Download - Microbenthos in a hypersaline tidal lagoon: factors ...Sediment cores (n = 3) with a surface area of 4 cm2 (as recommended by Eaton & Moss 1966) were taken from each of the benthic

AQUATIC MICROBIAL ECOLOGYAquat Microb Ecol

Vol. 38: 53–69, 2005 Published January 21

INTRODUCTION

Microbenthic communities play a major role in theflux of carbon in shallow aquatic environments. Thesecomplex microbial communities are vertically strati-fied, and the organisms growing herein are subjectedto steep physico-chemical microgradients (Jørgensenet al. 1983, Revsbech et al. 1983). Biomass, speciescomposition and physical structure of microbenthiccommunities are highly variable. These communitiescan vary from those inhabiting the surface of non-cohesive sediments, where the microbial community isinvisible to the naked eye, to very well-developed andconspicuous microbial mats that form a continuousbiofilm on the sediment surface. Given the trophody-namic importance of these kinds of microbial commu-

nities in intertidal flats and other shallow aquatic envi-ronments and their role in the exchange of nutrients atthe water-sediment interface, it is necessary to under-stand the ecological factors that determine their abun-dance and structure in nature. Moreover, biomassdensity, species composition and the physical structureof microbenthic communities are likely to stronglyaffect the overall nutrient cycling. Similarly, the physi-cal stability of sediment is also affected by speciescomposition: aggregation and compactation may occurby exopolymer secretion and entangling of filamen-tous cyanobacteria, while mechanical disturbance ofthe sediment increases with the abundance of meio-and macrofauna (Giere 1993). The presence of specificphysiological adaptations in some taxonomic groupsallows them to compete favourably in harsh environ-

© Inter-Research 2005 · www.int-res.com*Email: [email protected]

Microbenthos in a hypersaline tidal lagoon: factorsaffecting microhabitat, community structure andmass exchange at the sediment-water interface

J. García de Lomas*, A. Corzo, C. M. García, S. A. van Bergeijk

Depto. Biología (Área Ecología), Facultad de Ciencias del Mar y Ambientales, Pol. Río San Pedro s/n, 11510 Puerto Real (Cádiz), Spain

ABSTRACT: Three benthic microbial communities from a hypersaline lagoon with tidal influencewere studied: a compact microbial mat, a ‘fluffy’ microbial mat and a non-cohesive diatom-dominatedsediment. In each community, qualitative and quantitative analyses of phototrophs, meio- and macro-fauna were done. Vertical profiles of oxygen, sulfide and pH were measured at the sediment-waterinterface using microelectrodes. Physico-chemical properties of the water column were alsoassessed. The most compact mat, dominated by Microcoleus chthonoplastes, had the highest photo-synthetic biomass, while meio- and macrofauna were nearly absent. The compact mat showed thesteepest physico-chemical microgradients and the highest fluxes of oxygen and sulfide. The ‘fluffy’mat had an areal amount of chl a similar to the compact mat and contained a higher abundance ofmeiofauna. This mat showed the deepest oxic layer. The diatom-dominated sediment comprised ahigh abundance of macrofauna, but meiofauna were very scarce. This community presented rela-tively smoother microgradients and intermediate values of internal fluxes of oxygen. Our resultsshow the close relationship between the structure of the microbenthic communities, net metabolism,and the exchange of mass at the water-sediment interface. Changes in biotic and abiotic factorsdetermined the spatial distribution of each microbial community.

KEY WORDS: Microbenthic community · Microbial mat · Meiofauna · Macrofauna · Microelectrodes ·Biotic factors · Abiotic factors

Resale or republication not permitted without written consent of the publisher

Aquat Microb Ecol 38: 53–69, 2005

ments. Nitrogen fixation (Dubinin et al. 1992), fermen-tation (Moezelaar et al. 1996), the use of sulfide as anelectron source (de Wit & van Gemerden 1989, van denHoek et al. 1995), and the presence of some UV-pro-tective substances (Garcia-Pichel & Castenholz 1991)have been demonstrated in some cyanobacteria,allowing these microorganisms to survive under verystressful and dynamic conditions (high irradiance,salinity, temperature). However, these physiologicaladaptations do not explain why microbial mats areusually absent in more moderate environmental condi-tions. Meio- and macrofauna can also play an impor-tant role in controlling microbial mat development incoastal environments. Under moderate environmentalconditions, e.g. coastal environments, meio- andmacrofauna reduce primary producer concentrationdue to a high grazing activity (Hargrave 1970, Connoret al. 1982), and thus these mild conditions prevent mataccretion (Jørgensen et al. 1983, Awramik 1984). It hasbeen classically postulated that extreme conditionslike high salinities in hypersaline environments (Pier-son et al. 1987, Wieland & Kühl 2000), periodic dryings(Lassen et al. 1992), or high temperatures from thermalsources (Castenholz 1984) can limit meio- and macro-fauna growth, allowing the development of microbialmats. However, few studies have quantified meio- andmacrofauna in microbial mats (Fenchel 1998, Pinckneyet al. 2003), or analyzed their possible influence on matstructure, metabolism, and the exchange of mass at thewater-sediment interface (Cullen 1973, Aller & Aller1992, Huettel & Gust 1992).

In the present study, we compare 3 different types ofbenthic communities growing in a small hypersalinelagoon with tidal influence, and therefore under thesame ecological conditions at a macroscopic scale. Thelagoon is within a protected area, the Natural Park‘Bahía de Cádiz’, where different types of micro-benthic communities, including microbial mats, canbe found. Microbenthic communities had not beenstudied previously in the Bay of Cádiz despite theirimportance in terms of surface cover in salt marshes,saltpans and intertidal sediments. This study aims toelucidate the ecological factors affecting the benthicmicrohabitat and its microbial community and deter-mine how these factors can influence the communitystructure, metabolism and mass transfer at the water-sediment interface.

MATERIALS AND METHODS

Area of study. The sampling site is a small (ca. 3 ha)and very shallow (depth < 1 m, mean values 0.15 to0.20 m) lagoon located inside the Natural Park ‘Bahíade Cádiz’, in the SW of Spain (36° 32’ 2” N, 6° 12’ 35” W).

It is connected to the San Pedro estuary by a creek,allowing a tidal input of marine water (Fig. 1), but theregime of exchange and renovation of water in thelagoon does not coincide with that of the estuary. Twodifferent areas can be distinguished according to tidalexchange and water renovation. An artificial tube thatlimits water flow connects these 2 sub-lagoons. Theside closer to the estuary, in the north (Site 1), hasrelatively moderate salinities (30–50) and higherspatial homogeneity, whereas the patchy lagoon inthe south (Site 2) has a lower water turnover rate, lead-ing to more fluctuating and extreme salinities (up to80–100, mean value 45). Site 2 could therefore be asso-ciated with a strong physical control: longer emersionperiods (up to 2 mo) and more frequent fluctuationswith higher amplitudes of temperature, salinity andother physico-chemical variables.

The present study was done in 2 phases. In the firstphase (April to June 2000), we did a general character-ization of Site 1 and Site 2 to understand why microbial

54

Fig. 1. Area of study in SW of Spain. Surface colonized by mi-crobial mats in the lagoon. Microbial mats were only presentin the zone furthest away from the estuary. Some sedimentsurface was only covered by mats during the summer

(June to September)

García de Lomas et al.: Microbenthic communities in a hypersaline lagoon

mats were mainly present at Site 2. Once the generalhydrological dynamics of the lagoon were understood,the study was focussed (April 2003) on 3 kinds ofmicrobenthic communities present at Site 2 to betterresolve what ecological factors, biotic and abiotic,determine their different structure. The smaller size ofSite 2 guarantees a higher uniformity at the macro-scopic scale in the physico-chemical and climatologicalvariables to which the microbenthic communities wereadapted. The studied communities were non-cohesivediatom-dominated sediment, a thick microbial matwith a fluffy appearance, and a compact microbial mat.

Hydrology and physico-chemical parameters of thewater column. Characterization of the lagoon at differ-ent temporal resolutions was performed by assessingthe hydrology and a number of physico-chemical vari-ables. Near daily measurements of water level,rain–evaporation balance, salinity, pH, water tempera-ture, and the concentration of dissolved oxygen, nitrateand ammonium were carried out from April to June2000. During a 12 h tidal cycle on 3 June 2000, waterlevel, O2, pH, and water temperature were monitoredevery 10 min, using a multiparameter logger. Irradi-ance, rain and evaporation data were acquired fromthe meteorological station located at Cádiz city (8 kmaway from the lagoon).

