The fate of organic carbon and nitrogen in ... - IngentaConnect

Journal of Marine Research, 45, 231-257, 1987

The fate of organic carbon and nitrogen in experimental marine sediment systems: Influence of bioturbation and anoxia by E. Kristensen1 and T. H. Blackburn2

ABSTRACT The decay rate of particulate organic carbon (PaC) and nitrogen (PON) was followed during 94 days in three homogenized sediment microcosms: I. With a natural density of the polychaete Nereis virens (NOx-cores); 2. Defaunated, with an aerobic water phase (Ox-cores); and 3. Defaunated, with an anaerobic water phase (An-cores). In all cores there was a marked preferential mineralization of paN compared to pac. The presence of Nereis increased the net decomposition of pac and paN 2.6 and 1.6 times relative to Ox-cores. Ventilation of burrow structures by the worms increased the flux of O2, TC02 and DIN across the sediment-water interface 2.5-3.5 times. This significantly decreased the pore water concentrations of TC02 and DIN. Similarly, nitrification and denitrification were stimulated 2.3-2.4 times due to nereid activity. Oxygen did not increase organic degradation: in fact, the decay of pac and PON was faster in An- than in Ox-cores, 1.5-1.6 and 1.2 times, respectively. Sulfate reduction, measured at the end of experiment, was surprisingly low in the aerobic NOx- and Ox-cores relative to An-cores. Net ammonium production measured at the end of the experiment agreed with the mean loss of paN for Ox- and An-cores, but was low for NOx-cores, suggesting that a high C:N substrate was being degraded in these cores at the end. An empirical model describing the temporal decay pattern of pac and PON is presented: the detritus in all cores were initially composed of two fractions (similar C:N); a readily degradable (-43%) and a low degradable (-57%) fraction. A substantial part ofthe degradable fraction in NOx-cores was used during the experiment, with nitrogen being mineralized preferentially. The mean C:N molar ratio of detritus used was 5.9, compared to a value of 15.5 determined at the end. The Ox- and An-cores, however, showed similar C:N ratios for the detritus used during the experiment (3.7 and 4.8) and that measured at the end (4.2 and 4.6). Presumably not all the low C:N detritus had yet been mineralized in these cores at the end of experiment.

1. Introduction Knowledge of temporal and spatial changes in the pool sizes of particulate organic carbon and nitrogen are important for understanding the course of decomposition processes in coastal marine sediments (Blackburn, 1980; Blackburn and Henriksen, 1983). The rate of organic matter decomposition and microbial growth is generally inversely correlated with the age and C:N molar ratio of the substrate (Blackburn, I. Institute 2. Institute

of Biology, University of Odense, DK-5230 Odense M, Denmark. of Ecology and Genetics, University of Aarhus, DK-gooD Aarhus C, Denmark. 231

232

Journal of Marine Research

[45, 1

1980; Linley and Newell, 1984). This is usually observed by a decrease in organic matter mineralization with depth (age) in sediments. Aerobic respiration with O2 as terminal electron acceptor takes place in the oxic surface layer, whereas sulfate reduction is most significant in the deeper, reduced sediment. Nitrate reduction (e.g. denitrification) dominates just below the oxic zone. In general, the two former processes are the quantitatively most important, each being responsible for -50% of the total respiration, while the latter process usually accounts for less than 10% (S~rensen et a/., 1979; J~rgensen and S~rensen, 1985). Other mineralization processes, i.e. manganese oxide reduction, ferric iron reduction and methanogenesis, are generally of unknown importance in the overall oxidation of organic matter in marine sediments (Fenchel and Blackburn, 1979). Oxygen depletion in bottom layers of coastal waters during warm summer months is a widespread and increasing phenomenon all over the world (J~rgensen, 1980; Seliger et al., 1985). Anthropogenic inputs of organic matter and nutrients, with a subsequent high sedimentation rate of highly degradable organic particles, are responsible for a high benthic oxygen demand, leading to anoxia. Sediment oxygen uptake usually exceeds aerobic respiration, because reduced metabolites like HS-, NH4 + and CH4 are reoxidized by oxygen at the oxic/anoxic interface (J~rgensen, 1983). Virtually nothing is known about changes in rates and distribution of microbial processes when the bottom turns anoxic. Laboratory experiments on reaction rates of algal material during oxic and anoxic decomposition, however, have shown little difference (Foree and McCarty, 1970; Fallon and Brock, 1979). Infaunal animals are known to affect both rates and spatial distribution of sediment processes by their feeding, burrowing and ventilation activities (e.g. Aller, 1982; Kristensen, 1986). Studies of infaunal effects on microbial reaction rates have largely been based on modelling of pore water solute distribution, exchange rates across the sediment-water interface and bacterial distributions (e.g. Aller and Yingst, 1985; Kristensen, 1984; 1985). Very few reports have dealt with direct measurements of microbial activities associated with infaunal animals (Kristensen et al., 1985). The experimental microcosms used in this study offer an opportunity to examine the influence of infaunal polychaetes and anoxia on carbon and nitrogen dynamics in sediments. Natural heterogeneity, together with sedimentation events, usually impedes this kind of experiment under in situ conditions. In order to determine the loss of organic matter quantitatively, we found it necessary to homogenize the sediment. Homogenization by sieving, freezing and mixing, however, destroys the original structure, chemical profiles, and faunal populations in the sediment. Results from our microcosms are, therefore, not directly comparable to those obtained in situ, and they should merely be regarded as idealized sediments without natural heterogeneity.

2. Materials and methods Sediment and specimens of Nereis virens were collected in the shallow mesohaline estuary, Norsminde Fjord, Denmark (Muus, 1967; J~rgensen and S~rensen, 1985).

1987]

Kristensen & Blackburn: Organic carbon & nitrogen in sediments

233

The sampling site was situated near the entrance of the estuary at a water depth of -0.3 m. Annual salinity and temperature range in the area was - I to 25°C and 8 to 23%0, respectively.

a. Sediment collection and experimental set-up The uppermost -5 cm of the sediment surface was dredged and sieved through a 1.5 mm mesh on location, in order to remove macrofauna and larger particles (i.e. shells and gravel). Simultaneously, average-sized individuals (-0.3 g wet wt.) of the polychaete Nereis virens were collected and kept separate from the sediment. Tanks containing sieved sediment to a depth of -15 cm and worms were transported to the laboratory for further treatment. Since the objective of this study was to examine the effects of nereid polychaetes and anoxia on microbial processes and flux rates, meiofaunal organisms present in the sediment were killed before the start of experiment. Otherwise the activities of growing macrofauna I larvae and other meiofauna would have affected the processes in the aerobic, but not in the anaerobic cores. These small animals were killed by freezing the sediment for 48 hours at -20°C before further use. Microscopy on sediment samples before and after the freezing procedure revealed that live larvae and meiofauna had disappeared following the treatment. Freezing was not expected to reduce the bacterial populations severely. After thawing, the sediment was homogenized by handmixing in the transport-tanks and compaction was allowed to proceed. Four days later, 9 identical 10 cm long cores were taken with 4.6 cm i.d. plexiglass corers; to 3 of these were added 3 preweighed individuals of Nerds virens (NOx-cores), 3 were used as aerobic controls without Nereis (Ox-cores) and 3 were supplied with an anoxic (N2 purged) water phase (An-cores). All cores were supplied with overlying water (20%0) to a depth of 5 cm and placed in darkness with continuous stirring. Temperature was 22°C. The 6 aerobic (Ox- and NOx-) cores were open to air, while the anaerobic were closed with rubber stoppers, allowing a 3 cm anoxic N 2 head-space.

