Introduction

Iodine plays an important role on atmospheric chemistry by destroying tropospheric ozone and forming new particles1,2,3, especially in the marine boundary layer. Among the iodine species, methyl iodide (CH3I) with its relatively high concentration and long lifetime (~7 days) in the atmosphere is thought to be the dominate volatile organic iodine compounds (VOICs) which works as the carrier of iodine atoms from seawater to the atmosphere4,5,6,7, although other VOICs, such as ethyl iodide (C2H5I), chloroiodomethane (CH2ClI), diiodomethane (CH2I2) and bromoiodomethane (CH2BrI)8,9,10,11, as well as inorganic iodine, such as hypoiodous acid (HOI) and I212, are also widely detected over oceans.

CH3I is usually considered to be derived from oceans13. Emission from photochemical reactions in the surface sea water is the dominant source of CH3I6,13. Biogenic activity of phytoplankton and macroalgae is also its important source14, especially in coastal regions15. Besides, terrestrial ecosystems, such as rice cultivation16,17, peatland and wetland18, also have contribution to atmospheric CH3I and are even comparable to oceanic emissions in some local areas19. Biomass burning releases a small quantity of CH3I, but its contribution is negligible on the global scale20. However, anthropologic activities like fossil fuel combustion and industrial emissions are not regards as the source of CH3I. Because the emission of CH3I varies with a wide range from different sources and in different regions and seasons, the estimated global flux of CH3I has great uncertainty21. More observations with comprehensive spatial and seasonal scales would help reduce the uncertainty.

There have been a considerable number of studies about CH3I over oceans or at coastal sites22,23,24,25,26,27,28,29,30. The typical concentrations of CH3I in the marine boundary layer were 0.1–5 pptv, with higher levels over coastal areas than remote oceans13. Yokouchi, et al.31 reported atmospheric CH3I concentrations on a wide scale including in the high, middle and low latitudes of the both hemispheres. Recently, through several ship-based observation, Ooki, et al.32 firstly mapped CH3I and some other VOICs in surface seawater from the Arctic to the Antarctic, especially in the Indian Ocean, Bering Sea and western Arctic Ocean. However, the data for the spatial distribution of CH3I in the marine boundary layer on a global scale is still limited, especially little study has been conducted over oceans in the high latitudes of the Northern Hemisphere like the central Arctic Ocean33, where abruptly sea ice change occurs due to global warming.

During the 28th Chinese Antarctic Research Expedition (CHINARE 11/12, November 2011–April 2012) and the 5rd Chinese Arctic Research Expedition (CHINARE 12, July–September, 2012), ambient air samples were collected in the marine boundary layer from the Arctic to the Antarctic, across more than 150° latitudes, along cruise path from the Norwegian Sea through the central Arctic Ocean, the Chukchi Sea, the western North Pacific Ocean, the eastern Indian Ocean and the southern Ocean to the maritime Antarctic. The results reveals the spatial distribution of CH3I over oceans on a global scale, as well as the potential sources and influencing factors of atmospheric CH3I and hence provides new constraint for the model simulating the impact of CH3I on climate change.

Results

Potential CH3I emission from ships

CH3I is usually not thought to originate from anthropogenic emissions. Exceptionally, Yokouchi, et al.10 found that when air mass derived from polluted continental areas, CH3I mixing ratios at Hateruma Island in the East China Sea (24.05°N, 123.8°E) increased concurrently, indicating possible anthropogenic sources. However, other influencing factors, such as biogenic emissions from macroalgae in coastal seas or terrestrial ecosystems, cannot be eliminated. For samples in this study, CO was simultaneously determined with CH3I. In the marine boundary layer, including over coastal regions where influenced by input from continents, the concentration of CO is usually not more than 150 ppbv34,35,36. Extremely high CO levels indicate probable pollution by ship emissions. Although the sampling site was on the foredeck of the ship and upwind from the exhaust plume, the pollution from the ship was not absolutely excluded due to diffusive emissions37. As presented in Table S1 in the supplementary materials, during both the CHINARE 11/12 and the CHINARE 12, the average, maximum and median mixing ratios of CH3I in samples with CO concentration above 150 ppbv (CO > 150 ppbv) were much higher than those in samples with CO concentration below or equal to 150 ppbv (CO ≤ 150 ppbv). Besides, the difference between CH3I concentrations in samples with CO > 150 ppbv and those with CO ≤ 150 ppbv during the both two cruises was significant (heteroscedastic t-test, P < 0.05).

