Geo-Mar Lett (2013) 33:299–310 DOI 10.1007/s00367-013-0329-z

ORIGINAL

A method for the calculation of anaerobic oxidation of methane rates across regional scales: an example from the Belt Seas and The Sound (North Sea–Baltic Sea transition) José M. Mogollón & Andrew W. Dale & Jørn B. Jensen & Michael Schlüter & Pierre Regnier

Received: 28 December 2012 / Accepted: 1 May 2013 / Published online: 25 May 2013 # Springer-Verlag Berlin Heidelberg 2013

Abstract Estimating the amount of methane in the seafloor globally as well as the flux of methane from sediments toward the ocean–atmosphere system are important considerations in both geological and climate sciences. Nevertheless, global estimates of methane inventories and rates of methane production and consumption through anaerobic oxidation in marine sediments are very poorly constrained. Tools for regionally assessing methane formation and consumption rates would greatly increase our understanding of

J. M. Mogollón (*) Department of Geosciences, Utrecht University, P.O. Box 80.021, 3508 TA Utrecht, The Netherlands e-mail: [email protected] A. W. Dale Helmholtz-Zentrum für Ozeanforschung Kiel (GEOMAR), Wischhofstr. 1-3, 24148 Kiel, Germany J. B. Jensen Geological Survey of Denmark and Greenland (GEUS), Øster Voldgade 10, 1350 Copenhagen, Denmark J. M. Mogollón : M. Schlüter Alfred-Wegener-Institut Helmholtz-Zentrum für Polar- und Meeresforschung, Am Handelshafen 12, 27570 Bremerhaven, Germany P. Regnier Département des Sciences de la Terre et de l’Environnement, Université Libre de Bruxelles, 50 av. F.D. Roosevelt, 1050 Bruxelles, Belgium

the spatial heterogeneity of the methane cycle as well as help constrain the global methane budget. In this article, an algorithm for calculating methane consumption rates in the inner shelf is applied to the gas-rich sediments of the Belt Seas and The Sound (North Sea–Baltic Sea transition). It is based on the depth of free gas determined by hydroacoustic techniques and the local methane solubility concentration. Due to the continuous nature of shipboard hydroacoustic measurements, this algorithm captures spatial heterogeneities in methane fluxes better than geochemical analyses of point sources such as observational/sampling stations. The sensibility of the algorithm with respect to the resolution of the free gas depth measurements (2 m vs. 50 cm) is proven of minor importance (a discrepancy of <10%) for a small part of the study area. The algorithm-derived anaerobic methane oxidation rates compare well with previous measured and modeling studies. Finally, regional results reveal that contemporary anaerobic methane oxidation in worldwide inner-shelf sediments may be an order of magnitude lower (ca. 0.24 Tmol year–1) than previous estimates (4.6 Tmol year–1). These algorithms ultimately help improve regional estimates of anaerobic oxidation of methane rates.

Introduction Methane, a potent greenhouse gas, is pervasively present in marine sediments. It is produced (methanogenesis) principally by organic matter degradation below the sulfate reduction zone and consumed mostly through anaerobic oxidation (AOM), whereby sulfate typically acts as the electron

300

acceptor (Jørgensen and Kasten 2006). AOM takes place in the sulfate–methane transition zone (SMTZ) and is driven by downward-diffusing sulfate from the water column and upward-diffusing methane from the methanogenic zone. By dampening methane fluxes from the sediment toward the water column, AOM plays a major role in the global methane cycle (Dale et al. 2008a; Regnier et al. 2011). Although most methane in sediments is present in solid form as methane hydrate below the shelf break (Kvenvolden et al. 1993; Dickens 2011), continental shelf sediments also harbor a large methane pool in dissolved and free gas forms (Fleischer et al. 2001; Regnier et al. 2011). In these sediments, due to comparatively high organic matter inputs (e.g., Jørgensen and Kasten 2006; Krumins et al. 2013 and references therein) sulfate is typically consumed within the upper few meters of the sediment, leading to relatively high rates of methanogenesis below the SMTZ and gas formation (e.g., Hinrichs and Boetius 2002; Regnier et al. 2011). By implication, the methane reservoir on the shelf is generally located much closer to the sediment–water interface than is the case on the slope (Regnier et al. 2011), and is thus likely to be sensitive to environmental fluctuations in the overlying water column induced by, for example, climate change or eutrophication. Since free gas bubbles scatter sound waves that penetrate the sediment and cause acoustic blanking (Anderson et al. 1998), the free gas depth (FGD, i.e., the sediment depth where a prominent gas layer first appears) can be detected through shipboard seismo-acoustic techniques, and can thus be mapped at a regional scale (e.g., Laier and Jensen 2007). In this context, Dale et al. (2009) developed an algorithm relating the FGD to the methane flux toward the SMTZ (FCH4|SMTZ) using a combination of numerical modeling and geochemical data from 19 stations in Aarhus Bay (Denmark). This algorithm is applicable to diffusiondominated, inner-shelf gassy sediments outside the hydrate stability zone (Bohrmann and Torres 2006). Since in most continental shelf sediments upward-diffusing methane is almost completely consumed by AOM (Regnier et al. 2011), FCH4|SMTZ approximates the depth-integrated AOM rate (∑RAOM). This FGD-AOM algorithm can thus be translated into a method for quantifying AOM at a regional scale. Although not all inner shelf sediments contain free gas, they are important environments for AOM and, where gas is detectable, they provide an ideal location to apply the FGD-AOM method. Here, the predictive algorithm of Dale et al. (2009) is used to derive regional AOM rate estimates in two extensively surveyed areas located at the North Sea–Baltic Sea transition (Laier and Jensen 2007). In this general region, methane cycling has been documented by, for example, Hinrichs and Boetius (2002), Dale et al. (2008a, b), Pimenov et al. (2010), and Ulyanova et al. (2012). The

Geo-Mar Lett (2013) 33:299–310

algorithm is extended to account for a wider range of methane solubility in inner shelf sediments as a function of pressure (P), temperature (T) and salinity (S), and is combined with GIS methods to perform spatially resolved calculations over an area covering ca. 13,500 km2. The results presented here are the first estimates of regional AOM rates based on the application of observed methane gas distributions, and highlight the importance of using this type of approach to help refine the significance of AOM in the global methane cycle.