b. Flux measurements Once weekly (-each 7th day) for 13 weeks the experimental cores were incubated to determine solute exchange between the sediment and overlying water. Before incubations, the "old" water was removed and samples were frozen for later Dissolved Inorganic Nitrogen (DIN) analysis ("7 day" samples). Fresh GF-C filtered water was gently added to the two aerobic (NOx- and Ox-) core-types, which were then closed with rubber stoppers, allowing no head-space. The An-cores were treated similarly, except that the added water previously was purged with N2 for 0.5 hour. After a 3-5 hour incubation period with continuous stirring in darkness, samples were taken for analysis of O2, TC02 (C02, HC03- and C03--), HS- and DIN (NH4+, N02- and N03 -) ("4 hour" samples). Samples for dissolved gases were analyzed immediately. Oxygen was determined in duplicate by the Winkler technique. TC02 was determined by the step-wise potentio-


234

[45, I

metric Gran-titration (Edmond, 1970). Dissolved HS- was measured by the spectrophotometric diamine method of Cline (1969). Samples for DIN were quickly frozen and analyzed in duplicates as soon as possible. The standard autoanalyzer methods of Solorzano (1969) for NH4 + and of Armstrong et al. (1967) for N02- and N03 - were used. Exchange rates were calculated from concentration changes during the "4 hour" incubation period, using the volume of water trapped above the sediment and given as mmol m-2d-1• Rates for NH/, N02 - and N03 - exchange, however, are presented as weighed means of "4 hour" and "7 day" samples.

c. Sediment treatment Before and after the 94 day experimental period, sediment samples were taken to determine sediment characteristics, pore water solute concentrations and microbial activities. Samples taken from the transport-tank after the freezing procedure represented the start condition. At the end of the experiment, all cores were cut into 0-1, 1-2, 2-4, 4-6, 6-8 and 8-10 cm sections. During this, worms were collected from the NOx-cores and weighed. The sediment organic content was determined as loss on ignition at 500°C (LO!). POC and PON were analyzed by a Hewlett-Packard 185B CHN-analyzer. Total solid carbonates were determined on pre-ignited samples by the CHN-analyzer. This procedure is based on the assumption that no carbonate is lost by ignition at 500°C (Dean, 1974; Kristensen and Andersen, in prep). Pore water was obtained by centrifugation for 10 min at 3000 rpm in double centrifuge tubes. The supernatant was analyzed for TC02, NH/, N02 - and N03 -. Pore water TC02 was analyzed in triplicate on a Beckman model 865 Infrared Gas Analyzer by the method of Salonen (1981). Samples for NH4 +, N02 - and N03 - were frozen and later analyzed in duplicate as described earlier.

d. Microbial activity /5 N-turnover. A modification of the method of Blackburn

(1979) was used. Portions (5.0 of sediment were added from a cut-off syringe, under nitrogen, to 50 ul 0.01 M 99.8% 15N-NH4Cl in glass scintillation vials, closed with rubber stoppers. Sediment in the vials was incubated for 0, 5, 7 and 9 days at 22°C. Incubation was stopped by the addition of 1.0 M KCI. After vigorous mixing, the KCl extraction was continued for 0.5 h, before centrifugation at 3000 rpm for 10 min. The concentration of NH4 + in extracts was measured, and the 15N-content was determined by emission spectrometry. The net rate of NH4 + -production was obtained from the change in NH/ concentration with time and expressed as nmol cm-3 sediment d-I• The total rate of NH4 + production and the rate of NH4 + incorporation into bacterial cells, was derived from the 15N dilution data. cm3)

Sulfate reduction. A modified version of the method of J~rgensen (1977) was used. Portions (3.0 cm3) of sediment were added from a cut-off syringe, under nitrogen, to

1987]


235

10 ul 35S04-- in 13 cm3 glass serum bottles, closed with rubber stoppers. The amount of label in the acid-volatile sulfide pool and in the original sulfate pool was measured after incubation for two days. The rate of sulfate reduction was calculated from the ratio of these two values and the sulfate concentration.

Potential nitrification. Nitrification

activity in the pooled sediment samples was measured in aerobic sediment slurries enriched with NH4 +. Duplicate samples (-2 g sediment each) were dark-incubated at 22°C in 120 ml serum bottles with 40 ml filtered 200;00sea water. The water was previously enriched with NH4CI and KH2P04 to concentrations of 500 and 50 ILM, respectively (Henriksen et al.. 1981). Slurries were kept aerobic by continuous shaking to assure a maximum nitrification rate and inhibit dissimilatory nitrate r~duction. Samples of 2 ml were taken initially and after 3, 6, and 10 h incubation. The potential nitrification was determined from the accumulation of N02 - and N03 - and expressed as nmol cm-3 sediment h-1•

Potential nitrate reduction and denitrification. Denitrification

and total nitrate reduction of the pooled samples were determined in anaerobic sediment slurries enriched with NO) and purged with C2H2• The addition of C2H2 inhibits the ultimate reduction of N20 to N2 in the denitrification process (S~rensen, 1978b). Duplicate samples (-2 g sediment each) were incubated at 22°C in 120 ml darkened serum bottles with 40 ml filtered sea water. This was previously enriched with KN03 to a concentration of 200 ILM and made anaerobic by bubbling with N2• The slurries were purged extensively with N2 and subsequently with C2H2 for a few minutes before the serum bottles were stoppered with butyl rubber caps. The bottles were shaken throughout the experiment. Sampling occurred initially and after 1,4 and 9 h. Water samples of 2 ml were obtained for determination of total nitrate reduction, defined as loss of N03- and expressed as nmol cm-3 sediment h-1• Denitrification was determined by the rate of N20 accumulation, taking 3 ml gas samples by evacuated glass vials (Venoject, Terumo Corp.), expressed as nmol N20-N cm-3 sediment h-'. Concentrations of N20 were analyzed by injection of 0.3 ml samples into a Packard 427 gas chromatograph equipped with a 65Ni Electron Capture Detector. Gases were separated on a 2 m x 3.2 mm Porapak Q column.

3. Results a. Visual observations Regular visual inspections of the sediments revealed that, from identical start conditions, the three core-types developed differently. After a few days the sediment in NOx-cores had an irregular 1-2 cm thick brown, possibly oxidized zone at the sediment-water interface. The Nereis-burrows extended the oxidized zone into the black reduced sediment by a 1-3 mm thick wall lining down to -8 cm depth. The sediment surface in these cores was influenced by the feeding and searching activities


236

[45, I

Table I. Initial and final porosity of sediment in the three core types. Depth in cores (cm)

Nereis

Aerobic

Anaerobic

Initial:

0-10

0.34

0.34

0.34

Final:

0-1 1-2 2-4 4-6 6-8 8-10

0.44 0.34 0.32 0.30 0.27 0.29

0.37 0.34 0.32 0.29 0.28 0.25

0.35 0.32 0.31 0.28 0.27 0.28

of the worms. All Ox-cores quickly developed a regular 6-8 mm thick oxidized brown zone in the top-sediment. After 3-4 weeks the lower part of this zone turned reddish from the precipitation of oxidized iron compounds. The uppermost 2 cm of reduced sediment, that had a grayish-black appearance, was underlayed by a distinct I cm thick very black zone. Below this zone and down to the bottom the color appeared less black, although darker than the uppermost reduced sediment. The conspicuous black zone at 2-3 cm depth was presumably dominated by iron monosulfide precipitation. The sediment in An-cores turned black from top to bottom almost immediately after the start of experiment. The color was similar to the reduced layers in the NOx-cores and the bottom part of the Ox-cores.

x 10-2 0

% OF ORY WEIGHT 3 ,

2

,

20

,

/~

2

~

'

:r ~

6

UJ

a 8

10

f! NO

,

80

A:;-'O Jo"X

/ / \!

\\ Ii

A~ox

•

/7 Ii

'J;\

•

01

/ \/!

x

•

; An

I

•

70

~\l

".0:

\~ • O.

