Environmental Pollution 316 (2023) 120614

Contents lists available at ScienceDirect

Environmental Pollution
journal homepage: www.elsevier.com/locate/envpol

Shellfish as a bioremediation tool: A review and meta-analysis

Giulia Filippini a, *, Katherine A. Dafforn a, b, Ana B. Bugnot c, d
a
School of Natural Sciences, Macquarie University, North Ryde, NSW, 2109, Australia
b
Sydney Institute of Marine Science, Mosman, NSW, 2088, Australia
c
School of Life and Environmental Sciences, The University of Sydney, Sydney, NSW, 2006, Australia
d
Commonwealth Scientific and Industrial Research Organisation, Brisbane, QLD, 4001, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Over the last century, human activities have increased the amount of nutrients inputs to terrestrial and aquatic
Bioremediation ecosystems. These activities have altered nitrogen (N) and phosphorus (P) cycling, causing substantial changes in
Meta-analysis ecosystem function such as provision of clean air and water. Strategies that reduce and remove excess nutrients
Shellfish
are urgently needed to remediate impacted systems. Reef-forming shellfish (oysters and mussels) can play a
Nutrient fluxes
Denitrification
crucial role in nutrient cycling, particularly in N removal from aquatic systems by providing substrate for mi­
Environmental factors crobial colonisation and enhancing microbial denitrification in the surrounding sediments. However, the po­
tential for shellfish to enhance nutrient cycling (and denitrification) will likely vary spatially and in response to
several environmental factors. Here, we used 1) a qualitative analysis to review nutrient processes occurring on
shellfish; and 2) a meta-analysis to evaluate the influence of shellfish on benthic metabolism and nutrient cycling
in surrounding sediments, and how that is influenced by environmental factors such as grain size, seasonality,
water body type, and tidal position. Overall, we found that shellfish increased oxygen consumption, with
3-
consequent release of ammonia (NH+ 4 ) and phosphate (PO4 ) from shellfish and their surrounding sediments.
These parameters did not depend on grain size, water body type and tidal height, but the release of PO3- 4 was
variable between seasons, being highest during summer and autumn. Shellfish presence also enhanced denitri­
fication measured as dinitrogen gas (N2) efflux on both reefs and sediments. Denitrification was highest in la­
goons; in sandy sediments; and during the warmest season (summer). Thus, our findings highlight that
environmental context can mediate the effects of shellfish reefs on sediment function. This information is
important for managers seeking to use these animals as an effective bioremediation tool.

1. Introduction remineralisation of nutrients worldwide due to processes occurring in

the widely distributed soft-sediment habitats (Middelburg et al., 1997;
Bioavailable nitrogen (N) and phosphorus (P) are limiting resources Nixon, 1981). In addition, biogenic habitats such as mangroves, salt­
in natural ecosystems; hence they control primary productivity and marsh, seagrass, and shellfish reefs also play key roles in the processing
underpin the biodiversity and function of ecosystems on Earth. How­ of nutrients in these systems (Fourqurean et al., 2012; Kellogg et al.,
ever, over the last century, human activities have increased the amount 2014; Smyth et al., 2016; Valiela and Cole, 2002). Notwithstanding their
of these bioavailable nutrients in terrestrial and aquatic systems through importance to the health of global ecosystems, these coastal systems are
deforestation, burning of fossil fuels, the use of fertilisers and organic often highly degraded due to eutrophication from excess nutrients in­
waste. These activities have altered N and P cycling at global scales, puts derived from fertilizers, urban sewage, industrial wastewater, and
causing substantial changes in ecosystem function (Steffen et al., 2015; agricultural runoff, which accumulate in sediments (Herbert, 1999;
Vitousek et al., 1997). For example, increased emissions of nitrous oxide Lehtoranta et al., 2009). Increased bioavailable N and P in these systems
and nitric oxide into the atmosphere have contributed to global warm­ has resulted in algal blooms, anoxia and losses in biodiversity (Anderson
ing, formation of photochemical smog and production of acid rain et al., 2002; Meyer-Reil and Köster, 2000; Nixon, 1995). Management
(Gruber and Galloway, 2008). strategies for the reduction of nutrient inputs have been in place globally
Coastal marine ecosystems are responsible for 50% of the for several decades, but this effort alone has not been able to reverse

* Corresponding author. Macquarie University, NSW, 2109, Australia.

E-mail address: [email protected] (G. Filippini).

https://doi.org/10.1016/j.envpol.2022.120614
Received 26 May 2022; Received in revised form 17 October 2022; Accepted 4 November 2022
Available online 7 November 2022
0269-7491/© 2022 Elsevier Ltd. All rights reserved.
G. Filippini et al. Environmental Pollution 316 (2023) 120614

nutrient enrichment and recover ecological functions (Duarte and biodeposition rates in mussels compared to oysters; and Grenz (1989)
Krause-Jensen, 2018). Additional strategies that promote the decom­ indicated that mussels’ biodeposits had higher organic content than
position of organic matter and remineralisation of nutrients are needed oysters’. Additionally, Hawkins et al. (1998) reported that oysters are
to recover the health of these degraded ecosystems. less efficient than mussels in the selection and digestion of organic
In particular, the loss of more than 85% of shellfish reefs globally has particles, resulting in lower rates of energy gain and nutrient assimila­
meant of decline of their capacity to regulate water quality and process tion. On the other hand, a recent study conducted by Barrett et al. (2022)
nutrients (Beck et al., 2011). Shellfish such as oysters and mussels are revealed that rates of nitrogen removal by area via bioextraction and
benthic-pelagic couplers, filtering nutrients suspended in the water sediment denitrification were similar in cultivated oysters and mussels.
column and transferring them to benthic communities by assimilation in Similarly, Ray and Fulweiler (2021) did not find significant differences
their tissue or the release of faeces and pseudofaeces (termed bio­ in the effects of two species of oysters (Crassostrea virginica and C. gigas)
deposits). Moreover, biodeposits that settle on the surface of surround­ on nutrient cycling. Meta-analysis can help reconcile these seemingly
ing sediments are rich in organic carbon (C), N and P. They promote contrasting results and understand whether these differences in nutrient
sediment microbial metabolism, thus increasing oxygen consumption, processing might be due to shellfish taxa or to the different environ­
the release of inorganic nutrients such as ammonia (NH+ 4 ), nitrates mental conditions of each study.
(NO-3) and phosphates (PO3- 4 ) (Christensen et al., 2003; Kellogg et al., Here, we combined qualitative and quantitative reviews to summa­
2013; Newell, 2004). While some of these inorganic nutrients are uti­ rize global knowledge of the effects of reef-forming shellfish, namely
lised by pelagic and benthic primary producers, others can be tempo­ oysters and mussels, on biogeochemical cycles in marine coastal eco­
rarily or permanently removed from the system through sediment burial systems. First, we reviewed the capacity of reef-forming shellfish to
and coupled nitrification-denitrification that produces dinitrogen gas drive denitrification on their shell and in tissues. Second, we (i) assessed
(N2) (Herbert, 1999; Newell et al., 2002). This form of N diffuses into the if reef-forming shellfish significantly alter metabolism and nutrient
atmosphere and thus it is unavailable for primary producers. Further­ cycling in surrounding sediments via meta-analysis; and (ii) determined
more, recent studies have shown that shellfish can enhance denitrifi­ how environmental factors such as grain size (muddy vs sandy sedi­
cation in marine coastal systems by hosting denitrifying bacteria in their ments), seasonality, water body type (i.e. lagoon, distributary channel,
guts and shells (Arfken et al., 2017; Caffrey et al., 2016; Ray et al., 2019; gulf and bay) and tidal position affect the interactions between reef-
Svenningsen et al., 2012). Hence, the cultivation and harvesting of filter forming shellfish and sediment function. We predicted that sediments
feeding bivalves has been recognised as a bioremediation method to surrounding shellfish would have higher sediment oxygen demand
remove excess nutrients (Gifford et al., 2007; Onorevole et al., 2018; (SOD), nutrient regeneration (fluxes releasing to the water column), and
Petersen et al., 2014). Moreover, shellfish restoration strategies aiming nitrogen removal than sediments far from shellfish. In addition, we ex­
to recover the ecosystem service they provide are burgeoning pected that different shellfish taxa would similarly affect benthic
worldwide. metabolism and nutrient cycling in surrounding sediments, as observed
Even though shellfish generally promote efficient nutrient cycling in Barrett et al. (2022) and Ray and Fulweiler (2021).
and enhance denitrification as outlined above, there is also some evi­
dence showing the opposite. For example, in hypoxic areas where the 2. Methods
exchange of water is low or with limited capacity to process nutrient
inputs, the introduction of reef-forming shellfish has been shown to 2.1. Literature search and inclusion criteria
contribute to further sediment eutrophication and overwhelm the sys­
tem by directly excreting NH+ 4 or enhancing its release from organic- We reviewed literature on the role of reef-forming shellfish (mussels
enriched sediments through the addition of biodeposits (Erler et al., and oysters) on benthic metabolism and nutrient cycling in marine
2017; Hoellein and Zarnoch, 2014; Lee et al., 2011; Mortazavi et al., systems using Scopus and Web of Science. The search terms included
2015). Moreover, in these areas where the water resident time is high, (‘oxygen*, SOD*, sediment-oxygen-demand*, oxygen-flux*, benthic-
the presence of extensive and densely packed shellfish (e.g. intensive metabolism*, nutrient*, nutrient-cycl*, nutrient-flux*’) and (‘shellfish*,
farms) can deplete not only the resultant phytoplankton but also mussel*, bivalve*, oyster*’) in (‘coast*, estuary*, marine*, lagoon*,
planktonic larvae biomass of numerous species, thereby reducing the gulf*’) not (‘freshwater*’). Searches were conducted during June
functioning and recruitment of wild populations while affecting benthic 2020–June 2021. Reference lists of these studies were checked to
communities due to excessive loading of organic material in the sur­ include articles not recovered by our literature search (i.e. theses or
rounding sediments (Jiang and Gibbs, 2005; McKindsey et al., 2011; dissertations). After excluding results from unrelated research areas and
Smaal et al., 2013). Thus, in order to introduce or re-introduce shellfish removing duplicate articles, we obtained 3773 records (Fig. S1). We
into coastal systems sustainably (where the ecological carrying capacity then screened titles and abstracts to select articles that evaluated the
of the area will not be exceeded), it is important that restoration and effects of shellfish on (i) benthic metabolism (SOD) as a measure of
aquaculture managers consider shellfish density/size as well as biological aerobic metabolism and chemical oxidation in the sediment
site-specific characteristics (e.g. nutrient levels, natural primary pro­ (Wang, 1980); and/or (ii) nutrient cycling as inorganic nutrient fluxes
ductivity, and hydrodynamic conditions). So far, no studies have - 3-
between sediments and water column (NH+ 4 , NO3, PO4 , N2 products
directly assessed the effects of environmental context on the capacity of from bacterial aerobic and anaerobic metabolism). As a result, 81 re­
shellfish to modulate nutrient cycling (Ray and Fulweiler, 2021). We cords were selected, including 53 studies that compared benthic pro­
propose that local conditions that affect sediment stabilisation and cesses in sediments surrounding shellfish reefs and aquaculture farms
oxygenation, such as wave exposure (Almroth et al., 2009), tidal regime with sediments far from them, and 22 studies that evaluated nutrient
(Joye, 1993; Laima et al., 2002), temperature (Herbert, 1999; Thamdrup cycling on the shellfish (Fig. S1).
et al., 1998) and sediment grain size (Benelli et al., 2019; Serpetti et al., In subsequent filtering, studies were excluded if fluxes were calcu­
2016), can affect the interactions between shellfish and surrounding lated without using incubation methods (i.e. Fick’s equation). This is
sediments, affecting nutrient cycling. Consequently, to support the because modelled fluxes only consider the diffusion coefficient and do
development of successful bioremediation tools, it is necessary to not account for volumes and sampling times that provide more accurate
disentangle environmental factors that influence how reef-forming measures for cross-study comparisons (Harmesa and Wahyudi, 2020;
shellfish affect nutrient cycling. Roth et al., 2019). We also excluded studies that measured denitrifica­
Moreover, the degree of efficiency with which reef-forming shellfish tion using the Isotope Pair Technique (IPT) method, as it requires the
process nutrients might also depend on taxonomic variation. For addition of N isotopes and their rates are generally underestimated (Eyre
example, Tenore and Dunstan (1973) showed increased filtration and et al., 2002; Ray et al., 2021). When several studies were based on the

