-
I Journal of Coastal Research I SI 1 40 T 64-78 | West Palm
Beach, Florida I Winter 2005
Oyster Reef Habitat Restoration:Relationships Between Oyster
Abundanceand Community Development based onTwo Studies in Virginia
and SouthCarolinaMark W. Luckenbacht, Loren D. Coent, P.G. Ross,
Jr.§, and Jessica A. Stephentt
tVirginia Institute ofMarine Science
College of William andMary
P.O. Box 350Wachapreague, VA
23480, [email protected]
tMarine ResourcesResearch Institute
South CarolinaDepartment ofNatural Resources
P.O. Box 12559Charleston, SC 29422,
[email protected].
sc.us
§Virginia Institute ofMarine Science
College of William andMary
P.O. Box 350Wachapreague, VA
23480, [email protected]
ttMarine ResourcesResearch Institute
South CarolinaDepartment ofNatural Resources
P.O. Box 12559Charleston, SC 29422,
[email protected].
state.sc.us
ABSTRACT
@@lguilloo9ek.assee
_M_ma
lt- _O__~
LUCKENBACH, M.W.; COEN, L.D.; ROSS, P.G. JR. and STEPHEN, JA.,
2005. Oyster reef habitat res-toration: relationships between
oyster abundance and community development based on two studies
inVirginia and South Carolina. Journal of Coastal Research, SI(40),
64-78. West Palm Beach (Florida), ISSN0749-0208.
Most Atlantic and Gulf coast U.S. states with an oyster fishery
have operated some form of oyoter reefenhancement program over the
past 50 years. Although programs were initially only directed at
oysterfisheries augmentation, recent emphasis has shifted to
include the restoration of their ecological functions.Furthermore,
many of these programs are managed by environmental organizations
or state agencies nottraditionally involved in fisheries management
or research, but rather in ecological restoration,
monitoring,and/or environmental education. A simple assessment of
shellfish meetings over the past five years, in-cluding the
inaugural Restore America's Estuaries meeting from which this paper
is derived, revealed morethan 300 presentations related to oyster
restoration, with fewer than 25% focused solely on oyster
fisheryrestoration. Unfortunately, many of those efforts lacked
well-defined "success criteria," with progress oftenjudged using
fisheries-based metrics ouch as market-sized (generally 75 mm or
3") oysters. Here we discussour findings as they relate to the
value of alterative restoration metrics and associated success
criteriausing data from two very different systems and approaches:
one conducted in Virginia's lower ChesapeakeBay (Rappahannock
River), based on data from a two-year program utilizing oubtidally
constructed reefsof different reef "scale," and the other a
long-term study in South Carolna focusing on intertidal reefs.
Foreach system, we compared newly created reef structures, relating
oyster abundance and size to residentspecies abundance and
biodiversity over time. Our results revealed positive correlations
between severalcommunity descriptors and the size and density of
oysters on the reefs. Of the 15 significant (and b mar-ginaly
insignificant) correlations observed out of a total of 78 examined
across both studies, al but onewere positive. The exception was for
epifaunal invertebrate diversity vs. oyster biomass on the
Rappahan-nock reefs. Despite these numnerous positive correlations,
none indicated that market-sized oysters are aprerequisite for
supporting an abundant and diverse community. For example,
intertidal oysters >75 mmin South Carolina typically make up
-
Evaluating Success Criteria and Reef Development for Oyster
Restoration
restore depressed oyster populations over the pastdecade since
the workshop on oyster restoration in1995 at the Virginia Institute
of Marine Science(LUCKENBACH et al, 1999). There is now
evidencethat oysters can (or could at historical abundances)control
phytoplankton abundance and alter estu-arine food webs by enhancing
benthic-pelagic cou-pling (DAME et al., 1984, 2001; DAME and
LIBES,1993; NEWELL, 1988; ULANowicz and TUTTLE, 1992;ROTHSCHILD et
al., 1994). Indeed, there is an in-creasing recognition that
top-down control of phy-toplankton abundance via oysters should be
an im-portant part of overall strategies to improve waterquality in
eutrophic estuaries (NEWELL, 1988;KAUFMAN and DAYTON, 1997;
PETERSON and LuB-CHENCO, 1997; JACKSON et al., 2001). Moreover,
oys-ters are quintessential "ecosystem engineers"(JoNEs et al.,
1994; LENIHAN, 1999), constructingbiogenic habitats that provide
refuges, nestingsites, and foraging grounds for a variety of
resi-dent and transient species (LENrHAN and GRA-BOWSKI, 1998;
BREITGURG, 1999; COEN et al, 1999b;EGGLESTON et al, 1999; HARDING
and MANN, 1999;POSEY et al, 1999). Several studies have
nowdemonstrated greater biodiversity associated withoyster reefs
than with adjacent sedimentary hab-itats (POSEY et al, 1999; COEN
and LUCKENBACH,2000; O'BEIRN et al, 2000). In fact, for many
es-tuaries along the Mid-Atlantic and Gulf coasts ofthe U.S.,
oyster reefs are the primary source ofhard substrate, and as such
they may supportunique assemblages of organisms. Further, thereis
evidence that oyster reefs contribute signifi-cantly to enhanced
production, not merely the con-centration of finfish and decapod
crustaceans (PE-TERSON et al., 2003).
Over the past decade, the number of oyster res-toration projects
along the U.S. Atlantic and Gulfof Mexico coasts has increased
dramatically, large-ly in an effort to restore one or more of the
poten-tially lost ecological services (see reviews above).Many of
these projects are being conducted bystate and federal agencies not
typically involved infisheries management, such as the Virginia
De-partment of Environmental Quality, the SouthCarolina Department
of Health and Environmen-tal Control, the National Oceanic and
AtmosphericAdministration (NOAA) Restoration Center, theNOAA
Coastal Services Center, and the U.S. ArmyCorps of Engineers. In
addition, numerous non-governmental organizations and
communitygroups (e.g, NY/NJ Baykeepers, Chesapeake BayFoundation,
South Carolina's Oyster Restoration
and Enhancement Program (or SCORE) withSouth Carolina Coastal
Conservation League andCoastal Conservation Association (CCA),
NorthCarolina Coastal Federation, and Tampa Bay-watch) are actively
involved in oyster reef resto-ration for ecological motives.
Funding sources foroyster restoration have also included agencies
andorganizations with environmental restorationagendas (e.g.,
NOAA's Community RestorationProgram and FishAmerica). One
indication of thisenhanced level of restoration activity is the
largenumber of papers presented by such groups on oys-ter
restoration at recent meetings of the Interna-tional Conference of
Shellfish Restoration, the Na-tional Shellfisheries Association,
and at the Ma-rine Benthic Ecology meeting. A review of the
pub-lished abstracts and program schedules for thosemeetings over
the past five years and the inau-gural meeting of Restore America's
Estuaries re-veals more than 300 presentations related to oys-ter
restoration with fewer than 25% focused onoyster fishery
restoration (LUCKENBACH, unpub-lished data).
Unfortunately, many of the projects referred toin the previous
section have limited data, with fewresults published to date in the
primary literature.In the absence of expressly stated success
criteria(COEN and LUCKENBACH, 2000) and directed mon-itoring, the
early success of oyster restoration pro-jects, unfortunately, tends
to be judged based sole-ly on either the abundance of market-sized
(typi-cally 75 mm or 3") oysters or fishery landings data,neither
of which may be crucial to achieving moreecologically based
restoration goals.
Previously, we (COEN and LUCKENBACH, 2000)stressed the
importance of developing metrics forevaluating the specific goals
of ecological restora-tion. It is clear from that earlier work that
a viableoyster population is a critical component of a suc-cessful
oyster reef restoration effort. However, itis not clear if
harvestable quantities of market-sized (75 mm shell height (SH))
oysters are a crit-ical requisite for restoration success. In
stateswhere minimum harvest sizes are regulated (e.g.,75 mm SH in
Virginia and elsewhere) oysters maybe reproductively capable and
populations sustain-able, with relatively low abundances of
market-sized animals. Studies have now begun to dem-onstrate that
not all of the "ecological services" ofoyster communities come only
after the oyster pop-ulations are well established. In several
recent re-views (COEN et al, 1999a, 1999b, 2000; COEN
andLUCKENBACH, 2000; BREITBURG et al, 2000) we sug-
Journal of Coastal Research, Special Issue No. 40, 2005
65
-
Luckenbach et al.
gested that oysters, although the key to a fullyfunctioning
oyster reef, are not necessary to pro-vide many of the ecosystem
services that, for ex-ample, artificial reefs provide, such as
structureand refugia (BREITBURG, 1999; EGGLESTON et al.,1999;
LEHNERT and ALLEN, 2002; GLANcY, 2003).Biogeochemical coupling
benefits as measured inthe "CREEK" Project in South Carolina (DAME
etal., 2000, 2001) showed that nekton feeding cou-pled nutrients
around structured sites without liveoysters. Hence, oysters may not
be critical, initial-ly, in judging success and should not be used
toimply failure early in the life of a restoration pro-gram.