60

'"

Vii

j'

PON

.~. 30,

I

poe

.01

1 I

LO! I

~

I

Figure l. Final profiles of Particulate Organic Nitrogen (PON), Carbon (PaC) and Loss On Ignition (LOI) in the sediment of Nereis (NOx)-, Aerobic (Ox)- and Anaerobic (An)-cores. The dashed lines represent the initial condition. Values are given as % of dry weight. S.D. (not shown) were in the range of 2-25%.

1987]


Table 2. Average decay constant, k = -dG/Gdt core types over 94 days.

237

(yr-1), for LOI, POC and PON in the three

Nereis

Aerobic

Anaerobic

0.529 0.475 0.861

0.056 0.183 0.553

0.293 0.277 0.621

b. Solid phase The porosity data show that during the experiment compaction had occurred below 4 cm depth in all three sediment types (Table 1). In the 0-1 cm section of NOx-cores, however, the worms were responsible for a 20% increase in porosity relative to the other cores. The bottom sediment Of Ox- and An-cores developed pockets of gas (from 8-20 cm3 per core) during the experiment. These consisted of ~90% N2 and ~ 10% CH4• No such gas production was observed in the NOx-cores. The initial organic content in the sediment was low (LOI at start: 0.69%) compared to coastal marine sediments in general. POC and PON accounted for 37 and 4% of the LOI, respectively. During the 94 day experimental period all sediments showed a reduction in LOI, POC and PON (Fig. 1). The loss occurred at all depths, except for an enrichment in the 0-1 cm section of the two aerobic sediment-types; NOx- and Ox-cores. No similar enrichment occurred in the An-cores. The overall loss of LOI, POC and PON was most dramatic in the NOx-cores (13.6, 12.2 and 22.1 %, respectively) followed by the An-cores (7.5, 7.1 and 15.9%, respectively) and finally the Ox-cores (1.4, 4.7 and 13.8%, respectively). Nitrogen was mineralized preferentially to carbon (kPON/ kPOC = 1.8-2.9, Table 2). This was reflected in an increased C:N molar ratio of the organic matter from 10.8 initially to 11.9-12.2 after 94 days. The mean mineralization rate of POC over the experimental period was 44.9, 17.3 and 26.3 mmol m-2d-1 for NOx-, Ox- and An-cores, respectively. For PON, the mineralization rates were 7.56, 4.68 and 5.45 mmol m-2d-1, respectively. Nereids which had been added to the sediment in NOx-cores showed a loss in biomass and numbers during the experiment (Table 3). The loss in biomass was more pronounced (74%) than the loss in numbers (44%). The C:N molar ratio of Nereis virens was 4.2.

Table 3. Density and biomass of Nereis virens in NOx-cores. (t

Density Biomass

Start = 0 d)

1807 48.5 1.41 0.33

(t

End 94 d)

=

1004 12.8 0.37 0.09

m-2 gdw m-2 mol C m-2 mol N m-2


238

Teo2•

o

4

6

8

10

[45, 1

pmol ml-1 12

14

16

18

20

22

2

::E

U 4

:r:

t w

6

o 8

10

Figure 2. Final pore water concentration of TeD2 (nmol ml-I pore water) plotted against depth for the three sediment types. The dashed line represents the initial concentration. c. Pore water solutes The concentration of TC02 in the pore water after 94 days was higher than the initial value in all sediment types, as shown in Figure 2. The integrated pool size (0-10 cm) in the series NOx-, Ox- and An-cores increased by the ratios 1.0:1.5:2.0 (Table 4). The relatively low TC02 concentration and shallow gradient in the upper 5 cm of NOx-core sediment suggests a high transport rate out of the sediment due to Nereis ventilation. The pattern of NH4 + accumulation in the three sediment types was similar to that for TC02• The concentrations of NH4 + in pore water and attached to sediment particles (exchangeable) are plotted against sediment depth for each of the sediment types in Figure 3. The pool of NH/ (pore water + exchangeable) increased in the order NOx-, Ox- and An-cores by the ratios 1.0:3.3:4.3 (Table 4). The gradients were much shallower in NOx-cores than for the other sediment types. There was always a higher concentration of NH4 + in the pore water than held by the particulates. The ratio exchangeable/pore-water was 0.72, 0.63 and 0.62 for NOx-, Ox- and An-cores, respectively. Nitrate concentrations in the pore water are shown in Figure 4. The sediment in An-cores contained no N03 - at the detection level employed (-0.5 nmol ml-I). In Ox-cores the concentrations were at or below the initial level. The NOx-cores, however, had some N03 - down to 2 cm depth, with a secondary peak at 7-8 cm. The overlying water in the aerobic NOx- and Ox-cores were generally rich in N03 - after one week incubation, -22 and 65 nmol ml-I, respectively compared to -2 nmol ml-I in

1987]

Kristensen

& Blackburn:

Organic carbon & nitrogen in sediments

NEREIS nmol

00

400

+

NH4 em

8JO

AEROBIC

-3

nmol

400

1200

+

NH4 em

800

ANAEROBIC

-3 1200

239

+

nmol NH4 em

o

400

800

-3

1200

2 E

~4 :r:

~

Cl.

uJ Cl

6 8 10

Figure 3. Final concentration of pore water NHt (closed symbols) and exchangeable NHt (open symbols) (nmol cm-3 sediment) plotted against depth for each of the three sediment types. The dashed lines represent the initial pore water concentration.

Table 4. Pool sizes (mmol m-2), rate measurements (mmol m~2d-l) and C/N ratios integrated from 0 to 10 cm depth in the experimental cores at day 94.

Pools a) TC02 b) NH/ c) HS-

Nereis

Aerobic

Anaerobic

211.31 32.37 ± 0.69 35.50 ± 17.00

312.34 107.32 ± 2.15 71.70± 13.80

414.22 139.78 ± 1.87 179.60 ± 36.20

Rates d) S04 -- -Reduction e) O2 Reduction f) CO2 Production g) Net NH4 + Prod. h) NH4 + Incorp.

0.47 46.60 47.54 3.07 3.06

0.16 2.70 2.72 0.12 0.64

0.24 ± 0.18 18.50 ± 1.90 18.98 ± 1.93 4.50 ± 0.37 2.80 ± 1.60

C/N Ratios i) In detritus used j) Calculated

5.94 15.49 ± 1.07

3.70 4.22 ± 0.55

± ± ± ± ±

8.53 0.00 17.06 3.68 1.67

± ± ± ± ±

2.69 0.00 5.38 0.30 2.68

4.83 4.64 ± 1.51

The TC02 pool (a) is the total dissolved CO2 + HC03- + C03--. The NH4+ pool (b) is derived from KCI extracts minus the recovered 15N (c) is the total acid-volatile HS- pool. Sulfate reduction (d) were from 35Sassays. Rates (d) and (e) are used to calculate (f). (g) and (h) were obtained from the 15Ndilution assays. (i) refers to: loss of POC/loss of paN, over 94 days. (j) is the calculated C/N ratio in the detritus that was utilized on day 94; it is derived from: C oxidized/net N mineralization.

