Thus, the effects of drift algae on the floral and faunal components of seagrass beds canbe very complex, and may depend on the physical and biological characteristics specificto the particular seagrass bed in question (Hull, 1987).

Sea urchins are the main invertebrate grazers of live seagrass material (Greenway,1976; Zieman, 1982; Valentine and Heck, 1991). The urchin Lytechinus variegatus(Lamarck) is commonly found in seagrass beds throughout the western Atlantic and Gulfof Mexico. Much work has been devoted to the ecology of this important grazer,

¨especially in turtlegrass (Thalassia testudinum Banks ex Konig) beds, where L.variegatus most commonly occurs at population densities less than 10 individuals per

22m (Moore et al., 1963; Engstrom, 1982; Keller, 1983; Oliver, 1987; Montague et al.,1988; McGlathery, 1995). Larger urchin populations, however, have been reported andcan have serious impacts on the seagrass community in which they reside. The mostdramatic example of this is the formation of grazing fronts — extremely dense

22aggregations (up to 600 individuals m ) of urchins capable of completely denuding´areas of seagrass as large as thousands of square meters (Camp et al., 1973; Macia and

Lirman, 1999). However, even populations of urchins far smaller than those of the22grazing fronts (20–40 urchins m ) can remove a large proportion of the seagrass

standing crop (Greenway, 1976, 1995; Valentine and Heck, 1991).The presence of either drift algae or urchin grazers can have important effects on the

seagrass community. The potentially synergistic effect of these factors, however, has notreceived much attention. This study uses caging experiments in the field to investi-gate the effects of both algae (when present) and urchin grazing on a subtropical sea-grass bed. Both the plant and animal components of the seagrass community are con-sidered.

´S. Macia / J. Exp. Mar. Biol. Ecol. 246 (2000) 53 –67 55

2. Methods and materials

2.1. Field site

The field site, West Point, is a dense subtidal Thalassia testudinum bed in northernBiscayne Bay, Florida, USA (Fig. 1). Temperature at the field site ranges annually from18 to 318C and salinity from 30 to 43‰, with a water depth of approximately 1 m. Thesea urchin Lytechinus variegatus is the most conspicuous invertebrate member of theseagrass community. Monthly surveys were conducted at the site from September 1995to September 1998. At each survey five parallel 50-m transects were laid out 5 m apart,

2and five 1-m quadrats were randomly selected along each transect. All urchins withinthese 25 quadrats were counted (except for the first two surveys, in which only 15 andten total quadrats (i.e. three and two transects), respectively, were counted).

2.2. Cage design

2Circular cages with an area of 2 m were used. Cages were constructed of 5 m lengthsof stiff polyethylene netting (Vexar) with a mesh size of 3.2 3 3.2 cm. The sides of thecages were 60 cm high. Each cage had four 90-cm long pieces of steel reinforcement bar(rebar) attached with cable ties at regular distances from each other. The rebar protruded

Fig. 1. Map of north Biscayne Bay, Florida showing the West Point field site.


approximately 25 cm from the bottom of the netting for burial in the sediment. Cagetops were made of thin (0.5 mm diameter) flexible polypropylene netting with a meshsize of 3.2 3 3.8 cm. The thinnest possible netting was used for the cage tops in order tominimize shading effects. This netting was sewn onto the top of the cage side withplastic weed trimmer line.

Several control treatments were employed to test for caging artifacts. Partial cageswere constructed from 3.8 m lengths of netting and four pieces of rebar, as above. Thesepartial cage controls did not form a completely enclosed circle, having instead anopening of approximately 1.2 m (25% of the circumference). Two types of partial cagecontrols were used: with a top as described above, and with no top. Top controls werealso deployed; these were constructed by attaching a top to four pieces of rebar alone(i.e. no cage sides). Unmanipulated controls consisted of plots marked only by rebar. Totest for cage effects on water movement, current speed immediately above the seagrasscanopy (approximately 30 cm from the bottom) was measured both inside and outside ofthe full cages. Red dye was released into the water column and the time to travel 50 cmwas measured. Dye within the cages was released just inside the cage and allowed totravel towards the opposite side. There was no significant difference between currentspeed inside and outside of full cages (ln-transformed data: t 5 1.75; df 5 10; p 5 0.11).

