Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

Contents lists available at SciVerse ScienceDirect

Journal of Experimental Marine Biology and Ecology

j ourna l homepage: www.e lsev ie r .com/ locate / jembe

Benthic primary production and bacterial denitrification in a Mediterraneaneutrophic coastal lagoon

M. Bartoli a,⁎, G. Castaldelli b, D. Nizzoli a, P. Viaroli a

a Department of Life Sciences, University of Parma, Parma, Italyb Department of Biology and Evolution, University of Ferrara, Ferrara, Italy

⁎ Corresponding author. Tel.: +39 0521 905048.E-mail address: marco.bartoli@unipr.it (M. Bartoli).

0022-0981/$ – see front matter © 2012 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.jembe.2012.09.011

a b s t r a c t

a r t i c l e i n f o

Article history:Received 8 March 2012Received in revised form 20 September 2012Accepted 22 September 2012Available online 31 October 2012

Keywords:DenitrificationLagoonMacroalgaeMicrophytobenthosNitrogen uptake

Microphytobenthos and macroalgal mats are simultaneously present in eutrophic lagoons and are expectedto have different direct and indirect effects on nitrogen related processes through uptake and inhibition orstimulation of microbial activity. To assess the relative contribution of different primary producer communi-ties and heterotrophic processes to benthic nitrogen cycling, we studied nitrogen uptake and bacterial deni-trification in the eutrophic Sacca di Goro lagoon (northern Italy). Benthic fluxes of oxygen and dissolvedinorganic nitrogen (DIN), and rates of nitrification-coupled (Dn) and water–nitrate (Dw) denitrificationwere measured every 30–45 days for one year at two shallow sites. Station Giralda is close to the main fresh-water inlet, has turbid waters and muddy-organic and bioturbated sediments with microphytobenthos(MPB). Station Gorino is brackish with muddy-sand sediments which are covered by macroalgal mats ofthe genus Ulva. Here, sediment patches with and without macroalgae (MA) were simultaneously studied.Sediments with MPB were net heterotrophic and regenerated large amounts of ammonium to the water col-umn. At Gorino, sediments with MA were net autotrophic through the year, and DIN fluxes were mainly con-trolled by macroalgal uptake. On an annual basis, denitrification rates were three fold higher at Giralda(2.27±0.06 mol N m−2 yr−1) than at Gorino (0.83±0.01 mol N m−2 yr−1), due to higher nitrate in thewater column and nitrification in surface sediments. At Gorino, denitrification was one order of magnitudelower than DIN uptake by macroalgae (10.39±1.30 mol N m−2 yr−1). Nevertheless, the differences be-tween denitrification rates in sediments with and without MA were unexpectedly negligible, showing thatthe denitrification capacity was not suppressed by macroalgal competition. Results from this study suggestthat in eutrophic lagoons nitrogen cycling seems less affected by MPB compared to more oligotrophic coastalwaters and that most of the available DIN flows through benthic macroalgae. However, Ulva is only a tempo-rary N-sink and most of its nitrogen pool can be either rapidly recycled or exported by tidal currents to theopen sea.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

Eutrophic coastal lagoons display high rates of both primary pro-duction and respiration; the balance between these opposite pro-cesses can be extremely variable over small spatial and temporalscales, with implications for oxygen availability and nutrient cycling(Diaz and Rosenberg, 2008; Risgaard-Petersen, 2003; Viaroli andChristian, 2003). Anthropogenic nitrogen loads to coastal areas aredemonstrated to increase the rates of primary production and organicmatter input to surface sediments and to alter the composition of thebenthic vegetation (Galloway et al., 2004; Middelburg and Levin,2009; Valiela et al., 1997). The relative biomass abundance and the ac-tivity of benthic vegetation, viz seagrasses and macroalgae, andmicrophytobenthos (Burkholder et al., 2007; McGlathery et al., 2007;Viaroli et al., 2008) can greatly influence the fluxes of nitrogen at the

rights reserved.

water–sediment interface (Bartoli et al., 2003; Dalsgaard, 2003;Sundbäck et al., 2000; Tyler et al., 2003; Welsh et al., 2000a). A largebody of literature has explored nitrogen cycling within seagrassmeadows, as rooted macrophytes are particularly vulnerable to the in-crease of reactive nitrogen (Burkholder et al., 2007; McGlathery et al.,2007). Healthy meadows are generally found in oligotrophic, wellflushed coastal areas, with elevated light penetration, and are gener-ally net autotrophic during the vegetative period (Welsh et al.,2000a). In areas with seagrass meadows, the main nitrogen pool iswithin sediments and nitrogen fluxes are mainly driven by rootand leaf uptake (Bartoli et al., 2008; Risgaard-Petersen et al., 1998).While marine plants compete with denitrifiers for nitrogen andhave scarce oxygen transport capacity to the roots, limiting subsur-face nitrification, they stimulate bacterial nitrogen fixation, that con-tributes to a variable fraction of the plant N requirement (Ottosen etal., 1999; Welsh, 2000b).

The interactions between microphytobenthos (MPB) and benthicnitrogen cycling were also explored in detail. In autotrophic sediments

42 M. Bartoli et al. / Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

with benthicmicroalgae, direct nutrient uptake in theMPB layer can re-tain inorganic nitrogen and significantly attenuate its regeneration. TheMPB activity can either stimulate or depress nitrification and denitrifi-cation, depending upon N availability (Bartoli et al., 2003; Risgaard-Petersen, 2003; Sundbäck et al., 2000). Photosynthesis by MPB at thesediment–water interface can expand the oxic sediment horizon witha potential positive effect on nitrifiers (Revsbech et al., 1981). Converse-ly, MPB metabolism removes NH4

+ and CO2 from the porewater, occa-sionally resulting in pH levels of 9–10, which inhibit nitrification and,consequently, the coupled denitrification (Focht and Verstraete, 1977).Furthermore, oxygen production and consumption at the sediment–water interface regulate the thickness of the microoxic surficial layer,and as a consequence the pathway of nitrate from thewater to the anoxicdenitrification zone (Risgaard-Petersen et al., 1994).

By comparison, some studies were addressed to the interactions be-tweenmacroalgal mats (MA) and benthic nutrient cycling (McGlatheryet al., 2004, 2007; Sundbäck and Miles, 2002; Sundbäck et al., 2003;Tyler et al., 2003; Villares and Carballeira, 2003) while a few studies ex-plored the interactions between the activity of macroalgae and denitri-fication (Dalsgaard, 2003; Krause-Jensen et al., 1999; LaMontagne et al.,2002; Trimmer et al., 2000a). Macroalgal blooms are primarily sustainedby the DIN supply from thewater column, although the nitrogen require-ment can be recovered from sediments, via intense recycling of both am-monium and dissolved organic nitrogen, or from internal nitrogen pools(Naldi and Viaroli, 2002; Sundbäck et al., 2003; Trimmer et al., 2000b;Tyler et al., 2003). Dalsgaard (2003) demonstrated that even if limitedto a short time lag (1–2 months), the transient accumulation ofmacroalgalmats in a shallow fjord had the potential to change the annualprimary productivity and nutrient budgets. He demonstrated thatmacroalgal production can switch from a heterotrophic, net nitrogen re-leasing sediment into an autotrophic, net N-sink one. In eutrophicwarm waters, MA may attain growth rates up to 0.3 d−1, which aremainly sustained by their capacity to take up and store dissolved inorgan-ic nitrogen (DIN) sources, especially nitrate (Naldi and Viaroli, 2002). Atpeak biomass, macroalgae such as Ulva sp. usually attain a standingstock of some hundred g m−2 as dry weight, with net nitrogen uptakerates ranging from 10 to 20 μmol g−1 h−1. Under these circumstances,most of DIN flux is likely controlled by MA (Viaroli et al., 2005), with aprobable suppression of nitrate loss via denitrification (Krause-Jensen etal., 1999; LaMontagne et al., 2002; Trimmer et al., 2000a). However, theMA biomass is only a temporary sink, which suddenly and rapidly be-comes a nitrogen source for the system when the MA mat collapses insummer, causing the so called “dystrophic crises” (Viaroli et al., 1996,2008).

The main hypotheses of this study are that in Mediterranean eutro-phic lagoons, where MA are present all year round, macroalgae drive amajor fraction of DIN fluxes, suppress N loss via denitrification andhave a stronger effect on the benthic metabolism compared to MPB.To test these hypotheses, we assessed primary production and nitrogenuptake by MPB and MA in the Sacca di Goro (Po River Delta, NorthernItaly) and we compared N retention by MPB and MA with N losses viadenitrification.