Alternatively, as suggested by COFN et al.(1999a) and BREITBURG
et al. (2000), large oystersand "mature" reefs may be critical to
achievingboth fisheries enhancement and restoration goals.In
discussing many of the challenges faced in at-tempting to meet
fisheries rehabilitation and eco-logical restoration goals for
oyster reefs, COEN andLUCKENBACH (2000) further stressed the
impor-tance of developing meaningful success measuresand
implementing rigorous monitoring programsto track progress towards
those goals. Thus, to bet-ter evaluate the success of ecological
restorationefforts, we need to develop a better understandingof the
relationship between oyster populationstructure and abundance and
any potential "eco-logical services" that we are seeking to
restore.
In response to this need, we present the resultsfrom two studies
conducted in different estuarieson the U.S. Atlantic coast. One
study in the Rap-pahannock River, a mesohaline sub-estuary of
theChesapeake Bay, is still ongoing and is addressingthe role of
spatial scale (ranging from a few metersto several kilometers) on
the development of oysterpopulations and associated fauna. The
other study,conducted along the central coast of South Caro-lina,
compared the development of oyster popula-tions and associated
assemblages on constructedreefs to adjacent natural intertidal
reefs over a six-year period between 1995 and 2001. Both
studiesafford the opportunity to relate the
reef-associatedassemblages of organisms to oyster densities
andpopulation size structure over time. Our objectivein examining
these relationships is to provide abasis for beginning to formulate
metrics for res-toration success that reflect the biodiversity
andhabitat goals of many projects and to make rec-ommendations for
future work.
Figure 1. Experimental reef restoration sites in the
Rap-pahannock River, Chesapeake Bay, Virginia.
METHODS
Rappahannock River, Virginia
Study Site and Reef Construction
This study was conducted at four sites in thelower portion of
the Rappahannock River, Virgin-ia, USA, which is a tributary of the
ChesapeakeBay (upriver-most site, Drumming Ground: N 37039.248', W
760 27.648' downriver-most site: MillCreek N 370 35.157', W 760
24.024', see Figure 1).Historically, this region of the
Rappahannock wasa highly productive oyster harvesting area,
withextensive natural reefs (HARGIs, 1999). The specif-ic sites
chosen for the construction of reefs in thisstudy formerly
supported viable oyster reefs thatthrough a combination of
over-fishing, disease,and habitat degradation had all but
disappeared.Reef bases were constructed in August 2000 byplacing
shell piles in arrays as shown in Figure 2.Core material for
individual mounds was com-prised of surf clam (Spisula solidissima)
shell thatwas capped off with a veneer (generally 10-20 cm)of clean
oyster shell. Materials were barged to thefour reef sites and
deployed via a crane and bucketrig, creating "upside-down egg
carton" shaped sub-tidal reefs elevated approximately 3 m above
sea-bed and 1-2 m below the water surface at meanlow water (Figure
2). Reefs ranged in area, from
Journal of Coastal Research, Special Issue No. 40, 2005
66
-
Evaluating Success Criteria and Reef Development for Oyster
Restoration
WA= SUwAVSQdIM)
t
Figure 2. Generalized aerial footprint of reefs denoting
in-tra-reef locations. Each circle represents a mound
approxi-mately 10 m diameter. Generalized side view of an
individ-ual shell mound shown for each replicate reef mound at
ar-row.
approximately 400 m2 to 8,000 m2 . Intra-reef lo-cations were
designated in relation to distancefrom reef edge along longitudinal
axes (Figure 2).A subsequent manuscript will address the
devel-opment of oyster populations and associated com-munities in
relation to scale. Here we present ourfindings relating the density
and population struc-ture of oysters to the development of
reef-associ-ated community assemblages. For more detail,
seeLUCKENBACH and Ross (2003).
Sampling Methods
Standing stocks of oysters were estimated fromdiver-collected
samples taken at all reefs. Fifty-one replicate 0.25 m x 0.25 m
quadrates were hap-hazardly placed onto randomly-selected
moundswithin ten reefs (number of samples partitioned byreef size,
i.e., 5 of the 6, 7 of the 12 and 8 of the20 mounds for small,
medium, and large "reefs,"respectively; see Figure 2 and LuCKENBACH
andRoss, 2003, for details). One sample was collectedfrom each
mound, as mounds were treated as rep-licates within a reef
treatment, such that multiplereplicate samples were taken at the
time of sam-pling (see Figure 2). All reef material was exca-vated
to a depth of 10 cm by divers and transport-ed to the surface in
fine mesh bags. All live oystersin each sample were counted and SH
(longest lipto hinge linear distance, the standard measure
foroyster) measured. Samples were collected in July2001, October
2001, and July 2002.
Figure 3. Inlet Creek study site in Charleston Harbor,South
Carolina. See COEN et al., 1999b, and COEN and LucK-ENBACH, 2000,
for additional details.
A sub-sample of 132 oysters covering the rangeof oyster SH
encountered was measured for drytissue biomass from the October
2001 sample. Theshear number of oysters encountered
prohibitedashing all oysters throughout the study, justifyingthe
use of a regression equation to estimate bio-mass. Biomass values
for oysters were computedfrom a regression of ash-free dry weight
on shellheight [biomass (mg) = 0.007 X shell height (inmm)2'5 614,
R2 = 0.8988, n = 132]. All sessile epi-fauna on the reef substrate
or on the oysters inquadrat samples were identified to the
lowestpractical taxon and reported as either densities(e.g., for
barnacles and tunicates) or percent cover(e.g, for bryozoans and
sponges).
Small resident mobile fishes and crustaceanswere sampled using
substrate baskets embeddedin the reef. Thirty-centimeter diameter
PVC pipewas cut into 15-cm lengths and one end coveredwith 1-mm
plastic mesh. Three 5-cm diameterholes were cut along the midline
of the PCV ringand covered with 1-mm mesh. Baskets were thenfilled
with clean oyster shells similar to those usedin the reef
construction and buried flush with thereef surface by divers. The
mesh bottom and holesin the sides permitted the exchange of
interstitialpore water with the surrounding reef, while thebasket
allowed the retrieval of intact sampleswhich retained the more
mobile reef residentssuch as blennies, gobies, and mud crabs.
DuringApril 2001, a total of 189 baskets were deployed
athaphazardly located positions on a subset of thereef crests (1
per randomly selected replicatemound; see above replicate
allocation for the quad-rats) due to logistics and time
constraints, versus
Journal of Coastal Research, Special Issue No. 40, 2005
67
-
65 Luckenbach et al.
the alternative of randomly picking the exact spoton each
subtidal reef, as we had no reason to thinkthe micro-location of a
basket on a small moundwould yield any bias. Divers retrieved the
abovereplicate baskets in July 2001 (n = 62), October2001 (n = 60),
and July 2002 (n = 24) from allcombinations of reef locations. The
small numberof samples retrieved during the last sampling pe-riod
reflects loss of gear due to the erosive forcesduring the 15 months
that the final baskets werein the field. In the laboratory, all
motile organismsin the baskets were thoroughly rinsed over a
1-mmsieve mesh to remove all motile organisms, whichwere fixed in
isotonic Normalin® and then trans-ferred to 70% ethanol. All
organisms were lateridentified to the lowest practical taxon,
enumer-ated, and, where appropriate, measured (decapodcrustaceans,
carapace width, and finfish, totallength). Taxa such as amphipods
and polychaeteswere not measured.