Organic matter content of the sediment and granu-lometry. Water content and the amount of organicmatter were determined in sediment cores (n = 3, innerdiameter [i.d.] = 6.7 cm) collected during April 2003from each microbenthic community at Site 2 accordingto Krumgalz & Fainshtein (1989). In order to analyzegrain-size composition, 3 cores were taken (n = 3,i.d. = 10.5 cm) from each benthic community. In thelaboratory, samples (upper 5 cm) were passed througha stack of sieves with 2.0, 1.0, 0.5, 0.25, 0.125, and0.063 mm mesh sizes, using tap water. Through thiswet sieving, salt is removed but faecal pellets are con-served, which can consolidate great amounts of finesediment. Desiccation of the sediment, which leadsto disintegration of the faecal pellets into their con-stituents (Giere 1993), was thus avoided. After the wetsieving, each size fraction was dried at 60°C for 72 hand subsequently weighed. A granulometric curvewas obtained as cumulative percentages starting withthe coarsest fraction. Several granulometric indiceswere calculated: sorting coefficient, graphic mean(median grain diameter, Md), and inclusive graphicskewness (Giere 1993).

Phototrophs and vertical distribution of pigments.Sediment cores (n = 3) with a surface area of 4 cm2 (asrecommended by Eaton & Moss 1966) were taken fromeach of the benthic communities at Site 2 in April 2003and immediately carried to the laboratory. Vertical dis-tributions of pigments were studied in these cores by

slicing thin sections (1 mm) of fresh sediment using arazor blade. A total of 10 sections were obtained fromeach core. As extractant, 100% methanol was used(Thompson et al. 1999). Samples were sonicated (5pulses of 30 s at 15 W) and extracted for 24 h at 4°C in thedark. Extracts were filtered and the absorption spectrumwas obtained between 350 and 1100 nm using a spec-trophotometer (Unicam UV/VIS UV2®). Pigment con-centrations (chlorophyll a [chl a] and bacteriochlorophylla) were estimated after Pierson et al. (1987) and resultswere expressed in µg of chl a g–1 of wet weight (ww).Taxonomic identifications of microalgae were done bylight microscopy according to Bourrelly (1970, 1972,1981), Rippka et al. (1979), Sournia (1986), Castenholz etal. (1989), Staley et al. (1989), Komárek & Anagnostidis(1986, 1989), Round et al. (1990) and Chrétiennot-Dinet(1990). A chl a:C ratio of 1:50 was used to transform chl aconcentration in C (Joint 1978).

Meiofauna and macrofauna abundance. Corestaken in April (2003) for meiofauna analyses (n = 3)were collected by using a 2.7 cm i.d. syringe with thelower end cut. Meiofauna from the upper 1 cm werequantified. Samples were filtered through a fine meshof 63 µm (after passing through a mesh of 500 µm)using filtered (0.45 µm) seawater. Subsequently, theywere fixed and mixed with 4% formaldehyde, contain-ing 1 g l–1 Rose Bengal. Numbers of individuals fromdifferent groups (Nematoda, harpacticoid Copepoda,Amphipoda, polychaetes, Ciliata and Diptera larvae)were hand-sorted under a stereoscopic microscope.

Cores destined for macrofauna analysis (n = 3, i.d. =10.5 cm) were taken from each benthic community inApril (2003). The upper 5 cm were sieved with in situwater through a mesh of 500 µm (Holme 1971). A softbrush was used to facilitate sample dislodgmentthrough the filter (Arias & Drake 1999). Major taxo-nomic groups (Diptera larvae, polychaetes, Lamelli-branchia and amphipods) were hand-sorted under astereoscopic microscope. Species were determinedafter Arias & Drake (1999).

Vertical profiles of oxygen, sulfide and pH. Fromeach microbenthic community a sample was taken inApril-May (2003) with a plexiglass core tube (i.d. =6 cm) and was used for microelectrode measurementsimmediately upon return to the laboratory. The coretube containing the sample was placed in a tempera-ture-controlled flow chamber (Lorenzen et al. 1995),which was filled with filtered water from the samplingsite (salinity = 43), and a flow of 0.55 cm s–1 was estab-lished. This was calculated as the rate of outflowdivided by the water-filled cross-area of the chamber(Lorenzen et al. 1995). The water depth inside thechamber was 1 cm. The cores were incubated at atemperature of 20 ± 1°C and a photon flux density of220 µmol m–2 s–1 using a halogen lamp (Novaflex,

55


World Precision Instruments) for at least 1 h before themeasurements were started. Steady-state profiles ofdissolved oxygen, sulfide and pH were measured usingmicroelectrodes. The microelectrodes were mountedin a micromanipulator and were lowered into thesamples with a 100 µm vertical step resolution.

Microelectrodes were purchased from Unisense®.The oxygen sensors (Revsbech 1989) had a tip dia-meter of 20 to 30 µm, a response time of 0.2 to 0.4 s, anda stirring sensibility 0.1 pHunit), it was necessary to correct the measured sulfideprofiles (Kühl & Jørgensen 1992), whose concentrationwas calculated as follows:

[H2S] = [Stot2–]�(1 + (K1K2�[H3O+]2) + (K1�[H3O+])) (1)

where [Stot2–] is the total sulfide concentration, H3O+ isthe hydrogen ion activity, calculated from measured pHvalues. K1 and K2 are the first and second dissociationconstants, respectively, of the sulfide equilibrium sys-tem: pK1 = 7.05 and pK2 = 17.1 (Kühl & Jørgensen 1992).

Flux calculations. From the obtained oxygen pro-files, oxygen fluxes across the sediment-water inter-face (J ) were calculated using the equation of Fick’sfirst law of diffusion (Kühl et al. 1996):

J (z) = –Do [– ∂C(z)�∂z] (2)

where Do is the free solution molecular diffusion coef-ficient of oxygen and [∂C(z)�∂z] is the linear oxygenconcentration gradient in the diffusive boundary layerabove the sediment surface. Because salinity was 43,we used a Do of 1.948 × 10–5 cm2 s–1, which is 8% lowerthan the value for freshwater (Broecker & Peng 1974,Li & Gregory 1974). A negative flux (Jup) indicates anet export of oxygen out of the sediment and repre-sents the net areal photosynthesis (Pn) of the microbialcommunity present; a positive flux indicates a netuptake of oxygen and represents the areal respiration(Rdark) (Table 1).

In order to calculate the oxygen flux inside the sedi-ment, equations also based on Eq. (2) were used (Kühlet al. 1996):

J ’down = – φ(z) · Ds(z) · [– ∂C(z)�∂z] = De · ∂C(z)�∂z (3)

φ(x) · Ds(x) = Do · [∂Cw(x)�∂x ] · [∂Cs(x)�∂x]–1 (4)

where J’down is the oxygen flux inside the mat, φ(z) is theporosity and Ds(z) and De are the apparent and the ef-fective diffusion coefficients, respectively, inside the

56

Table 1. Oxygen and sulfide fluxes and other parameters calculated in the 3 benthic communities studied. The variables usedis the same as in Kühl et al. (1996). UL: values calculated in the upper layer of the fluffy mat, but not used for calculations. DL: values expected in the downward layer, where a dense and compact band of Microcoleus chthonoplastes was found, similar

in the compact mat. –: undetectable values within the upper 10 mm

Parameter Variables Units Sediment Fluffy mat Compact mat

Areal net photosynthesis, light Jup = Pn µmol O2 cm–2 h–1 0.2805 0.1403 0.3506Flux of O2 inside the mat, light J ’down µmol O2 cm–2 h–1 0.075 0.026a 0.1177Areal respiration, dark Rdark = Jdown µmol O2 cm–2 h–1 0.1503 0.1252 0.1001Volumetric dark respiration Rdark/Zdark µmol O2 cm–3 h–1 1.5 0.69 1.4Flux of H2S, light µmol H2S cm–2 h–1 – – 0.0874Flux of H2S, dark µmol H2S cm–2 h–1 – – 0.0798Oxic layer thickness, light Zlight mm 4.9 10 2Oxic layer thickness, dark Zdark mm 1 1.8 0.7Diffusivity (porosity × effective UL:0.9209diffusion coeficient)

φ · Ds (×105) cm2 s–1 0.7891 DL:0.4898 0.4898

Photoautotrophic biomass Chl a mg chl a m–2 172 689 721Photosynthetic quotient P/B d–1 0.093 0.012 0.028Photosynthetic quotient Pn/chl a µg C µg–1 chl a h–1 0.196 0.024 0.058aThis value was used for calculating J ’down


sediment. ∂C(z)�∂z is the concentration gradient in thelower limit of euphotic zone. The product φ × Ds wascalculated as the ratio of the concentration gradients ofoxygen concentration in water and sediment (Cw andCs, respectively), multiplied by the molecular diffiusioncoefficient in fresh water (Do) (Revsbech 1989, Glud etal. 1995) (Eq. 4). For the determination of diffusivity, thesediment was biologically inactivated by addingformaldehyde (4% v/v final concentration) and chemi-cally oxidized (see Glud et al. 1995). Oxygen profileswere measured in the dark. In this way, values of φ ×Ds = 0.7891 × 10–5, 0.9209 × 10–5 and 0.4898 × 10–5 cm2

s–1 were obtained for the sediment, the fluffy mat andthe compact mat, respectively (Table 1). However, a φ ×Ds of 0.4898 × 10–5 cm2 s–1 was used for the fluffy mat,because at the depth where J ’down was calculated adense layer of Microcoleus chthonoplastes was present.

