Abstract

The sea surface microlayer is the interfacial boundary layer between the marine environment and the troposphere. Surface microlayer samples were collected during a fjord mesocosm experiment to study microbial assemblage dynamics within the surface microlayer during a phytoplankton bloom. Transparent exopolymer particles were significantly enriched in the microlayer samples, supporting the concept of a gelatinous surface film. Dissolved organic carbon and bacterial cell numbers (determined by flow cytometry) were weakly enriched in the microlayer samples. However, the numbers of Bacteria 16S rRNA genes (determined by quantitative real-time PCR) were more variable, probably due to variable numbers of bacterial cells attached to particles. The enrichment of transparent exopolymer particles in the microlayer and the subsequent production of a gelatinous biofilm have implications on air–sea gas transfer and the partitioning of organic carbon in surface waters.

Introduction

The sea surface microlayer is the physical boundary layer between the ocean and the atmosphere. Physicochemically distinct relative to subsurface waters, the sea surface microlayer is a unique marine ecosystem that is often called the neuston (Liss & Duce, 2005; Marshall & Burchardt, 2005). A recent study has shown the enrichment of transparent exopolymer particles (TEP) in the sea surface microlayer compared with underlying subsurface water (Wurl & Holmes, 2008). TEP are ubiquitous gels in the oceans, produced in surface waters from the coagulation of phytoplankton-derived dissolved polysaccharides (Alldredge et al., 1993; Verdugo et al., 2004). Gel particles in the water column are readily colonized by bacterial cells, with attached cells accounting for up to 68% of the total bacterial assemblage (determined by DAPI counts) (Alldredge et al., 1993; Passow & Alldredge, 1994; Mari & Kiørboe, 1996). TEP are ‘sticky’ and as such they facilitate the aggregation of particles (e.g. faecal pellets, phytoplankton, zooplankton food webs) in the water column, therefore playing a critical role in marine snow formation and are thus important in the biogeochemical cycling of carbon and nitrogen pools (Alldredge et al., 1993; Verdugo et al., 2004). An enrichment of TEP in the sea surface microlayer may also result in similar aggregation processes taking place, giving the surface film properties analogous to aggregates in the water column (Cunliffe & Murrell, 2009).

It is hypothesized that an enrichment of TEP in the sea surface microlayer will result in a gelatinous surface film (Sieburth, 1983; Wurl & Holmes, 2008) and this film may therefore support a relatively high proportion of attached bacterial cells compared with free living bacterial cells (Cunliffe & Murrell, 2009). The purpose of this study was to determine TEP, dissolved organic carbon (DOC) and bacterial cell numbers by flow cytometry and Bacteria 16S rRNA gene copy number in the same surface microlayer samples and compare them with those in cognate subsurface water samples collected during an induced phytoplankton bloom mesocosm experiment. This characterization of the fjord surface film is discussed in the context of role of the film in air–sea gas transfer, marine biogeochemistry and microbial ecology.

Materials and methods

Sample collection

Samples were collected from a fjord phytoplankton bloom mesocosm experiment at the University of Bergen Marine Biological Station, 20 km south of Bergen, Norway (May 2008). The mesocosms were filled (2474 L) with prefiltered (∼300 μm) water from Raunafjorden and mixed continually with submerged aquarium pumps. Nutrients (16 μM NaNO3 and 1 μM KH2PO4) were added at day 0 and the development of the bloom in subsurface water (sampling depth 0.75 m) was monitored daily over a period of 11 days at the same time (09:00 hours) by phytoplankton cell counting using flow cytometry (M. Cunliffe et al., unpublished data) (Fig. 1). Samples were also collected at the middle (day 5) and end of the experiment (day 10) (Fig. 1) from the surface microlayer using a metal mesh screen with a sampling depth of ∼400 μm (Cunliffe et al., 2009). Water (250 mL) from the mesh screen was collected into a sterile bottle and then filtered through a Sterivex-GS filter unit (pore size 0.2 μm; Millipore). Cognate subsurface water samples collected using a syphon (sampling depth 0.75 m) were also processed in the same way.

1

Development of the phytoplankton bloom in a fjord mesocosm was determined by flow cytometry counting of phytoplankton cells in subsurface waters (0.75 m depth). The white circles are control mesocosms and the grey circles are nutrient amended mesocosms (n=3; ±SD). The surface microlayer was also sampled on day 5 and day 10.