2
G. Filippini et al. Environmental Pollution 316 (2023) 120614

same data, duplicate data was excluded. This resulted in a total of 18 if effects were driven by any single data point (Borenstein et al., 2021).
studies evaluating nutrient cycling on the shellfish reefs (Appendix A). Funnel plots indicated no publication bias for all variables assessed
However, due to the lack of consistency in the controls used between (Fig. S2). Furthermore, results from sensitivity analysis conducted for
studies (e.g. water only, bare sediments, dead oysters, shell biofilm), each response variable showed that results reflected patterns from a
meta-analyses were not possible. For this reason, we assessed general series of datapoints, rather than being influenced by single datapoints
patterns of nutrient cycling on shellfish by calculating the proportion of (Table S1).
studies reporting influx or efflux for each variable considered (SOD, We tested how the different shellfish species, functional group
- 3-
NH+4 , NO3, PO4 , N2) and mean fluxes to assess shellfish contribution to (oyster and mussel) and habitat type (farm vs reef) affected SOD and
nutrient cycling. Means were obtained from comparable studies that nutrient fluxes in surrounding sediments with a multi-level model using
measured fluxes in μmol m− 2 h− 1 (the most common unit reported), function ‘rma.mv’ from the metaphor package (Viechtbauer, 2010). Each
with values reported in μmol g− 1 h− 1 or in μmol individ− 1 h− 1 excluded. model included a fixed factor (shellfish species = 6 levels; functional
In contrast, there was a general agreement between studies on the group = 2 levels; and habitat type = 2 levels) and a random factor
control used when assessing biogeochemistry on sediments around (‘trial’) nested in ‘study identity’ to account for non-independence of
reefs, as most of them included sediments far from reefs as controls. multiple data points extracted from the same study (Noble et al., 2017).
Hence, meta-analyses were done on 32 studies (Appendix A) after Models were run without the intercept to test whether there were sig­
removing studies where: (a) a control site was not present or was not nificant differences between treatment and controls at each level of the
appropriate (i.e. presence of macroalgae/seagrass on the sediment); (b) factor considered. To evaluate for statistically significant differences
sediments were manipulated with organic material or biodeposit addi­ among shellfish species, functional groups and habitat types, we con­
tions; (c) replicates were not present; and/or (d) data were reported as a ducted a post-hoc test by using the function ‘ghlt’ in multcomp package
correlation or were not readable and/or extractable. (Hothorn et al., 2008). Specifications of the models used are summarised
in equation S1.
2.2. Data extraction for meta-analysis of benthic processes in sediments We then assessed if these patterns were affected by the environ­
surrounding shellfish mental context, including sediment grain size (sandy vs muddy), sea­
sonality (summer, autumn, winter and spring), water body type
For each of the 32 studies in the meta-analyses, we extracted sample (distributary channel, lagoon, and gulf/bay), and tidal zone (intertidal
size, mean and standard deviation (SD) for oxygen and nutrient fluxes in and subtidal) by adding them to the models described above - (Equation
sediments surrounding shellfish (treatment) and in control sediments at S2). Interactions between environmental factors and shellfish taxa and
least 15 m away from them. When data such as means and errors were habitat types could not be investigated due to insufficient data. To
not reported in the text or tables of the study, we extracted those from measure how seasons might affect SOD and nutrient cycling in sedi­
plots using WebPlotDigitizer version 4.4. If a study had multiple mea­ ments surrounding shellfish, studies conducted in locations charac­
sures through time or space, we extracted them all (the effect of multiple terised by only two seasons (i.e. tropics) were excluded. Since tropical
measures extracted from the same study was accounted for in the models locations do not present the same four seasons as temperate areas, but
as specified below). If a study reported a range of sample sizes, we used a rather only wet and dry, it was deemed inappropriate to include studies
conservative approach by selecting the lower sample size reported (e.g. conducted in tropical areas with those from temperate regions. More­
if “n = 3–4”, we used 3); while, if a study did not specify sample size, we over, an assessment of the effect of seasons in tropical zones was not
assumed a sample size equal to 2 (minimum number of replicates for possible due to a low number (two) of such studies. Orchard plots were
calculating mean and error). All nutrient fluxes were converted to μmol used to visualise effect sizes of different categories within each envi­
m− 2 h− 1 to allow comparisons between studies. In addition, all standard ronmental moderator using function ‘orchard_plot’ through the OrchaRd
errors (SE) were converted to SD. package (Noble, 2021). As there were no differences between shellfish
To determine how environmental factors affect the interactions be­ species and functional groups or between habitat types (except for NH+ 4
tween shellfish and sediment function, we also recorded for each study fluxes; see results below), effects of environmental factors on the SOD
the (i) sediment type (muddy or sandy) – i.e. when mud or sand is ≥ 50 and nutrient fluxes were analysed combining all taxonomic groups.
weight percent of the sediment, (ii) the season when the experiment Post-hoc tests showed that there were differences between habitats only
took place, (iii) water body type of the system (distributary channel, for NH+ 4 fluxes, but as studies assessing NH4 fluxes were mostly done in
+

