Successful initial restoration of oyster habitat in the lower Hudson River Estuary, United States

Abstract

The eastern oyster (Crassostrea virginica), has experienced dramatic declines throughout its range, and most U.S. states along the Atlantic and Gulf of Mexico coasts have restoration programs. Oyster restoration in the Hudson River Estuary has been a focus of management agencies for over two decades. The present project was initiated in 2013 after sampling for replacement of the Tappan Zee bridge determined live oysters would be lost during bridge construction, thus requiring mitigation measures. Preliminary studies identified three areas for pilot studies to inform the design of full-scale efforts. A pilot study involving deployment of oyster shell-filled gabions and two styles of Reef Balls, found all substrates supported oyster recruitment and growth but there were higher oyster densities on gabions. Some of the gabions, however, failed structurally necessitating a need for new design. For full-scale restoration, a total of 422 modified gabions and 881 Mini-Bay style Reef Balls were deployed across three sites which totaled approximately 2.4 ha (6 ac) in seafloor area. Both substrate types were heavily colonized by oysters and several other species at all three sites, essentially duplicating the findings of the pilot study. The final monitoring event in 2020 indicated a total of approximately 5.8 million live oysters were on the substrates deployed in 2018. This mitigation effort was the largest oyster habitat restoration project in, or north of, the New York Harbor region in recent decades based on restoration area and several oyster metrics of early success.

Full text

RESEARCH ARTICLE
Successful initial restoration of oyster habitat in the
lower Hudson River Estuary, United States
Raymond Grizzle1,2 , James Lodge3, Krystin Ward1, Katie Mosher4, Fred Jacobs5, Justin Krebs5
The eastern oyster (Crassostrea virginica), has experienced dramatic declines throughout its range, and most U.S. states along
the Atlantic and Gulf of Mexico coasts have restoration programs. Oyster restoration in the Hudson River Estuary has been a
focus of management agencies for over two decades. The present project was initiated in 2013 after sampling for replacement of
the Tappan Zee bridge determined live oysters would be lost during bridge construction, thus requiring mitigation measures.
Preliminary studies identified three areas for pilot studies to inform the design of full-scale efforts. A pilot study involving
deployment of oyster shell-filled gabions and two styles of Reef Balls, found all substrates supported oyster recruitment and
growth but there were higher oyster densities on gabions. Some of the gabions, however, failed structurally necessitating a need
for new design. For full-scale restoration, a total of 422 modified gabions and 881 Mini-Bay style Reef Balls were deployed
across three sites which totaled approximately 2.4 ha (6 ac) in seafloor area. Both substrate types were heavily colonized by oys-
ters and several other species at all three sites, essentially duplicating the findings of the pilot study. The final monitoring event
in 2020 indicated a total of approximately 5.8 million live oysters were on the substrates deployed in 2018. This mitigation effort
was the largest oyster habitat restoration project in, or north of, the New York Harbor region in recent decades based on res-
toration area and several oyster metrics of early success.
Key words: ecosystem services, gabions, mitigation, reef balls, shell
Implications for Practice
•This project demonstrates the value of empirical pilot
studies for designing full-scale oyster restoration
activities.
•This project also documents the impact that a broad coa-
lition of partners knowledgeable in various components
of the overall project can play in achieving success.
•The initial success of the full-scale restoration effort indi-
cates excellent potential for additional oyster restoration
in the lower Hudson River as well as additional topics
for future research.
•The finding of wide variability of restoration success met-
rics among the study sites underscores the importance of
site selection and suggests more needs to be learned about
the distribution of live natural reefs and recruitment pat-
terns in the study area.
Introduction
The eastern oyster (Crassostrea virginica) has experienced dra-
matic declines in many areas (Beck et al. 2011; zu Ermgassen
et al. 2012), and most U.S. coastal states along the Atlantic
and Gulf of Mexico have restoration programs (Bersoza
et al. 2018). Some programs focus on restoration for human har-
vest, others are aimed at restoring the ecosystem services oysters
provide, which include water filtration, habitat provision,
benthic–pelagic coupling, and refugia from predation (Coen
et al. 2007; Coen & Humphries 2017). In the Hudson River
Estuary (HRE), the emphasis over the past two decades has been
on these services, and restoration efforts have been highly col-
laborative involving a wide variety of partners (Levinton &
Doall 2011; Lodge et al. 2015).
The eastern oyster has existed in the HRE for at least six
millennia, their reefs at times covering hundreds of hectares of
seafloor area in the Tappan Zee area (Snow 1972; Bell
et al. 2006). The oyster’slong-term history in the area includes
at least two major declines, both associated with extreme
warm–cool cycles (Carbotte et al. 2004). Declines since the
1700s, however, largely have been attributed to human over-
harvesting, pollution, and in recent decades, oyster diseases
(Franz 1982; Kirby 2004; Beck et al. 2011). By the mid-
1900s, only small, isolated populations of living oysters were
known to occur in the New York Harbor region (Franz 1982;
MacKenzie 1984). Recent studies, however, found live oyster
Author contributions: JL, KM, FJ, JK, RG conceived and designed the research; RG,
KW, JL, KM performed the field studies; RG, KW processed and analyzed the samples
and the data; KM, KW contributed materials and analysis tools; RG, JL wrote and
edited the manuscript.
1
Jackson Estuarine Laboratory, University of New Hampshire, 85 Adams Point Road,
Durham, NH 03824, U.S.A.
2
Address correspondence to Raymond Grizzle, email ray.grizzle@unh.edu
3
Hudson River Foundation, 17 Battery Place # 915, New York, NY 10004, U.S.A.