2.3. Experimental design

Caging experiments were performed twice, in winter and summer. Experimental22 22 22treatments were 0 urchins per m , 10 urchins m (20/cage), and 20 urchins m

(40/cage). During the winter and fall months, the field site experiences a bloom of driftalgae (primarily Laurencia spp. and Dictyota spp.). These algae disappear in the warmermonths (Irlandi, unpublished data). To account for the effects of these drift algae, thewinter experiment included algae and no-algae treatments (2algae) for each of theexperimental urchin densities. Thus, the total number of treatments were: for the

22summer, seven (0, 10, and 20 urchins per m , unmanipulated control, partial cage 122top, partial cage 2 top, top only); for the winter, ten (0 per m 1 algae, 0 2 algae,

10 1 algae, 10 2 algae, 20 1 algae, 20 2 algae, unmanipulated control, partial cage 1

top, partial cage 2 top, top only).There were six replicates for each treatment. The cages were allocated, using a

random number generator, to positions on a rectangular grid (winter: 10 3 6 points;summer: 7 3 6) with 1-m intersection intervals. After the cages and control treatmentswere set out, preliminary surveys were conducted within each replicate. Epifaunalinvertebrates visible to the naked eye were visually identified and counted over a period

2of approximately 15 min. Shoot counts were conducted in a haphazardly located 0.04-mquadrat within each experimental plot. The quadrat was permanently marked withcolored flags so that it could be followed throughout the duration of the experiment.After the initial survey, the appropriate number of urchins was added or removed fromeach cage. The size range of the added urchins, which were collected from a nearby sitein Biscayne Bay, was similar to that of the population at West Point. The number ofurchins in cage control plots was not altered. In the winter experiment, algae wereremoved from those cages designated as no-algae. Tops were sewn on immediately after


addition or removal of urchins and algae. Summer cages were checked every 2–4 daysduring the experiment. Winter cages could not be visited as often because of weatherconditions, but were checked approximately once a week. Most cages had fewer thanfive escaped urchins at any given inspection, but at six inspections there was one cagewith more than 15 urchins missing. All missing urchins were replaced as necessary.Invading algae in the no-algae treatments were also removed as necessary (approximate-ly once every 10 days). Replacement of algae in 1 algae treatments was required inonly five instances.

Each experiment ran for 6 weeks: August–September 1997 (summer) and January–February 1998 (winter). Cages were established in the same general area on bothoccasions. At the end of the 6 weeks, each plot was resurveyed. Visual counts ofepifaunal invertebrates were repeated, as well as shoot counts within the permanentlymarked quadrat. Within each plot a core (15 cm diameter) was taken to a depth of 25 cm.Prior to penetrating the sediment the shoots to be included in the sample weremanipulated into the corer to ensure effective collection of all aboveground biomass.Core samples were sieved through a 0.5-mm mesh sieve. Seagrass samples wereseparated into live below- and above-ground biomass (including epiphytic algae), driedto constant weight at 708C, and weighed. All infaunal macroinvertebrates collected inthe cores were identified and counted.

2.4. Data analysis

The methods employed for surveying invertebrates were effective only for infaunaand for large, slow-moving epifauna (e.g. gastropods, sponges, echinoderms); highlymobile species, such as crustaceans, were not included in the study. The four mostcommon species were chosen as representative of the invertebrate population. Becauseof the low population densities found for most species (see Results), the non-parametricKruskal–Wallis test was used to compare abundances among treatments at the end of theexperiment. Given the lack of mobility of sponges, their abundance was counted at thebeginning and end of the experiment and compared for each treatment using a pairedt-test.

Assumptions of normality and homoscedasticity were tested with the Shapiro–Wilktest and the Bartlett test, respectively (a 5 0.05), prior to all parametric statisticalanalyses. Because shoot counts were conducted within the same permanently markedquadrat at the beginning and end of the experiments, these data were comparedindividually for each treatment with a paired t-test. To test for cage effects, above- andbelow-ground seagrass biomass of the unmanipulated control and the three cage controltreatments were compared with a one-way ANOVA. Because lack of a significantdifference among the cage controls indicated that there was no cage effect on seagrassbiomass (see Results), the cage control treatments were not included in further analyses.For the summer experiment, a one-way ANOVA comparing biomass in unmanipulatedcontrol plots and the three urchin density treatments was performed. For the winter

22experiment, a two-way ANOVA was used, with urchin density (0, 10, or 20 per m )and algae (presence or absence) as fixed factors. Control plots could not be included inthis analysis because algae were not manipulated within them, thus they lacked


appropriate replication for the algae factor. Because no algal effects were detected (seeResults), the 1 and 2 algae treatments were pooled into their respective urchindensities, and a one-way ANOVA was then used to compare these to the unmanipulatedcontrols.