Transient fishes associated with the reefs weresampled using
gill nets. Although nets were setduring 2001, only data collected
during two sam-pling efforts in May and June 2002 were used
foranalysis. As previously mentioned, we also usedoyster population
data from this period for analy-ses. Nets were 9 m long by 3 m high
and rigged tofish from the seabed up (i.e. sinking rigged net).Nets
utilizing 6.3 cm and 7.5 cm stretch mesh wereused during 2002 and
were randomly allocatedthroughout sampling periods. Anchored
monofila-ment gill nets were deployed for 3 h on all reefsizes at
both inner and outer reef locations whenapplicable. Sets were
repeated so that all locationswere sampled with both mesh sizes
during bothflood and ebb tidal cycles within sampling
periods.Although the majority of gill net sampling oc-curred
between dawn and dusk, one sample effortthat included all scale
treatments was undertakenduring the night. Nets were randomly
allocated tospecific locations witbin each region of the reefs.This
resulted in over 200 individual sets (seeLUCKENBACH and Ross, 2003
for more details). Af-ter 3 hr, nets were retrieved and fish were
iden-tified, enumerated, measured, and released ashort distance
from the reefs. In some cases, dueto high catches, processing of
samples had to beundertaken after all nets were harvested and
tak-en to a remote location.
Statistical Analyses
Temporal patterns in oyster abundance and bio-mass, along with
abundances of selected species
and community metrics, are presented graphicallyfor all sites
combined. One-way ANOVAs wereused to test the effect of reef site
on oyster abun-dance and biomass for July 2002 samples only (n= 4).
Tukey's Multiple Comparisons were subse-quently utilized to
elucidate reef site differences(SOKAL and ROHLF, 1981).
Furthermore, fordata from this sample period, we tested for
differ-ences between individual reefs (n = 10) indepen-dent of
geographic location. Both analyses weremeant to provide some
background regarding thegeneral oyster populations prior to
subsequent cor-relation analyses that are undertaken at the
in-dividual reef level. Oyster reefs were constructedin 2000, which
resulted in missing oyster settle-ment for that year, so no
measured settlement wasquantified until Fall 2001. We chose to use
the2002 sampling only because it represented 2001recruitment and
mortality, along with recruitmentand mortality through July 2002,
therefore paint-ing a more appropriate picture of existing
oysterpopulations. This is important because these werenew reefs
just developing oyster populations dur-ing the course of this
study. Size frequency datafor oysters from this same sampling date
are pre-sented graphically by site.
Spearman product-moment correlation coeffi-cients related total
oyster abundance, abundanceof age-class-two oysters, and biomass to
the abun-dances and biomass of dominant reef-associatedspecies to
selected community descriptors. Com-parisons included abundance and
biomass (forribbed mussels (Guekinsia demissa) only) of dom-inant
(based on measured abundance in thisstudy), reef-associated species
in logical groups, aswell as total abundance, species richness, and
di-versity (Shannon-Weiner Diversity Index) ofbroader taxonomic
groupings. Broad functionalgroupings were (1) epifaunal
invertebrates, (2) res-ident finfish, and (3) transient finfish.
Dominantspecies/group specific analyses were (1) for at-tached
epifauna, barnacles (Balanus spp.), (2) fordecapod crustaceans, mud
crabs (Xanthidae), (3)for bivalves, ribbed mussels, (4) for
resident fin-fish, skilletfish (Gobiesox strumosus), and (5)
fortransient finfish, white perch (Morone american-us). For
transient species, in addition to the dom-inant species, striped
bass (Morone saxatilis) datawere included because of their
management im-portance along the Atlantic seaboard. We deemedthese
functional groups to be the most logical as-semblage that we were
able to quantitatively sam-ple in this study.
Journal of Coastal Research, Special Issue No. 40, 2005
68 Luckenbach et al.
-
Evaluating Success Criteria and Reef Development for Oyster
Restoration
Spearman correlations were computed usingmeans for individual
reefs across intra-reef loca-tions for a given sampling period. For
epifaunaland transient finfish assemblages, we analyzeddata from
the Summer 2002 sampling period. Forthe resident finfish and
crustaceans, we used datafrom the Fall 2001 sampling period because
someof the later samples were lost during processing.We tested the
null hypothesis that these correla-tion coefficients did not differ
from zero using t-tests (SoKAL and ROHLF, 1981). All data sets
weretested for normality using the Kolmogorov-Smir-nov test (SAS
Institute Inc., 1990) and homosce-dasticity using Hartley's Fm_
Test (SOKAL andROHLF, 1981). Many data sets did not meet
theseassumptions, especially for equal variances.Therefore,
Spearman's rank correlations were uti-lized for comparisons. All
data analyses were car-ried out using SAS, except Hartley's F-Max
testswhich were computed manually according to So-KAL and ROHLF
(1981). A report (LuCKENBACH andRoss, 2003) and a subsequent
manuscript have orwill address variations in relation to reef size
andlocation within the reef.
Inlet Creek, South Carolina
Study Site and Background
The South Carolina studies were conducted inInlet Creek, a
tributary adjacent to CharlestonHarbor, as part of a larger study
examining thedevelopment of intertidal oyster reefs in relationto
reef age and season at sites with differing ad-jacent development
(COEN et at, 1999b). We alsocompared these intensively studied
sites to nu-merous sampling sites from across the state. Forthe
purpose of this overview, we focus only on thethree experimental
reefs constructed in InletCreek (N 320 47.93', W 790 49.73') and
the threeadjacent, natural reef areas (see Figure 3). The de-sign
and construction of these reefs, as well as pre-liminary physical
descriptions of the site, havebeen reported in detail in WENNER et
al, 1996;COEN et al, 1999b; COEN and LUCKENBACH, 2000,so we present
only a brief overview here. In Oc-tober 1994, three replicate 24-M2
experimentalreefs were constructed on an intertidal oyster
flatwithin Inlet Creek. Each experimental reef wascomprised of 156
plastic trays (0.46 m x 0.31 m x0.11 m) filled with clean oyster
shell and arrangedin a 6 x 26 array, with each reef paired with
anadjacent natural reef of similar size (see Figure 2in COEN et
al., 1999b). Here we discuss the results
for Inlet Creek only, the more undeveloped of thetwo sites at
the time of study. Oyster samplingbegan on experimental reefs after
initial recruit-ment in the late spring to early summer
1995,whereas resident sampling began earlier in March1995 (COEN et
al, 1999a, 1999b; COEN and LucK-ENBACH, 2000). Natural oyster
sampling in InletCreek started in 1997. Additional work
includedtransients, disease, and environmental samplingduring the
overall study from 1995-2001 (seeCOEN et alt 1999a, 1999b). However
for this paper,only a portion of the overall dataset was used.
Sampling Methods
Resident fauna (defined here as those organismsremaining within
the shell matrix when exposedat low tide) on the experimental reefs
were sam-pled by removing three randomly selected "quad-rats" (=
trays) sampled only once from each of thethree experimental reefs.
We rinsed the materialon a 0.5-mm sieve and retained all
organismscaught on the sieve. All oyster shell was thorough-ly
sorted and all live oysters counted and mea-sured. For the natural
oyster residents, we sam-pled using quadrates randomly placed on
adjacentoyster reefs outside of the paired natural reef tominimize
disturbance of other sampling. All organ-isms, including oysters,
were excavated to a depthof 11 cm and removed. Macrofauna were
enumer-ated to the lowest practical taxon for the first fouryears.
Thereafter, only decapod crabs and musselswere identified and
counted due to logistical con-straints. Macrofaunal biomass was
quantified us-ing wet weights (see COEN et al., 1999b) for
specifictaxonomic groupings (i.e. "Decapod Crabs" and"Shrimp,"
"Amphipods," "Isopods," "Polychaetes,"and "Mussels") throughout the
entire project du-ration. Sampling for reef residents began in
March1995, five months after the experimental reefswere constructed
and prior to any recruiting oys-ters, with the reefs sampled
bimonthly during thefirst year and quarterly during the second
year.The more frequent sampling allowed for better ini-tial
resolution as reefs began to receive both oysterand resident
recruits. Over the next three years,1997-1999, sampling was reduced
to summer andwinter samples, collected from the experimentaland
natural reefs as described above. For 2000-2001, resident samples
were collected only duringthe winter due to logistical (primarily
funding andmanpower) constraints over time (see above also).