240

[45,1

Journal of Marine Research NOj,

o

10

20

30

nmol ml-1 40

50

60

70

NOx

::c

lll. W

o

8

10

Figure 4. Final pore water concentration of NO) (nmol ml-1 pore water) plotted against depth for NOx- and Ox-sediment. The dashed line represents the initial concentration.

the weekly added new water. This implies a high rate of nitrate production at the sediment-water interface. The resulting steep gradients (Fig. 4) suggests, however, a downward transport and consumption of N03 - at depth in the sediment. N02 - was low in the pore water and is included in the NO] - data. d. Sediment-water fluxes O2 and Te02• The temporal patterns of O2 and TC02 exchange across the sedimentwater interface showed a similar trend in the three core types (Fig. 5). The NOx-cores started with high and variable flux rates followed by a gradual decline to the end of experiment. The conspicuous peaks of both O2 and TC02 flux at day 21 were caused by the replacement of 3 dead worms with live ones. Newly added worms to a system like ours will cause an instantaneous increase in the exchange of reduced and oxidized compounds during burrow construction. The subsequent decline was most pronounced for TC02, i.e. at day 92 the flux of O2 and TC02 in these cores was % and V3, respectively of that observed initially. The Ox-cores showed a similar exchange pattern, but the overall rates of both O2 and TC02 were only 1/3 of that in the NOx-cores. Since no O2 exchange occurred in An-cores, the product of sulfate reduction, HS-, was used to describe the sediment-water flux of electron acceptors. The efflux of HS- in An-cores showed a pattern like that of TC02 and should have been stoichimetrically equivalent to the rate of TC02 flux, if all the HS- escaped from the sediment (which was unlikely). In contrast to NOx- and Ox-cores, the An-cores exhibited a gradually increased metabolic activity from day 14 to day 42 for HS- and from day 14 to day 49 for TC02• Later, fluxes of both compounds declined; for TC02 to the level obtained at day 14, and for HS- to 1/4 of the initial rate.

1987]

Kristensen

& Blackburn:


241

250

200

••

"0

N

'e

150

o e e

100

50

a

. 20

, 40

60

80

100

DAYS

80

60

40

20

o

20

40

60

80

100

DAYS

Figure 5. (A) Efflux of Te02 from the sediment plotted against time for the three core types. (8) Oxygen uptake by the sediment for NOx- and Ox-cores and HS- efflux for An-cores plotted against time. The HS- flux is converted to O2 equivalents by multiplying with a factor 2 (2 O2 molecules are needed to oxidize one HS- molecule). Values in (A) and (8) are obtained from weekly "4 hour" incubations.

242


[45, 1

11

• 10

"0

'E

\/ •

... N

Inorgonic

N -flux

A Nereis

NHZ

9

"0

E E

2

o

100 DAYS

B

4 Aerobic

3

2

• •

a

100 DAYS

Figure 6. Release of NHt, NOi and NO; from the sediment in (A) Nereis-cores, (B) Aerobic-cores, and (C) Anaerobic-cores plotted against time. Values are weighted means of "4 hour" and "7 day" incubations.

1987]

Kristensen

& Blackburn: Organic carbon & nitrogen in sediments

243

6

5

3

2

o

20

40

60

80

100

DAYS

Figure 6. (Continued)

DIN. The flux patterns of dissolved inorganic nitrogen in NOx- and Ox-cores were similar. During the first weeks the exchange was dominated by a huge effiux of NH4 + (Fig. 6A,B). On day 7, NH/ flux from NOx-cores was 2.5 times that of Ox-cores. After a 14 day lag phase N02 - and NO) - were produced. Simultaneously the NH4 + flux decreased to very low levels. The flux of N02 - was only significant for a period of 5-10 days, after which NO) - represented the sole oxidized inorganic nitrogen flux. The NO) - flux in NOx-cores remained almost constant after day 20. In Ox-cores, however, the NO) - showed a decline from a maximum flux at day 20 to only 15% of that rate at day 92. All inorganic nitrogen in the An-cores was in the form of NH4 + (Fig. 6C). Here the flux showed a maximum rate at day 7, which was comparable to that of Ox-cores on the same day. Subsequently, the flux in An-cores gradually declined and finally reached 60% of the maximum rate. The total DIN flux in the aerobic NOx- and Ox-cores showed a dramatic decrease during the first 35 days, reaching a low but relatively constant level thereafter (Fig. 7). The sharp increase observed in NOx-cores from day 14 to 21 was caused by the replacement of dead nereids with live ones. The mean DIN flux, averaged over the experimental period, showed that NOx- and Ox-cores exhibited rather low rates, i.e. 3.07 and 0.89 mmol m-2d-1 compared to An-cores, 3.78 mmol m-2d-1• Most of the DIN flux in NOx-cores, however, occurred during the first 21 days.


244

[45, I

12

total DIN - flux 11 10 6 "'j"0 " ljl

E (5 E E

5 I,

3 2

\I\NO' •

\~ ---.~~ \.. .....-'-...../"'-. Ox

'\0'8==!:::::e:::::8::::.--./· o~

0

20

1,0

60

......• 00

80

100

DAYS

Figure 7. Release of total DIN from the sediment in the three core types plotted against time.

e. Anoxic mineralization The rates of NH4 + production and incorporation at the end of experiment did not show any particular pattern of distribution, when plotted against depth in the sediment (Fig. 8). The rate of incorporation was always lower than the rate of production; this resulted in a positive net production of NH4 + for all sediment types. There was a very high S.E. about the mean values, and these S.E. have not been shown. The S.E. of the integrated values, expressed on an areal basis, are shown in Table 4. As would be anticipated, there is a large degree of uncertainty for the An-core values. In general there was a very poor recovery of 15NH4+ in these experiments. This was around 35%, but sometimes increased on incubation, which indicated that 15NH4+ had gone into a pool, from which it could slowly reappear. The net NH4 + production rates for all the core types gave statistically satisfactory values (Table 4). The final rates of sulfate reduction and the concentration of acid-volatile sulfide with sediment depth have not been presented graphically. The values for NOx- and Ox-cores were unusually low, and only the integrated rates are given in Table 4.

f Nitrification Potential activity in the uppermost layers of the aerobic NOx- and Ox-core sediment were high at the end of experiment compared to the An-cores (Fig. 9A). In NOx-core

1987]

Kristensen & Blackburn: Organic carbon & nitrogen in sediments NEREIS -1 -3 cm day 80

AEROBIC nmol NH4cm-3day-1 40 0 80

•

2

•

245

ANAEROBIC + -3 -1 nmol NH4 cm day

o

40

80

•

• E 4

u

~ :I: I- 6 a..

•

UJ

• •

0

0

8 10

•

0

• o

Figure 8. Final rates of total NHt mineralization (closed symbols) and net rates of NHt production (open symbols) plotted against depth for each of the three sediment types. The area between the lines represents the rate of NHt incorporation into bacterial cells. sediments the final nitrification activity was considerably above the initial rate down to 4 em, with a maximum rate at 0-1 cm. Below 4 cm there was no increase over the initial activity. Final rates in sediment from Ox-cores were only significantly above the start level at 0-1 em. An-core sediment exhibited final rates of potential nitrification that were at or below the initial level at all sediment depths.

g. Nitrate reduction and denitrification The three sediment types showed more or less similar patterns of nitrate reduction (Fig. 9B). Determinations of nitrate reduction, however, was subject to a high degree of variance due to the large pool of NO) - in the slurries (200 nmol cm-). The rate of NO) - removal during the incubations «10%) was close to the resolution limit of the analytical procedure used. All sediments showed a distinct maximum rate at 0-1 cm, with the An-sediment being significantly highest. Below 1 cm depth a gradual decrease occurred, reaching the initial level at the bottom layer, with Ox-cores having the highest rates and An-cores the lowest. The observed potential denitrification rates exhibited profiles different from those of nitrate reduction in all sediment types (Fig. 9C). Relatively low rates were observed at 0-1 cm, with the NOx-cores being least active. Maximum rates were observed at some depth; at 1-2 cm for Ox- and An-cores and 2-4 cm for NOx-cores. They all showed a gradual decline below these depths, reaching the initial level at 7 cm. At 0-1 cm the ratios denitrification/nitrate-reduction were low, 0.07-0.35, while values of 0.68-1.17 appeared in intermediate depths (1-8 cm). In the bottom layer (8-10 cm) the ratios decreased to 0.23-0.43 for the three sediment types. By integrating over the entire depth the ratios in the three sediment types were similar, 0.60-0.62.

[45, I


246

o

40

20

60

potential

60

100

nitrification

A 10

nmol cm-3 h-1

o

40

80

120

=z-~ .~,

160

~o.

2

4

6

8

10

rr' ~. /J /N°Yoxl

nn;':

A

'I.