1

Development of the phytoplankton bloom in a fjord mesocosm was determined by flow cytometry counting of phytoplankton cells in subsurface waters (0.75 m depth). The white circles are control mesocosms and the grey circles are nutrient amended mesocosms (n=3; ±SD). The surface microlayer was also sampled on day 5 and day 10.

Chemical analysis

TEP in the samples were quantified spectrophotometrically using the alcian blue dye-binding assay (Passow & Alldredge, 1995). Filtered membranes were stained with 500 μL alcian blue solution [0.02% (w/v) alcian blue in 0.06% (v/v) acetic acid] and then rinsed with deionized water before the stained membrane was soaked in 6 mL 80% (v/v) sulphuric acid for 2 h. The absorbance of the solution was then measured at 787 nm in a 1 cm cuvette and values expressed as equivalents of the commercial polysaccharide Xanthan Gum (Sigma).

DOC was quantified in the sample filtrate by high temperature catalytic oxidation using a VPN (Shimadzu) apparatus (Sugimura & Suzuki, 1988).

Bacterial abundance analysis

Bacterial abundance was determined using two approaches, flow cytometry and quantitative real-time PCR (Q-PCR) of Bacteria 16S rRNA genes. Flow cytometry was performed with unfiltered samples using a Becton Dickinson FACScalibur benchtop flow cytometer (BD Bioscience). DNA was extracted from the membranes in a sucrose buffer using lysozyme, proteinase K, SDS and phenol–chloroform (Cunliffe et al., 2008). The resuspended DNA was diluted in molecular grade water to a concentration of 30 ng μL−1 and stored at −20 °C. Enumeration of Bacteria 16S rRNA genes in the extracted DNA was performed using primers 338F and 518R with minor modifications from Einen et al. (2008) using Power SYBR® Green (Applied Biosciences) and the reactions run on an ABI PRISM 7000 (Applied Biosciences) Q-PCR machine. All samples were run in triplicate and the standard curve, made from a dilution series of Escherichia coli genomic DNA ranging from 75 to 7.5 × 10−5 ng, had an R2 value of 0.96.

Statistical analysis

anova was used to identify statistical significance in the data (n=3; three replicate control mesocosms and three replicate nutrient amended mesocosms; P<0.05). Where significant differences were seen, Tukey's test was used to compare data within a defined set. Both anova and Tukey's test were performed using spss statistical software (SPSS).

Results and discussion

TEP

TEP concentrations ranged between 720 and 880 μg xanthan equivalent (eq.) L−1 (mean 813) on day 5 and between 900 and 960 μg xanthan eq. L−1 (mean 920) at day 10 in the subsurface water samples collected from the control mesocosms. There was very little increase in the concentration of TEP in the subsurface water samples collected from the nutrient amended mesocosms, ranging from 920 to 1000 μg xanthan eq. L−1 (mean 947) on day 5 and from 1040 to 1200 μg xanthan eq. L−1 (mean 1147) on day 10 (Fig. 2a).

2

(a) TEP, (b) DOC, (c) bacterial cell abundance (determined by flow cytometry) and (d) Bacteria 16S rRNA gene copies were quantified in surface microlayer samples and compared with cognate subsurface water samples (n=3; ±SD).

2

(a) TEP, (b) DOC, (c) bacterial cell abundance (determined by flow cytometry) and (d) Bacteria 16S rRNA gene copies were quantified in surface microlayer samples and compared with cognate subsurface water samples (n=3; ±SD).

TEP concentrations in the surface microlayer samples were significantly higher (P≤0.04) compared with subsurface water samples, ranging from 1440 to 3280 μg xanthan eq. L−1 (mean 2400) at day 5 and from 1920 to 3040 μg xanthan eq. L−1 (mean 2373) at day 10 in the control mesocosms and from 2080 to 2240 μg xanthan eq. L−1 (mean 2133) and from 2000 to 3760 μg xanthan eq. L−1 (mean 2773) at day 5 and day 10, respectively, in the nutrient-amended mesocosms (Fig. 2a).