Journal of Coastal Research, Special Issue No. 40, 2005
69
-
70 Luckenbach et al.
soE 40
g30 D.-Insgs G.-R.f201 A. , ?2k 1 M A20 10 .o020
I 10 19 2a 37 45 55S 1asw1oJ.1
, 40
& 20 Xla 0- -0.0,10 %120253t.r
I 10 1a 2a 37 40 059sawdo'5)
00
&;3 l 920 1 "?72,,2s7
I IC ,, 00 * , 6 5 1 10 1a 20 37 40 00
LFigure 4. Oyster abundance and size distribution at thefour
reef sites in the Rappahannock River, Chesapeake Bay.Data are from
samples taken during summer 2002 and arepooled across all reefs
(sizes and intra-reef locations) withina site. For total abundance,
Drumming Ground = TempleBay = Parrofs Rock > Mill Creek (Tukey's
StudentizedRange Test, P - 0.05).
0, ,52
00,2001 7.12001 2002
ao
1'2 7. C .0 0 0 0 2 . 1 0 .M..
t3m 1-i__
00 001 ,N201 !;- =
O i n o or O O 7. 0 2 2 5 0 0 , m o 0 2 0 0
10
14
S - M0 0 1002000 025 0 00
Statistical Analyses
Two-way ANOVAs (PCSAS 8.2) were used totest for the effects of
time (categorical dependingon particular sampling frequency) and
reef type onoyster abundance and mean shell height. All
as-sumptions were tested prior to statistical analyses.Mean
abundances and oyster size frequencies fromthe experimental and
natural reefs are presentedgraphically for the January sampling
period onlyfor 1997 through 2001 for simplicity here. Al-though we
sampled and measured oysters from ex-perimental reefs from 1995, we
did not begin tomeasure adjacent natural oyster populations
until1997, whereas resident sampling began soon afterthe reefs were
constructed in March 1995. Sincewe are using both for comparison,
we show the1997-2001 data ouly. We computed Pearson prod-uct-moment
correlation coefficients (SigmaStat2.0) relating mean oyster
abundance and meanshell height to the abundances and biomass
ofdominant reef-associated species and to various"community
descriptors" for the experimental andnatural reefs. Data from
January 1998 were se-lected for this analysis because that
representedthe latest time period for which complete data
onepifaunal abundance and diversity were availablefor the
study.
RESULTS
Rappahannock River, Virginia
Two years after construction, reefs at the foursites in the
lower Rappahannock River differed
2foWK M1= II210=
Figure 5. Temporal patterns of (A) oyster abundance, (B)oyster
biomass, (C) epifaunal abundance, (D) epifaunal di-versity, (E)
Geukensia demissa abundance, (F) Balanus spp.abundance, and (G)
xanthid crab abundance on the reefs inthe Rappahannock River,
Chesapeake Bay, Virginia. Valuesare means ± SE by reef site.
both in abundance and size structure of oysterpopulations
(Figure 4). Total oyster density variedsignificantly between sites
(F = 5.82, df = 3, P =0.0018), with Parrot's Rock, Drumming
Ground,and Temple Bay reefs all having greater meandensities than
Mill Creek reef (see Figure 4). How-ever, no significant
differences were observed be-tween individual reefs independent of
geographic"location" (F = 1.94, df = 9, P = 0.0731). Reefsalso
differed in densities of 1-2 year class oysters(i.e. those with SH
Ž 20 mm), which are shown inthe size distribution plots in Figure
4. Prior to re-cruitment in the summer of 2001, oyster abun-dances
at all of the reef sites were zero (Figure5A). Mean oyster
abundances peaked in fall 2001at approximately 350 oyster/M2
following recruit-ment during summer 2001 and fell slightly
bysummer 2002. Biomass of oysters on the reefs in-creased
throughout this period (Figure 5B), aswould be expected with newly
recruiting oysterpopulations on these reefs.
Epifaunal abundances were dominated by bar-
Journal of Coastal Research, Special Issue No. 40, 2005
Ik30'), ,
5 0 .OpO.0.
000 ,,0,5rZ3
10s
70 Luckenbach et al.
-
Evaluating Success Criteria and Reef Development for Oyster
Restoration
nacles in the genus Balanus, especially during thefirst summer
when densities exceeded 14,000/M2
(Figure 5F). Exclusive of barnacles, epifaunalabundances changed
little over the course of thestudy (Figure 5C). Overall diversity
of epifauna in-creased throughout the time period (Figure 5D),both
as a result of an increase in species richnessand a decline in
barnacle densities (Figure 5F).Prominent members of the epifaunal
assemblage,in addition to barnacles, included bivalves (Maco-ma
balthica, Mulinia lateralis, Mya arenaria, Geu-kensia demissa
(Figure 5E), Mytilus edulis, and Pe-tricola pholadiformis),
solitary and colonial tuni-cates, an ectoproct (Membranipora
tenuis), a ser-pulid polychaete (Hydroides dianthus), andxanthid
crabs (Figure 5G). In addition, the reefssupported seasonally
abundant and diverse as-semblages of macroalgae.
Within a single season, correlation coefficientsbetween oysters
(mean total abundance, meanabundance of year class 2 oysters, and
biomassacross reef sites) and various community descrip-tors and
dominant species varied considerably (Ta-ble 1). The only
significant correlation we foundwith total oyster abundance was
with the abun-dance of skillet fish, which showed a very
strongpositive correlation, and the total abundance ofresident
finfish, which showed a positive correla-tion. Similar significant
positive correlations wereobserved with the abundance of year class
2 oys-ters. In addition to these two parameters, barnacleabundance,
ribbed mussel biomass, and epifaunaldiversity were also
significantly correlated withoyster biomass, with total epifaunal
abundanceonly marginally insignificant (P = 0.054, see Table1).
Interestingly, the only negative correlation ob-served was between
oyster biomass and epifaunaldiversity. It is important to note here
that while alarge number of samples and individual organismswere
part of these analyses, the correlations wereconducted using means
for each individual reefand thus the tests for significance, with
only 8 de-grees of freedom, had relatively low power (HOENIGand
HEIsEY, 2001).
Inlet Creek, South Carolina
Overall during the study, we collected over 87resident and 60
transient species associated withour reefs (COEN et al., 1999a).
Oyster abundanceon the experimental reefs at Inlet Creek
increasedduring the period from January 1997 tbrough Jan-uary 2001,
but means (±SE) remained well below
an
nuk
r�
H
� aon0 .�acoa-u00��
0
-0
a a
'-a
an72C.
S.H-C) 0oo cq0
.0- a
.�su0
SaC.
)oau0
.2 aII
0.ac o ao o .�0 H
) aa -�
u
a 0
0000'� u�
-� -gii
Co. �
a �Z a15
.0
0
H
,I.0
H0~0
.0
~20
a
0i
I0)
co
ria0
H
.0
S t'Cs
AE
H
aIE-I 3
C.0
I5
H
10
.~d .2
a)
C.
+ 0 c Cotf CD 1:,-)LO co C 'CDC qo OC .- Co66 0 ,0 00 C
0 CoCo~ Co
Io Co00
c-nc OCos CD CO
Co OCo00 Co L66t 66Cc
CD CD Co C C C66 0q 00> uD o
CD X C tO
CT° CT°66 00
CD tbCD 1140 Co
66
D CO CoO CD sCoCo CoD CDok
I 0 I Co
Co Hncs cq
CD O,00
CoCO 0
V
cq cs to-9X o C oD C
v
C-c CD CoCo o
CoH c oCU - c-o1oo CD C) co r ellCo0 O OC CoH
CD CD CD CD CD O
1 1
O CD co 1 ) CC
6 6 C66 6 6
CO Co100
Co CoI
m c-001n-i r-~
I
c0 10 i +
'tN CoO ,
Co O C1 0
0) 0: CD OCoO CO
Ci 1- ! C i U OOCo C CH 1= C0 0 D
CqI
CoC-- Cd~- OC eo:
oo o Coo rco
cO oo O r
0 C C) I DC
0-l C c-c
Journal of Coastal Research, Special Issue No. 40, 2005
71
.0
2-
II
-
Luckenbach et al.
2I8�
'i
I
0-
IU i J~~~~~~~~~~~anuay 1997I iRiL n=~~~~~M 247
12D
15
12 *aw.. o l i r r2i31~F 19
12 NLni11a2 0
B
23
9 1lli111titl lb i7ligllWl37Jl r . .23
Is Jna2012 n= 2o- 0 1 1'1
a1 EI I L I4A.EEIHIEIEI1L 0 ,..