2. Material and methods

2.1. Study sites

The study was performed at two field sites within the eutrophicSacca di Goro lagoon (Viaroli et al., 2006). Station Giralda (44° 49′ N12° 17′ E) is located in the western part of the lagoon, close to theinlet of turbid, nutrient-rich freshwater from the Po di Volano canal;water depth averages 70 cm(Fig. 1). Sediments are generally bioturbat-ed by surface (i.e. Corophium insidiosum) and deep (i.e. Neanthessuccinea) burrowers and consist of a soft mud colonised by benthic di-atoms. Station Gorino (44° 48′ N 12° 19′ E) has a mean depth of60 cm and muddy-sand sediments. It is located in the eastern area of

the Sacca di Goro, where drifting mats of the macroalga Ulva sp. devel-op. Long termmonitoring of this site suggests a strong control of waterhydrochemistry by MA, with inorganic nitrogen drops and oxygenpeaks during the spring, when macroalgae exhibit maximum growthrates (Viaroli et al., 2006). Station Giralda and station Gorino aremicro-tidal, with average daily depth variations of ±30 cm.

2.2. Sampling programme

Measurements were performed in the framework of a European pro-ject (NICE, Nitrogen Cycling in Estuaries, contract MAS3-CT96-0048),according to an experimental approach definedduring an intercalibrationworkshop and detailed in a protocol method handbook (Dalsgaard et al.,2000). Undisturbed sediment samples were collected from stationGiralda on 03.26.97, 05.10.97, 06.18.97, 07.29.97, 09.5.97, 11.11.97,12.10.97, 01.13.98 and 02.17.98. Sediment samples were collected withcores of 2 different dimensions (diameter×length), each for a distinctpurpose: sediment characterisation (5×30 cm, n=3) and fluxmeasure-ments, (8×40 cm, n=6 to 10). At Gorino sediments were collected inpatches with floating mats of Ulva sp. and in areas devoid of macroalgaeon 04.03.97, 06.10.97, 08.21.97, 09.18.97, 10.16.97, 11.26.97, 01.27.98and 03.10.98. Sediment characterisation and flux measurements in sedi-ment without MA was performed as detailed for station Giralda (samecores, same replicates) while fluxes in sediments covered with MAwere determined in squared Plexiglas chambers (n=6) (see below).Sediment in the cores and chambers used for flux measurementswere levelled to a thickness of about 10 cm; the water column overly-ing sediments had a depth of about 30 cm. Water volume in coreswith MA-free sediments was about 1.5 l while that in chambers withMA was about 12 l.

At both sampling sites about 100 l of water were collected onevery sampling campaign for core maintenance, preincubation andincubation procedures.

2.3. Sediment features

Benthicmicroalgal biomasswasmeasured in triplicate as chlorophyll-a (Chl-a) concentration in the top 0.5 cm of sediment and determinedspectrophotometrically after extraction with 90% acetone (Lorenzen,1967). Bulk density was determined as the ratio between wet weightand volume (typically 5 ml) of sediment. Organic matter content (OM)was measured as percentage of weight loss by ignition (450 °C, 2 h)from dried sediment.

Biomass of macroalgae was estimated by random positioning of aplastic frame over the sediment surface (30×30 cm, n=5), harvestingof enclosed vegetal matter, drying at 60 °C after removal of ephiphytesand other organisms and weighing.

2.4. Flux measurements

Fluxes of DIN and O2 across the sediment–water interface and de-nitrification rates were measured during light and dark incubations ofintact sediment cores and flux chambers. Sediments with MPB fromstations Giralda and Gorino were collected by a hand corer andmaintained in cylindrical Plexiglas liners; water stirring was ensuredby a 4 cm long Teflon-coated magnetic stirring bar, suspended 6 cmabove the sediment surface and driven by an external rotating mag-net at 40 rpm. Three to five cores were incubated in the light andthree to five cores were incubated in the dark. Sediment samplingin areas covered with MA was done with a hand held box corermade of 1 mm thick steel plate, fitting precisely Plexiglas flux cham-bers (20×20×40 cm, w×d×h, n=6). The corer was pushed 10 to15 cm into the sediment, dug out with a shovel, transferred underwa-ter in the flux chamber and then carefully removed. When removingthe corer a gap of 2 to 3 mm was left between the sediment and fluxchamber walls, however, the sediment block expanded horizontally,

* Giralda

* Gorino

Fig. 1. Location of the sampling sites in the Sacca di Goro lagoon (Po River Delta, Northern Italy).

43M. Bartoli et al. / Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

filling this gap within a few seconds. This incubation system, intro-duced by T. Dalsgaard, differs from standard incubation cores as a larger,rather undisturbed sediment portion covered with a macroalgal mat isexplored. Furthermore, water stirring in chambers is created by diffusersthat allow an unidirectional water flow across the algal mat surface andbetter reproduce in situ water circulation and the behaviour of chemicalspecies in a benthic system with macroalgae compared to traditionalcores (for details see Dalsgaard et al., 2000; Dalsgaard, 2003). Macroalgaeused in the experiments were collected from the study site and trans-ferred into incubation chambers simulating in situ biomass. Preliminarytestswhich incubatedMA-free sediments from station Gorinowith cylin-drical cores and flux chambers demonstrated that the same fluxes wereobtained using both incubation systems. Three chambers were incubatedin the light and three chambers were incubated in the dark.

All sediment cores and chambers, whether containing MPB or MA,were double wrapped with aluminium foil in order to prevent lightpenetration to the sediment from the side and were incubated theday after the sampling at in situ temperature. Incubations in thelight were performed reproducing the average irradiance of eachsampling period, by means of halogen lamps and screens. Irradiancewas measured at the water surface with a LI-COR 192 sensor. As theheight of the water in the cores was ~30 cm, corresponding to halfwater column at both sites, the irradiance at the sediment surfacewas higher than the daily average. Measurements in the light werethus probably reflecting benthic photosynthetic activity at peakdaily irradiance and calculated average fluxes during the light phaseprobably overestimate those in situ. Before starting the incubationsthe cores and the chambers were maintained submersed in largetanks containing vigorously mixed in situ water, without the top lidand with the stirring system on to allow water exchange with the in-cubation tank. The day after the sampling, the water in the tanks wasexchanged and incubations were started by lowering the water in thetanks just below the top of the cores/chambers and by sealing them

with transparent Plexiglas floating lids provided with a samplingport. Floating lids were fitting precisely cores and chambers innerwalls and were about 3 cm thick so that a stagnant water film be-tween the lids and the core/chamber wall prevented significant gasexchange with the atmosphere.

Water samples (60 ml) were collected and replaced with water fromthe incubation tank 3–4 times at regular intervals from the overlyingwater using plastic syringes. Each sampling and replacement of waterwas equivalent to ~4 and ~0.5% of the total water incubated in coresand chambers, respectively. Samples for O2 determinations (approxi-mately 20 ml) were transferred to 7 ml glass vials (Exetainer, Labco,High Wycombe, UK) flushing the vial volume ~3 times to avoid atmo-spheric contamination and Winkler reagents were added immediately(Strickland and Parsons, 1972). Samples for NH4

+, NO3− and NO2

− deter-minations were filtered through Whatman GF/F glass fibre filters,transferred to plastic vials and frozen. NH4

+ was determined spectro-photometrically using salicylate and hypochlorite in the presence ofsodium nitroprussiate (Bower and Holm-Hansen, 1980). NO3

− wasdetermined after reduction to NO2

− in the presence of cadmium,and NO2

− was determined spectrophotometrically using sulphanil-amide and N-(1-naphtyl)ethylendiamine (Golterman et al., 1978).In all the incubations water–sediment fluxes of O2, NH4

+, NO2− and

NO3− were measured as concentration changes in the water with

time according to Eq. (1).

Flux ¼ α � Vð Þ=A ð1Þ

where:

Flux is expressed in μmol or mmol of the species of interest m−2 h−1;

α slope of the linear regression of concentration (μM or mM)versus time (h);

44 M. Bartoli et al. / Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

A area of sediment surface in core/chamber (m2);V volume of water in core/chamber (l).

A flux from the water column to the sediment is negative while aflux from the sediment to the water column is considered as positive.Total incubation time varied with the water temperature between aminimum of 2 and a maximum of 5 h, in order to keep oxygen con-centration within ±20% of initial value.

Daily rates were calculated multiplying hourly rates in the light andin the dark by the average number of light and dark hours in each sam-pling period and then summing the obtained amounts. Annual rateswere obtained by multiplying daily values by the number of days ineach sampling period and then summing values in each period. Stan-dard deviation of hourly averages was propagated accordingly.

Rates of net (NBP) and gross (GBP) benthic carbon fixation byMPB and MA were calculated from oxygen fluxes measured in thelight assuming a photosynthetic quotient for microphytobenthos(O2 evolution:CO2 uptake in the light) of 1.2 (Schramm et al., 1984;Wetzel and Likens, 1991). Rates of theoretical N assimilation by pri-mary producers were calculated for sediments with MPB and MAfrom net and gross primary productivity assuming for MPB a C/Nratio of 9 (Sundbäck et al., 2000 and references therein) and usingmeasured C/N molar ratios for Ulva (Viaroli et al., 2005).