3. Results

The long-term average population density of L. variegatus at the field site (September221995–September 1998) was 1.460.08 (S.E.) urchins per m (Fig. 2). Average monthly

22urchin density at the site ranged from 0.5 to 2.2 m . There was no significant differencein shoot density before and after the experiment in any of the cage control treatments,indicating that there was no cage effect for this variable (Table 1). This was true for bothwinter and summer experiments. In the summer none of the urchin treatments showedsignificant differences in shoot density before and after the 6 weeks (Table 1). The

2220-urchins m treatment, however, was only marginally non-significant, with a p value22 22of 0.056. In the winter, the urchin exclusion (0 m 6algae) and 10 m 2 algae

treatments did not result in significant changes in shoot density, while shoot density in22 22 22the 10 m 1 algae, 20 m 2 algae, and 20 m 1 algae treatments did decrease

significantly (Table 1).No cage effects were detected in either the winter or summer experiments for above-

Fig. 2. Long-term population density of the sea urchin Lytechinus variegatus at West Point, Biscayne Bay,Florida. Data points represent monthly means; error bars are standard error.


Table 1Results of paired t-tests comparing shoot counts in permanently marked quadrats at the beginning and end of

athe experiment

Treatment Initial Final t p

Summer 1997Control 21.565.0 22.864.6 0 1.00Top 22.765.3 26.064.3 2 0.84 0.441Partial 1 top 28.369.6 26.364.0 0.50 0.642Partial 2 top 28.068.0 22.063.5 1.35 0.235

220 m 26.067.9 20.364.1 0.99 0.3662210 m 30.0612.3 28.367.0 0.53 0.6182220 m 25.868.7 17.864.0 2.48 0.056

Winter 1998Control 22.765.5 18.365.4 1.11 0.320Top 25.865.1 26.765.3 2 0.26 0.802Partial 1 top 19.369.4 23.065.9 2 1.48 0.200Partial 2 top 28.264.3 24.565.9 1.09 0.327

220 m 1 algae 24.863.3 20.364.8 1.59 0.174220 m 2 algae 18.063.3 23.865.3 2 1.16 0.297

2210 m 1 algae 29.564.1 19.862.5 3.61 0.0152210 m 2 algae 16.062.6 16.762.0 2 0.23 0.8302220 m 1 algae 30.066.0 10.862.9 4.02 0.0102220 m 2 algae 16.061.0 11.362.1 2.59 0.049

a 22Initial and final shoot count data presented as number of shoots per 0.04 m (6S.E.). See text forexplanation of treatments. Assumptions of normality and homoscedasticity were met. All df 5 5.

or below-ground seagrass biomass (Table 2). In the winter, higher urchin densitysignificantly decreased aboveground T. testudinum biomass, but there was no effect ofalgae on seagrass biomass, either above- or below-ground (Table 3). Winter below-ground biomass was not affected by urchin grazing. Because there was no effect ofalgae, the 1 and 2 algae treatments were combined into their respective urchin

22densities and the three treatments (0, 10, and 20 urchins m ) were compared to theunmanipulated controls with a one-way ANOVA. This comparison showed a significantdifference in aboveground biomass among the four treatments (df 5 41; F 5 6.42;

22p 5 0.001). The 20 m treatment had significantly lower biomass than the control and 022m treatments (Fig. 3a; Tukey–Kramer test). In the summer, urchin grazing, regardless

Table 2aOne-way ANOVA for cage effects for both above- and below-ground seagrass biomass

Control Top Partial 1 top Partial 2 top F p

Aboveground, winter 382.9667.8 383.4696.2 371.46126.5 363.4639.2 0.40 0.752Aboveground, summer 266.4629.7 190.2664.0 234.0661.9 171.8647.6 0.36 0.786Belowground, winter 302.4645.3 312.7642.5 201.3643.2 339.9626.1 2.29 0.109Belowground, summer 184.5633.0 120.8622.9 143.4620.7 207.1635.3 1.85 0.170

a 22Values given are in g dry weight m (6S.E.). Treatments included in ANOVA were: unmanipulatedcontrol plots, tops only, partial cages with tops, and partial cages without tops. Assumptions of both normalityand homoscedasticity were met. All df 5 23.


Table 3aTwo-way ANOVA for T. testudinum biomass in the winter experiment

Source df MS F p

Aboveground biomassUrchin density 2 50.67 8.21 0.001Algae 1 10.35 1.67 0.205Urchin 3 algae 2 0.52 0.08 0.919Error 30 6.17 – –

Belowground biomassUrchin density 2 2.41 0.54 0.590Algae 1 9.35 2.08 0.160Urchin 3 algae 2 1.65 0.37 0.697Error 30 4.50 – –

a Cage control treatments were not included in the analysis because no cage effects were detected (see Table2). Assumptions of both normality and homoscedasticity were met.

of density, had no significant effect on seagrass biomass (Fig. 3b; aboveground: df 5 23;F 5 1.34; p 5 0.289; belowground: df 5 23; F 5 0.13; p 5 0.939).