The enrichment of TEP in the sea surface microlayer has only recently been considered (Kuznetsova et al., 2005; Wurl & Holmes, 2008). Wurl & Holmes (2008) quantified TEP in estuarine and oceanic samples collected from South East Asia. The TEP concentrations determined are comparable with those in this study (Table 1), especially from estuarine samples where biological activity was highest. Other studies have reported lower TEP concentrations (Table 1) and this may reflect the level of biological activity in the mesocosms in this study.

1

Averaged concentrations of TEP in surface microlayer (SML) and subsurface water (SS) samples from this study compared with previous studies

 Sample details SML or SS Depth TEP (μg xanthan eq. L−1) Reference ± SD Range Norwegian fjord mesocosm SML ∼400 μm 2420 ± 651 1400–3760 This study Norwegian fjord mesocosm SS 0.75 m 957 ± 139 720–1200 This study Jonor Strait (estuarine) (Jan–May 2007) SML ∼50 μm 1731 ± 1340 177–3831 Wurl & Holmes (2008) Jonor Strait (estuarine) (Jan–May 2007) SS 1 m 2264 ± 2535 281–8026 Wurl & Holmes (2008) Singapore Strait (oceanic) (Jan–May 2007) SML ∼50 μm 638 ± 349 94–1297 Wurl & Holmes (2008) Singapore Strait (oceanic) (Jan–May 2007) SS 1 m 512 ± 288 129–1073 Wurl & Holmes (2008) Monterey Bay (July 1993) SS 0–10 m 271 ± 30 81–310 Passow & Alldredge (1995) Monterey Bay (July 1993) SS 10–50 m 59 ± 12 46–63 Passow & Alldredge (1995) Norwegian fjord (May 1992) SS 0–18 m 190 ± 53 100–255 Passow & Alldredge (1995) Norwegian fjord (May 1992) SS 21–63 m 191 ± 45 125–250 Passow & Alldredge (1995)
 Sample details SML or SS Depth TEP (μg xanthan eq. L−1) Reference ± SD Range Norwegian fjord mesocosm SML ∼400 μm 2420 ± 651 1400–3760 This study Norwegian fjord mesocosm SS 0.75 m 957 ± 139 720–1200 This study Jonor Strait (estuarine) (Jan–May 2007) SML ∼50 μm 1731 ± 1340 177–3831 Wurl & Holmes (2008) Jonor Strait (estuarine) (Jan–May 2007) SS 1 m 2264 ± 2535 281–8026 Wurl & Holmes (2008) Singapore Strait (oceanic) (Jan–May 2007) SML ∼50 μm 638 ± 349 94–1297 Wurl & Holmes (2008) Singapore Strait (oceanic) (Jan–May 2007) SS 1 m 512 ± 288 129–1073 Wurl & Holmes (2008) Monterey Bay (July 1993) SS 0–10 m 271 ± 30 81–310 Passow & Alldredge (1995) Monterey Bay (July 1993) SS 10–50 m 59 ± 12 46–63 Passow & Alldredge (1995) Norwegian fjord (May 1992) SS 0–18 m 190 ± 53 100–255 Passow & Alldredge (1995) Norwegian fjord (May 1992) SS 21–63 m 191 ± 45 125–250 Passow & Alldredge (1995)

TEP enrichment factors (EF) in this study ranged between 2.3 and 2.9 (Table 2). These are higher than those reported by Wurl & Holmes (2008) in oceanic (mean 1.3) and estuarine (mean 1.8) samples and may be a result of the calm conditions emulated in the mesocosms (M. Cunliffe et al., unpublished data). Nevertheless, the enrichment of TEP in the sea surface microlayer reported in this study and previous studies (Kuznetsova et al., 2005; Wurl & Holmes, 2008) indicates a biogenic gelatinous surface microlayer film (Sieburth, 1983; Wurl & Holmes, 2008; Cunliffe & Murrell, 2009). The presence of a biogenic surface microlayer film at the air–sea interface will have a profound effect on gas transfer between the ocean and the troposphere, suppressing the gas transfer velocity between the two systems (Frew et al., 1990).