0

'i-L

potential

nitrate

-200

reduction

B

Figure 9. Final rates of (A) potential nitrification, (B) potential nitrate reduction and (C) potential denitrification, plotted against depth for the three sediment types. The dashed lines represent tJ1einitial rates.

4. Discussion Carbon budgets for the three sediment systems, expressed as daily means, are presented in Table 5. Net mineralization during the experiment was significant in all systems. Loss of pac integrated from 0 to 10 cm was stimulated 2.6 and 1.6 fold in NOx- and An-cores, respectively, relative to Ox-cores. The pac decay constants of 0.183-0.475 yr-1 (Table 2) were similar to those previously reported for -other

1987]

Kristensen

& Blackburn:


247

nmol cm-3 h-1

o

20 •

40

__

60

80

.tl

:--

~~

An

-~7

2

~NOX 4

6

~l ~

8

~

-F-+

Y

po'oo,to'

''"it"""""

(N20-NI

:

C

I

10

Figure 9. (Continued)

sediment systems (Cammen, 1975; J~rgensen, 1977; Westrich and Berner, 1984). The dissolution of solid carbonates caused by metabolic acidification, i.e. 5-10% of the loss in POC (Table 5, c), was of minor importance. Unfortunately, we did not measure pools and fluxes of dissolved organic carbon (DOC), but these are generally found to be low compared to POC and TC02 (Andersen and Hargrave, 1984; Hines and Jones, 1985). The total loss of carbon from the sediments (Table 5, e) was similar to the observed mean rate of O2 consumption, suggesting a TCOd02 flux ratio around 1. This ratio is much lower than that obtained by using the directly measured TC02 flux (i.e. 2.4-3.4). The incubation methods used are probably responsible for an anomalous high diffusional TC02 exchange. After the "7 day" sampling, when the overlying Table 5. Budget for carbon in cores, expressed as mean rates (mmol m-2 d-1) over a 94 day period. Mean oxygen and sulfide fluxes are included for comparison. Nereis a. ~POC in sediment b. ~POC in Nereis c. CaCO) dissolution d. ~TC02 in pore water

-44.9 -11.1 -2.4 1.6

Aerobic

Anaerobic

-17.3

-26.3

-1.7 2.6

-1.2 3.7

e. Loss from sediment (a + b + c + d) f. Observed TC02 flux

-56.8 -186.5

-16.4 -52.6

-23.8 -64.0

g. Not accounted for (f - e)

-129.7

-36.2

-40.2

h. Observed O2 flux i. Observed HS - flux

55.1

21.8 7.9

248


[45, 1

water in the cores was renewed to initiate the "4 hour" incubation, the gradients of TC02 at the sediment-water interface was increased dramatically. The TC02 production calculated from O2 uptake, assuming a TC02/02 flux ratio of 1, can provide an estimate of TC02 build-up in the overlying water over a 7 day period. The calculations are, however, based on the following assumptions; O2 uptake is not affected by the incubation method used and the flux of CO2 across the water-air interface is negligible. The last assumption is obviously false, but the equilibration of supersaturated sea water with atmosphere CO2 is known to be rather slow (Stumm and Morgan, 1970). Accordingly, TC02 in the overlying water after 7 days potentially would reach 9.5 and 4.9/Lmol ml-I for NOx- and Ox-cores, respectively, compared to a concentration of 1.8 /Lmol ml-1 in the added water. The erroneously high "4 hour" TC02 fluxes are, therefore, useless in budget calculations (Table 5, f, g). High TCOd02 flux ratios of 2-4 are found occasionally in sediments (Hargrave and Phillips, 1981; Andersen and Hargrave, 1984; Kepkay and Andersen, 1985), which has been attributed to anaerobic mineralization (i.e. sulfate reduction) with subsequent sulfide precipitation as FeS and FeS2 at depth. In sediments, where the sulfide is fully reoxidized by oxygen at the sediment-water interface, the TC02/02 flux ratio more frequently lies around 1 (Raine and Patching, 1980; Anderson et al., 1986). It was expected that there would have been a more reduced environment in the present sediments in the sequence NOx-, Ox- to An-cores. This was observed both by visual inspections and by the increase in HS- and NH/ pools (Table 4). The NOx-cores were relatively oxidized because ventilatory pumping of Nereis transported O2 deep into the sediment via the burrows (Kristensen, 1985). However, most of sediment in these cores still appeared to be anoxic at the end of experiment. The Ox-cores should not be particularly oxidized, since the access of O2 was limited by diffusion, and O2 should not have penetrated more than -3 mm below the surface (Revsbech et al., 1980). The surprisingly low rates of sulfate reduction measured at the end of experiment in both NOx- and Ox-cores (equivalent to 2-3% of the mean POC loss) might indicate that they were relatively less reduced than An-cores (Table 4). Sulfate reduction measured in the An-sediment was equivalent to 65% of the mean loss of POC. Sulfate reduction is generally found to be responsible for -50% of the overall and -90% of the anaerobic mineralization in coastal sediments (J~rgensen, 1983). The present sulfate reducing activity may be underestimated by failure to measure non-acid-volatile sulfur forms (SO and FeS2)' which is shown to be a more important route for S-- than the acid-volatile forms (H2S and FeS) in less reduced sediments (Howarth and J~rgensen, 1984). Another possible cause for the low activity in the NOx- and Ox-cores is, that during the sulfate reduction assay, sediment samples from these less reduced cores might have suffered more from oxygen-exposure, which was to some extent inevitable because of the very small sample size (3 cm3). Respiration by the nereids in NOx-cores was directly responsible for 40% of the total system O2 uptake (Kristensen, 1983). The animals, however, appeared to be starved, as observed by the relatively high loss of Nereis biomass (Table 3). The metabolism of

1987]


249

worm tissue accounted for -20% of the total loss of carbon from the NOx-cores. Consequentially, feeding was responsible for -20% of the POC loss, which is similar to that previously reported (Billen, 1978; Kristensen, 1985). The remaining stimulation of POC loss by the worms in NOx-cores, was caused by enhanced microbial activity in the sediment. Polychaete burrow structures are generally organic rich sites of high microbial activity (Aller and Yingst, 1985; Kristensen et al.. 1985). Heterotrophic activity is stimulated by grazing, mucus additions and facilitated transport of solutes in and out of the sediment. However, the exposure of sulfide and NH4 + to oxygen may also stimulate chemolithoautotrophic activity. Aller and Yingst (1978) have proposed that increased mineralization rates, due to infaunal animals, would mainly be caused by a stimulation of sulfate reduction in the sediment. This was apparently not the case in our experimental systems, although sulfate reduction was doubled in NOx-cores relative to Ox-cores (Tabie 4). The facilitated solute transport due to Nereis ventilation was clearly observed by a 2.5 and 3.5 fold increase in O2 and TC02 flux in NOx-cores relative to Ox-cores (Fig. 5, Table 5). Similarly, the profiles of pore water TC02 showed reduced levels, although higher than initially, in NOx-cores relative to the others (Fig. 2). The 50% higher rate of POC loss in An-cores relative to Ox-cores was reflected in the pools of TC02 (Table 4). This suggests that mineralization is faster in the absence of oxygen and nitrate reduction. Under reduced conditions, a close coupling between hydrolytic-fermentation processes and sulfate reduction may result in an extensive degradation of organic matter. Anderson et al. (1986), on the other hand, found a decreased mineralization rate, measured as TC02 production, when oxygen was depleted in an undisturbed sediment. Their benthic flux chambers, however, are difficult to compare with our experimental systems. The decrease in TC02 production when their flux chambers turned anoxic may very well be due to death of the abundant macrofauna that were present. Aerobic decomposition is known to be energetically more beneficial than anaerobic (e.g. Claypool and Kaplan, 1974). In order to maintain the same level of metabolic activity as aerobic organisms, sulfate reducers should be expected to oxidize a larger amount of organic carbon. During 02-mediated respiration, most of the free energy of the respired organic matter is available to the respiring organisms. During dissimilatory sulfate reduction, on the other hand, only about 25% of the energy is available to the organisms. About 75% of the free energy is conserved in reduced inorganic sulfur compounds (Howarth and Teal, 1980). The exchange of O2 and TC02 reflected the differences in POC loss between the three systems (Fig. 5). The temporal pattern may partially explain the apparent deficit in measured sulfate reduction. The high initial flux rates could have been due to preferential breakdown of the more labile fractions of organic matter, such as dead meiofauna. These organisms, which were killed by the freezing treatment, were composed ofvery easily decomposable organic matter (Westrich and Berner, 1984). At the end of experiment, when sulfate reduction was determined, only the more refractory components were left. This was seen in the decrease of O2 and TC02 flux to