Table 1Main features of the benthic environment at the two sampling sites. OM(sedimentary organ-ic matter) and density were measured in the upper 2 cm sediment layer. Chl a (sedimentchlorophyll a)wasmeasured in the upper 0.5 cm sediment layer. Values refer tomean annu-al values and standard deviation (n=27 and 24 at Giralda and Gorino, respectively).Macrofauna groups include surface and subsurface deposit feeders (SDF and SSDF, respec-tively), surface feeders (SF) and scrapers (SC).

Sedimenttypology

OM (%) Density(g cm−3)

Chl a(mg m−2)

Primaryproducers

Macrofauna

Giralda Muddy 6.4±1.6 1.15±0.11 36±15 MPB SDF, SSDFGorino Muddy

sand4.4±0.9 1.24±0.08 17±4 MPB, MA SDF, SSDF,

SF, SC

2.5. Denitrification measurements

The isotope pairing technique (IPT) was used to measure denitri-fication (light and dark rates) on the same set of cores and chambersused for solute fluxes (Nielsen, 1992). Incubations for solute fluxesand denitrification were sequential and both performed the dayafter the sampling. The water in the tank was renewed between thetwo incubations, and the open cores and chambers were submersedfor a couple of hours. The IPT allows for differentiation of denitrifica-tion of nitrate diffusing to the anoxic sediment from the water col-umn (Dw) and denitrification of nitrate produced within thesediment due to nitrification (Dn); total denitrification (Dt) is thesum of Dw and Dn.

Methodological concerns were raised about the IPT, in particularwhen the technique is applied to sediments where denitrificationand anammox coexist. Anammox cannot be discriminated from deni-trification as a source of N2 and makes invalid the assumptions onwhich IPT calculations are based, in particular that the addition of15NO3

− does not affect the production of 14N–N2 (Risgaard-Petersenet al., 2003). However, different studies showed that anammox cancontribute significantly to N2 production in coastal shelves and deepseas (Dalsgaard et al., 2005; Trimmer et al., 2006), but that it repre-sents a minor fraction of N2 production in coastal eutrophic ecosys-tems (Burgin and Hamilton, 2007; Koop-Jakobsen and Giblin, 2009).At stations Giralda and Gorino calculated 28N2 production was inde-pendent from 15NO3

− additions in all experiments carried out withthe concentration series approach. The contribution of anammox toN2 production was thus considered negligible (Risgaard-Petersen etal., 2003) and the original calculations proposed by Nielsen (1992)were considered reliable.

At the beginning of the experiment, the water in the tank waslowered just below the top of the cores/chambers and a water subsam-ple (5 ml) was taken from each core for NO3

− concentration measure-ment. 15NO3

−, from a stock 15 mM Na15NO3 solution was then addedto the water phase of each of the replicate cores/chambers. Labelled ni-trate was added to have a final 15N atom% of at least 30%. Within 5 minof the addition of 15NO3

− another water samplewas collected from eachcore to calculate the 14N/15N ratio in the NO3

− pool; the cores and cham-bers were then closed with floating lids and the incubation started. Atthe end of the incubation, 5 to 10 ml of ZnCl2 (7 M) was added to thewater phase of each core and chamber, and then whole sediment and

water volumes were gently mixed. An aliquot of the slurry was trans-ferred to a 12.5 ml gas-tight vial; 14N15N and 15N15N abundance in N2

were analysed bymass spectrometry at the National Environmental Re-search Agency, Silkeborg, Denmark (Risgaard-Petersen and Rysgaard,1995).

At the end of denitrificationmeasurements the biomass of MAwasquantified. Macroalgae were carefully removed from each chamber,rinsed, dried at 60 °C and weighed. Also benthic macrofauna was re-covered and identified from each incubated core and chamber, bysieving sediments (0.5 mm sieve).

3. Results

3.1. General features of the sampling sites

General features of the benthic environment at the two samplingsites are reported in Table 1. At station Giralda macrofauna was com-posed of surface deposit feeders, dominated in numbers by Streblospioshrubsolii, C. insidiosum, Polydora ciliata and N. succinea, feeding onfine detritus at the sediment surface, and subsurface deposit feeders,dominated by Capitella capitata. At Gorino the presence of macroalgaecreated a higher spatial and trophic heterogeneity and contributed toselect a more diversified macrobenthic community. Surface depositfeeders were dominated in numbers by P. ciliata, C. insidiosum,S. shrubsolii, and N. succinea, while SSDF included C. capitata. Moreover,Musculista sennhousia represented the group of suspension feederswhile Microdeutopus gryllotalpa, Gammarus spp. and Dexamine spinosarepresented the group of scrapers, feeding on particles attached tomac-rophytes. At GorinoMAmats were present in all sampling periods withAugust 1997 as only exception, when macroalgae were not found. Thebiomass of Ulva showed an erratic pattern, with a late winter peak(478±60 gdw m−2, Fig. 2). We estimated that when macroalgaewere present they covered more than 80% of the sediment surface.

The range of variation of water salinity was 10–22 psu (annual aver-age 15 psu) at Giralda and 20–26 psu, (annual average 22 psu) at Gorino.Concentration of dissolved inorganic nitrogen at Giralda averaged116 μM and almost double that at Gorino (60 μM). At both sites nitriteconcentrations were low and nitrate represented on average 46and 48% of total dissolved inorganic nitrogen at Giralda and Gorino,respectively.

At the two sampling stations physico-chemical parametersexhibited a strong seasonality (Fig. 3). At Giralda water temperaturevaried between 7 and 26 °C; ammonium and nitrate had similar con-centrations in the water column and displayed a similar pattern withminimum values in summer (~20 μM) and winter peaks (>100 μM).At Gorino water temperature varied between 9 and 27 °C; the seasonalpattern of nitrate concentration was similar to that at Giralda but thevariation range was smaller (11–54 μM). The seasonal pattern of am-monium (10–45 μM) was different, with relatively higher values inlate summer–autumn months (Fig. 3).

03/97 06/97 09/97 12/97 03/980

100

200

300

400

500

600U

lva

spp.

bio

mas

s (g

dw

m-2

)

Fig. 2. Average biomass (±standard deviation, n=5) of Ulva spp. at station Gorino.

Giralda, MPB

03/97 06/97 09/97 12/97 03/98

-10

-5

0

5

10

mm

ol O

2 m

-2d-1

mm

ol O

2 m

-2h-1

-80

-60

-40

-20

0

20

40

60

80NBPGBP

BRdaily

45M. Bartoli et al. / Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

3.2. Oxygen fluxes

At station Giralda net benthic production (NBP) was >0 only inJuly and September 1997 and in February 1998. On a daily basis ben-thic oxygen respiration (BR) measured in the dark (from −1.51±0.15 to −5.02±0.24 mmol O2 m−2 h−1) was not balanced by anequivalent oxygen production in the light, with February 1998 asthe only exception, resulting in negative benthic oxygen daily bud-gets in 8 out of 9 sampling dates (Fig. 4). Gross benthic production(GBP) varied from a minimum of 0.65 to a maximum of 5.84 mmolO2 m2 h−1.

At station Gorino during light incubations sediment respiration gen-erally prevailed over MPB production, resulting in negative oxygenfluxes in almost all sampling dates. BR varied between −1.66±0.27and −6.60±1.10 mmol O2 m−2 h−1, with absolute values higherthan NBP, allowing the calculation of GBP rates (between 0.2±2.0and 4.45±1.18 mmol O2 m−2 h−1). At Gorino, MA-free sediments

03/97 06/97 09/97 12/97 03/98

Tem

pera

ture

(°C

)T

empe

ratu

re (

°C)

0

10

20

30

40

Irra

dian

ce (

mol

pho

tons

m-2

d-1)

Irra

dian

ce (

mol

pho

tons

m-2

d-1)

0

20

40

60

80

0

50

100

150

200temperature irradiance

NO3-

NO2-

NH4+

03/97 06/97 09/97 12/97 03/980

10

20

30

40

0

20

40

60

80

[NH

4+],

[NO

2- ], [N

O3- ],

µM

[NH

4+],

[NO

2- ], [N

O3- ],

µM

0

20

40

60

80temperature irradiance

NO3-

NO2-

NH4+

Giralda

Gorino

Fig. 3. Seasonal variations of irradiance (average value measured at the water surfaceat each sampling period), water temperature and dissolved inorganic nitrogen at thetwo sampling sites.

were heterotrophic in all sampling dates, with highest oxygen demandcalculated in September 1997 (−56.9±4.33 mmol O2 m−2 d−1). In-cubations with sediment and MA generally resulted in NBP>0, withSeptember 1997 as only exception; positive values varied between8.05±2.17 and 32.3±3.17 mmol O2 m−2 h−1. Dark oxygen demandpeaked in June 1997 with −10.22±1.09 mmol O2 m−2 h−1 and wasgenerally lower (as absolute value) than NBP, resulting in positivedaily oxygen budgets in most sampling dates (Fig. 4). Sediment withMA displayed elevated rates of GBP, up to 40 mmol m−2 h−1. Overall,on an annual basis, sediments with benthic microalgae at Giralda andGorino were net heterotrophic (−11.16±1.24 and −19.00±0.59 mol O2 m−2 y−1) while sediments with benthic macroalgae atGorino were net autotrophic (24.35±2.28 mol O2 m−2 y−1).