2

Averaged surface microlayer enrichment factors of TEP, DOC, bacterial cell counts from flow cytometry and Bacteria 16S rRNA gene copy numbers

 TEP DOC Cell counts 16S rRNA gene copies SD SD SD SD Day 5 control 2.9 0.9 1.4 0.2 1.2 0.5 1.4 0.5 Day 5 nutrient amended 2.3 0.0 1.1 0.2 1.4 0.7 1.5 0.9 Day 10 control 2.6 0.5 1.5 0.6 0.9 0.3 11.6 6.5 Day 10 nutrient amended 2.5 1.0 1.2 0.3 1.1 0.1 1.8 0.2
 TEP DOC Cell counts 16S rRNA gene copies SD SD SD SD Day 5 control 2.9 0.9 1.4 0.2 1.2 0.5 1.4 0.5 Day 5 nutrient amended 2.3 0.0 1.1 0.2 1.4 0.7 1.5 0.9 Day 10 control 2.6 0.5 1.5 0.6 0.9 0.3 11.6 6.5 Day 10 nutrient amended 2.5 1.0 1.2 0.3 1.1 0.1 1.8 0.2

Possible routes of TEP into the surface microlayer have been studied using a laboratory system, which used natural TEP collected from the Santa Barbara Channel (California) (Azetsu-Scott & Passow, 2004). The study showed that TEP ascend the water column and become enriched in surface waters. Possible causes of the ascension of TEP into the microlayer include particle density and the surface-active nature of TEP (Azetsu-Scott & Passow, 2004).

DOC

The concentrations of DOC ranged between 136 and 141 μM C (mean 137) at day 5 and between 189 and 223 μM C (mean 202) at day 10 in the subsurface water samples collected from the control mesocosms, and between 156 and 203 μM C (mean 176) at day 5, and between 208 and 311 μM C (mean 264) at day 10 in the subsurface water samples collected from the nutrient amended mesocosms (Fig. 2b).

In the surface microlayer samples, the concentrations of DOC ranged between 174 and 222 μM C (mean 198) at day 5 and between 205 and 409 μM C (mean 288) at day 10 in the control mesocosms and between 178 and 260 μM C (mean 199) at day 5 and between 275 and 338 μM C (mean 302) at day 10 in the nutrient-amended mesocosms. The concentrations of DOC in this study are within the same range as those reported previously for estuarine and oceanic samples (Wurl & Holmes, 2008). EF for DOC in all the surface microlayer samples were low, ≤1.5 (Table 2). Low DOC EF have been reported previously from a variety of sample sites and range between <1.5 and 2 (Carlson, 1983; Williams et al., 1986; Wurl & Holmes, 2008).

TEP in the surface microlayer may facilitate the formation of aggregates (Cunliffe & Murrell, 2009). In this study, during sample filtration, aggregates were visibly more abundant on the filter membranes from the surface microlayer samples compared with filter membranes from the subsurface water samples (Fig. 3). Azetsu-Scott & Passow (2004) have shown that TEP-rich aggregates can ascend in the water column. Importantly, the attachment of positively buoyant TEP to a particle, which would normally descend can cause the particle to ascend. This is possible with natural marine snow-sized aggregates (Azetsu-Scott & Passow, 2004), therefore the aggregates enriched in the surface microlayer samples in this study (Fig. 3) could be marine snow-type aggregates.

3

Example of the visible enrichment of aggregates on the filter membranes of surface microlayer (SML) samples compared with subsurface water (SS) samples collected on day 10.

3

Example of the visible enrichment of aggregates on the filter membranes of surface microlayer (SML) samples compared with subsurface water (SS) samples collected on day 10.

Marine aggregates are very porous, with up to 99% of their volume being interstitial fluid (Alldredge & Gotschalk, 1988). Because of microbial processes within aggregates (i.e. the conversion of particulate organic matter to dissolved organic matter), the concentration of DOC within the interstitial fluid can be up to two orders of magnitude higher than in the surrounding seawater (Alldredge, 2000). However, aggregate-derived DOC contributes little to seawater DOC (Alldredge, 2000), as aggregates may not be very leaky, and due to tight coupling between DOC production and DOC uptake by attached bacterial cells (Goldthwait et al., 2005). Furthermore, Goldthwait et al. (2005) showed, postfiltration, a substantial amount of interstitial DOC remains within the aggregate and is not released into the filtrate. This suggests that quantification based on filtrate DOC considerably underestimates total DOC in the surface microlayer.