In order to determine the reason for the increase of CH3I mixing ratios in samples with CO > 150 ppbv, based on 7-day air mass back trajectories (BTs), we split these samples into three groups: ocean origin (OO), land origin (LO) and Antarctic origin (AO) as the same way of our previous study38. Air mass of OO samples only transported over oceans during the past 7 days, whereas air mass of AO and LO samples passed through continental Antarctica and other continents, respectively. The sampling sites when CO > 150 ppbv included both coastal regions and remote oceans (Figure S1). The mean levels of CH3I in LO, AO and OO samples when CO > 150 ppbv were 1.0 ± 0.69, 1.2 ± 0.93 and 2.1 ± 3.2 pptv (mean ± standard deviation, SD, the same below), respectively (Fig. 1a). There was no significant difference (heteroscedastic t-test, P > 0.05) among the three groups of samples. Emissions from macroalgae in coastal regions or terrestrial ecosystems could not explain the abruptly increased CH3I concentrations in OO samples when CO > 150 ppbv. Ship emission was probably caused the increase of CH3I concentrations in OO samples. Besides, owing to the existence of iodine in fossil fuel, such as petroleum39 and coal40, the combustion of fossil fuel may be a potential source of atmospheric CH3I. Direct measurement of combustion exhaust in further studies will be in favor for confirming this source.

Figure 1
figure 1

Box-and-whisker plots of CH3I concentrations in ocean origin (OO), land origin (LO) and Antarctic origin (AO) samples with CO concentration (a) below or equal to 150 ppbv and (b) above 150 ppbv during the CHINARE 11/12 and the CHINARE 12. The lower and upper boundaries of the box represent the 25th and the 75th percentiles, respectively; the whiskers below and above the box indicate the minimum and maximum, respectively; the line within the box marks the median; the dot represents the mean.

Full size image

Spatial distribution

In order to avoid the disturbance from ship emission, only samples with CO ≤ 150 ppbv were selected to discuss the spatial distribution of CH3I concentration during the CHINARE 11/12 and CHINARE 12 (Fig. 2, Table 1). The mixing ratios of CH3I ranged from 0.17 to 2.9 pptv, with a mean of 0.56 ± 0.41 pptv and a median of 0.46 pptv. The mean levels of CH3I for LO, OO and AO samples were 0.80 ± 0.61, 0.53 ± 0.37 and 0.46 ± 0.19 pptv, respectively (Fig. 1b). CH3I concentrations in LO samples were significantly higher than those in OO and AO samples (P < 0.05). Emissions form macroalgaes in coastal regions can cause high CH3I concentrations in LO samples, while low sea surface temperature (SST) and sea ice coverage may depress CH3I production and sea-air exchange5,41 and thus resulted in low CH3I levels in AO samples. Saiz-Lopez, et al.13 summarized mixing ratios of CH3I in the marine boundary layer ranged as a mean level of 1.6 pptv and a median level of 1.2 pptv over coastal regions (corresponding to LO and AO samples in this study) and a mean level of 0.87 pptv and a median level of 0.70 pptv over open oceans (corresponding to OO samples in this study). It demonstrated that CH3I mixing ratios in the marine boundary layer in 2011–2012 were relatively lower than previous observations. It may be relevant to the SST-related decadal anomalies of CH3I emissions30. Similarly, in some coastal sites, such as Happo Ridge (36.7°N, 137.8°E), Hateruma Island (24.1°N, 123.8°E) and Cape Grim (40.4°S, 144.6°E), evident downtrend of CH3I concentrations were observed from 201030. It should be pointed out that CH3I concentrations show seasonal variation, with different patterns in different latitudes31. In this study, due to limited observation time in each region through ship-based research, seasonal variation cannot be investigated. But the sampling time will be considered when discussing spatial distribution. Detailed results with sampling information are listed in Table S2.