12 R - I N=E
0 10 20 30 40 E0 E0 70 E0 50 100 110 123 123 140
Shell length (mm)
Figure 6. Oyster size (abundance)-frequency distributionsfor
natural and experinental population samples collectedfrom Inlet
Creek, Charleston Harbor, South Carolina, resi-dent samples (n = 9
for each date x reef type). The verticallines of each plot
represent overall mean size for each reeftype; solid is for oysters
on the experimental reefs, dashedrepresents oysters on natural
reefs. (Note: Values overlap forJanuary 2000.)
0. 00.2 202.200200000400 0 26
1200007 ai
0003 C0 000n .d.602w 0660.60
a-0:
IMI0200 w=D 0
s60 c. 2060000202202024
rc 07 00I SO 3
s -D
0234 60602202200026,2.0 l
2006 2007 200- 0000 =S0 0002
Figure 7. Temporal patterns for (A) mean oyster abun-dance, (B)
mean oyster height, (C) mean epifaunal abun-dance, (D) epifaunal
diversity, (E) mean Geukensia demissaabundance, (F) mean xanthid
crab abundance, (G) meanEu-rypanopeus depressus abundance, and (H)
Panopeus herbstiiabundance for experimental (dashed lines) and
natural (solidlines) reefs in Inlet Creek, Charleston Harbor, South
Caro-lina. All values are means ± SE by reef type (experimentalvs.
natural) except for D, epifaunal diversity.
(e.g., January 2001, mean no. 497 ± 282) densitiesfound on
adjacent natural reefs at Inlet (January2000 and 2001, means from
861-1646/M2 ) or anyof the other sites we have sampled across the
state(Figures 6 and 7A). For comparison, mean densi-ties across
South Carolina ranged from a low of500/M2 (±88) to over 6,436/in2
(±500), for samplescollected by us from 1997 to 2002. A two-way
AN-OVA revealed significant effects for reef type (F =201.80, P
< 0.0001), time of sampling (F = 3.68,P = 0.0022), and the
interaction term (F = 3.37,F = 0.0043) for oyster abundance. Oyster
size fre-quency distributions were similar on the experi-mental and
natural reefs, with natural reefs at In-let having more oysters
above 75 mm SH (Figures6 and 7B), resulting in marginally
insignificantdifferences (P = 0.0581) in mean SH of oysters
be-tween the reef types.
South Carolina has no harvest size limit, sothese differences
are not relevant for the resource.Prominent members of the resident
faunal assem-blage (numerically) included polychaetes
(Nereissuccinea and Streblospio benedicti), several mus-
sels (Brachidontes exustus and Geukensia demissa),a gastropod
(Creedonia succinea), mites, and a per-acarid (Gammarus palustris).
In addition to theabove taxa, natural reefs also had large
numbersof the gastropod ectoparasite Boonea impressa andthe xanthid
crab Eurypanopeus depressus. Epifau-nal abundance and diversity
measures are onlyavailable for this site for the years
1996-1998;during January 1997, total epifaunal abundanceon the
experimental reefs was similar to that onthe natural reefs because
of high abundances ofgastropods (primarily C. succinea), mussels
(B. ex-ustus and G. demissa), and acarinids (mites), butdiversity
was lower (Figures 7C and D).
Ribbed mussel and xanthid crab (total of 14spp.) abundances
showed similar patterns, al-though densities on the natural reefs
exceededthose on the experimental reefs until 2001 (Fig-ures 7E and
F). Initially high abundances of thexanthid crab Eurypanopeus
depressus on the nat-ural reefs declined between 1998 and 1999;
abun-dances remained comparatively low during the
Journal of Coastal Research, Special Issue No. 40, 2005
72
-
Evaluating Success Criteria and Reef Development for Oyster
Restoration
Table 2. Correlations between oysters (abundance and height),
dominant taxa, and community metrics for reefs in Inlet Creek,South
Carolina, January 1998. S = Species Richness, H' = Shannon-TWeiner
Diversity Index, r = Pearson product momentcoefficients, p =
Probability of r = 0. Sample size, n = 9 per site for each in A and
B.
Epifaunal Invertebrates Xanthid sp.
Oyster Mean # Total Xanthid G. demissa G. demissa E. depres- P.
herbstiiC. virginica species S H' Total Abun. Biomass Abun. Abun.
Biomass sus Abun. Abun.
A. Experinental Reefs
Abundancer 0.103 0.513 0.055 -0.118 0.761 0.775 -0.218 0.706
0.605p 17.67 0.791 0.158 0.888 0.762 0.017 0.014 0.574 0.034
0.085
Mean Heightr 0.636 0.070 0.832 0.599 0.374 0.285 0.542 -0.046
0.722p 0.065 0.858 0.005 0.089 0.321 0.457 0.132 0.906 0.028
B. Adjacent Natural ReefsAbundance
r -0.036 0.225 0.661 0.105 0.495 0.745 0.092 0.475 -0.183p 18.78
0.927 0.560 0.053 0.788 0.175 0.021 0.814 0.196 0.637
Mean Heightr 0.203 -0.169 -0.084 -0.285 0.218 -0.144 -0.309
-0.021 0.608p 0.600 0.664 0.829 0.457 0.572 0.712 0.418 0.958
0.083
study on the experimental reefs (Figure 7G). Incontrast, the
xanthid crab Panopeus herbstii abun-dances varied in a similar
manner on both reeftypes throughout the study (Figure 7H).
Correlation coefficients between oysters (meanabundance and
shell height) and community met-rics and dominant species revealed
several signif-icant positive relationships on the experimentaland
natural reefs (Table 2). For the experimentalreefs, the abundance
of the ribbed mussel Geuken-sia demissa and the xanthid crab
Eurypanopeusdepressus varied positively with total oyster
abun-dance, while total numbers of epifaunal inverte-brates and the
abundance of the xanthid crab Pan-opeus herbstii were positively
correlated with oys-ter height (Table 2A). On the adjacent natural
oys-ter reefs (of unknown age), the only significantcorrelations
were between overall oyster abun-dance versus total epifaunal
densities or overallmussel abundance (Table 2B). No significant
neg-ative correlations were observed between eitheroyster abundance
or size and any of the other var-iables.
DISCUSSION
Understanding the relationship between the de-velopment of
oyster populations and other reef-as-sociated organisms is a key
element in evaluatingthe ecological success of oyster reef
restoration ef-forts. The two studies outlined here were each
de-
signed with different specific goals in mind; nev-ertheless,
they provide an opportunity to examineseveral aspects of community
development in re-lation to oyster populations on reefs from very
dif-ferent systems. The Rappahannock reefs are rel-atively large,
subtidal mounds extending severalmeters above the seabed. In
contrast, the SouthCarolina reefs, both natural and experimental,
arerelatively small by comparison, located entirelywithin the
intertidal zone, and generally extend10-30 cm above the upper
sediment surface; al-though oysters extend 1-3 m or more from the
lowintertidal to upper intertidal as reefs. The reefs inthe two
systems are also at very different stagesof development, with less
than two years since con-struction for the Rappahannock River reefs
com-pared to from six to seven years for the South Car-olina
intertidal reefs. Differences in reef morphol-ogy, experimental
design, and sampling tech-niques make direct statistical
comparisons of datafrom the two systems inappropriate.
However,consideration of patterns within each systemshould make the
general conclusions informativewithin a region or reef type (i.e.
subtidal or inter-tidal).
Central to our objective here is to ask whetheror not successful
ecological restoration of oysterreefs is dependent upon various
oyster populationattributes such as "abundance" or "size" (as
mea-sured here by shell height or indirectly as bio-
Journal of Coastal Research, Special Issue No. 40, 2005
73
-
Luckenbach et al.
mass). The reefs at the Rappahannock River sitewere still too
young to assess patterns in relationto market-sized oysters, but
they do allow us toexamine the relationship between the early
devel-opment of oyster populations and reef-associatedorganisms.
The reef bases in the RappahannockRiver were constructed in August
2000 after nat-ural oyster recruitment had occurred. Samplestaken
during the early summer of 2001, prior tothe peak of oyster
recruitment in the region, foundno oysters on the reef substrate.
By the summer2002, two age classes of oysters were evident onall of
the reefs at densities ranging from 77 to 277oysters/m2 . Even at
the highest of these densities,oysters do not monopolize the space
on the originalsubstrate material and are considerably less
abun-dant than on natural and other restored reefs fromthe
Chesapeake Bay (O'BEIRN et al., 2000). Nu-merous epifauna,
especially barnacles, recruited inlarge numbers to all of the reefs
prior to any oysterrecruitment occurring in 2001 (Figure 5).
