in situ oxygen dynamics and carbon turnover in an … set up to provide data on local weather...
TRANSCRIPT
MARINE ECOLOGY PROGRESS SERIESMar Ecol Prog Ser
Vol. 566: 49–65, 2017doi: 10.3354/meps12016
Published February 27
INTRODUCTION
Sand- and mudflats represent productive transitionzones at the land−sea interphase and provide impor-tant ecosystem services (Lotze et al. 2005, Reise2005). In both settings, microphytobenthos (MPB) arethe key primary producers and their activity regu-lates a range of biogeochemical processes and func-
tions (Cammen 1991, Middelburg et al. 2000, Sund-bäck et al. 2000). This includes modulating O2 avail-ability, the sediment redox status, and sediment sta-bility and permeability through production of exo-polymeric substances (Paterson 1989, Andersen et al.2010), which are all factors that influence rates andpathways for nutrient regeneration and carbon turn-over (De Wit et al. 1989, Risgaard-Petersen 2003, Bar-
© The authors 2017. Open Access under Creative Commons byAttribution Licence. Use, distribution and reproduction are un -restricted. Authors and original publication must be credited.
Publisher: Inter-Research · www.int-res.com
*Corresponding author: [email protected]
In situ oxygen dynamics and carbon turnover inan intertidal sediment (Skallingen, Denmark)
Eva Walpersdorf1, Michael Kühl2,3, Bo Elberling1, Thorbjørn J. Andersen1, Birger U. Hansen1, Morten Pejrup1, Ronnie N. Glud4,5,6,*
1Department of Geosciences and Natural Resource Management, University of Copenhagen, Øster Voldgade 10, 1350 Copenhagen K, Denmark
2Marine Biological Laboratory, University of Copenhagen, Strandpromenaden 5, 3000 Helsingør, Denmark3Plant Functional Biology and Climate Change Cluster, University of Technology Sydney, Ultimo, NSW 2007, Australia
4University of Southern Denmark, Institute of Biology and Nordic Center for Earth Evolution (NordCEE), Odense M, Denmark5Department of Ocean and Environmental Sciences, Tokyo University of Marine Science and Technology, 4-5-7 Konan,
Minato-ku, Tokyo 108-8477, Japan6Scottish Association for Marine Science, Dunstaffnage Marine Laboratory, PA37 1QA Dunbeg, UK
ABSTRACT: Intertidal areas are considered to be productive habitats, but due to the challengingand dynamic environment, in situ assessments of their performance remain scarce. Here, weaimed to quantify the production and turnover of organic material in an intertidal sediment acrossseveral consecutive diel/tidal cycles and to evaluate the importance of key drivers affecting thecommunity performance. Time-series measurements of the oxygen (O2) microdistribution across 2sites at a silty sandflat documented extreme variability, which was mainly driven by light avail-ability and flow conditions. Diffusion dominated the interstitial solute transport, but advectiveporewater transport was observed during 15% of the 9 d study period. Photosynthetic activitynever exhibited light inhibition and average daytime net photosynthesis ranged between 1.01 and11.15 mmol m−2 d−1 depending on the daily light availability, while the assessed gross primary pro-duction was ∼80% higher. When irradiance exceeded 17.2 mol photons m−2 d−1, daytime net auto-trophy more than balanced the heterotrophic activity during the night (average night-time respi-ration, ERn: −4.01 mmol m−2 d−1). However, despite intense primary production, the integratedactivity during the target autumn period was net heterotrophic with an average net ecosystemmetabolism of −2.21 (span: −7.93 to 1.48) mmol O2 m−2 d−1. This study highlights the extreme tem-poral and spatial variation of intertidal sediments and the importance of accounting for natural insitu dynamics to correctly assess their performance and contribution to system production.
KEY WORDS: In situ O2 dynamics · Transecting microprofiler · Benthic primary production ·P/E relation · Flow · Net ecosystem metabolism · Intertidal
OPENPEN ACCESSCCESS
50
toli et al. 2003). However, intertidal sediments arealso very dynamic due to highly fluctuating environ-mental conditions, where inundation, light, tempera-ture, salinity, and particle redistribution can changeon timescales of minutes to months depending ondiurnal, tidal, and seasonal cycles. In high-energycoastal settings, tidal sediments are dominated bysand and advective porewater transport that facili-tate efficient biocatalytic transformation of trappedorganic material (Ehrenhauss & Huettel 2004, Ehren-hauss et al. 2004, Cook et al. 2007, Anschutz et al.2009). In more protected areas, the tidal flats aredominated by diffusion-controlled silty sedimentswith high carbon content.
A major challenge working in these complex het-erogeneous and dynamic environments is to deter-mine primary production and respiration withoutaltering the environmental controls. Reliable meas-urements have to account for spatial and temporalvariations in light, temperature, inundation, hydro-dynamics, and biological factors that are all respond-ing to phase-shifted diel and tidal cycles. Most stud-ies quantifying the carbon turnover of intertidal flatshave employed sediment enclosure approaches thatare bound to affect many of the controls regulatingbenthic activity, and which have difficulties in cap-turing the activity over complete tidal cycles (e.g.Spilmont et al. 2007, Migné et al. 2009). Therefore,carbon budgets for intertidal environments are oftenbased on measurements conducted under artificialsettings without accounting for natural variability.Only few integrative and non-invasive in situ studiesof intertidal sediments have been conducted in eitherhighly permeable sands (e.g. de Beer et al. 2005,Werner et al. 2006) or in muddy sediments (Denis &Desreumaux 2009, Denis et al. 2012). But even thesedetailed investigations have only partly been able todetermine the metabolic state of their respective set-tings, and it is still not clear whether intertidal sedi-ments are sources or sinks for CO2 (e.g. Cook et al.2004, Klaassen & Spilmont 2012). In the presentstudy, we applied an autonomous transecting profil-ing instrument equipped with an array of microelec-trodes to quantify the in situ microscale O2 distribu-tion in silty sandflats over a series of consecutive dieland tidal cycles. This approach enabled us to quan-tify benthic activity during concurrent ambientchanges in inundation, light availability, tempera-ture, flow conditions, and bed level. The data wereused to non-invasively quantify the production andmineralization of organic material at in situ condi-tions, and to evaluate the relative importance of keycontrols for the benthic O2 dynamics of intertidal silty
sands. Findings are discussed in the context of avail-able studies on carbon turnover in temperate inter-tidal sediments.
MATERIALS AND METHODS
Study site and sediment characteristics
The investigations were carried out from 26 Sep-tember to 4 October 2007 (autumn) on an intertidalsandflat in the northern part of the Danish WaddenSea in the shelter of the Skallingen barrier spit
Mar Ecol Prog Ser 566: 49–65, 2017
0 8 24 m
Stn A Stn B
Stn A
Stn B
9°8°E
55°N
8°E 9°
55°32’N
8°20’E
GERMANY
DENMARK
Esbjerg
North Sea
FANØ
km0 1 2 A
B
C
Fig. 1. (A,B) Study area (from Andersen 2001) showing thepresence of mud flats (black), mixed mud flats (dark grey),sand flats (light grey) and the position of the field site (redsquare) at low tide. (C) The transect off the coast with 2 mainmeasurement locations: Stn A, 32U 0453592, 6154129 (UTMcoordinate system) and Stn B (with profiler and acousticDoppler velocimeter [ADV]), 0453609, 6154099. Stn A re-mains water covered during low tide (blue horizontal line
indicates the water level at low tide)
51
(Fig. 1). Freshwater input from the river Varde re -sulted in salinity variations between 23 and 30 athigh tide. The area is microtidal, and maximumwater depths ranged between 0.3 and 0.9 m depend-ing on tidal stage and wind impact. Inundation timevaried between 3.5 and 7 h, but even during low tidethe surface sediment remained water-saturated andcovered with a thin layer of water. A climate stationwas set up to provide data on local weather condi-tions. Irradiation at the site was recorded from the af-ternoon of 1 October onwards throughout the fieldmeasurements, while irradiance data for the preced-ing period were taken from a nearby climate station(15 km). The recorded downwelling short-wave irra-diance (in W m−2) was converted into photosyntheti-cally active radiation (PAR; 400 to 700 nm) in units ofµmol photons m−2 s−1 by calibration against parallellight measurements taken with a light meter (LI-250;LiCor) equipped with an underwater photon irradi-ance sensor (Li 192A; LiCor) positioned at the sedi-ment surface. Measurements were conducted at 2 lo-cations, 8 and 24 m from the marsh edge (Stn A: 32U0453592, 6154129 and Stn B: 32U 0453609, 6154099;UTM coordinate system) to assess any potentialmesoscale difference in sediment performance. Thesediment at both stations consisted of slightly muddy(∼5% silt and clay), fine to medium well-sorted sili-cate sand grains with a mean grain size (geometricmean, method of moments) of 194 to 271 µm in theupper 15 cm (determined using forward laser scatter-ing by use of a Malvern Mastersizer/E, 1 core, 10 cmdiameter). A total of 2 cores station−1 (3.6 cm diame-ter) were taken for the determination of porosity andtotal organic carbon (TOC). Porosity was calculatedfrom the weight loss of wet sediment after drying andthe measured sediment density. Sediment perme-ability was not measured during the campaign, butthe site was revisited in August 2010 and sedimentpermeability was quantified by the constant head ap-proach (Klute & Dirksen1986) and amounted to(mean ± SD) 1.0 × 10−12 ± 2.5 × 10−13 m2 (n = 3) and 3.8× 10−12 ± 6.3 × 10−13 m2 (n = 3) at Stns A and B, respec-tively. The sediments were therefore characterizedby relatively low permeability (Forster et al. 2003).