3.3. Inorganic nitrogen fluxes

Sediments with benthic microalgae at Giralda and Gorino were aNH4

+ source in most incubations in the dark and in the light; the ac-tivity of MPB resulted in only a slight attenuation of ammonium re-lease (Fig. 5). On a daily basis sediments at Gorino were always net

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

mm

ol O

2 m

-2h-1

mm

ol O

2 m

-2h-1

mm

ol O

2 m

-2d-1

mm

ol O

2 m

-2d-1

Gorino, MPB

-10

-5

0

5

10

-100

-50

0

50

100

Gorino, MA

-20

0

20

40

60

-200

-100

0

100

200

300

400

NBPGBP

BRdaily

NBPGBP

BRdaily

Fig. 4. Benthic fluxes of dissolved oxygen measured in the light (NBP) and in the dark(BR), calculated gross benthic production (GBP) and net daily fluxes (average values±standard deviation, n=3 to 5, are presented). At station Giralda primary producerswere represented only by benthic microalgae (MPB) while at station Gorino bothMPB and benthic macroalgae (Ulva spp., MA) were represented. Note different scalesin the graphs.

Giralda, MPB

03/97 06/97 09/97 12/97 03/98 03/97 06/97 09/97 12/97 03/98 03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

µmol

NH

4+m

-2h-1

µmol

NH

4+m

-2h-1

µmol

NH

4+m

-2h-1

mm

ol N

H4+

m-2

d-1

mm

ol N

O2- m

-2d-1

mm

ol N

O3- m

-2d-1

mm

ol N

O3- m

-2d-1

mm

ol N

O3- m

-2d-1

mm

ol N

O2- m

-2d-1

mm

ol N

O2- m

-2d-1

µmol

NO

2- m-2

h-1

µmol

NO

3- m-2

h-1µm

ol N

O3- m

-2h-1

µmol

NO

3- m-2

h-1

µmol

NO

2- m-2

h-1µm

ol N

O2- m

-2h-1

mm

ol N

H4+

m-2

d-1m

mol

NH

4+m

-2d-1

-1500

-1000

-500

0

500

1000

1500

-20

-10

0

10

20

lightdarkdaily -2000

-1000

0

1000

2000

-30

-20

-10

0

10

20

30

-400

-200

0

200

400

-10

-5

0

5

10

lightdarkdaily

lightdarkdaily

Gorino, MPB

-1000

-500

0

500

1000

-20

-10

0

10

20

lightdarkdaily

-1000

-500

0

500

1000

-15

-10

-5

0

5

10

15

-200

-100

0

100

200

-4

-2

0

2

4

lightdarkdaily

lightdarkdaily

Gorino, MA

-3000

-2000

-1000

0

1000

2000

3000

-40

-20

0

20

40

lightdarkdaily -4000

-2000

0

2000

4000

-60

-40

-20

0

20

40

60

-300

-200

-100

0

100

200

300

-4

-2

0

2

4

lightdarkdaily

lightdarkdaily

Fig. 5. Benthic fluxes of dissolved inorganic nitrogen (NH4+, NO2

− and NO3−) measured in the light and in the dark and calculated daily rates (average values±standard deviation, n=3 to 5, are presented).

46M.Bartoliet

al./JournalofExperim

entalMarine

Biologyand

Ecology438

(2012)41

–51

Table 2Daily rates of total denitrification (Dw+Dn) measured at station Giralda and at stationGorino (average values±standard deviation, n=3 to 5, are reported).

Giralda Gorino

MPB MPB MA

Sampling date Dtot(mmol N m−2 d−1)

Sampling date Dtot(mmol N m−2 d−1)

03/26/1997 6.29±1.02 04/03/1997 1.92±0.22 0.94±0.0805/10/1997 28.2±3.30 06/10/1997 0.81±0.19 1.46±0.3706/18/1997 6.87±1.14 08/21/1997 2.27±0.31 1.05±0.2307/29/1997 1.92±0.18 09/18/1997 3.79±0.37 3.15±0.7009/05/1997 1.95±0.31 10/19/1997 1.03±0.14 0.59±0.1111/11/1997 1.75±0.43 11/25/1997 2.03±0.44 2.36±1.0312/10/1997 3.67±0.86 01/27/1998 3.96±0.43 5.76±1.6101/13/1998 1.68±0.38 03/10/1998 3.91±0.85 2.99±0.6202/17/1998 2.38±0.81

47M. Bartoli et al. / Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

NH4+ producing, with rates between 3.27 and 11.71 mmol N m−2 d−1;

at station Giralda the picture was similar, with a net NH4+ uptake mea-

sured only in February 1998 (−5.02±1.14 mmol N m−2 d−1).Overall, NH4

+ regeneration was comparable at the two sites, with1.92±0.70 and 1.74±0.28 mol N m−2 y−1 at Giralda and Gorino,respectively. Ammonium efflux from sediments was not apparent inthe presence of Ulva, with net uptake in all light and in most dark in-cubations, peaking in October 1997 (above 2000 μmol N m−2 h−1,light flux). On an annual basis sediment with Ulva trapped some5.02±1.14 mol N–NH4

+m−2 y−1.At the two stations nitrite fluxes were about one order of magnitude

lower than those of nitrate (Fig. 5). At Giralda sediments were producing

Giralda, MPB

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

µmol

N m

-2h-1

mm

ol N

m-2

d-1m

mol

N m

-2d-1

mm

ol N

m-2

d-1

µmol

N m

-2h-1

µmol

N m

-2h-1

0

200

1000

1200

0

1

2

3

4

20

25

lightdarkdaily

Dn

Gorino, MPB

0

50

100

150

200

0

1

2

3

4

5

6

lightdarkdaily

Dn

Gorino, MA

0

50

100

150

200

250

300

0

1

2

3

4

5

6

lightdarkdaily

Dn

Fig. 6. Rates of denitrification of nitrate produced within sediments by nitrification (Dn) anhourly rates measured in the light and in the dark and daily rates (average values±standa

NO2− in most sampling dates and on an annual basis the net flux to the

water column was 0.64±0.20 mol N m−2 y−1; at Gorino, NO2−

fluxesfrom and to the water column were nearly balanced and annual rateswere small (0.06±0.07 and −0.04±0.09 mol N m−2 y−1, in incuba-tions with MPB and MA, respectively).

Nitrate fluxesweremore erratic and likely influenced by nitrate avail-ability in the water column and pore water concentration gradients(MA-free sediments) orUlvauptake. AtGiralda dailyfluxeswerenegativein spring and winter and positive during summer, while at Gorino dailyfluxes did not show a clear seasonal pattern and were mostly negative,with a peak in June over −3000 μmol N m−2 h−1 measured in thelight in sediments with Ulva. Annual rates were −1.33±0.76, −0.97±0.30 and −5.33±0.63 mol N m−2 y−1 at Giralda, Gorino (MPB) andGorino (MA), respectively.

3.4. Denitrification rates

At station Giralda denitrification rates were highly variablethrough the year, with daily values between 1.68±0.38 and 28.2±3.30 mmol N m−2 d−1 (Table 2). Measured hourly values reachedtheir maximum (about 1100 μmol N m−2 h−1) in May 1997, whilstduring the rest of the year rates were lower, between 70 and260 μmol N m−2 h−1 (Fig. 6). The peak of total denitrificationcorresponded with the peak in water column nitrate concentrationbut Dn, rather than Dw, was the dominant component during bothlight and dark incubations (Fig. 6). These results can be explained interms of high nitrate production rates, sustained by the availabilityof ammonium and by high density of burrowing macrofauna, at thatdate mainly represented by C. insidiosum (2388±683 ind.m−2). At

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

03/97 06/97 09/97 12/97 03/98

µmol

N m

-2h-1

µmol

N m

-2h-1

µmol

N m

-2h-1

mm

ol N

m-2

d-1m

mol

N m

-2d-1

mm

ol N

m-2

d-10

50

100

150

200

250

300

0

2

4

6

8

10

Dw

0

50

100

150

200

250

300

0

1

2

3

4

Dw

0

50

100

150

200

0

1

2

3

4

Dw

d of nitrate diffusing to anoxic sediments from the water column (Dw). Graphs reportrd deviation, n=3 to 5, are presented).