lagoon, and gulf/bay), and (iv) position relative to tidal height (inter­ farms (more details in results), it was not possible to test for moderators
tidal and subtidal) – where available. Based on the morphological for each habitat type. Hence, they were also combined.
characteristics of the water bodies in the analysed studies, we defined
and grouped them into the following categories: (a) distributary channel 3. Results
- water system connected to the sea/ocean through a river/s; (b) lagoon -
coastal water body separated, intermittently, or constantly connected to The majority of studies (94%) included in the qualitative and meta-
the sea through one or more narrow mouth/s; and (c) gulf and bay – analysis were within temperate areas, with only three studies reported
coastal area of ocean/sea partly surrounded by land. The categorization from tropical zones (French Polynesia). Results showed geographical
employed was based on a visual recognition using Google Maps. When bias as most studies were conducted in the Northern hemisphere (85%),
general information about the sediment type or specifics of grain size with the majority of these being from North America (61%) and Europe
distribution was not reported in the studies, we characterised the site (34%, Fig. S3). Of the studies measuring fluxes on shellfish, 52% focused
from information in other studies done at the same location. on oysters and the remaining 48% on mussels. Similarly, 57% of studies
For each data point extracted, we used Hedge’s standard mean dif­ assessing fluxes on sediments around shellfish were associated with
ference (SDM) to estimate effect sizes and variance of fluxes (Hedges, oysters (70% of which were conducted in North America), and 43% with
1981). Forest plots were constructed for each response variable using mussels (of which 60% were undertaken in Europe, Fig. S4). Overall,
the ‘forest’ function in metafor package (Viechtbauer, 2010) to show the C. virginica and Mytilus galloprovincialis were the most studied species,
individual effect size and sample variance of each study. All analyses making up 35% and 15% of the studies respectively. These were fol­
were done in R version 4.1.0. lowed by C. gigas and Mytilus edulis (both representing 14% of studies,
We assessed publication bias for each response variable by checking Fig. S3). While 70% of the studies were undertaken in shellfish aqua­
whether there was a significant correlation between effect sizes (SMD) culture sites, the remainder were completed in natural or restored
and sample sizes using funnel plots (Zwetsloot et al., 2017). Sensitivity shellfish reefs. Interestingly, more than 67% of the studies assessing
- 3-
analyses were also performed using the ‘leave-one-out’ method to detect sediment oxygen demand, NH+ 4 , NO3 and PO4 fluxes were done in

3
G. Filippini et al. Environmental Pollution 316 (2023) 120614

aquaculture farms, while 79% of studies assessing denitrification were were similar to sediments far from the reefs (− 0.09 ± 0.61; p-value =
done on shellfish reefs (Fig. S4). 0.881). Although differences between shellfish aquaculture and reefs
were found for this nutrient (Table S5), moderators were assessed by
3.1. Benthic metabolism and nutrient fluxes on shellfish reefs and in pooling the data together since there were only five studies found for
surrounding sediments reefs (Fig. S4).

More than 70% of the 18 studies in our qualitative review that 3.2. Environmental moderators of shellfish effects on benthic metabolism
evaluated nutrient cycling on shellfish measured metabolism (oxygen and nutrient fluxes in surrounding sediments
consumption) and the excretion of NH+ 4 . Oxygen consumption had a
magnitude of 2213.69 ± 480.94 μmol m− 2 h− 1. About 30% of these Sediments surrounding shellfish reefs and farms had greater SOD and
studies considered the efflux and influx of NO-3 and PO34 , while 20%
-
release of NH+4 than control sediments in sandy and muddy sediments,
focused on N2 (Table S2). All 18 studies reported positive fluxes of NH+ 4, intertidal and subtidal zones and seasons (Figs. 2 and 3, Table 1).
and N2 from shellfish (i.e. efflux, or release from shellfish to water col­ However, results varied between water body types. Both SOD and
umn), at 842.76 ± 174.82 and 302 ± 125.90 μmol N2 m− 2 h− 1 respec­ release of NH+ 4 in sediments surrounding shellfish were significantly
tively. In contrast, there were mixed results for NO-3, with half the greater than controls in lagoons and gulf/bays, but not in distributary
datapoints (4 out of 8) indicating an influx, and the other half reporting channels, although these differences might be influenced by a lack for
an efflux (Table S3). Mean and SD calculations for NO-3 and PO3- 4 were power in distributary channels given the lower number of studies per­
not possible due to the different units of measurement used between formed in this water body (Figs. 2 and 3, Table 1). By contrast, NO-3
studies (i.e. some standardised fluxes by habitat area or volume, while influxes in sediments near and far from shellfish were similar in all the
environmental contexts (Fig. 4, Table 1). Moreover, the release of PO34
-
some others by dry weight).
From the 32 studies considered for the meta-analyses, more than was found to be greater in sediments surrounding shellfish than control
40% assessed benthic metabolism and the release of NH+ 4 on sediments sediments regardless of sediment type or tidal zone (Fig. 5, Table 1), but
surrounding shellfish reefs and farms, about 30% measured NO-3 and only in summer and autumn and in lagoon and gulf/bay systems
PO34 fluxes, and 25% evaluated denitrification (Table S4). Overall, meta-
-
(although one study was found in a distributary channel, Fig. 5, Table 1).
3-
analyses showed that SOD and NH+ 4 , PO4 and N2 effluxes were greater in N2 effluxes were 0.68 ± 0.32 times greater in sandy sediments sur­
sediments near than far from shellfish (Fig. 1, Table 1). Specifically, rounding oysters than control sediments, but did not differ in muddy
sediments surrounding shellfish had 1.38 ± 0.19 times greater SOD than sediments (Fig. 6, Table 1). The release of N2 from sediments sur­
3-
control sediments. NH+ 4 , PO4 and N2 effluxes were 1.49 ± 0.29, 1.26 ± rounding oysters was 0.99 ± 0.39 times greater than controls in sum­
0.28, 0.59 ± 0.25 times greater in sediments close to than those far from mer, while no differences were found during the other seasons (Fig. 6,
shellfish reefs/farms. By contrast, NO-3 was taken up at similar rates by Table 1). N2 effluxes from sediments around oysters were also greater
sediments near and far from shellfish (Fig. 1; Table 1). Post-hoc tests did than control sites in lagoons and intertidal systems (Fig. 6, Table 1).
not show differences between different shellfish species and functional However, the lack of effects for many of these groups might be due to the
groups (oysters and mussels, Table S5); hence the assessment of mod­ low number of studies available.
erators was done for all taxa combined. On the other hand, differences
between shellfish farms and reefs were only found for NH+ 4 fluxes 4. Discussion
(Table S5). Specifically, the release of NH+ 4 from sediments surrounding
shellfish aquaculture was significantly higher than control sites (1.80 ± The continuing eutrophication of coastal marine ecosystems is a
0.29; p-value < 0.001); while in sediments surrounding reefs NH+ 4 fluxes global issue that needs to be addressed with urgency. Biogenic habitats

- 3-
Fig. 1. Effect size (SMD ± CI) of sediment oxygen demand (SOD), ammonia (NH+ 4 ) flux, nitrate (NO3) flux, phosphate (PO4 ), dinitrogen gas (N2) in sediments
surrounding shellfish (oyster only studies were found for N2 fluxes) when compared to sediments far from shellfish. Dotted line shows an effect size equal to 0. Only p-
values that are statistically significant (p-value < 0.05) are shown in bold.

4
G. Filippini et al. Environmental Pollution 316 (2023) 120614

have been shown to play an essential role in nutrient processing through

Meta-analysis comparing benthic metabolism and nutrient fluxes between sediments near and far from shellfish from all studies (overall effect size) and at different sediment type, season, water body type, and tidal type,
4 ), nitrate (NO3 ), phosphate (PO4 ), and dinitrogen gas (N2) fluxes. Distributary

z-value
assimilation (filter feeding/roots absorption), burial and enhanced

− 0.41
2.32

2.16

2.54

2.21

2.53
0.86

1.08
0.40

0.37

0.71
0.00
denitrification. For example, mangrove forests were shown to release N2
from 1.9 to 9.25 μmol N2 m− 2 h− 1 (Alongi, 1996; Kristensen et al., 1998;
Rivera-Monroy et al., 1995). Higher denitrification rates (> 50 μmol N2

0.25
0.49
0.32
0.58
0.56
0.39
0.55
0.32
0.64
1.24
0.90
0.27
m− 2 h− 1) have also been found in salt marsh and seagrass environments

SE
(Aoki and McGlathery, 2018; Caffrey et al., 2007; Garcias-Bonet et al.,
Estimate 2018; Sousa et al., 2012; Velinsky et al., 2017). In this study, we esti­

− 0.38
mated that denitrification rates on shellfish can be at least an order of
0.59
0.42
0.68
0.63
0.23
0.99
0.20
0.70
0.45
0.00

0.69
magnitude greater than those observed in many other biogenic habitats,
N2

with a mean of 302 ± 125.90 μmol N2 m− 2 h− 1. Likewise, we estimated

that the presence of oysters stimulates the release of up to 60% more N2
z-value

4.46
3.56
2.49

3.53
2.91
3.78

2.08
3.50
2.63
from the surrounding sediments, similar to that found in a previous
1.49
1.91

0.13
study (Ray and Fulweiler, 2021). The increases in SOD by 1.38 times
close to shellfish are evidence of a surge in aerobic microbial processes,
3-

supporting the mentioned increases in nutrient regeneration. These re­

0.28
0.39
0.44
0.66
0.75
0.43
0.51
0.37
1.34
0.55
0.33
0.58
SE

sults highlight that shellfish are important for removing N from coastal
ecosystems because denitrification can be enhanced both by microbes
on shellfish and by shellfish stimulating microbial activity in nearby
Estimate