Photosynthetic and respiratory quotients of 1 wereused to express photosynthesis and respiration rate inC units when necessary. Net carbon fixation rate dur-ing the day was calculated from the areal net photo-synthesis rate, and the aerobic carbon mineralizationduring the night from the respiration rate in the dark,assuming a light:dark cycle of 12:12 h light:dark. Thedaily net primary production was calculated by sub-tracting carbon mineralization from carbon fixation.

Sulfide fluxes inside the fluffy and the compact matswere calculated using Eq. (3), assuming Ds (H2S) =1.39 × 10–5 cm2 s–1 as the apparent diffusion coefficientof sulfide (Kühl & Jørgensen 1992) and the porosityclose to the unit and constant with depth.

Statistical analysis. Differences in chl a, meio- andmacrofauna among microbenthic communities (meanvalues) were tested by analysis of variance (ANOVA).The Tukey test was used after ANOVA to further testdifferences between 2 means (Zar 1984). Two-factorANOVA was used to compare vertical distributionsof chl a and oxygen among communities, using Stat-graphics®Plus 5.0. In this analysis, the effects on oxy-gen or chl a of the factors type of community and depth(as dependent variables) were statistically evaluated.

RESULTS

Hydrology and physico-chemical parameters of thewater column

Tidal influence in the lagoon was clearly noted at Site1, where the highest water levels coincided with thehighest levels of high tide (Fig. 2a). However, the waterlevel was not only controlled by tides. The negativerain–evaporation balance in May and June (Fig. 2b) re-sulted in a decrease in water level and a considerable in-crease in salinity, which was most pronounced at Site 2

(Fig. 2c). During the monitoring period, nitrate concen-tration remained below 1.05 µM at both stations, whileammonium varied between 2.5 and 10 µM without aclear trend (data not shown). Measurements of waterlevel, dissolved O2, pH, and temperature during a tidalcycle (short-term measurements, Fig. 2) also revealedthat tidal influence was greater at Site 1 than at Site 2,where the input of tidal water is hardly observed (Fig.2e). A sharp change was observed in water level and wa-ter temperature at Site 1 at the moment of tidal input(Fig. 2e,f,h). Evidence of a daily cycle in both O2 and pHwere observed at both sites (Fig. 2f,g). O2 in the watercolumn was undersaturated in the morning (20 to 50%)and supersaturated during the afternoon (100 to 150%).

Spatial distribution of benthic microbial communities

Spatial distribution of different communities in the la-goon was not homogeneous (Fig. 1). Microbial mats cov-ered a large area at Site 2 (about 50% in winter andabout 65% in summer), where tidal influence was lowerand therefore more prone to seasonal salinity changesand periodic desiccation. Microbial mats were also pre-sent in small areas closer to the San Pedro estuary. Thesesites were located well above sea level, and tidal waterinput was restricted to the biggest tides. The aspect ofsurface and physical structure of the microbenthic com-munities studied at Site 2 differed considerably (Fig. 3).Moreover, a clear zonation was observed in the spatialdistribution of these communities along a transect fromthe center of the lagoon to the emerged shore (Fig. 4).The inner part of the lagoon was covered by lax diatom-dominated sediment without mat structure (Fig. 3a) withan overlaying water column of about 15 cm. This sedi-ment had a loose appearance and its surface comprisedplenty of tubes from amphipods and polychaetes. Afluffy microbial mat covered a band closer to the shore-line at a mean depth of 6 cm and showed a 4 to 5 mmthick surface, fluffy layer (Fig. 3b). Compact-laminatedmicrobial mats (Fig. 3c) mainly appeared in the lagoonfringes (about 3 cm depth), where the emersion-floodepisodes were more frequent and/or water salinityreached higher values. Each of these 3 communitiesrepresented different autotrophic biomass inhabiting thesediment surface. To emphasise this, we measured totalchl a per unit area in a short transect along a steep shore(Fig. 5). The mats covered wider fringes where the slopewas less pronounced, as they appeared in the water levelfluctuation zone, i.e. in the fringe that is not permanentlybut only periodically flooded.

The sediments at Site 2 were mainly composed of finesand (Md = 0.19 mm). Some differences were observedbetween communities (Fig. 6), such as a higher meangrain size in both microbial mats (Md = 0.23 and 0.35 mm

57


for fluffy and compact mats, respectively). However, thiscould be an artifact due to retention of portions oftangled cyanobacterial filaments in the biggest sieve.

Types of microbenthic community andtaxonomic composition

Three different microbenthic communities werefound at Site 2. Benthic pennate diatoms dominated

the non-cohesive sediment located in the innermostpart of the lagoon as the main taxonomic group amongphototrophs. Pleurosigma, Nitzschia, Amphora andNavicula genera appeared frequently. Cyanobacteriawere much scarcer than diatoms, but few and isolatedcolonies of Oscillatoria sp. and Spirulina sp. (Oscillato-riales), with gliding motility, were observed in the lax,light-brown surface of the sediment.

Fluffy mats usually presented several well-definedlayers. Different taxonomic groups appeared vertically

58

b

d

Site 1

Site 2

Site 1

Site 2

a

Near daily assessment(April to June 2000)

f

g

h

e

Short-term assessment(3 June 2000)

1

0.8

0.6

0.4

0.2

0

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

0

140

100

60

20

8.2

8.1

8.0

7.9

7.8

7.7

15

10

5

0

–5

–10

120

100

80

60

40

20

3000

2000

1000

4.0

3.5

3.0

2.5

Wat

er le

vel (

m)

Sal

init

y

Dai

ly ir

rad

ianc

e(1

0 K

J m

–2)

Rai

n–ev

apo

rati

on

bal

ance

(mm

or

l m–2

)

— April — — May — — June —10 20 30 10 20 30 10 20 30

10:00 13:00 16:00 19:00 22:00Time (h)

pH

35

30

25

20

15

Te

mp

era

ture

(°C

)

Hei

ght

of

the

hig

h ti

de

(m)

Wat

er le

vel (

m)

O2

(% s

atur

atio

n)c

Fig. 2. (a–d) Long-term and (e–h) short-term changes in physico-chemical variables of the 2 locations of the lagoon. Monitoring of(a) water level, (b) rain–evaporation balance, (c) salinity and (d) daily irradiance was carried out on a nearly daily basis from Aprilto June 2000. Assessment of (e) water level, (f) dissolved oxygen, (g) pH and (h) water temperature was done over a 12 h tidal

cycle (3 June 2000). Dashed line in (a) represents the tidal height at the nearest reference station (Cadiz)


segregated in the mat, forming different coloured lay-ers. Pennate diatoms, mainly Nitzschia spp., Naviculaspp. and Amphora spp., dominated the upper goldenlayer. Striatella unipunctata was also abundant atsome isolated points. Fluffy tufts (Fig. 3b) were formedby the diatoms previously mentioned and some fila-mentous cyanobacteria, like Oscillatoria sp. Below thisfluffy, golden layer, a green band dominated by cyano-bacteria was found, with Microcoleus chthonoplastes(Oscillatoriales) as the dominant species. Gloeocapsaand Chroococcus turgidus (order Chroococcales) werealso found. A thin white band of ca. 100 µm dominatedby the filamentous, motile colourless sulfur bacteriumBeggiatoa sp. was observed below the cyanobacteriallayer. A 200 µm thick purple layer and a thin hard crustcould be observed before reaching the black sediment.