4
Billion Oyster Project, Governors Island, 10 South Street, Slip 7, New York,
NY 10004, U.S.A.
5
AKRF, Inc., 7250 Parkway Drive, Suite 210, Hanover, MD 21076, U.S.A.
© 2023 Society for Ecological Restoration.
doi: 10.1111/rec.14077
Supporting information at:
http://onlinelibrary.wiley.com/doi/10.1111/rec.14077/suppinfo
Restoration Ecology 1of11

populations in the HRE though their spatial extent and condition
were not well-characterized.
Surveys from 1998 to 2004 mapped extensive dead reefs,
some buried under 10 m of soft sediments, in the Tappan Zee
area, and collected a few live oysters (Bell et al. 2006). More
recent research demonstrated the viability of juvenile and adult
oysters in several areas of the New York Harbor region, includ-
ing the lower HRE (Medley 2010; Levinton & Doall 2011;
Levinton et al. 2013). And a field experiment assessing the suc-
cess of small, constructed reefs at five sites in the region found
the highest natural recruitment at the Tappan Zee site, suggest-
ing the presence of live reefs in the general area (Grizzle
et al. 2013). Most recently, seafloor surveys conducted as part
of the Tappan Zee bridge replacement planning and environ-
mental review process found live oysters in several areas and
determined that bridge construction would result in either tem-
porary or permanent loss of oyster habitat. As a result, the
New York Department of Environmental Conservation required
the New York Thruway Authority to provide compensatory mit-
igation of the loss by construction (restoration) of new oyster
habitat.
Oysters are sedentary, broadcast spawners that release their
sperm and eggs into the water column where fertilization occurs.
The resulting larvae develop in the water column for two or
more weeks, then settle onto a hard substrate and metamorphose
into juvenile oysters called “spat.”Spat that survive and develop
for a few months are considered recruits to the existing popula-
tion. Restoration of oyster habitat typically at a minimum
involves deploying hard substrate suitable for settlement of wild
oyster larvae, particularly in areas thought to be substrate-
limited, followed by laboratory-reared juvenile oysters if natural
recruitment is also considered limited (see review by Coen &
Humphries 2017). Previous studies in the Tappan Zee area had
indicated some level of natural recruitment to experimental sub-
strates (Carthan & Levinton 2013; Grizzle et al. 2013), but more
extensive studies were needed. Thus, the mitigation process
requiring construction (restoration) of new oyster habitat
included formation of an Oyster Work Group (OWG) in 2013
consisting of researchers, regulators, consultants, and other
stakeholders, to guide the overall effort (Gann et al. 2019).
The OWG recommended a tiered approach overall where the
results at each level informed the design of the next—an adap-
tive approach. The major findings of the Tiers 1 and 2 studies
(2014–2015) were: (1) live oysters were collected at seven of
nine study sites and (2) very low salinities were recorded at
times at all sites. Densities of live oysters ranged from 0 to
30/m
2
at the nine study sites, and live oysters ranged from spat
(<25 mm) to large adults (115 mm). Although the water quality
data indicated surprisingly low salinity (<1 psu) and dissolved
oxygen (<1 mg/L) levels at times, the oyster data indicated that
at least three sites should be considered for further studies
because of high recruitment.
The present paper describes the subsequent 5-year
(2015–2020) collaborative restoration effort consisting of:
(1) a 3-year pilot study to assess three sites and three substrates
suitable for oyster recruitment; (2) a full-scale construction (res-
toration) of oyster habitat at three sites; and (3) assessment of
full-scale restoration success for 2 years post-construction. We
conclude with a discussion focusing on gaps in our knowledge
for design of future oyster habitat restoration in the HRE.
Methods
Study Area
All studies were conducted in subtidal waters of the Tappan Zee
portion of the HRE (Fig. 1), which has a combination of charac-
teristics suitable for oyster reef development: oligohaline salin-
ity regime, circulation patterns that potentially result in
favorable larval transport, and widespread firm substrates
(mainly sand and gravel) suitable for oyster recruitment
(Starke et al. 2011). Maximum water depths in the study area
are approximately 11 m below mean low water and there are
extensive shoal areas (Starke et al. 2011). Bottom types include
amix of mud, sand, and cobble size sediments (Princeton
Hydro 2015). Although oysters have been observed in the inter-
tidal zone in the New York Harbor region (Medley 2010), they
mainly occur in subtidal waters in the study area. Molluscan
shellfish harvest is prohibited in the lower HRE due to water
quality conditions. Thus, the present project was aimed at oyster
habitat restoration for ecosystem services.
Pilot Study
The pilot study consisted of deployment of test substrates,
assessment of oyster reef development, and monitoring of water
quality. The experimental design was a 3 (sites) 3 (substrate
types) factorial. Three replicates of three test substrates: metal
gabion cages containing oyster shells, small “Lo-Pro”Reef
Balls, and larger “Mini-Bay”Reef Balls were deployed at Sites
1, 5, and 8 (Fig. 1) on 22 June, 2015. Both substrate types pro-
vided opportunities for students to participate in their construc-
tion and sampling (Fig. 2). Datasondes with sensors for
salinity, temperature, and dissolved oxygen were also deployed
in 2015 and 2016 at the three sites. Recruitment was determined
by deploying spat collectors at Sites 1, 5, 8, and 0, in June 2015
and July 2016, and monitored monthly until October each year.
Somewhat different sampling methods were used to quantify
oyster metrics during the 3 years of the pilot study (see Lodge
et al. 2017 for details), but all were extractive methods
(Baggett et al. 2014). Gabions were sampled by removing a con-
sistent volume of shell “cultch”material and counting and mea-
suring shell height on a subset of the live oysters on the shell.