1.26
1.37
1.01
0.98
1.44
1.53
1.49
1.39
0.18
1.14
1.17
1.52

sediments.
PO3-
4
−

Interestingly, qualitative results showed that shellfish can uptake

large amounts of oxygen (~2000 μmol m− 2 h− 1) and excrete NH+ 4 at
~800 μmol m− 2 h− 1. Moreover, they can increase the release of NH+ 4 and
z-value

− 0.77
− 1.09

− 0.27
− 0.90
− 1.45
− 0.17

− 1.05
0.03
0.21
0.95

0.51
0.56

PO34 from sediments to the water column by up to 1.49 times. If optimal

-

biological and environmental conditions exist (i.e. an aerobic environ­

including showing estimate (Hedge’s g), standard error (SE), and z-value for variables of sediment oxygen demand (SOD), ammonia (NH+

-
ment adjacent to an anaerobic one), the generated NH+ 4 /NO3 is con­
0.34
0.45
0.51
1.73
0.06
0.51
0.64
0.44
1.62
0.49
0.47
0.47

verted into N2 by coupled nitrification-denitrification reactions (Kellogg

SE

et al., 2014; Newell, 2004). Inorganic N that is not transformed will

diffuse into the water column, where it can drive phytoplankton growth
Estimate

and could potentially lead to algal blooms, eutrophication and anoxic

− 0.26
− 0.49

− 0.14
− 0.57
− 0.64
− 0.28

− 0.49
0.02
0.36
0.04

0.25
0.26

events, especially in systems where the mixing is limited or where the

NO-3

oxygen is already somewhat depleted. Previous studies have shown that

shellfish may be able to counter this increased phytoplankton biomass
z-value

through filtration (Cranford et al., 2011; Soto and Mena, 1999), but this
5.21
4.25
3.30
3.41
2.50
4.00
2.78
3.13

4.18
5.49
2.34
0.95

top-down control can have both positive and negative impacts on

ecosystem services depending on site-specific characteristics (Smaal
et al., 2019; van der Schatte Olivier et al., 2020). Positive effects are
0.29
0.38
0.41
0.69
0.50
0.37
0.48
0.38
1.29
0.47
0.37
0.44

likely to occur at optimal shellfish density in more enriched and shallow

SE

areas, where shellfish filtration can act as a biofilter mitigating eutro­

phication and improving water quality (Jansen et al., 2019; Lindahl,
Estimate

2011; Schroder et al., 2014). By reducing turbidity and increasing light

1.49
1.61
1.36
2.34
1.25
1.48
1.32
1.20
1.23
1.97
2.01
1.03
NH+
4

penetration, shellfish can enhance benthic productivity of important

habitats like seagrass (Sharma et al., 2016; Smyth et al., 2018; Wall
et al., 2008), which in turn improves water quality (by reducing the
stands for distributary channel. Significant differences are outlined in bold.

z-value

7.70
4.48
6.57
2.27
3.92
4.86
3.32
5.83

5.29
5.57
5.25

availability of nutrients for phytoplankton) and provides food and

0.71

habitat for many species (Newell et al., 2002; Petersen et al., 2019;
Sandoval-Gil et al., 2016; Zertuche-González et al., 2009). Additionally,
through the filtration of phytoplankton, shellfish accumulate nutrients
0.18
0.24
0.26
0.50
0.39
0.29
0.38
0.23
0.65
0.31
0.25
0.27
SE

(N and P) that can temporarily be subtracted and permanently removed

from coastal environments if/when harvested (Carmichael et al., 2012;
Cerco and Noel, 2007). Interestingly, a recent study has shown that
Estimate

1.38
1.10
1.69
1.13
1.52
1.41
1.22
1.35
0.46
1.63
1.36
1.41

shellfish can remove 314–581 kg N ha− 1 yr− 1 via harvesting, which

SOD

provides a higher contribution in the reduction of N than sediment

denitrification beneath farms (Barrett et al., 2022). On the other hand, in
oligotrophic systems or in areas with greater flushing (highly hydrody­
Overall effect size

namic systems), large densities of shellfish can extremely deplete

Distributary

phytoplankton, resulting in an increase of non-filtered primary pro­

Intertidal
Gulf/Bay
Summer

Subtidal
Autumn

ducers (picoplankton) and a reduction of food for themselves and for

Lagoon
Muddy

Winter
Spring
Sandy

other filter-feeders (Grant and Pastres, 2019; Smaal et al., 2013). This
would therefore shift benthic communities from filter to deposit feeders,
while reducing recruitment of wild shellfish and other species (Cranford,
Water body type

2019; Gibbs, 2007; van Stralen and Dijkema, 1994). Furthermore, Fer­
Sediment type

riss et al. (2019) indicated that shellfish aquaculture can reduce seagrass
Seasonality

Tidal type

density, biomass and reproduction in areas directly underneath and

Table 1

immediately surrounding farms, mainly due to space and light limita­

tion. Thus, to achieve successful bioremediation, it is essential to

5
G. Filippini et al. Environmental Pollution 316 (2023) 120614

Fig. 2. Effect size (SMD ± CI and Prediction interval) of sediment oxygen demand (SOD) in sediments surrounding shellfish when compared to sediments far from
shellfish for the four selected environmental factors: (a) sediment type, (b) seasonality, (c) water body type, and (d) tidal type. Dotted line shows an effect size equal
to 0. Individual effect size points on the right side of the dotted line indicates a greater SOD from the sediments surrounding shellfish, while individual effect size
points on the left side suggests a less SOD. Point estimates, 95% confidence intervals and 95% prediction intervals are represented by shaded points, bold black error
bars and black horizontal lines. Only p-values that are statistically significant (p-value < 0.05) are shown in bold.

consider environmental characteristics of the site and set a threshold on shellfish reefs. Similarly, the majority of studies focusing on benthic
metabolism, NO-3 and PO34 fluxes were undertaken in shellfish aqua­
-
the scale of restoration/introduction (i.e. the correct density of shellfish)
that will positively affect interactions between shellfish and surrounding culture and in oyster reefs for denitrification. Thus, we suggest that
habitats. studies assessing benthic metabolism and nutrient cycling in both
While restored shellfish reefs and aquaculture farms may have shellfish aquaculture and reefs need to expand the variables measured
different impacts on sediment functions due to their structural and (including shellfish density and reef/farm size) to understand the dif­
ecological differences, we found that this was only true for NH+ 4 fluxes. ferences between them.
Findings indicated that aquaculture farms increased NH+ 4 fluxes in sur­ Although oyster and mussel species differ in their capacity to ingest,
rounding sediments, while this was not observed for shellfish reefs. This assimilate and reject various quantities and qualities of food particles,
difference might be due to shading of farms preventing the recycle of resulting in different compositions of their biodeposits (Grenz, 1989;
nutrients via primary productivity, and/or to higher shellfish density Hawkins et al., 1998; Tenore and Dunstan, 1973), no differences were
and size in aquaculture sites. As mentioned, a larger density of shellfish detected in benthic metabolism and nutrient regeneration of sediments
might lead to overgrazing of phytoplankton, which can result in a higher surrounding oyster and mussel species. This result is in accordance with
deposition of organic material in the surrounding sediments (Weitzman, our hypothesis and with two previous studies which indicated that
3-
2019). The large accumulation of biodeposits can result in higher NH+ 4 sediment effluxes of NH+ 4 and PO4 were comparable among oyster
release through mineralisation as well as depletion of oxygen and species, and that cultivated oysters and mussels similarly enhance
release of H2S near and under shellfish farms (Callier et al., 2009; sediment denitrification (Barrett et al., 2022; Ray and Fulweiler, 2021).
McKindsey et al., 2011). It should be noted that these results might be However, we could not assess whether mussels would affect denitrifi­
due to a low power to detect effects of reefs, given that only a small cation in the same way. This is because mussel studies typically estimate
number of studies (n = 5) addressing NH+ 4 cycling were conducted in N2 fluxes using the IPT method, which produces results that are not

6
G. Filippini et al. Environmental Pollution 316 (2023) 120614

Fig. 3. Effect size (SMD ± CI and Prediction interval) of ammonia (NH+ 4 ) flux in sediments surrounding shellfish when compared to sediments far from shellfish for
the four selected environmental factors: (a) sediment type, (b) seasonality, (c) water body type, and (d) tidal type. Further details as in Fig. 2.