A total of 34 macroinvertebrate species were identified from the site, but the vast22majority were present in very small numbers ( < 1 per m ). Overall macrofaunal

abundance (including echinoderms, molluscs, polychaetes and sponges) was 144.5615.322individuals m in the summer, and 117.7612.0 in the winter. The four most common

species were the epifaunal gastropod Lithopoma americanum (Turbinidae), the spongeHaliclona permollis (Haliclonidae), and the infaunal polychaetes Marphysa sanguinea(Eunicidae) and Asychis elongata (Maldanidae). The density of H. permollis, averaged

22 22over all treatments, was 0.24 m 60.07 in the summer and 0.31 m 60.07 in thewinter. Results of the paired t-tests indicated that for no treatment, in either winter orsummer, was there a change in the density of H. permollis (all p . 0.05). Similarly, nosignificant differences were found in numbers of L. americanum, A. elongata, or M.sanguinea (Table 4).

4. Discussion

4.1. Effects of urchin grazing on seagrass population

The seagrass bed at West Point harbors a small, but relatively stable, population of L.variegatus. Exclusion of urchins had no effect on either the shoot density or biomass ofT. testudinum, indicating that the natural levels of grazing at West Point are notsufficient to have a significant impact on the biomass or shoot density of the seagrass

´population. This result agrees with the findings of Cebrian et al. (1998), who found thatmoderate grazing of seagrasses, as mimicked experimentally by clipping, has little, ifany, negative effect on leaf growth rate. The present study, however, took place over a


Fig. 3. Biomass of Thalassia after 6 weeks at experimental urchin densities. (a) Winter experiment (August–September 1997): bars represent pooled 1 and 2 algae treatments. Aboveground values with different lettersare significantly different from each other. Belowground biomass did not differ among treatments. (b) Summerexperiment (January–February 1998): no significant differences were detected among any of the treatments foreither above- or below-ground biomass.

period of only 6 weeks. Further experiments are needed to determine if the pattern found´here holds true over longer time periods (Macia, manuscript in preparation).

Two seagrass population variables, biomass and shoot density, were used to assess theeffects of urchin grazing and drift algae at the study site. These two variables, however,did not necessarily react similarly to the experimental treatments. In the summer,experimentally increased urchin grazing did not significantly affect shoot density. At 20

22urchins m , however, the difference in shoot count was only marginally non-significant


Table 4aThe common macroinvertebrates of West Point, Biscayne Bay, Florida

22 2Species Season Density ([ m ) x p

L. americanum Summer 0.0960.04 6.32 0.390Winter 0.1560.04 5.68 0.770

M. sanguinea Summer 53.765.8 7.21 0.302Winter 67.369.7 6.41 0.698

A. elongata Summer 33.067.6 12.38 0.054Winter 61.9611.4 16.77 0.053

a Lithopoma americanum is an epifaunal gastropod; Marphysa sanguinea and Asychis elongata are infaunalpolychaetes. Population densities are averaged over all treatments, and given in mean6S.E. Results of theKruskal–Wallis test comparing final invertebrate densities among all treatments are presented.

22( p 5 0.056). This suggests that, for this time of year, 20 urchins m is the criticalurchin density above which the seagrass shoot density will be negatively affected.During the winter months, this critical density is less than that of the summer. Shoot

22count data indicate that at 10 urchins m , in the presence of the naturally occurringalgae, shoot density is significantly decreased. Winter critical urchin density for negative

22effects on shoot number is therefore approximately 10 urchins m , half of thecorresponding summer value. Thus, the effects of grazing on T. testudinum shoot densityare more pronounced in the winter than in the summer.

Estimated critical urchin densities for negative effects on biomass were higher thanthose for shoot density. Aboveground biomass was affected by urchin grazing only in

22the winter, and then only at the highest density tested (20 urchins m ). Biomass at thisurchin density, however, was not significantly different than the biomass at 10 urchins

22m , indicating that critical urchin density for biomass effects in the winter is22somewhere between 10 and 20 urchins m . Biomass in the summer did not decline

22significantly even at 20 urchins m . Therefore, critical urchin density for biomass22effects in the summer must be at some value higher than 20 per m . Although the

critical urchin densities for shoot number and aboveground biomass differ, bothvariables indicate that grazing on T. testudinum has a more pronounced effect in thewinter. This conclusion is similar to that found by Valentine and Heck (1991), who usedcages to study L. variegatus grazing effects in the northern Gulf of Mexico. Their studyindicated that approximately twice as high an urchin abundance is required to overgraze