The compact mat usually appeared to be associatedwith the upper border of the lagoon, and it was thinner

than the fluffy mats. A diatom layer was not observed.The photosynthetic community was mainly composedof the filamentous cyanobacteria Microcoleus chthono-plastes, which formed a dense and highly compact mat(Fig. 3c). Below this band, a thin white band composedof Beggiatoa and a thin purple layer could be observedbefore reaching the black sediment.

Vertical distribution of pigments and organic matter

The vertical structure of the photosynthetic commu-nity was significantly different among the 3 benthiccommunities (p < 0.001, 2-factor ANOVA). In the sedi-ment and the compact mat, maximum chl a concentra-tions were found in the uppermost layer, while in thefluffy mat they were found between 3 and 5 mm depth(Fig. 7a). This subsurface chl a maximum coincided

59

Fig. 3. Surficial appearance of the 3 benthic communities studied. (a) Sediment without mat, where some polychaete tubes canbe observed (arrows), (b) fluffy microbial mat, and (c) compact-laminated microbial mat with a gastropod (Hydrobia ulvae)

migrating during flooding (arrow)


with the presence of a dense layer of Microcoleuschthonoplastes at this depth. Mean values were alsosignificantly different (p < 0.025, ANOVA) amongcommunities (14.7 µg chl a g–1 ww in the sediment and63.9 and 66.2 µg chl a g–1 ww for the fluffy and thecompact mats, respectively), but significant differ-ences only existed between the sediment and bothmats (p < 0.001, Tukey test). However, maximum sur-face values in the compact mat were 2-fold higher thanin the subsurface maximum present in the fluffy mat,and even 4 to 5 times higher when the upper layers ofboth mats were compared (Fig. 7a). Maximum bac-teriochlorophyll a concentrations were found belowthe maximum chl a concentrations in the 3 communi-ties, but they were much higher in the compact matthan in the sediment and the fluffy mat (Fig. 7b).

Differences in the vertical structure of microbenthiccommunities were evident in their organic matter andwater contents (Fig. 8). A lower organic content waspresent in the sediment (Fig. 8a), while both mats(fluffy and compact) showed greater amounts of

organic content (Fig. 8a), mainly associated with thesurface, where diatoms and cyanobacteria accumulate.Water content analysis also showed some differencesbetween the benthic communities studied, with ahigher amount of water in the fluffy mat (Fig. 8b),which could be related to its plenty-of-tufts fluffystructure.

Vertical microgradients of O2, sulfide and pH

Vertical distribution of O2 in light significantly dif-fered among communities (p < 0.001, 2-factor ANOVA).The highest O2 penetration and oxygen concentrationswere detected in the fluffy microbial mat, where anoxygen subsurface maximum (values up to 1100 µM)was found at 3.5 to 6 mm depth. The sediment and thecompact mat presented very similar maximum concen-trations of oxygen, of ca. 800 µM (Fig. 9a), despite hav-ing quite different chl a concentrations (Fig. 7); theseO2 maxima were shallower than in the fluffy mat, andthey were located at similar depths. However, O2penetration was 2-fold higher in the diatom-dominatedsediment compared with the compact microbial mat.Maximum oxygen concentrations in both the fluffy andthe compact mats were associated with a layer domi-nated by Microcoleus chthonoplastes. In the dark, theoxic layer was strongly reduced in all 3 communities,

60

low water levelhigh water level

compactmats

sediment(no microbial mat)

fluffy matsFig. 4. Zonation of the 3 dif-ferent microbenthic commu-nities along a shore profile

High water level

Low water level

0

20

40

60

80

0 20 40 60 80 100

Emergedzone

most

Water levelfluctuation zone

(shore)Immersedzone

100 cm

of thetime

Fig. 5. Benthic chlorophyll a distribution along a steep shore.Each point corresponds to the mean value of 3 samples. Error

bars represent ±SD

2 1 0.25 0.1250.5 0.0630

80

60

40

20

100

Sediment

Fluffy mat

Compact mat

Cum

ulat

ive

wei

ght

%

Grain size (mm)

Fig. 6. Granulometric curves of the 3 microbenthic communi-ties studied


reaching very similar depths: 1 mm in the sedimentand 0.7 mm in the compact mat (Fig. 9d). The maxi-mum difference between light and dark conditionsoccurred in the fluffy mat, with a drastic reductionfrom 10 to 1.7 mm.

Areal net photosynthesis rate calculated from the O2profiles at the diffusive boundary layer in light repre-sents the oxygen exportation to the overlying water(Jup), and was highest for the compact mat (0.35 µmolO2 cm–2 h–1). A somewhat lower rate was estimated forthe sediment (0.2805 µmol O2 cm–2 h–1) and the lowestrate was estimated for the fluffy mat (0.1403 µmol O2cm–2 h–1) (Table 1). The oxygen exportation obtained inthe compact mat could be expected by taking intoaccount that chl a concentration was highest in thiscommunity. However, minimum values found in thefluffy mat were unexpected, as maximum oxygen con-centrations of up to 1100 µM were found (see ‘Discus-sion’). The flux of O2 produced in the communityphotic layer that diffuses downward (J ’down) was high-est for the compact mat (0.112 µmol O2 cm–2 h–1), indi-cating that both the highest net photosynthesis and thehighest oxygen consumption below the photic layeroccurred in this mat. The lowest J ’down values wereestimated in the fluffy mat (0.03 µmol O2 cm–2 h–1)(Table 1). These results are again surprising, consider-ing that both microbial mats were similar in the layerwhere J ’down was calculated. A layer of Microcoleuschthonoplastes above the Beggiatoa layer and the pur-ple bacteria was found at 7 to 9 mm in the fluffy matand at 1.5 to 2 mm in the compact mat, where J ’downwas calculated for each community. Areal respirationin the dark (Rdark) was maximum for the diatom-dominated sediment (0.15 µmol O2 cm–2 h–1), some-what lower for the fluffy mat (0.12 µmol O2 cm–2 h–1),and the lowest rate was calculated in the compact mat(0.10 µmol O2 cm–2 h–1) (Table 1).

In the light, pH profiles corresponded with oxygenprofiles, with the highest pHs, generally, at thesame depths as the highest oxygen concentrations(Fig. 9a,c). Maximum pH values of 9.6 and 10.5 weremeasured in the fluffy and the compact mats, respec-tively (Fig. 9c). In the dark, pH remained constant in boththe sediment and the fluffy mat, while in the compactmat a relatively steep decrease was observed (Fig. 9f).

Sulfide was not detected in the upper 9 mm in thediatom-dominated sediment (Fig. 9b,e). In the fluffy mat,some sulfide was measured from 8 mm downward onlyin the dark (Fig. 9e). In the compact mat, increasing con-centrations with depth (up to 500 µM) were found in bothlight and dark. However, the oxygen-sulfide transitionlayer was pushed downward in light as the result ofphotosynthetic O2 production. (Fig. 9b,e). Apparent ratesof sulfate reduction for the compact mat, calculated fromthe sulfide profiles, reached 0.084 and 0.0798 µmolcm–2 h–1 in light and dark, respectively (Table 1).

The degree of compactness of each community isclearly reflected in the product φ × Ds, which expressdiffusivity (Table 1). As expected, minimum values ofφ × Ds were found in the compact mat. The highestvalue of φ × Ds was observed in the fluffy mat, reveal-ing that the compactness of the fluffy mat was evenlower (0.921 × 10–5 cm2 s–1) than that of the sediment(0.789 × 10–5 cm2 s–1). The water content in the upper2 mm and O2 penetration depth were positively corre-lated with diffusivity (Fig. 10a,b). Meiofaunal abun-dance and diffusivity seems to be positively relatedalthough large changes in diffusivity between thecompact mat and the diatom-dominated sedimentwere not matched by a similar increase in meiofauna(Fig. 10c). However, the large abundance of macro-fauna in the diatom-dominated sediment might beresponsible for the relatively high diffusivity measuredin this community (Fig. 11a).