comparable to the N2/Ar technique (Eyre et al., 2002). Despite this, our However, these variables were also found to be significantly higher in
results suggest that the introduction of either mussels or oysters into more open systems (i.e. gulf and bay), suggesting that the presence of
degraded systems can enhance benthic metabolism and nutrient cycling shellfish affects nutrient fluxes in sediments at a range of hydrodynamic
at similar rates. conditions. Similarly, results from this study did not show differences
3-
It is important to note that these results may have been influenced by between tidal level for SOD and nutrient fluxes (NH+ 4 and PO4 ), high­
geographical location since the majority of studies on mussels and lighting that the presence of shellfish can have a significant effect on
oysters were undertaken in the Northern hemisphere (in Europe and sediment functions in both intertidal and subtidal systems. Thus, even if
North America respectively), with minimal research conducted in other sediments in intertidal systems are subjected to greater environmental
countries such as Australia and South Africa. Moreover, Chesapeake pressure due to drastic changes in light intensity, temperature, salinity,
Bay, Baltic Sea and Northern Adriatic Sea (where most of the studies oxygen availability, or hydrodynamics during high and low tides (Hou
were completed) are among the coastal areas most impacted by nutrient et al., 2007; Nielsen et al., 1990; Rysgaard et al., 1999), the increase in
loadings in the world (Malone and Newton, 2020). While the presence of organic material through shellfish biodeposits might still promote mi­
shellfish in these areas has been shown to significantly change nutrient crobial metabolism and nutrient regeneration in these systems. Alter­
cycling, this may not always be the case for areas characterised by fewer nately, this increase in SOD and nutrient fluxes could be due to excretion
nutrient inputs (Jansen et al., 2012; Stenton-Dozey et al., 2011). and bioturbation activity of an increased infauna presence (such as de­
Therefore, research focusing on how benthic metabolism and nutrient posit feeding polychaetes) in sediments around shellfish (Bugnot et al.,
cycling are affected by either mussels or oysters under a variety of 2022; Giles et al., 2006). In terms of the seasonal differences observed in
environmental conditions is needed to shed light on the global relevance nutrient fluxes (release of PO3- 4 and N2), results showed expected pat­
of these results. terns, as benthic metabolism and nutrient cycling are influenced by the
As expected, SOD and nutrient regeneration (releases of NH+ 4 and increase in water temperatures during warmer seasons (Akomeah and
3-
PO4 ) increased considerably in lower energy systems (i.e. lagoons), Lindenschmidt, 2017; Malinverno and Martinez, 2015). On the other
probably due to the higher accumulation of biodeposits in the sur­ hand, our results have shown that shellfish increased SOD and the
rounding sediments (Mazouni et al., 1996; Nizzoli et al., 2005). release of NH+ 4 from surrounding sediments in all seasons. Thus, this

7
G. Filippini et al. Environmental Pollution 316 (2023) 120614

Fig. 4. Effect size (SMD ± CI and Prediction interval) of nitrate (NO-3) flux in sediments surrounding shellfish when compared to sediments far from shellfish for the
four selected environmental factors: (a) sediment type, (b) seasonality, (c) water body type, and (d) tidal type. Further details as in Fig. 2.

result might suggest that mineralisation of organic material in sediments better identify environmental conditions that can enhance bioremedia­
around shellfish might be more influenced by physico-chemical vari­ tion. Additionally, as we still know little of the mechanisms by which
ables other than temperature (i.e. quality, quantity of OM or bio­ shellfish affect surrounding sediments (if it is via biodeposit additions or
deposits) or by the presence of infauna found to be greater in sediments if their presence changes microbial biodiversity), we suggest that these
around shellfish (Bugnot et al., 2022; Giles et al., 2006). analyses should be linked with microbial communities and gene markers
Meta-analysis results showed that oysters enhance denitrification in of environmental DNA to fully understand how shellfish affect nutrient
sandy lagoons. These findings are not surprising as the deposition of cycling (Lindemann et al., 2016; Thomsen and Willerslev, 2015).
labile organic matter via biodeposits in sediments can be enhanced in In summary, this is the first study showing that, overall, environ­
more enclosed systems, where the mixing and resuspension of organic mental factors such as season, sediment composition and water body
material is limited (Hartstein and Stevens, 2005; Humphries et al., type can modulate the effects of shellfish reefs on sediment function,
2016). Moreover, sandy sediments are generally characterised by high with consequences for biogeochemical cycling. Moreover, the effects of
permeability and great oxygenation, hence the deposition of biodeposits shellfish on sediment function depend on shellfish density, reef/farm
can stimulate microbial metabolism and subsequently decrease size and other site-specific conditions which can tip the scale towards
oxygenation, generating favourable oxic/anoxic transitions for deni­ negative effects on ecosystem services. Hence, in order to efficiently use
trifying processes (Ahmerkamp et al., 2017; Cornwell et al., 1999). shellfish to remediate excess nutrients and improve water quality, these
Although denitrification in muddy sediments and other water bodies parameters need to be considered and further studies ae needed to
considered (gulf/bay and distributary channel) presented a lower establish thresholds in shellfish densities and restoration/farm size.
number of studies (potentially masking evidence of denitrification in Currently, site selection procedures for marine ecosystem restoration
these environmental context), results suggest that introducing oysters have been limited to consider the environmental context suitable for the
into lagoons or other systems with sandy sediments can efficiently survival of the restored organism (Beseres Pollack et al., 2013; Lester
remove N from these environments, thereby improving water quality. et al., 2020). However, the results of this study showed that the recovery
Future research could consider variables such as shellfish size, density, of ecosystem services such as nutrient regeneration depends on envi­
reef/culture age, light intensity, and water quality/nutrient gradients to ronmental context, hence site selection should also include information

8
G. Filippini et al. Environmental Pollution 316 (2023) 120614

Fig. 5. Effect size (SMD ± CI and Prediction interval) of phosphate (PO3-

4 ) flux in sediments surrounding shellfish when compared to sediments far from shellfish for
the four selected environmental factors: (a) sediment type, (b) seasonality, (c) water body type, and (d) tidal type. Further details as in Fig. 2.

3-
about the environmental conditions that maximise the desired efflux (NH+4 and PO4 ), and denitrification (N2) in the surrounding sed­
ecosystem services. Here, we propose that projects aiming to maximise iments. Although, the success of this tool does not necessarily depend on
water quality and N removal delivered by shellfish should include environmental context (i.e. sediments and water body types) with
3- -
nutrient levels (e.g. high levels of natural primary productivity), sedi­ regards to SOD and nutrient cycling (NH+ 4 , PO4 and NO3); bioremedi­
ment grain size (e.g. sandy sediments), hydrodynamics (e.g. lagoons ation projects that aim to remove N could improve outcomes by intro­
systems), and depth (e.g. shallow water) for site selection. Similarly, ducing oysters in lagoons and in systems with sandy sediments. We also
planning for areas suitable for the development of aquaculture farms are suggest that the planning of sites and spatial scales of shellfish restora­
limited to assessing areas with low potential for ecological impacts, tion and development of aquaculture farms should consider the envi­
while providing good economic and social benefits (Stelzenmüller et al., ronmental aspects identified in this study to maximise ecosystem
2017; UNESCO-IOC and European Commission, 2021). If we aim to services provided by shellfish (i.e. nutrient remediation and improve­
maximise the co-benefits of shellfish aquaculture, then planning should ment of water quality).
also consider prioritising areas with the environmental conditions that
maximise the delivery of ecosystem services (Gentry et al., 2020). Ul­ Credit author statement
timately, strategies that set clear objectives can plan ahead and incor­
porate the knowledge gathered in this study to maximise the delivery of Giulia Filippini: Conceptualization, Methodology, Formal analysis,
ecosystem services by shellfish. Data curation, Visualization, Writing – original draft, Writing – review &
editing. Katherine A. Dafforn: Conceptualization, Methodology, Super­
5. Conclusions vision, Writing – review & editing. Ana B. Bugnot: Conceptualization,
Methodology, Formal analysis, Supervision, Writing – review & editing.
This study showed that reef-forming shellfish (both oysters and
mussels) can potentially act as bioremediation tools to mitigate eutro­
Funding sources
phication. In the cases analysed in this review and meta-analysis,
shellfish significantly increased oxygen demand, inorganic nutrient
This research did not receive any specific grant from funding

9
G. Filippini et al. Environmental Pollution 316 (2023) 120614

Fig. 6. Effect size (SMD ± CI and Prediction interval) of dinitrogen gas (N2) flux in sediments surrounding shellfish when compared to sediments far from shellfish
for the four selected environmental factors: (a) sediment type, (b) seasonality, (c) water body type, and (d) tidal type. Further details as in Fig. 2.

agencies in the public, commercial, or not-for-profit sectors. GF was Copenhagen) and Dr Nina Schäfer (Sydney Institute of Marine Science)
supported by an iMQRES PhD Scholarship awarded through the School for their help during statistical analysis.
of Natural Sciences, Macquarie University.
Appendix A. Supplementary data

Declaration of competing interest

Supplementary data to this article can be found online at https://doi.
org/10.1016/j.envpol.2022.120614.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
References
the work reported in this paper.
Ahmerkamp, S., Winter, C., Krämer, K., Beer, D.D., Janssen, F., Friedrich, J., Kuypers, M.
Data availability M.M., Holtappels, M., 2017. Regulation of benthic oxygen fluxes in permeable
sediments of the coastal ocean. Limnol. Oceanogr. 62 (5), 1935–1954.
Akomeah, E., Lindenschmidt, K.-E., 2017. Seasonal Variation in Sediment Oxygen
Data will be made available on request. Demand in a Northern Chained River-Lake System. Water 9, 254.
Almroth, E., Tengberg, A., Andersson, J.H., Pakhomova, S., Hall, P.O.J., 2009. Effects of
resuspension on benthic fluxes of oxygen, nutrients, dissolved inorganic carbon, iron
Acknowledgements and manganese in the Gulf of Finland, Baltic Sea. Continent. Shelf Res. 29 (5–6),
807–818.
A special thanks is given to Anne Lyngholm Sørensen (University of