22aboveground T. testudinum biomass in the summer ( . 40 urchins m ) than in the22winter ( , 20 urchins m ). The results of Valentine and Heck (1991) pertain to

complete defoliation of the seagrasses, however; even their lowest urchin density (1022urchins m ) was enough to cause significant decreases in biomass at all times of the

year.At West Point, the critical urchin density required for negative effects on aboveground

22biomass ( 4 20 and 10–20 urchins m in the summer and winter, respectively)22appears to be higher than that for shoot density (20 and 10 urchins m in the summer

and winter, respectively). In other words, shoot density is affected by urchin grazingbefore effects on biomass become apparent. This pattern may be a result of urchingrazing activity causing an increase in T. testudinum shoot-specific productivity.


Increased productivity could manifest itself as longer or more numerous leaves pershoot, either of which would produce greater amounts of biomass. In this way,aboveground biomass may remain unaffected by grazing even though shoot density isdecreasing. A grazing-induced increase in productivity has in fact been documented for aT. testudinum population in the Caribbean island of St. Croix, which exhibits increasedspecific growth rate when subject to moderate levels of turtle or urchin grazing (Ziemanet al., 1984). Studies from terrestrial grasslands suggest several mechanisms throughwhich grazing can increase plant productivity. These mechanisms include: increasedrates of photosynthesis stimulated by the loss of photosynthetic products; reallocation ofnutrient reserves from the root system; increased light penetration resulting fromdecreased self-shading effects; and nutrient input in the form of grazer feces /urine(McNaughton, 1979 and references therein; Milchunas and Lauenroth, 1993). Of themechanisms mentioned above, fertilization of seagrasses by urchin feces may be

´particularly important to T. testudinum (Macia, manuscript in preparation).The results of this study suggest that the vulnerability of seagrasses to grazing

depends on the level of grazing pressure that the seagrass population normallyexperiences. Seagrass beds exposed to relatively low grazing rates, such as at WestPoint, may be more resistant to short-term increases in grazing rates. Critical urchindensity for negative effects on seagrass biomass estimated from this study is higher than

22the 10 urchins m found by Valentine and Heck (1991) in the Gulf of Mexico. Thiscontrast may be a result of the different grazing pressures exerted naturally on these twoseagrass beds. In the Gulf of Mexico site, urchin density ranged from 12 to 25 urchins

22 22m , whereas in Biscayne Bay the urchin density is only 1.4 per m . Field studies fromJamaica show that T. testudinum subjected to repeated cropping every 70 days exhibits adecline in biomass after the sixth and seventh harvests (Greenway, 1974). Valentine andHeck (1991) stated that conditions for overgrazing are common at their site. Periodicovergrazing may make this T. testudinum population more susceptible to grazing effectsthan the West Point population, which appears to suffer virtually no impact from thenaturally low grazing pressure. Using caging experiments, Keller (1983) investigatedanother Jamaican population of L. variegatus with a relatively low population density

22(3.7 per m ). On all but one of the sampling dates, experimental densities of 16 urchins22m had no significant effect on aboveground T. testudinum biomass. This situation is

similar to the present study, where a seagrass population with low natural urchinabundance appears to be relatively resistant to experimentally increased grazingpressure.

4.2. Effects of drift algae and interaction with grazing

Removal of algae in the urchin exclusion cages did not have a positive effect onseagrass shoot density, indicating that drift algae alone do not appear to negatively affectthis parameter. Similarly, drift algae did not affect seagrass biomass. For shoot density,however, there was an interaction between drift algae and increased urchin density.Removal of algae had a positive effect on T. testudinum shoot density in the 10 urchins

22m treatment: cages without algae did not experience a significant decrease in shootcount, while those with algae did. Although shoot density declined both with and


22without algae in the 20 urchins m treatment, the effect was more pronounced for the1 algae treatment, as shown by the lower p-value in the paired t-test.

Hull (1987) discussed the potential effects, both positive and negative, of benthicmacroalgal mats on the surrounding community. Among these are reduced watervelocity, flushing and oxygen exchange between the sediment and water column. Whilethese factors may not be of great importance to T. testudinum, which thrives in anoxicsediments (Zieman, 1982), competition for light can be very important (Short et al.,1995). The drift algal mats at the study site are very thick and can completely obscurelarge seagrass patches of many square meters in area (personal observation). Neverthe-less, the results of the present study suggest that the annual algal bloom alone has nonegative effects on T. testudinum shoot density or biomass. Other studies, however, haveshown that if such patches remain in place for sufficiently long periods of time (e.g. 6months) they can cause substantial loss of aboveground biomass and create bare patches(Holmquist, 1997).