Regional setting The Belt Seas and The Sound connect the Baltic Sea to the North Sea (Fig. 1). Through these waterways, dense saline waters from the North Sea flow into the Baltic Sea, while overlying brackish waters originating from river runoff within the Baltic Sea flow out into the North Sea. Ultimately, this water exchange produces a bottom-water salinity (S) gradient that ranges from S=27 in the northern part of the study region to S=11 in the southern part (Fig. 2a). The water depth is <35 m (in large parts even shallower), equivalent to a maximum pressure of 4.7 bar, and the bottom-water mean annual temperature ranges from 280–283 K (Fig. 2b). Bottom sediment type varies from organic-poor sand to larger expanses of organic-rich mud (e.g., Hermansen and Jensen 2000; Leipe et al. 2011). The deposition of organic-rich material has steadily taken place over the past 10,000 years, when the latest marine transgression flooded the region, eventually leading to marine conditions ca. 8,000 years ago (Jensen and Bennike 2009). Due to the (1) overall shallow water depths, (2) commonly high organic carbon content, and (3) long time span facilitating deposition, methanogenesis is widespread in these sediments (Dale et al. 2008b; Mogollón et al. 2012). A free gas phase has been observed at many locations, generally 2 to 4 m below the sediment surface (Laier and Jensen 2007; Dale et al. 2008b). Aarhus Bay is situated in the Belt Seas (northwestern part of the study area, Fig. 1). This semi-enclosed bay of 26 m maximum depth has received considerable attention during recent geochemical (Dale et al. 2008b) and geological studies (Jensen and Bennike 2009). Based on seismo-acoustic surveys, these works revealed the presence of a 5.4 km2 gassy area with a FGD ranging between 0.25 and 8 m depth. Bottom-water temperature (ca. 283 K) and salinity (24.9– 25.0) vary little across Aarhus Bay.

Methods Regional AOM rates were estimated by multiplying calculated methane fluxes toward the SMTZ (FCH4|SMTZ) with

Geo-Mar Lett (2013) 33:299–310 Fig. 1 Bathymetry of the Belt Seas and The Sound (after Hermansen and Jensen 2000) along a North Sea–Baltic Sea continuum (inset). Red box Aarhus Bay

301

Denmark North Sea Baltic Sea

Aarhus Bay

Sweden

Study Area 56o0’N

The Sound Belt Seas

Water depth (m) 0-1 1-2 2-3 3-4 4-5

55o0’N

Germany 10o0’E

the seafloor surface area where free gas is present. In the original study of Dale et al. (2009), FCH4|SMTZ (nmol cm–2 day–1) was calculated using a power law relationship: F CH4 jSMTZ ¼ ca FGDcb

ð1Þ

where FGD is the free gas depth (cm), and ca , cb are parameters that depend on the methane solubility concentration at the FGD, C CH4*|FGD (Table 1). While this empirical relation successfully predicts measured methane fluxes, it is important to note that it does not provide specific details about the pathways of methane formation and consumption, nor about the fate of dissolved inorganic carbon and sulfide produced during AOM. It simply represents a tool for quantifying AOM rates in gassy sediments assuming that the methane profile is at or near steady-state conditions. Furthermore, the premise behind this algorithm is that, in general terms, methane dynamics are ultimately reflected in the FGD. In the present study, Eq. 1 was generalized for a continuous set of methane solubility concentrations by finding the best linear fit for ca and cb with respect to C CH4*|FGD according to the discrete values found in Dale et al. (2009; Table 1): ca ¼ 7:37⋅10



C CH4 FGD

þ 1:91⋅103

ð2Þ

cb ¼ 5:250⋅10−3 C CH4 FGD þ 1:014

ð3Þ

2

Several algorithms have been published for calculating the methane solubility concentration (C CH4*) at different T, P,

11o0’E

12o0’

5-6 6-7 7-8 8-9 9 - 10 10 - 11 11 - 12 12 - 13 13 - 14 14 - 15 15 - 16 16 - 17 17 - 18 18 - 19 19 - 20

20 - 21 21 - 22 22 - 23 23 - 24 24 - 25 25 - 26 26 - 27 27 - 28 28 - 29 29 - 30 30 - 31 31 - 32 32 - 33 33 - 34 34 - 35

13o0’E

and S. The third-order polynomial regression in Dale et al. (2008b) is accurate to within 3% of the Duan et al. (1992) model. The assessments of Mogollón et al. (2009, 2011) are accurate to within 1% of the thermodynamic calculations of Duan and Mao (2006). However, the P, T, S ranges of those field studies are quite narrow. Here, C CH4* is calculated over a range applicable to shallow freshwater settings grading into fully marine settings on the continental shelf (P=1.1–20 bar corresponding to a depth range of ca. 0–200 m, T=273–300 K, S=0–35), by fitting the solubility values of Duan and Mao (2006) through the following regression:   C ¼C ⋅eð−ASÞ ð4Þ CH4

CH4

S¼0

where

C CH4 S¼0



¼ aP2 þ bP þ c ⋅T 2 þ dP2 þ eP þ f ⋅T
þ gP2 þ hP þ i

and A ¼ BT −c ; B ¼ jP−k þ l; C ¼ mP−n þ o where A, B, and C are quasi-empirical constants, and values for a–o are listed in Table 2. The methane solubility concentrations was first estimated for freshwater conditions (S=0), which provided the constants a–i in the above equation. These constants were determined by assuming a quadratic fit with respect to temperature (Fig. 3a), in turn assumed to depend on pressure also through a quadratic equation (not shown). A salinity increase was observed to lead to

302

Geo-Mar Lett (2013) 33:299–310

Fig. 2 a Bottom-water salinity and b mean annual bottomwater temperatures in the study areas