Table 1 Range, mean (±SD) and median of CH3I (pptv) in different regions during the CHINARE 11/12 and the CHINARE 12.
Full size table
Figure 2
figure 2

The spatial distribution of CH3I concentrations in the marine boundary layer when the concentration of CO below 150 pptv during the CHINARE 11/12 and the CHINARE 12.

Base map is from ArcGIS 10.0 software (http://www.esri.com).

Full size image

CH3I in the marine boundary layer is principally produced through photochemical reaction in the sea surface, which is affected by solar radiation intensity and dissolved organic carbon (DOC) concentration in sea surface water5,42. The transport of CH3I from sea to air is controlled by SST26,41. Therefore, previous studies over the Pacific Ocean and Atlantic Ocean revealed obviously decreasing trend of CH3I concentrations with increasing latitudes29,31. In this study, samples in the low latitudes (30°N–30°S) were collected in the spring and autumn. According to Yokouchi, et al.31, no pronounced seasonal variation was observed in this region. CH3I concentration in the low latitudes ranged from 0.20 to 1.4 pptv, with a mean of 0.62 ± 0.43 pptv and a median of 0.54 pptv. The mean and median concentrations of CH3I in the low latitudes were higher than those in the high latitudes (60°–90°) of both hemispheres and those in the middle latitudes (30°–60°) of the Southern Hemisphere, but they were lower than those in the middle latitudes in the Northern Hemisphere (Table 1). Previous studies also indicated that the concentration of CH3I near the equator is slightly suppressed31,41. The chemical loss of CH3I in the seawater and marine boundary layer is mainly through the nucleophilic substitution reaction with chloride (Cl−) whose rate depends on temperature13,43. Thereby, although the production and emission of CH3I in the low latitudes is the rapidest, the accumulation amount of atmospheric CH3I may be weakened due to high loss rate. Moreover, intense convection in the tropical latitudes44 will accelerate the dilution of CH3I and cause the decrease of CH3I concentration in the boundary layer31. Moreover, seasonal variation of CH3I concentration with a peak value during our sampling time in 30°–60°N is also a reason causing the lower concentrations in the low latitudes than in the middle latitudes of the Northern Hemisphere (see below).

The highest CH3I concentrations, with the mean of 1.1 ± 1.0 pptv and the median of 0.80 pptv, were found in the middle latitudes of the Northern Hemisphere where samples were collected in the summer. In the middle latitudes, CH3I concentrations reach the peak in the summer and early autumn and reach the tough in the winter31. High solar radiation and sea-surface DOC in the middle latitude in the summer can promote the photochemical emission of CH3I13. Besides, most samples in the middle latitudes of the Northern Hemisphere were collected in coastal regions where the emission from large algae can further enhance CH3I concentrations. In the western North Pacific Ocean (including the Sea of Okhotsk and the Bering Sea), CH3I concentrations ranged from 0.28 to 2.9 pptv, with a mean of 0.91 ± 0.83 and a median 0.74 pptv, which were comparable to the results in the early-middle autumn reported by Yokouchi, et al.25 and the results in the summer reported by Yokouchi, et al.31 Correspondingly, CH3I concentrations in the seawater in September–October in this region also showed high levels32. In the middle latitudes of the Southern Hemisphere, most samples were collected in the spring and autumn and CH3I concentrations were lower than those in the middle latitudes of the Northern Hemisphere, with a mean of 0.60 ± 0.33 pptv and a median of 0.51 pptv. Over the Australian adjacent Sea, CH3I concentrations ranged from 0.20 to 1.3 pptv, with an average of 0.51 ± 0.31 pptv and a median of 0.51 pptv, which were near to the data reported by Yokouchi, et al.31 at Cape Grim, a coastal site of Australia (a mean of ~0.60 pptv and a median ~0.64 pptv) in the same months (March and November). Unlike previous observation over the Southern Ocean in the summer45, during which the peak value reached up to 2.6 pptv, CH3I concentrations in this study ranged from 0.24 to 0.95 pptv, with a mean of 0.51 ± 0.19 pptv and a median of 0.49 pptv. Similarly, Ooki, et al.32 reported that in the autumn, CH3I concentrations in the seawater in the Southern Ocean dropped compared with those in lower latitudes.

