distribution, flux and photoproduction of carbon monoxide in the bohai and yellow seas

10

Click here to load reader

Upload: jian

Post on 06-Apr-2017

216 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

Marine Chemistry 168 (2015) 104–113

Contents lists available at ScienceDirect

Marine Chemistry

j ourna l homepage: www.e lsev ie r .com/ locate /marchem

Distribution, flux and photoproduction of carbon monoxide in the Bohaiand Yellow Seas

Bao-Zhen Zhao a,b, Gui-Peng Yang a,b,⁎, Huixiang Xie c, Xiao-Lan Lu a,b, Jian Yang a,b

a Key Laboratory of Marine Chemistry Theory and Technology, Ministry of Education, Ocean University of China, Qingdao 266100, Chinab College of Chemistry and Chemical Engineering, Ocean University of China, Qingdao 266100, Chinac Institut des sciences de la mer de Rimouski, Université du Québec à Rimouski, Rimouski, Québec G5L 3A1, Canada

⁎ Corresponding author at: College of Chemistry andUniversity of China, Qingdao 266100, China. Tel.: +8666782657.

E-mail address: [email protected] (G.-P. Yang).

http://dx.doi.org/10.1016/j.marchem.2014.11.0060304-4203/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 13 April 2014Received in revised form 30 September 2014Accepted 11 November 2014Available online 18 November 2014

Keywords:Carbon monoxideSea-to-air fluxPhotoproductionBohai SeaYellow Sea

The distribution, sea-to-air flux, and photoproduction of carbon monoxide (CO) were determined in the BohaiSea (BS) and the Yellow Sea (YS) during September, 2010. The concentrations of CO in the surface water variedfrom 0.09 to 6.81 nmol L−1 with an average of 1.05 nmol L−1. Surface water at most sampling stations wassupersaturated with CO with a mean saturation factor of 2.3. The hourly and daily sea-to-air fluxes of CO wereestimated to be 76.6 nmol m−2 h−1 and 0.65 μmol m−2 d−1, respectively. By extrapolation, the emission ofCO from the study area accounted for 0.1% of the global oceanic emission. Coupled optical–photochemicalmodeling based on measured spectral CO apparent quantum yields arrived at a total CO photoproduction of78.5 Gg CO-C yr−1 in the study area. Using CO as a proxy for dissolved inorganic carbon and biolabile carbonphotoproducts, the total photomineralization rate of dissolved organic matter was estimated to be 2.4 TgC yr−1, representing 2.0% of the primary production in the study area. Our results show that photomineralizationis a relatively small but significant term in the organic carbon cycle of coastal waters.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Carbonmonoxide (CO) has long been of biogeochemical interest be-cause of its key role in regulating atmospheric concentration of hydroxylradicals (•OH) (Derwent, 1995; Thompson, 1992; Yang et al., 2010).About 90% of atmospheric CO reacts with •OH to form carbon dioxide(CO2) (Fichot and Miller, 2010), which as a greenhouse gas contributesmost to the global warming. Simultaneously, the reaction of •OH withmethane, which is another more efficient greenhouse gas, is a majorpathway to remove the •OH from the atmosphere. Through competitionfor atmospheric •OH, CO indirectly affects the concentration of atmo-spheric methane and thus acts as an indirect greenhouse gas (Evansand Puckrin, 1995; Thompson, 1992).

The ocean has long been recognized as a source of atmospheric CO(Stubbins et al., 2006a; Yang et al., 2010; 2011), albeit with large uncer-tainties in its source strength.Marine emissions of CO are poorly definedand become increasingly controversial. The estimates of CO emissionsmade during the last several decades span two orders of magnitude,ranging from 4 to 600 Tg CO-C yr−1 (Bates et al., 1995; Conrad et al.,1982; Erickson, 1989; Khalil and Rasmussen, 1990; Prather et al.,2001; Stubbins et al., 2006a; Zuo and Jones, 1995). Little consideration

Chemical Engineering, Ocean532 66782686; fax: +86 532

of coastal and estuary areas, whose role on the global CO emissionshould not be neglected (Day and Faloona, 2009; Yang et al., 2011;Zafiriou et al., 2008), increases the uncertainty of CO emission inmarineenvironments.

CO is the second most abundant inorganic carbon product of chro-mophoric dissolved organic matter (CDOM) photochemistry (Millerand Zepp, 1995; Mopper and Kieber, 2000). The direct evidence for amajor photochemical CO source in seawater came from observationsof a pronounced diurnal cycle in sea-surface CO concentrations. CO con-centration in seawater shows complex spatial and temporal variationsas a result of interactions among photoproduction (Yang et al., 2011;Zafiriou et al., 2003; Zuo and Jones, 1995) and microbial consumption(Xie et al., 2005; Yang et al., 2010; Zafiriou et al., 2003), air–sea gas ex-change (Bates et al., 1995; Stubbins et al., 2006a; Yang et al., 2010; Zuoand Jones, 1995) and physical mixing (Gnanadesikan, 1996; Johnsonand Bates, 1996). Estimates of global open-ocean CO photoproductioncover an enormous range (30–820 Tg CO-C yr−1) (Moran and Zepp,1997; Valentine and Zepp, 1993; Zuo and Jones, 1995), with a relativelyprecise amount of 30–70 Tg CO-C yr−1 (Fichot and Miller, 2010;Stubbins et al., 2006b; Zafiriou et al., 2003). Moreover, CO has been sug-gested as a key proxy to evaluate the photoproduction of CO2 andbiolabile organic carbon (Miller and Zepp, 1995; Miller et al., 2002;Mopper and Kieber, 2000), which together account for the majority ofCDOM photodegradation products (Stubbins et al., 2006b). Thus amore precise quantification of CO photoproduction rates becomes in-creasingly important, particularly in coastal waters where only a few

Page 2: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

105B.-Z. Zhao et al. / Marine Chemistry 168 (2015) 104–113

studies on CO photoproduction have been conducted (Day and Faloona,2009; Yang et al., 2011; Zhang and Xie, 2012).

The Bohai Sea (BS) is a semi-enclosed shelf sea of the northwest Pa-cific Ocean, with a surface area of 7.7 × 104 km2 and an average depth of18.7 m. The area with a depth of less than 30 m constitutes 95% of thetotal area of the sea (Mao et al., 2008). The Yellow Sea (YS) is commonlyconsidered to be among the broadest marginal seas in the world (Yanget al., 2003). It is a semi-enclosed sea of the northwest Pacific Oceanwith a surface area of 38 × 104 km2 and an average depth of 44 m,surrounded by the west coast of the Korean Peninsula and the eastcoast of China and connected to the East China Sea (ECS) in the southand to the BS in the north (Sun, 2006).