Epifau-nal abundances (exclusive of barnacles) increasedonly
slightly over time, as oyster abundance andbiomass increased, while
barnacle abundances de-clined with time and presumably reef
develop-ment.
Though not presented in the Results section, theabundances of
transient finfish averaged across allof the reefs over time
revealed a strong seasonalpattern, but no clear inter-annual
pattern thatcould be related to oyster abundance or biomass.This
study did not include any natural "control"reefs for comparison, so
we are unable to relate thevarious descriptors of the reef
assemblages to "nat-ural reefs" and must rely on comparisons with
oys-ter abundance and "size" (most common measure-ment is shell
height or calculated biomass) acrossexperimental reefs. This is in
large part due to thefact that there are few or no healthy reefs
for com-parison as there are in the South Carolina study.
In contrast to the observation for Virginia thatlarval supply
may play a significant role in resto-ration success without the
significant addition of"seed" oysters to jump start reef oyster
popula-tions, South Carolina restoration success appearsto be
simply the result of the addition of the lim-iting substrate,
oyster shell. The experimentalreefs in Inlet Creek, South Carolina,
did have ex-tensive natural reefs for comparison, and over thetime
period from 1995 to 2000 they failed to con-verge with the natural
reefs, as measured by ei-ther total oyster density or abundance of
market-sized >75 mm oysters observed on the natural
reefs. Using our historical statewide South Caro-lina data,
oysters >75 mm typically make up lessthan 10% of all reef
oysters, with a maximum of18% at only two of the sites to date.
Also, althoughthe filtering capacity of a mature oyster reef maynot
have been reached due to low oyster initialdensities, mussels
recruited to reefs in large num-bers (as high as 1,500/M2 ),
potentially providingpreviously unrealized benefits, not noted for
oysterreefs before (COEN et al, 1999). Nesting sites forresident
fish are also critical, as is a complexthree-dimensional structure
for the associateddecapod crabs (GRANT and McDONALD,
1979;BREITBURG, 1999; COEN et al., 1999; MEYER andTOWNSEND, 2000;
GRABOWKSI, 2002; GLANcY et al,2003). Although oysters are not
necessary to es-tablish some of these ecological benefits,
sustain-ability over time and augmentation of these ben-efits
(e.g., increased habitat) does require estab-lishment of oyster
populations.
Total epifaunal abundance and epifaunal diver-sity (available
only for the period from 1996-1998)did not show temporal patterns
(with increasingage) related to either oyster abundance (=
density)or "size" (shell height or biomass) over the sametime
period, but epifaunal abundance on the ex-perimental reefs did
approach that found on thenatural reefs in January 1997, largely as
a resultof gastropods (2 spp.) and mussels (2 spp.) recruit-ing in
large numbers to the experimental reefs.Temporal patterns of
abundance were similar forsome species on the experimental and
naturalreefs (e.g., Panopeus herbstii) and different for oth-ers
(e.g., Geukensia demissa), though the latter didhave similar
abundances during 2001 when oysterabundances on the two reef types
began to con-verge.
We used correlation rather than regression inanalyzing
relationships between oysters and vari-ous components of the reef
assemblage, becausewe were lacking specific information
regardingcause and effect relationships among the assem-blage
entities and the observed oyster populations.We can hypothesize
that positive relationshipsmight be associated with: (1) increased
habitatheterogeneity, (2) the provision of refuges, (3)availability
of nesting sites for resident fishes(BREITBURG, 1999), and (4)
enhanced benthic-pe-lagic coupling. Conversely, competitive
interac-tions for (1) space and (2) food, as well as (3) ex-clusion
of some species from refuges, could resultin negative relationships
between oysters and oth-er species. Alternatively, there may be no
direct
Journal of Coastal Research, Special Issue No. 40, 2005
74
-
Evaluating Success Criteria and Reef Development for Oyster
Restoration
causal relationship, and the parameters may co-vary in relation
to some other factor (e.g., local wa-ter quality, larval supply, or
food availability). Itis informative that of the 15 significant
(and 5marginally insignificant) correlations out of a totalof 78
examined (or 19% significant) that we ob-served between oysters and
the evaluated com-munity descriptors or dominant species acrossboth
studies and various reef types, all but onewere positive, the
exception being epifaunal inver-tebrate diversity in relation to
oyster biomass onthe Rappahannock reefs. For the Virginia
subtidalreefs, the most consistent pattern observed was apositive
relationship between resident finfish (to-tal abundance and G.
strumosus) and all measuresof oyster density (total abundance, year
class 2abundance, and biomass). For the South Carolinastudy, three
of the seven (or 43%) sigruficant in-vertebrate correlations (Table
2A) observed for theexperimental reefs were with total oyster
density,while only one of the seven was significant for nat-ural
reefs (Table 2B). For mean height, only twoof the seven and none of
the seven were significantfor the experimental and natural reefs,
respective-ly (Table 2). For the experimental reefs where oys-ter
density is gradually increasing, key communitymetrics such as total
xanthid crabs and the meanabundance of either the mussel G. demissa
or thexanthid crab E. depressus are potentially valuableindicators
of reef progress.
Abundances for some species varied differentlywith oyster
abundance or size/biomass dependingon whether they were viewed over
temporal orspatial scales. For instance, on the reefs in
theRappahannock, the abundance of Balanus spp. de-clines sharply
between the summers of 2001 and2002, during which time mean oyster
biomass in-creases from 0 to 18 g/m2 (Figures 5B and F);however,
when we examine the relationship be-tween oyster biomass and
barnacle abundanceduring that last sampling period, we observe a
sig-nificant positive relationship (Table 1). Similarly,on both the
experimental and natural reefs inSouth Carolina, the abundance of
Panopeus herbs-tii is strongly inversely related to mean oyster
size(as directly measured here) between 1996 and2001 (compare
Figures 7B and H). Yet, when weexamine the relationship between
oyster shellheight and P herbstii abundance at a single sam-pling
period (January 1998), we find a significantpositive correlation on
the experimental reefs (Ta-ble 2A) and a similar, though marginally
insignif-icant, pattern on the natural reefs (Table 2B). Po-
tential explanations for these discrepancies be-tween temporal
and spatial patterns include: (1)that in either one or both cases
that the organismsin question and oysters co-vary in response
tosome other factor(s), and (2) that the magnitude oftemporal
variations in oyster abundance or "size"(shell height or biomass)
over the course of thesestudies exceeds those observed at any one
timeacross treatment replicates and thus represents amore robust
test of the effects of oysters.
A correlational approach alone will not suffice totruly evaluate
the relationships between oysterpopulations and the ecological
fuinctions of re-stored oyster reefs. We still need to develop a
bet-ter understanding of specific interactions betweenspecies. For
instance, we did not observe a consis-tent relationship between
either the temporal orspatial patterns of oysters and xanthid crabs
ateither study site, despite the fact that xanthidcrabs are an
important predator on small oysters,while also providing a refugia
for the same crabsand other resident finfish (COEN et al., 1999a;
GRA-BowsKi, 2002, in press). Direct and indirect effectsof
predator-prey interactions among reef-associat-ed organisms along
with their relationship to oys-ter population structure, need to be
clarified. Fur-ther, the consequences of the competing roles
thatoysters play in facilitating the establishment ofsome species
by providing hard substrate, and incompeting with many of those
same species forspace and food, are not well understood. Many
res-ident fish require large, clean, and dead articulat-ed shells
for complex life histories (BREITBURG,1999; COEN et al, 1999a).
Additionally, we suspectthat there are numerous aspects of reef
morphol-ogy, location within respect to tidal range, and po-sition
in the landscape affecting the developmentof reef communities that
have yet to be clarified(GRABOWSKI, 2002, in press).
As PALMER et al. (1997) point out, choosing res-toration
endpoints is a crucial and often difficulttask in ecological
restoration. Habitat restorationsuccess should not be dependent
solely on thegrowth/survival of the restored species (CRAFT etal,
1999). A focus only on the resource (eg., har-vestable oysters,
fishing mortality) will miss pos-sibly critically important
measures of reef resto-ration success (e.g., benthic pelagic
coupling, hy-drodynamic effects). In a similar vein, we need
tobetter understand how feedbacks work in healthyand degraded
systems and whether their restora-tion results in alternative
states not predicted
Journal of Coastal Research, Special Issue No. 40, 2005
75
-
76 Luckenbach et
al.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
from past observations and recent work (SUDING,2003).