Bacterial abundance

Bacterial cell abundance in the subsurface water samples at day 5 ranged between 2.8 × 105 and 8.0 × 105 cells mL−1 (mean 5.5 × 105) in the control mesocosms and between 1.1 × 105 and 9.6 × 105 cells mL−1 (mean 6.6 × 105) in the nutrient-amended mesocosms. By day 10, there evidently had been a significant increase (P=0.004) in cell abundance in the nutrient-amended mesocosms, with a range of 7.6 × 105–1.8 × 106 cells mL−1 (mean 1.4 × 106) but not in the control mesocosms, with a range of 5.1 × 105–8.9 × 105 cells mL−1 (mean 6.9 × 105) (Fig. 2c).

Change in bacterial cell abundance in the surface microlayer samples followed the same pattern as bacterial cell abundance in the subsurface water samples (Fig. 2c). At day 5, cell abundance in the control mesocosm samples had a range of 5 × 105–7.1 × 105 cells mL−1 (mean 6 × 105) and cell abundance in the nutrient-amended mesocosm samples had a range of 6.1 × 105–1 × 106 cells mL−1 (mean 7.7 × 105). As with bacterial cell abundance in subsurface water samples, by day 10, there was a significant increase (P=0.001) in cell abundance in the nutrient-amended mesocosm samples, with a range of 8 × 105–1.9 × 106 cells mL−1 (mean 1.5 × 106) compared with cell abundance in the control mesocosm samples, with a range of 4 × 105–1 × 106 cells mL−1 (mean 6.1 × 105) (Fig. 2c). Bacterial cell abundance EF values in the surface microlayer were low, ≤1.4 (Table 2).

The number of Bacteria 16S rRNA gene copies in the subsurface water samples on day 5 ranged between 6.3 × 105 and 1 × 106 copies mL−1 (mean 9.5 × 105) in the control mesocosms and between 1.3 × 106 and 1.5 × 106 copies mL−1 (mean 1.4 × 106) in the nutrient-amended mesocosms. Numbers increased substantially on day 10 (Fig. 2d), with gene copy numbers in the control mesocosm samples ranging between 1.1 × 1010 and 1.5 × 1010 copies mL−1 (mean 1.3 × 1010) and in the nutrient-amended mesocosms ranging between 5.1 × 1010 and 6.4 × 1010 copies mL−1 (mean 5.9 × 1010).

As with bacterial cell abundance determined by flow cytometry, Bacteria 16S rRNA gene copy numbers in the surface microlayer samples generally followed the same pattern of change as the number of gene copies in the subsurface water samples (Fig. 2d). At day 5, the number of gene copies ranged between 1.2 × 106 and 1.23 × 106 copies mL−1 (mean 1.2 × 106) in the control mesocosm samples and between 7.9 × 105 and 3.5 × 106 copies mL−1 (mean 2.1 × 106) in the nutrient-amended mesocosm samples. Gene copies increased at day 10, with numbers ranging between 5.9 × 1010 and 2.3 × 1011 copies mL−1 (mean 1.5 × 1011) in the control mesocosms and between 9.6 × 1010 and 1.3 × 1011 copies mL−1 (mean 1.1 × 1011) in the nutrient-amended mesocosm samples. There was a significant enrichment of gene copies (P=0.021) in the surface microlayer samples compared with cognate subsurface samples at day 10 (Fig. 2d). Therefore, the Bacteria 16S rRNA gene copy EF for the control mesocosms at day 10 were substantially higher at 11.6 compared with all others (Table 2).

Low EF values for bacterial cell counts determined by flow cytometry have been reported previously, suggesting weak enrichment of bacterial cells in the sea surface microlayer (Agogue et al., 2004, 2005; Joux et al., 2006). Copy numbers of Bacteria 16S rRNA genes in the surface microlayer samples in this study corroborated this observation for most of the samples, i.e. low but not statistically significant enrichment in the surface microlayer. However, there was a significant enrichment of Bacteria 16S rRNA genes in the microlayer of the control mesocosm samples at day 10. Flow cytometry will count the free-living bacterial cell pool, whereas Q-PCR of genes from DNA extracted from filtered membranes will enumerate both the free-living and attached bacterial cell pools. Previous studies of TEP in subsurface waters have reported a broad range of values for the percentage of attached bacterial cells, 0.5–68% of the total number of cells (Alldredge et al., 1993; Mari & Kiørboe, 1996). This suggests that a high degree of variability exists between the attached and free-living bacterial cell pools in relation to TEP and this variability may account for the observations reported here.