To date, only a few studies have been conducted regarding CO inChina's marginal seas. Lu et al. (2010b) determined the concentrationsof CO in the North Yellow Sea (NYS) by headspace analysis. Yang et al.(2010 and 2011) reported distribution, flux, photoproduction, and bio-logical consumption of CO in the YS and the ECS. In this study we pres-ent data of CO from seawater and atmosphere samples covering almostthe entire area of the BS and YS (Fig. 1A). The objectives of this study areto understand the spatial distributions of CO, estimate the CO emissioninto the atmosphere, and quantify the photoproduction of CO in the BSand YS.

2. Methods

2.1. Sampling

Samples were collected in the BS and YS on board the R/V “DongFang Hong 2” during September 8–22, 2010. The locations of samplingstations are shown in Fig. 1A.

The surface seawater samples were taken with 12-L Niskin bottlesdeployed on standard conductivity–temperature–depth (CTD) rosettes.Fifty-milliliter acid-cleaned glass syringesfittedwith a three-wayTeflonvalve were used to sub-sample the Niskin bottles, following theprocedures recommended by Zafiriou et al. (2008) to minimize sam-pling artifacts. Once collected, the samples were immediately analyzedfor CO surface concentration ([CO]surf).

To minimize the influence of ship emissions, atmospheric COsamples were collected on the top deck of the ship, facing the wind,with 50 mL gastight glass syringes when the ship was decelerating.Then the samples were analyzed for CO atmospheric concentration([CO]atm).

Samples for dissolved organic carbon (DOC) and chlorophyll a (Chl-a)were immediately filtered through glass-fiber filters (Whatman GF/F,precombusted at 500 °C for 3 h). The filtrate for the analyses of DOCwas collected into the precombusted 40 mL glass vials and storedat−20 °C. The filters were stored at−20 °C for the analysis of Chl-a.

Samples for determining the apparent quantum yields (AQYs) of COphotoproduction were sequentially passed through 0.8-μm and 0.2-μmpolyethersulfonemembrane filters contained in an ArcoPak™ 500 Cap-sule filtration unit (Pall, USA), transferred into acid-cleaned 5-L clearglass bottles, and stored in the dark at 4 °C. After the cruise, the sampleswere brought back to a land-based laboratory in Qingdao and irradiatedwithin two months of sample collection.

2.2. Irradiation

Samples were re-filtered with 0.2-μm polyethersulfone membranes(Pall, USA), purged in the darkwith CO-free air tominimize backgroundCO concentration, and then siphoned into gastight quartz-windowedcells. The cell was filled from the bottom and overflowed with ~50 mLof sample before closing. The samples were aligned vertically in ablack, temperature-regulated aluminum block holder and irradiated ata constant temperature (24 ± 0.1 °C) using a SUNTEST CPS solar simu-lator equipped with a 1.5-kW xenon lamp (Atlas, Germany). SuccessiveSchott long band-pass cutofffilters ofWG280,WG295,WG305,WG320,

WG345, GG395, GG435 and GG495 (numbers are nominal 50% trans-mission cutoff wavelengths) were placed at the base of the irradiationchamber to obtain eight spectral treatments (Johannessen and Miller,2001; Zhang et al., 2006). Spectral irradiance under each cutoff filterwas measured with an ILT-900R UV–VIS spectroradiometer (Interna-tional Light Technologies, USA). Parallel incubations in the dark wereconducted to correct for potential dark production of CO.

2.3. Analysis

Both atmospheric and water samples were analyzed using a modi-fied ta3000 reduction gas analyzer (Trace Analytical, Ametek, USA).Water samples were preprocessed using a headspace method beforeanalysis (Lu et al., 2010b; Xie et al., 2002). Briefly, 6 mL of high-puritynitrogen was introduced into the sample-filled syringe with 44 mL ofseawater. The syringe was shaken for 5 min at a constant rotational ve-locity (120 r min−1) and then the equilibrated headspace air wasinjected into the gas analyzer through a syringe filter holder (Millipore)that contained a water-impermeable 0.2 μm Nuclepore Teflon filter toprevent liquid water from entering the chromatographic columns ofthe analyzer (Xie et al., 2002). Atmospheric samples were directlyinjected into the analyzer for quantification. The system was calibratedwith a gaseous CO standard (nominal concentration: 0.993 ppm byvolume in nitrogen; analytical accuracy: ±5%, State Center for StandardMatter, China). The CO standard was approved by China State Bureau ofTechnical Supervision and was a certified reference material with astandard reference material no. 060152.

Absorbance spectra of CDOMwere recorded from 200 to 800 nm at1-nm increments using a UV-2550 UV–VIS spectrometer (Shimadzu)fitted with 10 cm quartz cells and referenced to Milli-Q water. A base-line correction was applied by subtracting the absorbance value aver-aged over a 5-nm interval around 685 nm from all the spectral values(Babin et al., 2003). Absorption coefficients (a) were calculated as:

a ¼ 2:303� Ar

ð1Þ

where A is the absorbance, and r is the cell's light path length in meters.The concentration of dissolved organic carbon (DOC) wasmeasured

using a TOC-VCPH carbon analyzer (Shimadzu, Japan) calibrated withpotassium biphthalate. The DOC consensus reference materials (CRM)from the Hansell Laboratory (University of Miami, USA) were used asreference materials. The relative standard deviation was less than 2%.

The concentration of Chl-awas fluorometrically measured with aF-4500 fluorescence spectrophotometer (Hitachi, Japan) after filtrationof 300mLof seawater ontoWhatmanGF/F glassfiberfilter and extractionin 90% acetone according to Parsons et al. (1984).

Wind speeds were measured at a height of 10 m with a 27600-4Xship-borne weather instrument (Young, USA).

2.4. Calculation

2.4.1. CO fluxAccording to Xie et al. (2002), CO concentration in the initial surface

seawater samples, [CO]surf (nmol L−1), could be calculated as:

CO½ �surf ¼ pma βpVw þ Vað Þ= RTVwð Þ ð2Þ

wherema is the equilibrated headspacemixing ratio (ppbv); β ((mLCO)(mLH2O)−1 atm−1) is the Bunsen solubility coefficient of CO, which is afunction of temperature and salinity (Wiesenburg and Guinasso, 1979);p is the atmospheric pressure (atm); Vw is the water sample size (mL);Va is the volume of the headspace air (mL); R is the gas constant(0.08206 atm L mol−1 K−1), and T is the temperature (K).

Page 3: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

A

B

Fig. 1. The sampling stations (A) and concentrations of CO in the surface water ([CO]surf) (B) in the BS and YS. The open circle denotes surface seawater sampling in the daytime while thesolid circle indicates surface seawater sampling in the nighttime.

106 B.-Z. Zhao et al. / Marine Chemistry 168 (2015) 104–113

The [CO]eq (expected dissolved CO concentration in equilibriumwithseawater) was calculated using the equation of:

CO½ �eq ¼ CO½ �atm � β� �

=Vm ð3Þ

where [CO]atm stands for the COmixing ratio of the atmosphere (ppbv),and Vm stands for the molar volume of CO at standard temperature andpressure, 25.094 L mol−1 (Lide, 1992; Stubbins et al., 2006a).

