- Email: [email protected]
Carbon monoxide distribution and microbial consumption in the Southern Yellow Sea
Estuarine, Coastal and Shelf Science 163 (2015) 125e133
Contents lists available at ScienceDirect
Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss
Carbon monoxide distribution and microbial consumption in the Southern Yellow Sea Wei-Lei Wang a, b, Gui-Peng Yang a, *, Xiao-Lan Lu a a Key Laboratory of Marine Chemistry Theory and Technology, Ocean University of China, Ministry of Education/Qingdao Collaborative Innovation Center of Marine Science and Technology, Qingdao 266100, China b Department of Earth System Science, University of California, Irvine, CA 92697, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 January 2015 Received in revised form 10 June 2015 Accepted 11 June 2015 Available online 20 June 2015
Two cruises were conducted in the Southern Yellow Sea (SYS) in late July and early August 2008 to study marine carbon monoxide (CO) distribution, sea-to-air flux, and microbial consumption. Surface water dissolved CO showed an apparently higher average concentration (31.9% higher) during the second cruise than during the first cruise. Surface water CO concentrations were found to follow an apparent diurnal variation, during which CO concentrations varied by a factor of 8e14. Atmospheric CO mixing ratios ranged from 194 to 596 ppbv with an average of 390 ppbv for the first cruise, and from 53 to 238 ppbv with an average of 124 ppbv for the second one. Average sea-to-air CO fluxes (W92) were estimated to be 4.58 mmol m2 d1 and 0.08 mmol m2 d1 for the first and second cruise, respectively. Incubation experiments were conducted at 12 stations during the second cruise; results showed that surface-water microbial CO consumption rate constants (Kco) ranged from 0.07 to 0.83 h1, with an average of 0.33 h1. Microbial CO uptake typically followed first-order reaction kinetics at most of the studied stations. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Carbon monoxide Southern Yellow Sea Gas exchange Microbial CO consumption
1. Introduction Atmospheric CO plays an important role in controlling atmospheric oxidation capacity through its reaction with hydroxyl radical (OH) (Derwent, 1995), thus imposes an impact on other reduced species, such as methane (Crutzen and Zimmermann, 1991; Thompson, 1992). CO oxidation in conjunction with the reduction of NOx has an influence on the abundance of tropospheric ozone that is a potential greenhouse gas and atmospheric oxidant (Dignon and Hameed, 1985; Stubbins et al., 2006). The major source of oceanic CO is in-situ photolysis of colored dissolved organic matter (CDOM); and its sink includes microbial consumption, sea-to-air exchange (Conrad and Seiler, 1980, 1982; Bates et al., 1995), and vertical mixing (Kettle, 1994, 2005; Najjar et al., 1995; Gnanadesikan, 1996; Johnson and Bates, 1996; Stubbins et al., 2006). The ocean, as a whole, is a source of atmospheric CO. Carbon monoxide (CO) in seawater has been studied intensively
* Corresponding author. College of Chemistry and Chemical Engineering, Ocean University of China, Qingdao 266100, China. E-mail address: [email protected] (G.-P. Yang). http://dx.doi.org/10.1016/j.ecss.2015.06.012 0272-7714/© 2015 Elsevier Ltd. All rights reserved.
worldwide: from initial laboratory experiments (Swinnerton et al., 1968; Xie et al., 2002), sea-to-air flux estimates (Stubbins et al., 2006; Yang et al., 2010, 2011), to the mechanisms of biological consumption (Xie et al., 2005; Zhang et al., 2008), and photoproduction (Zhang et al., 2006; Yang et al., 2011). However, there are only a few papers reporting CO distribution, and biological CO consumption in marginal seas around China. The objective of this study is to supply CO data to the global CO biogeochemical cycle database. The Yellow Sea (YS) is divided into northern YS and southern YS (SYS) by a line between Chengshanjiao in Shandong Peninsula and Jangsan point in Korean Peninsula. The bathymetry along the western coasts (China side) in the SYS is characterized by a wide shallow water, Subei Bank (Lü et al., 2010). The SYS has a total surface area of 30.9 104 km2 and an average depth of 46 m. The hydrographic character of this region is mainly defined by a circulation system that includes the Yellow Sea Cold Water Mass, the Yellow Sea Warm Current, and the Yellow Sea Coastal Currents on the western side (Feng et al., 1999; Yuan et al., 2008). With the Jiaozhou Bay and the Haizhou Bay located on the west of the SYS, the SYS is not only one of the most concentrated regions of human activities and economic development, but also an intense
126
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
interaction zone of land, sea, and atmospheric processes. Owing to its shallow depth, the SYS has no deep-water mass exchange, leading to relatively poor self-purification ability and hydrological conditions, which makes the SYS a quite complicated area. In this paper, we reported the distribution of CO in surface water and in the atmosphere of the SYS in July and August 2008. 2. Material and methods 2.1. Sampling
Water temperature and salinity data were obtained from the CTD. Wind speeds were measured at a height of 10 m by ship-borne weather instrument (27600-4X, Young®, US). 2.3. Calculation of CO flux Sea-to-air CO fluxes were calculated according to the gas exchange model of Liss (1973), which assumes that sea-to-air flux (F) is proportional to the product of the concentration difference (CobsCeq) across the air-sea interface and a transfer velocity (k):
F ¼ k Cobs Ceq
The studied area in the SYS is presented in Fig. 1. A total of 55 sea surface samples were collected on board R/V “Dong Fang Hong 2” during two cruises conducted respectively in July 22e26 and August 5e14, 2008. Water samples were collected using 12-L Niskin bottles mounted on a Seabird CTD-Sonde Rosette supplying salinity, temperature, and depth data. Blanks of our sampling bottles were estimated to be at picomolar levels due to the fact that the bottles were held in a waterproof shelter that provided a low-light environment to ensure the Niskin bottles' acceptability (Zhang et al., 2006; Xie et al., 2009). Atmospheric CO samples were collected on the top deck of the ship (10 m above sea level) with a 50-ml dry, acid-cleaned, and gastight glass-only syringe when the ship was underway. In this way, the exhausts of the ship would not significantly contaminate the atmospheric CO samples.
where T is temperature in Celsius degree. Cobs is the observed concentration of dissolved CO, which is calculated according to Xie et al. (2002). Briefly, the initial CO concentration ({CO}eq in ml CO/ ml H2O) is calculated, assuming a mass balance:
2.2. CO analysis and incubation
fCOgeq ¼ 106 ma ðb p Vw þ Va Þ=Vw
Both atmospheric and seawater CO samples were immediately (normally less than half an hour) analyzed onboard after collection following the methods described in Xie et al. (2002) and Lu et al. (2010). Briefly, water samples were transferred from Niskin bottles into 50-ml acid-cleaned glass-only syringes using a Teflon tube coupled with a three-way Nylon valve. Sampling syringes were rinsed with sample water for three times, including one bubblefree flushing, before the final drawing. Surface water samples (average depth: 0.5 m) were analyzed as soon as being sampled with no delay to avoid any possible production or consumption. A precise ratio between sample water and CO-zero headspace air was set using a device described in Xie et al. (2002). Phase equilibration was established by shaking the sampling syringe for 4 min with a shaker (GJ-8175, Feihuang®, Guangdong, China). After that, headspace sample was injected into a RGA3 reduction gas analyzer (Trace Analytical Model ta3000R Gas Analyzer, Ametek®, USA). A 0.2 mm Nuclepore Teflon water-impermeable (13-mm diameter) filter fitted in a filter holder (Millipore®) was used here to prevent any liquid water from being injected into the analyzer. Atmospheric samples were injected into the analyzer immediately after sampling. Calibration was conducted every few hours with a commercial standard CO gas (nominal concentration: 299 ppbv in zerograde air, analytical accuracy: ±5%, DaTe®, Dalian, China) that was a certified reference material with the standard reference material No. 060152, and was approved by China State Bureau of Technical Supervision. In the incubation experiments, surface-water was incubated in a 200 ml acid-cleaned Al-foil wrapped all-glass syringe at a temperature ±1 C of surface water temperature. A 50 ml acid-cleaned glass syringe was used to get sub-samples from the 200 ml incubation syringe. The first sample was sampled and measured immediately after each incubation; and the second one was analyzed half an hour later after the first sample. The following subsamples were analyzed at appropriate intervals depending on CO consumption rate. The method precisions for repeated analyses of water samples, atmospheric samples, and incubation samples were about 5%, 3%, and 7%, respectively, in routine sample analysis.
where the transfer velocity k was calculated using the quadratic k/ wind speed relationship established by Wanninkhof (1992) (W92), and was adjusted by multiplying (ScCO/660)0.5, where ScCO is the Schmidt number for CO. The ScCO in seawater was calculated according to Zafiriou et al. (2008):
ScCO ¼ 0:0553T 3 þ 4:3825T 2 140:07T þ 2134
where ma is equilibrated headspace mixing ratio (ppmv); b is the Bunsen solubility coefficient of CO, which varies as a function of temperature and salinity (Wiesenburg and Guinasso, 1979); p is atmospheric pressure (atm) of dry air; Vw is the volume of water sample (ml); Va is the volume of the headspace air (ml). Cobs (nM) is calculated using the following equation:
. Cobs ¼ 109 p fCOgeq ðR TÞ where R is the gas constant (0.08206 atm L mol1 K1); and T is the temperature (K). Ceq is the air-equilibrated seawater CO concentration.