0 5 10

Denmark

20

30

a

40 km

Sweden Aarhus Bay o

56 0’N

The Sound Belt Seas

Salinity (-) 24.0 - 25.0 23.0 - 24.0 22.0 - 23.0 21.0 - 22.0 20.0 - 21.0 19.0 - 20.0 18.0 - 19.0 17.0 - 18.0 16.0 - 17.0 15.0 - 16.0 14.0 - 15.0 13.0 - 14.0 12.0 - 13.0 11.0 - 12.0 < 11.0 13o0’E

> 30.0 29.0 - 30.0 28.0 - 29.0 27.0 - 28.0 26.0 - 27.0 25.0 - 26.0 55o0’N

Germany 10o0’E

11o0’E 0 5 10

Denmark

12o0’E 20

30

b

40 km

Sweden Aarhus Bay o

56 0’N

The Sound Belt Seas

Temperature (K) 282.90-283.15 282.65-282.90 282.40-282.65 282.15-282.40 281.90-282.15 281.65-281.90 281.40-281.65 281.15-281.40 280.90-281.15 280.65-280.90 280.40-280.65 280.15-280-40 < 280.15

o

55 0’N

Germany o

10 0’E

a quasi-linear C CH4* decrease with temperature at any given pressure value (e.g., Fig. 3b). The exponential constant A in Eq. 4 decreased with temperature through a power-law relationship (Fig. 3c). Likewise, the constants B, C in Eq. 4 depend on pressure through a power law (Fig. 3d). Solubility calculations using Eq. 4 compare favorably with the thermodynamic data of Duan and Mao (2006), with the error only exceeding 1.1% at temperatures greater than 297 K and pressures lower than

o

o

11 0’E

o

12 0’E

13 0’E

5 bar (Fig. 4). For the purpose of the regional extrapolations, the resulting C CH4* values were converted to molarity by Table 1 Values of ca and cb for different C CH4* (mM; from Dale et al. 2009)

C CH4* (mM)

ca

cb

5 7 9

5,516 7,216 8,463

1.038 1.056 1.059

Geo-Mar Lett (2013) 33:299–310

Parameter

Value

a b c d e

−4.372018 • 10–9 8.278747 • 10–7 −1.478806 • 10–8 2.699539 • 10–6 −5.148083 • 10–4

f g h i j k l m n o

6.761301 • 10–6 −4.205982 • 10–4 8.132632 • 10–2 −7.307755 • 10–4 2.269174 • 103 7.794562 • 10–1 2.611000 • 103 1.186056 • 10–1 6.026738 • 10–1 2.314900 • 100

multiplying the obtained values with the in situ pore water density, which was calculated based on the in situ T and S using standard equations (McCutcheon et al. 1993). The resulting relation yields a nomogram (Fig. 5) analogous to the one presented in Dale et al. (2009), whereby FCH4|SMTZ increases with increasing C CH4*|FGD (a function of P, T, S) and with decreasing FGD. The methane flux can thus be calculated for any pair of C CH4*|FGD and FGD values.

2.4

a

4.0

CH4*|S=0 (mol/kg) . 10-2

CH4*|P = 10 bar (mol/kg) . 10-2

4.5

P = 20 bar

3.5 3.0

P = 15 bar

2.5 2.0

P = 10 bar

1.5 1.0

P = 5 bar

0.5

P = 1.2 bar

0.0

275

280

285

290

295

T = 274 K

2.0 1.8

T = 281 K

1.6

T = 285 K

1.4

T = 292 K T = 300 K

1.2 1.0

300

b

2.2

0

5

10

15

20

25

30

35

Salinity (-)

Temperature (K) 5.8

2.33 4.6

5.7

c

5.6

A (-) . 10-3

2.34 2.35

5.5

4.2

5.4 5.3 5.2

P = 20 bar

5.1 5.0

2.36 4.0 -0.6026738

C = 0.1186056 P

3.8

2.37 2.38

B = 2269.174 P

3.4

P = 1.2 bar

+ 2.3149

3.6 -0.7794562

+ 2611.0

2.39 2.4

4.9

3.2

4.8

3.0

2.41 2.42

4.7 4.6

d

4.4

2.8 275

280

285

290

Temperature (K)

295

300

0

5

10

Pressure (bar)

15

2.43 20

C (-)

Fig. 3 Methane solubility calculations for ranges of pressure (P), salinity (S), and temperature (T) for shelf sediments: open symbols thermodynamically calculated values (Duan and Mao 2006). a Comparison of methane solubility concentrations at S=0 and variable P, T using Duan and Mao (2006, symbols) and Eq. 3 (lines). b Comparison of methane solubility concentrations at P=10 bar and variable S, T for Duan and Mao (2006, symbols) and Eq. 3 (lines). c Variable A (Eq. 3) dependency on P, T. Solid symbols Extracted best fits (based on error minimizing regressions), lines calculated approximations. d Variables B and C as a function of P. Solid symbols Extracted best fits (based on error minimizing regressions), lines power function approximation of these fits. Note that the C axis has been inverted for clarity

Spatially resolved C CH4*|FGD were calculated from the T, P, and S maps presented in Figs. 1 and 2, and used in conjunction with FGD maps to regionally calculate FCH4 |SMTZ over the Belt Seas and The Sound. For pressure calculations, bathymetry data (Fig. 1) were obtained from the Leibniz Institute for Baltic Sea Research in Warnemünde (http://www.io-warnemuende.de/topography-of-the-balticsea.html; T. Seifert et al., unpublished data, 2001 Baltic Sea Science Congress, Stockholm, poster no. 147). Bottomwater salinity data for the year 2006 were downloaded from the HELCOM website, which hosts information from the B A L A N C E p r o j e c t ( h t t p : / / w w w. h e l c o m . f i / G I S / BalanceData/en\_GB/main/). A yearly averaged bottomwater temperature map was constructed based on historical bottom-water temperature data from ICES (http:// geo.ices.dk/; Fig. 2). Only a low-resolution FGD map (2-m intervals) is available for the Belt Seas and The Sound. Laier and Jensen (2007) generated this FGD map by compiling a large array of archived seismic data at different frequencies from various sources. In Aarhus Bay, additional surveys using a 0.4– 10 kHz X-star full spectrum sonar, a 1–10 kHz chirp, a 0.6– 2 kHz boomer, and a 800–1,200 Hz channel sparker (Jensen and Bennike 2009; Dale et al. 2009) provided a local map where FGDs were binned into 50-cm sediment depth intervals (higher resolution). The sensitivity of the calculations with respect to FGD resolution was thus tested in