10
G. Filippini et al. Environmental Pollution 316 (2023) 120614

Alongi, D.M., 1996. The dynamics of benthic nutrient pools and fluxes in tropical Grenz, C., 1989. Quantification et destinée de la biodéposition en zones de production
mangrove forests. J. Mar. Res. 54, 123–148. conchylicole intensive en Méditerranée.
Anderson, D.M., Glibert, P.M., Burkholder, J.M., 2002. Harmful algal blooms and Gruber, N., Galloway, J.N., 2008. An Earth-system perspective of the global nitrogen
eutrophication: nutrient sources, composition, and consequences. Estuaries 25 (4), cycle. Nature 451 (7176), 293–296.
704–726. Harmesa, H., Wahyudi, A.J., 2020. Two Approaches to Measure Trace Metals Fluxes at
Aoki, L.R., McGlathery, K.J., 2018. Restoration enhances denitrification and DNRA in Sediment-Water Interface: Sediment Porewater Profile and Benthic Incubation.
subsurface sediments of Zostera marina seagrass meadows. Mar. Ecol. Prog. Ser. 602, ASEAN J. Sci. Technol. Dev. 37 (3).
87–102. Hartstein, N.D., Stevens, C.L., 2005. Deposition beneath long-line mussel farms.
Arfken, A., Song, B., Bowman, J.S., Piehler, M., 2017. Denitrification potential of the Aquacult. Eng. 33 (3), 192–213.
eastern oyster microbiome using a 16S rRNA gene based metabolic inference Hawkins, A.J.S., Bayne, B.L., Bougrier, S., Héral, M., Iglesias, J.I.P., Navarro, E.,
approach. PLoS One 12 (9), e0185071. Smith, R.F.M., Urrutia, M.B., 1998. Some general relationships in comparing the
Barrett, L.T., Theuerkauf, S.J., Rose, J.M., Alleway, H.K., Bricker, S.B., Parker, M., feeding physiology of suspension-feeding bivalve molluscs. J. Exp. Mar. Biol. Ecol.
Petrolia, D.R., Jones, R.C., 2022. Sustainable growth of non-fed aquaculture can 219 (1), 87–103.
generate valuable ecosystem benefits. Ecosyst. Serv. 53, 101396. Hedges, L.V., 1981. Distribution theory for glass’s estimator of effect size and related
Beck, M.W., Brumbaugh, R.D., Airoldi, L., Carranza, A., Coen, L.D., Crawford, C., estimators. J. Educ. Stat. 6 (2), 107–128.
Defeo, O., Edgar, G.J., Hancock, B., Kay, M.C., Lenihan, H.S., Luckenbach, M.W., Herbert, R.A., 1999. Nitrogen cycling in coastal marine ecosystems. FEMS (Fed. Eur.
Toropova, C.L., Zhang, G., Guo, X., 2011. Oyster reefs at risk and recommendations Microbiol. Soc.) Microbiol. Rev. 23 (5), 563–590.
for conservation, restoration, and management. J. Biosci. 61 (2), 107–116. Hoellein, T.J., Zarnoch, C.B., 2014. Effect of eastern oysters (Crassostrea virginica) on
Benelli, S., Bartoli, M., Ribaudo, C., Fano, E.A., 2019. Contrasting effects of an alien sediment carbon and nitrogen dynamics in an urban estuary. Ecol. Appl. 24 (2),
worm on benthic N cycling in muddy and sandy sediments. Water 11 (3), 465. 271–286.
Beseres Pollack, J., Yoskowitz, D., Kim, H.C., Montagna, P.A., 2013. Role and value of Hothorn, T., Bretz, F., Westfall, P., 2008. Simultaneous inference in general parametric
nitrogen regulation provided by oysters (Crassostrea virginica) in the Mission-Aransas models. Biometrical Journal 50, 346–363.
Estuary, Texas, USA. PloS one 8 (6), e65314. Hou, L.J., Liu, M., Xu, S.Y., Ou, D.N., Yu, J., Cheng, S.B., Lin, X., Yang, Y., 2007. The
Borenstein, M., Hedges, L.V., Higgins, J.P., Rothstein, H.R., 2021. Introduction to Meta- effects of semi-lunar spring and neap tidal change on nitrification, denitrification
Analysis. John Wiley & Sons. and N2O vertical distribution in the intertidal sediments of the Yangtze estuary,
Bugnot, A.B., Dafforn, K.A., Coleman, R.A., Ramsdale, M., Gibbeson, J.T., Erickson, K., China. Estuar. Coast. Shelf Sci. 73 (3), 607–616.
Vila-Concejo, A., Figueira, W.F., Gribben, P.E., 2022. Linking habitat interactions Humphries, A.T., Ayvazian, S.G., Carey, J.C., Hancock, B.T., Grabbert, S., Cobb, D.,
and biodiversity within seascapes. Ecosphere 13 (4), e4021. Strobel, C.J., Fulweiler, R.W., 2016. Directly measured denitrification reveals oyster
Caffrey, J.M., Hollibaugh, J.T., Mortazavi, B., 2016. Living oysters and their shells as aquaculture and restored oyster reefs remove nitrogen at comparable high rates.
sites of nitrification and denitrification. Mar. Pollut. Bull. 112 (1–2), 86–90. Front. Mar. Sci. 3, 74.
Caffrey, J.M., Murrell, M.C., Wigand, C., McKinney, R., 2007. Effect of nutrient loading Jansen, H.M., Strand, Ø., van Broekhoven, W., Strohmeier, T., Verdegem, M.C., Smaal, A.
on biogeochemical and microbial processes in a New England salt marsh. C., 2019. Feedbacks from filter feeders: review on the role of mussels in cycling and
Biogeochemistry 82 (3), 251–264. storage of nutrients in oligo- meso- and eutrophic cultivation areas. In: Smaal, A.C.,
Callier, M.D., Richard, M., McKindsey, C.W., Archambault, P., Desrosiers, G., 2009. Ferreira, J.G., Grant, J., Petersen, J.K., Strand, Ø. (Eds.), Goods and Services of
Responses of benthic macrofauna and biogeochemical fluxes to various levels of Marine Bivalves: Cham. Springer International Publishing, pp. 143–177.
mussel biodeposition: an in situ “benthocosm” experiment. Mar. Pollut. Bull. 58 Jansen, H.M., Strand, Ø., Verdegem, M., Smaal, A., 2012. Accumulation, release and
(10), 1544–1553. turnover of nutrients (C-N-P-Si) by the blue mussel Mytilus edulis under oligotrophic
Carmichael, R.H., Walton, W., Clark, H., 2012. Bivalve-enhanced nitrogen removal from conditions. J. Exp. Mar. Biol. Ecol. 416, 185–195.
coastal estuaries. Can. J. Fish. Aquat. 69 (7), 1131–1149. Jiang, W., Gibbs, M.T., 2005. Predicting the carrying capacity of bivalve shellfish culture
Cerco, C.F., Noel, M.R., 2007. Can oyster restoration reverse cultural eutrophication in using a steady linear food web model. Aquaculture 244 (1), 171–185.
Chesapeake Bay? Estuar. Coast. 30 (2), 331–343. Joye, S.B., 1993. Spatial and Temporal Patterns of Nitrogen Fixation and Denitrification
Christensen, P.B., Glud, R.N., Dalsgaard, T., Gillespie, P., 2003. Impacts of longline in the Intertidal and Subtidal Sediments of a Mediterranean-type Estuary: Tomales
mussel farming on oxygen and nitrogen dynamics and biological communities of Bay. The University of North Carolina at Chapel Hill.