61

a) Chlorophyll a b) Bacteriochlorophyll a

0 50 100 150 200

0

2

4

6

8

10

Sediment

Fluffy mat

Compact mat

0 10 20 30 40 50

Fig. 7. Vertical distribution of(a) chlorophyll a and (b) bacte-riochlorophyll a in the 3 com-munities studied. Each pointcorresponds to the mean valueof 3 samples. Error bars re-present ±SD. ww: wet weight


Percentage

Dep

th (m

m)

0 20 40 60 80

0

2

4

6

8

10

SedimentFluffy matCompact mat

0 20 40 60 80 100

a) Organic matter b) % water

Fig. 8. Vertical distribution of (a) organic matter content and (b) water in the 3 communities studied. Each point corresponds to the mean value of 3 samples. Error bars represent ±SD

0 200 600 1000 0 200 400 600 8 9 10

Overlying water

Sediment

Fluffy mat

Compact mat

surface

Overlying water

surface

–2

0

2

4

6

8

–2

0

2

4

6

8

Dep

th (m

m)

Dep

th (m

m)

Fig. 9. Steady-state profiles of oxygen, H2S and pH in the 3 benthic communities studied (a–c) illuminated with 220 µmol photonsm–2 s–1 and (d–f) in darkness. Each profile is the mean of 2 to 5 measurements. Error bars are not presented for clarity


Meiofauna and macrofauna

Several different taxonomic groups of meiofaunalorganisms were found: nematodes, harpacticoid cope-pods, amphipods (juvenile), polychaetes and largeciliates (>300 µm). In the fluffy mat, meiofaunal abun-dance was very high (mainly nematodes and harpacti-coids) in comparison to sediment and the compact mat,where meiofaunal abundance was very low (Fig. 11a).The concentration of nematodes was 8-fold higher inthe fluffy mat than in the sediment, while harpacticoidswere only found in the fluffy mat (0.51 × 106 ind. m–2).Abundances of nematodes and harpacticoids weresignificantly different among the different communi-ties (p < 0.001, ANOVA).

Four different groups of macroorganisms were found:dipteran larvae (Chironomus salinarius and Halo-cladius varians), polychaetes (Capitella capitata andNereis diversicolor), Lamellibranchia (Cerastodermaglaucum) and amphipods (Cymadusa filosa, Micro-deutopus gryllotalpa and Melita palmata) (Fig. 11b).No living gastropods were found, but a large numberof empty shells from Hydrobia minoricensis and H.ulvae were present (Fig. 11c), especially in the baresediment, indicating that this group can temporarilycolonize this habitat. Polychaetes were only present insediment (1.17 × 104 ind. m–2), while dipteran larvaewere as abundant in the sediment as in the fluffy mat

63

50

60

70

80

90

100

2

4

6

8

10

12

Diffusivity

O2

pene

trat

ion

(mm

)W

ater

con

tent

(ave

rage

%)

Mei

ofau

na (i

nd. c

m–2

)

0

50

100

150

200

250

300

4 986 75 10

a y = 27.663 * e (1.0971e + 05x)R = 0.8481

b y = 0.131 * e (4.553e + 05x)R = 0.97855

c y = 3.9461 * e (3.9596e + 05x)R = 0.88015

S

FM

CM

FM

FM

S

SCM

CM

Fig. 10. Correlations between diffusivity and (a) water content,(b) oxygen penetration depth, and (c) meiofauna in the 3benthic communities studied. Exponential fits (fine line) arerepresented in all cases. CM: compact mat; FM: fluffy mat;

S: sedimentFig. 11. Abundance of (a) meiofauna and (b) macrofauna perm2 and (c) empty shells from Hydrobia ulvae and H. minori-censis. Each bar represents the mean value of 3 samples.

Error bars represent ±SD


(2 × 103 ind. m–2) (Fig. 11b). However, the maindipteran species in the sediment was the red C. sali-narius, while H. varians (green in colour) onlyappeared in the fluffy mat. Amphipods were muchmore abundant in sediment (ca. 1.6 × 104 ind. m–2) thanin the fluffy mat (3.5 × 103 ind. m–2). The most strikingresult was the absence of macrofauna in the compactmat. Abundances of amphipods, polychaetes, dipteranlarvae, and numbers of Hydrobia spp. shells all dif-fered significantly among the 3 communities (p <0.003, ANOVA). Abundances of Lamellibranchia didnot differ among communities (p = 0.26).

DISCUSSION

Types of microbenthic communities and spatialdistributions

Well-developed microbial mats were mainly presentat Site 2. Sites 1 and 2 differed considerably regardingphysical forcing. Direct tidal influence at Site 1 con-trolled the water level and physico-chemical proper-ties of the water column (Fig. 2). This regulation wasapparent at short (within a day) and longer (weeks andmonths) time scales. In particular, daily tides keptsalinity changes quite constant at Site 1, while salinityat Site 2 increased from April to the middle of June.Water balance over time was more complicated at Site2, where the entry of water from the San Pedro estuarywas restricted to the very high tides and thereforechanges in the rain–evaporation balance control thewater level and salinity in this part of the lagoon thatusually dries out in summer. These observations sug-gest that microbial mats develop either in sedimentsprone to episodic emersion or seasonal dry periods, orwhere hypersaline conditions are dominant for at leastpart of the year. Both frequent emersion periods andhigh salinity are the major differences between Sites 1and 2 (Fig. 2a,c). The restriction of microbial mats toextreme environments has been repeatedly shown(Castenholz 1984, Pierson et al. 1987, Lassen et al.1992, Wieland & Kühl 2000). However, it is unclearwhether the association between microbial mats andextreme conditions is a consequence of specific physi-ological adaptation of filamentous cyanobacteria dom-inating microbial mats, and/or because extreme condi-tions are limiting for grazers. Physiological adaptationsto extreme abiotic factors could allow cyanobacteria tooutcompete other taxonomic groups of primary pro-ducers like diatoms. However, the absence of micro-bial mats in more moderate conditions like thosepresent at Site 1 is generally explained to be a resultof colonization by meio- and macrofauna (Hargrave1970, Connor et al. 1982). An experimental microcosm

study reported that microbial mats only developedwhen potential grazers were excluded (Fenchel 1998).

Microbenthic communities found at Site 2, distin-guishable to the naked eye (Fig. 3), were spatially dis-tributed according to a very clear zonation pattern(Figs. 4 & 5). From the lagoon center, the diatom-dominated sediment, the fluffy microbial mat, and thecompact microbial mat covered concentric zoneslocated at progressively shallower water depths and,therefore, under increasing physical stress. Thesemicrobial communities were different in taxonomiccomposition, vertical organization, standing stock ofautotrophic biomass, and also supported differentabundances of meio- and macrofauna (Figs. 5, 7 & 11).

Pennate diatoms dominated the oxygenic photosyn-thetic component of the sediment community of theinner part of the lagoon. Although some mat-formingmicroorganisms like Oscillatoria sp. and Spirulina sp.were present as isolated short colonies, this microbialcommunity did not constitute mat consistency. Thesediment grains, mainly fine sand, were not stronglybound either by mucus secretions or by the entangledframework of filamentous cyanobacteria as shown bythe granulometric curve (Fig. 6). Standing stocks ofphotoautotrophic biomass and total organic mattercontent in sediment were the lowest of the communi-ties studied (Figs. 7 & 8). These values were somewhatlarger than the range of concentrations found in manytidal flats in the upper first centimeter: 46 to 94 mg chla m–2 in Netarts bay, USA (Davis & McIntire 1983), 11to 240 (mean ca. 110) mg chl a m–2 in Tagus Estuary,Portugal (Brotas et al. 1995) and ca. 3 to 20 mg chl am–2 in the Molenplaat, The Netherlands (Barranguet etal. 1997). Likely, the reduced erosion rate of sedimentsurface due to little water movement in this shelteredenvironment in comparison to a tidal flat could explainthe relatively high photoautotrophic biomass of thisdiatom-dominated microbial community.