Although the naturally-occurring drift algal blooms do not appear to have significanteffects on the T. testudinum population, when combined with increased grazing pressurethere are synergistic effects that can greatly decrease seagrass shoot density. Increasedgrazing pressure exacerbates the negative effects of the drift algae (or vice versa). Whensubject to both algal competition and increased urchin grazing, T. testudinum shootdensity experiences a greater decline than when there is only an increase in grazingactivity. Thus it appears that the non-significant impact of the naturally occurring urchinpopulation is important to the persistence of the West Point seagrass community.Increased grazing pressure from a larger urchin population, coupled with the negativeeffects of algae, could be so detrimental as to prevent full recovery of the seagrasspopulation from the annual algal bloom.

4.3. Effects on the invertebrate population

The numbers of invertebrates found in this study are relatively low in comparison toother seagrass bed studies, which often report invertebrate abundances in the thousands

22per m (Santos and Simon, 1974; Brook, 1978; Gore et al., 1981; Greening andLivingston, 1982; Virnstein et al., 1983; Bauer, 1985; Virnstein and Howard, 1987;Valentine and Heck, 1993; Greenway, 1995). One factor undoubtedly contributing to thelow observed faunal numbers in the present study is sampling design. The samplingprotocol used in this study could not effectively include crustaceans, which are amongthe most common inhabitants of seagrass beds (Orth, 1973; Gore et al., 1981; Lewis,1984; Holmquist et al., 1989). Even with the absence of crustacean numbers taken intoaccount, however, the abundance of invertebrates at the study site is still considerablylower than that found in other studies. The reasons for this are unclear, and await theresults of further studies.

Four invertebrate species, other than L. variegatus, were commonly found at WestPoint. None of these species showed a substantial response to increased urchin grazing,even when such grazing had significant effects on the seagrass. This result is notsurprising for the sponge Haliclona permollis, as L. variegatus does not generally feedupon sponges. The herbivorous gastropod Lithopoma americanum, however, resides on


the seagrass blades, presumably feeding on epiphytes (Emerson and Jacobson, 1976). Asthere was no decrease in aboveground biomass of T. testudinum in the summerexperiment, the lack of an effect on L. americanum is not surprising. In the winter,however, there was a decrease in seagrass biomass in the high urchin density treatment.Nevertheless, a concurrent decline in the abundance of L. americanum did not occur.The population density of L. americanum found during the winter was very low 2

220.1560.04 individuals m . It is possible that such a small population could maintainitself on what seagrass remained, despite the significant losses in aboveground biomass.

The polychaetes Marphysa sanguinea and Asychis elongata feed primarily on detritus(Day, 1967; Fauchald and Jumars, 1979; Prevedelli, 1992). Neither species was affectedby the experimental treatments. As part of the infauna, these animals would notexperience any direct impact from the algae or the urchins, which are primarilyaboveground grazers. It is also unlikely that, over the short duration of the experiment (6weeks), increased grazing pressure would affect the amount of detritus available belowthe surface of the sediment, where these species feed. Long-term exposure to increasedurchin grazing activity, however, could potentially increase the amount of organic

´material in the sediment via deposition of feces (Macia, manuscript in preparation).These could eventually be worked into the sediment and benefit the infaunal detritivores.

5. Conclusions

The T. testudinum population at West Point in Biscayne Bay, Florida, experiences lowlevels of grazing from a small but relatively stable population of the sea urchinLytechinus variegatus. This natural grazing pressure does not appear to have significantconsequences for either the biomass or shoot density of the seagrass population.Resistance of T. testudinum to grazing is higher in the summer than in the winter. Shootdensity is more vulnerable to urchin grazing than aboveground biomass, possibly as aresult of grazing-induced increases in seagrass productivity. The overall vulnerability ofa seagrass population to grazing may depend on the level of grazing to which it isnormally exposed. Seagrass beds, such as at West Point, with small herbivorepopulations may be more resistant to temporarily increased grazing than beds thatnormally experience higher grazing pressure. The normal presence of drift algal bloomsdoes not appear to be detrimental to T. testudinum, but when combined with increasedgrazing pressure can have serious negative effects on the shoot density of the seagrass.At West Point, a site with relatively low faunal abundances, increased urchin grazingappears to have neither positive nor negative effects on the sampled invertebratecommunity.