There was a marked increase in the number of Bacteria 16S rRNA gene copies between day 5 and day 10 (Fig. 2d). Flow cytometry differentiated between high nucleic acid (HNA) and low nucleic acid bacterial cells based on cell-specific SYBR Green DNA staining (M. Cunliffe et al., unpublished data). The relative abundance of HNA bacterial cells in all mesocosms increased significantly (P≤0.007) from day 5 to day 10 (Fig. 4). HNA bacterial cells are probably the active fraction of the total bacterial assemblage (Vaque et al., 2001). A higher proportion of HNA bacterial cells at day 10 may contribute to a higher number of 16S rRNA genes at day 10, because, Bacteria groups which respond rapidly to increased resource availability may contain higher numbers of rRNA genes (Klappenbach et al., 2000).

4

Increase on day 10 in the relative abundance of HNA bacterial cells and the decrease in the relative abundance of low nucleic acid (LNA) bacterial cells (n=3).

4

Increase on day 10 in the relative abundance of HNA bacterial cells and the decrease in the relative abundance of low nucleic acid (LNA) bacterial cells (n=3).

Conclusions

This study contributes to the characterization of the sea surface microlayer, specifically in the context of microbial ecology. The enrichment of TEP in the surface microlayer in this and previous studies (Kuznetsova et al., 2005; Wurl & Holmes, 2008) continues to empirically support the hypothesis of a biogenic gelatinous surface microlayer film (Sieburth, 1983). The presence of a TEP-rich film indicates that microbial degradation of TEP in the microlayer is slower than the rate of TEP ascension into the microlayer from the water column below. Furthermore, the gelatinous film may act as an important reservoir of primary produced organic carbon from the water column. DOC in the water of the surface microlayer is weakly enriched. However, if the surface microlayer film is an organic aggregate then future work needs to consider interstitial fluids within the film matrix. Bacterial numbers, on the whole, also appear weakly enriched in the microlayer. However, there is evidence from the enumeration of Bacteria 16S rRNA genes that some variation exists and these bacterial hot spots could be associated with TEP.

Acknowledgements

This work was supported by the Natural Environment Research Council (UK) through the project – SOLAS Bergen Mesocosm experiment (NE/E011446/1), which is a part of the NERC–Surface Ocean Lower Atmosphere Study (SOLAS) directed programme. We thank all the people involved in the project who helped with the preparation and sampling of the mesocosms. We also thank Steve Mowbray at the University of Edinburgh for DOC analysis.

References

Agogue
H
Casamayor
EO
Joux
F
et al. (
2004
)
Comparison of samplers for the biological characterization of the sea surface microlayer
.
Limnol Oceanogr-Meth

2
:
213
225
.
Agogue
H
Casamayor
EO
Bourrain
M
Obernosterer
I
Joux
F
Herndl
GJ
Lebaron
P
(
2005
)
A survey on bacteria inhabiting the sea surface microlayer of coastal ecosystems
.
FEMS Microbiol Ecol

54
:
269
280
.
Alldredge
AL
(
2000
)
Interstitial dissolved organic carbon (DOC) concentrations within sinking marine aggregates and their potential contribution to carbon flux
.
Limnol Oceanogr

45
:
1245
1253
.
Alldredge
AL
Gotschalk
C
(
1988
)
In situ settling behaviour of marine snow
.
Limnol Oceanogr

33
:
339
351
.
Alldredge
AL
Passow
U
Logan
BE
(
1993
)
The abundance and significance of a class of large, transparent organic particles in the ocean
.
Deep-Sea Res Pt I

40
:
1131
1140
.
Azetsu-Scott
K
Passow
U
(
2004
)
Ascending marine particles: significance of transparent exopolymer particles (TEP) in the upper ocean
.
Limnol Oceanogr

49
:
741
748
.
Carlson
DJ
(
1983
)
Dissolved organic materials in surface microlayers: temporal and spatial variability and relation to sea state
.
Limnol Oceanogr

28
:
415
431
.
Cunliffe
M
Murrell
JC
(
2009
)
The sea surface microlayer is a gelatinous biofilm
.
ISME J

3
:
1001
1003
.
Cunliffe
M
Schafer
H
Harrison
E
Cleave
S
Upstill-Goddard
RC
Murrell
JC
(
2008
)
Phylogenetic and functional gene analysis of the bacterial and archaeal communities associated with the surface microlayer of an estuary
.
ISME J