Closer to the lagoon fringe, the diatom-dominatedcommunity was replaced by a fluffy microbial mat. Thisseems to be mainly a transition community between thediatom-dominated sediment and the compact microbialmat dominated by cyanobacteria. The 2 major charac-teristics of the fluffy mat were the presence of a diatomlayer with some Oscillatoria sp. filaments on a deeperMicrocoleus band, and the fluffy consistency of the di-atom layer. Diatom species found in the fluffy mat weresimilar to those found in the diatom-dominated sedi-ment, suggesting a possible colonization from sedi-ment. The presence of a golden diatom layer oncyanobacteria-dominated microbial mats is relativelycommon (Jørgensen et al. 1983, Pierson et al. 1987,Lassen et al. 1992), but the ecological factors that deter-mine its presence are not clear. It is likely that in oursystem, diatom colonization of the compact microbial

64


mats at the lagoon fringe is limited by the lower resis-tance of diatoms to drought stress compared to mat-forming cyanobacteria. Although the diatom layer isfrequently observed in microbial mats, the extremefluffy consistency found here is less common. It is likelythat such a fragile 3D structure can persist only underquiescent hydrodynamic conditions, such as those thatoccur within the lagoon. This structure considerablyincreases the fractal dimension at the water-sedimentinterface, therefore increasing the surface for massexchange. Biofilms grown under conditions of low flowvelocity spontaneously develop a spongy architecturewith voids and channels where mass transfer is nolonger exclusively by molecular diffusion (Røy et al.2002). In addition, its 3-dimensional structure poten-tially creates many new ecological microniches (Raf-faelli & Hawkins 1996), which may explain the muchhigher abundance of meiofauna in the fluffy microbialmats compared with the diatom-dominated sedimentand the compact microbial mat. Alternatively, meio-faunal activity could also have contributed to theobserved structure (Huettel & Gust 1992), which mayremain because of low shear stress and turbulence ex-isting at Site 2.

The compact microbial mat, dominated by Micro-coleus chthonoplastes, was restricted to the lagoonfringe, where periodic emersion and desiccation aremore likely. M. chthonoplastes has a variety of physio-logical adaptations to drought stress: the production ofUV-protective substances (Garcia-Pichel & Castenholz1991), compatible solutes (Karsten 1996), capsularexopolymeric substances (De Winder et al. 1999), and ahigher competitive success at high temperature rela-tive to diatoms (Waterman et al. 1999) are some exam-ples. In addition, the very low abundance of meio- andmacrofauna in the lagoon fringe likely contributed tothe preservation of the compact mat structure. Belowthe M. chthonoplastes layer, the vertical communitystructure of both types of mats were very similar.Despite having very different vertical distributions ofphotoautrophic biomass (Fig. 7), both types of matshad similar concentrations of chl a per surface unit.

Meio- and macrofaunal abundance probably alsoinfluences microbenthic community composition inevery zone. Benthic fauna are in turn critically con-trolled by hydric stress. Macrofauna was almost absentfrom the area covered by compact microbial mats, butreached maximal densities in the innermost part of thelagoon covered by loose sediment (Fig. 11). Althoughat the time of sampling no living gastropods werefound, the number of empty Hydrobia spp. shellssuggests either a preference for the permanently im-mersed zone, or an accumulation of corpses in deeperareas where water remains for a longer time. In con-trast, meiofauna were more abundant in the fluffy mat,

with nematodes and harpacticoid copepods being themost abundant taxonomic groups. Both macrofaunaand meiofauna abundances were low in the lagoonfringe where the water level was lower and the risk ofdesiccation higher.

Community structure and function

Differences in the vertical structure and speciescomposition of the microbenthic communities consid-erably affected their net metabolism and the massexchange at the water-sediment interface. These dif-ferences were manifested in the vertical profiles of O2,pH and H2S among the 3 types of communities. Micro-electrode measurements in the 3 communities weredone under identical experimental conditions, includ-ing temperature, irradiance, and flow velocity, whichallowed for direct comparison of measured profiles,and the process rates derived from them. The differentstructures of the 3 communities were reflected in themeasured diffusivities. In the compact mat, diffusivitywas the lowest, while in the fluffy mat it was the high-est. Assuming the same oxygen concentration in thewater column, oxygen penetration depth in the darkdepends directly on diffusivity and inversely on theoxygen consumption rate within the sediment (Revs-bech et al. 1980, Rasmussen & Jørgensen 1992). Insummary, the fluffy mat was characterized by the high-est diffusivity coefficients, O2 penetration and O2 con-sumption rates, while the compact mat showed thelowest ones.

The oxygen penetration depth in the light insediments inhabited by photosynthetic communitiesdepends strongly on incident photon flux density. Inthe light, photosynthetic oxygen production consider-ably increases the oxygen availability in the sedimentand the oxygen penetrates down to deeper layers(Jørgensen et al. 1983). It is possible that photon fluxdensity used in this study (220 µmol photons m–2 s–1)was not saturating, but since it was identical duringlight measurements for the 3 communities, it is inter-esting to compare the differences found between darkand light. The fluffy microbial mat showed the biggestdifferences between oxygen penetration depth in thelight (10 mm) and in the dark (1.8 mm). This 8.2 mmdifference represents an important contribution of theoxygenic photosynthetic community to the volume ofaerobic sediment during daytime, where the oxidationof organic matter is carried out by aerobic respira-tion. Lower values were measured for the diatom-dominated sediment (3.9 mm) and the compact micro-bial mat (1.3 mm), which were similar to those reportedin the literature for these types of communities (Jør-gensen et al. 1983, Wieland & Kühl 2000). At a given

65


photon flux density, oxygen penetration depends onthe net photosynthetic rate in the photic layer, theoxygen consumption rate (including both biotic andabiotic processes) in the deepest layer, and on thesediment diffusivity. The structure of the fluffy micro-bial mat was responsible for a high diffusivity andprobably led to a considerably deeper penetration oflight, which caused a much deeper O2 penetration inthe light than for the other communities.

Below the oxic layer, mineralization of the organicmatter depends upon the availability of electronacceptors other than O2. In marine sediments, sulphatereduction is the major anaerobic processes to completethe oxidation of organic matter (Jørgensen 1982).Our results show important differences among the 3benthic microbial communities regarding dissolvedsulphide concentration within the sediment (Fig. 10).Sulphide was not detected in the diatom-dominatedsediment in the upper 9 mm in the light or in the dark.In the fluffy mat, small amounts of H2S were onlydetected below 8 mm in the dark when photosyntheticactivity was suppressed. However, high H2S concen-trations were measured in the compact mat below theoxic layer both in the light and dark. Sulphate reduc-tion in coastal sediments is mainly controlled by tem-perature and the input of organic matter (Moeslund etal. 1994). Since the experimental conditions for themicroelectrodes measurement were identical, temper-ature cannot explain the observed differences in H2Sconcentration. Therefore, the absence of H2S in thediatom-dominated sediment is likely related to itslower organic-matter content (Fig. 8). Different reoxi-dation rates could also occur among communities,since the O2 penetration depth was also different inlight. Dissolved sulphide is not usually found close tothe surface in low-organic coastal sediments, wherethe reoxidation with Fe and Mn is rapid (Moeslund etal. 1994, Thamdrup et al. 1994). However, the usualpresence of H2S in microbial mats is due to the highaccumulation of biomass per unit of volume and totheir relatively high primary production rates (Cohen& Rosenberg 1989, Kühl & Jørgensen 1992). Both con-ditions were observed in the compact microbial mat,where areal net photosynthesis rate and oxygenicphotosynthetic biomass was higher than in other sedi-ment types (Table 1, Fig. 7). The fluffy mat containedonly slightly lower integrated chl a and a larger con-tent of organic matter (Figs. 7 & 8); however, areal netphotosynthesis rate was the lowest (Table 1). In thefluffy and compact mats, the presence and the positionof the H2S-O2 interface was associated with the whitelayer dominated by the sulphide oxidizing bacteriaBeggiatoa. This layer was more developed in thecompact mat and was absent in the diatom-dominatedsediment, where no H2S was detected.

Food chains from microbenthic communities

Primary production by microphytobenthos is theessential energy and organic carbon source for the het-erotrophic community inhabiting the lagoon. Besidesbeing a very shallow environment, the mean chl aconcentrations in the water column during the sameperiod shown in Fig. 2 were 3.05 and 1.45 µg chl a l–1

for Sites 1 and 2, respectively (results not shown).Assuming a mean depth of 0.5 m, the integrated chl aconcentrations in the water column, 1.52 and 0.72 mgchl a m–2, were more than 2 orders of magnitude lowerthan those measured in the benthic communities stud-ied, ranging from 172 to 721 mg m–2 for the diatom-dominated community and the compact microbial mat,respectively. Phytoplankton biomass was likely limitedby the low inorganic nutrient concentration in thewater column. Therefore, the balance between photo-synthesis and respiration by the benthic microbialcommunity seems primarily responsible for the ob-served daily changes in O2 saturation and pH in thewater column at both sites (Fig. 2). However, it is diffi-cult to assess the relative contribution of each commu-nity, since they have different chl a concentrations andmeio- and macrofaunal abundances.