Sediments for the analysis of TOC were freeze-dried, ground, and acidified with 1 mol l−1 HCl to re -move inorganic carbon. Subsequently, TOC was de -termined in sediment washed and dried at 105°Cusing a LECO TruSpec carbon nitrogen determinator.Samples for chlorophyll a (chl a) were taken with cut-off syringes (1.9 and 2.8 cm inner diameter), slicedimmediately in the field and stored frozen until analy-sis. Chl a was determined by extraction in 96%
ethanol for 24 h and spectrophotometry (Biochrom Li-bra S12) at 665 and 750 nm and again 2 to 3 min afteracidification with 1 mol l−1 HCl to degrade all chloro-phyll into phaeopigments. Chl a and phaeo pigmentcontent was calculated according to Lorenzen (1967).Macrofauna present in the upper 10 cm of the sedi-ment was sampled in a 10 cm diameter core at eachsite to give a qualitative impression of the communitycomposition only. The fauna, retained on a 500 µmsieve, was dominated by Hydrobia ulvae and nema-todes, but at Stn A small juvenile polychaetes (Nereissp. and Scolelepis sp.) and a few bi valves (Macomasp.) of variable sizes were also ob served. Sedimentcharacteristics are summarized in Table 1.
Hydrodynamics and bed level change
The 3-dimensional (3D) flow velocity at Stn B wasmeasured in a 0.25 cm3 water volume situated 6.5 cmabove the bed by a SonTek/YSI 10 MHz acousticDoppler velocimeter (ADV) at 10 Hz; average datafrom periods of 30 s were stored every 10 min. Waterdepth, salinity, temperature, and turbidity weremeasured with an Aanderaa RCM9 current meterwith sensors placed 10 cm above the bed.
Walpersdorf et al.: Intertidal benthic photosynthesis, drivers and NEM
Depth Stn Stn (cm) A B
Mean grain sizea (µm) 0−1 208 194
Mud (%) 4.5 4.6
Porosity 0−0.5 0.48 0.46 0.5−1 0.52 0.41
TOCb (%) 0−0.5 0.20 0.13 0.5−1 0.18 0.11
Chlorophyll a (µg g−1) 0−0.2 14.4c 17.7 ± 2.5 (n = 7)
0.2−0.4 12.2c 16.0 ± 1.2 (n = 7)
Hydrobia ulvae (ind. m−2) 0−10 1020 382
Nereis sp. (ind. m−2) 0−10 1150 0
Scolelepis sp. (ind. m−2) 0−10 255 0
Nematodes (ind. m−2) 0−10 16 100 7670aGeometric mean as calculated by the method ofmoments
bTotal organic carbon, salt free dry weightc114–117 mg chl a m−2 (range, n = 2) in total contained inthe upper 4 mm at Stn A
Table 1. Selected sediment characteristics of the study site.Numbers represent single determinations (grain size) orthe average of 2 measurements; standard deviation is
given for chl a
Mar Ecol Prog Ser 566: 49–65, 201752
In situ sensor measurements of dissolved oxygen
In situ measurements of O2 concentration versusdepth profiles in the sediment were done with atransecting microprofiler (Glud et al. 2009a). Theinstrument was placed on the tidal flat during lowtide with the microsensors initially positioned 0.5 to1 cm above the sediment surface. The profilingmodule was equipped with 2 or 3 O2 sensors and acustom-made 2-wire resistivity sensor (1 mm tipdiameter) for detection of the sediment surface posi-tion. The O2 sensors were Clark-type microsensors(Revsbech 1989) with a sensing tip diameter of~10 µm, a t90 (the time it takes for the sensor signalto reach 90% of the final value) response time of<2 s, and a stirring sensitivity <2% (Gundersen etal. 1995). The microsensor response was linearlycalibrated from sensor readings in the overlyingwater (sampled over the duration of the inundationperiod), and in the anoxic part of the sediment. Dis-solved O2 concentration in the ambient seawaterwas determined on-site by Winkler titration (Grass -hoff et al. 1999). Vertical profiling into the sedimentwas performed in steps of 200 µm covering a totaldistance of 2 cm; at each depth, the signals werelogged. After the profiler had moved the microsen-sor tips back to the start position above the sedimentsurface and prior to the next measurement, a motormoved the sledge with the instrument horizontally7 mm prior to the next set of profile measurements.This distance eliminated any potential oxygenationand other artefacts from holes created during previ-ous measurements and thus al lowed undisturbeddepth profiles of O2. Each set of O2 depth profileswas recorded within 20 min with a programmedresting time of 2 min prior to the next profilingevent. These measurements were conducted duringdeployment times of 14 to 24 h, yielding up to 60profiles sensor−1 covering a total lateral distance ofmaximally 47.5 cm; the distance between the indi-vidual O2 microsensors was 2 to 4 cm. The tempera-ture at the position of the microprofiler was re -corded in air and in the surface sediment (+1, −1,−2, −5 cm) using Tinytag data loggers and zero-offcalibrated thermistor probes. Data were recordedevery 30 s during the deployments.
During low tide, additional O2 microprofiles weremeasured manually to complement the automatedprofiling lander measurements. For these measure-ments, the microsensors were positioned by use of amanually operated micromanipulator (MM33; März -häuser), while the sensor signal was read from apicoammeter (PA2000; Unisense).
Modelling of diffusive O2 exchange
Diffusive O2 fluxes in the sediment were calculatedfrom complete in situ profiles measured under quasisteady-state conditions using the software packagePROFILE (Berg et al. 1998). This procedure was pre-ferred rather than the more usual approach based onthe concentration gradient within (or just below) thediffusive boundary layer (Jørgensen & Revsbech1985) due to the relatively coarse resolution at whichthe profiles were obtained. The relative position ofthe sediment surface for the respective profiles wasidentified from a distinct break in the slope of theconcentration profile (Glud et al. 1995). The profileanalysis software calculates the rate of net produc-tion or consumption as a function of depth togetherwith the flux across the sediment−water interface.The procedure is based on a control volume ap -proach of equally spaced zones, and uses a series ofleast square fits to adapt the model to the measuredconcentration profile, followed by comparisons ofthese fits through statistical F-testing. This enablesselection of the simplest production−consumptionprofile that best reproduces the measured concentra-tion profile. Boundary conditions chosen for diffusivetransport modelling were the O2 concentration andflux at the bottom of the calculation domain. Sedi-ment diffusivity (Ds) was assumed constant withdepth and was calculated as: Ds = D / (1 + 3 [1 − ϕ])(Iversen & Jørgensen 1993), where D is the moleculardiffusion coefficient of O2 in water corrected for insitu temperature and salinity (Li & Gregory 1974) andϕ is the porosity of the surface sediment (Table 1). Forthe iteration of finding the best fits, maximum devia-tion when accepting a calculated minimum was set to0.001%, the level of significance (p-value) used inthe F-statistics was 0.01. Daylight O2 production wascalculated from a production/light (P/E) relationshipestablished for the entire study period according tothe formula P = Pmax {1 − [exp(−αE /Pmax)]} + R (Plattet al. 1980), where P = net production (in mmol m−2
d−1), Pmax = photosynthetic capacity (in mmol O2 m−2
d−1), α = initial slope of net production at subsaturat-ing light levels, E = incident light at the sediment sur-face (in µmol photons m−2 s−1), R = dark flux (in mmolO2 m−2 d−1).
Light penetration depth
Light measurements in the sediment were performedwith a fiber-optic scalar irradiance microprobe with asmall diffusing/light-scattering sphere (<100 µm dia -
Walpersdorf et al.: Intertidal benthic photosynthesis, drivers and NEM 53
meter) cast at the tapered tip (Rickelt et al. 2016). Themicroprobe was coupled to a custom-built fiber-opticlight meter measuring PAR (400 to 700 nm) (Kühl et al.1997). The linear detector re sponse was calibrated us-ing the sensor signal in seawater at known down-welling irradiance and in darkness.
Light profiles within the sediment were measuredin the laboratory on subsamples of water-saturatedsurface sediment from the site. A fiber-optic tungstenhalogen lamp equipped with a collimating lens(Schott KL-2500; 3100 K) illuminated the sedimentsample vertically from above, while the scalar irradi-ance microprobe was mounted in a motorized micro-manipulator (Unisense) and was inserted into thesediment at an angle of 40° relative to the incidentlight to prevent self-shading. Motor control and re -cording of the sensor readings was performed usingthe software package Profix (PyroScience). A total of3 to 4 profiles down to 4.2 mm depth were measuredin each sample at a depth resolution of 200 µm, with5 s waiting time at each depth and signal integrationover 1 s.
The attenuation coefficient for photon scalar irradi-ance, K0(PAR), within the photic zone was calculatedas: K0(PAR) = −d [lnE0(PAR)]/dz, where z is depth,by a linear regression on ln-transformed data overthe respective linear depth interval (Kühl 2005).
RESULTS
Benthic oxygen dynamics and regulating factors
Time-series of microprofiles covering >76 h and 7tidal cycles demonstrated an extensive spatio-tempo-ral variation in the sediment O2 distribution (Figs. 2 &3). Parallel sensor measurements reflected the sameoverall pattern, but also revealed microscale patchi-ness (Figs. 2 & 3).