In the high latitudes of the Northern Hemisphere and the Southern Hemisphere, samples were mostly collected in the summer and CH3I concentrations were in relatively low levels as a whole. The median levels in the two regions were both only 0.42 pptv. CH3I concentrations over the Arctic Ocean, including the Chukchi Sea and the Central Arctic Ocean, stayed in very low levels (Fig. 2), with medians of 0.22 and 0.24 pptv, respectively. Accordingly, the lowest CH3I concentrations in the seawater on a global scale were also found in the Chukchi Sea and the west part of the Central Arctic Ocean32. Especially, all the CH3I concentrations over the Central Arctic Ocean were below 0.5 pptv (Table 1). Low SST (about 0 °C during our sampling in the Central Arctic Ocean) may reduce the photochemical and biogenic production of CH3I. Moreover, sea-air exchange is also depressed by low SST and the coverage of sea ice. However, the results reported by Yokouchi, et al.33 in the Chukchi Sea and the west part of the Central Arctic Ocean in the autumn were much high than our data, with a range of 0.33–0.85 pptv and a mean of 0.52 pptv. This disparity may be caused by seasonal variation. In the high latitudes of both hemispheres, CH3I concentrations display a minimum in the summer and a maximum in the winter31. This seasonal pattern is caused by limited local emission throughout the year and more photolytic decomposition in the summer due to intense solar radiation in the high latitudes. Dramatically, over the Norwegian Sea, the average and median concentrations were as high as 0.92 ± 0.52 pptv and 0.91 pptv, respectively, which were higher than those over all the other seas. Although located in the high latitudes, the photochemical production rate and sea-air flux of CH3I in the Norwegian Sea are probably in high levels, because the average SST was about 10 °C during our sampling. In addition, biogenic emission may also have an important contribution owing to high oceanic primary production in the summer in the Norwegian Sea46. Overall, atmospheric CH3I concentrations in the Antarctic were relatively low, but higher than those over the Arctic Ocean. This spatial distribution pattern is also consistent with that for CH3I in seawater32. The mean and median concentrations of CH3I over the coastal region of the West Antarctic (including the Antarctic Peninsular and the Drake Passage) were slightly higher than those over the East Antarctic (Table 1). It may be caused by more oceanic emission in the West Antarctic due to higher SST in this region (2.3 °C in average) than in the East Antarctic (0 °C in average) during sampling.

Discussion

Role of sea surface temperature (SST)

The concentration of CH3I over oceans is mainly influenced by oceanic production, sea-air exchange and chemical loss in the seawater and atmosphere. Generally, high SST will promote the photochemical production of CH3I and sea-air exchange5,41 and thus correspond to high CH3I mixing ratios in the marine boundary layer. However, the relationship between CH3I concentration in the air and SST was not linear. As showed in Figure S2, atmospheric concentrations of CH3I increased with increasing SST and reached a peak at 10–15 °C, but rapidly deceased and stay at low levels at SST of 15–30 °C and increased with SST above 30 °C. This trend coincides well with the pattern between SST and CH3I in the seawater reported by Ooki, et al.32, that CH3I concentrations show a peak at SST of ~15 °C and a tough at SST of ~25 °C. In the Norwegian Sea, SST ranged from 6.1 to 14 °C and showed significantly positive correlation (R = 0.62, P < 0.05) with atmospheric CH3I concentration (Table 2). However, in the other regions, no obvious relationship between CH3I and SST was found. It may be due to complex influencing factors of atmospheric CH3I other than SST. For instance, in the Norwegian Sea, CH3I also well correlated with colored dissolved organic matter (CDOM), which is a component of DOC and can partly reflect its level (Table 2).