Given their differences in photoautotrophic biomass,the 3 communities are likely to have different contribu-tions to the total primary production of the system.In addition, the fraction of the fixed C that will be re-cycled within each microbial community or exported tohigher trophic levels seems to differ between commu-nities as well. Comparison of net photosynthesic car-bon fixation rate and aerobic respiration rate in thedark provides insight into the fate of fixed C for eachcommunity and how much will be available for highertrophic levels. In the diatom-dominated sediment, netcarbon fixation during the light period was 40.1 µg Ccm–2 d–1, while dark respiration was 21.6 µg C cm–2 d–1

and thus the daily net primary production was 18.5 µgC cm–2 d–1. This daily net primary production did notaccumulate in the sediment because both photosyn-thetic biomass and organic matter content in the sedi-ment were the lowest of the 3 communities. Given theabundance of macrofauna in the diatom-dominatedsediment, it is likely that a large fraction of the dailynet primary production is channelled to higher trophiclevels. The flow of organic carbon to anaerobic miner-alization was apparently low because no H2S wasdetected in this community. The turnover rate for thephotosynthetic component (P:B) can be approachedby dividing the areal net photosynthesis rate by thephotoautotrophic biomass, both expressed in carbonunits. The P:B ratio for the oxygenic photosyntheticcomponent was 0.09 d–1, the highest of the 3 com-munities investigated.

66


Net photosynthetic carbon fixation in the fluffymicrobial mat was relatively low, 20.2 µg C cm–2 d–1,although this community had an areal standing stockof chl a (689 mg chl a m–2) similar to that of the compactmat (721 mg chl a m–2). This low rate is probably theresult of respiration by the abundant meiofauna inhab-iting mostly the upper diatom layer. Dark respirationwas 18 µg C cm–2 d–1, and therefore the daily net pri-mary production was only 2.2 µg C cm–2 d–1. Given theabundance of meiofauna, they seem to be the majorconsumers of primary production in the diatom layer inthe fluffy mat. In addition, high numbers of meiofaunamay indirectly stimulate microphytobenthic produc-tion by enhancing nutrient availability and thus allow-ing a deeper penetration of light (Fenchel & Straarup1971, Jørgensen & Des Marais 1986). Direct exploita-tion by meiofauna of primary production at the level ofthe Microcoleus chthonoplastes band, where the chl amaximum in this community was detected, is moreunlikely due to periodic anoxia. Because meiofaunaare known to be mainly restricted to the oxic layer insediments (Knox 2000), it is likely that the existence ofdaily anaerobic conditions during the night at the levelof the M. chthonoplastes layer prevented it from beinggrazed by meiofauna. The destiny of the fixed carbonin this layer is to either accumulate in the sedimentor be mineralized anaerobically. The presence of sul-phide in the dark indicates the activity of sulphate-reducing bacteria. In addition, the fluxes of O2 at themat–water interface might have been underestimateddue to the fluffy architecture of the diatom layer. Thestructure of this layer, which includes many voids andchannels, might have prevented molecular diffusionfrom being the dominant mass-transfer mechanism.We have some evidence of this because our attempts tomeasure gross primary production inside the fluffydiatom layer by the dark-light shift technique failed.This technique is based on Fick’s second law of molec-ular diffusion (Kühl et al. 1996). If this was the case,both net photosynthesis rate and respiration in darkmight have been underestimated.

The compact microbial mat presented the highestareal net photosynthesis rate. On a daily basis, the netcarbon fixation rate was 50.5 µg C cm–2 d–1. Only afraction of this C was mineralized during the night:14.4 µg C cm–2 d–1. Since meio- and macrofauna werealmost absent in the area covered by the compactmicrobial mat, the daily net primary production(36.1 µg C cm–2 d–1) was used in mat growth and/oroxidized by anaerobic processes. The importance ofsulphate reduction as a mechanism of anaerobicmineralization of the organic carbon was evident fromthe high H2S concentration found in this mat and,structurally, in the presence of a very conspicuousBeggiatoa layer.

Acknowledgements. This work was financially supportedby grants MAT2000-0261, REN2002-01281/MAR, from theMinisterio de Ciencia y Tecnología, Spain. We acknowledgeJ. A. Muñoz, from Observatorio Meteorológico, for providingclimatological data. The authors thank the reviewers fortheir helpful suggestions, which substantially improved themanuscript.

LITERATURE CITED

Aller RC, Aller Y (1992) Meiofauna and solute transport inmarine muds. Limnol Oceanogr 37:1018–1033

Arias A, Drake P (1999) Fauna acuática de las salinas delParque Natural Bahía de Cádiz. EGMASA, Seville

Awramik SM (1984) Ancient stromatolites and microbialmats. In: Cohen Y, Castenholz RW, Halvorson HO (eds)Microbial mats: stromatolites. AR Liss, New York, p 1–22

Barranguet C, Herman PMJ, Sinke JJ (1997) Microphyto-benthos biomass and community composition studied bypigment biomarkers: importance and fate in the carboncycle of a tidal flat. J Sea Res 38:59–70

Bourrelly P (1970) Les algues d’eau douce. Initiation à la sys-tématique. I: Les algues bleues et rouges, les eugléniens,peridiniens et cryptomonadines. N Boubée & Cie, Paris

Bourrelly P (1972) Les algues d’eau douce. Initiation à la sys-tématique. II: Les algues vertes. N Boubée & Cie, Paris

Bourrelly P (1981) Les algues d’eau douce. Initiation à lasystématique. III: Les algues jaunes et brunes. Chryso-phycées, phéophycées, xanthophycées et Diatomées. NBoubée & Cie, Paris

Broecker WS, Peng TH (1974) Gas exchange rate betweensea and air. Tellus 26:21–35

Brotas V, Cabrita T, Portugal A, Serodio J, Catarino F (1995)Spatio-temporal distribution of the microphybenthic bio-mass in intertidal flats of Tagus Estuary (Portugal). Hydro-biologia 300/301:93–104

Castenholz RW (1984) Composition of hot spring microbialmats: a summary. In: Cohen Y, Castenholz RW, HalvorsonHO (eds) Microbial mats: stromatolites. AR Liss, NewYork, p 39–58

Castenholz RW, Gherna GL, Lewin R, Rippka R, WaterburyJB, Whitton BA (1989) Oxygenic photosynthetic bac-teria. In: Staley JT, Bryant MP, Pfennig N, Holt JG(eds) Bergey’s manual of systematic bacteriology, Vol 3.Williams & Wilkins, Baltimore, MD, p 1710–1806

Chrétiennot-Dinet MJ (1990) Atlas du phytoplancton marin,Vol III: Chlorarachniophycées, chlorophycées, cryptophy-cées, euglenophycées, eustigmatophycées, prasinophycées,prymnésiophycées, rhodophycées et tribophycées. CentreNational de la Recherch Scientifique, Paris

Cohen Y, Rosenberg E (1989) Microbial mats: physiologi-cal ecology of benthic microbial communities. AmericanSociety for Microbiology, Washington, DC

Connor MS, Teal JM, Valiela I (1982) The effect of feeding bymud snails, Ilyanass obsoleta (Say), on the structure andmetabolism of a laboratory benthic algal community.J Exp Mar Biol Ecol 65:29–45

Cullen DJ (1973) Bioturbation in superficial marine sedimentsby interstitial meiobenthos. Nature 242:323–324

Davis MW, McIntire CD (1983) Effects of physical gradientson the production dynamics of sediment-associated algae.Mar Ecol Prog Ser 13:103–114

De Winder B, Staats N, Stal LJ, Paterson DM (1999) Carbo-hydrate secretion by phototrophic communities in tidalsediments. J Sea Res 42:131–146

de Wit R, van Gemerden H (1989) Growth responses of the

67


cyanobacterium Microcoleus chthonoplastes with sulfideas an electron donor. In: Cohen Y, Rosenberg E (eds)Microbial mats: physiological ecology of benthic microbialcommunities. American Society for Microbiology, Wash-ington, DC, p 320–325

Dubinin AV, Gerasimenko LM, Zavarzin GA (1992) Nitrogenfixation by cyanobacterium Microcoleus chthonoplastesfrom hypersaline lagoons of Lake Sivash. Mikrobiologya61:593–597

Eaton JW, Moss B (1966) The estimation of numbers andpigment content in epipelic algal populations. LimnolOceanogr 11:584–595

Epping EH, Khalili A, Thar R (1999) Photosynthesis and thedynamics of oxygen consumption in a microbial mat ascalculated from transient oxygen microprofiles. LimnolOceanogr 44:1936–1948