250

% and


[45, 1

1/), respectively, of that initially observed. Such temporal flux patterns have been observed frequently in experimental systems (Kelly and Nixon, 1984; Kepkay and Andersen, 1985; Andersen and Kristensen, in prep). Sulfate reduction was probably limited by the availability of organic matter when measured finally. This can very well explain the 35% deficit in the An-cores, but probably not entirely that in the other cores. Nitrogen budgets for the three systems, expressed as daily means, are summarized in Table 6. Decay constants for PON, i.e. 0.553-0.861 yr-1 (Table 2), were 2-3 times higher than for POCo The inter-core pattern was similar, but the stimulation in NOxand An-cores relative to Ox-cores were less dramatic, i.e. 1.6 and 1.2 fold, respectively. Similarly, Aller and Yingst (1985) estimated that sediment net NH4 + production in the presence of a variety of infaunal animals was stimulated 1.1-1.8 fold relative to defaunated controls. The supply, by Nereis virens excretion, to the loss of nitrogen from the sediment, estimated from Kristensen (1984), was 30% of the total. The contribution of metabolized Nereis virens tissue, however, accounted for 25% of the total loss of nitrogen (Table 6, b), leaving only 5% originating from ingested detritus. This is less than previously reported for infaunal populations, i.e. 10-20% (Billen, 1978; Blackburn and Henriksen, 1983; Kristensen, 1985), suggesting that the worms were nitrogen limited in our system and therefore lost weight during the experiment. Rates of net NH4 + production measured at the end of the experiment (i.e. 3.07,4.50 and 3.68 mmol m-2d-1 for NOx-, Ox- and An-cores, respectively, Table 4) had some similarity to the values of mean PON loss per day (i.e. 7.56,4.68 and 5.45, respectively, Table 6, a). The latter, however, were higher particularly for NOx-cores. The decline in reactivity of the organic matter was clearly observed from the temporal pattern of NH4 + flux in the An-cores (Fig. 6C). The mean loss of nitrogen in An-cores was in agreement with the measured mean rates of NH4 + flux (Table 6, d, e). The dramatic temporal changes of NH/, N02 - and NO) - flux in NOx- and Ox-cores, where a decline in NH4 + flux was accompanied by an increased flux of N02 - and NO) -, indicated the growth of nitrifying bacteria. Nitrifiers, which are known to grow slowly (Kaplan, 1983), needed a 14 day lag phase before measura:ble nitrification occurred. A succession was observed with NH4 + oxidizers appearing first, followed immediately by N02 - oxidizers. This may be a consequence of substrate availability, NH4 + being abundant from start and N02 - appearing later when sufficient NH4 + has been oxidized. Potential rates of nitrification showed that only the two aerobic core systems exhibited any increased activity over the initial level, and this was primarily in the oxidized surface layers of the sediment (Fig. 9A). Nereis has a significant effect on nitrifying populations, since the activity in the NOx-cores was much higher than in the Ox-cores. According to the estimates from Table 6, nitrification was stimulated 2.3 fold by the presence of Nereis. Kristensen et al. (1985) has shown that the walls of nereid burrows are sites of very high potential nitrification activity. They estimated

1987]


251

Table 6. Budget for nitrogen in cores, expressed as mean rates (mmol m-2 d-') over a 94 day period. a. IlPON in sediment b. IlPON in Nereis c. IlDlN in pore water

Nereis -7.56 -2.61 -0.01

Aerobic -4.68

Anaerobic -5.45

0.78

1.13

d. Loss from sediment (a + b + c) e. Observed DIN flux

-10.18 -3.07

-3.90 -0.89

f. Denitrification (d - e) g. Nitrate red. (f x 1/0.61)* h. Obs. NO] - flux

-7.11 -11.66 -0.77

-3.01 -4.93 -0.51

i. Nitrification - (g + h)

12.43

-4.32 -3.78

5.44

*The nitrate reduction estimate is calculated from the denitrification rate, using the average ratio between the two processes found from the potential assays (denitr/nred = 0.61).

that burrow walls, despite a relatively low oxygen penetration into the wall linings, accounted for 10-70% of the in situ bulk sediment nitrification. The dramatic decrease in total DIN flux in NOx- and Ox-cores (Fig. 7), caused by a high rate of denitrification in the sediment with subsequent N210ss to the atmosphere, was synchronous with the appearance of measureable levels of N02and N03-. Despite the differences in DIN flux rates, the potential activity of nitrate reduction and denitrification appeared unaffected by the various treatments (Fig. 9B, C). The An-cores showed similar potential rates as the aerobic cores, confirming that a wide variety of heterotrophic microorganisms are capable of dissimilatory nitrate reduction under appropriate conditions (Knowles, 1982). The rates of total nitrate reduction, shown in Table 6, g, are calculated using the denitrification estimate (Table 6, f), assuming a ratio; denitrification/nitrate reduction = 0.61. This ratio, which is derived from the potential slurry assays integrated from 0 to 10 em, is expected to be representative for the intact cores. Similarly, reports on undisturbed sediments have revealed ratios of 0.27-0.57 (S~rensen, 1978a; Nishio et al., 1982). Despite a similarity in potential rates, the higher availability of NO] - in NOx-cores was responsible for a 2.4 fold stimulation in denitrification and nitrate reduction relative to Ox-cores (Table 6, f, g). This is in agreement with Kristensen et al. (1985), who found that the increased availability of N03 - in burrow walls, caused by nitrification and Nereis ventilatory O2 and NO] - transport, made these structures responsible for 50-80% of the bulk denitrification in coastal sediments. By use of appropriate conversion factors (J~rgensen and S~rensen, 1985) the rates of denitrification (x 5/4) and nitrate reduction (x 2) estimated in Table 6 were found to be responsible for mineralization of 18.0 and 7.6 mmol C m-2d-1 in NOx- and Ox-cores, respectively.