The α (CO saturation factor) was calculated using the equation of:

α ¼ CO½ �surf = CO½ �eq: ð4Þ

Then, the sea-to-air flux of CO (F) (nmol m−2 h−1) was calculatedaccording to the following equations:

F ¼ k CO½ �surf− CO½ �eq� �

ð5Þ

Page 4: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

107B.-Z. Zhao et al. / Marine Chemistry 168 (2015) 104–113

where k is the gas transfer coefficient (m h−1). Here k was calculatedwith themethod of Edson et al. (2011), which provides amore accurateparameterization over a wind speed range from 0 to 18 m s−1 and alsoconsiders the influence of bubble-mediated exchange on gas transfer:

k ¼ 0:029u3 þ 5:4� �

� Sc=660ð Þ−1=2 ð6Þ

where u represents the wind speed (m s−1) measured at the height of10 m. The Schmidt number (Sc) of CO was obtained from the study ofZafiriou et al. (2008).

2.4.2. CO photoproductionThe apparent quantum yield of CO photoproduction is defined as the

number of CO molecules photochemically produced divided by thenumber of photons absorbed by CDOMat a givenwavelength. The num-ber of photons absorbed by CDOM at wavelength λ in the irradiationcells (Qa,λ, mol photons s−1 nm−1) was calculated according to themethod of Hu et al. (2002):

Qa;λ ¼ Q0;λ � acdom;λ=at;λ� �

� S� 1− exp −at;λ � l� �h i

ð7Þ

where S is the cross-section (m2) of the illuminated area and l is thepath-length (m) of the irradiation cells, Q0, λ is the photo flux (molphotons m−2 s−1 nm−1) just below the front window, and at, λ is thetotal absorption coefficient (m−1), the sum of the absorption byCDOM (acdom, λ, m−1) and water (aw, λ, m−1). Values of aw, λ weretaken from Buiteveld et al. (1994) and Pope and Fry (1997). A Matlab-coded, iterative curve-fitting method was employed to derive the COAQY spectra (Johannessen and Miller, 2001; Zhang et al., 2006;Ziolkowski, 2000), assuming that the AQY decreases with wavelengthin a quasi-exponential manner (Zhang et al., 2006):

AQY λð Þ ¼ m1 � expm2

m3 þ λ

� �ð8Þ

where m1, m2 and m3 are the fitting parameters and λ represents thewavelength (nm).

Then the CO photoproduction rate (PCO (μmol (CO) m−2 d−1)) inthe surface water was calculated using the equation modified fromZafiriou et al. (2003):

PCO ¼Z600

280

irradiance� attention1þ2 � acdom;λ=at;λ � AQY λð Þdλ ð9Þ

where irradiance is the global spectral solar irradiance (molphotons m−2 d−1 nm−1), and attenuation1+ 2 denotes the correctionfactors for reflection of light by cloud and water surface (0.69 and0.94, Yang et al., 2011). Herewe adopted the SimpleModel of the Atmo-spheric Radiative Transfer of the Sunshine spectral irradiance (SMARTS)(Gueymard, 1995; 2001) to derive the spectral solar irradiance, cover-ing the full range of photochemically active radiation (280–600 nm)with a spectral resolution of at least 0.5 nm.

3. Results and discussion

3.1. Atmospheric CO

Atmospheric mixing ratios ([CO]atm) of the BS and YS varied from168 to 2740 ppbv with an average of 642 ppbv (SD = 433 ppbv, n =61), as shown in Table 1 and Fig. 2C. The highest CO mixing ratio(2740 ppbv) was observed at station 45 beside the oil-drilling platformin the Bohai Bay,while the lowest ratio (168ppbv) appeared at station 8far away from land. The highest CO mixing ratio was 16 times higherthan the lowest one as a result of the effect of human activities. The at-mospheric CO mixing ratios displayed a decreasing trend from inshore

to offshore. A similar phenomenon was found in the YS and the ECSby Yang et al. (2010 and 2011).

In a previous study in the same area from 21 April to 4 May 2010,Zhang (2011) found that the atmospheric CO mixing ratios rangedfrom 215 to 850 ppbv with an average of 414 ppbv (SD = 140 ppbv,n = 69). Obviously, our results were on average 55% higher thanthose reported by Zhang (2011), probably due to seasonal variations.The similar phenomenonwas observed byXie et al. (2009) in the south-eastern Beaufort Sea in spring and autumn.

Yang et al. (2010) reported that the atmospheric COmixing ratios inthe South Yellow Sea (SYS) and the ECS ranged from 165 to 671 ppbvwith a mean value of 297 ppbv (SD = 133 ppbv, n = 23) in lateautumn,whichweremuch lower than ourmeasurements due plausiblyto the different study areas. The SYS (mean value: 423 ppbv) exhibitedhigher CO mixing ratios than the ECS (mean value: 252 ppbv) (Yanget al., 2010). Based on the result of Zhang (2011) (mean value:941 ppbv in the BS, N339 ppbv in the NYS, N333 ppbv in the SYS) andour own data (mean value: 1019 ppbv in the BS, N686 ppbv in theNYS, N462 ppbv in the SYS), it could be found that the BS displayedhigher COmixing ratios than the YS (theNYS N the SYS) both in autumnand spring. Therefore, the spatial distributions of CO mixing ratiosover the China seas in autumn displayed the following order: theBS N the YS N the ECS.

3.2. Dissolved CO in surface seawater

Concentrations of CO in the surface water ([CO]surf) ranged from0.09 to 6.81 nmol L−1 with an average of 1.05 nmol L−1 (SD =1.18 nmol L−1, n = 61) (Table 1, Figs. 1B and 2C). The maximum con-centration was observed at station 19 probably due to the effect ofthe Yangtze River diluted water and the sampling time (10:59). TheYangtze River diluted water might bring a great amount of terrestrialCDOM which had been demonstrated to be more efficient at photo-chemically producing CO than marine CDOM (Yang et al., 2011; Zepp,2003; Zhang et al., 2006). The station 19 sampled in daytime (10:59)relative to stations 25 and 26 closer to the Yangtze River estuary sam-pled in nighttime (18:12 and 20:38) might have much higher irradi-ance. That enhanced the photoproduction of CO. The minimumconcentrationwas observed at station 50 near the Yellow River estuary.The runoff of the YellowRiver carried lots of silt, and thus thehigher tur-bidity in the estuary (The Ministry of Water Resources of the People'sRepublic of China, 2010; Yu, 2012; Zhai et al., 2005) reduced thephotoproduction of CO. Additionally, the sampling time at this stationwas nighttime, which inhibited the process of photoproduction. Whatis more, the difference of wind speeds between station 19 (1.2 m s−1)and station 50 (7.8 m s−1) (Table 1, Fig. 2D) increased the differenceof those concentrations. However, the DOC concentration at station 19(1.90mg C L−1) (Table 1, Fig. 2B) was slightly lower than that at station50 (1.96 mg C L−1) (Table 1, Fig. 2B), demonstrating that high surfaceCO concentration might not be attributed to high DOC concentration.This observation is in agreement with previous study by Yang et al.(2011).