B (-) . 103

Table 2 Values of a–o constants used for the C CH4* (mol kg–1) calculation (Eq. 3) under given P (in bar), T (in K) and S (dimensionless) conditions

303

304

Geo-Mar Lett (2013) 33:299–310

Temperature (K) 1

275

280

285

290

295

300 Error (%)

3

Pressure (bar)

5

1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1

7 9 11 13 15 17 19

Fig. 4 Percentage relative error for values of C CH4* calculated using Eq. 3 versus the algorithm of Duan and Mao (2006)

Aarhus Bay by comparing the results of the low- and highresolution FGD maps. In the study region, sediments with mud (<63 μm) contents exceeding 10% have been found to contain the highest amount of organic matter (Leipe et al. 2011). Thus, for interpretation purposes, the surficial distribution of sediments likely to host significant amounts of methane was delineated by combining all sediments with mud contents exceeding 10% (cf. the mud, sandy mud, and muddy sand areas depicted in Hermansen and Jensen 2000), referred to in this paper as potentially methanogenic sediments or areas. All surface areas were calculated using the projection WGS1984 UTM32N.

Results The FCH4|SMTZ distribution calculated over the 5.4 km2 gassy area in Aarhus Bay is shown in Fig. 6 for both the 2-m (low resolution) and 50-cm (high resolution) FGD intervals. Note that in Aarhus Bay the methane solubility Solubility concentration (mM) Free gas depth (FGD) (cm)

200 300 400 500 600 700 800

5

6

7

8

9

10 11 12 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0

Methane flux to SMTZ (nmol cm-2 day-1)

100

4

Fig. 5 Methane fluxes to the SMTZ based on the local methane solubility concentration and the free gas depth as predicted by Eqs. 1 and 2

variations mostly reflect the change in water depth, which ranges from 18 to 26 m. Since it is assumed that FGD and ∑RAOM are equal, Fig. 6 provides also the spatial distribution of depth-integrated AOM rates. The high-resolution FGD map predicts a regional AOM rate of 0.496 and 0.779 Mmol year–1, depending on whether the deeper or shallower FGD values (respectively) are used for each 50cm depth interval (Fig. 6). By contrast, regional AOM rates for the lowresolution FGD map span a significantly larger range, between 0.356 Mmol year –1 (for the deeper FGD values) and 1.32 Mmol year–1 (for the shallower FGD values). Yet, the calculated AOM rates deviate by only ca. 8% (0.599 Mmol year–1 for the high-resolution map, and 0.553 Mmol year–1 for the low-resolution map) if the FGD is specified at the average depth of each bin. Therefore, while the reduction in the range of predicted AOM rates favors the application of more detailed FGD maps, the results also show that the low-resolution map provides reasonable average values. In the Belt Seas and The Sound, a large shallow seismo-acoustic database was compiled by Laier and Jensen (2007) to produce a regional 2-m interval FGD map (not shown). The studied area of the Belt Seas extends over 11,997 km2, where surficial, potentially methanogenic sediments cover 43% (5,113 km2, Table 3) and gas covers 12% (1,398 km 2, Table 4). In The Sound, the study area extends over 1,499 km2, with the potentially methanogenic and gassy sediments covering 34% (509 km2, Table 3) and 14%, respectively (204 km2, Table 4). A small percentage of gassy areas coincide with surficial sandy sediments (<10% mud contents; areal coverage of 5% for the Belt Seas and 2% for The Sound). The likelihood of gas in sandy layers is small, due to their low organic matter content and limited ability to trap gas bubbles (Jensen et al. 2002). Although this gas could potentially originate in deeper organic-rich layers that are surficially overlain by sandy layers, the most likely reason for the apparent presence of gas in these sediments may be digitizing and/or mapping errors. Figure 7 shows the areal distribution of FCH4|SMTZ for the entire study area. While the fjords located along the western part of the Belt Seas are generally gas- and mud-rich, the FCH4|SMTZ at these locations is low (<10 nmol cm–2 day–1) due to the shallow water depths and, consequently, low C CH4*|FGD. By contrast, at greater water depths in the Belt Seas and The Sound, extensive gas bodies are present where the FCH4|SMTZ typically exceeds 15 nmol cm–2 day–1. When calculated over the entire area, the results indicate that a total of 218 and 31.5 Mmol year–1 are consumed by AOM in the gassy sediments of the Belt Seas and The Sound, respectively (Table 5).

Geo-Mar Lett (2013) 33:299–310 Fig. 6 Maps of the free gas depth (FGD, a, c) and methane fluxes to the SMTZ (b, d) at both low resolution (2-m interval, a, b) and high resolution (50-cm interval, c, d) in Aarhus Bay. Solid black lines Bathymetry in m; for location of study area, see red box in Fig. 2. Note the separate scales for FGD in a and c