Ceq ¼ ðCOa bÞ=M where COa is the CO mixing ratio of the atmosphere (ppbv); M is the molar volume of CO at standard temperature and pressure, 25.094 L mol1 (Lide, 1992; Stubbins et al., 2006). 3. Results and discussion 3.1. Atmospheric CO Atmospheric CO mixing ratios in the investigated area ranged from 194.8 to 596.9 ppbv with an average of 390.4 ppbv for the first cruise, which was higher than those reported in the literature. For instance, Yang et al. (2010) reported that average CO mixing ratio was 297 ppbv in the overlapped study area in November 2007. Lam et al. (2000) reported that average CO mixing ratio in autumn was 313 ppbv at their monitoring site Cape D'Aguilar (22.2 N, 114.3 E, 60 m above sea level). Stubbins et al. (2006) found that mean CO mixing ratio was 151 ppbv over the ocean from Montevideo, Uruguay (35 S, 55 W) to Grimsby, UK (54 N, 0 W) in April and May 2000. The higher atmospheric CO mixing ratio observed during the
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
127
Fig. 1. Locations of the sampling stations in the Southern Yellow Sea for the first cruise (upper), with extended stations in the East China Sea for the second cruise (lower).
first cruise were probably due to the fact that the experienced high wind speed intensified terrestrial influence from the mainland of China, which is believed to be a major CO source in the study areas. In contrast, average atmospheric CO mixing ratio during the second cruise ranged from 53.5 to 238.5 ppbv with an average of
126.9 ppbv, which was lower than that obtained during the first cruise, and was even lower than that observed over the open sea (e.g., Stubbins et al., 2006). The low atmospheric CO mixing ratio during the second cruise might be attributed to the low wind speeds (average: 0.45 m/s), which reduced the terrestrial influence
128
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
at the same time reduced sea-to-air gas exchange. 3.2. Dissolved CO in seawater The CO data obtained from the surface water of the SYS, together with water depth (D), surface water temperature (T), salinity (S), wind speed (u), time of sampling (t), and sea-to-air flux measured during the two cruises, are listed in Tables 1 and 2, respectively. Surface seawater CO concentrations in the investigated area ranged from 0.10 to 7.33 nmol L1, with a mean of 1.91 nmol L1 for the first cruise, and from 0.09 to 11.8 nmol L1 with a mean of 2.52 nmol L1 for the second cruise, respectively. CO concentrations obtained in these two studies were much higher than those reported in Yang et al. (2010) in almost the same area in November 2007, when the surface water CO concentrations ranged from 0.13 to 2.30 nmol L1, with an average of 0.68 nmol L1. The difference in CO concentrations between the current study and Yang et al. (2010) could be due to many reasons, such as the different sampling seasons, since temperature (average sea surface temperature 27.3 vs. 21.2 C) was higher and light was stronger in present study (summer vs. winter) than those in Yang et al. (2010). High temperature and strong light would promote CO production in seawater (Conrad et al., 1982; Zhang et al., 2006). The difference might also be due to the different biological activities, which had been verified by the low microbial CO consumption rate constants measured in the same cruise (see section 3.4). From Tables 1 and 2, it can be seen that both average and maximum CO concentrations in the second cruise are higher than those in the first cruise. If we made an assumption that biological consumption, the major CO sink in the ocean, during the two cruises were in the similar level, the difference in CO concentrations might be due to the fact that the wind speeds during the first cruise (the mean wind speed was 5.6 m/s) were higher than those during the second cruise (the mean wind speed was 0.45 m/s). The higher wind speed resulted in a deeper mixed layer, a greater seato-air emission, and ultimately the lower surface seawater CO concentrations during the first cruise. CO concentrations were binned, based on the sampling time, into eight three-hour-long classes. Averaged concentration along
with standard deviation of each class was plotted against the time of a day to get average diurnal variations (Fig. 2). Fig. 2a and b show the diurnal variations for the first cruise and the second cruise, respectively. It is clear that CO diurnal variations of the two cruises showed a similar pattern, with daytime maximal values being 8e14 times higher than nighttime minimal values. CO concentrations remained basically constant throughout the night, indicating a dynamic production-consumption balance. The highest CO concentrations generally appeared from 15:00 to 18:00 (local time), which was several hours later than the solar radiation maximum. The huge ranges and distribution patterns observed in diurnal variations were caused by the competition between photoproduction and microbial consumption (Day and Faloona, 2009), which are believed to be the main source and sink of oceanic CO, respectively. It is interesting to find from Fig. 2a and b that nighttime CO concentrations during the two cruises seemed equal, but the daytime CO concentrations had a large disparity, with the second cruise exhibiting higher CO concentrations. The same surveying season and area, to some extent, ensured that the production and consumption were at a similar level. The daytime CO concentration difference possibly indicated that other processes such as sea-to-air emissions also played a role in controlling oceanic CO biogeochemical cycle. Other researchers have reported CO diurnal variation. Conrad et al. (1982) observed a diurnal variation in euphotic ocean water and claimed that the variation of CO concentrations was a product of photochemical production (source) of CO from the dissolved organic carbon (DOC), microbial oxidation, and sea-to-air flux (sink). Jones (1991) also investigated the diurnal variation of CO in seawater and found that diurnal variation could occur under sunny conditions. The variation factors of CO concentrations in the diurnal cycle were reported to be 5e7 by Stubbins et al. (2006). Our observations were in good agreement with the literature. 3.3. Sea-to-air flux of CO CO saturation factor (a), which is derived from Cobs divided by Ceq, ranged from 0.54 to 25.90, with an average of 6.15 during the
Table 1 Description of sampling stations along with depth (D), temperature (T), salinity, wind speed (u), time of sampling (t), surface [CO], CO mixing ratio (MR), CO saturation factor (a), and sea-to-air CO flux (F) in the Southern Yellow Sea during the first cruise. Station
Location
D (m)
T ( C)
S
u (m/s)
t
[CO] (nM)
MR (ppbv)
a
Flux (mmol/m2/d)
D9 C8 C6 C4 C2 C1 F1 F2 B1 B3 B5 B7 B9 B11 E2 A6 A4 A2 A1 G2 G1 G0 FB2 Mean
35.25E, 35.25E, 35.25E, 35.25E, 35.25E, 35.25E, 35.00E, 34.73E, 34.50E, 34.50E, 34.50E, 34.50E, 34.50E, 34.50E, 34.00E, 33.50E, 33.50E, 33.50E, 33.50E, 33.50E, 34.00E, 34.25E, 34.66E, NA
73.0 64.0 45.0 43.0 33.0 33.0 33.0 25.0 18.0 20.0 20.0 51.6 63.0 73.0 68.0 45.0 37.7 25.0 18.0 18.0 18.0 20.0 21.0 37.6
25.62 25.83 26.04 26.13 25.39 24.46 24.21 22.89 22.73 19.21 22.61 25.60 26.32 26.16 26.20 25.14 25.30 24.99 26.75 26.82 26.17 23.11 23.24 24.82
30.99 30.59 30.64 30.18 30.23 30.94 30.68 30.12 29.98 30.71 30.22 31.27 31.21 31.62 31.05 31.30 31.33 30.92 30.10 30.35 31.00 29.63 30.79 30.69
7.7 5.3 5.0 3.4 3.3 6.3 6.2 7.9 6.4 3.7 6.0 6.5 7.2 5.9 3.1 3.0 4.3 4.8 6.4 8.0 6.6 6.8 5.2 5.61
08:39 11:34 14:31 17:19 20:02 21:34 23:08 01:05 02:48 05:42 08:15 11:02 13:46 16:48 20:17 23:23 02:11 05:02 06:35 08:49 11:04 13:32 16:17 NA
1.27 2.32 1.12 1.91 0.70 0.50 0.10 0.66 0.15 0.84 1.86 4.95 4.41 3.64 3.42 1.36 1.07 0.33 0.28 0.80 0.92 3.95 7.33 1.91
344.3 368.0 244.6 240.3 287.6 271.5 240.9 194.8 233.1 310.4 521.0 596.9 581.7 559.2 456.9 421.1 555.7 397.4 364.5 350.1 477.4 577.2 384.0 390.4
4.99 8.57 6.23 10.8 3.28 2.51 0.54 4.63 0.85 3.68 4.84 11.34 10.32 8.84 10.20 4.39 2.62 1.11 1.04 3.11 2.61 9.30 25.90 6.15
6.53 6.27 2.57 2.20 0.57 1.26 0.34 3.28 0.11 0.78 5.36 20.65 22.77 12.35 3.26 1.01 1.32 0.08 0.05 3.88 2.72 16.64 19.51 5.76
NA: not available.