Reef Balls were sampled by counting and measuring all live
oysters in replicate quadrats (0.025 m
2
or 0.01 m
2
), or all
live oysters on the outside of each Reef Ball when density was
low. Analysis of variance (ANOVA) was used to assess the
effects of site, substrate type, and interactions on oyster metrics
(mainly size and density). Size–frequency plots of shell height
distributions were also constructed.
Full-Scale Restoration
Problems during the pilot study led to a new design for the
gabions that included an internal cavity that made them more
Restoration Ecology2of11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

durable and potentially increased habitat provision (see below),
and design of a “clustered”deployment for all substrates for the
full-scale restoration effort. Mini-Bay Reef Balls and modified
gabions were used for full-scale restoration which was initiated
by deploying replicates of the two substrate types at three sites
(Figs. 1&3). Based on the results of earlier surveys, and the
pilot study, Sites 1 and 8 were judged to have the highest prob-
ability of successful restoration. A more limited effort was
added at a third location (the glove [=Site 0]), as there was
already a live oyster reef near that site.
The gabions were modified from the pilot study design by add-
ing an internal cavity to provide space for fish and other
Figure 1. Sites for pilot study (1, 5, and 8), and for full-scale restoration (0 [Glove], 1, and 8). Red polygons indicate previously mapped historical oyster reefs.
Sources: Esri, DeLorme, NAVTEQ, TomTom, Intermap, increment P Corp., GEBCO, USGS, FAO, NPS, NRCAN, GeoBase, IGN, Kadaster NL, Ordnance
Survey, Esri Japan, METI, Esri China (Hong Kong), swisstopo, and the GIS User Community.
Restoration Ecology 3of11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

organisms, and to increase their structural integrity. This design
also increased the space available for larvae to penetrate and
reduced the amount of recycled shell required for the project.
Gabions were constructed of 12.5 mm rolled round bar frames
with the resulting cube lined internally with 12.5-gauge,
24.5 mm mesh wire, and were filled with seasoned (dried onshore
for a minimum of 12 months) oyster and clam shell collected from
local restaurants (see Fig. 3for dimensions of substrates). Site
0encompassed an area of 0.03 ha (0.07 ac) and consisted of one
cluster of 54 Reef Balls and one cluster of 36 gabions. Site
1 encompassed an area of 1.35 ha (3.35 ac) and consisted of
414 Reef Balls in 15 clusters and 193 gabions in 11 clusters. Site
8 encompassed an area of 1.04 ha (2.57 ac) and consisted of
413 Reef Balls in 15 clusters and 193 gabions in 11 clusters. All
substrates were deployed in water depths greater than 3 m at mean
lower low water during 10–31 July, 2018.
Oyster size and density were characterized by annual sam-
pling in fall 2019 and 2020, thus providing three summer
recruitment periods (2018, 2019, and 2020). After the substrates
were removed from the water, the number and size (shell height
measured with calipers or ruler to nearest 1 mm) of individual,
live oysters were determined. For the Reef Balls, in 2019, if
the number of oysters and oyster spat for the entire Reef Ball
was less than 50, all oysters on the exterior and interior surfaces
were counted and measured. If the number of oysters for the
entire reef ball was greater than 50, individual live oysters in
four replicate 0.04 m
2
(20 cm 20 cm) quadrats placed ran-
domly at multiple locations on the exterior of the reef ball were
measured. In 2020, all counts and measurements were made
using duplicate 0.04 m
2
quadrats placed at random locations
on opposite sides of each Reef Ball. For gabions in both years,
a section of the wire mesh from the tops of the gabion cages in
two areas was opened using wire cutters and two 0.04 m
2
(20 cm 20 cm) quadrats were placed haphazardly. All shells
were excavated from the upper 2 cm, all live oysters were
counted, and shell height (to nearest 1 mm) was measured.
Oyster metrics from both years were compared graphically
but only the data from the final dataset (2020) were assessed sta-
tistically. The effects of site and substrate type on oyster size and
density were compared among sites as well as the interaction of
Figure 2. Retrieval of substrates involved divers attaching lines, then winching onboard for processing (photos by Katie Mosher).
Restoration Ecology4of11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

substrate and site in separate ANOVAs with post hoc Tukey’s
tests. The normality and homogeneity of oyster density and shell
height datasets were visually confirmed. Two outliers were
removed (Gabion treatment in Site 8) to improve normality of
the oyster density dataset. Data analysis and visualization was
conducted in R 2022.07.2 with dplyr,stats,ggplot, and patch-
work packages (Wickham 2016; R Core Team 2013;
Pedersen2023; Wickham et al. 2023).
Results
Pilot Study
Details on the results of the pilot study are in Lodge et al. (2017).
Here, we provide a summary of only the major findings relevant
for design of full-scale restoration. Live oysters were found on
all three deployed substrate types in all 3 years, with total mean
densities by site ranging from 7 to 405/m
2
. On the final sampling
in 2017, live oyster density on the gabions (all three sites com-
bined) was greater than 10-fold greater than either Reef Ball
type. Both Reef Ball types had similar densities. The combined
oyster size dataset (all substrates and sites) clearly showed
1 year class in 2015, 2 in 2016, and 3 in 2017. Overall, these data
indicated successful recruitment for all 3 years.