coastal sediments. Aquaculture 218 (1-4), 567–588. Kellogg, M., Cornwell, J., Owens, M., Paynter, K., 2013. Denitrification and nutrient
Cornwell, J.C., Kemp, W.M., Kana, T.M., 1999. Denitrification in coastal ecosystems: assimilation on a restored oyster reef. Mar. Ecol. Prog. Ser. 480, 1–19.
methods, environmental controls, and ecosystem level controls, a review. Aquat. Kellogg, M.L., Smyth, A.R., Luckenbach, M.W., Carmichael, R.H., Brown, B.L.,
Ecol. 33 (1), 41–54. Cornwell, J.C., Piehler, M.F., Owens, M.S., Dalrymple, D.J., Higgins, C.B., 2014. Use
Cranford, P., Ward, J., Shumway, S., 2011. Bivalve Filter Feeding: Variability and Limits of oysters to mitigate eutrophication in coastal waters. Estuar. Coast Shelf Sci. 151,
of the Aquaculture Biofilter. Shellfish aquaculture and the environment 81–124. 156–168.
Cranford, P.J., 2019. Magnitude and extent of water clarification services provided by Kristensen, E., Jensen, M.H., Banta, G.T., Hansen, K., Holmer, M., King, G.M., 1998.
bivalve suspension feeding. In: Smaal, A.C., Ferreira, J.G., Grant, J., Petersen, J.K., Transformation and transport of inorganic nitrogen in sediments of a southeast Asian
Strand, Ø. (Eds.), Goods and Services of Marine Bivalves: Cham. Springer mangrove forest. Aquat. Microb. Ecol. 15 (2), 165–175.
International Publishing, pp. 119–141. Laima, M., Brossard, D., Sauriau, P.G., Girard, M., Richard, P., Gouleau, D., Joassard, L.,
Duarte, C.M., Krause-Jensen, D., 2018. Intervention options to accelerate ecosystem 2002. The influence of long emersion on biota, ammonium fluxes and nitrification in
recovery from coastal eutrophication. Front. Mar. Sci. 5, 470. intertidal sediments of Marennes-Oléron Bay, France. Mar. Environ. Res. 53 (4),
Erler, D.V., Welsh, D.T., Bennet, W.W., Meziane, T., Hubas, C., Nizzoli, D., Ferguson, A.J. 381–402.
P., 2017. The impact of suspended oyster farming on nitrogen cycling and nitrous Lee, J.S., Kim, Y.T., Shin, K.H., Hyun, J.H., Kim, S.Y., 2011. Benthic nutrient fluxes at
oxide production in a sub-tropical Australian estuary. Estuar. Coast. Shelf Sci 192, longline sea squirt and oyster aquaculture farms and their role in coastal ecosystems.
117–127. Aquacult. Int. 19 (5), 931–944.
Eyre, B., Rysgaard, S., Dalsgaard, T., Christensen, P., 2002. Comparison of isotope Lehtoranta, J., Ekholm, P., Pitkänen, H., 2009. Coastal eutrophication thresholds: a
pairing and N2:Ar methods for measuring sediment denitrification—assumption, matter of sediment microbial processes. Ambio 38 (6), 303–308.
modifications, and implications. Estuar. Coast 25, 1077–1087. Lester, S.E., Dubel, A.K., Hernán, G., McHenry, J., Rassweiler, A., 2020. Spatial planning
Ferriss, B.E., Conway-Cranos, L.L., Sanderson, B.L., Hoberecht, L., 2019. Bivalve Principles for marine ecosystem restoration. Front. Mar. Sci. 7, 328.
aquaculture and eelgrass: a global meta-analysis. Aquaculture 498, 254–262. Lindahl, O., 2011. Mussel Farming as a Tool for Re-eutrophication of Coastal Waters:
Fourqurean, J.W., Kendrick, G.A., Collins, L.S., Chambers, R.M., Vanderklift, M.A., 2012. Experiences from Sweden. Shellfish Aquaculture and the Environment, pp. 217–237.
Carbon, nitrogen and phosphorus storage in subtropical seagrass meadows: examples Lindemann, S., Zarnoch, C.B., Castignetti, D., Hoellein, T.J., 2016. Effect of eastern
from Florida Bay and Shark Bay. Mar. Freshw. Res. 63, 967–983. oysters (Crassostrea virginica) and seasonality on nitrite reductase gene abundance
Garcias-Bonet, N., Fusi, M., Ali, M., Shaw, D.R., Saikaly, P.E., Daffonchio, D., Duarte, C. (nirS, nirK, nrfA) in an urban estuary. Estuar. Coast 39 (1), 218–232.
M., 2018. High denitrification and anaerobic ammonium oxidation contributes to Malinverno, A., Martinez, E.A., 2015. The effect of temperature on organic carbon
net nitrogen loss in a seagrass ecosystem in the central Red Sea. Biogeosciences 15 degradation in marine sediments. Sci. Rep. 5 (1), 17861.
(23), 7333–7346. Malone, T.C., Newton, A., 2020. The globalization of cultural eutrophication in the
Gentry, R.R., Alleway, H.K., Bishop, M.J., Gillies, C.L., Waters, T., Jones, R., 2020. coastal ocean: causes and consequences. Front. Mar. Sci. 670.
Exploring the potential for marine aquaculture to contribute to ecosystem services. Mazouni, N., Gaertner, J.-C., Deslous-Paoli, J.-M., Landrein, S., Geringer
Rev. Aquacult. 12 (2), 499–512. d’Oedenberg, M., 1996. Nutrient and oxygen exchanges at the water–sediment
Gibbs, M.T., 2007. Sustainability performance indicators for suspended bivalve interface in a shellfish farming lagoon (Thau, France). J. Exp. Mar. Biol. Ecol. 205
aquaculture activities. Ecol. Indicat. 7 (1), 94–107. (1), 91–113.
Gifford, S., Dunstan, R.H., O’Connor, W., Koller, C.E., MacFarlane, G.R., 2007. Aquatic McKindsey, C., Archambault, P., Callier, M., Olivier, F., 2011. Influence of suspended and
zooremediation: deploying animals to remediate contaminated aquatic off-bottom mussel culture on the sea bottom and benthic habitats: a review. Can. J.
environments. Trends Biotechnol. 25 (2), 60–65. Zool. 89, 622–646.
Giles, H., Pilditch, C.A., Bell, D.G., 2006. Sedimentation from mussel (Perna canaliculus) Meyer-Reil, L.-A., Köster, M., 2000. Eutrophication of marine waters: effects on benthic
culture in the Firth of Thames, New Zealand: impacts on sediment oxygen and microbial communities. Mar. Pollut. Bull. 41 (1), 255–263.
nutrient fluxes. Aquaculture 261 (1), 125–140. Middelburg, J.J., Soetaert, K., Herman, P.M.J., 1997. Empirical relationships for use in
Grant, J., Pastres, R., 2019. Ecosystem models of bivalve aquaculture: implications for global diagenetic models: deep Sea Research Part I. Oceanographic Research Papers
supporting goods and services. In: Smaal, A.C., Ferreira, J.G., Grant, J., Petersen, J. 44 (2), 327–344.
K., Strand, Ø. (Eds.), Goods and Services of Marine Bivalves: Cham. Springer Mortazavi, B., Ortmann, A.C., Wang, L., Bernard, R.J., Staudhammer, C.L., Dalrymple, J.
International Publishing, pp. 507–525. D., Carmichael, R.H., Kleinhuizen, A.A., 2015. Evaluating the impact of oyster