123.00N 122.50N 122.00N 121.50N 121.00N 120.75N 120.67N 120.58N 120.50N 121.00N 121.50N 122.00N 122.50N 123.00N 123.00N 123.00N 122.50N 122.00N 121.75N 121.50N 121.25N 121.00N 120.85N
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
129
Table 2 Description of sampling stations along with depth (D), temperature (T), salinity, wind speed (u), time of sampling (t), surface [CO], CO mixing ratio (MR), CO saturation factor (a), and sea-to-air CO flux (F) in the Southern Yellow Sea during the second cruise. Station
Location
D (m)
T ( C)
S
u (m/s)
t
[CO] (nM)
MR (ppbv)
a
Flux (mmol/m2/d)
A2 A4 A6 A8 B1 B3 B4 B6 D1 D2 D4 D6 D8 D9 E2 F0401 F0403 F0405 F0407 F0501 F0502 F0504 F0506 F0507 F0508 F0601 F0603 G3 G5 G7 H1 H2 Mean
35.99E, 35.67E, 35.33E, 35.00E, 35.00E, 35.00E, 35.00E, 35.00E, 34.50E, 34.50E, 34.50E, 34.50E, 34.50E, 34.50E, 33.83E, 31.63E, 31.97E, 32.33E, 32.71E, 31.50E, 31.42E, 31.25E, 31.06E, 30.96E, 30.84E, 32.46E, 31.67E, 33.50E, 33.50E, 33.50E, 33.00E, 32.50E, NA
30.9 41.5 55.3 73.1 37.4 36.1 46.3 61.5 18.2 19.8 21.0 51.8 68.8 68.8 68.2 34.0 41.5 43.8 80.4 26.4 42.0 48.7 50.5 60.2 72.6 92.5 68.1 28.9 45.0 69.0 19.8 25.1 48.4
23.9 26.1 26.8 27.3 30.1 30.8 26.2 27.4 24.8 23.1 26.8 25.9 27.7 27.6 27.8 27.1 26.9 28.9 29.1 26.5 27.7 28.7 29.6 28.8 28.9 28.7 29.0 25.5 26.9 27.8 25.8 26.1 27.3
30.79 30.47 30.45 30.78 29.81 30.81 30.77 30.82 27.63 28.99 30.29 30.33 31.44 31.31 31.20 21.41 32.58 30.34 28.79 26.98 26.17 32.64 33.70 33.75 32.17 28.91 28.61 31.58 30.94 31.24 31.06 30.97 30.24
0.2 0.5 0.4 0.3 0.6 0.1 0.8 0.9 0.5 0.3 0.3 0.3 0.5 0.4 0.4 0.2 0.6 0.6 0.5 0.3 0.5 0.5 0.7 0.5 0.4 0.4 0.6 0.9 0.4 0.1 0.5 0.2 0.5
00:35 04:37 09:13 14:01 01:19 23:16 21:26 17:58 06:17 07:39 11:22 14:59 18:44 20:25 01:17 00:44 05:34 10:51 16:40 02:22 00:29 19:30 14:03 10:48 07:34 19:48 01:08 11:13 07:42 03:58 15:47 18:54 NA
0.91 0.11 3.97 8.98 7.80 0.41 0.58 11.80 0.09 1.41 2.59 3.16 1.69 1.05 0.18 1.63 1.66 5.18 2.47 0.44 0.24 3.68 3.36 1.29 0.39 2.70 0.43 3.91 0.70 0.23 7.21 0.35 2.52
146.3 53.5 63.2 138.7 63.8 56.3 66.4 111.0 59.6 59.4 69.5 184.1 211.2 225.9 212.4 101.8 102.5 112.0 115.6 91.2 89.2 84.2 80.8 NA 87.1 NA 85.9 238.5 211.9 202.0 202.8 207.8 124.5
8.48 2.79 85.90 88.80 174.01 10.40 11.90 145.11 2.00 29.81 50.92 23.30 10.63 6.15 1.12 20.90 22.30 65.72 29.80 6.64 3.73 61.72 60.10 NA 6.38 NA 7.06 22.12 4.53 1.59 48.42 2.30 33.81
0.003 0.002 0.070 0.090 0.344 0.000 0.037 1.078 0.001 0.013 0.026 0.030 0.044 0.016 0.000 0.007 0.064 0.218 0.071 0.004 0.005 0.107 0.196 NA 0.006 NA 0.016 0.327 0.010 0.000 0.192 0.001 0.056
121.02N 121.67N 122.33N 123.00N 120.67N 121.33N 121.66N 122.33N 120.50N 120.67N 121.32N 122.00N 122.67N 123.00N 123.45N 122.66N 123.52N 124.42N 125.39N 122.33N 122.71N 123.61N 124.72N 125.24N 125.85N 125.85N 125.85N 122.33N 123.00N 123.67N 122.08N 122.15N
NA: not available.
first cruise, and from 1.12 to 174.3 with a mean of 33.82 during the second cruise. All surface water CO concentrations were supersaturated compared with atmospheric CO mixing ratios during the two cruises, except for those measured at stations F1 and B1 in the first cruise, where saturation factors were less than one. Water samples at these two stations were collected at night (Table 1) when CO concentration was low during a diurnal cycle (Fig. 2a and b). In general, the saturation factors during the second cruise were much higher than those during the first cruise, which was caused by the higher surface water CO concentrations and the lower atmospheric CO mixing ratios during the second cruise. The saturation factors measured in the current study were higher than those in previous study, for example, the average saturation factor in the second cruise was approximately 10 times the average reported in Yang et al. (2010). The reported saturation factors in the coastal and marginal seas ranged from 3 to 100 (Swinnerton at al., 1969, 1970; Seiler and Junge, 1970; Lamontagne et al., 1971; Conrad et al., 1982; Butler et al., 1987). The average saturation factors of both the first and second cruises fell within this reported range. However, the lowest value of the first cruise was less than the lower limit of the literature value; while the highest value of the second cruise was greater than the upper limit of the literature value. The high saturation factors of the second cruise indicated either high production of CO or decreased microbial consumption. Based on the method introduced by Stubbins et al. (2006) and Yang et al. (2010), average sea-to-air emissions were estimated to be 4.58 mmol m2 d1 (W92) for the first cruise, and 0.08 mmol m2 d1 (W92) for the second cruise, respectively. The sea-to-air flux during the first cruise was higher than those obtained from the North Atlantic Ocean (2.2 ± 1.5 mmol m2 d1) and
the North Global Ocean (1.9 ± 1.3 mmol m2 d1) (Stubbins et al., 2006). It was also greater than our previous result obtained in November 2007, which was 1.82 mmol m2 d1 by the W92 equation (Yang et al., 2010). Additionally, it is interesting to find that the average saturation factor in the second cruise is about 5 times greater than that in the first cruise, whereas the average sea-to-air flux of CO in the first cruise is around 60 times greater than that in the second cruise. These paradoxical distributions are caused by difference in wind speed that is quadratically related to gas transfer velocity (k). This comparison emphasizes the role played by wind speed when the W92 equation is used to calculate the sea-to-air flux. 3.4. Microbial consumption rate of CO In the second cruise, whole water incubation method (see Xie et al., 2005 for detail) was applied to determine microbial CO consumption rate. The incubation experiments showed that CO concentration decreased quickly at the beginning of incubation, and slowed down over time. Decay curves at most of the stations showed exponential decay patterns. Previous studies also reported first order exponential decay kinetic pattern. For example, Xie et al. (2005) reported that CO concentrations decreased exponentially with incubation time in the Delaware Bay, the NW Atlantic, and the Beaufort Sea. Yang et al. (2010) added additional supports to the exponential decay pattern with data collected in the East China Sea. However, at two stations A8 and F0507, as shown in Fig. 3, CO concentrations decreased linearly with incubation time. Linear decay patterns were reported in Xie et al. (2005) and Zhang (2008). Zhang (2008) pointed out that linear decay patterns were probably
130
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
Fig. 2. Variation of dissolved CO concentrations in surface seawater along with the time of day. (a). Diurnal variations of the first cruise. (b). Diurnal variations of the second cruise.
caused by special microbe species. Fig. 3 indicates that the microbial CO consumption rate constants, Kco, ranged from 0.07 to 0.83 h1, with an average of 0.33 h1. Our results are apparently lower than those previously measured in the East China Sea and the SYS (from 0.15 to 2.14 h1 with an average of 0.80 h1), but comparable to those obtained in open oceans. For instance, Xie et al. (2005) found that Kco ranged from 0.071 to 0.91 h1 with a mean value of 0.26 h1 in the NW
Atlantic. Day and Faloona (2009) reported similar microbial consumption rates (0.05e0.80 h1) in the California Current upwelling system. 4. Conclusions In this work, we reported the distribution, flux, and biological consumption of CO in the SYS. Some important conclusions were
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
drawn as follows. This studied area, which has a strong seasonal cycle, together with the apparently different wind-speeds experienced during the two cruises, offered us an opportunity to study the factors that could influence surface water CO concentration. Comparison
131
between these two cruises together with comparison with previous research results (Yang et al., 2010) revealed that surface water CO concentration followed seasonal, diurnal, and day-to-day variations. These variations were controlled jointly by a combination of light intensity, biological CO consumption, temperature, and wind-
Fig. 3. CO concentrations versus incubation time at selected incubation stations.