Full-Scale Restoration
When considering datasets from the full-scale restoration
(2019–2020), the 3-year pilot study (2015–2017; Lodge
et al. 2017), and a preliminary recruitment study in the same area
(2015–2016; AKRF 2016a,2016b)—substantial annual spat
sets were recorded throughout the study area for 6 consecutive
years. And monthly monitoring of spat collectors during the
pilot study indicated larval setting usually occurred between
mid-August and mid-September. These are important findings
for future restoration activities because they suggest that only
addition of appropriate substrates during late spring/early sum-
mer will likely be necessary for restoration success.
A total of 37 Reef Balls and 20 gabions were monitored from
the three sites in 2019, and sampling occurred on 6 days over
two periods (30 September–2 October, and 29–30 October;
approximately 14 months after deployment). In 2020, a total of
36 Reef Balls and 20 gabions were monitored from the three sites,
and sampling occurred on 5 and 6 October, approximately
26 months after deployment. Sampling occurred in the fall so
newly recruited oysters (spat from the typical summer spawn)
would be detected. Both substrate types were heavily colonized
by oysters and several other taxa at all three sites, essentially dupli-
cating the general findings of the 3-year pilot study (Lodge
et al. 2017;Fig.4). At Site 0, oysters had achieved approximately
100% areal coverage of both substrates in some areas, and some
oyster clusters projected greater than 10 cm above the substrate
surface. Less oyster areal coverage and vertical height occurred
at Sites 1 and 8, but development at both sites was substantial.
Site Differences in Oyster Metrics. There were marked dif-
ferences in oyster density among the sites in both years, and both
Figure 3. Top: shell-filled gabions and Mini-Bay Reef Balls used in full-scale construction. Bottom: deployment of the two substrate types, gabions and Mini-
Bay style Reef Balls in July 2018 (photos by Brian DeGasperis/NYSDEC).
Restoration Ecology 5of11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

annual datasets (2019 and 2020) showed the same by-site trends
for live oyster density: Site 0 greater than Sites 1 and 8 (Figs. 4&5).
In 2020, mean oyster densities across both substrates were greatest at
Site 0 and similar between the other two sites (F
[2,102]
=17.30,
p< 0.001; Table S1), resulting in approximately 40% greater
densities at Site 0. Mean shell height was approximately 25%
greater at Site 0 than between the other two sites with similar
oyster sizes (F
[2,104]
=34.05, p< 0.001).
Substrate Differences in Oyster Metrics. The effect of sub-
strate type on oyster density also indicated similar relative trends
in 2019 and 2020 (Fig. 5). There was no difference in oyster den-
sity between the Reef Balls and gabion treatments across sites
(t
1,103
=3.85, p=0.053), however oyster shell heights were
greater on Reef Balls (t
1,105
=3.97, p=0.049). It should be
noted that the greater densities on the gabions compared to the
Reef Balls during the pilot study (see above) may indicate that
the magnitude of the differences between the two substrates
decreases as populations develop.
Oyster Metrics by Site and Substrate. There were no signif-
icant interaction effects in 2020 between site and substrate types
on both mean oyster density (F
[2,99]
=1.43, p=0.245) and
shell height (F
[2,101]
=0.66, p=0.52), suggesting that the
overall trends for the main effects of site and substrate type did
not vary across the three sites.
Oyster Size Distributions by Site and Substrate. As already
noted, three size/year classes of oysters were expected on the
substrates sampled in October 2020 because their deployment
in July 2018 allowed for three summer reproduction and
recruitment periods. Size–frequency plots of the 2019 data indi-
cated two size/year classes and the 2020 data indicated three
classes, but perhaps only at Site 0 where the largest oysters in
2020 exceeded 120 mm shell height (Fig. 6). Data from the
other sites and substrates were more variable, which may indi-
cate variations in growth and/or survival. One major difference
in patterns between the 2 years was strong small size classes
(10–15 mm) at Sites 1 and 8 in 2019 but not 2020 (Fig. 6).
Assessment of Overall Mitigation Effort. Table 1summarizes
the results of the overall mitigation effort with respect to abun-
dances of live oysters on the substrates. There were a total of
approximately 5.8 million live oysters on the deployed sub-
strates at the three mitigation sites in October 2020. Although
most of the oysters were recent recruits (Fig. 6), substantial
numbers of second and third year individuals (many >100 mm
shell height) were present indicating good survival and growth
for the 2.3-year development period.
Water Quality Trends. The present project also provided addi-
tional data on surprisingly robust oyster performance in waters
of very low salinity (see AKRF 2021 for details). During the
5-year study period, spring, and early summer (April–July)
salinity measurements were often less than 5 practical salinity
unit (PSU) and sometimes less than 1. The observed strong oys-
ter performance was surprising for such extreme conditions.
Discussion
The present mitigation effort was the largest oyster restoration
project in, and north of, the New York Harbor region in recent
decades. The final sampling in 2020 indicated a total of
Figure 4. Typical substrates retrieved from the three full-scale restoration sites in 2019 (left) and 2020 (right). Note greater oyster habitat development on both
substrate types at Site 0 compared to other sites.
Restoration Ecology6of11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

approximately 5.8 million live oysters had recruited to the
deployed substrates. The total permitted restoration area of the
three sites was approximately 2.4 ha (6.0 ac) but the Reef Balls
and gabions covered less than 5% of the total permitted area.