305 o

o

10 24’E

o

10 26’E

o

10 28’E

10 30’E

0.2-2.0 2.0-4.0 >4.0

o

56 7’N

o

o

10 24’E

o

10 26’E

10 28’E

a

Free gas depth (FGD, m)

o

10 30’E

b o

56 7’N

56o6’N

o

56 6’N

18

20

22

24

18

26

56o5’N

20

22

24

26

o

56 5’N

0

1

2

0

km

1

2

km

d

c o

o

56 7’N

56 7’N

56o6’N

56o6’N

18

20

22

24

26

18

o

20

22

24

26

56o5’N

56 5’N

0 o

10 24’E

1

2

0

km

o

10 26’E

o

o

10 28’E

10 30’E

Free gas depth (FGD, m) 0.2-0.5 0.5-1.0 1.0-1.5 1.5-2.0 2.0-2.5

2.5-3.0 3.0-3.5 3.5-4.0 4.0-4.5 4.5-5.0

Discussion Factors controlling methane flux Within gassy sediments, the FGD ultimately provides the major control over the methane and sulfate fluxes toward the SMTZ (Fig. 8). As the FGD shoals, the distance between the FGD and the sediment–water interface becomes smaller, leading to larger methane and sulfate fluxes to the SMTZ (Fig. 8). The response is not linear, since the fluxes will increase exponentially as the FGD approaches the interface. With respect to C CH4*|FGD, gassy sites located at deeper water depths require higher dissolved methane concentrations in order to reach saturation. Consequently, for locations with the same FGD but different water depths, the deeper sediments (higher C CH4*|FGD) will be characterized by steeper geochemical gradients and, therefore, a higher methane flux (Fig. 8). The predicted FCH4|SMTZ based on Eqs. 2 and 3 for the expected range of temperatures and pressures in the Belt Seas follows this expected pattern

5.0-5.5 5.5-6.0 6.0-6.5 6.5-7.0 7.0-7.5

o

10 24’E

1

2 o

10 26’E

km 10o28’E

10o30’E

FCH4|SMTZ using mid-point FGD (nmol cm-2 day-1) 7.5-8.0 8.0-8.5 8.5-9.0 9.0-9.5 >9.5

0-5 5 - 10 10 - 12 12 - 14 14 - 16

16 - 18 18 - 20 20 - 22 22 - 24 24 - 26

26 - 28 28 - 30 30 - 35 35 - 40 40 - 50

50 - 60 60 - 80 80 - 100 100 - 120 120 - 150

(Fig. 7). The effects of FGD variations on the local C CH4*|FGD are negligible in comparison to the effects on FCH4|SMTZ. For example, an increase in the FGD from 2 m to 4 m over a water depth of 20 m would change the Table 3 Areal distribution of potentially methanogenic sediments (mud (<63 μm) content >10%) in the study areas according to the binned water depth. The percentage of potentially methanogenic sediments with respect to total binned area is given in parentheses Water depth (m)

Belt Seas (km2)

The Sound (km2)

0–5 5–10 10–15 15–20 20–25 25–30 30–35 Total

375 (14%) 948 (34%) 1,305 (52%) 1,129 (55%) 952 (70%) 309 (74%) 95.4 (87%) 5,113 (43%)

2.88 (0.78%) 46.6 (11%) 168 (54%) 106 (60%) 178 (91%) 8.51 (95%) 0 (−) 509 (34%)

306 Table 4 Distribution of gassy sediments within the study areas according to water depth. Percentages in parentheses represent the areal coverage of gassy sediments with respect to potentially methanogenic areas (non-italics) and total binned areas (italics)

Geo-Mar Lett (2013) 33:299–310

Water depth (m)

Belt Seas (km2)

The Sound (km2)

0–5 5–10 10–15 15–20 20–25 25–30 30–35 Total

140 (37%, 5%) 228 (24%, 8%) 458 (35%, 18%) 319 (28%, 15%) 207 (22%, 15%) 45.5 (15%, 11%) 0.0166 (0.002%, 0.001%) 1,398 (27%, 12%)

0.0234 (0.81%, 0.0063%) 10.9 (23%, 2%) 96.5 (57%, 31%) 49.3 (47%, 28%) 43.2 (24%, 22%) 3.65 (43%, 41%) 0 (−) 204 (40%, 14%)

C CH4*|FGD from 5.23 mM to 5.54 mM (for T=281 K and S=20). Concomitantly, this FGD increase would decrease F CH 4 | SM TZ much more significantly (from 2.31 to 1.12 nmol cm–2 day–1) due to the decrease in the methane and sulfate gradients, even when accounting for the minor increase in C CH4*|FGD. Mud and gas distribution In the study region, the degradation of organic matter in sediments deposited during the Holocene represents the shallow source for both dissolved and gaseous methane (Jensen et al. 1999; Laier and Jensen 2007; Jensen and Bennike 2009; Mogollón et al. 2012). Table 3 shows that the fraction of the seafloor covered by potentially methanogenic deposits increases with water depth. Although this trend is consistent at all the study locations, patterns of fine-grained sediment Fig. 7 Map of methane fluxes to the SMTZ for the Belt Seas and The Sound study areas (light green). Gray Locations of potentially methanogenic sediments. Note that high fluxes (purple) are essentially restricted to Aarhus Bay

Denmark

deposition may be more complex and governed by variations in the direction and intensity of bottom-water currents caused by interactions with morphological features such as sills. In The Sound, for example, all of the mud and gas accumulation occurs on the northern side of Saltholm Island largely due to the barrier that the island creates for material migrating with southbound bottom-water currents. The areal extent of gassy sediments is restricted to less than half of the total potentially methanogenic areas (Table 4, Fig. 7), generally where these sediments exceed 4–5 m in thickness (Jensen and Bennike 2009). These results are consistent with previous studies showing that mud thickness is the most important parameter controlling the occurrence of free gas in the Baltic Sea (Moros et al. 2002; Thießen et al. 2006; Mogollón et al. 2012). In sediments with thicker Holocene mud sequences, methanogenesis can lead to dissolved methane accumulations that surpass the local C CH4* and form a free gas phase. 0 5 10

20

30

40 Kilometers

Sweden Aarhus Bay 56o0’N

Saltholm The Sound Potentially methanogenic areas

Study areas

55o0’N

Germany 10o0’E

> 30

18-20

28-30

16-18

26-28

14-16

24-26

12-14

22-24

10-12

20-22

5-10 <5

Belt Seas 11o0’E

Methane flux to SMTZ (nmol cm-1 day-2)

12o0’E

13o0’E

Geo-Mar Lett (2013) 33:299–310

307

Table 5 Areal AOM rates in the different study areas according to water depth using the midpoint value of the FGD interval. Values in parenthesis were calculated assuming the end-point values of the FGD interval