Table 2 Correlation coefficients of CH3I with the sea surface temperature (SST), wind speed (WD), sea surface salinity (SSS), colored dissolved organic matter (CDOM), concentrations of isoprene, α-pinene and carbon monoxide (CO) and 1-day to 7-day chlorophyll-a exposure in different regions during the CHINARE 11/12 and the CHINARE 12*.
Full size table

Role of biogenic emission indicated by chlorophyll-a

Because CH3I in the marine boundary layer is affected by biogenic emission, its concentration should have relativity with chlorophyll-a. Previous studies reveals that CH3I does not present well correlation with in-situ chlorophyll-a, but significantly correlated with chlorophyll-a exposure along back trajectories over the past several days45,47. According to Lai, et al.45, chlorophyll-a data from NASA was taken at 6-hourly position along the back trajectories to calculate chlorophyll-a exposure over 1–7 days prior to reach the sampling site. As presented in Table 2, over the Norwegian Sea, CH3I concentration was significantly correlated with 1-day, 3-day and 4-day chlorophyll-a exposure, indicating that both local source and long-range transport of biogenic emissions affect CH3I in this region. CH3I also had significant relationship with 4-day to 7-day chlorophyll-a exposure in the Chukchi Sea and with 7-day chlorophyll-a exposure in the Australian Adjacent Sea, indicating the influence from long-range transport of biogenic emission; while significant correlation between CH3I and 1-day chlorophyll-a exposure in the Southern Ocean, indicating the contribution of local biogenic emission. In the Norwegian Sea and Southeast Asian Sea, CH3I also present significantly positive correlation with α-pinene released by phytoplankton48 (Table 2), indicating joint sources. For instance, prochlorococcus is a major biogenic source of CH3I49 and its spatial distribution, high levels in the low and middle latitudes, also agrees with that of CH3I. Meanwhile, prochlorococcus presents high α-pinene emission rate48. However, over the two seas, no significant relationship was found between CH3I and isoprene, another important biogenic volatile organic compound (BVOC) species over oceans50. This discrepancy may be due to many other functional types of phytoplankton with high isoprene emission rate, such as synechococcus, haptophytes and diatoms species51.

Role of sea surface salinity and other physical factors

Significantly negative correlations between CH3I concentration and sea surface salinity (SSS) was found in the coastal region of the West Antarctic (R = −0.57, P < 0.05). Similarly, the decreasing trend of CH3I concentration in the seawater with the increasing SSS was observed at Kiel Fjord in the Baltic Sea52. High SSS value means high Cl− in the seawater and thus causes more CH3I depleted in the seawater. However, the parameters link to production and sea-air exchange of CH3I, such as SST, wind speed (WD) and CDOM did not present obvious relationship with atmospheric CH3I in the West Antarctic. It indicated that rather than emission, chemical loss dominated CH3I concentrations in this region. Besides these factor, in the western North Pacific Ocean, dramatically, CH3I showed significant correlation with CO (R = 0.73, P < 0.05), indicating combustion emissions. CH3I concentrations in LO samples (0.80–2.9 pptv) were much higher than those in OO samples (0.31–0.68 pptv), further suggesting the input from continents. This region is heavily affected by biomass burning in the East Siberia53, which can also emit CH3I54. Besides, anthropogenic sources like fossil fuel burning possibly also played a role on CH3I over oceans10. In the Central Arctic Ocean, the Barents Sea and the coastal region of the East Antarctic Ocean, no parameter was found to be significantly correlated with CH3I concentration. It may be caused by the offset of the role of each factor and too low CH3I levels in these seas.