The prime factor affecting benthic O2 distributionwas light availability. The sediment surface exhibiteda light dependent O2 production and the O2 concentra-tion occasionally reached values up to ~950 µmol l−1
(3.0 times air saturation) during mid-day flood tide.This activity enhanced the O2 concentrations in theoverlying water up to ~340 µmol l−1 (~120% air satura-tion). However, O2 concentrations as high as ~650 µmoll−1 O2 (~230% air saturation) were en coun tered in theoverlying water during low tide, when the sedimentwas covered by a layer of ~1−2 cm of almost stagnantwater. During the night, respiration and re-oxidationprocesses reduced the thickness of the oxic zone, andthe O2 concentration in the overlying water declined to
~200 µmol l−1 (~70% air saturation). Beside this overallpattern, short-term variations in solar irradiance dueto changing cloud cover quickly led to changes in theO2 distribution within the sediment. For instance, themaximum O2 porewater concentration at high tideduring the afternoon of 29 September declined by~300 µmol l−1 within 22 min in response to dense cloudcover, which reduced the ambient irradiance at thesediment surface from 930 to 340 µmol photons m−2 d−1
(Figs. 2 & 3C). An example of the opposite dynamic wasobserved at Stn A on 3 October during low tide, whenthe O2 concentration of the overlying water increasedby~200µmoll−1within22minasambientlightlevelsin-creasedfrom190to280µmolphotonsm−2d−1 (Fig. 2).
While the salinity-induced changes in O2 solubilitywere small, temperature variation affected the O2
availability. During the study period, temperatures atthe sediment−water interface varied between 9.6and 17.1°C, which roughly corresponds to a solubilitydriven change in O2 concentration of 40 µmol l−1 atsaturation. Generally, minimum and maximum tem-peratures were reached at ebb-tide, but variationduring a single tidal cycle only led to a maximum sol-ubility driven change of 20 µmol O2 l−1. Thus, maxi-mally 3 to 20% of the observed change in dissolvedO2 concentration could be attributed to salinity ortemperature-induced changes in O2 solubility.
Apart from light and temperature, other physicalfactors also affected the benthic O2 distribution andconcentration. Despite low O2 concentration at thesediment surface during night-time, these periodsalso reflected the deepest O2 penetration depth (OPD)(Figs. 2 & 3B). Such apparently counter-intuitive ob-servations were caused by transient tide-induced ad-vective flow and hydraulic-driven percolation ofwater through the surface sediments, especially atnight when the diatom coverage typically becomesless distinct due to downward cell migration (deBrouwer & Stal 2001, Du et al. 2010). This most likelyenhanced the permeability of the sediment surfaceduring night-time, as observed in other studies (e.g.Westall & Rincé 1994, Cahoon 1999, Paterson &Hagerthey 2001). During such periods, the benthic O2
availability appeared to be strongly affected by theambient average flow (Fig. 4). In these instances, con-vection could have contributed to the observed pore-water mixing, but we have no direct means to assessthis potential effect. Suspension of the fluffy sedimentmaterial was only observed occasionally, and thenjust as the rising tide reached the study area or duringperiods where flow exceeded 0.1 m s−1 (Fig. 4). In par-allel with the microsensor measurements, the ADVcontinuously recorded the relative position of the bed
Mar Ecol Prog Ser 566: 49–65, 201754
level at a given location with a precision of ±1 mm. Ingeneral, bed levels were constant, but in a few inci-dences we observed a stepwise shift of 1 or 2 mmwhich also was reflected in rapid O2 changes as in-duced by micro scale sedimentation and erosion(Fig. 2). Fauna-induced advective O2 transport as ex-pressed by subsurface maxima or abrupt erratic fluc-tuations in the O2 concentration distribution duringprofiling (Glud et al. 2003, Jørgensen et al. 2005, de
Beer et al. 2005) was occasionally observed (Figs. 2 &3D). Based on this, we estimated that only 3% of themicro profiles were affected by macrofauna activity.
Diffusion was the prevailing interstitial transportmode in the targeted silty sands as judged by the factthat most of the time the obtained microprofiles couldbe modelled by PROFILE using pure diffusive inter-stitial solute transport (Fig. 3; R2 ≥ 0.99). As evaluatedfrom the entire data set, these conditions prevailed
13:17 15:53 18:29 21:05 23:41 02:17 04:53 07:29 10:05
Time of day
13:17 15:53 18:29 21:05 23:41 02:17 04:53 07:29 10:05
0
200
400
600
800
1000
Light
OPD
1000
800
600
400
200
0
0.0
0.2
0.4
0.6
0.8
1.0
12:27 13:39 14:51 16:03 17:16 18:28 19:40 20:53 22:05
Wat
er le
vel (
m)
Time of day
Water level
Mean current
1.0
0.8
0.6
0.4
0.2
0.0
Light
OPD_1
OPD_2
12:27 22:05 20:53
0
2
4
6
8
10
12
19:40 18:28 17:16 16:03 14:51 13:39
0.15
0.10
0.05
0
Water level Mean current
Ligh
t (µ
mol
pho
tons
m–2
s–1)
Time of day (h:min)
Wat
er le
vel (
m)
Stn B 26 Sep Stn B 29/30 Sep
Dep
th (m
m)
Time of day (h:min)
A B
C
D E
F G
O2
pen
etra
tion
dep
th (m
m)
Cur
rent
(m s
–1)
O2
(µm
ol l–1
)O
2 (µ
mol
l–1)
Dep
th (m
m)
Fig. 2 (continued on next page). Temporal variation in O2 distribution at the sediment−water interphase monitored at 4 time in-tervals. At one occasion, only 1 sensor was deployed (A), while the other measurements were done with 2 parallel sensors(B,C; H,J; and I,K). Concurrent measurements of light availability and O2 penetration in the sediment are shown in (D,E,L,M),
while (F,G,N,O) show water levels and current speed at the measuring site. Note differences in scale
Walpersdorf et al.: Intertidal benthic photosynthesis, drivers and NEM
85% of the time. Advection only contributed to thebenthic O2 dynamics during periods of intensedrainage and higher flow velocities (Figs. 3B & 4).Fig. 4 depicts the typical response to increased bot-tom water flow, with the significant increase in OPDonly visible during high tide periods under dark orvery low-light conditions when average current in -creased (>0.05 m s−1) due to elevated wind speeds(>7 m s−1).
During the study period, the sediment was alwayswater-saturated and vertical PAR attenuation meas-ured in the laboratory was well described with a
mono-exponential attenuation function using aK0(PAR) value of −2.19 mm−1 (±0.04, R2 = 0.99). Max-imum light penetration (<0.01% of surface irradi-ance) was 4.2 mm under maximal incident irradiancelevels of ~750 µmol photons m−2 s−1. However, scalarirradiance at the sediment surface increased to~930 µmol photons m−2 s−1 due to intense light scat-tering in the sand grains (Fig. 3C). Applying the min-imum and maximum in situ irradiance measurementsduring our campaign with the K0 value found in thelaboratory measurements, we calculated photic zonethickness (defined as the depth where only 1% of
55
0.00
0.05
0.10
0.15
13:54 16:40 19:25 22:10 00:56 03:41 06:27 09:12 11:5811:39 14:31 17:22 20:13 23:04 01:55 04:46 07:37 10:29
Time of day
Water level Mean current
0
2
4
6
8
10
120
200
400
600
800
1000
1.0
0.8
0.6
0.4
0.2
0.0
1000
800
600
400
200
0
11:39 10:29 07:37
0
2
4
6
8
10
12
04:46 01:55 23:04 20:13 17:22 14:31 13:54 16:40 19:25 22:10 00:56 03:41 06:27 09:12 11:58
0.15
0.10
0.05
0
Light
OPD_1
OPD_2
J
Stn B 2/3 Oct Stn A 3/4 Oct
H I
K
L M
N O
Ligh
t (µ
mol
pho
tons
m–2
s–1)
Time of day (h:min)
Wat
er le
vel (
m)
Dep
th (m
m)
Time of day (h:min)O
2 p
enet
ratio
n d
epth
(mm
)C
urre
nt (m
s–1)
O2
(µm
ol l–1
)O
2 (µ
mol
l–1)
Dep
th (m
m)
Fig. 2 (continued)
56 Mar Ecol Prog Ser 566: 49–65, 2017
–5
–4
–3
–2
–1
0
10 100 200 300 400
22:4323:0500:3402:04Model
–5
–4
–3
–2
–1
0
10 200 400 600 800
12:5613:4114:0214:25ModelLight 14:02
Low
tid
e
29/30 Sep1 Oct
Hig
h tid
e
LightA B
C D
Darkness
Dep
th (m
m)
–0.2 0.0 0.2 0.4
14:02
–0.2 0.0 0.2 0.4
00:34
–0.4 0.0 0.4 0.8
13:41–5
–4
–3
–2
–1
0
–0.2 0.0 0.2 0.4
22:43–5
–4
–3
–2
–1
0
–0.2 0.0 0.2 0.4
12:56
O2 (nmol cm–3 s–1)
O2 (nmol cm–3 s–1) O2 (nmol cm–3 s–1)
O2 (nmol cm–3 s–1)
Dep
th (m
m)
–12
–10
–8
–6
–4
–2
0
20 100 200 300
19:1921:3407:3607:58Model
–5
–4
–3
–2
–1
0
10 200 400 600 800 1000
15:1915:4216:05modelLight 15:19Light 15:42
29 Sep
30 Sep
2 Oct
–0.2 0.0 0.2 0.4
15:42
–0.2 0.0 0.2 0.4
16:05–5
–4
–3
–2
–1
0
–0.2 0.0 0.2 0.4
15:19–5
–4
–3
–2
–1
0
–0.2 0.0 0.2 0.4
19:19
O2 (µmol l–1), irradiance (µmol photons m–2 s–1) O2 (µmol l–1)
Dep
th (m
m)
Dep
th (m
m)
O2 (µmol l–1), irradiance (µmol photons m–2 s–1) O2 (µmol l–1)
Fig. 3. Characteristic O2 depth distribution at the main study site during (A,B) low tide and (C,D) high tide at different times ofthe day and different light availability. (B,D) Night-time profiles showed the occurrence of hydraulic percolation (B), advectionor fauna ventilation (D) increasing the O2 penetration considerably. Volumetric production and consumption rates of O2 during
periods of diffusive solute transport were modelled using PROFILE (Berg et al. 1998) (R2 ≥ 0.99)
57
surface irradiance remained) to vary from 0.9 to3.3 mm during daytime. This corresponded well tothe zone of net-photosynthesis modelled by PROFILEas suming diffusive conditions, where these modelledvalues ranged between 0.5 and 3.0 mm with an aver-age value of 1.3 ± 0.5 mm (n = 80). Chl a levels of thesurface sediment were relatively high, reaching val-ues of 14.4 µg chl a g−1 at Stn A and 17.7 ± 2.5 µg chl ag−1 (mean ± SD) at Stn B (n = 7) (Table 1), and did notvary sig nificantly during the investigation period.These relatively high values are consistent with thehigh photo trophic activity, but clearly the sedimen-tary chl a content does not necessarily reflect activeMPB biomass.