132
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
Fig. 3. (continued).
speed (depth of mixed layer and sea-to-air flux). Surface water CO concentration was also subject to strong spatial variations, as indicated by the large error bars in Fig. 2. Atmospheric CO mixing ratio in the studied area was dominated by terrestrial influence, and thus varied dynamically with wind direction and speed. The fact that both the lowest and the highest biological CO consumption
rate constants were out of reported range revealed the complexity of the studied area. The paradoxical distributions of saturation factor and sea-to-air flux observed in these two cruises emphasized the important role played by wind speed in calculating sea-to-air flux. Even though the sea-to-air CO flux in the second cruise was suppressed by the unusually low wind speed, the SYS was still an
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
atmospheric CO source. Acknowledgments We thank the captain and crew of the R/V “Dong Fang Hong 2” for their support and help during the two cruises. This work was financially supported by the National Natural Science Foundation of China (No. 40976043), the Changjiang Scholars Programme, Ministry of Education of China and the Taishan Scholars Programme of Shandong Province, China. References Bates, T.S., Kelly, K.C., Johnson, J.E., Gammon, R.H., 1995. Regional and seasonal variations in the flux of oceanic carbon monoxide to the atmosphere. J. Geophys. Res. 100 (D11), 23093e23101. Butler, J.H., Jones, R.D., Garber, J.H., Gordon, L.I., 1987. Seasonal distributions and turnover of reduced traces gases and hydroxylamine in Yaquina Bay, Oregon. Geochim. Cosmochim. Acta 51, 697e706. Conrad, R., Seiler, W., 1980. Photo-oxidative production and microbial consumption of carbon monoxide in seawater. FEMS Microbiol. Lett. 9 (1), 61e64. Conrad, R., Seiler, W., 1982. Utilization of traces of carbon monoxide by aerobic oligotrophic microorganisms in ocean, lake and soil. Arch. Microbiol. 132 (1), 41e46. Conrad, R., Seiler, W., Bunse, G., Giehl, H., 1982. Carbon monoxide in seawater (Atlantic Ocean). J. Geophys. Res. 87 (C11), 8839e8852. Crutzen, P.J., Zimmermann, P.H., 1991. The changing photochemistry of the troposphere. Tellus 43AB, 136e151. Day, D.A., Faloona, I., 2009. Carbon monoxide and chromophoric dissolved organic matter cycles in the shelf waters of the northern California upwelling system. J. Geophys. Res. 114 (C01006) http://dx.doi.org/10.1029/2007JC004590. Derwent, R.G., 1995. Air chemistry and terrestrial gas emissions: a global perspective. Philos. Trans. R. Soc. Lond. Ser. A 351, 205e217. Dignon, J., Hameed, S., 1985. A model investigation of the impact of increase in anthropogenic NOx emissions between 1967 and 1980 on tropospheric ozone. J. Atmos. Chem. 3, 491e506. Feng, S.Z., Li, F.Q., Li, S.Q., 1999. Marine Science Introduction. Higher Education Press, Beijing, China, pp. 434e498 (in Chinese). Gnanadesikan, A., 1996. Modeling the diurnal cycle of carbon monoxide: sensitivity to physics, chemistry, biology, and optics. J. Geophys. Res. 101 (C5), 12177e12191. Johnson, J.E., Bates, T.S., 1996. Sources and sinks of carbon monoxide in the mixed layer of the tropical South Pacific Ocean. Glob. Biogeochem. Cycl. 10 (2), 347e359. Jones, R.D., 1991. Carbon monoxide and methane distribution and consumption in the photic zone of the Sargasso Sea. Deep Sea Res. 38 (6), 625e635. Kettle, A.J., 1994. A Model of the Temporal and Spatial Distribution of Carbon Monoxide in the Mixed Layer (M.Sc. thesis). Woods Hole Oceanographic Institution, Institute of Technology Joint Program in Oceanography, Woods Hole, Mass, 146pp. Kettle, A.J., 2005. Diurnal cycling of carbon monoxide (CO) in the upper ocean near Bermuda. Ocean. Model. 8, 337e367. Lam, K.S., Wang, T.J., Chan, L.Y., Wang, T., Harris, J., 2000. Flow patterns influencing the seasonal behavior of surface ozone and carbon monoxide at a coastal site near Hong Kong. Atmos. Environ. 35, 3121e3135. Lamontagne, R.A., Swinnerton, J.W., Linnenbom, V.J., 1971. Nonequilibrium of carbon monoxide and methane at the air-sea interface. J. Geophys. Res. 76, 5117e5121.
133
Lide, D.R., 1992. The Handbook of Chemistry and Physics. CRC Press, Ann Arbor. Liss, P.S., 1973. Processes of gas exchange across an airewater interface. Deep Sea Res. 20, 221e238. Lu, X.L., Yang, G.P., Wang, X.M., Wang, W.L., Ren, C.Y., 2010. Determination of carbon monoxide in seawater by headspace analysis. Chin. J. Anal. Chem. 38 (3), 352e356. Lü, X., Qiao, F., Xia, C., Wang, G., Yuan, Y., 2010. Upwelling and surface cold patches in the Yellow Sea in summer: effects of tidal mixing on the vertical circulation. Cont. Shelf Res. 30, 620e632. Najjar, R.G., Erickson III, D.J., Madronich, S., 1995. Modeling the air-sea fluxes of gases formed from the decomposition of dissolved organic matter: carbonyl sulfide and carbon monoxide. In: Zepp, R.G., Sonntag, C. (Eds.), The Role of Nonliving Organic Matter in the Earth's Carbon Cycle. Wiley & Sons, New York, pp. 107e132. Seiler, W., Junge, C., 1970. Carbon monoxide in the atmosphere. J. Geophys. Res. 75, 2217e2226. Stubbins, A., Uher, G., Kitidis, V., Law, C., Upstill-Goddard, R.C., Woodward, E.M.S., 2006. The open-ocean source of atmospheric carbon monoxide. Deep Sea Res. 53, 1685e1694. Swinnerton, J.W., Linnenbom, V.J., Cheek, C.H., 1968. A sensitive gas chromatographic method for determining carbon monoxide in seawater. Limnol. Oceanogr. 13 (1), 193e195. Swinnerton, J.W., Linnenbom, V.J., Cheek, C.H., 1969. Distribution of methane and carbon monoxide between the atmosphere and natural waters. Environ. Sci. Technol. 3, 836e838. Swinnerton, J.W., Linnenbom, V.J., Lamontagne, R.A., 1970. Ocean: a natural source of carbon monoxide. Science 167 (3920), 984e986. Thompson, A.M., 1992. The oxidizing capacity of the Earth's atmosphere: probable past and future changes. Science 256, 1157e1165. Wanninkhof, R., 1992. Relationships between wind speed and gas exchange over the ocean. J. Geophys. Res. 97 (C5), 7373e7382. Wiesenburg, D.A., Guinasso, N.L., 1979. Equilibrium solubilities of methane, carbon monoxide, and hydrogen in water and seawater. J. Chem. Eng. Data 24, 356e360. Xie, H., Andrews, S.S., Martin, W.R., Miller, J., Ziolkowski, L., Taylor, C.D., Zafiriou, O.C., 2002. Validated methods for sampling and headspace analysis of carbon monoxide in seawater. Mar. Chem. 77, 93e108. Xie, H., Zafiriou, O.C., Umile, T.P., Kieber, D.J., 2005. Biological consumption of carbon monoxide in Delaware Bay, NW Atlantic and Beaufort Sea. Mar. Ecol. Prog. Ser. 290, 1e14. Xie, H., Belanger, S., Demers, S., Vincent, W.F., Papakyriakou, T.N., 2009. Photobiogeochemical cycling of carbon monoxide in the southeastern Beaufort Sea in spring and autumn. Limnol. Oceanogr. 54 (1), 234e249. Yang, G.P., Wang, W.L., Lu, X.L., Ren, C.Y., 2010. Distribution, flux and biological consumption of carbon monoxide in the Southern Yellow Sea and the East China Sea. Mar. Chem. 122, 74e82. Yang, G.P., Ren, C.Y., Lu, X.L., Liu, C.Y., Ding, H.B., 2011. Distribution, flux, and photoproduction of carbon monoxide in the East China Sea and Yellow Sea in spring. J. Geophys. Res. 116, C02001. http://dx.doi.org/10.1029/2010JC006300. Yuan, D.L., Li, J.R., Hu, D.X., 2008. Cross-shelf circulation in the Yellow and East China Seas indicated by MODIS satellite observations. J. Mar. Syst. 70, 134e149. Zafiriou, O.C., Xie, H., Nelson, N.B., Najjar, R.G., Wang, W., 2008. Diel carbon monoxide cycling in the upper Sargasso Sea near Bermuda at the onset of spring and in midsummer. Limnol. Oceanogr. 53, 835e850. Zhang, Y., Xie, H., Chen, G.H., 2006. Factors affecting the efficiency of carbon monoxide photoproduction in the St. Lawrence estuarine system (Canada). Environ. Sci. Technol. 40, 7771e7777. Zhang, Y., 2008. Identity and Dynamics of the Microbial Community Responsible for Carbon Monoxide Oxidation in Marine Environments (Ph.D. thesis). Ocean University of China, Qingdao, pp. 125e155.