Fenchel T (1998) Formation of laminated cyanobacterial matsin the absence of benthic fauna. Aquat Microb Ecol 14:235–240

Fenchel T, Straarup BJ (1971) Vertical distribution of photo-synthetic pigments and the penetration of light in marinesediments. Oikos 22:172–182

Garcia-Pichel F, Castenholz RW (1991) Characterization andbiological implications of scytonemin, a cyanobacterialsheath pigment. J Phycol 27:395–409

Giere O (1993) Meiobenthology. The microscopic fauna inaquatic sediments. Springer-Verlag, Berlin

Glud RN, Jensen K, Revsbech NP (1995) Diffusivity in surficialsediments and benthic mats determined by use of a com-bined N2O-O2 microsensor. Geochim Cosmochim Acta 59:231–237

Hargrave BT (1970) The effect of deposit-feeding amphipodon the metabolism of benthic microflora. Limnol Oceanogr15:21–30

Holme NA (1971) Macrofauna sampling. In: Holme NA,McIntyre AD (eds) Methods for the study of marine ben-thos. Blackwell Scientific Publications, Oxford, p 80–130

Huettel M, Gust G (1992) Impact of bioroughness on inter-facial solute exchange in permeable sediments. Mar EcolProg Ser 89:253–267

Jeroschewski P, Steuckart C, Kühl M (1996) An amperometricmicrosensor for the determination of H2S in aquatic envi-ronments. Anal Chem 68:4351–4357

Joint IR (1978) Microbial production of an estuarine mudflat.Estuar Coast Mar Sci 7:185–195

Jørgensen BB (1982) Mineralization of organic matter inthe sea bed—the role of sulphate reduction. Nature 296:643–645

Jørgensen BB, Des Marais DJ (1986) A simple fiberopticmicroprobe for high resolution light measurements:application in marine sediment. Limnol Oceanogr 31:1376–1383

Jørgensen BB, Revsbech NP, Cohen Y (1983) Photosynthesisand structure of benthic microbial mats: microelectrodeand SEM studies of four cyanobacterial communities.Limnol Oceanogr 28:1075–1093

Karsten U (1996) Growth and organic osmolites of geographi-cally different isolates of Microcoleus chthonoplastes(Cyanobacteria) from benthic microbial mats: response tosalinity change. J Phycol 32:501–506

Knox GA (2000) The ecology of seashores. CRC Press, NewYork

Komárek J, Anagnostidis K (1986) Modern approach to theclassification system of cyanophytes 2-Chroococcales.Arch Hydrobiol Suppl 73:157–226

Komárek J, Anagnostidis K (1989) Modern approach to theclassification system of cyanophytes 4-Nostocales. Arch

Hydrobiol Suppl 82:247–345Krumgalz B, Fainshtein G (1989) Trace metal contents in cer-

tified reference sediments determined by nitric acid diges-tion and atomic absorption spectrometry. Anal Chim Acta218:335–340

Kühl M, Jørgensen BB (1992) Microsensor measurements ofsulfate reduction and sulfide oxidation in compact micro-bial communities of aerobic biofilms. Appl Environ Micro-biol 58:1164–1174

Kühl M, Steuckart C, Eickert G, Jerochewski P (1998) A H2Smicrosensor for profiling biofilms and sediments: applica-tion in an acidic lake sediment. Aquat Microb Ecol 15:201–209

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

Lassen C, Ploug H, Jørgensen BB (1992) Microalgal photo-synthesis and spectral scalar irradiance in coastal marinesediments of Limfjorden, Denmark. Limnol Oceanogr 37:760–772

Li YH, Gregory S (1974) Diffusion of ions in sea water and indeep sea sediments. Geochim Cosmochim Acta 38:703–714

Lorenzen JR, Glud RN, Ploug H, Ramsing NB (1995) Impact ofmicrosensor caused changes in diffusive boundary layerthickness on O2 profiles and photosynthetic rates in ben-thic communities of microorganisms. Mar Ecol Prog Ser119:237–241

Moeslund L, Thamdrup B, Jørgensen BB (1994) Sulfur andiron cycling in a coastal sediment: radiotracer studies andseasonal dynamics. Biogeochemistry 27:129–152

Moezelaar R, Bijvank SM, Stal LJ (1996) Fermentation andsulfur reduction in the mat-building cyanobacteriumMicrocoleus chthonoplastes. Appl Environ Microbiol 62:1752–1758

Pierson B, Oesterle A, Murphy GL (1987) Pigments, lightpenetration, and photosynthetic activity in the multi-layered microbial mats of Great Sipewisset Salt Marsh,Massachusetts. FEMS Microbiol Ecol 45:365–376

Pinckney JL, Carman KR, Lumsden SE, Hymel SN (2003)Microalgal-meiofaunal trophic relationships in muddyintertidal estuarine sediments. Aquat Microb Ecol 31:99–108

Raffaelli D, Hawkins S (1996) Intertidal ecology. Chapman &Hall, London

Rasmussen H, Jørgensen BB (1992) Microelectrode studies ofseasonal oxygen uptake in a coastal sediment: role ofmolecular diffusion. Mar Ecol Prog Ser 81:289–303

Revsbech NP (1989) An oxygen microelectrode with a guardcathode. Limnol Oceanogr 34:474–478

Revsbech NP, Jørgensen BB (1986) Microelectrodes: their usein microbial ecology. In: Marshall KC (ed) Advances inmicrobial ecology, 9. Plenum, New York, p 293–352

Revsbech NP, Sorensen J, Blackburn TH, Lomholt JP (1980)Distribution of oxygen in marine sediments measured withmicroelectrodes. Limnol Oceanogr 25:403–411

Revsbech NP, Jørgensen BB, Blackburn TH, Cohen Y (1983)Microelectrode studies of photosynthesis and O2, H2S, andpH profiles of a microbial mat. Limnol Oceanogr 28:1062–1074

Rippka R, Deruelles J, Waterbury JB, Herdman M, Stanier RY(1979) Generic assigments, strain histories and propertiesof pure cultures of Cyanobacteria. J Gen Microbiol 111:1–61

Round FE, Crawford RM, Mann DG (1990) The diatoms.Biology and morphology of the genera. CambridgeUniversity Press, Cambridge

68


Røy H, Hüttel M, Jørgensen BB (2002) The role of small-scalesediment topography for oxygen flux across the diffusiveboundary layer. Limnol Oceanogr 47:837–847

Sournia A (1986) Atlas du phytoplancton marin, Vol I. Intro-duction, cyanophycées, dictyochophycées, dinophycéeset raphidophycées. Centre National de la RechercheScientifique, Paris

Staley JT, Bryant MP, Pfennig N, Holt JG (1989) Bergey’smanual of systematic bacteriology, Vol 3. Williams &Wilkins, Baltimore, MD

Thamdrup B, Fosing H, Jørgensen BB (1994) Manganese, ironand sulfur cycling in a coastal marine sediment, AarhusBay, Denmark. Geochim Cosmochim Acta 58:5115–5129

Thompson RC, Tobin ML, Hawkins SJ, Norton TA (1999)Problems in extraction and spectrophotometric determi-

nation of chlorophyll from epilithic microbial biofilms: to-wards a standard method. J Mar Biol Assoc UK 79:551–558

Van den Hoek C, Mann DG, Jahns HM (1995) Algae. Anintroduction to phycology. Cambridge University Press,Cambridge

Waterman F, Hillebrand H, Gerdes G, Krumbein WE, Som-mer U (1999) Competition between benthic cyanobacteriaand diatoms as influenced by different grain sizes andtemperatures. Mar Ecol Prog Ser 187:77–87

Wieland A, Kühl M (2000) Short-term temperature effects onoxygen and sulfide cycling in a hypersaline cyanobacter-ial mat (Solar Lake, Egypt). Mar Ecol Prog Ser 196:87–102

Zar JH (1984) Biostatistical analysis, 2nd edn. Prentice-Hall,Englewood Cliffs, NJ

69

Editorial responsibility: Kevin Carman, Baton Rouge, Louisiana, USA

Submitted: January 6, 2004; Accepted: October 22, 2004Proofs received from author(s): January 12, 2005

Download - Microbenthos in a hypersaline tidal lagoon: factors ...Sediment cores (n = 3) with a surface area of 4 cm2 (as recommended by Eaton & Moss 1966) were taken from each of the benthic

Top Related