Acknowledgements

The establishment, maintenance, and break-down of over 100 cages and experimentalplots would not have been possible without the invaluable field assistance of thefollowing people, whose curiosity about working with cages has undoubtedly been


forever satisfied: B. Orlando, P. Biber, L. Kaufman, T. Jones, E. Irlandi, D. Lirman, M.Brown, A. Morales, and J. Wiley. The comments of M. Harwell, E. Irlandi, J. Leal, A.Szmant, N. Voss, an anonymous reviewer and especially M. Robinson helped greatly inshaping my rough early drafts into this final version. I am also very grateful to S.Schultze for his help in the identification of polychaetes. Financial support for this workwas provided by NOAA Coastal Ocean Program Award [NA37RJ0149, The Sanibel-Captiva Shell Club, and the RSMAS Anonymous Donor Award. This research issubmitted in partial fulfillment of the requirements for the PhD degree at the Universityof Miami. [RW]

References

Bauer, R.T., 1985. Diel and seasonal variation in species composition and abundance of caridean shrimps(Crustacea, Decapoda) from seagrass meadows on the north coast of Puerto Rico. Bull. Mar. Sci. 36,150–162.

Bell, S.S., Hall, M.O., 1997. Drift macroalgal abundance in seagrass beds: investigating large-scaleassociations with physical and biotic attributes. Mar. Ecol. Prog. Ser. 147, 277–283.

Brook, I.M., 1978. Comparative macrofaunal abundance in turtlegrass (Thalassia testudinum) communities inSouth Florida characterized by high blade density. Bull. Mar. Sci. 28, 212–217.

Camp, D.K., Cobb, S.P., van Breedveld, J.F., 1973. Overgrazing of seagrasses by a regular urchin, Lytechinusvariegatus. Bioscience 23, 37–38.

´Cebrian, J., Duarte, C.M., Agawin, N.S.R., Merino, M., 1998. Leaf growth response to simulated herbivory: acomparison among seagrass species. J. Exp. Mar. Biol. Ecol. 220, 67–81.

Day, J.H., 1967. A Monograph of the Polychaeta of Southern Africa. Part 1. Errantia, British Museum ofNatural History, London.

Emerson, W.K., Jacobson, M.K., 1976. The American Museum of Natural History Guide to Shells, Knopf,New York.

Engstrom, N.A., 1982. Immigration as a factor in maintaining populations of the sea urchin Lytechinusvariegatus (Echinodermata: Echinoidea) in seagrass beds on the southwest coast of Puerto Rico. Stud.Neotrop. Fauna Environ. 17, 51–60.

Fauchald, K., Jumars, P., 1979. The diet of worms: a study of polychaete feeding guilds. Oceanogr. Mar. Biol.Annu. Rev. 17, 193–284.

Gore, R.H., Gallaher, E.E., Scotto, L.E., Wilson, K.A., 1981. Studies on decapod crustacea from the IndianRiver Region of Florida. Estuar. Coast. Shelf Sci. 12, 485–508.

Greening, H.S., Livingston, R.J., 1982. Diel variation in the structure of seagrass-associated epibenthicmacroinvertebrate communities. Mar. Ecol. Prog. Ser. 7, 147–156.

¨Greenway, M., 1974. The effects of cropping on the growth of Thalassia testudinum (Konig) in Jamaica.Aquaculture 4, 199–206.

Greenway, M., 1976. The grazing of Thalassia testudinum in Kingston Harbour. Jamaica. Aquat. Bot. 2,117–126.

Greenway, M., 1995. Trophic relationships of macrofauna within a Jamaican seagrass meadow and the role ofthe echinoid Lytechinus variegatus (Lamarck). Bull. Mar. Sci. 56, 719–736.

Hansen, K., Kristensen, E., 1997. Impact of macrofaunal recolonization on benthic metabolism and nutrientfluxes in a shallow marine sediment previously overgrown with macroalgal mats. Estuar. Coast. Shelf Sci.45, 613–628.

Holmquist, J.G., 1994. Benthic macroalgae as a dispersal mechanism for fauna: influence of a marinetumbleweed. J. Exp. Mar. Biol. Ecol. 180, 235–251.

Holmquist, J.G., 1997. Disturbance and gap formation in a marine benthic mosaic: influence of shiftingmacroalgal patches on seagrass structure and mobile invertebrates. Mar. Ecol. Prog. Ser. 158, 121–130.


Holmquist, J.G., Powell, G.V.N., Sogard, S.M., 1989. Decapod and stomatopod assemblages on a system ofseagrass-covered mud banks in Florida Bay. Mar. Biol. 100, 473–483.