48 M. Bartoli et al. / Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

Giralda there were no systematic differences between rates of Dn orDw measured in the light and in the dark; Dn represented up to85% of total denitrification (May 1997) and on an annual basis it rep-resented on average 52% of Dtot.

At Gorino daily rates of total denitrification varied also consid-erably through the year, by a factor 5 in MA-free sediments(0.81–3.96 mmol N m−2 d−1) and by a factor 10 in the presenceof macroalgae (0.6–5.8 mmol N m−2 d−1) (Table 2). Rates tendedto be lower than those measured at Giralda, in particular in spring,while they peaked in January 1998 in both sediments with andwithout MA (Table 2). As evidenced for station Giralda there wereno systematic differences between rates of Dn or Dw measured inthe light and in the dark, both in the presence or absence ofmacroalgae (Fig. 6). Denitrification of nitrate produced via nitrifi-cation within sediments represented a variable fraction of total de-nitrification, averaging 38 and 47% in MA-free sediments and insediments with macroalgae, respectively.

On an annual basis rates of total denitrification were about 3 foldhigher at Giralda (2.27±0.06 mol N m−2 y−1) than at Gorino, wherenitrogen loss via denitrification was comparable when measured insediments devoid (0.85±0.02 mol N m−2 y−1) or with Ulva (0.83±0.01 mol N m−2 y−1).

4. Discussion

4.1. Benthic production and nitrogen uptake by benthic micro andmacroalgae

In our experimental setup, during light incubations, we havereproduced at the water surface the average irradiance of each sam-pling period. As the water column overlying sediment in cores andchambers was nearly half of that at the sampling sites, irradiance atthe sediment interface was higher in experimental compared to aver-age in situ conditions and probably closer to that at midday. We as-sumed constant rates during the light period; integrated dailyvalues of net and gross primary production by MPB are as a conse-quence overestimates of in situ values. Still, they are in the lowerrange of values reported for other estuarine or shallow coastal envi-ronments (Risgaard-Petersen, 2003; Sundbäck et al., 2000). At stationGiralda low photosynthetic activity by benthic microalgae was aprobable consequence of elevated turbidity associated to the fresh-water inflow from the Po di Volano Canal and limited light penetra-tion, even in the shallow water column overlying sediments duringincubations (Viaroli et al., 2006). At station Gorino, low chlorophylla concentrations is surface sediments and low benthic productionby MPB are likely due to the shading effect of drifting macroalgalmats (Valiela et al., 1997).

Table 3Annual oxygen, dissolved inorganic nitrogen and N2 fluxes measured via intact coresand chambers incubation and inorganic nitrogen uptake calculated from net andgross benthic production (average values±standard deviation, n=3 to 5, arepresented).

Flux (mol m−2y−1) Giralda MPB Gorino MPB Gorino MA

Measured viaincubation

O2 −11.16±1.24 −19.00±0.59 24.35±2.28

Measured viaincubation

NH4+ 1.92±0.70 1.74±0.28 −5.02±1.14

NO2− 0.64±0.20 0.06±0.07 −0.04±0.09

NO3− −1.33±0.76 −0.97±0.30 −5.33±0.63

DIN 1.23±1.05 0.83±0.41 −10.39±1.30Calculated fromO2 data

Net DINuptake

0.26±0.14 0.004±0.004 6.52±1.38

Gross DINuptake

1.36±0.84 0.89±0.41 10.18±2.31

Measured viaIPT

N2 2.27±0.06 0.85±0.02 0.83±0.01

AtGiralda, calculatedNBP varied between 0 and 360 mgC m−2 d−1

while GBP varied between 59 and 754 mg C m−2 d−1; at Gorino NBPvaried between 0 and 12 mg C m−2 d−1 while GBP varied between48 and 629 mg C m−2 d−1. Integrating NBP and GBP rates over theentire period of this study we calculated a net and gross carbon fixa-tion by MPB of 28 and 145 g C m−2 y−1 at Giralda and of 0.4 and96 g C m−2 y−1 at Gorino.

Net productionmeasured in the presence of MAwas one order ofmagnitude higher compared to that of MPB and varied between0.75 and 3.67 g C m−2 d−1, whilst GBP varied between 1.60 and4.41 g C m−2 d−1. These productivity values fall between rangespreviously reported for macrophyte communities (Bartoli et al.,2008; Dalsgaard, 2003; Welsh et al., 2000a). The benthic systemwith MA, on a yearly basis, displayed an high GBP to BR ratio and wasnet autotrophic; net and gross carbon fixation by the sediment+Ulvacommunity was 562 and 855 g C m−2 y−1, respectively.

At Giralda, N uptake calculated from NBP varied between 0 and326 μmol N m−2 h−1 while that calculated from GBP varied between60 and 541 μmol N m−2 h−1. At Gorino, nitrogen uptake in sedi-ments with benthic microalgae calculated from net primary produc-tivity values was 0 in 7 out of 8 sampling dates while that calculatedfrom GBP rates varied between 19 and 412 μmol N m−2 h−1. Theo-retical nitrogen uptake in sediments with Ulva varied between 958and 3850 μmol N m−2 h−1 (from NBP) and between 2045 and4619 μmol N m−2 h−1 (from GBP). Integrating theoretical N assim-ilation calculated from NBP over the entire period of this study gives0.26±0.14 and 0.004±0.004 mol N m−2 y−1 for Giralda andGorino, respectively (sediments with benthic microalgae) and 6.52±1.38 mol N m−2 y−1 for sediments with Ulva at Gorino (Table 3). The-oretical DIN assimilation calculated from GBP was instead 1.36±0.84and 0.89±0.41 mol N m−2 y−1 (Giralda andGorino, respectively, sed-iments with MPB) and 10.18±2.31 mol N m−2 y−1 (Gorino, sedi-ments with MA).

During the period of this investigation the pool of nitrogen in thestanding stock ofUlvawas between aminimumof 0.11 and amaximumof 1.02 mol N m−2 (annual average 0.44 mol N m−2) and was >40times higher than that immobilised in benthic algae. Assuming formicrophytobenthos a C/Chl a ratio of 30 and a C/N ratio of 9(Sundbäck et al., 2000 and references therein) we calculated that inthe period of this study the pool of N retained in themicrophytobenthosbiomass was between 3.3 and 19.7 mmol N m−2 (annual average10.5 mmol N m−2) at Giralda and between 2.8 and 5.8 mmol N m−2

(annual average of 4.4 mmol N m−2) at Gorino.Nitrogen turnover inUlva varied between 3 and 54 days (annual av-

erage 17.5 days), N turnover in microphytobenthos was faster, with asimilar range at the two sites (1–21 days) and annual averages of 7.8and 7.2 days at Gorino and Giralda, respectively. These N turnoverrates are similar to those previously reported for microphytobenthosand seagrass communities (Nielsen et al., 2004; Risgaard-Petersen etal., 1998; Sundbäck and McGlathery, 2005).

4.2. Benthic respiration and denitrification rates

Sediments at stations Giralda and MA-free sediments at stationGorino displayed elevated dark oxygen demand, with measuredrates (from −1.51 to −6.60 mmol O2 m−2 h−1) in the higherrange of those reported for other estuaries (Nizzoli et al., 2007). Theannual average (−71 mmol O2 m−2 d−1, pooled data from the twosites) more than doubles the average calculated from about 50 estua-rine areas (−34 mmol O2 m−2 d−1, Hopkinson and Smith, 2005).High benthic respiration is sustained by labile carbon inputs to sur-face sediments from both allochthonous and autochthonous sourcesand by elevated densities of macrofauna. At Giralda sediments werealways bioturbated and the organic load was probably allochthonousand generated from settled particles of fluvial origin, as phytoplank-ton blooming in the nutrient-rich terminal reach of the Volano

49M. Bartoli et al. / Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

Canal. At Gorino, the sedimentary load of organic matter was deter-mined by settled fragments of Ulva, as reflected by the diversifiedmacrofauna community found at this site. High variability of oxygenconsumption occurs from site to site within the same estuary,depending upon water temperature, salinity, organic matter contentof sediments, its macromolecular quality and also the density ofbioturbating fauna. We acknowledge that with 3 to 5 core replicateswe have explored only 150–250 cm2 of sediment surface, and therisk for large variability associated to patchy distribution of benthicmacrofauna is elevated (Glud and Blackburn, 2002). This problem isless relevant when using large chambers as those employed for sedi-ments with MA, where explored surface for light and dark fluxes wasabout 1200 cm2.