252

Journaf of Marine Research

[45, 1

This is equivalent to 40-44% of the total poc loss in the two systems. These values are about 10 times higher than are usually observed in coastal sediments (S~rensen et af., 1979; J~rgensen and S~rensen, 1985). The very high denitrification rates were probably a result of the high nitrification activity in these sediments, resulting in nitrate concentrations as high as 20-70 nmol ml-1 in the overlying water. The estimated actual nitrification was 5-10 times higher than that usually observed in situ (Henriksen et af., 1981). The ratios of exchangeable:porewater NH4 + (0.62-0.72) was unusually low in comparison to the value of 1.3 obtained by Mackin and Aller (1984) in a wide variety of sediment types. Possibly the artificiality of the present systems contributed to a change in the adsorption capacity of the sediment. The pools ofNH4 + showed a similar pattern to that of TC02, with very high pools in the defaunated Ox- and An-cores (Fig. 3). These high pools may be caused by the absence of perturbing meiofauna and macrofauna in the uppermost layers of the sediment, leaving molecular diffusion to be the sole transport process. The much lower level of NH4 + in NOx-cores was due to the improved transport of this compound out of the sediment via the ventilation current and due to removal by nitrification in the burrow walls. A model description of net detrital breakdown in the microcosms is illustrated in Figure 10. The idea is that a certain fraction (-43%) of the detritus is readily degradable at day 0 (Fig. lOB). The residual low or nondegradable fraction of similar C:N molar ratio should be composed of highly refractory compounds (Westrich and Berner, 1984). The mean C:N ratios of the detritus used in Ox- and An-cores (3.7 and 4.8) were very similar to those measured on day 94 (4.2 and 4.6). Accordingly, the net C:N ratio of the detritus that was degraded initially was very low (-4) in all cores, indicating that detritus of a very high nitrogen content was mineralized. However, the mean C:N ratio of the detritus that was used in the NOx-cores during the 94 day period was 5.9 (Table 4), and the ratio calculated from the measured net carbon and nitrogen mineralization rates at the end of experiment was 15.5. This could be explained by an initial rapid breakdown of low C:N detritus in the degradable fraction in these cores. When this was exhausted, after 40-60 days (Fig. lOA), the remaining readily degradable detritus had a higher C:N. The decline observed in the rate of TC02 and O2 flux occurred at this time (Fig. 5). Presumably not all the low C:N detritus had yet been mineralized in Ox- and An-cores finally (Fig. lOA). The C:N ratio of the detritus used in these cores should have approached that of NOx-cores, if the experiment was allowed to proceed beyond 94 days. Blackburn (1980) made similar observations for natural sediments, where newly sedimented low C:N material was mineralized at the surface, whereas high C:N material was mineralized in the lower (older) sediment layers. The very low C:N ratios in the organic material which had disappeared over the 94 days incubation, may partially be explained by the participation of bacterial biomass in the degradative process. Depending on the efficiency of carbon incorporation by heterotrophs (Linley and Newell, 1984) and the flux of reduced inorganic compounds

1987]

Kristensen & Blackburn:


253

18 16

A

14 12

CoN IN NEREIS SEDIMENT

d

;:::

«

10

« ...J

8

:l: z

6

n:: n:: a

u

-

, ANAEROBIC _,_,-

.-....... .--'"

.------

AEROBIC 2

0

10

20

40

30

50

60 DAYS

70

80

90

100

110

120

B DEGR __ I1l ...J

o

a a.

NERE IS - CORE

LOWDEGR N

DEGR LOW DEGR

Nr--. _ '---

-

43%

57%

C

C

-

i

z

••

22%

N IS USED

u

12%

C IS USED

u

:z

-

« C)

n:: a

C.N 11

CoN 11 DAY

0

MEAN CoN IN DETRITUS USED 5.9

CoN

CoN

16

11

DAY 94

Figure 10. (A) C:N ratios in detritus used in Nereis-, Aerobic- and Anaerobic-cores and bulk C:N ratio in Nereis-core sediment during the 94 day experimental period. Symbols indicate the measured values. Curves are drawn by hand according to the temporal pattern predicted. The dashed lines represent the expected C:N pattern beyond the experimental period. (B) Net changes in pools of organic carbon and nitrogen in Nereis-cores during the 94 day experimental period. The detritus was initially (day 0) composed of a readily degradable fraction (-43%) and a low degradable fraction (-57%), both with a molar C:N ratio of II. At day 94, 12 and 22% (C:N = 5.9) of the organic carbon and nitrogen had been used, which accounted for 28 and 51% of the degradable fractions. The final C:N ratio in the degradable detritus was 16, providing a bulk sediment ratio of 12. a variable quantity of the metabolized carbon will be incorporated into heterotrophic and chemolithoautotrophic bacterial biomass. Some evidence for this incorporation is the net gain of organic detritus at the oxic/anoxic interface in the two aerobic NOxand Ox-cores (Fig. I), since neither sedimentation nor photosynthesis occurred (Kepkay et at .. 1979; Howarth, 1984). The observed high potential activity of nitrifiers


254

[45, 1

in the top layers of these cores at the end of experiment confirms the contribution by chemolithoautotrophs (Fig. 9A). If the structural components of the bacterial cells, which have a high C:N ratio, tended to accumulate after the cells had died, an increase in the residual organic detritus is expected. This would result in an apparently low C:N in the organic detritus oxidized. This concept of a qualitative change in organic detritus with time, due to new material being created rather than to original material being altered, should be considered in degradative models (Westrich and Berner, 1984).

s.

Conclusions

From the present experiments it may be concluded that aerobic conditions in the overlying water do not increase the overall rate of organic degradation in the sediment, as seen in the higher rates of detrital decomposition in the An- relative to the Ox-cores. Perhaps this should not be so unexpected since a variety of anaerobic processes are known to be rapid when supplied with sufficient substrate, e.g. fermentative degradation of fodder in ruminants, beer production and lactic fermentation of milk. The high rate of decomposition commonly observed in the oxic part of sediments compared to deeper anoxic parts, should be due to higher degradability of newly settled organic matter at the sediment-water interface. The presence of an abundant infauna, e.g. Nereis spp., will increase the overall mineralization, nitrification and denitrification processes as well as solute transport in sediments by a factor of 2-3. The more rapid degradation of organic detritus in the NOx-sediment must be attributed to activities other than simple introduction of oxygen into the otherwise anoxic sediment. Some of these activities have been discussed. It is probable that most of the readily degradable low C:N detritus had disappeared from the Nereis-sediments after 94 days incubation. This was not the case for the Ox- and An-sediments. It is expected that, in due course, the degradation rate in all three sediment systems studied would tend to approach a low common level, dictated by the availability of degradable material. Acknowledgments. study.

We thank Tove Wiegers for her assistance in the analytical part of this

REFERENCES Aller, R. C. 1982. The effects of macrobenthos on chemical properties of marine sediment and overlying water, in. Animal-Sediment Relations, P. L. McCall and M. J. S. Tevesz, eds., Plenum, NY, 53-102. Aller, R. C. and J. Y. Yingst. 1978. Biogeochemistry of tube-dwellings: A study of the sedentary polychaete Amphitrite ornata (Leidy). J. Mar. Res., 36, 201-254. -1985. Effects of the marine deposit-feeders Heteromastus filiformis (Polychaeta), Macoma balthica (Bivalvia), and Tellina texana (Bivalvia) on averaged sedimentary solute transport, reaction rates, and microbial distributions. J. Mar. Res., 43. 615-645. Andersen, F. C/>.and B. T. Hargrave. 1984. Effects of Spartina detritus enrichment on aerobic/anaerobic benthic metabolism in an intertidal sediment. Mar. Ecol. Prog. Ser., 16. 161-171.

1987]