Hull, S.C., 1987. Macroalgal mats and species abundance: a field experiment. Estuar. Coast. Shelf Sci. 25,519–532.

Keller, B.D., 1983. Coexistence of sea urchins in seagrass meadows: an experimental analysis of competitionand predation. Ecology 64, 1581–1598.

Lewis, F.G., 1984. Distribution of macrobenthic crustacean associated with Thalassia, Halodule and bare sandsubstrata. Mar. Ecol. Prog. Ser. 19, 101–113.

´Macia, S., Lirman, D., 1999. Destruction of Florida Bay seagrasses by a grazing front of sea urchins. Bull.Mar. Sci. 65, 593–601.

McGlathery, K.J., 1995. Nutrient and grazing influences on a subtropical seagrass community. Mar. Ecol.Prog. Ser. 122, 239–252.

McNaughton, S.J., 1979. Grazing as an optimization process: grass–ungulate relationships in the Serengeti.Am. Nat. 113, 691–703.

Milchunas, D.G., Lauenroth, W.K., 1993. Quantitative effects of grazing on vegetation and soils over a globalrange of environments. Ecol. Monogr. 63, 327–366.

Montague, J.R., Ortiz, L., Arguelles, A., Millan, J.M., Cardoch, L., 1988. Density and dispersion estimates forsea urchins in a south Florida seagrass community. Fla. Sci. 51, 19–22.

Moore, H.B., Jutare, T., Bauer, J.C., Jones, J.A., 1963. The biology of Lytechinus variegatus. Bull. Mar. Sci.13, 23–53.

Ogden, J.C., 1976. Some aspects of herbivore–plant relationships on Caribbean reefs and seagrass beds. Aquat.Bot. 2, 103–116.

Ogden, J.C., 1980. Faunal relationships in Caribbean seagrass beds. In: Phillips, R.C., McRoy, C.P. (Eds.),Handbook of Seagrass Biology, Garland Press, New York, pp. 173–198.

Oliver, G.D., 1987. Population Dynamics of Lytechinus variegatus. MS thesis, University of Miami.Orth, R.J., 1973. Benthic infauna of eelgrass, Zostera marina, beds. Chesapeake Sci. 14, 258–269.Prevedelli, D., 1992. Influence of temperature and diet on the larval development and growth of juveniles

Marphysa sanguinea (Montagu) (Polychaeta, Eunicidae). In: Dauvin, J.C., Laubier, L., Reish, D.J. (Eds.),Proceedings of the Fourth International Polychaete Conference, pp. 521–526.

Santos, S.L., Simon, J.L., 1974. Distribution and abundance of the polychaetous annelids in a south Floridaestuary. Bull. Mar. Sci. 24, 669–689.

Short, F.T., Burdick, D.M., Kaldy, J.E., 1995. Mesocosm experiments quantify the effects of eutrophication oneelgrass, Zostera marina. Limnol. Oceanogr. 40, 740–749.

Valentine, J.F., Heck, K.L., 1991. The role of sea urchin grazing in regulating subtropical seagrass meadows:evidence from field manipulations in the northern Gulf of Mexico. J. Exp. Mar. Biol. Ecol. 154, 215–230.

Valentine, J.F., Heck, K.L., 1993. Mussels in seagrass meadows: their influence on macroinvertebrateabundance and secondary production in the northern Gulf of Mexico. Mar. Ecol. Prog. Ser. 96, 63–74.

Virnstein, R.W., Carbonara, P.A., 1985. Seasonal abundance and distribution of drift algae and seagrasses in themid-Indian River Lagoon, Florida. Aquat. Bot. 23, 67–82.

Virnstein, R.W., Howard, R.K., 1987. Motile epifauna of marine macrophytes in the Indian River Lagoon,Florida. II. Comparisons between drift algae and three species of seagrasses. Bull. Mar. Sci. 41, 13–26.

Virnstein, R.W., Mikkelson, P.S., Cairns, K.D., Capone, M.A., 1983. Seagrass beds versus sand bottoms: thetrophic importance of their associated benthic invertebrates. Fla. Sci. 46, 363–381.

Zieman, J.C., 1982. The ecology of the seagrasses of south Florida: a community profile, US Fish and WildlifeService, Biol. Serv. Program FWS/OBS-82/25.

Zieman, J.C., Iverson, R.L., Ogden, J.C., 1984. Herbivory effects on Thalassia testudinum leaf growth andnitrogen content. Mar. Ecol. Prog. Ser. 15, 151–158.

the effects of sea urchin grazing and drift algal blooms on a subtropical seagrass bed community

Documents