Sediments at Giralda and MA-free sediments at Gorino had compa-rable dark respiration, with annual averages overlapping (2.86±1.21and 3.05±1.75 mmol O2 m−2 h−1, respectively). If MA are included,benthic respiration rates at Gorino increase by 130%, with an annual av-erage of 6.98 mmol O2 m−2 h−1. Within the Sacca di Goro lagoon,higher rates of dark oxygen uptake are reported only for sediments col-lected within areas licenced for molluscs farming (up to 20 mmolO2 m−2 h−1, Bartoli et al., 2001; Nizzoli et al., 2005).

In terms of molar ratios, denitrification rates measured at the twosites represented on average a small fraction of sediment oxygen de-mand. At Giralda, in 8 out of 9 samplings, total denitrification variedbetween 2 and 7% of BR. The sampling of May 1997 was an exception,as total denitrification represented about 67% of dark oxygen fluxes.Such peak is probably due to sudden increase of the density ofC. insidiosum, that has a demonstrated role in enhancing oxygentransfer to surface sediments and nitrification coupled denitrification(Pelegrí et al., 1994). At Gorino, in all samplings, denitrification repre-sented from b1 to 8% of BR. Oxygen consumed via nitrification, calculat-ed as the sum of 2Dn+2NOx

− efflux, represented a variable fraction ofBR, in particular at Giralda, where it varied between 1 and 80% (annualaverage 25%). At Gorino nitrification represented a smaller fraction ofBR (between b1 and 10%, with an annual average of 6%).

We estimated carbon oxidation by denitrification assuming 1.25 C:1N (mol:mol) (Richards, 1965) to evaluate the contribution of denitrifi-cation to organic matter mineralization. We assumed for stationsGiralda and Gorino a respiratory quotient (RQ=TCO2:O2 dark fluxes)close to unity (Hargrave et al., 2008). Total denitrification rates mea-sured were responsible for an extremely variable production of inor-ganic carbon, between 78 and 1440 μmol C m−2 h−1 at Giralda(average of pooled data 300 μmol C m−2 h−1) and between 38 and210 μmol C m−2 h−1 at Gorino (average of pooled data 117 μmolC m−2 h−1). Assuming that BR, corrected for the amount of oxygenused by nitrifiers, is a proxy for organic carbon mineralization, we esti-mated that the fraction of carbon oxidation due to denitrification aver-aged 19% and 7% at Giralda and Gorino, respectively. Similar results,showing higher relative importance of denitrification as a mineraliza-tion path in the more fresh estuarine area, are reported by Yoon andBenner (1992) and Laursen and Seitzinger (2002).

Eyre and Ferguson (2009) defined denitrification efficiency (DE) asthe ratio of inorganic nitrogen efflux from sediment as N2 to thetotal efflux of DIN+N2. They noticed that highest DE (~70%)characterised sediments with inorganic carbon fluxes between 500and 1500 μmol m−2 h−1, under a moderate organic carbon enrich-ment. As in the Sacca di Goro we generally measured respirationrates higher than 1500 μmol O2 m−2 h−1 we expected much lowerDE and elevated N recycling. Denitrification efficiencies calculatedwith data from Giralda and Gorino (only sediments without MA)were very erratic and did not correlate with oxygen demand; DE an-nual average was 59% at Giralda and 42% at Gorino. Values varied atboth sites between a minimum of 5%, calculated in autumn 1997,and a maximum of 100% (May, July, December 1997 and February1998 at Giralda; August 1997 and January 1998 at Gorino). Weargue that at those sites studied by Eyre and Ferguson denitrification

was mainly supported by Dn while denitrification of water columnnitrate was of minor importance due to very low NO3

− concentrations.Under such circumstances an organic enrichment stimulates benthicrespiration and ammonium production via ammonification, but simul-taneously limits nitrification, resulting in higher NH4

+ efflux and lowerdenitrification coupled to nitrification. In the Sacca di Goro, nitrate inthe water column was abundant at both sites and organic enrichmenthad a minor effect on macrofauna, composed by tolerant species, andon Dn. At organic-rich sites with elevated concentrations of NO3

− inthe water column as is the Sacca di Goro, a thinner layer of oxic sedi-ment can stimulate Dw, resulting in higher N2fluxes. The erratic patternof denitrification efficiency is thus the probable consequence ofco-occurring situations and may depend, among other factors, on ni-trate levels in the water, organic input, life cycle of benthic macrofauna,water temperatures and rates of ammonification. Denitrification effi-ciency drops in eutrophic environments when processes such as nitrateammonification become favoured over denitrification, a demonstratedsituation occurring in the Sacca di Goro in the sediments below musselropes (Nizzoli et al., 2006).

4.3. Annual net fluxes of oxygen and inorganic nitrogen and denitrificationrates

On an annual basis sediments with MPB were net heterotrophic atboth sites, with a more negative oxygen balance at Gorino (Table 3).Heterotrophy was coupled to net release of DIN to the water columnat both sampling sites, mostly sustained by ammonium regeneration.As the activity of MPB was low, calculated net inorganic nitrogen up-take at the sediment–water interface was only a minor nitrogen sinkand had a negligible role as a trap for regenerated nutrients (Table 3).Nitrogen removal via denitrification largely exceeded theoretical ni-trogen uptake by MPB at both sites and was a significant N sink inparticular at station Giralda. Here, calculated annual denitrificationrate was approximately three fold higher than that calculated at Gorinolikely due to higher nitrate concentrations in the water column, higherbioturbation by burrowing fauna and higher nitrification. Denitrificationis not considered a significant sink of nitrogen in shallow lagoons withphotosynthetically active benthic populations when external nutrientloading is low to moderate as primary producers outcompete bacteriafor DIN (Rysgaard et al., 1995). Risgaard-Petersen (2003) suggested thatautotrophic sediments with MPB display lower rates of Dn than net het-erotrophic sediments. Sediments at station Gorino and Giralda were netheterotrophic and denitrification rates were demonstrated to be muchhigher that theoretical N uptake by benthic microalgae.

Sediments with MA were net autotrophic to hyperautotrophic,with oxygen production largely exceeding dark respiration andwere a large nitrogen sink (Table 3). Net annual fluxes of ammoniumand nitrate were both directed towards the sediment+MA benthicsystem and comparable in magnitude. Overall, measured DIN fluxesfell between theoretical nitrogen uptake estimated from net andgross MA production (Table 3). Annual inorganic nitrogen fluxesmeasured at station Giralda and Gorino are in the higher extreme ofthose reported in analogous studies (Dalsgaard, 2003; Sundbäck etal., 2000). Tyler et al. (2003) performed an annual study at four sta-tions within an estuary where dissolved inorganic and organic nitro-gen fluxes across sediments with benthic microalgae and macroalgaewere simultaneously measured. They have found that net autotrophicsediments were a net sink of nitrogen due to uptake at the interface.Net heterotrophic sediments on the contrary released nitrogen thatwas taken up in the benthic system by macroalgal mats, preventingits diffusion into the water column. They have estimated that Nregenerated from sediments accounted for 27–75% of macroalgal up-take. In the Sacca di Goro lagoon our annual estimates suggest thatdespite elevated regeneration from sediments, released nitrogenaccounted for less than 20% of nitrogen requirements by Ulva, mean-ing that most nitrogen was taken from the water column. Large

50 M. Bartoli et al. / Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

amounts of nitrogen are retained temporarily in macroalgal tissues,that act as a short-term nutrient buffer; however, most of this nitro-gen is released during senescence over periods of several weeks toseveral months (Tyler et al., 2001). In the Sacca di Goro, DIN uptakeby MA was equally divided into ammonium and nitrate uptake,reflecting similar availability of the two ions in the water column.

Annual nitrogen removal via denitrification in sediments with MAwas about 10 fold lower than rates of nitrogen uptake by macroalgaebut was similar to the amount measured in MA-free sediments. Thismeans thatmacroalgal uptake did not significantly influence nitrogen re-moval via denitrification, probably due to ample availability of inorganicnitrogen at the study site. There are a few studies reporting rates of deni-trification in sediments covered bymacroalgae. LaMontagne et al. (2002)and Trimmer et al. (2000b) found low rates of denitrification in sedi-ments beneath macroalgae, likely due to reducing organic sedimentsand inhibitory effect of sulphide to nitrifiers and denitrifiers (Sloth et al.,1995; Viaroli et al., 1996). Results from the present study suggest thatrates of denitrification were lower at the more marine station Gorino,likely due to a combination of factors including sediment redox, nitrateavailability and bioturbation, but not due to the presence of macroalgae.

Rates of denitrification in temperate coastal bays range between 2.1and 16.7 μmol N m−2 h−1 (measured in rootedmacrophyte meadows),between 12 and 55 μmol N m−2 h−1 (sediment beneath macroalgalmats) and between 0.02 and 60 μmol N m−2 h−1 (sediments with ben-thic microalgae) (McGlathery et al., 2007). Although dentrification ratesreported in the present study are elevated if compared to values reportedin other estuarine areas, they are comparatively lower than uptake ratesby Ulva and effluxes from sediments.