Declines in Crassostrea virginica abundancesthroughout much of
the U.S. Atlantic coast havehad important fisheries and ecological
implica-tions (NEWELL, 1988; KAUFMAN and DAYTON, 1997;PETERSON and
LUBCHENCO, 1997; COEN et al,1999a; JACKSON et al., 2001; PETERSON
et al, 2003).Fisheries restoration is, undoubtedly, a
desirablerestoration endpoint and the explicit goal of nu-merous
restoration efforts. However, it is restora-tion of lost ecological
functions provided by oysterreefs that has been the focus of most
recent efforts.While our results reveal positive correlations
be-tween the diversity and abundance of reef-associ-ated species
and the abundance and direct or in-direct measure of oyster "size"
(shell height or bio-mass), they do not indicate that market-sized
oys-ters are requisite for supporting an abundant anddiverse
community. Although South Carolina ex-perimental reefs have not
converged with the nat-ural reefs, even after six years, using
numbers ofoysters or some measure of vertical complexity ofoyster
clusters versus the natural reefs, they arepersisting with slowly
increasing oyster popula-tions, and they support a diverse
assemblage ofresident organisms. Similarly, the Rappahannockreefs,
at the time of final sampling, were only twoyears old and did not
yet support any market-sizedoysters. However, they did support
resident andtransient community assemblages, many of whichwere
positively correlated with the abundance andsize (shell height or
biomass) of oysters on thereefs. Until we develop a more thorough
under-standing of the individual species interactions andmechanisms
linking oyster population structure tothe composition and diversity
of reef communities,we suggest that oyster abundance and some
mea-sure of "size" (age) structure provides a quantita-tive measure
of restoration success, but harvest-able quantities of market-sized
oysters are not re-quired for achieving some level of ecological
res-toration.
In the future, we need to develop and evaluaterestoration
progress by using standard criteriathat can be applied to projects
or programs beingconducted over a wide geographic range. In
somecases, it may be easier and more cost effective tomeasure
surrogate or indirect benefits (e.g., filter-ing, habitat use) than
to focus on the oyster pop-ulations alone. For example, seston
uptake mightbe able to estimate the total effect of the oysterreef
(constructed or natural) community on water
quality by quantifying the amount of the water col-umn that is
cleared of seston (GRIZZLE and LuTz,1989). This is one of the major
"ecosystem servic-es" often touted in the oyster restoration
litera-ture, but rarely quantified.
ACKNOWLEDGMENTS
We wish to thank F. Holland, D. Bushek, R.Mann, and two
anonymous reviewers, among oth-ers, for significant contributions
throughout thework. The Virginia Rappahannock work (MWLand PGR) was
supported by grants from the Vir-ginia Graduate Sea Grant
Consortium and theVirginia Oyster Heritage Program. We are
espe-cially grateful to J. Wesson for constructing thereefs and to
G. Arnold, A. Birch, S. Bonniwell, J.Nestlerode, B. Parks, L.
Sorabella and S. Spearsfor assistance in the field and lab. This is
Contri-bution #2598 from the Virginia Institute of MarineScience,
College of William and Mary.
For the South Carolina component (LDC andJAS), we cannot list
all of the many individualswho were involved in the development,
collection,and analyses of the data from the Oyster Re-search
Program (ORP). However, we are' espe-cially grateful to R. Giotta,
D. Knott, B. Stender,E. Wenner, W. Post. W. Hegler, K.
Hammer-strom, B. Conte, J. McAlister, and M. Bolton-Warberg for lab
and field assistance. The SouthCarolina portion of this research
was funded bygrants from the National Oceanic and Atmo-spheric
Administration (NOAA) South CarolinaSea Grant Consortium
(#NA86RG0052), theSouth Carolina Marine Recreational FisheriesStamp
Program, and the South Carolina Depart-ment of Natural Resources
through its MarineResources Research Institute. This is
Contribu-tion #541 from the Marine Resources ResearchInstitute,
SCDNR.
LITERATURE CITED
BREITBURG, D., 1999. Are three dimensional structuresand healthy
oyster populations the keys to an ecolog-ically interesting and
important fish community? In:LUCKENBACH, M.W., MANN, R. and WESSON,
J.A. (eds.),Oyster Reef Habitat Restoration: A Synopsis and
Syn-thesis of Approaches. Gloucester Point, Virginia: Vir-ginia
Institute of Marine Science Press, pp. 239-250.
BREITBURG, D.L.; COEN, L.D.; LUCKENBAcH, M.W.; MANN,R.; PosEy,
M. and WEssON, J.A., 2000. Oyster ReefRestoration: Convergence of
Harvest and Conserva-tion Strategies. Journal of Shellfish
Research, 19, 371-378.
COEN, L.D. and STEPHEN, J.A., 2002. An examination of
Journal of Coastal Research, Special Issue No. 40, 2005
76 Luckenbach et aL.
-
Evaluating Success Criteria and Reef Development for Oyster
Restoration
oyster reefs as biologically-critical estuarine ecosys-tems,
South Carolina Sea Grant: Annual Report, 1
6 p.COEN, L.D. and LUCKENBACH, M.W., 2000. Developing
success criteria and goals for evaluating shellfish hab-itat
restoration: ecological function or resource exploi-tation?
Ecological Engineering, 15, 323-343.
COEN, L.D.; LUCKENBACH, M.W. and BREITBURG, D.L.,1999a. The role
of oyster reefs as essential fish habi-tat: a review of current
knowledge and some new per-spectives. In: BENAKA, L.R. (ed.), Fish
Habitat: Essen-tial Fish HabitatAnd Rehabilitation. Bethesda,
Mary-land: American Fisheries Society, Symposium 22,
pp.438-454.
COFN, L.D.; KNOTT, D.M.; WENNER, E.L.; HADLEY, N.H.;RINGwoOD,
A.M. and BOBO, M.Y., 1999b. South Caro-lina intertidal oyster reef
studies: design, sampling,and focus for evaluating habitat value
and function.In: LUCKENBACH, M.W.; MANN, R. and WESSON, J.A.(eds.),
Oyster Reef Habitat Restoration:A Synopsis andSynthesis of
Approaches. Gloucester Point, Virginia:Virginia Institute of Marine
Science Press, pp. 133-158.
CRAFT, C.; READER, J.; SAcco, J.N. and BROOME, S.W.,1999.
Twenty-five years of ecosystem development ofconstructed Spartina
alterniflora (Loisel) marshes.Ecological Applications, 9,
1405-1419.
DAME, R.F., 1999. Oyster reefs as components in estua-rine
nutrient cycling: incidental or controlling? In:LUCKENBACH, M.W.;
MANN, R. and WESSON, J.A. (eds.),Oyster Reef Habitat Restoration: A
Synopsis and Syn-thesis of Approaches. Gloucester Point, Virginia:
Vir-ginia Institute of Marine Science Press, pp. 267-280.
DAM,E, R.F.; ZINGMARK, R.G. and HAsKINs, E., 1984. Oys-ter reefs
as processors of estuarine materials. Journalof Experimental Marine
Biology and Ecology, 164,147-159.
DAME, R.F. and LIBES, S., 1993. Oyster reefs and nutri-ent
retention in tidal creeks. Journal of ExperimentalMarine Biology
and Ecology, 171, 251-258.
DAME, D.; BUSHEK, D.; ALLEN, D.; EDWARDS, D.; GREGORY,L.;
LEWITUS, A.; CRAwFORD, S.,; KOEPFLER, E.; CORBETT,C.; KJERVE, B.
and PRINS, T., 2000. The experimentalanalysis of tidal creeks
dominated by oyster reefs: thepre-manipulation year. Journal of
Shellfish Research,19, 361-370.
DAME, R.F.; BUSHEK, D. and PRINS, T., 2001. The role
ofsuspension feeders as ecosystem transformers in shal-low coastal
environments. In: REISE, K. (ed.), TheEcology of Sedimentary
Coasts. Berlin: Springer-Ver-lag, pp. 11-37.
EGGLESTON, D.B.; ELIS, W.E.; ETHERINGTON, L.L.; DAHL-GREN, C.P.
and PosEy, M.H., 1999. Organism respons-es to habitat fragmentation
and diversity: habitat col-onization by estuarine macrofauna.
Journal of Exper-imental Marine Biology and Ecology, 236,
107-132.