Even so, they provided almost an acre (approximately 14% of
the total area) of “hard substrate”potentially suitable for oyster
colonization because of their structural complexity. Both sub-
strates project about a meter above the seafloor and provide
nearly 3 m
2
of surface area per unit. The gabions have much
more small-scale surface area than the Reef Balls due to the
rough surface provided by the oyster and clam shells, but this
was not accounted for in any of the calculations. Reef Balls have
been used extensively in many areas, and their effectiveness for
habitat restoration in various respects is well-documented
(http://www.reefball.org). In contrast, shell-filled gabions have
only recently begun to be used for oyster restoration
(e.g. Safak et al. 2020) and their initial effectiveness as well as
durability and sustainability in the long term remain to be thor-
oughly tested. The similar short-term effectiveness of the two
substrate types suggests that—as has been demonstrated in
many studies (see Bersoza et al. 2018 for review)—larvae settle
and oyster populations develop on many types of hard substrate.
Nonetheless, we are aware of no long-term (>5 years) studies
that compare different substrate types. Additional monitoring
is needed to assess the long-term success of our project.
Our study also provided strong evidence of the potential for
additional restoration of oyster habitat in the general area of
the Tappan Zee involving only the addition of appropriate hard
substrates. Work on several topics, however, is needed to move
the process forward. Foremost is the need for more information on
the spatial extent of live oysters and bottom types potentially suit-
able for oyster restoration. As discussed in the Introduction sec-
tion, bottom surveys in the early 2000s (Carbotte et al. 2004;
Bell et al. 2006), and more recently as part of the preliminary stud-
ies for the present mitigation project (AKRF 2016a,2016b), pro-
vided evidence that substantial live oyster populations existed in
the lower HRE. None of these studies, however, provided infor-
mation sufficient to characterize the spatial extent of live oysters.
Thus, additional acoustic surveys coupled with extensive bottom
sampling are needed to adequately characterize the spatial extent
and condition of live oyster habitat in the lower HRE.
The Tappan Zee area was the location of a historical (1950s)
major commercial oyster industry that involved transferring
adult oysters from Long Island Sound onto approximately
5000 acres of leased bottom in the HRE with the aim of enhanc-
ing spawning and local recruitment, followed by removing
“seed”(juvenile) oysters presumably produced by the trans-
ferred adults, for transplanting to growing areas in Long Island
Sound (Bromley 1954). The overall process was perhaps pat-
terned after the long-recognized notion that recruitment is
Figure 5. Mean (1 SE) oyster densities (top) and oyster shell height (bottom) by site and substrate for 2019 and 2020.
Restoration Ecology 7of11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

positively related to proximity to broodstock (Brooks 1891,
pp 204–205). More recent research in Chesapeake Bay has also
demonstrated the importance of proximity to broodstock oysters
for restoration success (Schulte et al. 2009; Schulte &
Burke 2014). Recent research in New Hampshire similarly
found that most successful oyster reef restoration projects were
less than 1 km from a healthy natural reef (Grizzle et al. 2021),
and experiments involving three natural reefs found that most
recruitment occurred less than 500 m from the reefs
(Atwood & Grizzle 2020). Previous studies in the Tappan Zee
area documented oyster recruitment but were not aimed at deter-
mining spatial gradients in lower reaches of the HRE (Carthan &
Levinton 2013; Kulp & Peterson 2016; McFarland &
Hare2018). Future mapping work focused on live oyster habitat
and natural recruitment patterns are clearly needed.
The oyster metrics from all study sites indicated surprisingly
robust performance in waters of very low salinity during the
warmer months. Previous field studies (Levinton et al. 2011;
Grizzle et al. 2013; McFarland & Hare 2018) and multi-factor
spatial modeling of restoration suitability for the eastern oyster
(Starke et al. 2011) identified the Tappan Zee area as potentially
providing a low-salinity refuge from disease and predators, but
with a tradeoff of higher mortality due to sporadic storm events
that could result in prolonged salinity below the oyster’s
Figure 6. Size–frequency distributions by site and substrate type for fall 2019 and 2020 oyster data, representing three recruitment periods.
Table 1. Abundances of live oysters in 2020 (2 years post-deployment) by site and substrate type. Estimated no. of live oysters =total no. of substrates
deployed surface area of each substrate mean live oyster density.
Substrate type
Total no.
of substrates
deployed
Surface area of
each substrate (m
2
)
Mean live oyster
density (no./m
2
)1SE
Estimated no.
of live oysters 1 SE
0 Reef Ball 54 2.74 2394 186 354,216 2,7521
0 Gabion 36 1.94 2844 275 198,625 1,9206
1 Reef Ball 414 2.74 1315 125 1,491,683 14,1795
1 Gabion 193 1.94 1702 165 637,263 6,1779
8 Reef Ball 413 2.74 1965 185 2,223,633 20,9350
8 Gabion 193 1.94 2311 522 865,285 19,5447
Total 5,770,705
Restoration Ecology8of11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

tolerance. During the present 5-year study period (pilot and full-
scale), spring, and early summer (April–July) salinity measure-
ments were often less than 5 PSU and sometimes near zero
(AKRF 2021), far below what is generally considered optimum
(Shumway 1996; Starke et al. 2011). There is a growing litera-
ture that documents tolerance of some populations of the eastern
oyster to low salinities and the genetic basis for such tolerance
(e.g. McCarty et al. 2022; Swam et al. 2022). This is an impor-
tant research topic relevant to the design of future restoration
projects in the HRE and elsewhere.
Additionally, our data suggested individual growth rates com-
parable to other studies in the region. Individual oyster growth
was not directly determined in our study, but it can be estimated
based on size-frequency data. Growth rates for shell height of
25–40 mm/year for the first few years have been reported in ear-
lier studies in the New York Harbor area (Cerrato 2006;Med-
ley2010; Levinton & Doall 2011; Levinton et al. 2013).