Contents lists available at ScienceDirect
Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss
Carbon monoxide distribution and microbial consumption in the Southern Yellow Sea Wei-Lei Wang a, b, Gui-Peng Yang a, *, Xiao-Lan Lu a a Key Laboratory of Marine Chemistry Theory and Technology, Ocean University of China, Ministry of Education/Qingdao Collaborative Innovation Center of Marine Science and Technology, Qingdao 266100, China b Department of Earth System Science, University of California, Irvine, CA 92697, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 January 2015 Received in revised form 10 June 2015 Accepted 11 June 2015 Available online 20 June 2015
Two cruises were conducted in the Southern Yellow Sea (SYS) in late July and early August 2008 to study marine carbon monoxide (CO) distribution, sea-to-air flux, and microbial consumption. Surface water dissolved CO showed an apparently higher average concentration (31.9% higher) during the second cruise than during the first cruise. Surface water CO concentrations were found to follow an apparent diurnal variation, during which CO concentrations varied by a factor of 8e14. Atmospheric CO mixing ratios ranged from 194 to 596 ppbv with an average of 390 ppbv for the first cruise, and from 53 to 238 ppbv with an average of 124 ppbv for the second one. Average sea-to-air CO fluxes (W92) were estimated to be 4.58 mmol m2 d1 and 0.08 mmol m2 d1 for the first and second cruise, respectively. Incubation experiments were conducted at 12 stations during the second cruise; results showed that surface-water microbial CO consumption rate constants (Kco) ranged from 0.07 to 0.83 h1, with an average of 0.33 h1. Microbial CO uptake typically followed first-order reaction kinetics at most of the studied stations. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Carbon monoxide Southern Yellow Sea Gas exchange Microbial CO consumption
1. Introduction Atmospheric CO plays an important role in controlling atmospheric oxidation capacity through its reaction with hydroxyl radical (OH) (Derwent, 1995), thus imposes an impact on other reduced species, such as methane (Crutzen and Zimmermann, 1991; Thompson, 1992). CO oxidation in conjunction with the reduction of NOx has an influence on the abundance of tropospheric ozone that is a potential greenhouse gas and atmospheric oxidant (Dignon and Hameed, 1985; Stubbins et al., 2006). The major source of oceanic CO is in-situ photolysis of colored dissolved organic matter (CDOM); and its sink includes microbial consumption, sea-to-air exchange (Conrad and Seiler, 1980, 1982; Bates et al., 1995), and vertical mixing (Kettle, 1994, 2005; Najjar et al., 1995; Gnanadesikan, 1996; Johnson and Bates, 1996; Stubbins et al., 2006). The ocean, as a whole, is a source of atmospheric CO. Carbon monoxide (CO) in seawater has been studied intensively
* Corresponding author. College of Chemistry and Chemical Engineering, Ocean University of China, Qingdao 266100, China. E-mail address: [email protected] (G.-P. Yang). http://dx.doi.org/10.1016/j.ecss.2015.06.012 0272-7714/© 2015 Elsevier Ltd. All rights reserved.
worldwide: from initial laboratory experiments (Swinnerton et al., 1968; Xie et al., 2002), sea-to-air flux estimates (Stubbins et al., 2006; Yang et al., 2010, 2011), to the mechanisms of biological consumption (Xie et al., 2005; Zhang et al., 2008), and photoproduction (Zhang et al., 2006; Yang et al., 2011). However, there are only a few papers reporting CO distribution, and biological CO consumption in marginal seas around China. The objective of this study is to supply CO data to the global CO biogeochemical cycle database. The Yellow Sea (YS) is divided into northern YS and southern YS (SYS) by a line between Chengshanjiao in Shandong Peninsula and Jangsan point in Korean Peninsula. The bathymetry along the western coasts (China side) in the SYS is characterized by a wide shallow water, Subei Bank (Lü et al., 2010). The SYS has a total surface area of 30.9 104 km2 and an average depth of 46 m. The hydrographic character of this region is mainly defined by a circulation system that includes the Yellow Sea Cold Water Mass, the Yellow Sea Warm Current, and the Yellow Sea Coastal Currents on the western side (Feng et al., 1999; Yuan et al., 2008). With the Jiaozhou Bay and the Haizhou Bay located on the west of the SYS, the SYS is not only one of the most concentrated regions of human activities and economic development, but also an intense
126
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
interaction zone of land, sea, and atmospheric processes. Owing to its shallow depth, the SYS has no deep-water mass exchange, leading to relatively poor self-purification ability and hydrological conditions, which makes the SYS a quite complicated area. In this paper, we reported the distribution of CO in surface water and in the atmosphere of the SYS in July and August 2008. 2. Material and methods 2.1. Sampling
Water temperature and salinity data were obtained from the CTD. Wind speeds were measured at a height of 10 m by ship-borne weather instrument (27600-4X, Young®, US). 2.3. Calculation of CO flux Sea-to-air CO fluxes were calculated according to the gas exchange model of Liss (1973), which assumes that sea-to-air flux (F) is proportional to the product of the concentration difference (CobsCeq) across the air-sea interface and a transfer velocity (k):
F ¼ k Cobs Ceq
The studied area in the SYS is presented in Fig. 1. A total of 55 sea surface samples were collected on board R/V “Dong Fang Hong 2” during two cruises conducted respectively in July 22e26 and August 5e14, 2008. Water samples were collected using 12-L Niskin bottles mounted on a Seabird CTD-Sonde Rosette supplying salinity, temperature, and depth data. Blanks of our sampling bottles were estimated to be at picomolar levels due to the fact that the bottles were held in a waterproof shelter that provided a low-light environment to ensure the Niskin bottles' acceptability (Zhang et al., 2006; Xie et al., 2009). Atmospheric CO samples were collected on the top deck of the ship (10 m above sea level) with a 50-ml dry, acid-cleaned, and gastight glass-only syringe when the ship was underway. In this way, the exhausts of the ship would not significantly contaminate the atmospheric CO samples.
where T is temperature in Celsius degree. Cobs is the observed concentration of dissolved CO, which is calculated according to Xie et al. (2002). Briefly, the initial CO concentration ({CO}eq in ml CO/ ml H2O) is calculated, assuming a mass balance:
2.2. CO analysis and incubation
fCOgeq ¼ 106 ma ðb p Vw þ Va Þ=Vw
Both atmospheric and seawater CO samples were immediately (normally less than half an hour) analyzed onboard after collection following the methods described in Xie et al. (2002) and Lu et al. (2010). Briefly, water samples were transferred from Niskin bottles into 50-ml acid-cleaned glass-only syringes using a Teflon tube coupled with a three-way Nylon valve. Sampling syringes were rinsed with sample water for three times, including one bubblefree flushing, before the final drawing. Surface water samples (average depth: 0.5 m) were analyzed as soon as being sampled with no delay to avoid any possible production or consumption. A precise ratio between sample water and CO-zero headspace air was set using a device described in Xie et al. (2002). Phase equilibration was established by shaking the sampling syringe for 4 min with a shaker (GJ-8175, Feihuang®, Guangdong, China). After that, headspace sample was injected into a RGA3 reduction gas analyzer (Trace Analytical Model ta3000R Gas Analyzer, Ametek®, USA). A 0.2 mm Nuclepore Teflon water-impermeable (13-mm diameter) filter fitted in a filter holder (Millipore®) was used here to prevent any liquid water from being injected into the analyzer. Atmospheric samples were injected into the analyzer immediately after sampling. Calibration was conducted every few hours with a commercial standard CO gas (nominal concentration: 299 ppbv in zerograde air, analytical accuracy: ±5%, DaTe®, Dalian, China) that was a certified reference material with the standard reference material No. 060152, and was approved by China State Bureau of Technical Supervision. In the incubation experiments, surface-water was incubated in a 200 ml acid-cleaned Al-foil wrapped all-glass syringe at a temperature ±1 C of surface water temperature. A 50 ml acid-cleaned glass syringe was used to get sub-samples from the 200 ml incubation syringe. The first sample was sampled and measured immediately after each incubation; and the second one was analyzed half an hour later after the first sample. The following subsamples were analyzed at appropriate intervals depending on CO consumption rate. The method precisions for repeated analyses of water samples, atmospheric samples, and incubation samples were about 5%, 3%, and 7%, respectively, in routine sample analysis.
where the transfer velocity k was calculated using the quadratic k/ wind speed relationship established by Wanninkhof (1992) (W92), and was adjusted by multiplying (ScCO/660)0.5, where ScCO is the Schmidt number for CO. The ScCO in seawater was calculated according to Zafiriou et al. (2008):
ScCO ¼ 0:0553T 3 þ 4:3825T 2 140:07T þ 2134
where ma is equilibrated headspace mixing ratio (ppmv); b is the Bunsen solubility coefficient of CO, which varies as a function of temperature and salinity (Wiesenburg and Guinasso, 1979); p is atmospheric pressure (atm) of dry air; Vw is the volume of water sample (ml); Va is the volume of the headspace air (ml). Cobs (nM) is calculated using the following equation:
. Cobs ¼ 109 p fCOgeq ðR TÞ where R is the gas constant (0.08206 atm L mol1 K1); and T is the temperature (K). Ceq is the air-equilibrated seawater CO concentration.