Water depth (m)

Belt Seas The Sound (Mmol year–1), mid (shallow–deep)

Total

0–5 5–10 10–15 15–20 20–25 25–30 30–35 Total

7.58 (11.3–5.77) 19.9 (29.7–15.1) 64.0 (95.7–48.4) 72.9 (10.9–54.8) 48.3 (73.5–36.1) 5.46 (8.24–4.08) 0.002 (0.003–0.002) 218 (328–164)

7.58 (11.3–5.78) 20.8 (31.0–15.7) 75.3 (113–56.9) 83.0 (125–62.4) 57.1 (87.0–42.8) 5.89 (8.89–4.41) 0.002 (0.003–0.002) 250 (373–188)

0.002 (0.002–0.001) 0.865 (1.29–0.656) 11.3 (16.8–8.50) 10.1 (15.2–7.62) 8.86 (13.4–6.65) 0.432 (0.653–0.323) 0 (0–0) 31.5 (47.3–23.7)

and 24.1%, 48.3%, 27.6% for potentially methanogenic sediments, in both cases along an increasing water depth gradient). This tendency indicates that potentially methanogenic sediment deposition is favored at water depths >10 m. Table 3 further shows that potentially methanogenic sediments cover more than half of the area located at water depths >10 m. Unlike potentially methanogenic sediments (Table 3), the percent coverage of gassy sediments does not correlate with water depth (Table 4). The prevalence of potentially

For the Belt Seas and The Sound combined, shallower water depths prevail and deeper waters are earmarked by a progressively smaller areal coverage: 47.1% of the combined study area has <10 m water depth, 37.5% has up to 10–20 m water depth, and 15.4% is deeper than 20 m. By contrast, the corresponding combined data for gassy sediments and potentially methanogenic sediments reveal distinct peaks in areal coverage for intermediate water depths of 10–20 m, and lower values at both shallower and deeper water depths (24.0%, 57.4%, 18.6% for gassy sediments,

800m

Concentration

a

b

Sediment

d

* CCH 4

FGD

30m

c

b

Sediment Depth

SO4

Water column

CH4 Acoustic Turbidity

c

d

CH4 Acoustic Turbidity

* CCH 4

CH4 Acoustic Turbidity

Fig. 8 a Chirp profile of Aarhus Bay gassy sediments (after Laier and Jensen 2007). Light gray Acoustic blanking. b–e Schematic representation of sulfate (green), dissolved methane (red), and methane solubility (blue) profiles for different FGD and C CH4* regimes: b deep FGD, deep water depth; c shallow FGD, moderate water depth; d shallow FGD, shallow water depth; e deep FGD, moderate water depth (theoretical shallower site with the same FGD as site b and thus not

SO4

CH4

*

CCH4

Sediment Depth

* CCH 4

SO4

e Sediment Depth

SO4

Concentration

FGD

Concentration

FGD

Concentration

Sediment Depth

FGD

Acoustic Turbidity

Acoustic Turbidity

shown on chirp profile). Light blue Water column, yellow sediment, gray acoustic blanking due to the presence of methane gas, open arrows changing magnitude of methane flux to SMTZ. The shallower FGD leads to an increase in the methane gradient and thus a higher diffusive methane flux to the SMTZ. Likewise, a higher C CH4*|FGD translates into a higher methane concentration at the FGD and thus a higher methane flux to the SMTZ

308

Regional AOM budgeting

10-1

ΣRAOM (mol m-2 yr-1)

methanogenic sediments favors gas formation at deeper water depths, but the higher solubility (due to the increased pressure) requires more methane production to form free methane gas. Thus, while thick Holocene mud layers are present at deeper water depths leading to high rates of methanogenesis, this may not be sufficient to overcome the increase in methane solubility. Consequently, the majority of gassy regions (57.4%) are located at water depths of 10–20 m leading to the highest areal AOM rates within this depth interval (Table 5).

Geo-Mar Lett (2013) 33:299–310

100

10-1

10-2

e ur er at Lit

e

So

un d (e B xc el l. A t S ar ea hu s sB ay ) Aa rh us Ba y

10-3

Th

The regions of gas occurrence in the study areas represent 12% and 14% of the total area in the Belt Seas and The Sound, respectively (Table 4). While only a portion of the area covered by potentially methanogenic sediments contains free gas, non-gassy sediments are either devoid of methane or produce methane fluxes to the SMTZ which are significantly smaller than gassy locations within a given basin (Regnier et al. 2011; Mogollón et al. 2012). It is thus reasonable to assume that an estimate for the regional AOM rates in the gassy areas represents a realistic lower estimate of the bulk methane oxidation over the entire study area. The predicted methane fluxes toward the SMTZ for the Belt Seas and The Sound compare well with previously compiled ∑RAOM rates for other passive shelf systems worldwide (Regnier et al. 2011), and also with previously reported measured ∑RAOM rates (Hinrichs and Boetius 2002) in the inner shelf (Fig. 9). Integrating over the entire region, the FGD algorithm leads to a regional AOM rate of 250 Mmol year–1. This calculation ignores non-gassy, potentially methanogenic areas. The presence of methane is unknown there and would require numerous local coring, geochemical analyses, and data interpolation to assess the regional AOM rates. Nevertheless, a regional AOM rate range of 0–640 Mmol year–1 can be conjectured based on the assumption that AOM is absent because non-gassy sediments are devoid of methane, and that the average ∑RAOM in non-gassy areas is equivalent to the average ∑RAOM in the gassy areas, respectively. In reality, it is highly likely that the value falls somewhere in between these two extreme end-members. If we assume that the AOM rates in these areas are similar to AOM rates estimated for the sediments with the deepest FGD in the regions, the spatially integrated AOM rate would represent about a third (79 Mmol year–1) of the regional AOM rates calculated in the gassy regions. Few studies have attempted to quantify AOM rates at the regional scale, and these have mostly relied on extrapolating point source data (e.g., Hinrichs and Boetius 2002; Wallmann et al. 2006). For instance, the rates calculated in this study represent about 30 times the amount of methane