Although rates can vary widely, using 35 mm/year as an approx-
imate mean value, the size-frequency data from Site 0 for both
years (for full-scale restoration) indicate 2-year classes in 2019
and 3 in 2020. Data from the other two sites, however, were more
variable which may indicate variations in growth and/or survival
rates among the sites and between the substrates.
Although the major focus of the monitoring effort was oysters,
other species occurred in the “fouling”communities (i.e. all spe-
cies that typically occur on hard substrates) that developed on
the experimental substrates, and two taxa (barnacles and mussels)
were at much higher densities than oysters (Lodge et al. 2020).
The timing of larval settlement among the fouling community
species could not be determined but based on the approximately
100% cover of barnacles (Balanus improvisus) and mussels
(Ischadium recurvum) observed in some areas it is possible they
could have inhibited oyster settlement. The relationships among
species in fouling communities is complex and includes the obvi-
ous competition for space as well as predation on settling larvae
(Kochman et al. 2008; Boudreaux et al. 2009;Barnes
et al. 2010). Further studies might yield relevant information rel-
evant to the timing of restoration activities such as substrate
deployment. Although our study was not designed to unravel
the effects of various species interactions, no potential causes
(e.g. known predators) for their declines were evident and there
were wide variations in seasonal and year-to-year abundances of
both taxa. Thus, the developing oyster populations on the exper-
imental substrates provided habitat for other sedentary species,
and presumably other motile species typically associated with
oyster reefs in the region (Peterson & Kulp 2013;Glenn
et al. 2020). The new fouling communities on the deployed sub-
strates also provided considerable, though unmeasured, water fil-
tration due to the substantial abundances of at least three filter
feeding groups (oysters, mussels, and barnacles). The provision
of these and other ecosystem services by the new oyster habitat
should be important topics for future research in the HRE.
Acknowledgments
The Hudson River Foundation, University of New Hampshire,
Billion Oyster Project, and AKRF, Inc. were the major partners
for the project. The New York State Thruway Authority pro-
vided funding and guidance for the project. New York
Harbor School students and instructors contributed design
expertise, fabrication, and handling of the gabions. Captain
Mike’s Diving Services, Inc. provided divers for substrate
retrieval. Billion Oyster Project’s Public Volunteer Program
workforce fabricated gabions using shell donated by dozens of
New York City restaurants. Captain Mike Abegg (Brooklyn
Marine Services) and their New York Harbor School alumni
crew led vessel operations for the pilot and monitoring phases.
The Arben Group deployed the Reef Balls and the Billion
Oyster Project’s gabions. Princeton Hydro conducted the initial
oyster and seafloor reconnaissance sampling. The NYSDEC and
the Oyster Work Group provided technical feedback on study
design and report review, and Prudent Engineering conducted
the Tier 1 seafloor characterization using side-scan sonar.
G. McKown provided the statistical analyses. Finally, we thank
J. Levinton and an anonymous reviewer for extensive construc-
tive critical reviews that greatly improved the manuscript.
LITERATURE CITED
AKRF (2016a) Tier 3 progress report, oyster research and restoration plan, new
NY bridge project. AKRF, Inc., New York. https://www.hudsonriver.org/
wp-content/uploads/2023/05/Tier-3-Progress-Report_March2016-Final.pdf
(accessed 15 December 2023)
AKRF (2016b)Tier 3 progress report, oyster research and restoration plan, new
NY bridge project. AKRF, Inc., New York. https://www.hudsonriver.org/
wp-content/uploads/2023/05/Tier-3-Progress-Report_November2016.pdf
(accessed 15 December 2023)
AKRF (2021) 2020 post-construction oyster monitoring final report. AKRF, Inc.,
New York. https://www.hudsonriver.org/wp-content/uploads/2021/09/
TZB-Final-Report.pdf (accessed 14 December 2023)
Atwood RL, Grizzle RE (2020) Eastern oyster recruitment patterns on and near
natural reefs: implications for the design of oyster reef restoration projects.
Journal of Shellfish Research 39:283–289. https://doi.org/10.2983/035.
039.0209
Baggett LP, Powers SP, Brumbaugh R, Coen LD, DeAngelis B, Greene J,
Hancock B, Morlock S (2014) Oyster habitat restoration monitoring and
assessment handbook. The Nature Conservancy, Arlington, Virginia.
http://www.oyster-restoration.org/wp-content/uploads/2014/01/Oyster-
Habitat-Restoration-Monitoring-and-Assessment-Handbook.pdf
Barnes BB, Luckenbach MW, Kingsley-Smith PR (2010) Oyster reef community
interactions: the effect of resident fauna on oyster (Crassostrea spp.) larval
recruitment. Journal of Experimental Marine Biology and Ecology 391:
169–177. https://doi.org/10.1016/j.jembe.2010.06.026
Beck MW, Brumbaugh RD, Airoldi L, Carranza A, Coen LD, Crawford C,
Defeo O, Edgar GJ, Hancock B, Kay MC (2011) Oyster reefs at risk and
recommendations for conservation, restoration, and management. Biosci-
ence 61:107–116. https://doi.org/10.1525/bio.2011.61.2.5
Bell RE, Flood RD, Carbotte SM, Ryan W, Mchugh CMG, Cormier M-H, et al.
(2006) Benthic habitat mapping in the Hudson River Estuary. Pages 51–64.
In: Levinton JS, Waldman JR (eds) The Hudson River Estuary. Cambridge
University Press, New York
Bersoza AH, Brumbaugh RD, Frederick P, Grizzle R, Luckenbach MW,
Peterson CH, Angelini C (2018) Restoring the eastern oyster: how much
progress has been made in 53 years? Frontiers in Ecology and Environmen-
tal Science 16:1–9. https://doi.org/10.1002/fee.1935
Boudreaux ML, Walters LJ, Rittschof D (2009) Interactions between native bar-
nacles, non-native barnacles, and the eastern oyster Crassostrea virginica.