5. Conclusions

This study has three main outcomes: the first deal with the minorrole of microphytobenthos on benthic nitrogen cycling in the eutro-phic Sacca di Goro lagoon, which can be addressed to water turbidityand to the presence of drifting macroalgae. The second outcome is therelevance of Ulva as main oxygen source and (temporary) nitrogensink in the lagoon, through its photosynthetic activity and nutrientuptake. The third important result is about a missing, not evident ormasked effect of primary producers on nitrogen losses via denitrifica-tion. This process is quantitatively relevant in the Sacca di Goro andnot suppressed by macroalgal activity probably due to the large avail-ability of inorganic nitrogen in the water column and in surfacesediments.

A final remark deals with the elevated temporal (and spatial) var-iability of measured processes in this eutrophic lagoon, a probableconsequence of multiple, interacting factors that have a complex reg-ulation of nitrogen cycling. Sudden blooms of macroalgae can switcha heterotrophic, net recycling system into an autotrophic, sink one aswell as short term increase of burrowing macrofauna can exponen-tially enhance nitrogen losses via denitrification. Experimental andmonitoring activities in complex systems as coastal lagoons shouldas a consequence be properly designed to catch the correct temporaland spatial scale of analysis.

Acknowledgements

Thiswork is a contribution to the Eloise programme (Eloise no. 361/6)within the frameworks of the NICE (contract no. MAS3-CT96-0048) pro-ject. We thank two anonymous reviewers that provided very helpfulcomments on the manuscript. [SS]

References

Bartoli, M., Nizzoli, D., Viaroli, P., Turolla, E., Castadelli, G., Fano, E.A., Rossi, R., 2001. Im-pact of Tapes philippinarum farming on nutrient dynamics and benthic respirationin the Sacca di Goro. Hydrobiologia 455, 203–212.

Bartoli, M., Nizzoli, D., Viaroli, P., 2003. Microphytobenthos activity and fluxes at the sed-iment–water interface: interactions and spatial variability. Aquat. Ecol. 37, 341–349.

Bartoli, M., Nizzoli, D., Castaldelli, G., Viaroli, P., 2008. Community metabolism andbuffering capacity of nitrogen in a Ruppia cirrhosa meadow. J. Exp. Mar. Biol. Ecol.360, 21–30.

Bower, C.E., Holm-Hansen, T., 1980. A salicylate-hypochlorite method for determiningammonia in seawater. Can. J. Fish. Aquat. Sci. 37, 794–798.

Burgin, A.J., Hamilton, S.K., 2007. Have we overemphasized the role of denitrification inaquatic ecosystems? A review of nitrate removal pathways. Front. Ecol. Environ. 5,89–96.

Burkholder, J.M., Tomasko, D.A., Touchette, B.W., 2007. Seagrasses and eutrophication.J. Exp. Mar. Biol. Ecol. 350, 46–72.

Dalsgaard, T., 2003. Benthic primary production and nutrient cycling in sediments withbenthic microalgae and transient accumulation of macroalgae. Limnol. Oceanogr.48, 2138–2150.

Dalsgaard, T., Nielsen, L.P., Brotas, V., Viaroli, P., Underwood, G.J.C., Nedwell, D.B.,Sundbäck, K., Rysgaard, S., Miles, A., Bartoli, M., Dong, L., Thornton, D.C.O.,Ottosen, L.D.M., Castaldelli, G., Risgaard-Petersen, N., 2000. Protocol Handbookfor NICE-Nitrogen Cycling in Estuaries: A Project Under the EU ResearchProgramme. Marine Science and Technology (MAST III), National EnvironmentalResearch Institute, Silkeborg, Denmark.

Dalsgaard, T., Thamdrup, B., Canfield, D.E., 2005. Anaerobic ammonium oxidation(anammox) in the marine environment. Res. Microbiol. 156, 457–464.

Diaz, R.J., Rosenberg, R., 2008. Spreading dead zones and consequences for marineecosystems. Science 321, 926–929.

Eyre, B.D., Ferguson, A.J.P., 2009. Denitrification efficiency for defining critical loads ofcarbon in shallow coastal ecosystems. Hydrobiologia 629, 137–146.

Focht, D.D., Verstraete, V., 1977. Biochemical ecology of nitrification and denitrification.Adv. Microb. Ecol. 1, 135–214.

Galloway, J.N., Dentener, F.J., Capone, D.G., Boyer, E.W., Howarth, R.W., Seitzinger, S.P.,Asner, G.P., Cleveland, C.C., Green, P.A., Holland, E.A., Karl, D.M., Michaels, A.F.,Porter, J.H., Townsend, A.R., Vorosmarty, C.J., 2004. Nitrogen cycles: past, present,and future. Biogeochemistry 70, 153–226.

Glud, R.N., Blackburn, N., 2002. The effect of chamber size on in situ benthic oxygen up-take measurements: a simulation study. Ophelia 56, 23–31.

Golterman, H.L., Clymo, R.S., Ohnstand, M.A.M., 1978. Methods for Physical and Chem-ical Analysis of Fresh Waters. I.B.P. Handbook Nr. 8. Blackwell, Oxford.

Hargrave, B.T., Holmer, M., Newcombe, C.P., 2008. Towards a classification of organicenrichment in marine sediments based on biogeochemical indicators. Mar. Pollut.Bull. 56, 810–824.

Hopkinson Jr., C.S., Smith, E.M., 2005. Estuarine respiration: an overview of benthic,pelagic and whole system respiration. In: delGiorgio, P.A., Williams, P.J.B. (Eds.),Respiration in Aquatic Ecosystems. Oxford University Press, Oxford.

Koop-Jakobsen, K., Giblin, A., 2009. Anammox in tidal marsh sediments: the role of sa-linity, nitrogen loading, and marsh vegetation. Estuaries Coasts 32, 238–245.

Krause-Jensen, D., Christensen, P.B., Rysgaard, S., 1999. Oxygen and nutrient dynamicswithin mats of the filamentous macroalga Chaetomorpha linum. Estuaries 22, 31–38.

LaMontagne, M., Astorga, V., Giblin, A.E., Valiela, I., 2002. Denitrification and the stoi-chiometry of nutrient regeneration in Waquoit Bay, Massachusetts. Estuaries 25,272–281.

Laursen, A.E., Seitzinger, S.P., 2002. The role of denitrification in nitrogen removal andcarbon mineralization in Mid-Atlantic Bight sediments. Cont. Shelf Res. 22,1397–1416.

Lorenzen, C.J., 1967. Determination of chlorophyll and phaeo-pigments: spectrophotometricequations. Limnol. Oceanogr. 12, 343–346.

McGlathery, K.J., Sundbäck, K., Anderson, I.C., 2004. The importance of primary producersfor benthic nitrogen and phosphorous cycling. In Estuarine nutrient cycling: the in-fluence of primary producers. Aquat. Ecol. Ser. 2, 231–261.

McGlathery, K.J., Sundbäck, K., Anderson, I.C., 2007. Eutrophication in shallow coastalbays and lagoons: the role of plants in the coastal filter. Mar. Ecol. Prog. Ser. 248,1–18.

Middelburg, J.J., Levin, L.A., 2009. Coastal hypoxia and sediment biogeochemistry.Biogeosciences 6, 1273–1293.

Naldi, M., Viaroli, P., 2002. Nitrate uptake and storage in the seaweed Ulva rigidaC. Agardh in relation to nitrate availability and thallus nitrate content in a eutro-phic coastal lagoon (Po River Delta, Italy). J. Exp. Mar. Biol. Ecol. 269, 65–83.

Nielsen, L.P., 1992. Denitrification in sediment determined from nitrogen isotopepairing. FEMS Microbiol. Ecol. 86, 357–362.

Nielsen, S.L., Banta, G.T., Pedersen, M.F. (Eds.), 2004. Estuarine Nutrient Cycling: The Influ-ence of Primary Producers. Kluwer Academic Publishers, Dordrecht, The Netherlands.

Nizzoli, D., Welsh, D.T., Bartoli, M., Viaroli, P., 2005. Impact of mussel (Mytilusgalloprovincialis) farming on oxygen consumption and nutrient recycling in a eutro-phic coastal lagoon. Hydrobiologia 550, 183–198.

Nizzoli, D., Welsh, D.T., Fano, E.A., Viaroli, P., 2006. Impact of clam and mussel farmingon benthic metabolism and nitrogen cycling, with emphasis on nitrate reductionpathways. Mar. Ecol. Prog. Ser. 315, 151–165.