11
G. Filippini et al. Environmental Pollution 316 (2023) 120614

(Crassostrea virginica) gardening on sediment nitrogen cycling in a subtropical Smyth, A.R., Geraldi, N.R., Thompson, S.P., Piehler, M.F., 2016. Biological activity
estuary. Bull. Mar. Sci. 91 (3), 323–341. exceeds biogenic structure in influencing sediment nitrogen cycling in experimental
Newell, R.I.E., 2004. Ecosystem influences of natural and cultivated populations of oyster reefs. Mar. Ecol. Prog. Ser. 560, 173–183.
suspension-feeding bivalve molluscs: a review. J. Shellfish Res. 23, 51–61. Smyth, A.R., Murphy, A.E., Anderson, I.C., Song, B., 2018. Differential effects of bivalves
Newell, R.I.E., Cornwell, J.C., Owens, M.S., 2002. Influence of simulated bivalve on sediment nitrogen cycling in a shallow coastal bay. Estuar. Coast 41 (4),
biodeposition and microphytobenthos on sediment nitrogen dynamics. A laboratory 1147–1163.
study: Limnol. Oceanogr. 47 (5), 1367–1379. Soto, D., Mena, G., 1999. Filter Feeding by the Freshwater Mussel, Diplodon Chilensis, as
Nielsen, L.P., Christensen, P.B., Revsbech, N.P., Sørensen, J., 1990. Denitrification and a Biocontrol of Salmon Farming Eutrophication: Aquaculture, vol. 171, pp. 65–81, 1.
oxygen respiration in biofilms studied with a microsensor for nitrous oxide and Sousa, A.I., Lillebø, A.I., Risgaard-Petersen, N., Pardal, M.A., Caçador, I., 2012.
oxygen. Microb. Ecol. 19 (1), 63–72. Denitrification: an ecosystem service provided by salt marshes. Mar. Ecol. Prog. Ser.
Nixon, S.W., 1981. Remineralization and nutrient cycling in coastal marine ecosystems. 448, 79–92.
In: Neilson, B.J., Cronin, L.E. (Eds.), Estuaries and Nutrients. Humana Press, Totowa, Steffen, W., Richardson, K., Rockström, J., Cornell, S.E., Fetzer, I., Bennett, E.M.,
NJ, pp. 111–138. Biggs, R., Carpenter, S.R., Vries, W.d., Wit, C.A.d., Folke, C., Gerten, D., Heinke, J.,
Nixon, 1995. Coastal marine eutrophication: a definition, social causes, and future Mace, G.M., Persson, L.M., Ramanathan, V., Reyers, B., Sörlin, S., 2015. Planetary
concerns. Ophelia 41 (1), 199–219. boundaries: guiding human development on a changing planet. Science 347 (6223),
Nizzoli, D., Welsh, D., Bartoli, M., Viaroli, P., 2005. Impacts of Mussel (Mytilus 1259855.
galloprovincialis) Farming on Oxygen Consumption and Nutrient Recycling in a Stelzenmüller, V., Gimpel, A., Gopnik, M., Gee, K., 2017. Aquaculture Site-Selection and
Eutrophic Coastal Lagoon. Hydrobiologia 550 (1), 183–198. Marine Spatial Planning: The Roles of GIS-Based Tools and Models. In Aquaculture
Noble, D., 2021. orchaRd: Orchard Plots and Prediction Intervals for Meta-Analysis. Perspective of Multi-Use Sites in the Open Ocean. Springer, Cham, pp. 131–148.
Noble, D.W.A., Lagisz, M., O’dea, R.E., Nakagawa, S., 2017. Nonindependence and Stenton-Dozey, J., Probyn, T., Busby, A., 2011. Impact of mussel (Mytilus
sensitivity analyses in ecological and evolutionary meta-analyses. Mol. Ecol. 26 (9), galloprovincialis) raft-culture on benthic macrofauna, in situ oxygen uptake, and
2410–2425. nutrient fluxes in Saldanha Bay, South Africa. Can. J. Fish. Aquat. Sci. 58,
Onorevole, K.M., Thompson, S.P., Piehler, M.F., 2018. Living shorelines enhance 1021–1031.
nitrogen removal capacity over time. Ecol. Eng. 120, 238–248. Svenningsen, N.B., Heisterkamp, I.M., Sigby-Clausen, M., Larsen, L.H., Nielsen, L.P.,
Petersen, J., Holmer, M., Termansen, M., Hasler, B., 2019. Nutrient Extraction through Stief, P., Schramm, A., 2012. Shell biofilm nitrification and gut denitrification
Bivalves. In Goods and Services of Marine Bivalves. Springer, Cham, pp. 179–208. contribute to emission of nitrous oxide by the invasive freshwater mussel Dreissena
Petersen, J.K., Hasler, B., Timmermann, K., Nielsen, P., Torring, D.B., Larsen, M.M., polymorpha (zebra mussel). Appl. Environ. Microbiol. 78 (12), 4505–4509.
Holmer, M., 2014. Mussels as a tool for mitigation of nutrients in the marine Tenore, K.R., Dunstan, W.M., 1973. Comparison of feeding and biodeposition of three
environment. Mar. Pollut. Bull. 82 (1–2), 137–143. bivalves at different food levels. Mar. Biol. 21 (3), 190–195.
Ray, N., Hancock, B., Brush, M., Colden, A., Cornwell, J., Labrie, M., Maguire, T., Thamdrup, B., Hansen, J.W., Jørgensen, B.B., 1998. Temperature dependence of aerobic
Maxwell, T., Rogers, D., Stevick, R., Unruh, A., Kellogg, M., Smyth, A., Fulweiler, W., respiration in a coastal sediment. FEMS (Fed. Eur. Microbiol. Soc.) Microbiol. Ecol.
2021. A review of how we assess denitrification in oyster habitats and proposed 25 (2), 189–200.
guidelines for future studies. Limnol. Oceanogr.: Methods 19 (10), 714–731. Thomsen, P.F., Willerslev, E., 2015. Environmental DNA – an emerging tool in
Ray, N.E., Fulweiler, R.W., 2021. Meta-analysis of oyster impacts on coastal conservation for monitoring past and present biodiversity. Biol. Conserv. 183, 4–18.
biogeochemistry. Nat. Sustain. 4 (3), 261–269. UNESCO-IOC and European Commission, 2021. MSPglobal International Guide on
Ray, N.E., Henning, M.C., Fulweiler, R.W., 2019. Nitrogen and phosphorus cycling in the Marine/Maritime Spatial Planning (IOC Manuals and Guides no 89). UNESCO, Paris,
digestive system and shell biofilm of the eastern oyster Crassostrea virginica. Mar. France, p. 152.
Ecol. Prog. Ser. 621, 95–105. Valiela, I., Cole, M.L., 2002. Comparative evidence that salt marshes and mangroves may
Rivera-Monroy, V.H., Twilley, R.R., Boustany, R.G., Day, J.W., Vera-Herrera, F., del protect seagrass meadows from land-derived nitrogen loads. Ecosystems 5 (1),
Carmen Ramirez, M., 1995. Direct denitrification in mangrove sediments in 92–102.
Terminos Lagoon, Mexico. Mar. Ecol. Prog. Ser. 126, 97–109. van der Schatte Olivier, A., Jones, L., Vay, L.L., Christie, M., Wilson, J., Malham, S.K.,
Roth, F., Wild, C., Carvalho, S., Rädecker, N., Voolstra, C.R., Kürten, B., Anlauf, H., El- 2020. A global review of the ecosystem services provided by bivalve aquaculture.
Khaled, Y.C., Carolan, R., Jones, B.H., 2019. An in situ approach for measuring Rev. Aquacult. 12 (1), 3–25.
biogeochemical fluxes in structurally complex benthic communities. Methods Ecol. van Stralen, M.R., Dijkema, R.D., 1994. Mussel culture in a changing environment: the
Evol. 10 (5), 712–725. effects of a coastal engineering project on mussel culture (Mytilus edulis L.) in the
Rysgaard, S., Thastum, P., Dalsgaard, T., Christensen, P.B., Sloth, N.P., 1999. Effects of Oosterschelde estuary (SW Netherlands). Hydrobiologia 282 (1), 359–379.
salinity on NH+ 4 adsorption capacity, nitrification, and denitrification in Danish Velinsky, D.J., Paudel, B., Quirk, T., Piehler, M., Smyth, A., 2017. Salt marsh
estuarine sediments. Estuaries 22 (1), 21–30. denitrification provides a significant nitrogen sink in Barnegat Bay, New Jersey.
Sandoval-Gil, J., Alexandre, A., Santos, R., Camacho-Ibar, V.F., 2016. Nitrogen uptake J. Coast Res. (78), 70–78, 10078.
and internal recycling in zostera marina exposed to oyster farming Eelgrass Potential Viechtbauer, W., 2010. Conducting meta-analyses in {R} with the {metafor} package.
as a Natural Biofilter. Estuar. Coast 39 (6), 1694–1708. J. Stat. Software 36, 1–48.
Schroder, T., Stank, J., Schernewski, G., Krost, P., 2014. The impact of a mussel farm on Vitousek, P.M., Aber, J.D., Howarth, R.W., Likens, G.E., Matson, P.A., Schindler, D.W.,
water transparency in the Kiel Fjord. Ocean Coast Manag. 101, 42–52. Schlesinger, W.H., Tilman, D.G., 1997. Human alteration of the global nitrogen
Serpetti, N., Witte, U.F.M., Heath, M.R., 2016. Statistical modeling of variability in cycle: sources and consequences. Ecol. Appl. 7 (3), 737–750.
sediment-water nutrient and oxygen fluxes. Front. Earth Sci. 4, 65. Wall, C.C., Peterson, B.J., Gobler, C.J., 2008. Facilitation of seagrass Zostera marina
Sharma, S., Goff, J., Moody, R., Byron, D., Heck, K., Powers, S., Ferraro, C., Cebrian, J., productivity by suspension-feeding bivalves. Mar. Ecol. Prog. Ser. 357, 165–174.
2016. Do restored oyster reefs benefit seagrasses? An experimental study in the Wang, W., 1980. Fractionation of sediment oxygen demand. Water Res. 14 (6), 603–612.
Northern Gulf of Mexico. Restor. Ecol. 24 (3), 306–313. Weitzman, J., 2019. Applying the ecosystem services concept to aquaculture: a review of
Smaal, A.C., Ferreira, J.G., Grant, J., Petersen, J.K., Strand, Ø., 2019. Correction to: approaches, definitions, and uses. Ecosyst. Serv. 35, 194–206.
goods and services of marine bivalves. In: Smaal, A.C., Ferreira, J.G., Grant, J., Zertuche-González, J.A., Camacho-Ibar, V.F., Pacheco-Ruíz, I., Cabello-Pasini, A.,
Petersen, J.K., Strand, Ø. (Eds.), Goods and Services of Marine Bivalves: Cham. Galindo-Bect, L.A., Guzmán-Calderón, J.M., Macias-Carranza, V., Espinoza-
Springer International Publishing. E1-E1. Avalos, J., 2009. The role of Ulva spp. as a temporary nutrient sink in a coastal
Smaal, A.C., Schellekens, T., van Stralen, M.R., Kromkamp, J.C., 2013. Decrease of the lagoon with oyster cultivation and upwelling influence. J. Appl. Phycol. 21 (6), 729.
carrying capacity of the Oosterschelde estuary (SW Delta, NL) for bivalve filter Zwetsloot, P.P., Van Der Naald, M., Sena, E.S., Howells, D.W., IntHout, J., De Groot, J.A.,
feeders due to overgrazing? Aquaculture 404, 28–34. Chamuleau, S.A., MacLeod, M.R., Wever, K.E., 2017. Standardized Mean Differences
Cause Funnel Plot Distortion in Publication Bias Assessments. Elife 6, e24260.

12