Our results agreedwell with the surface seawater CO concentrationsavailable in the literature. Zhang and Xie (2012) reported that the sur-face CO concentrations of the upper and the lower estuary of the St.Lawrence in autumn were 1.70 and 1.10 nmol L−1, respectively. Ourvalues are a little lower than those observed by Zhang (2011) in thesame area (the BS and YS) in spring, where the surface CO concentra-tions ranged from 0.19 to 3.57 nmol L−1 with an average of1.24 nmol L−1 (SD = 0.79 nmol L−1, n = 69). Xie et al. (2009) andZhang and Xie (2012) also found that the surface CO concentrationswere higher in spring than in autumn in the southeastern BeaufortSea and the St. Lawrence estuary.

In order to evaluate the effect of sampling time on the surface COconcentration,meandiurnal CO concentrationswere derived by poolingthe mean concentrations from all 1-h sections (Fig. 3) (Stubbins et al.,

Page 5: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

Table 1Environmental conditions of stations sampled for atmosphere [CO]atm and surface seawater [CO]surf along with the fitting parameters for AQY.

Station Time Watertemperature (°C)

Salinity(‰)

Wind speed(m/s)

Chl-a(μg L−1)

DOC(mg C L−1)

[CO]atm(ppbv)

[CO]surf(nmol L−1)

a350(m−1)

SUVA254

(m2 g−1 C)AQY (mol CO mol photons−1)

m1 m2 m3

1 2157 23.697 30.345 7.5 0.62 1.94 325 0.29 0.47 1.87 2.641E−10 2584.8 −107.32 2347 23.751 30.092 9.2 0.55 2.55 227 0.193 0418 23.925 30.554 7.0 0.56 2.84 228 0.214 0854 23.973 29.999 2.9 0.63 1.75 279 1.17 0.38 1.82 2.656E−10 2567.9 −107.55 1305 24.968 30.100 10.3 0.38 1.77 420 2.706 1614 26.644 31.224 10.6 0.61 1.96 288 1.947 2005 27.434 31.544 8.6 0.30 Null 193 0.398 2316 27.497 31.383 8.2 0.41 1.93 168 0.559 0222 24.499 30.690 2.5 0.49 2.45 316 0.2410 0618 24.715 30.291 Null 0.59 1.70 507 0.45 0.31 1.94 2.654E−10 2526.3 −107.711 1508 24.335 30.324 6.0 0.84 1.89 418 0.82 0.37 1.73 2.662E−10 2467.2 −107.612 1122 24.758 30.305 6.4 1.10 2.01 334 0.9113 0747 24.279 30.852 7.8 0.80 1.84 609 0.67 0.52 2.16 2.625E−10 2577.5 −107.814 0425 24.321 30.894 6.8 0.82 Null 430 0.3315 0126 26.255 29.453 5.1 1.09 1.96 426 0.36 0.82 2.54 2.647E−10 2500.7 −107.816 2124 26.291 29.908 7.3 0.53 2.17 213 0.2517 1729 22.644 30.861 0.3 1.52 1.89 444 0.75 0.43 1.91 2.671E−10 2429.2 −107.818 1419 27.988 30.466 2.5 1.49 1.72 881 1.4819 1059 27.521 30.026 1.2 1.81 1.90 989 6.8120 0737 26.239 29.785 3.9 4.16 1.89 1155 0.8621 0515 25.974 29.978 6.9 4.95 1.76 807 0.2822 0820 25.971 31.026 6.7 0.51 1.95 402 0.49 0.35 1.67 2.630E−10 2551.9 −108.023 1144 24.032 30.856 3.7 1.12 1.93 409 0.4624 1337 25.208 30.744 2.6 0.94 1.77 514 1.39 0.52 1.97 2.636E−10 2538 −108.225 1812 25.766 30.269 1.5 1.29 1.71 419 0.4226 2038 24.91 30.888 2.4 1.53 1.47 348 0.41 0.53 1.96 2.642E−10 2571.9 −107.727 2342 25.239 31.067 2.3 3.12 1.80 337 1.8228 0217 25.302 30.774 3.7 1.33 1.76 439 0.50 0.38 1.50 2.621E−10 2581.4 −107.829 0453 26.861 28.611 4.8 2.83 1.92 876 0.6530 0528 24.745 30.189 2.1 0.30 1.38 1606 1.11 0.35 2.29 2.644E−10 2548.6 −107.931 0710 25.022 30.682 4.8 0.82 1.55 777 1.0632 0922 24.282 30.661 3.3 0.89 1.54 844 1.3033 1105 23.529 30.599 1.1 2.52 1.51 965 2.0334 1645 25.95 30.904 0.9 0.45 1.55 638 2.15 0.34 2.32 2.643E−10 2476.1 −108.335 1824 26.102 31.182 1.4 Null 1.41 448 0.8036 1958 25.573 31.118 0.6 0.63 1.26 554 0.5637 2124 26.257 31.293 2.1 0.29 1.28 656 0.4638 2248 25.628 31.338 6.3 0.49 2.22 602 4.86 0.29 1.25 2.659E−10 2542 −107.539 0412 22.536 27.017 1.6 7.42 1.53 688 0.57 0.85 3.33 2.637E−10 2492.5 −107.640 0112 23.124 20.381 1.2 4.42 1.39 669 0.2241 2241 23.957 28.467 3.6 1.39 1.72 517 0.6142 2023 22.465 30.747 3.7 6.81 1.78 567 1.0043 1858 23.907 30.595 4.7 1.71 1.88 706 0.42 0.72 2.92 2.675E−10 2360.2 −107.744 0829 21.761 30.845 6.5 0.85 2.13 379 0.19 1.15 3.56 2.654E−10 2481.4 −107.745 0915 21.016 31.058 1.9 0.54 1.97 2740 3.7246 1111 20.67 30.693 0.8 0.47 1.76 2028 3.3547 1250 19.874 30.191 2.2 0.91 1.52 1139 1.6648 1421 21.655 29.214 0.9 2.84 1.50 879 1.18 0.83 3.53 2.662E−10 2338.1 −107.949 2305 21.488 30.872 8.1 10.15 2.32 340 0.2450 2139 22.581 30.748 7.8 1.26 1.96 398 0.09 0.74 2.93 2.671E−10 2420 −108.551 1529 24.131 28.371 12.0 4.27 1.96 481 0.2752 1409 24.748 28.726 11.6 3.27 2.27 426 0.6853 1258 25.039 29.003 9.4 1.56 2.14 525 0.6254 1034 25.188 30.746 5.4 1.58 1.85 1100 0.80 0.70 2.85 2.677E−10 2385 −108.155 1040 24.03 26.778 2.6 3.56 1.66 1106 2.6856 1411 24.299 30.305 2.8 0.94 1.76 1046 1.2557 0026 22.637 28.683 5.5 1.97 1.65 645 0.6058 0210 22.701 27.858 5.2 4.72 1.75 699 0.6259 0411 23.335 30.162 3.2 1.09 1.42 566 0.4160 0723 23.705 29.676 3.4 2.29 1.60 625 0.71 0.53 2.59 2.663E−10 2541 −107.461 1727 23.875 29.911 12.1 1.86 1.86 397 0.20

Null was not measured.