Kristensen

& Blackburn: Organic carbon & nitrogen in sediments

255

Anderson, L. G., P. O. J. Hall, A. Iverfeldt, M. M. Rutgers van der Loeff, B. Sundby and S. F. G. Westerlund. 1986. Benthic respiration measured by total carbonate production. Limnol. Oceanogr., 31.319-329. Armstrong, F. A. J., C. R. Stearns and J. D. H. Strickland. 1967. The measurement of upwelling and subsequent biological processes by means of the Technicon Auto-analyser and associated equipment. Deep-Sea Res., 14, 381-389. Billen, G. 1978. A budget of nitrogen recycling in North Sea sediments off the Belgian coast. Estuar. Coast. Mar. Sci., 7. 127-146. Blackburn, T. H. 1979. Methods for measuring rates of NHt turnover in anoxic marine sediments, using a 15N-NHt dilution technique. Appl. Environm. Microbiol., 37. 760-765. -1980. Seasonal variations in the rate of organic-N mineralization in anoxic marine sediments, in Biogeochimie de la Matiere Organique a L'interface Eau-Sediment Marin. Edition du CNRS, Paris, 173-183. Blackburn, T. H. and K. Henriksen. 1983. Nitrogen cycling in different types of sediments from Danish waters. Limnol. Oceahogr., 28, 477-493. Cammen, L. M. 1975. Accumulation rate and turnover time of carbon in a salt marsh sediment. Limnol. Oceanogr., 20. 1012-1014. Claypool, G. E. and I. R. Kaplan. 1974. The origin and distribution of methane in marine sediments, in Natural Gases in Marine Sediments, I. R. Kaplan, ed., Plenum, NY, 99-140. Cline, J. D. 1969. Spectrophotometric determination of hydrogen sulfide in natural waters. Limnol. Oceanogr., 14. 454-458. Dean, W. E., J r. 1974. Determination of carbonate and organic matter in calcareous sediments and sedimentary rocks by loss on ignition: Comparison with other methods. J. Sed. Petrol., 44, 242-248. Edmond, J. M. 1970. High precision determination of titration alkalinity and total carbon dioxide content of sea water by potentiometric titration. Deep-Sea Res., 17. 737-750. Fallon, R. D. and T. D. Brock. 1979. Decomposition of blue-green algal (cyanobacteria) blooms in Lake Mendota, Wisconsin. Appl. Environm. Microbiol., 37. 820-830. Fenchel, T. and T. H. Blackburn. 1979. Bacteria and Mineral Cycling, Academic Press, London. Foree, E. G. and P. L. McCarty. 1970. Anaerobic decomposition of algae. Environm. Sci. Technol., 4. 842-849. Hargrave, B. T. and G. A. Phillips. 1981. Annual in situ carbon dioxide and oxygen flux across a subtidal marine sediment. Estuar. Cstl. Shelf Sci., 12. 725-737. Henriksen, K., J. I. Hansen and T. H. Blackburn. 1981. Rates of nitrification, distribution of nitrifying bacteria, and nitrate fluxes in different types of sediment from Danish waters. Mar. BioI., 61,299-304. Hines, M. E. and G. E. Jones. 1985. Microbial biogeochemistry and bioturbation in the sediments of Great Bay, New Hampshire. Estuar. Cstl. Shelf Sci., 20, 729-742. Howarth, R. W. 1984. The ecological significance of sulfur in the energy dynamics of salt marsh and coastal marine sediments. Biogeochemistry, I, 5-27. Howarth, R. W. and B. B. J~rgensen. 1984. Formation of 35S-labelled elemental sulfur and pyrite in coastal marine sediments (Limfjorden and Kysing Fjord, Denmark) during short-term 35S04- reduction measurements. Geochim. Cosmochim. Acta, 48. 1807-1818. Howarth, R. W. and J. M. Teal. 1980. Energy flow in a salt marsh ecosystem: The role of reduced inorganic sulfur compounds. Am. Nat., 116. 862-872. J~rgensen, B. B. 1977. The sulfur cycle of a coastal marine sediment (Limfjorden, Denmark). Limnol. Oceanogr., 22.814--831. -1980. Seasonal oxygen depletion in the bottom waters of a Danish fjord and its effect on the benthic community. Oikos, 34. 68-76.

256 --


[45, 1

1983. Processes at the sediment-water interface, in The Major Biogeochemical Cycles and Their Interactions, B. Bolin and R. B. Cook, eds., John Wiley & Sons, Chichester, 477-509. J~rgensen, B. B. and J. S~rensen. 1985. Seasonal cycles of O2, NO] and SO~- reduction in estuarine sediments: The significance of an NO] reduction maximum in spring. Mar. Ecol. Prog. Ser., 24, 65-74. Kaplan, W. A. 1983. Nitrification, in Nitrogen in the Marine Environment, E. J. Carpenter and D. G. Capone, eds., Academic Press, NY, 139-191. Kelly, J. R. and S. W. Nixon. 1984. Experimental studies of the effect of organic deposition on the metabolism of a coastal marine bottom community. Mar. Ecot Prog. Ser., 17,157-169. Kepkay, P. E. and F. (/). Andersen. 1985. Aerobic and anaerobic metabolism of a sediment enriched with Spartina detritus. Mar. Ecol. Prog. Ser., 2/. 153-161. Kepkay, P. E., R. C. Cooke and J. A. Novitsky. 1979. Microbial autotrophy: A primary source of organic carbon in marine sediments. Science, 204, 68-69. Knowles, R. 1982. Denitrification. Microb. Rev., 46, 43-70. Kristensen, E. 1983. Ventilation and oxygen uptake by three species of Nereis (Annelida:Polychaeta). II. Effects of temperature and salinity changes. Mar. Ecol. Prog. Ser., /2, 299-306. -1984. Effect of natural concentrations on nutrient exchange between a polychaete burrow in estuarine sediment and the overlying water. J. Exp. Mar. BioI. Ecol., 75, 171-190. -1985. Oxygen and inorganic nitrogen exchange in a Nereis virens (Polychaeta) bioturbated sediment-water system. J. Cstl. Res., /, 109-116. -1986. Benthic fauna and biogeochemical processes in marine sediments: Microbial activities and fluxes, in Nitrogen Cycling in Coastal Marine Sediments, T. H. Blackburn and J. S~rensen, eds., John Wiley & Sons, Chichester, (in press). Kristensen, E., M. H. Jensen and T. K. Andersen. 1985. The impact of polychaete (Nereis virens Sars) burrows on nitrification and nitrate reduction in estuarine sediments. J. Exp. Mar. BioI. Ecol., 85, 75-91. Linley, E. A. S. and R. C. Newell. 1984. Estimates of bacterial growth yields based on plant detritus. Bull. Mar. Sci., 35, 409--425. Mackin, J. E. and R. C. Aller. 1984. Ammonium adsorption in marine sediments. Limnol. Oceanogr., 29,250---257. Muus, B. J. 1967. The fauna of Danish estuaries and lagoons. Distribution and ecology of dominating species in the shallow reaches of the mesohaline zone. Medd. Danm. Fisk. Havunders., 5. 1-316. Nishio, T., I. Koike and A. Hattori. 1982. Denitrification, nitrate reduction and oxygen consumption in coastal and estuarine sediments. Appl. Environm. Microbiol., 43, 648-653. Raine, R. C. T. and J. W. Patching. 1980. Aspects of carbon and nitrogen cycling in a shallow marine environment. J. Exp. Mar. BioI. Ecol., 47, 127-139. Revsbech, N. P., J. S~rensen, T. H. Blackburn and J. P. Lomholt. 1980. Distribution of oxygen in marine sediments measured with microelectrodes. Limnol. Oceanogr., 25, 403-411. Salonen, K. 1981. Rapid and precise determination of total inorganic carbon and some gases in aqueous solutions. Water Res., /5,403--406. Seliger, H. H., J. A. Boggs and W. H. Biggley. 1985. Catastrophic anoxia in the Chesapeake Bay in 1984. Science, 228, 70---73. Solorzano, L. 1969. Determinations of ammonia in natural waters by the phenolhypochlorite method. Limnol. Oceanogr., /4. 799-801. S~rensen, J. 1978a. Capacity for denitrification and reduction of nitrate to ammonia in a coastal marine sediment. Appl. Environm. Microbiol., 35,301-305. -1978b. Denitrification rates in a marine sediment as measured by the acetylene inhibition technique. Appl. Environm. Microbiol., 36, 139-143.

1987]

Kristensen & Blackburn:


257

Stirensen, J., B. B. Jtirgensen and N. P. Revsbech. 1979. A comparison of oxygen, nitrate and sulfate respiration in coastal marine sediments. Microb. EcoI., 5. 105-115. Stumm, W. and J. J. Morgan. 1970. Aquatic Chemistry, Wiley-Interscience, New York, 583 pp. Westrich, J. T. and R. A. Berner. 1984. The role of sedimentary organic matter in bacterial sulfate reduction: The G model tested. LimnoI. Oceanogr., 29. 236-249.

Received: 25 August. 1986: revised: 5 December, 1986.