Autotrophic activity in the surface sediment
The depth-integrated net O2 production modelledfrom concentration profiles measured under differentirradiance levels during the investigation periodexhibited a typical saturation curve that could beapproximated with a hyperbolic tangent functionassuming no photoinhibition (Platt et al. 1980, Fig. 5).Microprofiles obtained manually and by the transect-ing lander showed very similar results, indicating
that the instrument did not affect the local microenvi-ronment. There was no difference between dataobtained at the 2 stations or at different tides, andthus data were compiled for further analysis (Fig. 5).Photosynthesis continued during high tide undercalm conditions and a clear water column. Migrationof activity zones (net primary production, NPP) wasnot indicated by consecutive in situ microprofilemeasurements during daytime. This indicates thatpotential vertical migration of MPB during daytimewas minor. The compensation irradiance for thedepth-integrated net photosynthesis at which thesystem shifts from heterotroph to autotroph (commu-nity compensation point, CCP) was 45 µmol photonsm−2 s−1, while Pmax was 38 mmol O2 m−2 d−1, and theirradiance at onset of saturation (Pmax/α) equalled220 µmol photons m−2 s−1. Analysis of 80 concentra-tion profiles showing oxygen production indicatedthat, on average, 55% of the produced O2 was
Walpersdorf et al.: Intertidal benthic photosynthesis, drivers and NEM
y = 347.03x2 – 5.84x + 2.00R2 = 0.56
0
1
2
3
4
5
6
7
0.00 0.04 0.08 0.12
O2
pen
etra
tion
dep
th (m
m)
Current (m s–1)
2 Oct evening3 Oct morning3 Oct evening4 Oct morning
Fig. 4. Measured O2 penetration depth as a function of meancurrent speed for 2 calm (2 October, morning; 3 October,evening) and 2 rough periods (3 October, evening; 4 Octo-ber, morning) when current speeds exceeded 0.1 m s−1 andmaximum velocities (not shown) reached 0.3 m s−1, at whichpoint the sediment began to be unstable. Two macrofauna-impacted and 1 very disturbed profile (3 October, evening)were omitted from the analysis. Values represent measuredO2 penetration depths with 2 sensors during high tide atnight or low irradiance levels <100 µmol photons m−2 s−1
(and mostly below the sediment compensation irradiance CCP = 45 µmol photons m−2 s−1)
0 200 400 600 800 1000
Surface irradiance (µmol photons m–2 s–1)
–20
0
20
40
O2
pro
duc
tion
(mm
ol m
–2 d
–1)
Fig. 5. Diffusive O2 exchange, modelled from 109 in situ O2
concentration profiles, as a function of downwelling irradi-ance fitted using the function: P (net primary production) =Pmax {1 − [exp(−αE/Pmax)]} + R (Platt et al. 1980), with Pmax
(photsynthetic capacity) = 38.3 mmol O2 m−2 d−1, α (initialslope at subsaturating light levels) = 0.177 mmol O2 m−2 d−1
[µmol photons m−2 s−1]−1 and R = −7.12 mmol O2 m−2 d−1 (R2 =0.95; n = 109). Black (grey) circles: low (high) tide productioncalculated from micro profile measurements at Stn B; whitecrosses: manually recorded profiles; white squares: low tidemeasurements at the water-covered Stn A; orange circles:net O2 exchange at the primary interphase; black crosses:daytime O2 consumption (ERd) below the photic zone. Thetrend-line is produced by: ERd = –[[ERd-max {1 − [exp(−αE/ERd-max)]} + R]], where ERd-max = 10.5 mmol m−2 d−1, α = 0.073
and R = 0.487 mmol m−2 d−1 (R2 = 0.44; n = 75)
58
released to the overlying water, while 45% was con-sumed in the deeper sediment layers (Fig. 5).
The average O2 consumption in deeper sedimentlayers was clearly stimulated by photosynthesis atthe sediment surface, as reflected by the volume-spe-cific O2 consumption rate (Rv) modelled for 109 in situmicroprofiles using PROFILE (Fig. 6). For instance, Rv
increased by up to ∼200% in the depth interval be -tween 0.1 and 0.2 mm during daytime. The strongestincrease was observed just below the average depthof the photic zone, and activity was in the same orderof magnitude as consumption rates in the upper mmof the sediment in darkness. This indicates highmicrobial activity and re-oxidation of reduced end-products of anaerobic carbon degradation at theoxic−anoxic interface.
In darkness, Rv varied between −0.024 and−0.079 nmol O2 cm−3 s−1 with maximal values at thesediment surface. During daytime, the values for thedeeper oxic sediment layers varied between −0.044and −0.072 nmol O2 cm−3 s−1 with the lowest activitylevels encountered in the deepest parts of the con-centration profiles. The error bars indicate a high lat-eral variability of Rv. Assuming that the average O2
consumption rate derived for the surface layers dur-ing periods where diffusion prevailed is similar to theO2 consumption rate in deeper layers that are occa-sionally oxygenated by large-scale physical advec-tion, we can estimate the importance of advection forO2 consumption. In general, advection extended theoxic volume of the non-photic sediment by 153%.However, periods of advection/percolation only oc -
cur red during 15% of the measuring time, andmainly at night. Consequently, the estimated advec-tive-driven O2 uptake calculated with an averagenight-time Rv of −0.055 nmol O2 cm−3 s−1 and an ex -tension of the oxic zone by ∼3.1 mm only accountedfor a daily average of −2.08 (−1.33 to −3.76) mmol m−2
d−1 or 22% of the integrated daily net O2 consump-tion for the study period (Table 2).
DISCUSSION
Benthic diatoms are regarded as significant pri-mary producers of intertidal ecosystems (MacIntyreet al. 1996, Underwood & Kromkamp 1999) yet quan-titative in situ estimates of their importance in naturaldynamic settings are few (Billerbeck et al. 2007,Denis & Desreumaux 2009, Denis et al. 2012). By ap -plying a transecting profiling instrument, we wereable to monitor in situ sediment O2 dynamics andenvironmental effects on benthic primary productionin intertidal fine-sand sediments during a 9 d period.These data enable a discussion on carbon turnover innatural and dynamic settings of a microtidal systemand the assessment of the key factors regulating theactivity.