108 B.-Z. Zhao et al. / Marine Chemistry 168 (2015) 104–113

2006a). The surface CO concentrations in the daytime (mean value:1.18 nmol L−1) were far higher than those in the nighttime (meanvalue: 0.59 nmol L−1) due to the CO photoproduction. Therefore, thesampling time has an important influence on the surface CO concentra-tion. Particularly, the maximum concentration sampling in the daytime(6.81 nmol L−1 at station 19) was higher than that sampling in thenighttime (4.86 nmol L−1 at station 38), while theminimumconcentra-tion sampling in the nighttime (0.09 nmol L−1 at station 50) was lower

than that sampling in the daytime (0.19 nmol L−1 at station 44). The COphotoproductionmight lead to the difference. Theminimum concentra-tion in the daytime (08:29)might be due to the beginning of productionafter an overnight microbial consumption and a high wind speed(6.5 m s−1, Table 1, Fig. 2D). The maximum concentration sampling inthe nighttime might attribute to low microbial consumption.

To evaluate the contribution of the variations of temperature andsalinity to CO concentration, variations of temperature-driven and

Page 6: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

20

22

24

26

28

30

Wate

r T

em

pera

ture

(oC

)

15

20

25

30

35

Sali

nit

y

A

0

3

6

9

12

Ch

l-a

(�g

L-1

)

1

2

3

4

DO

C (

mg

C L

-1 )

B

0

2

4

6

8

[CO

] surf (

nm

ol

L-1

)

0

2

4

6

8

10

12

Satu

rati

on

facto

r

C

0

500

1000

1500

2000

2500

3000

[C

O] a

tm (

pp

bv

)

0 5 10 15 20 25 30 35 40 45 50 55 60 65

0

200

400

600

800

1000

Flu

x (

nm

ol

m -2

h-1

)

Station Number

0

3

6

9

12

15

Win

d s

peed

(m

s-1

)

D

Fig. 2. Seawater temperature (solid square) and salinity (open circle) (A), chlorophyll a (solid square) and dissolved organic carbon (DOC) (open circle) (B), concentrations of CO inatmosphere (open circle) and in surface seawater (solid square) and saturation factor (solid triangle) (C), sea-to-air flux of CO (solid square) and wind speed (open circle) (D) in theBS and YS.

109B.-Z. Zhao et al. / Marine Chemistry 168 (2015) 104–113

salinity-driven CO concentrationwere calculatedwith themethod of Luet al. (2010a). However, temperature-driven and salinity-driven varia-tions only accounted for 4.2% and 0.7% of the CO concentration in thestudy region, respectively.

Linear regression analyses were processed between the surface COconcentrations and other environmental parameters such as Chl-a andDOC with an attempt to identify the possible factors which directlyaffected the variation of sea surface CO. However, no significant relation-ships appeared between the concentrations of CO and these parameters,which might be attributed to the complex environmental conditionsin the study area. Yang et al. (2011) also reported similar results inthe ECS and the YS.

3.3. Sea-to-air flux of CO

CO atmost stations (about 80%)was supersaturatedwith the satura-tion factor varying from 0.3 to 11.0 (mean: 2.3, SD = 2.3, n = 61)(Fig. 2C). The majority of the undersaturated stations were sampled atnight when the CO concentrations were lower than their local 24-h av-erages. Regionally, the majority of the undersaturated stations were lo-cated in the BS and NYS, due to higher atmospheric CO mixing ratios.Additionally, the undersaturation of CO at some stations (e.g., 49 and50) (Fig. 1A) was mainly due to the low surface CO concentrations,which resulted from the inhibition of photoproduction because oflight limit or/and high turbidity. The mean saturation factor in this

Page 7: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 01 02 03 04 05-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0Night Mean=0.59Day Mean=1.18

[CO

] surf (

nm

ol

L-1

)

Time

Fig. 3. Variation in the average CO concentrations ([CO]surf) with the time. Error bar stands for mean standard deviation of the average concentration in each hour.

110 B.-Z. Zhao et al. / Marine Chemistry 168 (2015) 104–113

study was similar to that reported by Yang et al. (2010) for the SYS andthe ECS in November (ranging from 0.6 to 15.4 with a mean of 3.4), butmuch lower than that of Yang et al. (2011) for the YS and ECS in May(from 1.1 to 50.4 with a mean of 16.0) as well as that of Zhang andXie (2012) for the St. Lawrence estuary in autumn (upper estuary:8.05; lower estuary: 8.11).

Hourly fluxes were calculated based on spot wind speeds and COconcentrations in the atmosphere and surface water. The sea-to-airflux ranged from −63.2 to 1020.9 nmol m−2 h−1, with an average of76.6 nmol m−2 h−1 (SD = 189.8 nmol m−2 h−1, n = 60) (Fig. 2D),exhibiting a considerable spatial variability. Because of theundersaturation, CO fluxes in several stations revealed negative values.However, the fluxes atmost stations (about 80%)were positive, demon-strating that the BS and YS were a net source of atmospheric CO inautumn.

The daily flux (μmolm−2 d−1) of CO in the BS and YSwas calculatedusing the method reported by Stubbins et al. (2006a) and Yang et al.(2011), based on the hourly flux (nmol m−2 h−1). These hourly fluxeswere split into individual days and hourly fluxes plotted against timeof day. Days without full coverage of diurnal variations were discardedleaving 4 d of data. The area under each daily curve was then calculated(OriginPro 8) to give daily CO emissions. Rates generated in this wayranged from 0.10 to 2.06 μmol m−2 d−1 with an average of0.65 μmol m−2 d−1 (SD = 0.81 μmol m−2 d−1, n = 4). These datawere consistent with those of Xie et al. (2009), which was0.35 μmol m−2 d−1 for the Mackenzie shelf and Amundsen Gulf areain autumn. Our fluxes were much lower than those of Yang et al.(2011),which ranged from1.23 to 18.60 μmolm−2 d−1with an averageof 6.67 μmol m−2 d−1 because of the lower saturation factors in thepresent study. In addition, our valueswere also lower than those report-ed by Zhang (2011) in the same area, where the daily flux of CO rangedfrom−2.45 to 10.82 μmol m−2 d−1 with an average of 2.13 μmol m−2-

d−1. This might be attributed to the lower saturation factors as well asthe lower wind speeds in this study.