Ceq ¼ ðCOa bÞ=M where COa is the CO mixing ratio of the atmosphere (ppbv); M is the molar volume of CO at standard temperature and pressure, 25.094 L mol1 (Lide, 1992; Stubbins et al., 2006). 3. Results and discussion 3.1. Atmospheric CO Atmospheric CO mixing ratios in the investigated area ranged from 194.8 to 596.9 ppbv with an average of 390.4 ppbv for the first cruise, which was higher than those reported in the literature. For instance, Yang et al. (2010) reported that average CO mixing ratio was 297 ppbv in the overlapped study area in November 2007. Lam et al. (2000) reported that average CO mixing ratio in autumn was 313 ppbv at their monitoring site Cape D'Aguilar (22.2 N, 114.3 E, 60 m above sea level). Stubbins et al. (2006) found that mean CO mixing ratio was 151 ppbv over the ocean from Montevideo, Uruguay (35 S, 55 W) to Grimsby, UK (54 N, 0 W) in April and May 2000. The higher atmospheric CO mixing ratio observed during the
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
127
Fig. 1. Locations of the sampling stations in the Southern Yellow Sea for the first cruise (upper), with extended stations in the East China Sea for the second cruise (lower).
first cruise were probably due to the fact that the experienced high wind speed intensified terrestrial influence from the mainland of China, which is believed to be a major CO source in the study areas. In contrast, average atmospheric CO mixing ratio during the second cruise ranged from 53.5 to 238.5 ppbv with an average of
126.9 ppbv, which was lower than that obtained during the first cruise, and was even lower than that observed over the open sea (e.g., Stubbins et al., 2006). The low atmospheric CO mixing ratio during the second cruise might be attributed to the low wind speeds (average: 0.45 m/s), which reduced the terrestrial influence
128
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
at the same time reduced sea-to-air gas exchange. 3.2. Dissolved CO in seawater The CO data obtained from the surface water of the SYS, together with water depth (D), surface water temperature (T), salinity (S), wind speed (u), time of sampling (t), and sea-to-air flux measured during the two cruises, are listed in Tables 1 and 2, respectively. Surface seawater CO concentrations in the investigated area ranged from 0.10 to 7.33 nmol L1, with a mean of 1.91 nmol L1 for the first cruise, and from 0.09 to 11.8 nmol L1 with a mean of 2.52 nmol L1 for the second cruise, respectively. CO concentrations obtained in these two studies were much higher than those reported in Yang et al. (2010) in almost the same area in November 2007, when the surface water CO concentrations ranged from 0.13 to 2.30 nmol L1, with an average of 0.68 nmol L1. The difference in CO concentrations between the current study and Yang et al. (2010) could be due to many reasons, such as the different sampling seasons, since temperature (average sea surface temperature 27.3 vs. 21.2 C) was higher and light was stronger in present study (summer vs. winter) than those in Yang et al. (2010). High temperature and strong light would promote CO production in seawater (Conrad et al., 1982; Zhang et al., 2006). The difference might also be due to the different biological activities, which had been verified by the low microbial CO consumption rate constants measured in the same cruise (see section 3.4). From Tables 1 and 2, it can be seen that both average and maximum CO concentrations in the second cruise are higher than those in the first cruise. If we made an assumption that biological consumption, the major CO sink in the ocean, during the two cruises were in the similar level, the difference in CO concentrations might be due to the fact that the wind speeds during the first cruise (the mean wind speed was 5.6 m/s) were higher than those during the second cruise (the mean wind speed was 0.45 m/s). The higher wind speed resulted in a deeper mixed layer, a greater seato-air emission, and ultimately the lower surface seawater CO concentrations during the first cruise. CO concentrations were binned, based on the sampling time, into eight three-hour-long classes. Averaged concentration along
with standard deviation of each class was plotted against the time of a day to get average diurnal variations (Fig. 2). Fig. 2a and b show the diurnal variations for the first cruise and the second cruise, respectively. It is clear that CO diurnal variations of the two cruises showed a similar pattern, with daytime maximal values being 8e14 times higher than nighttime minimal values. CO concentrations remained basically constant throughout the night, indicating a dynamic production-consumption balance. The highest CO concentrations generally appeared from 15:00 to 18:00 (local time), which was several hours later than the solar radiation maximum. The huge ranges and distribution patterns observed in diurnal variations were caused by the competition between photoproduction and microbial consumption (Day and Faloona, 2009), which are believed to be the main source and sink of oceanic CO, respectively. It is interesting to find from Fig. 2a and b that nighttime CO concentrations during the two cruises seemed equal, but the daytime CO concentrations had a large disparity, with the second cruise exhibiting higher CO concentrations. The same surveying season and area, to some extent, ensured that the production and consumption were at a similar level. The daytime CO concentration difference possibly indicated that other processes such as sea-to-air emissions also played a role in controlling oceanic CO biogeochemical cycle. Other researchers have reported CO diurnal variation. Conrad et al. (1982) observed a diurnal variation in euphotic ocean water and claimed that the variation of CO concentrations was a product of photochemical production (source) of CO from the dissolved organic carbon (DOC), microbial oxidation, and sea-to-air flux (sink). Jones (1991) also investigated the diurnal variation of CO in seawater and found that diurnal variation could occur under sunny conditions. The variation factors of CO concentrations in the diurnal cycle were reported to be 5e7 by Stubbins et al. (2006). Our observations were in good agreement with the literature. 3.3. Sea-to-air flux of CO CO saturation factor (a), which is derived from Cobs divided by Ceq, ranged from 0.54 to 25.90, with an average of 6.15 during the
Table 1 Description of sampling stations along with depth (D), temperature (T), salinity, wind speed (u), time of sampling (t), surface [CO], CO mixing ratio (MR), CO saturation factor (a), and sea-to-air CO flux (F) in the Southern Yellow Sea during the first cruise. Station
Location
D (m)
T ( C)
S
u (m/s)
t
[CO] (nM)
MR (ppbv)
a
Flux (mmol/m2/d)
D9 C8 C6 C4 C2 C1 F1 F2 B1 B3 B5 B7 B9 B11 E2 A6 A4 A2 A1 G2 G1 G0 FB2 Mean
35.25E, 35.25E, 35.25E, 35.25E, 35.25E, 35.25E, 35.00E, 34.73E, 34.50E, 34.50E, 34.50E, 34.50E, 34.50E, 34.50E, 34.00E, 33.50E, 33.50E, 33.50E, 33.50E, 33.50E, 34.00E, 34.25E, 34.66E, NA
73.0 64.0 45.0 43.0 33.0 33.0 33.0 25.0 18.0 20.0 20.0 51.6 63.0 73.0 68.0 45.0 37.7 25.0 18.0 18.0 18.0 20.0 21.0 37.6
25.62 25.83 26.04 26.13 25.39 24.46 24.21 22.89 22.73 19.21 22.61 25.60 26.32 26.16 26.20 25.14 25.30 24.99 26.75 26.82 26.17 23.11 23.24 24.82
30.99 30.59 30.64 30.18 30.23 30.94 30.68 30.12 29.98 30.71 30.22 31.27 31.21 31.62 31.05 31.30 31.33 30.92 30.10 30.35 31.00 29.63 30.79 30.69
7.7 5.3 5.0 3.4 3.3 6.3 6.2 7.9 6.4 3.7 6.0 6.5 7.2 5.9 3.1 3.0 4.3 4.8 6.4 8.0 6.6 6.8 5.2 5.61
08:39 11:34 14:31 17:19 20:02 21:34 23:08 01:05 02:48 05:42 08:15 11:02 13:46 16:48 20:17 23:23 02:11 05:02 06:35 08:49 11:04 13:32 16:17 NA
1.27 2.32 1.12 1.91 0.70 0.50 0.10 0.66 0.15 0.84 1.86 4.95 4.41 3.64 3.42 1.36 1.07 0.33 0.28 0.80 0.92 3.95 7.33 1.91
344.3 368.0 244.6 240.3 287.6 271.5 240.9 194.8 233.1 310.4 521.0 596.9 581.7 559.2 456.9 421.1 555.7 397.4 364.5 350.1 477.4 577.2 384.0 390.4
4.99 8.57 6.23 10.8 3.28 2.51 0.54 4.63 0.85 3.68 4.84 11.34 10.32 8.84 10.20 4.39 2.62 1.11 1.04 3.11 2.61 9.30 25.90 6.15
6.53 6.27 2.57 2.20 0.57 1.26 0.34 3.28 0.11 0.78 5.36 20.65 22.77 12.35 3.26 1.01 1.32 0.08 0.05 3.88 2.72 16.64 19.51 5.76
NA: not available.