Effect of advective flow and fauna on O2 dynamics
Low-energy conditions prevailed during our inves-tigations, which is typical for the targeted area that ischaracterized by relatively fine-grained sedimentswith a fluffy layer of muddy fine sand at the surface(Vinther et al. 2004). Despite relatively low sedimentpermeability (1.0 × 10−12 to 3.8 × 10−12 m2) and prevail-ing diffusive solute transport in the interstitial pore -waters, advective porewater transport was ob servedwhen current speeds exceeded 0.05 m s−1 (Fig. 4). Inone instance (4 October, Stn A), the wind speedreached >12 m s−1 and the average flow velocity in-creased to 0.1 m s−1 during incoming tide. This in-duced persistent advective porewater transport andrising concentrations of suspended matter in the over-lying water from 26 to 77 mg l−1. During this period,the O2 penetration depth increased by 105% to an av-erage (± SD) of 4.1 ± 1.2 mm, leading to a 158% in -crease in the O2 consumption rate for a period of 4.8 h.Integrated over the daily period, this event alone in-creased total net O2 consumption for this day by 23%.During the study period, advective porewater trans-port occurred 15% of the time. If we include activitiesin zones occasionally oxygenated by advection/
Mar Ecol Prog Ser 566: 49–65, 2017
–0.10–0.08–0.06–0.04–0.020.00
0–0.05
0.05–0.1
0.1–0.15
0.15–0.2
0.2–0.25
0.25–0.3
0.3–0.35
Dep
th (c
m)
Night Day
Volumetric O2 consumption (nmol cm s–1)
Fig. 6. Weighted average volumetric O2 consumption rates(Rv) in the surface sediment (nmol cm−3 s−1) including con-sumption rates from 80 daytime (white bars) and 29 night-time profiles (grey bars). The modelled Rv in light only in-cluded the sections below sediment layers that exhibited netproduction, while rates in darkness included the entire oxic
zone. Error bars: SE (n = 22 to 104)
59
percolation or convection, the daily benthic O2 uptakewas enhanced by −2.08 mmol m−2 d−1 and the net eco-system metabolism (NEM) during the investigationperiod becomes slightly negative with an average of−2.21(1.48 to −7.93) mmol O2 m−2 d−1 (Table 2, Fig. 7).The estimate on the relative importance of advec-tion/percolation/convection for the overall O2 con-sumption rates is, however, conservative. Mainlysince the Rv of the deeper sediment layers, given ac-cumulated reduced substances (i.e. iron sulfide, FeS)below the prevailing oxic zone, may be higher thanthe applied values derived from the surface sediments.Several intertidal and subtidal studies have shownthat porewater advection markedly increases OPD,O2 consumption, and aerobic mineralization rates (deBeer et al. 2005, Billerbeck et al. 2006, Werner et al.2006, Cook et al. 2007, Berg et al. 2013; see alsoour Table 3). However, these investigations targetedhighly permeable sediment, with high interstitial sup-ply of organic matter and O2 facilitated by a well-de-veloped rippled surface topography. At our microtidalstudy site, characterized by silty sand, the occasional
Walpersdorf et al.: Intertidal benthic photosynthesis, drivers and NEM
Date Irradiance O2 (mmol m−2 d−1) Tidal maximum Water Inundation(dd/mm) Mol photons Light ERn ERd Adv./Percol. NPP GPP NEM (daytime h) level time (m−2 d−1) (h) (night h) (light h) (24 h)a (light h) (light h) (24 h) Ebb Flood max. (m) (h d−1)
26/09 21.36 11.2 −3.67 −4.50 −1.50 11.15 18.85 1.48 09:29 03:34 0.55 12.2 22:19 15:39 0.84
27/09 19.90 11.0 −3.73 −4.54 −2.24 11.08 18.74 0.57 10:04 04:09 0.48 9.7 23:09 16:09 0.48
28/09 3.25 7.1 −4.30 −1.41 −1.33 2.42 5.63 −4.62 10:31 04:28 −0.14 0 23:03 16:51 −0.09
29/09 9.37 8.7 −4.19 −2.99 −1.41 6.27 11.63 −2.32 10:31 05:11 −0.11 5.9 23:59 17:54 0.62
30/09 13.06 9.6 −3.98 −3.74 −2.87 8.30 14.68 −2.30 12:09 06:19 0.81 13.5 18:24 0.9
01/10 11.87 10.4 −3.92 −4.02 −1.33 8.44 15.44 −0.83 00:59 06:24 0.54 9.5 12:49 19:09 0.47
02/10 17.30 10.7 −3.78 −3.99 −1.57 9.65 16.66 0.31 01:29 07:24 0.33 8.2 13:34 19:59 0.33
03/10 6.31 9.7 −3.98 −2.99 −2.75 5.45 11.13 −4.26 01:54 07:59 0.5 8.5 14:39 20:54 0.25
04/10 2.07 5.3 −4.59 −0.59 −3.76 1.01 2.89 −7.93 02:39 08:59 0.4 9.7 15:19 22:01 0.41
Average 11.6 9.3 −4.01 −3.20 −2.08 7.08 12.85 −2.21SD 7.0 2.0 0.30 1.38 0.87 3.61 5.60 3.01
a Additional O2 consumption due to tidal advection/percolation per day mainly at night, causing an average extension of theoxic zone by ∼3.1 mm (see ‘Results: Autotrophic activity in the surface sediment’ for details)
Table 2. Daily irradiance, light hours (h) with a light intensity above the community compensation point (CCP), diffusive O2 ex-change (ERn: night-time respiration, ERd: daytime respiration below the photic zone, NPP and GPP: net and gross primary produc-tion during light hours, advection/percolation and NEM: net ecosystem metabolism per 24 h during a period of consolidation.Fluxes were calculated for every minute of each day of the field campaign in September and October 2007 using a hyperbolic tan-gent function fitted to the production vs. irradiance curve. For further reference, times of peak ebb and flood tide and maximum
water level during inundation are included
–10
–8
R2 = 0.93
–6
–4
–2
0
2
4
0 4 8 12 16 20 24
NE
M (m
mol
O2
m–2
d–1)
( )
Irradiance (mol photons m–2 d–1)
Fig. 7. Relationship between 24 h time-integrated down-welling irradiance and net ecosystem O2 metabolism (NEM)including tidal advection under calm conditions. The 4thOctober (diamond in parentheses) was excluded from theanalysis because of stormy conditions prevailing during the
first high tide period of the day
60 Mar Ecol Prog Ser 566: 49–65, 2017
short-term delivery of O2 to greater depth as inducedby percolation and advection (convection) cannot tobe expected to stimulate aerobic respiration to thesame extent. More energetic conditions would pre-sumably have further enhanced the O2 uptake, espe-cially if high flow velocities were combined with waveaction and increased sediment re suspension (Vintheret al. 2004). Such conditions were not observed duringour study, but high-energy events can have quantita-tively large and long-lasting impacts on sediment re-oxidation, respiration, and photosynthesis (Jørgensen1996, Cook et al. 2007, Zetsche et al. 2012).
Besides hydrodynamics, faunal activity can alsoen hance the O2 exchange and the benthic OPD sig-nificantly. About 3% of the measured O2 micropro-files were visibly affected by macrofaunal activityand bioirrigation, which could have stimulated thebenthic O2 consumption (Glud et al. 2003). However,as only 2 sediment cores were recovered for qualita-tive assessment of fauna occurrence, a precise quan-titative assessment of the importance of faunal respi-ration and irrigation for the overall oxygen budgetwas not possible. But given the low macrofaunalabundance in these dynamic settings, the fauna-mediated O2 consumption was probably of minorimportance.
Primary production of an intertidal sand flat underlow-energy conditions
The established empirical relationship between so -lar irradiance and photosynthesis (P/E curve) (Fig. 5)is based on numerous in situ O2 microprofiles andincludes the natural spatial and temporal variabilityin environmental drivers. It was therefore possible tomake a relatively robust estimate of the in situ ben-thic primary production during the study period.Integrated over the daylight hours of the respectivedays, the NPP ranged between 1.01 and 11.15 mmolO2 m−2 d−1 depending on the light availability, withan average NPP of 7.08 mmol O2 m−2 d−1. The ecosys-tem respiration (ER) in darkness, ERn, integratedover the number of night hours ranged be tween−3.67 and −4.59 mmol m−2 d−1 with an average of−4.01 mmol O2 m−2 d−1 (Table 2).
To assess gross primary production (GPP) of ben-thic communities it is commonly assumed that ER re -mains light-independent (e.g. Spilmont et al. 2007,Denis & Desreumaux 2009, Glud et al. 2009b). Apply-ing this procedure to our data with GPP = NPP + |ER|(where |ER| includes respiration in darkness for thedepth of the photic zone and respiration below the
photic zone as measured during daytime) led to anaverage GPP of 1.31 mmol O2 m−2 h−1. Integrated forthe number of daylight hours, the GPP ranged be -tween 2.89 and 18.85 mmol O2 m−2 d−1 with an aver-age of 12.85 mmol O2 m−2 d−1 (Table 2). However, thisvalue must represent a minimum, given that ER inthe photic zone is presumably elevated during day-time (Kühl et al. 1996, Brotas et al. 2003, Glud et al.2009b) as also indicated by a stimulated daytime res-piration (ERd) below the photic zone (see Figs. 5 & 6).It is difficult to assess how much respiration was ele-vated within the photic zone during daytime, butprevious estimates range between 35 and 80%(Fenchel & Glud 2000).
Based on the microprofile measurements obtainedduring periods without porewater advection, the sed-iment was, on all days, net autotrophic during day-light, but integration over the diel periods (24 h) re -vealed that the system was almost at metabolicbalance with a slightly negative NEM of −0.13 mmolO2 m−2 d−1. As outlined above, including high energyperiods with porewater advection, NEM becameeven more heterotrophic with a net O2 exchange rateof −2.21 mmol m−2 d−1 (Table 2)
For maintaining balance between benthic het-erotrophy and autotrophy during calm periods, a min-imum of ∼12.0 mol photons m−2 d−1 was required. Ac -counting for advective transport processes (al thoughonly occurring 15% of the time), this value increasedto 17.2 mol photons m−2 d−1. The close correlation be-tween NEM and the daily integrated incident irradi-ance demonstrates the importance of light as the maincontrol factor for net system metabolism of intertidalsediments (Fig. 7). In conclusion, the system was al-most at metabolic balance during the autumn studyperiod. However, we cannot extrapolate the activitylevels beyond the study period, and the community-based P/E relationship presumably changed acrossseasons due to differences in light availability, tem-perature, grazing pressure, and sediment dynamics(Barranguet et al. 1998, Hancke & Glud 2004, Hubas& Davoult 2006, Migné et al. 2009). However, the pre-sented data document an ex tensive spatio-temporalvariation in O2 distribution that was mainly related tolight availability (i.e. phototrophic activity) duringdaylight hours. In addition, advection, resuspension,and flow dynamics strongly impacted the benthic ac-tivity. Macrofauna activity and temperature dynamicsduring the study period were presumably of minorimportance for benthic O2 exchange. However, all to-gether the data highlight the importance of applyingnon-invasive measuring approaches to correctly as-sess in situ productivity of intertidal systems.