Based on the daily CO emission rates, the whole BS and YS wereestimated to emit 0.3 Gg CO-C in the autumn. This estimate was consid-erably lower than the previously reported value in the spring (1.1 GgCO-C) (Zhang, 2011), which might be due to seasonal variations in theseawater and atmospheric CO concentrations. If the seasonal variationswere ignored, the annual CO flux estimate would be 2.8 Gg CO-C on thebasis of the result of Zhang (2011) and our data, accounting for approx-imately 0.1% of the global oceanic CO emission of 3.7 Tg CO-C yr−1

(Stubbins et al., 2006a). However, the CO emissionmay have significant

seasonal variations, further investigation will be required in differentseasons in order to improve the annual CO flux estimate.

3.4. Photoproduction of CO

AQY spectra for CO photoproduction are shown in Fig. 4 and the cor-responding AQY parameters are given in Table 1. Our CO AQY spectra inthe BS and YS are comparable with those obtained from other estuarineand fresh waters (Fig. 5). Using 330 nm as an example, the values of COAQY were 2.7 × 10−5 mol CO mol photons−1 in the Delaware Estuary(White et al., 2010) and 2.5 × 10−5mol COmol photons−1 in the Pacific(Zafiriou et al., 2003), in good agreement with our results in the BS andYS (2.2 × 10−5 mol CO mol photons−1). Across the wavelength range(280–360 nm), our CO AQY spectra are higher than those in the Pacific(Zafiriou et al., 2003). The seawater in the study area contained moreterrestrial CDOM than Pacific water. These observations suggestedthat the two precursors (terrestrial CDOM and marine CDOM) had dif-ferent efficiencies at photochemically producing CO. The terrestrialCDOM was more prone to photolysis than marine algae-derived one,as discussed by Yang et al. (2011), Zhang et al. (2006) and Zuo andJones (1997). At the wavelength N360 nm, CO AQY in this study areawas similar to that in the Delaware Estuary (White et al., 2010).

Following previous studies (Zafiriou et al., 2003; Ziolkowski andMiller, 2007), we fitted our individual AQY spectra to a compositeequation (Fig. 6), as shown below:

AQY λð Þ ¼ 2:627� 10−10 � exp2505:6

λ−108:5

� �r ¼ 0:9999; p b 0:01: ð10Þ

Using the SMARTS2-derived irradiances and the AQYs describedabove, we estimated the CO photoproduction rate in the study area tobe 23.8 μmol m−2 d−1. This result is much lower than those obtainedby Zhang (2011) (54.6 μmol m−2 d−1) in the BS and YS in spring andYang et al. (2011) (50.8 μmolm−2 d−1) in the ECS and the YS in spring.However, thedifference in the rates between the spring and the autumnwas not as large as that in AmundsenGulf (45.8 μmolm−2 d−1 in springand 3.0 μmol m−2 d−1 in autumn, respectively) (Xie et al., 2009). Incomparison, the photoproduction rate of COwas almost 37 times higherthan its sea-to-air flux (0.65 μmol m−2 d−1) in the study area. Theresult indicated that in the BS and YS the major sink of CO might bemicrobial consumption rather than the sea-to-air emission in theautumn. According to the area of the BS and YS, the CO photoproductionwas calculated to be 12.0 Gg CO-C in autumn. If the seasonal variations

Page 8: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

250 300 350 400 450 500 550 600 650

1E-8

1E-7

1E-6

1E-5

1E-4

1E-3

01

04

10

11

13

15

17

22

24

26

28

30

34

38

39

43

44

48

50

54

60 (nm)

AQ

Y (

mo

l C

O m

ol

ph

oto

ns

-1)

Average

λ

Fig. 4. CO AQY spectra in the BS and YS.

111B.-Z. Zhao et al. / Marine Chemistry 168 (2015) 104–113

were ignored, the annual CO photoproduction in the BS and YSwould be78.5 Gg CO-C on the basis of the result of Zhang (2011) and our data,accounting for about 0.2% of the global oceanic CO photoproduction(40.6 Tg CO-C yr−1) (Fichot and Miller, 2010). This result implied thesignificant role of the study area in global oceanic CO photochemicalprocess.

3.5. Implications for the coastal carbon cycle

CO is the second most abundant and the most precisely measuredCDOM product, and can be used as a proxy for other major photoprod-ucts like dissolved inorganic carbon (DIC) and biolabile organic carbon.Miller and Zepp (1995) reported that DIC is produced fromCDOMabout

250 300 350 400

1E-8

1E-7

1E-6

1E-5

1E-4

1E-3

AQ

Y (

mo

l C

O m

ol

ph

oto

ns

-1)

G

Z

W

T

Fig. 5. Comparison of AQY spectra in this study with previously published AQY spectra. The AQfrom Zafiriou et al. (2003) and that for estuary is from White et al. (2010).

20 times as fast as CO in coastal waters near theMississippi River. MillerandMoran (1997) found that themolar ratio of DIC to CO production is15 and that of DIC to biolabile organic carbon released from thephotochemistry of CDOM is approximately 1.

Adopting the ratios reported by Miller and Moran (1997), we esti-mated the total photomineralization of DOC in the BS and YS to be2.4 Tg C yr−1, occupying 2.0% of the mean annual primary productionof this system (122.0 Tg C yr−1) (Zhu et al., 1993). Our result agreeswell with a previous estimate that the photomineralization of DOCaccounted for 2.5% of the primary production in the shelf waters ofthe northern California upwelling system (Day and Faloona, 2009).Our study suggests that photomineralization plays a significant role inthe regional carbon cycle in the BS and YS.

450 500 550 600 650

(nm)

ao and Zepp (1998) S=0

afiriou et al. (2003) Open Ocean

hite et al. (2010) S=21

his Study S=27.0-31.3 T=24 oC

λ

Y spectrum for freshwater is from Gao and Zepp (1998) while that for Pacific blue water is

Page 9: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

250 300 350 400 450 500 550 600 650

1E-8

1E-7

1E-6

1E-5

1E-4

1E-3

AQ

Y (

mo

l C

O m

ol

ph

oto

ns

-1)

(nm)

AQY=2.627e-10*exp(2505.6/( -108.5))

r

2=0.9999 p<0.01

λ

Fig. 6. The equation fitted between the mean AQY and λ in the BS and YS.

112 B.-Z. Zhao et al. / Marine Chemistry 168 (2015) 104–113

4. Summary and conclusions

This study reports the distribution, sea-to-air flux and photo-production rate of CO in the BS and YS. The atmospheric CO mixingratio ranged from 168 to 2740 ppbv (mean: 642 ppbv) and revealed thefollowing order: the BS N the YS N the ECS in autumn. Surface water COconcentration varied from 0.09 nmol L−1 to 6.81 nmol L−1 with anaverage of 1.05 nmol L−1. CO was supersaturated at most samplingstations and the BS and YS were a net source of atmospheric CO. Ourresults suggest that coastal areas should be counted in assessing theglobal oceanic CO emission.