Bulletin of Marine Science 84:43–57
Restoration Ecology 9of11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

Bromley A (1954) The oyster and the brothers flower. The New York State Con-
servationist 8:4–9
Brooks WK (1891) The oyster, a popular summary of a scientific study. Maryland
paperback bookshelf edition. Vol 1996. The Johns Hopkins University
Press, Baltimore, Maryland. https://doi.org/10.5962/bhl.title.11479
Carbotte SM, Bell RE, Ryan WBF, McHugh C, Slagle A, Nitsche F,
Rubenstone J (2004) Environmental change and oyster colonization within
the Hudson River estuary linked to Holocene climate. Geo-Marine Letters
24:212–224. https://doi.org/10.1007/s00367-004-0179-9
Carthan R, Levinton JS (2013) Recruitment of oysters within the Hudson River
estuary. Section III. Pages 1–28. In: Fernald SH, Yozzo D, Andreyko H
(eds) Final reports of the Tibor T. Polgar fellowship program, 2012. Hud-
son River Foundation, New York
Cerrato RM (2006) Long-term and large-scale patterns in the benthic communi-
ties of New York Harbor. Pages 242–265. In: Levinton JS, Waldman JR
(eds) The Hudson River estuary. Cambridge University Press, New York.
https://doi.org/10.1017/CBO9780511550539.020
Coen LD, Brumbaugh RD, Bushek D, Grizzle R, Luckenbach MW, Posey MH,
Powers SP, Tolley SG (2007) As we see it: ecosystem services related to
oyster restoration. Marine Ecology Progress Series 341:303–307. https://
doi.org/10.3354/meps341303
Coen LD, Humphries AT (2017) Oyster-generated marine habitats: their services,
enhancement, restoration and monitoring. Pages 274–294. In: Routledge
handbook of ecological and environmental restoration. Routledge, New York
Franz DR (1982) An historical perspective on mollusks in lower New York Har-
bor, with emphasis on oysters. Pages 181–197. In: Meyer GF
(ed) Ecological stress and the New York bight: science and management.
Estuarine Research Federation, Columbia, South Carolina
Gann GD, McDonald T, Walder B, Aronson J, Nelson CR, Jonson J, et al. (2019)
International principles and standards for the practice of ecological restora-
tion. Second edition. Restoration Ecology 27:S1–S46. https://doi.org/10.
1111/rec.13035
Glenn M, Mathieson A, Grizzle R, Burdick D (2020) Seaweed communities in
four subtidal habitats within the Great Bay estuary, New Hampshire: oyster
farm gear, oyster reef, eelgrass bed, and mudflat. Journal of Experimental
Marine Biology and Ecology 524:151307. https://doi.org/10.1016/j.
jembe.2019.151307
Grizzle R, Ward K, Konisky R, Greene J, Abeels H, Atwood R (2021) Oyster reef
restoration in New Hampshire, U.S.A.: lessons learned during two decades
of practice. Ecological Restoration 39:260–273. https://doi.org/10.3368/er.
39.4.260
Grizzle R, Ward K, Lodge J, Suszkowski D, Mosher-Smith K, Kalchmayr K,
Malinowski P (2013) Oyster restoration research project final technical
report, phase I: experimental oyster reef development and performance
results. 2009–2012. http://www.hudsonriver.org/download/ORRP_Phase1.
2013.pdf (accessed 14 December 2022)
Kirby MX (2004) Fishing down the coast: historical expansion and collapse of oyster
fisheries along continental margins. Proceedings of the National Academy of
Sciences 101:13096–13099. https://doi.org/10.1073/pnas.0405150101
Kochman J, Buschbaum C, Volkenbom N, Reise K (2008) Shift from native mus-
sels to alien oysters: differential effects of ecosystem engineers. Journal of
Experimental Marine Biology and Ecology 364:1–10. https://doi.org/10.
1016/j.jembe.2008.05.015
Kulp RE, Peterson BJ (2016) Evaluating the impact of mesopredators on oyster
restoration in the New York metropolitan region. Journal of Shellfish
Research 35:801–807. https://doi.org/10.2983/035.035.0408
Levinton JS, Doall M (2011) Guiding oyster restoration: growth, condition and
spawning success of experimental populations of oysters throughout New
York-New Jersey Harbor. Final report to the Hudson River Foundation,
New York. http://hudsonriver.org/wp-content/uploads/library/Levinton_
006_07A_final_report.pdf (accessed 14 December 2022)
Levinton J, Doall M, Allam B (2013) Growth and mortality patterns of the eastern
oysters Crassostrea virginica in impacted waters in coastal waters in New
York, U.S.A. Journal of Shellfish Research 32:417–427. https://doi.org/10.
2983/035.032.0222
Levinton J, Doall M, Ralston D, Starke A, Allam B (2011) Climate change, pre-
cipitation and impacts on an estuarine refuge from disease. PLoS ONE 6:
e18849. https://doi.org/10.1371/journal.pone.0018849
Lodge, J., Grizzle, R., Coen, L., Mass Fitzgerald, A., Comi, M., Malinowski, P.