2
:
776
789
.
Cunliffe
M
Harrison
E
Salter
M
Schafer
H
Upstill-Goddard
RC
Murrell
JC
(
2009
)
Comparison and validation of sampling strategies for the molecular microbial ecological analysis of surface microlayers
.
Aquat Microb Ecol
, in press.
Einen
J
Thorseth
IH
Øverås
L
(
2008
)
Enumeration of Archaea and Bacteria in seafloor basalt using real-time quantitative PCR and fluorescence microscopy
.
FEMS Microbiol Lett

282
:
182
187
.
Frew
NM
Goldman
JC
Dennett
MR
Johnson
AS
(
1990
)
Impact of phytoplankton-generated surfactants on air–sea gas-exchange
.
J Geophys Res-Oceans

95
:
3337
3352
.
Goldthwait
SA
Carlson
CA
Henderson
GK
Alldredge
AL
(
2005
)
Effects of physical fragmentation on remineralization of marine snow
.
Mar Ecol-Prog Ser

305
:
59
65
.
Joux
F
Agogue
H
Obernosterer
I
Dupuy
C
Reinthaler
T
Herndl
GJ
Lebaron
P
(
2006
)
Microbial community structure in the sea surface microlayer at two contrasting coastal sites in the northwestern Mediterranean Sea
.
Aquat Microb Ecol

42
:
91
104
.
Klappenbach
JA
Dunbar
JM
Schmidt
TM
(
2000
)
rRNA operon copy number reflects ecological strategies of Bacteria
.
Appl Environ Microb

66
:
1328
1333
.
Kuznetsova
M
Lee
C
Aller
J
(
2005
)
Characterization of the proteinaceous matter in marine aerosols
.
Mar Chem

96
:
359
377
.
Liss
PS
Duce
RA
(
2005
) The sea surface and global change. Cambridge University Press, UK.
Mari
X
Kiørboe
T
(
1996
)
Abundance, size distribution and bacterial colonization of transparent exopolymer particles (TEP) during spring in the Kattegat
.
J Plankton Res

18
:
969
986
.
Marshall
HG
Burchardt
L
(
2005
)
Neuston: Its definition with a historical review regarding its concept and community structure
.
Arch Hydrobiol

164
:
429
448
.
Passow
U
Alldredge
AL
(
1994
)
Distribution, size and bacterial colonization of transparent exopolymer particles (TEP) in the ocean
.
Mar Ecol-Prog Ser

113
:
185
198
.
Passow
U
Alldredge
AL
(
1995
)
A dye-binding assay for the spectrophotometric measurement of transparent exopolymer particles (TEP)
.
Limnol Oceanogr

40
:
1326
1335
.
Sieburth
JM
(
1983
)
Microbiological and organic-chemical processes in the surface and mixed layers
.
Air–Sea Exchange of Gases and Particles
(
Liss
PS
Slinn
WGN
, eds), pp.
121
172
.
Reidel Publishers Co.
, Hingham, MA.
Sugimura
Y
Suzuki
Y
(
1988
)
A high temperature catalytic oxidation method of non-volatile dissolved organic carbon in seawater by direct injection of liquid samples
.
Mar Chem

24
:
105
131
.
Vaque
D
Casamayor
EO
Gasol
JM
(
2001
)
Dynamics of whole community bacterial production and grazing losses in seawater incubations as related to the changes in the proportions of bacteria with different DNA content
.
Aquat Microb Ecol

25
:
163
177
.
Verdugo
P
Alldredge
AL
Azam
F
Kirchman
DL
Passow
U
Santschi
PH
(
2004
)
The oceanic gel phase: a bridge in the DOM-POM continuum
.
Mar Chem

92
:
67
85
.
Williams
PM
Carlucci
AF
Henrichs
SM
Vanvleet
ES
Horrigan
SG
Reid
FMH
Robertson
KJ
(
1986
)
Chemical and microbiological studies of sea-surface films in the southern gulf of California and off the west-coast of Baja-California
.
Mar Chem

19
:
17
98
.
Wurl
O
Holmes
M
(
2008
)
The gelatinous nature of the sea-surface microlayer
.
Mar Chem

110
:
89
97
.
Editor: Aharon Oren