The photoproduction rate of CO in the BS and YS in the autumn wasestimated to be 23.8 μmol m−2 d−1. Using CO as a proxy for DIC andbiolabile organic carbon photoproduction, the total photomineralizationrate of DOC was estimated to be 2.4 Tg C yr−1, accounting for 2.0% ofthe mean annual primary production in the BS and YS. Our resultindicates that the photochemistry of CO is a significant term in the organiccarbon cycle of the study area.

Acknowledgments

We thank the captain and crew of the R/V “Dong Fang Hong 2” fortheir help and cooperation during the field investigation. This workwas supported by the National Natural Science Foundation of China(Grant Nos. 40976043 and 41320104008), the Changjiang ScholarsProgram, Ministry of Education of China, and the Taishan ScholarsProgram of Shandong Province.

References

Babin, M., Stramski, D., Ferrari, G.M., Claustre, H., Bricaud, A., Obolensky, G., Hoepffner, N.,2003. Variations in the light absorption coefficients of phytoplankton, nonalgal parti-cles and dissolved organic matter in coastal waters around Europe. J. Geophys. Res.108 (C7). http://dx.doi.org/10.1029/2001JC000882.

Bates, T.S., Kelly, K.C., Johnson, J.E., Gammon, R.H., 1995. Regional and seasonal variationsin the flux of oceanic carbon monoxide to the atmosphere. J. Geophys. Res. Atmos.100 (D11), 23093–23101.

Buiteveld, H., Hakvoort, J.M.H., Donze, M., 1994. The optical properties of pure water. In:Jaffe, J.S. (Ed.), SPIE Proceedings on Ocean Optics XII. The Society of Photo-Optical In-strumentation Engineers, pp. 174–183.

Conrad, R., Seiler, W., Bunse, G., Giehl, H., 1982. Carbon monoxide in seawater (AtlanticOcean). J. Geophys. Res. Oceans 87 (C11), 8839–8852.

Day, D.A., Faloona, I., 2009. Carbon monoxide and chromophoric dissolved organic mattercycles 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. London, Ser. A 351, 205–217.

Edson, J.B., Fairall, C.W., Bariteau, L., Zappa, C.J., Cifuentes-Lorenzen, A., Mcgillis, W.R.,Pezoa, S., Hare, J.E., Helmig, D., 2011. Direct covariance measurement of CO2 gastransfer velocity during the 2008 Southern Ocean Gas Exchange Experiment: windspeed dependency. J. Geophys. Res. 116 (C00F10). http://dx.doi.org/10.1029/2011JC007022.

Erickson III, D.J., 1989. Ocean to atmosphere carbon monoxide flux: global inventory andclimate implications. Global Biogeochem. Cycles 3, 305–314.

Evans, W.F.J., Puckrin, E., 1995. An observation of the greenhouse radiation associatedwith carbon monoxide. Geophys. Res. Lett. 22 (8), 925–928.

Fichot, C.G., Miller, W.L., 2010. An approach to quantify depth-resolved marine photo-chemical fluxes using remote sensing: application to carbon monoxide (CO)photoproduction. Remote Sens. Environ. 114, 1363–1377.

Gao, H.Z., Zepp, R.G., 1998. Factors influencing photoreactions of dissolved organic matterin a coastal river of the southeastern United States. Environ. Sci. Technol. 32,2940–2946.

Gnanadesikan, A., 1996. Modeling the diurnal cycle of carbon monoxide: sensitivity tophysics, chemistry, biology, and optics. J. Geophys. Res. 101 (C5), 12177–12191.

Gueymard, C., 1995. SMARTS, a simplemodel of the atmospheric radiative transfer of sun-shine: algorithms and performance assessment. Professional Paper FSEC-PF-270-95 .Florida Solar Energy Center.

Gueymard, C., 2001. Parameterized transmittance model for direct beam and circumsolarspectral irradiance. Sol. Energy 71 (5), 325–346.

Hu, C.M., Muller-Karger, F.E., Zepp, R.G., 2002. Absorbance, absorption coefficient, and ap-parent quantum yield: a comment on common ambiguity in the use of these opticalconcepts. Limnol. Oceanogr. 47, 1261–1267.

Johannessen, S.C., Miller, W.L., 2001. Quantum yield for the photochemical production ofdissolved inorganic carbon in seawater. Mar. Chem. 76, 271–283.

Johnson, J.E., Bates, T.S., 1996. Sources and sinks of carbon monoxide in themixed layer ofthe tropical South Pacific Ocean. Global Biogeochem. Cycles 10 (2), 347–359.

Khalil, M.A.K., Rasmussen, R.A., 1990. The global cycle of carbon-monoxide: trends andmass balance. Chemosphere 20 (1–2), 227–242.

Lide, D.R., 1992. The Handbook of Chemistry and Physics. CRC Press, Ann Arbor.Lu, X.L., Yang, G.P., Song, G.S., Zhang, L., 2010a. Distributions and fluxes of methyl chloride

and methyl bromide in the East China Sea and the Southern Yellow Sea in autumn.Mar. Chem. 118, 75–84.

Lu, X.L., Yang, G.P., Wang, X.M., Wang, W.L., Ren, C.Y., 2010b. Determination of carbonmonoxide in seawater by headspace analysis. Chin. J. Anal. Chem. 38, 352–356 (InChinese with English abstract).

Page 10: Distribution, flux and photoproduction of carbon monoxide in the Bohai and Yellow Seas

113B.-Z. Zhao et al. / Marine Chemistry 168 (2015) 104–113

Mao, X.Y., Jiang, W.S., Zhao, P., Gao, H.W., 2008. A 3-D numerical study of salinity varia-tions in the Bohai Sea during the recent years. Cont. Shelf Res. 28 (19), 2689–2699.

Miller, W.L., Moran, M.A., 1997. Interaction of photochemical and microbial process in thedegradation of refractory dissolved organic matter from a coastal marine environ-ment. Limnol. Oceanogr. 42, 1317–1324.

Miller, W.L., Zepp, R.G., 1995. Photochemical production of dissolved inorganic carbonfrom terrestrial organic matter: significance to the oceanic organic carbon cycle.Geophys. Res. Lett. 22, 417–420.

Miller,W.L., Moran, M.A., Sheldon,W.M., Zepp, R.G., Opsahl, S., 2002. Determination of ap-parent quantum yield spectra for the formation of biologically labile photoproducts.Limnol. Oceanogr. 47, 343–352.

Mopper, K., Kieber, D.J., 2000. Marine photochemistry and its impact on carbon cycling.In: deMora, S., Demers, S., Vernet, M. (Eds.), The Effects of UV Radiation in theMarineEnvironment. Cambridge University Press, Cambridge, pp. 101–129.

Moran, M.A., Zepp, R.G., 1997. Role of photoreactions in the formation of biologically la-bile compounds from dissolved organic matter. Limnol. Oceanogr. 42, 1307–1316.