123.00N 122.50N 122.00N 121.50N 121.00N 120.75N 120.67N 120.58N 120.50N 121.00N 121.50N 122.00N 122.50N 123.00N 123.00N 123.00N 122.50N 122.00N 121.75N 121.50N 121.25N 121.00N 120.85N
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
129
Table 2 Description of sampling stations along with depth (D), temperature (T), salinity, wind speed (u), time of sampling (t), surface [CO], CO mixing ratio (MR), CO saturation factor (a), and sea-to-air CO flux (F) in the Southern Yellow Sea during the second cruise. Station
Location
D (m)
T ( C)
S
u (m/s)
t
[CO] (nM)
MR (ppbv)
a
Flux (mmol/m2/d)
A2 A4 A6 A8 B1 B3 B4 B6 D1 D2 D4 D6 D8 D9 E2 F0401 F0403 F0405 F0407 F0501 F0502 F0504 F0506 F0507 F0508 F0601 F0603 G3 G5 G7 H1 H2 Mean
35.99E, 35.67E, 35.33E, 35.00E, 35.00E, 35.00E, 35.00E, 35.00E, 34.50E, 34.50E, 34.50E, 34.50E, 34.50E, 34.50E, 33.83E, 31.63E, 31.97E, 32.33E, 32.71E, 31.50E, 31.42E, 31.25E, 31.06E, 30.96E, 30.84E, 32.46E, 31.67E, 33.50E, 33.50E, 33.50E, 33.00E, 32.50E, NA
30.9 41.5 55.3 73.1 37.4 36.1 46.3 61.5 18.2 19.8 21.0 51.8 68.8 68.8 68.2 34.0 41.5 43.8 80.4 26.4 42.0 48.7 50.5 60.2 72.6 92.5 68.1 28.9 45.0 69.0 19.8 25.1 48.4
23.9 26.1 26.8 27.3 30.1 30.8 26.2 27.4 24.8 23.1 26.8 25.9 27.7 27.6 27.8 27.1 26.9 28.9 29.1 26.5 27.7 28.7 29.6 28.8 28.9 28.7 29.0 25.5 26.9 27.8 25.8 26.1 27.3
30.79 30.47 30.45 30.78 29.81 30.81 30.77 30.82 27.63 28.99 30.29 30.33 31.44 31.31 31.20 21.41 32.58 30.34 28.79 26.98 26.17 32.64 33.70 33.75 32.17 28.91 28.61 31.58 30.94 31.24 31.06 30.97 30.24
0.2 0.5 0.4 0.3 0.6 0.1 0.8 0.9 0.5 0.3 0.3 0.3 0.5 0.4 0.4 0.2 0.6 0.6 0.5 0.3 0.5 0.5 0.7 0.5 0.4 0.4 0.6 0.9 0.4 0.1 0.5 0.2 0.5
00:35 04:37 09:13 14:01 01:19 23:16 21:26 17:58 06:17 07:39 11:22 14:59 18:44 20:25 01:17 00:44 05:34 10:51 16:40 02:22 00:29 19:30 14:03 10:48 07:34 19:48 01:08 11:13 07:42 03:58 15:47 18:54 NA
0.91 0.11 3.97 8.98 7.80 0.41 0.58 11.80 0.09 1.41 2.59 3.16 1.69 1.05 0.18 1.63 1.66 5.18 2.47 0.44 0.24 3.68 3.36 1.29 0.39 2.70 0.43 3.91 0.70 0.23 7.21 0.35 2.52
146.3 53.5 63.2 138.7 63.8 56.3 66.4 111.0 59.6 59.4 69.5 184.1 211.2 225.9 212.4 101.8 102.5 112.0 115.6 91.2 89.2 84.2 80.8 NA 87.1 NA 85.9 238.5 211.9 202.0 202.8 207.8 124.5
8.48 2.79 85.90 88.80 174.01 10.40 11.90 145.11 2.00 29.81 50.92 23.30 10.63 6.15 1.12 20.90 22.30 65.72 29.80 6.64 3.73 61.72 60.10 NA 6.38 NA 7.06 22.12 4.53 1.59 48.42 2.30 33.81
0.003 0.002 0.070 0.090 0.344 0.000 0.037 1.078 0.001 0.013 0.026 0.030 0.044 0.016 0.000 0.007 0.064 0.218 0.071 0.004 0.005 0.107 0.196 NA 0.006 NA 0.016 0.327 0.010 0.000 0.192 0.001 0.056
121.02N 121.67N 122.33N 123.00N 120.67N 121.33N 121.66N 122.33N 120.50N 120.67N 121.32N 122.00N 122.67N 123.00N 123.45N 122.66N 123.52N 124.42N 125.39N 122.33N 122.71N 123.61N 124.72N 125.24N 125.85N 125.85N 125.85N 122.33N 123.00N 123.67N 122.08N 122.15N
NA: not available.
first cruise, and from 1.12 to 174.3 with a mean of 33.82 during the second cruise. All surface water CO concentrations were supersaturated compared with atmospheric CO mixing ratios during the two cruises, except for those measured at stations F1 and B1 in the first cruise, where saturation factors were less than one. Water samples at these two stations were collected at night (Table 1) when CO concentration was low during a diurnal cycle (Fig. 2a and b). In general, the saturation factors during the second cruise were much higher than those during the first cruise, which was caused by the higher surface water CO concentrations and the lower atmospheric CO mixing ratios during the second cruise. The saturation factors measured in the current study were higher than those in previous study, for example, the average saturation factor in the second cruise was approximately 10 times the average reported in Yang et al. (2010). The reported saturation factors in the coastal and marginal seas ranged from 3 to 100 (Swinnerton at al., 1969, 1970; Seiler and Junge, 1970; Lamontagne et al., 1971; Conrad et al., 1982; Butler et al., 1987). The average saturation factors of both the first and second cruises fell within this reported range. However, the lowest value of the first cruise was less than the lower limit of the literature value; while the highest value of the second cruise was greater than the upper limit of the literature value. The high saturation factors of the second cruise indicated either high production of CO or decreased microbial consumption. Based on the method introduced by Stubbins et al. (2006) and Yang et al. (2010), average sea-to-air emissions were estimated to be 4.58 mmol m2 d1 (W92) for the first cruise, and 0.08 mmol m2 d1 (W92) for the second cruise, respectively. The sea-to-air flux during the first cruise was higher than those obtained from the North Atlantic Ocean (2.2 ± 1.5 mmol m2 d1) and
the North Global Ocean (1.9 ± 1.3 mmol m2 d1) (Stubbins et al., 2006). It was also greater than our previous result obtained in November 2007, which was 1.82 mmol m2 d1 by the W92 equation (Yang et al., 2010). Additionally, it is interesting to find that the average saturation factor in the second cruise is about 5 times greater than that in the first cruise, whereas the average sea-to-air flux of CO in the first cruise is around 60 times greater than that in the second cruise. These paradoxical distributions are caused by difference in wind speed that is quadratically related to gas transfer velocity (k). This comparison emphasizes the role played by wind speed when the W92 equation is used to calculate the sea-to-air flux. 3.4. Microbial consumption rate of CO In the second cruise, whole water incubation method (see Xie et al., 2005 for detail) was applied to determine microbial CO consumption rate. The incubation experiments showed that CO concentration decreased quickly at the beginning of incubation, and slowed down over time. Decay curves at most of the stations showed exponential decay patterns. Previous studies also reported first order exponential decay kinetic pattern. For example, Xie et al. (2005) reported that CO concentrations decreased exponentially with incubation time in the Delaware Bay, the NW Atlantic, and the Beaufort Sea. Yang et al. (2010) added additional supports to the exponential decay pattern with data collected in the East China Sea. However, at two stations A8 and F0507, as shown in Fig. 3, CO concentrations decreased linearly with incubation time. Linear decay patterns were reported in Xie et al. (2005) and Zhang (2008). Zhang (2008) pointed out that linear decay patterns were probably
130
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
Fig. 2. Variation of dissolved CO concentrations in surface seawater along with the time of day. (a). Diurnal variations of the first cruise. (b). Diurnal variations of the second cruise.
caused by special microbe species. Fig. 3 indicates that the microbial CO consumption rate constants, Kco, ranged from 0.07 to 0.83 h1, with an average of 0.33 h1. Our results are apparently lower than those previously measured in the East China Sea and the SYS (from 0.15 to 2.14 h1 with an average of 0.80 h1), but comparable to those obtained in open oceans. For instance, Xie et al. (2005) found that Kco ranged from 0.071 to 0.91 h1 with a mean value of 0.26 h1 in the NW
Atlantic. Day and Faloona (2009) reported similar microbial consumption rates (0.05e0.80 h1) in the California Current upwelling system. 4. Conclusions In this work, we reported the distribution, flux, and biological consumption of CO in the SYS. Some important conclusions were
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
drawn as follows. This studied area, which has a strong seasonal cycle, together with the apparently different wind-speeds experienced during the two cruises, offered us an opportunity to study the factors that could influence surface water CO concentration. Comparison
131
between these two cruises together with comparison with previous research results (Yang et al., 2010) revealed that surface water CO concentration followed seasonal, diurnal, and day-to-day variations. These variations were controlled jointly by a combination of light intensity, biological CO consumption, temperature, and wind-
Fig. 3. CO concentrations versus incubation time at selected incubation stations.