Fig. 9 Ranges of depth-integrated AOM rates (∑RAOM) obtained from 2-m interval FGD maps for the present study areas, compared to literature values for other passive shelf areas extracted from Regnier et al. (2011, symbols) and Hinrichs and Boetius (2002, bar)

consumed yearly at large submarine volcanoes such as the Dvurechenskii mud volcano in the Black Sea (Wallmann et al. 2006), a location considered to be a major laboratory for methane cycling (Bohrmann et al. 2003). The study area is thus a region of significant methane production and consumption. For diffusive, inner-shelf systems, a global methane consumption rate of 4.6 Tmol year–1 has been proposed based on ∑RAOM rates compiled at five sites, four of which are near or within the Belt Seas (Hinrichs and Boetius 2002). The results of the present study show that this value might be overestimated mainly because the average ∑RAOM of these five sites was assumed representative of global innershelf sediments. By contrast, extrapolating the results in this study to an inner shelf area of 13×106 km2 (Hinrichs and Boetius 2002), global AOM consumption rates in inner shelf sediments would reach 0.24 Tmol year–1, or an order of magnitude lower. This value could be somewhat higher if the non-gassy methanogenic sediments would be included in the calculation but would most likely remain under 0.5 Tmol year–1. This estimate reflects the heterogeneity of methane fluxes in gassy sediments and the presence of large areas on the shelf devoid of significant amounts of organic matter (e.g., non-accumulating relict sands cover roughly 70% of the worldwide continental shelf; Emery 1968; Krumins et al. 2013). In the future, further regional FGD

Geo-Mar Lett (2013) 33:299–310

studies in other shallow depositional settings could expand the range of the FGD algorithm and help better constrain global AOM rates in continental shelf sediments.

Conclusions Seismo-acoustic techniques represent a cost-effective way of surveying large (10–10,000 km2) areas not only from a geophysical perspective but also, as shown in this study, to provide a means for estimating geochemical fluxes on a regional basis. The FGD represents a potential geophysical indicator for the globally relevant AOM process in diffusive inner-shelf sediments. This methodology can account for spatial heterogeneities which cannot be addressed through point-source extrapolation of rates. In this study, regional AOM rates of 250 Mmol year–1 were estimated for the Belt Seas and The Sound, the inlets of which connect the North Sea to the Baltic Sea. The largest fluxes occur in gassy areas with deep bathymetries due to the higher methane solubility concentrations. Although the corresponding FGD maps relied on a 2-m interval resolution, AOM rates derived from a 50-cm and a 2-m FGD proved similar for a smaller sector comprising Aarhus Bay. Results here suggest that some previous studies may overestimate global AOM in inner shelves. This stresses the importance of developing global FGD maps not just in the inner shelf but also in the outer shelf and upper slope which could help further calibrate the FGD-AOM algorithm and extend its global AOM rate estimates. Moreover, new techniques which can better capture regional methane fluxes in locations where no free gas is detected need to be developed in order to better estimate rate heterogeneities when attempting global methane budget calculations. Acknowledgements We wish to thank Kerstin Jerosch for her help with the grain size classification system. Gerald Dickens, an anonymous reviewer and the editor provided constructive comments that improved the paper. This study was funded by NWO Vidi Award #864.05.007: Marine methane flux and climate change: from biosphere to geosphere, by the European Community’s Seventh Framework Programme (FP/2007–2013) under grant agreement number 217246 made with the joint Baltic Sea research and development programme BONUS, and by the government of the Brussels-Capital Region (Brains Back to Brussels award to P. Regnier). Further funding is acknowledged from the European Union’s Seventh Framework Program (FP7/2007-2013) under grant agreement no 283080, project GEOCARBON.

References Anderson AL, Abegg F, Hawkins JA, Duncan ME, Lyons AP (1998) Bubble populations and acoustic interaction with the gassy floor of Eckernförde Bay. Cont Shelf Res 18:1807–1838. doi:10.1016/ S0278-4343(98)00059-4