61
Cross-study comparisons of NEM in intertidal sediments
Despite the importance of tidal environments forproduction and degradation of organic material, theiroverall trophic status is poorly established (Denis &Desreumaux 2009, Klaassen & Spilmont 2012). How-ever, provided that in situ dynamics are taken intoaccount, the net turnover of organic carbon can berobustly assessed by quantifying NEM via either O2
or CO2 exchange as integrated on a 24 h basis(Staehr et al. 2012). This provides an integratedmeasure of phototrophic and heterotrophic activity,including re-oxidation of reduced end-products fromquantitatively important anaerobic metabolic path-ways (Fenchel & Glud 2000). However, a cross-studycomparison is complicated by the use of differentmeasuring procedures and often an insufficientamount of information for calculating NEM.
Here, we have tried to standardize calculationsfrom available studies to provide such cross-studyana lysis of ER, NPP, GPP, and most importantly NEMof different temperate intertidal systems (Table 3).We analysed data from 12 comparable studies oftidally influenced sediments, of which 7 were con-ducted during the autumn (Table 3, values are givenin O2 concentrations). The ER varied more than 100-fold, from −0.06 mmol m−2 h−1 in fine sandy sediments(Roscoff Aber Bay, France) to approx. −7.5 mmol m−2
h−1 in permeable sandy sediments (Spiekeroog, Sylt,Germany). Clearly, the ER is highly dependent onadvective porewater transport and the ability of thepermeable sediments to act as a biocatalytic filter(Huettel et al. 2003). However, maintaining an ad -vective supply of organic material and O2 for ele-vated respiratory activity requires frequent resettingof the sediment by major resuspension/storm eventsto avoid clogging of the pore space, which otherwisewould reduce sediment permeability (Zetsche et al.2011, 2012). Nevertheless, the available studies inhighly permeable settings (Sylt, Spiekeroog; Table 3)only exhibited relatively low seasonal variability,with an average (± SD) NPP of 2.13 ± 1.26 mmol m−2
h−1. In contrast, a megatidal exposed sandy beach atWime reux (Spilmont et al. 2005, not included inTable 3) was characterized by a relatively low ben-thic metabolism with high variability during a 3 yrsurvey (GPP: 1.45 ± 3.40 mmol m−2 h−1; ER: 0.14 ±0.16 mmol m−2 h−1; mean ± SD), highly depending onmeteorological conditions. Calm conditions favouredmicroalgae blooms while windy periods favouredheterogeneous deposition of planktonic detritalmaterial (foam) causing short events of maximum
metabolic rates with a net productivity of up to 17.81mmol O2 m−2 h−1 or a net consumption of −1 mmol O2
m−2 h−1 (not shown). The im pact on the local carbonbudget was low, but is assumed to be significant at alarger spatial scale.
In low-energy intertidal systems with muddy andpresumably diffusion-controlled sediments, the aver-age ER amounted to −0.91 mmol m−2 h−1 (7 studies,Table 3). Most of these studies were conducted dur-ing autumn and exhibited an average daytime NPPrate of 0.61 mmol m−2 h−1 (5 studies). The single ob -servations ranged between −3.80 and 9.36 mmol m−2
h−1. However, a few long-term studies indicated con-siderable variability during specific days and alsobetween days as well as a seasonal maximum in pri-mary production during spring and summer (Cancheand Tagus estuaries, Table 3).
All sites exhibited considerable GPP, ranging from0.49 mmol m−2 h−1 at Neuharlinger Siel (Germany)(under light limitation during high tide) to 24.3 mmolm−2 h−1 in the Tagus Estuary (Portugal) on a brightsummer day. Overall, the few available studies did notindicate any specific relation between GPP and sedi-ment texture or any seasonality. Given the publishedinformation, NEM could only be extracted from 6 ofthe available studies and ranged from −140 mmol m−2
d−1 in the highly permeable sandy sediments at Sylt(Germany) (Table 3) to 9.71 mmol m−2 d−1 at a mudflatof the Seine Estuary (France). The ER, NPP, GPP, aswell as NEM derived for our micro tidal area inSkallingen exhibit net-heterotrophic conditions andfall well within the broad span of rates found in inter-tidal in situ studies conducted under calm conditionsduring autumn.
Generally, our review of relevant studies (Table 3)indicated net heterotrophy, with the exception beingsandy sites under low-energy conditions in theRoscoff Aber Bay (France), where sediments werefound to be autotrophic throughout the year (Hubas &Davoult 2006). Overall, the available data therebysuggest that despite considerable primary production,tidally influenced sediments appear to be net hetero-trophic with an average NEM value reaching−105 mmol m−2 d−1 for highly permeable sands and−3.00 mmol m−2 d−1 for sediments under low-energyconditions in autumn. The heterotrophic activity mustbe sustained by external organic matter delivered bytidal and riverine inflow and/or local primary produc-tion from the preceding summer months. However, itmust be emphasized that cross-seasonal extrapolationis problematic. Most available studies poorly resolveseasonal dynamics and tend not to include high ener-getic periods during storms or floods. Indeed, previous
Walpersdorf et al.: Intertidal benthic photosynthesis, drivers and NEM
62
studies have concluded that inmany cases it is difficult to resolveif intertidal settings are sourcesor sinks for C and nutrients (e.g.Middelburg et al. 1996, Cook etal. 2004, Cai 2011, Klaassen &Spilmont 2012). More long-termand seasonal non-invasive in situinvestigations are required to re-solve this question. Recently,aquatic eddy correlation (AEC,e.g. Berg & Huettel 2008, McGin-nis et al. 2014) has proven itselfas a very powerful, non-invasivetechnique for resolving benthicO2 ex change and metabolism inchallenging coastal environments(e.g. Glud et al. 2010, Attard et al.2014, Long et al. 2015). The ap-proach, however, will only workduring inundation, but combinedapplication of AEC and continu-ous auto nomous in situ micropro-filing measurements, as realizedhere, might represent an idealprocedure for resolving NEM indynamic intertidal settings and toassess the importance of such en-vironments for marine carbon cy-cling in general.
Acknowledgements. This study wasconducted within the framework ofthe project ‘Oxygen availability con-trolling the dynamics of buried or-ganic carbon pools and greenhousegas emissions’ financed by the DanishCouncil for Independent Research(FNU). E.W. and R.N.G. were addi-tionally supported by the FNU project0602-02276B, the Commission for Sci-entific Research in Greenland (KVUG,GCRC6507), the UK National Envi-ronmental Research Council (NERC,NE/ F1018612/ 1 and the EuropeanRe search Council (ERC-adG-2014-669947 HADES-ERC). We thank AnniGlud for providing the sensors and in-troduction in sensor maintenance,Paul Christiansen, Hanna Ferdinand,Mikkel Fruergaard and LasseKirkegaard for technical assistanceand help during field work. We ap-preciate the constructive anonymousreviews that helped us improving themanuscript.
Mar Ecol Prog Ser 566: 49–65, 2017
Loc
atio
n
T
emp
erat
ure
S
easo
n
E
R
NP
P
GP
P
N
EM
Tro
ph
ic
Par
amet
er/
Sou
rce
(
°C)
(m
mol
m−
2h
−1 )
(m
mol
m−
2h
−1 )
(m
mol
m−
2h
−1 )
(m
mol
m−
2d
−1 )
stat
e
met
hod
San
d
Wad
den
Sea
/ S
ylt
N
A
Dif
f. s
eas.
−7.
29 t
o −
4.38
2.92
7.2
9 to
10.
21
−
140
to −
70
H
O2/
MS
1
A
utu
mn
21.0
± 0
.7
Su
mm
er
−
7.88
to
−3.
30
O2/
MS
2
−
1.45
2.7
9
4.24
A
O
2/C
H
3
Wad
den
Sea
11–2
2
A
utu
mn
−7.
29 t
o −
1.50
O
2/M
S
4
Sp
iek
eroo
g
−
0.62
0.6
7
1.28
A
O
2/C
H
3
Ros
coff
Ab
er B
ay M
ediu
m s
and
NA
A
utu
mn
x
−0.
06 ±
0.0
0
0
.79
± 0
.00
0
.85
± 0
.00
0.
29 t
o 1.
91
A
C
O2/
CH
5, 6
Fin
e sa
nd
−
0.21
± 0
.01
1.8
1 ±
0.0
1
2.0
2 ±
0.0
2
0.39
to
5.33
A
CO
2/C
H
5
, 6
San
d/M
ud
Wad
den
Sea
/ S
kal
lin
gen
9.6
–17.
1
Au
tum
nx
−
0.27
± 0
.01
0.7
1 ±
0.2
9
1.3
1 ±
0.4
1
−
7.93
to
1.48
H
O
2/M
S
Th
is s
tud
yW
add
en S
ea/
Neu
har
lin
ger
Sie
l
NA
A
utu
mn
−
2.1
−1.
61
0
.49
H
O
2/C
H
3R
osco
ff A
ber
Bay
NA
A
utu
mn
x
−0.
68 ±
0.0
1
1
.01
± 0
.01
1
.70
± 0
.03
−10
.0 t
o −
4.07
H
C
O2/
CH
5, 6
Mu
dT
agu
s E
stu
ary
18
–31
Su
mm
er
−2.
4
3.0
to
5.1
A
O
2/C
H
7
−0.
9
3.4
to
6.5
8.7
to 2
4.3
O
2/M
S
Can
che
Est
uar
y
8.6
–22
Sp
rin
g
−1.
80 t
o −
1.08
−
1.44
to
9.36
0.7
4 to
12.
1
A−
H
O2/
MS
8
6.9
–19.
2
−
3.5
to −
0.4
−3.
80 t
o 6.
74
1.29
(0.
25 t
o 7.
49)
A
−H
O
2/M
S
9
Mon
t S
ain
t M
ich
el B
ay
N
A
Au
tum
nx
−
1.24
ito
−0.