132
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
Fig. 3. (continued).
speed (depth of mixed layer and sea-to-air flux). Surface water CO concentration was also subject to strong spatial variations, as indicated by the large error bars in Fig. 2. Atmospheric CO mixing ratio in the studied area was dominated by terrestrial influence, and thus varied dynamically with wind direction and speed. The fact that both the lowest and the highest biological CO consumption
rate constants were out of reported range revealed the complexity of the studied area. The paradoxical distributions of saturation factor and sea-to-air flux observed in these two cruises emphasized the important role played by wind speed in calculating sea-to-air flux. Even though the sea-to-air CO flux in the second cruise was suppressed by the unusually low wind speed, the SYS was still an
W.-L. Wang et al. / Estuarine, Coastal and Shelf Science 163 (2015) 125e133
atmospheric CO source. Acknowledgments We thank the captain and crew of the R/V “Dong Fang Hong 2” for their support and help during the two cruises. This work was financially supported by the National Natural Science Foundation of China (No. 40976043), the Changjiang Scholars Programme, Ministry of Education of China and the Taishan Scholars Programme of Shandong Province, China. References Bates, T.S., Kelly, K.C., Johnson, J.E., Gammon, R.H., 1995. Regional and seasonal variations in the flux of oceanic carbon monoxide to the atmosphere. J. Geophys. Res. 100 (D11), 23093e23101. Butler, J.H., Jones, R.D., Garber, J.H., Gordon, L.I., 1987. Seasonal distributions and turnover of reduced traces gases and hydroxylamine in Yaquina Bay, Oregon. Geochim. Cosmochim. Acta 51, 697e706. Conrad, R., Seiler, W., 1980. Photo-oxidative production and microbial consumption of carbon monoxide in seawater. FEMS Microbiol. Lett. 9 (1), 61e64. Conrad, R., Seiler, W., 1982. Utilization of traces of carbon monoxide by aerobic oligotrophic microorganisms in ocean, lake and soil. Arch. Microbiol. 132 (1), 41e46. Conrad, R., Seiler, W., Bunse, G., Giehl, H., 1982. Carbon monoxide in seawater (Atlantic Ocean). J. Geophys. Res. 87 (C11), 8839e8852. Crutzen, P.J., Zimmermann, P.H., 1991. The changing photochemistry of the troposphere. Tellus 43AB, 136e151. Day, D.A., Faloona, I., 2009. Carbon monoxide and chromophoric dissolved organic matter cycles in the shelf waters of the northern California upwelling system. J. Geophys. Res. 114 (C01006) http://dx.doi.org/10.1029/2007JC004590. Derwent, R.G., 1995. Air chemistry and terrestrial gas emissions: a global perspective. Philos. Trans. R. Soc. Lond. Ser. A 351, 205e217. Dignon, J., Hameed, S., 1985. A model investigation of the impact of increase in anthropogenic NOx emissions between 1967 and 1980 on tropospheric ozone. J. Atmos. Chem. 3, 491e506. Feng, S.Z., Li, F.Q., Li, S.Q., 1999. Marine Science Introduction. Higher Education Press, Beijing, China, pp. 434e498 (in Chinese). Gnanadesikan, A., 1996. Modeling the diurnal cycle of carbon monoxide: sensitivity to physics, chemistry, biology, and optics. J. Geophys. Res. 101 (C5), 12177e12191. Johnson, J.E., Bates, T.S., 1996. Sources and sinks of carbon monoxide in the mixed layer of the tropical South Pacific Ocean. Glob. Biogeochem. Cycl. 10 (2), 347e359. Jones, R.D., 1991. Carbon monoxide and methane distribution and consumption in the photic zone of the Sargasso Sea. Deep Sea Res. 38 (6), 625e635. Kettle, A.J., 1994. A Model of the Temporal and Spatial Distribution of Carbon Monoxide in the Mixed Layer (M.Sc. thesis). Woods Hole Oceanographic Institution, Institute of Technology Joint Program in Oceanography, Woods Hole, Mass, 146pp. Kettle, A.J., 2005. Diurnal cycling of carbon monoxide (CO) in the upper ocean near Bermuda. Ocean. Model. 8, 337e367. Lam, K.S., Wang, T.J., Chan, L.Y., Wang, T., Harris, J., 2000. Flow patterns influencing the seasonal behavior of surface ozone and carbon monoxide at a coastal site near Hong Kong. Atmos. Environ. 35, 3121e3135. Lamontagne, R.A., Swinnerton, J.W., Linnenbom, V.J., 1971. Nonequilibrium of carbon monoxide and methane at the air-sea interface. J. Geophys. Res. 76, 5117e5121.
133
Lide, D.R., 1992. The Handbook of Chemistry and Physics. CRC Press, Ann Arbor. Liss, P.S., 1973. Processes of gas exchange across an airewater interface. Deep Sea Res. 20, 221e238. Lu, X.L., Yang, G.P., Wang, X.M., Wang, W.L., Ren, C.Y., 2010. Determination of carbon monoxide in seawater by headspace analysis. Chin. J. Anal. Chem. 38 (3), 352e356. Lü, X., Qiao, F., Xia, C., Wang, G., Yuan, Y., 2010. Upwelling and surface cold patches in the Yellow Sea in summer: effects of tidal mixing on the vertical circulation. Cont. Shelf Res. 30, 620e632. Najjar, R.G., Erickson III, D.J., Madronich, S., 1995. Modeling the air-sea fluxes of gases formed from the decomposition of dissolved organic matter: carbonyl sulfide and carbon monoxide. In: Zepp, R.G., Sonntag, C. (Eds.), The Role of Nonliving Organic Matter in the Earth's Carbon Cycle. Wiley & Sons, New York, pp. 107e132. Seiler, W., Junge, C., 1970. Carbon monoxide in the atmosphere. J. Geophys. Res. 75, 2217e2226. Stubbins, A., Uher, G., Kitidis, V., Law, C., Upstill-Goddard, R.C., Woodward, E.M.S., 2006. The open-ocean source of atmospheric carbon monoxide. Deep Sea Res. 53, 1685e1694. Swinnerton, J.W., Linnenbom, V.J., Cheek, C.H., 1968. A sensitive gas chromatographic method for determining carbon monoxide in seawater. Limnol. Oceanogr. 13 (1), 193e195. Swinnerton, J.W., Linnenbom, V.J., Cheek, C.H., 1969. Distribution of methane and carbon monoxide between the atmosphere and natural waters. Environ. Sci. Technol. 3, 836e838. Swinnerton, J.W., Linnenbom, V.J., Lamontagne, R.A., 1970. Ocean: a natural source of carbon monoxide. Science 167 (3920), 984e986. Thompson, A.M., 1992. The oxidizing capacity of the Earth's atmosphere: probable past and future changes. Science 256, 1157e1165. Wanninkhof, R., 1992. Relationships between wind speed and gas exchange over the ocean. J. Geophys. Res. 97 (C5), 7373e7382. Wiesenburg, D.A., Guinasso, N.L., 1979. Equilibrium solubilities of methane, carbon monoxide, and hydrogen in water and seawater. J. Chem. Eng. Data 24, 356e360. Xie, H., Andrews, S.S., Martin, W.R., Miller, J., Ziolkowski, L., Taylor, C.D., Zafiriou, O.C., 2002. Validated methods for sampling and headspace analysis of carbon monoxide in seawater. Mar. Chem. 77, 93e108. Xie, H., Zafiriou, O.C., Umile, T.P., Kieber, D.J., 2005. Biological consumption of carbon monoxide in Delaware Bay, NW Atlantic and Beaufort Sea. Mar. Ecol. Prog. Ser. 290, 1e14. Xie, H., Belanger, S., Demers, S., Vincent, W.F., Papakyriakou, T.N., 2009. Photobiogeochemical cycling of carbon monoxide in the southeastern Beaufort Sea in spring and autumn. Limnol. Oceanogr. 54 (1), 234e249. Yang, G.P., Wang, W.L., Lu, X.L., Ren, C.Y., 2010. Distribution, flux and biological consumption of carbon monoxide in the Southern Yellow Sea and the East China Sea. Mar. Chem. 122, 74e82. Yang, G.P., Ren, C.Y., Lu, X.L., Liu, C.Y., Ding, H.B., 2011. Distribution, flux, and photoproduction of carbon monoxide in the East China Sea and Yellow Sea in spring. J. Geophys. Res. 116, C02001. http://dx.doi.org/10.1029/2010JC006300. Yuan, D.L., Li, J.R., Hu, D.X., 2008. Cross-shelf circulation in the Yellow and East China Seas indicated by MODIS satellite observations. J. Mar. Syst. 70, 134e149. Zafiriou, O.C., Xie, H., Nelson, N.B., Najjar, R.G., Wang, W., 2008. Diel carbon monoxide cycling in the upper Sargasso Sea near Bermuda at the onset of spring and in midsummer. Limnol. Oceanogr. 53, 835e850. Zhang, Y., Xie, H., Chen, G.H., 2006. Factors affecting the efficiency of carbon monoxide photoproduction in the St. Lawrence estuarine system (Canada). Environ. Sci. Technol. 40, 7771e7777. Zhang, Y., 2008. Identity and Dynamics of the Microbial Community Responsible for Carbon Monoxide Oxidation in Marine Environments (Ph.D. thesis). Ocean University of China, Qingdao, pp. 125e155.