2015. Final report of the NOAA/WCS regional partnership grant, New
York. https://www.hudsonriver.org/wp-content/uploads/2017/10/HRF-
NOAA-WCS-Final-Report-Web-Version.pdf (accessed 14 December
2023)
Lodge J, Grizzle R, Ward, K, Malinowski P (2017) Final report –tier 3 Tap-
pan Zee Bridge oyster restoration pilot study. Fred Jacobs, AKRF,
Inc. https://www.hudsonriver.org/wp-content/uploads/2017/10/HRF-
Tappan-Zee-Final-Report_2017-12-21.pdf (accessed 14 December
2023)
Lodge J, Grizzle R, Ward K, Mosher K, Burmester L (2020) Progress Report #1 –
preliminary data assessment, the governor Mario M. Cuomo/new NY
bridge project at Tappan zee oyster habitat restoration study –oyster mon-
itoring. Submitted to Fred Jacobs, AKRF, Inc. https://www.hudsonriver.
org/wp-content/uploads/2021/09/TZB-Final-Report.pdf (accessed 11
December 2023)
MacKenzie CL Jr (1984) A history of oystering in Raritan Bay, with environmen-
tal observations. Pages 37–66. In: Pacheco AL (ed) Raritan Bay its multiple
uses and abuses. Technical Series Report 30. U.S. Department of Com-
merce, NOAA, National Marine Fisheries Service, NEFC, Sandy Hook
Laboratory, Sandy Hook, New Jersey
McCarty AJ, Allen SK Jr, Plough1 LV (2022) Genome-wide analysis of acute
low salinity tolerance in the eastern oyster Crassostrea virginica and poten-
tial of genomic selection for trait improvement. G3 12:jkab368. https://doi.
org/10.1093/g3journal/jkab368
McFarland K, Hare MP (2018) Restoring oysters to urban estuaries: redefin-
ing habitat quality for eastern oyster performance near new York City.
PLoS ONE 13:e0207368. https://doi.org/10.1371/journal.pone.
0207368
Medley TL (2010) Wild oysters, Crassostrea virginica, in the Hudson River estu-
ary: growth, health and population structure. PhD dissertation. The City
University of New York, New York, https://doi.org/10.1080/09602011.
2010.517688
Pedersen T (2023) patchwork: the composter of plots. https://patchwork.data-
imaginist.com,https://github.com/thomasp85/patchwork (accessed 2
December 2023)
Peterson B, Kulp R (2013) Investigating ecological restoration: enhancement of fish-
eries due to the presence of oyster reefs in the Hudson River 2011–2012.
http://www.oyster-restoration.org/wp-content/uploads/2012/06/HRF-final-
report_Peterson_2013-small.pdf (accessed 14 December 2023)
Princeton Hydro (2015) Tier 2 oyster sampling report, Tappan Zee Bridge Hud-
son River crossing project, Westchester and Rockland counties, New York.
AKRF, Inc., Hanover, Maryland
R Core Team (2013) R: a language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. http://www.R-
project.org/
Safak I, Norby PL, Dix N, Grizzle RE, Southwell M, Veenstra JJ, et al. (2020)
Coupling breakwalls with oyster restoration structures enhances living
shoreline performance along energetic shorelines. Ecological Engineering
158:106071. https://doi.org/10.1016/j.ecoleng.2020.106071
Schulte DM, Burke RP (2014) Recruitment enhancement as an indicator of oyster
restoration success in the Chesapeake Bay. Ecological Restoration 32:434–
440. https://doi.org/10.3368/er.32.4.434
Schulte DM, Burke RP, Lipcius RN (2009) Unprecedented restoration of a native
oyster metapopulation. Science 325:1124–1128. https://doi.org/10.1126/
science.1176516
Shumway S (1996) Environmental factors. Pages 467–514. In: Kennedy VS,
Newell RIE, Eble AF (eds) The eastern oyster Crassostrea virginica. Mary-
land Sea Grant College, University of Maryland System, College Park
Snow DR (1972) Rising sea level and prehistoric cultural ecology in northern
New England. American Antiquity 37:211–221. https://doi.org/10.2307/
278207
Restoration Ecology10 of 11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

Starke A, Levinton JS, Doall M (2011) Restoration of Crassostrea virginica
(Gmelin) to the Hudson River, U.S.A.: a spatiotemporal modelling
approach. Journal of Shellfish Research 30:671–684. https://doi.org/10.
2983/035.030.0309
SwamLM,LaPeyreMK,CallamBR,LaPeyreJF (2022) Local populations of east-
ern oyster from Louisiana differ in low-salinity tolerance. North Ameri can Jour-
nal of Aquaculture 84:381–391. https://doi.org/10.1002/naaq.10248
Wickham H (2016) ggplot2: elegant graphics for data analysis. https://ggplot2.
tidyverse.org (accessed 2 December 2023)
Wickham H, Francois R, Henry L, Vaughan D (2023) dplyr: a grammar of data manip-
ulation. https://dplyr.tidyverse.org (accessed 2 December 2023)
zu Ermgassen PSE, Spalding MD, GrizzleRE, Brumbaugh RD (2012) Quantifying the
loss of marine ecosystem service: filtrationbytheeasternoysterinUSestuaries.
Estuaries and Coasts 36:36–43. https://doi.org/10.1007/s12237-012-9559-y
Supporting Information
The following information may be found in the online version of this article:
Table S1. Compiled Student ttests and ANOVAs with post hoc Tukey’s tests on the
final (2020) dataset for oyster density and oyster shell height by site, substrate, and the
interaction of site and substrate.
Coordinating Editor: Michael Sievers Received: 23 May, 2023; First decision: 30 August, 2023; Revised: 30 November,
2023; Accepted: 3 December, 2023
Restoration Ecology 11 of 11
Oyster restoration in the Hudson
1526100x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/rec.14077, Wiley Online Library on [21/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License