309 Bohrmann G, Torres ME (2006) Gas hydrates in marine sediments. In: Schulz HD, Zabel M (eds) Marine geochemistry, 2nd edn. Springer, Berlin, pp 481–512 Bohrmann G, Ivanov M, Foucher J-P, Spiess V, Bialas J, Greinert J, Weinrebe W, Abegg F, Aloisi G, Artemov Y, Blinova V, Drews M, Heidersdorf F, Krabbenhoft A, Klaucke I, Krastel S, Leder T, Polikarpov I, Saburova M, Schmale O, Seifert R, Volkonskaya A, Zillmer M (2003) Mud volcanoes and gas hydrates in the Black Sea: new data from Dvurechenskii and Odessa mud volcanoes. Geo-Mar Lett 23(3/4):239–249. doi:10.1007/s00367-003-0157-7 Dale AW, Van Cappellen P, Aguilera DR, Regnier P (2008a) Methane efflux from marine sediments in passive and active margins: estimations from bioenergetic reaction-transport simulations. Earth Planet Sci Lett 265:329–344. doi:10.1016/j.epsl.2007.09.026 Dale AW, Aguilera DR, Regnier P, Fossing H, Knab NJ, Jørgensen BB (2008b) Seasonal dynamics of the depth and rate of anaerobic oxidation of methane in Aarhus Bay (Denmark) sediments. J Mar Res 66:127–155 Dale AW, Regnier P, Van Cappellen P, Fossing H, Jensen JB, Jørgensen BB (2009) Remote quantification of methane fluxes in gassy marine sediments through seismic survey. Geology 37(3):235– 238. doi:10.1130/G25323A.1 Dickens GR (2011) Down the Rabbit Hole: toward appropriate discussion of methane release from gas hydrate systems during the Paleocene-Eocene thermal maximum and other past hyperthermal events. Clim Past 7:831–846. doi:10.5194/cp-7-831-2011 Duan Z, Mao S (2006) A thermodynamic model for calculating methane solubility, density and gas phase composition of methanebearing aqueous fluids from 273 to 523 K and from 1 to 2000 bar. Geochim Cosmochim Acta 70(13):3369–3386. doi:10.1016/ j.gca.2006.03.018 Duan Z, Møller N, Greenberg J, Weare JH (1992) The prediction of methane solubility in natural waters to high ionic strength from 0 to 250°C and from 0 to 1600 bar. Geochim Cosmochim Acta 56(4):1451–1460 Emery KO (1968) Relict sediments on continental shelves of world. AAPG Bull 52:445–464 Fleischer P, Orsi TH, Richardson MD, Anderson AL (2001) Distribution of free gas in marine sediments: a global overview. Geo-Mar Lett 21(2):103–122. doi:10.1007/s003670100072 Hermansen B, Jensen J (2000) Digital sea bottom sediment map around Denmark. Geol Soc Denmark and Greenland 68 Hinrichs K, Boetius A (2002) The anaerobic oxidation of methane: new insights in microbial ecology and biogeochemistry. In: Wefer G, Billet D, Hebbeln D, Jørgensen BB, Schlüter M, Van Weering TCE (eds) Ocean margin systems. Springer, Berlin, pp 457–477 Jensen JB, Bennike O (2009) Geological setting as background for methane distribution in Holocene mud deposits, Århus Bay, Denmark. Cont Shelf Res 29(5/6):775–784. doi:10.1016/ j.csr.2008.08.007 Jensen JB, Bennike O, Witkowski A, Lemke W, Kuijpers A (1999) Early Holocene history of the southwestern Baltic Sea: the Ancylus Lake stage. Boreas 28(4):437–453. doi:10.1111/j.15023885.1999.tb00233.x Jensen JB, Kuijpers A, Bennike O, Laier T, Werner F (2002) New geological aspects for freshwater seepage and formation in Eckernförde Bay, western Baltic. Cont Shelf Res 22(15):2159– 2173. doi:10.1016/S0278-4343(02)00076-6 Jørgensen BB, Kasten S (2006) Sulfur cycling and methane oxidation. In: Schulz HD, Zabel M (eds) Marine geochemistry, 2nd edn. Springer, Berlin, pp 271–309 Krumins V, Gehlen M, Arndt S, Van Cappellen P, Regnier P (2013) Dissolved inorganic carbon and alkalinity fluxes from coastal marine sediments: model estimates for different shelf environments and sensitivity to global change. Biogeosciences 10:371– 398. doi:10.5194/bg-10-371-2013

310 Kvenvolden KA, Ginsburg GD, Soloviev VA (1993) Worldwide distribution of subaquatic gas hydrates. Geo-Mar Lett 13(1):32–40. doi:10.1007/BF01204390 Laier T, Jensen JB (2007) Shallow gas depth-contour map of the Skagerrak-western Baltic Sea region. Geo-Mar Lett 27(2/ 4):127–141. doi:10.1007/s00367-007-0066-2 Leipe T, Tauber F, Vallius H, Virtasalo J, Uścinowicz S, Kowalski N, Hille S, Lindgren S, Myllyvirta T (2011) Particulate organic carbon (POC) in surface sediments of the Baltic Sea. Geo-Mar Lett 31(3):175–188. doi:10.1007/s00367-010-0223-x McCutcheon S, Martin J, Barnwell TJ (1993) Water quality. In: Maidment D (ed) Handbook in hydrology, chap 11. McGrawHill, New York Mogollón JM, L’Heureux I, Dale AW, Regnier P (2009) Methane gasphase dynamics in marine sediments: a model study. Am J Sci 309:189–220. doi:10.2475/03.2009.01 Mogollón JM, Dale AW, L’Heureux I, Regnier P (2011) Impact of seasonal temperature and pressure changes on methane gas production, dissolution, and transport in unfractured sediments. J Geophys Res-Biogeosci 116, G03031. doi:10.1029/2010JG001592 Mogollón JM, Dale AW, Fossing H, Regnier P (2012) Timescales for the development of methanogenesis and free gas layers in recently-deposited sediments of Arkona Basin (Baltic Sea). Biogeosciences 9:1915–1933. doi:10.5194/bg-9-1915-2012

Geo-Mar Lett (2013) 33:299–310 Moros M, Lemke W, Kuijpers A, Endler R, Jensen JB, Bennike O, Gingele F (2002) Regressions and transgressions of the Baltic basin reflected by a new high-resolution deglacial and postglacial lithostratigraphy for Arkona Basin sediments (western Baltic Sea). Boreas 31(2):151–162. doi:10.1080/030094802320129953 Pimenov NV, Ulyanova MO, Kanapatsky TA, Veslopolova EF, Sigalevich PA, Sivkov VV (2010) Microbially mediated methane and sulfur cycling in pockmark sediments of the Gdansk Basin, Baltic Sea. Geo-Mar Lett 30(3/4):439–448. doi:10.1007/s00367-010-0200-4 Regnier P, Dale AW, Arndt S, LaRowe D, Mogollón JM, Van Cappellen P (2011) Quantitative analysis of anaerobic oxidation of methane (AOM) in marine sediments: a modeling perspective. Earth Sci Rev 106:105–130. doi:10.1016/j.earscirev.2011.01.002 Thießen O, Schmidt M, Theilen F, Schmitt M, Klein G (2006) Methane formation and distribution of acoustic turbidity in organic-rich surface sediments in the Arkona Basin, Baltic Sea. Cont Shelf Res 26(19):2469–2483. doi:10.1016/j.csr.2006.07.020 Ulyanova M, Sivkov V, Kanapatskij T, Sigalevich P, Pimenov N (2012) Methane fluxes in the southeastern Baltic Sea. Geo-Mar Lett 32(5/6):535–544. doi:10.1007/s00367-012-0304-0 Wallmann K, Drews M, Aloisi G, Bohrmann G (2006) Methane discharge into the Black Sea and the global ocean via fluid flow through submarine mud volcanoes. Earth Planet Sci Lett 248(1/ 2):545–560. doi:10.1016/j.epsl.2006.06.026