24e
−
17.4
to
−12
.9
H
CO
2/C
H
10
Au
tum
n
−0.
14
0.
54 ±
0.1
5
1
1
Sei
ne
Est
uar
y
N
A
Au
tum
nx
−
1.44
± 0
.04
0.9
7 ±
0.1
0
2.4
1 ±
0.1
4
−
2.78
to
9.71
A
−H
CO
2/C
H
12
Tab
le 3
. Com
par
ison
of r
ecen
t in
sit
ust
ud
ies
of O
2/C
O2
flu
xes
(hou
rly
nig
ht r
esp
irat
ion
[ER
]; n
et a
nd
gro
ss p
rim
ary
pro
du
ctio
n [N
PP,
GP
P])
an
d m
etab
olic
sta
te (n
et e
co-
syst
em m
etab
olis
m,
NE
M)
of i
nte
rtid
al s
edim
ents
. F
or c
omp
aris
on,
rate
s in
mm
ol m
−2
d−
1or
nm
ol c
m−
2s−
1h
ave
bee
n c
onve
rted
to
hou
rly
rate
s, a
nd
mg
C t
o m
mol
C,
wh
ich
is a
pp
roxi
mat
ely
equ
al t
o m
mol
O2.
For
Ros
coff
Ab
er B
ay, M
ont
Sai
nt
Mic
hel
Bay
an
d t
he
Sei
ne
Est
uar
y, r
ates
hav
e b
een
cal
cula
ted
usi
ng
th
e fo
rmu
las
giv
en in
Hu
bas
et
al. (
2006
), M
ign
é et
al.
(20
09)
and
Sp
ilm
ont
et a
l. (
2006
). W
her
e p
ossi
ble
, hou
rly
and
dai
ly r
ates
are
giv
en f
or t
he
sam
e p
erio
d a
s in
th
is s
tud
y (2
6 S
epte
mb
er t
o4
Oct
ober
, au
tum
nx )
. NA
: not
ind
icat
ed, A
: au
totr
oph
, H: h
eter
otro
ph
, MS
: mic
rose
nso
rs, C
H: c
ham
ber
s, e
and
i: c
onsu
mp
tion
du
rin
g e
mer
sion
an
d im
mer
sion
. Sou
rces
:(1
) de
Bee
r et
al.
(200
5); (
2) W
ern
er e
t al.
(200
6); (
3) B
ille
rbec
k e
t al.
(200
7); (
4) B
ille
rbec
k e
t al.
(200
6); (
5) H
ub
as e
t al.
(200
6); (
6) H
ub
as &
Dav
oult
(200
6); (
7) B
rota
s et
al.
(2
003)
; (8)
Den
is &
Des
reu
mau
x (2
009)
; (9)
Den
is e
t al
. (20
12);
(10
) M
ign
é et
al.
(20
09);
(11
) D
avou
lt e
t al
. (20
09);
(12
) S
pil
mon
t et
al.
(20
06)
63
LITERATURE CITED
Andersen TJ (2001) Seasonal variations in erodibility of twotemperate, microtidal mudflats. Estuar Coast Shelf Sci53: 1−12
Andersen TJ, Lanuru M, van Bernem C, Pejrup M, Rieth-mueller R (2010) Erodibility of a mixed mudflat domi-nated by microphytobenthos and Cerastoderma edule,East Frisian Wadden Sea, Germany. Estuar Coast ShelfSci 87: 197−206
Anschutz P, Smith T, Mouret A, Deborde J, Bujan S, PoirierD, Lecroart P (2009) Tidal sands as biogeochemical reac-tors. Estuar Coast Shelf Sci 84: 84−90
Attard KM, Glud RN, McGinnis DF, Rysgaard S (2014) Sea-sonal rates of benthic primary production in a Greenlandfjord measured by aquatic eddy correlation. LimnolOceanogr 59: 1555−1569
Barranguet C, Kromkamp J, Peene J (1998) Factors control-ling primary production and photosynthetic characteris-tics of intertidal microphytobenthos. Mar Ecol Prog Ser173: 117−126
Bartoli M, Nizzoli D, Viaroli P (2003) Microphytobenthosactivity and fluxes at the sediment-water interface: interactions and spatial variability. Aquat Ecol 37: 341−349
Berg P, Huettel M (2008) Monitoring the seafloor using thenon-invasive eddy correlation technique: integratedbenthic exchange dynamics. Oceanography (Wash DC)21: 164−167
Berg P, Risgaard-Petersen N, Rysgaard S (1998) Interpreta-tion of measured profiles in sediment pore water. LimnolOceanogr 43: 1500−1510
Berg P, Long MH, Huettel M, Rheuban JE and others (2013)Eddy correlation measurements of oxygen fluxes in per-meable sediments exposed to varying current flow andlight. Limnol Oceanogr 58: 1329−1343
Billerbeck M, Werner U, Polerecky L, Walpersdorf E, deBeer D, Huettel M (2006) Surficial and deep pore watercirculation governs spatial and temporal scales of nutri-ent recycling in intertidal sand flat sediment. Mar EcolProg Ser 326: 61−76
Billerbeck M, Røy H, Bosselmann K, Huettel M (2007) Ben-thic photosynthesis in submerged Wadden Sea intertidalflats. Estuar Coast Shelf Sci 71: 704−716
Brotas V, Risgaard-Petersen N, Serodio J, Ottosen L, Dals-gaard T, Ribeiro L (2003) In situ measurements of photo-synthetic activity and respiration of intertidal benthicmicroalgal communities undergoing vertical migration.Ophelia 57: 13−26
Cahoon LB (1999) The role of benthic microalgae in neriticecosystems. Oceanogr Mar Biol Annu Rev 37: 47−86
Cai WJ (2011) Estuarine and coastal ocean carbon paradox: CO2 sinks or sites of terrestrial carbon incineration?Annu Rev Mar Sci 3: 123−145
Cammen LM (1991) Annual bacterial production in relationto benthic microalgal production and sediment oxygenuptake in an intertidal sandflat and an intertidal mudflat.Mar Ecol Prog Ser 71: 13−25
Cook PLM, Revill AT, Clementson LA, Volkman JK (2004)Carbon and nitrogen cycling on intertidal mudflats of atemperate Australian estuary. III. Sources of organic mat-ter. Mar Ecol Prog Ser 280: 55−72
Cook PLM, Wenzhöfer F, Glud RN, Janssen F, Hüttel M(2007) Benthic solute exchange and carbon mineraliza-tion in two shallow subtidal sandy sediments: impact of
advective pore-water exchange. Limnol Oceanogr 52: 1943−1963
Davoult D, Migné A, Créach A, Gevaert F, Hubas C, SpilmontN, Boucher G (2009) Spatio-temporal variability of inter-tidal benthic primary production and respiration in thewestern part of the Mont Saint-Michel Bay (Western Eng-lish Channel, France). Hydrobiologia 620: 163−172
de Beer D, Wenzhöfer F, Ferdelman TG, Boehme SE andothers (2005) Transport and mineralization rates in NorthSea sandy intertidal sediments, Sylt-Rømø Basin, Wad-den Sea. Limnol Oceanogr 50: 113−127
de Brouwer JFC, Stal LJ (2001) Short-term dynamics inmicrophytobenthos distribution and associated extracel-lular carbohydrates in surface sediments of an intertidalmudflat. Mar Ecol Prog Ser 218: 33−44
De Wit R, Jonkers HM, van den Ende FP, van Gemerden H(1989) In situ fluctuations of oxygen and sulphide in mar-ine microbial sediment ecosystems. Neth J Sea Res 23: 271−281
Denis L, Desreumaux PE (2009) Short-term variability ofintertidal microphytobenthic production using an oxy-gen profiling system. Mar Freshw Res 60: 712−726
Denis L, Gevaert F, Spilmont N (2012) Microphytobenthicproduction estimated by in situ oxygen microprofiling: short-term dynamics and carbon budget implications.J Soils Sediments 12: 1517−1529
Du GY, Son M, An S, Chung IK (2010) Temporal variation inthe distribution of microphytobenthos in intertidal flats ofthe Nakdong River estuary, Korea. Estuar Coast Shelf Sci86: 62−70
Ehrenhauss S, Huettel M (2004) Advective transport anddecomposition of chain-forming planktonic diatoms inpermeable sediments. J Sea Res 52: 179−197
Ehrenhauss S, Witte U, Bühring SI, Huettel M (2004) Effectof advective pore water transport on distribution anddegradation of diatoms in permeable North Sea sedi-ments. Mar Ecol Prog Ser 271: 99−111
Fenchel T, Glud RN (2000) Benthic primary production andO2−CO2 dynamics in a shallow water sediment: spatialand temporal activity. Ophelia 53: 159−171
Forster S, Bobertz B, Bohling B (2003) Permeability of sandsin the coastal areas of the southern Baltic Sea: mapping agrain-size related sediment property. Aquat Geochem 9: 171−190
Glud RN, Jensen K, Revsbech NP (1995) Diffusivity in surfi-cial sediments and benthic mats determined by use of acombined N2O-O2 microsensor. Geochim CosmochimActa 59: 231−237
Glud RN, Gundersen JK, Røy H, Jørgensen BB (2003) Sea-sonal dynamics of benthic O2 uptake in a semienclosedbay: importance of diffusion and faunal activity. LimnolOceanogr 48: 1265−1276
Glud RN, Stahl H, Berg P, Wenzhöfer F, Oguri K, Kitazato H(2009a) In situ microscale variation in distribution and con-sumption of O2: a case study from a deep ocean marginsediment (Sagami Bay, Japan). Limnol Oceanogr 54: 1−12
Glud RN, Woelfel J, Karsten U, Kühl M, Rysgaard S (2009b)Benthic microalgal production in the Arctic: applied meth-ods and status of the current database. Bot Mar 52: 559−571
Glud RN, Berg P, Hume A, Batty P, Blicher ME, Lennert K,Rysgaard S (2010) Benthic O2 exchange across hard-bottom substrates quantified by eddy correlation in asub-Arctic fjord. Mar Ecol Prog Ser 417: 1−12