Nizzoli, D., Bartoli, M., Cooper, M., Welsh, D.T., Underwood, G.J.C., Viaroli, P., 2007. Im-plications for oxygen, nutrient fluxes and denitrification rates during the earlystage of sediment colonisation by the polychaete Nereis spp. In four estuaries. Estu-arine Coastal Shelf Sci. 75, 125–134.

Ottosen, L.D.M., Risgaard-Petersen, N., Nielsen, L.P., 1999. Direct and in directmeasurements of nitrification and denitrification in the rhizosphere of aquaticmacrophytes. Aquat. Microb. Ecol. 19, 81–91.

Pelegrí, S.P., Nielsen, L.P., Blackburn, T.H., 1994. Dentrification in estuarine sedimentstimulated by the irrigation activity of the amphipod Corophium volutator. Mar.Ecol. Prog. Ser. 105, 285–290.

51M. Bartoli et al. / Journal of Experimental Marine Biology and Ecology 438 (2012) 41–51

Revsbech, N.P., Jorgensen, B.B., Brix, O., 1981. Primary production of microalgae in sed-iments measured by oxygen microprofile, H14CO3 fixation, and oxygen exchangemethods. Limnol. Oceanogr. 26, 717–730.

Richards, F.A., 1965. Anoxic basins and fjords. In: Ryley, J.P., Skirrow, G. (Eds.), Chemi-cal Oceanography. Academic Press, London.

Risgaard-Petersen, N., 2003. Coupled nitrification–denitrification in autotrophic andheterotrophic estuarine sediments: on the influence of benthic microalgae. Limnol.Oceanogr. 48, 93–105.

Risgaard-Petersen, N., Rysgaard, S., 1995. Nitrate reduction in sediments and water-logged soil measured by 15N techniques. In: Alef, Nannipieri (Eds.), Methods inApplied Soil Microbiology and Biochemistry. Academic Press.

Risgaard-Petersen, N., Rysgaard, S., Nielsen, L.P., Resvsbech, N.P., 1994. Diurnal varia-tion of denitrification and nitrification in sediments colonized by benthicmicrophytes. Limnol. Oceanogr. 39, 573–579.

Risgaard-Petersen, N., Dalsgaard, T., Rysgaard, S., Christensen, P.B., Borum, J.,McGlathery, K., Nielsen, L.P., 1998. Nitrogen balance of a temperate eelgrass Zosteramarina bed. Mar. Ecol. Prog. Ser. 174, 281–291.

Risgaard-Petersen, N., Nielsen, L.P., Rysgaard, S., Dalsgaard, T., Meyer, R.L., 2003. Applicationof the isotope pairing technique in sediments where anammox and denitrification co-exist. Limnol. Oceanogr. Methods 1, 63–73.

Rysgaard, S., Christensen, P.B., Nielsen, L.P., 1995. Seasonal variation in nitrification anddenitrification in estuarine sediment colonized by benthic microalgae andbioturbating infauna. Mar. Ecol. Prog. Ser. 126, 111–121.

Schramm, W., Gualberto, E., Orosco, C., 1984. Release of dissolved organic matter frommarine tropical reef plants: temperature and desiccation effects. Bot. Mar. 27,71–77.

Sloth, N.P., Blackburn, H., Hansen, L.S., Risgaard-Petersen, N., Lomstein, B., 1995. Nitro-gen cycling in sediments with different organic loading. Mar. Ecol. Prog. Ser. 116,163–170.

Strickland, J.D.H., Parsons, T.R., 1972. A Practical Handbook of Sea-water Analysis, 2ndedition. J Fish Res Bd Canada.

Sundbäck, K., McGlathery, K., 2005. Interactions between benthic macroalgal andmicroalgal mats. Macro- and Microorganisms in Marine Sediments: Coastal andEstuarine Studies, 60.

Sundbäck, K., Miles, A., 2002. Role of microphytobenthos and denitrification for nutri-ent turnover in embayments with floating macroalgal mats: a spring situation.Aquat. Microb. Ecol. 30, 91–101.

Sundbäck, K., Miles, A., Göransson, E., 2000. Nitrogen fluxes, denitrification and the roleof microphytobenthos in microtidal shallow-water sediments: an annual study.Mar. Ecol. Prog. Ser. 200, 59–76.

Sundbäck, K., Miles, A., Hulth, S., Pihl, L., Engström, P., Selander, E., Svenson, A., 2003.Importance of benthic nutrient regeneration during initiation of macroalgalblooms in shallow bays. Mar. Ecol. Prog. Ser. 246, 115–126.

Trimmer, M., Nedwell, D.B., Sivyer, D.B., Malcolm, S.J., 2000a. Seasonal benthic organicmatter mineralization measured by oxygen uptake and denitrification along atransect of the inner and outer River Thames estuary, UK. Mar. Ecol. Prog. Ser.197, 103–119.

Trimmer, M., Nedwell, D.B., Sivyer, D.B., Malcolm, S.J., 2000b. Seasonal organicmineralisation and denitrification in intertidal sediments and their relationshipto the abundance of Enteromorpha sp. and Ulva sp. Mar. Ecol. Prog. Ser. 203, 67–80.

Trimmer, M., Risgaard-Petersen, N., Nicholls, J.C., Engstrom, P., 2006. Direct measure-ments of anaerobic ammonium oxidation (anammox) and denitrification in intactsediment cores. Mar. Ecol. Prog. Ser. 326, 37–47.

Tyler, A.C., McGlathery, K.J., Anderson, I.C., 2001. Macroalgal mediation of dissolved or-ganic nitrogen fluxes in a temperate coastal lagoon. Estuarine Coastal Shelf Sci. 53,155–168.

Tyler, A.C., McGlathery, K.J., Anderson, I.C., 2003. Benthic algae control sediment–watercolumn fluxes of organic and inorganic nitrogen compounds in a temperate lagoon.Limnol. Oceanogr. 48, 2125–2137.

Valiela, I., McLelland, J., Hauxwell, J., Behr, P.J., Hersh, D., Foreman, K., 1997. Macroalgalblooms in shallow estuaries: controls and ecophysiological and ecosystem conse-quences. Limnol. Oceanogr. 42, 1105–1118.

Viaroli, P., Christian, R.R., 2003. Description of trophic status, hyperautotrophy and dys-trophy of a coastal lagoon through a potential oxygen production and consumptionindex — TOSI: Trophic Oxygen Status Index. Ecol. Indic. 3, 237–250.

Viaroli, P., Bartoli, M., Bondavalli, C., Christian, R.R., Giordani, G., Naldi, M., 1996.Macrophyte communities and their impact on benthic fluxes of oxygen, sulphideand nutrients in shallow eutrophic environments. Hydrobiologia 329, 105–119.

Viaroli, P., Bartoli, M., Azzoni, R., Giordani, G., Mucchino, C., Naldi, M., Nizzoli, D., Tajé,L., 2005. Nutrient and iron limitation to Ulva blooms in an eutrophic coastal lagoon(Sacca di Goro, Italy). Hydrobiologia 550, 57–71.

Viaroli, P., Giordani, G., Bartoli, M., Naldi, M., Azzoni, R., Nizzoli, D., Ferrari, I., Zaldívar,Comenges, J.M., Bencivelli, S., Castaldelli, G., Fano, E.A., 2006. The Sacca di Goro lagoonand an arm of the Po River. In: Wangersky, P.J. (Ed.), The Handbook of EnvironmentalChemistry: Estuaries, vol. 5/H, pp. 197–232.

Viaroli, P., Bartoli, M., Giordani, G., Naldi, M., Orfanidis, S., Zaldivar, J.M., 2008. Communityshifts, alternative stable states, biogeochemical controls and feedbacks in eutrophiccoastal lagoons: a brief overview. Aquat. Conserv. 18, 105–117.

Villares, R., Carballeira, A., 2003. Seasonal variation in the concentrations of nutrients intwo green macroalgae and nutrient levels in sediments in the Rı ́as Baixas (NWSpain). Estuarine Coastal Shelf Sci. 58, 887–900.

Welsh, D.T., 2000b. Nitrogen fixation in seagrass meadows: regulation, plant–bacteriainteractions and significance to primary productivity. Ecol. Lett. 3, 58–71.

Welsh, D.T., Bartoli, M., Nizzoli, D., Castaldelli, G., Riou, S., Viaroli, P., 2000a. Deni-trification, nitrogen fixation, community primary productivity and inorganic-N and oxygen fluxes in an intertidal Zostera noltii meadow. Mar. Ecol. Prog.Ser. 208, 65–77.

Wetzel, R.G., Likens, G.E., 1991. Limnological Analyses, 2nd ed. Springer-Verlag.Yoon, W.B., Benner, R., 1992. Denitrification and oxygen consumption in sediments of

two south Texas estuaries. Mar. Ecol. Prog. Ser. 90, 157–167.