GLANCY, T.P.; FRAZIER, T.K.; CiCHRA, C.E. and LINDBERG,W.J.,
2003. Comparative patterns of occupancy bydecapod crustaceans in
seagrass, oysters, and marshedge habitats in a Northeast Gulf of
Mexico estuary.Estuaries, 26, 1291-1301.
GRABOwsKI, J.H., 2002. The influence of trophic inter-actions,
habitat complexity, and landscape setting oncommunity dynamics and
restoration of oyster reefs.
Chapel Hill, North Carolina: The University of NorthCarolina,
Ph.D. thesis, 169 pp.
GRABowsKI, J.H., 2004. Habitat complexity disruptspredator-prey
interactions but not the trophic cascadein oyster-reef communities.
Ecology, in press.
GRANT, J. andMcDoNALD, J., 1979. Desiccation toleranceof
Eurypanopeus depressus (Smith) (Decapoda: Xan-thidae) and the
exploitation of macrohabitat. Estuar-ies, 2,172-177.
GRIZZLE, R.E. and LUTZ, R.A., 1989. A statistical modelrelating
horizontal seston fluxes and bottom sedimentcharacteristics to
growth of Mercenaria mercenaria.Marine Biology, 102, 95-105.
HARDING, J.M. and MANN, R., 1999. Fish species richnessin
relation to restored oyster reefs, Piankatank River,Virginia.
Bulletin Marine Science, 65, 289-300.
HARGIS, W.J., JR., 1999. The evolution of the Chesapeakeoyster
reef system during the Holocene epoch. In:LUCKENBACH, M.W., MANN,
R., and WEssoN, J.A. (eds.),Oyster Reef Habitat Restoration: A
Synopsis and Syn-thesis of Approaches. Gloucester Point, Virginia:
Vir-ginia Institute of Marine Science Press, pp. 5-24.
HOENIG, J.M. and HEISEY, D.M., 2001. The Abuse of Pow-er: The
Pervasive Fallacy of Power Calculations forData Analysis. The
American Statistician, 55, 1-6.
JACKSON, J.B.C.; KIRBY, M.X.; BERGER, W.H.; BJORNDAL,K-A. and
BOTSFORD, L.W., 2001. Historical overfishingand the recent collapse
of coastal ecosystems. Science,293, 629-638.
JONES, C.G.; LAWTON, J.H. and SHACHAK, M., 1994. Or-ganisms as
ecosystem engineers. Oikos, 69, 373-386.
KAUFMAN, L. and DAYTON, P.K., 1997. Impacts of marineresource
extraction on ecosystem services and sustain-ability. In: DAILY,
G.C., (ed.), Nature's Services: SocietalDependence on Natural
Ecosystems. Washington, D.C.:Island Press, pp. 275-293.
LEHNERT, R.L. and ALLEN, D.M., 2002. Nekton use ofsubtidal
oyster shell habitats in a Southeastern U.S.estuary. Estuaries, 25,
1015-1024.
LENIHAN, H.S., 1996. Physical-biological coupling on oys-ter
reefs: hydrodynamics, sedimentation, and the pro-duction of
oysters. Chapel Hill, North Carolina: TheUniversity of North
Carolina, Ph.D. thesis, 171 pp.
LENIHAN, H.S., 1999. Physical-biological coupling on oys-ter
reefs: how habitat structure influences individualperformance.
Ecological Monographs, 69, 251-275.
LENIHAN, H.S. and GRABowsKI, J.H., 1998. Recruitmentto and
Utilization of Oyster Reef Habitat by Fishes,Shrimps, and Blue
Crabs: An Experiment with Eco-nomic Analysis. Beaufort, North
Carolina: NationalMarine Fisheries Service, National Research
CouncilFinal Report No. 197.
LENIHAN, H.S.; MIcHELI, F.; SHELTON, S.W. and PETER-SON, C.H.,
1999. The influence of multiple environ-mental stressors on
susceptibility to parasites: an ex-perimental determination with
oysters. Limnologyand Oceanography, 44, 910-924.
LUCKENBACH, M.W.; MANN, R. and WESSON, J.A., (eds.),1999. Oyster
Reef Habitat Restoration: A Synopsis andSynthesis of Approaches.
Gloucester Point, Virginia:Virginia Institute of Marine Science
Press, 366p.
LUCKENBACH, M.W and Ross, P.G., 2003. An Experimen-tal
Evaluation of the Effects of Scale on Oyster Reef
Journal of Coastal Research, Special Issue No. 40, 2005
77
-
Luckenbach et al.
Restoration. Charlottesville, Virginia: Final Report toVirginia
Sea Grant, 120 pp.
MEYER, D.L. and TOWNSEND, E.C., 2000. Faunal utiliza-tion of
created intertidal oyster reef (Crassostrea vir-ginica) in the
southeastern United States. Estuaries,23, 34-45.
NWELL, R.IE., 1988. Ecological changes in ChesapeakeBay: are
they the result of overharvesting the Ameri-can oyster, Crassostrea
virginica? In: LYNCH, M.P. andKROME, E.C., (eds.). Understanding
the Estuary: Ad-vances in Chesapeake Bay Research. Gloucester
Point,Virginia: Chesapeake Research Consortium, Publica-tion 129
CBP/TRS 24/88, pp. 536-546.
O'BEiRN, F.X.; LUCKENBACH, M.W.; NESTLERODE, J.A. andCoATEs,
G.M., 2000. Toward design criteria in con-structed oyster reefs:
oyster recruitment as a functionof substrate type and tidal height.
Journal of ShellfishResearch, 19, 387-396.
PALMER, M.A.; AMBROSE, R.F. and POFF, N.L., 1997. Eco-logical
theory and community restoration ecology. Res-toration Ecology, 5,
291-300.
PETERSON, C.H.; GRABoWSKI, J.H. and POWERS, S.P.,2003. Estimated
enhancement of fish production re-sulting from restoring oyster
reef habitat: quantitativevaluation. Marine Ecology Progress
Series, 264, 251-266.
PETERSON, C.H. and LUBCHENCO, J., 1997. Marine eco-system
services. In: DAiLY, G.C. (ed.), Nature's Servic-
es: Societal Dependence On Natural Ecosystems. Wash-ington,
D.C.: Island Press, pp. 177-194.
POSEY, M.H.; ALPFHN, T.D.; POWELL, C.M. and TOWNSEND,E., 1999.
Use of oyster reefs as habitat for epibenthicfish and decapods. In:
LUCKENBACH, M.W.; MANN, R.and WESSON, J.A. (eds.), Oyster Reef
Habitat Restora-tion: A Synopsis and Synthesis of Approaches.
Glouces-ter Point, Virginia: Virginia Institute of Marine Sci-ence
Press, pp. 229-237.
RoTHsCHILD, B.J.; AuLT, J.S.; GouLLRTQuER, P. and HER-AL, M.,
1994. Decline of the Chesapeake Bay oysterpopulation: a century of
habitat destruction and over-fishing. Marine Ecology Progress
Series, 111, 29-39.
SAS INSTITuTE, INc. STAFF, 1990. SAS Procedures Guide,Version 6,
Third Edition, Cary, North Carolina: SASInstitute, Inc.
SOKAL, R.R. and ROHLF, F.J., 1981. Biometry, Second Edi-tion.
New York: W.H. Freeman and Company, 859p.
SuDING, K.N., GRoss, K.L. and HOUSEMAN, G.R., 2003.Alternative
states and positive feedbacks in restora-tion ecology. Trends in
Ecology and Evolution, 19, 46-53.
ULANOWICZ, R.D. and TurrLE, J.H., 1992. The trophicconsequences
of oyster stock rehabilitation in Chesa-peake Bay. Estuaries, 15,
298-306.
WENNER, E.; BEATTY, H.R. and COEN, L.D., 1996. A quan-titative
system for sampling nekton on intertidal oys-ter reefs. Journal of
Shellfish Research, 15, 769-775.
Journal of Coastal Research, Special Issue No. 40, 2005
78
-
COPYRIGHT INFORMATION
TITLE: Oyster Reef Habitat Restoration: Relationships
BetweenOyster Abundance and Community Development basedon Two
Studies in Virginia and South Carolina
SOURCE: J Coast Res Special Issue no40 Wint 2005WN:
0534907132006
The magazine publisher is the copyright holder of this article
and itis reproduced with permission. Further reproduction of this
article inviolation of the copyright is prohibited.
Copyright 1982-2005 The H.W. Wilson Company. All rights
reserved.