Grasshoff K, Kremling K, Ehrhardt M (1999) Methods of sea-water analysis. Wiley-VCH, Weinheim
Walpersdorf et al.: Intertidal benthic photosynthesis, drivers and NEM
Mar Ecol Prog Ser 566: 49–65, 2017
Gundersen JK, Glud RN, Jørgensen BB (1995) Havbundensiltomsætning. Havforskning fra Miljøstyrelsen, Hav-90.Faglig Rapport No. 57. Environmental ProtectionAgency, Copenhagen
Hancke K, Glud RN (2004) Temperature effects on respira-tion and photosynthesis in three diatom-dominated ben-thic communities. Aquat Microb Ecol 37: 265−281
Hubas C, Davoult D (2006) Does seasonal proliferation ofEnteromorpha sp. affect the annual benthic metabolismof a small macrotidal estuary? (Roscoff Aber Bay, France).Estuar Coast Shelf Sci 70: 287−296
Hubas C, Davoult D, Cariou T, Artigas LF (2006) Factorscontrolling benthic metabolism during low tide along agranulometric gradient in an intertidal bay (Roscoff AberBay, France). Mar Ecol Prog Ser 316: 53−68
Huettel M, Røy H, Precht E, Ehrenhauss S (2003) Hydrody-namical impact on biogeochemical processes in aquaticsediments. Hydrobiologia 494: 231−236
Iversen N, Jørgensen BB (1993) Diffusion coefficients of sul-fate and methane in marine sediments: influence ofporosity. Geochim Cosmochim Acta 57: 571−578
Jørgensen BB (1996) Material flux in the sediment. In: Jør-gensen BB, Richardson K (eds) Eutrophication in coastalmarine ecosystems. American Geophysical Union, Wash-ington, DC, p 115−136
Jørgensen BB, Revsbech NP (1985) Diffusive boundary lay-ers and the oxygen uptake of sediment and detritus. Lim-nol Oceanogr 30: 111−122
Jørgensen BB, Glud RN, Holby O (2005) Oxygen distribu-tion and bioirrigation in Arctic fjord sediments (Svalbard,Barents Sea). Mar Ecol Prog Ser 292: 85−95
Klaassen W, Spilmont N (2012) Inter-annual variability ofCO2 exchanges between an emersed tidal flat and theatmosphere. Estuar Coast Shelf Sci 100: 18−25
Klute A, Dirksen C (1986) Hydraulic conductivity and diffu-sivity: laboratory methods. In: Klute A (ed) Methods ofsoil analysis, Part 1. Physical and mineralogical methods.American Society of Agronomy, Madison, WI, p 687−700
Kühl M (2005) Optical microsensors for analysis of microbialcommunities. Methods Enzymol 397: 166−199
Kühl M, Glud RN, Ploug H, Ramsing NB (1996) Microenvi-ronmental control of photosynthesis and photosynthesis-coupled respiration in an epilithic cyanobacterial biofilm.J Phycol 32: 799−812
Kühl M, Lassen C, Revsbech NP (1997) A simple light meterfor measurements of PAR (400 to 700 nm) with fiber-opticmicroprobes: application for P vs E0(PAR) measurementsin a microbial mat. Aquat Microb Ecol 13: 197−207
Li YH, Gregory S (1974) Diffusion of ions in sea water anddeep sea sediments. Geochim Cosmochim Acta 88: 708−714
Long MH, Berg P, McGlathery KJ, Zieman JC (2015) Sub-tropical seagrass ecosystem metabolism measured byeddy covariance. Mar Ecol Prog Ser 529: 75−90
Lorenzen CJ (1967) Determination of chlorophyll andpheo-pigments: spectrophotometric equations. LimnolOceanogr 12: 343−346
Lotze HK, Reise K, Worm B, van Beusekom J and others(2005) Human transformations of the Wadden Sea eco-system through time: a synthesis. Helgol Mar Res 59: 84−95
MacIntyre HL, Geider RJ, Miller DC (1996) Microphytoben-thos: the ecological role of the ‘Secret Garden’ of unvege-tated, shallow water marine habitats. I. Distribution, abun-dance and primary production. Estuaries 19: 186−201
McGinnis DF, Sommer S, Lorke A, Glud RN, Linke P (2014)Quantifying tidally-driven benthic oxygen exchangeacross permeable sediments: an aquatic eddy correlationstudy. J Geophys Res Oceans 119: 6918−6932
Middelburg JJ, Klaver G, Nieuwenhuize J, WielemakerA, de Haas W, Vlug T, van der Nat JFWA (1996)Organic matter mineralization in intertidal sedimentsalong an estuarine gradient. Mar Ecol Prog Ser 132: 157−168
Middelburg JJ, Barranguet C, Boschker HTS, Herman PMJ,Moens T, Heip CHR (2000) The fate of intertidal micro-phytobenthos carbon: an in situ 13C-labeling study. Lim-nol Oceanogr 45: 1224−1234
Migné A, Spilmont N, Boucher G, Denis L and others (2009)Annual budget of benthic production in Mont Saint-Michel Bay considering cloudiness, microphytobenthosmigration, and variability of respiration rates with tidalconditions. Cont Shelf Res 29: 2280−2285
Paterson DM (1989) Short-term changes in the erodibility ofintertidal cohesive sediments related to the migratorybehaviour of epipelic diatoms. Limnol Oceanogr 34: 223−234
Paterson DM, Hagerthey SE (2001) Microphytobenthos incontrasting coastal ecosystems: biology and dynamics.In: Reise K (ed) Ecological comparisons of sedimentaryshores. Ecological Studies, Vol 151. Springer-Verlag,Berlin, p 105−125
Platt T, Gallegos CL, Harrison WG (1980) Photoinhibition ofphotosynthesis in natural assemblages of marine phyto-plankton. J Mar Res 38: 687−701
Reise K (2005) Coast of change: habitat loss and transforma-tions in the Wadden Sea. Helgol Mar Res 59: 9−21
Revsbech NP (1989) An oxygen microelectrode with a guardcathode. Limnol Oceanogr 34: 474−478
Rickelt LF, Lichtenberg M, Trampe E, Kühl M (2016) Fiber-optic probes for small-scale measurements of scalar irra-diance. Photochem Photobiol 92: 331−342
Risgaard-Petersen N (2003) Coupled nitrification-denitrifi-cation in autotrophic and heterotrophic estuarine sedi-ments: on the influence of benthic microalgae. LimnolOceanogr 48: 93−105
Spilmont N, Migné A, Lefebvre A, Artigas LF, Rauch M,Davoult D (2005) Temporal variability of intertidal ben-thic metabolism under emersed conditions in an exposedsandy beach (Wimereux, eastern English Channel,France). J Sea Res 53: 161−167
Spilmont N, Davoult D, Migné A (2006) Benthic primaryproduction during emersion: in situ measurements andpotential primary production in the Seine Estuary (Eng-lish Channel, France). Mar Pollut Bull 53: 49−55
Spilmont N, Migné A, Seuront L, Davoult D (2007) Short-term variability of intertidal benthic community produc-tion during emersion and the implication in annualbudget calculation. Mar Ecol Prog Ser 333: 95−101
Staehr PA, Testa JM, Kemp WM, Cole JJ, Sand-Jensen K,Smith SV (2012) The metabolism of aquatic ecosystems: history, applications, and future challenges. Aquat Sci74: 15−29
Sundbäck K, Miles A, Göransson E (2000) Nitrogen fluxes,denitrification and the role of microphytobenthos inmicrotidal shallow-water sediments: an annual study.Mar Ecol Prog Ser 200: 59−76
Underwood GJC, Kromkamp J (1999) Primary productionby phytoplankton and microphytobenthos in estuaries.Adv Ecol Res 29: 93−153
64
Walpersdorf et al.: Intertidal benthic photosynthesis, drivers and NEM 65
Vinther N, Christiansen C, Bartholdy J, Sørensen C, Lund-Hansen LC (2004) Sediment transport across a tidal dividein the Danish Wadden Sea. Geogr Tidsskr 104: 71−86
Werner U, Billerbeck M, Polerecky L, Franke U, Huettel M,Van Beusekom JEE, de Beer D (2006) Spatial and tempo-ral pattern of mineralization rates and oxygen distribu-tion in a permeable intertidal sandflat (Sylt, Germany).Limnol Oceanogr 51: 2549−2563
Westall F, Rincé Y (1994) Biofilms, microbial mats and
microbe-particle interactions: electron microscope obser-vations from diatomaceous sediments. Sedimentology41: 147−162
Zetsche E, Paterson DM, Lumsdon DG, Witte U (2011) Tem-poral variation in the sediment permeability of an inter-tidal sandflat. Mar Ecol Prog Ser 441: 49−63
Zetsche E, Bulling MT, Witte U (2012) Permeability of inter-tidal sandflats: impact of temporal variability on sedi-ment metabolism. Cont Shelf Res 42: 41−50
Editorial responsibility: Erik Kristensen, Odense, Denmark
Submitted: July 21, 2016; Accepted: December 13, 2016Proofs received from author(s): February 12, 2017