Parsons, T.R., Maita, Y., Lalli, C.M., 1984. A Manual of Chemical and Biological Methods forSeawater Analysis. Pergamon Press.

Pope, R.M., Fry, E.S., 1997. Absorption spectrum (380–700 nm) of pure water II. Integrat-ing cavity measurements. Appl. Opt. 36, 8710–8723.

Prather, M., Ehhalt, D., Dentener, F., Derwent, R., Dlugokencky, E., Holland, E., Isaksen, I.,Katima, J., Kirchhoff, V., Matson, P., Midgley, P., Wang, M., 2001. Atmospheric chem-istry and greenhouse gases. (Chapter 4) In: Houghton, J.T., Ding, Y., Griggs, D.J.,Noguer, M., van der Linden,, P.J., Dai, X., Maskell, K., Johnson, C.A. (Eds.), ClimateChange 2001: The Scientific Basis. Contribution of Working Group 1 to the Third As-sessment Report of the Intergovernmental Panel on Climate Change. Cambridge Uni-versity Press, Cambridge, pp. 239–287.

Stubbins, A., Uher, G., Kitidis, V., Law, C.S., Upstill-Goddard, R.C., Woodward, E.M.S., 2006a.The open-ocean source of atmospheric carbon monoxide. Deep Sea Res. Part II 53,1685–1694.

Stubbins, A., Uher, G., Law, C.S., Mopper, K., Robinson, C., Upstill-Goddard, R.C., 2006b.Open-ocean carbonmonoxide photoproduction. Deep Sea Res. Part II 53, 1695–1705.

Sun, X.P., 2006. Coastal Areas of China. Ocean Press, Beijing, pp. 4–5 (In Chinese).The Ministry ofWater Resources of the People's Republic of China, 2010. Bulletin of China

River Sediment. China Water & Power Press, Beijing, p. 69 (In Chinese).Thompson, A.M., 1992. The oxidizing capacity of the Earth's atmosphere: probable past

and future changes. Science 256, 1157–1165.Valentine, R.L., Zepp, R.G., 1993. Formationof carbonmonoxide from thephotodegradationof

terrestrial dissolved organic carbon in naturalwaters. Environ. Sci. Technol. 27, 409–412.White, E.M., Kieber, D.J., Sherrard, J., Miller, W.L., Mopper, K., 2010. Carbon dioxide and

carbon monoxide photoproduction quantum yields in the Delaware Estuary. Mar.Chem. 118, 11–21.

Wiesenburg, D.A., Guinasso, N.L., 1979. Equilibrium solubilities of methane, carbon mon-oxide and hydrogen in water and sea water. J. Chem. Eng. Data 24, 356–360.

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 monoxidein seawater. Mar. Chem. 77, 93–108.

Xie, H.X., Zafiriou, O.C., Umile, T.P., Kieber, D.J., 2005. Biological consumption of carbonmon-oxide in Delaware Bay, NW Atlantic and Beaufort Sea. Mar. Ecol. Prog. Ser. 290, 1–14.

Xie, H.X., Bélanger, S., Demers, S., Vincent, W.F., Papakyriakou, T.N., 2009.Photobiogeochemical cycling of carbon monoxide in the southeastern Beaufort Seain spring and autumn. Limnol. Oceanogr. 54 (1), 234–249.

Yang, S.Y., Jung, H.S., Lim, D.I., Li, C.X., 2003. A review on the provenance discrimination ofsediments in the Yellow Sea. Earth-Sci. Rev. 63 (1–2), 93–120.

Yang, G.P., Wang, W.L., Lu, X.L., Ren, C.Y., 2010. Distribution, flux and biological consump-tion of carbon monoxide in the Southern Yellow Sea and the East China Sea. Mar.Chem. 122, 74–82.

Yang, G.P., Ren, C.Y., Lu, X.L., Liu, C.Y., Ding, H.B., 2011. Distribution, flux andphotoproduction 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.

Yu, J., 2012. Seasonal Variation and Distribution of Suspended Sediment in the Yellow Sea.(M. Sc. Thesis), Ocean University of China, China (In Chinese).

Zafiriou, O.C., Andrews, S.S., Wang, W., 2003. Concordant estimates of oceanic carbonmonoxide source and sink processes in the Pacific yield a balanced global “blue-water” CO budget. Global Biogeochem. Cycles 17 (1), 1015–1027.

Zafiriou, O.C., Xie, H., Nelson, N.B., Najjar, R.G., Wang, W., 2008. Diel carbon monoxide cy-cling in the upper Sargasso Sea near Bermuda at the onset of spring and in midsum-mer. Limnol. Oceanogr. 53, 835–850.

Zeep, R.G., 2003. Solar UVR and aquatic carbon, nitrogen, sulfur and metal cycles. In:Helbling, E.W., Zagarese, H. (Eds.), UV Effects in Aquatic Organisms and Ecosystems.The Royal Society of Chemistry, pp. 137–184.

Zhai, S.K., Zhang, H.J., Fan, D.J., Yang, R.M., Cao, L.H., 2005. Corresponding relationship be-tween suspended matter concentration and turbidity on Changjiang Estuary and ad-jacent sea area. Acta Sci. Circumst. 25 (5), 693–699 (In Chinese with English abstract).

Zhang, C., 2011. Distribution, Flux, Photoproduction and Biological Consumption of CarbonMonoxide in sea Water. (M. Sc. Thesis), Ocean University of China, China (In Chinese).

Zhang, Y., Xie, H.X., 2012. The sources and sinks of carbon monoxide in the St Lawrenceestuarine system. Deep Sea Res. Part II 81–84, 114–123.

Zhang, Y., Xie, H.X., Chen, G.H., 2006. Factors affecting the efficiency of carbon monoxidephotoproduction in the St. Lawrence estuarine system (Canada). Environ. Sci.Technol. 40 (24), 7771–7777.

Zhu, M.Y., Mao, X.H., Lv, R.H., Sun, M.H., 1993. The chlorophyll-a and primary productionin the Yellow Sea. J. Oceanogr. Huanghai Bohai Seas 11 (3), 38–51 (In Chinese withEnglish abstract).

Ziolkowski, L.A., 2000. Marine Photochemical Production of Carbon Monoxide. (M. Sc.Thesis) Dalhousie University.

Ziolkowski, L.A., Miller, W.L., 2007. Variability of the apparent quantum efficiency of COphotoproduction in the Gulf of Maine and Northwest Atlantic. Mar. Chem. 105,258–270.

Zuo, Y., Jones, R.D., 1995. Formation of carbon-monoxide by photolysis of dissolved ma-rine organic material and its significance in the carbon cycling of the oceans.Naturwissenschaften 82 (10), 472–474.

Zuo, Y., Jones, R.D., 1997. Photochemistry of natural dissolved organic matter in lake andwetland waters—production of carbon monoxide. Water Res. 31, 850–858.