Experimental Methods

Sampling

Ambient air samples were collected between the East China Sea to the coastal regions of Antarctica (35°N–70°S) during the CHINARE 11/12 and between the East China Sea to the Arctic Ocean (37°N–88°N) during the CHINARE 12. The sampling site located upwind on the upper-most deck of the icebreaker Xuelong. 2-L electro-polished stainless steel canisters, which can keep gases in it out of light and avoid photochemical reaction, were used to collect ambient air samples. Prior to the cruises, the canisters were cleaned and evacuated. Each sampling lasted for about 5 minutes. After sampling, the canisters were then stored in a dark and thermostatic room at 4 °C during the cruises. After the cruises, samples were sent to the Guangzhou Institute of Geochemistry (GIG), Chinese Academy of Sciences, for analysis immediately. All the analysis was done within 6 months after collection for the CHINARE 11/12 samples and within 3 months for the CHINARE 12 samples. Yokouchi, et al.50 reported that CH3I does not show significant decline in canisters 6-month after sampling.

Chemical analysis

A Model 7100 pre-concentrator (Entech Instruments Inc., California, USA) coupled with an Agilent 5973 N gas chromatography-mass selective detector/flame ionization detector (GC-MSD/FID, Agilent Technologies, USA) was used to analyse volatile organic compounds (VOCs) in the canister samples. Details of the analytical procedure were previously described55. Briefly, 500 mL air sample in the canister was first concentrated in a liquid-nitrogen cryogenic trap at −160 °C. Then, pure helium transferred the trapped VOCs to a secondary trap at −40 °C with Tenax-TA as an adsorbent. During these two processes, the majority of H2O and CO2 were removed. The secondary trap was then heated to transfer the target VOCs to a third cryo-focus trap at −170 °C by helium. The third trap was heated rapidly to transfer the VOCs into the GC-MSD/FID system. Helium was used with a HP-1 capillary column (60 m × 0.32 mm × 1.0 μm, Agilent Technologies, USA) as carrier gas and then divided in two ways: The first was a PLOT-Q column (30 m × 0.32 mm × 2.0 μm, Agilent Technologies, USA) followed by FID detection. The second was a 65 cm × 0.10 mm I.D stainless steel line followed by MSD detection. The GC oven temperature was initially set at −50 °C for 3 min and increased to 10 °C at 15 °C min−1, then 120 °C at 5 °C min−1, then 250 °C at 10 °C min−1 and remaining at 250 °C for 10 min. The MSD was used in selected ion monitoring (SIM) mode and the ionization method was electron impact ionization (EI). CH3I concentrations were obtained from the signal of MSD. CO in the canister was separated by a packed column (5 Å Molecular Sieve 60/80 mesh, 3 m × 1/8 in.), converted to CH4 by a Ni-based catalyst and then analyzed by an Agilent 6890 gas chromatograph equipped with an FID.

Quality Control and Assurance

Before the cruises, all canisters were cleaned at least five times by filling and evacuating with humidified zero air. In order to check any possible contamination in the canisters, all canisters were evacuated after the cleansing procedures, re-filled with pure nitrogen, stored in the laboratory for at least 24 h and then the same methods as field samples were used to analyse VOCs and ensure that no target compounds were found or that they were under the method detection limit. CH3I was identified based on its retention times and mass spectra and quantified by a mixture standard with 0.58 pptv of CH3I in it from the Rowland/Blake laboratory at the University of California, Irvine56. The comparison of quantification between the GIG laboratory and the Rowland/Blake laboratory had been conducted through duplicate samples57. The relative measurement deviations were within 4% for CH3I. Before sample analysis, the analytical system was checked daily with a one-point calibration. If the results were beyond ±10% of the initial calibration curve, recalibration was performed. The detection limit for CH3I in this study is ~0.10 pptv.

Data of chlorophyll-a, sea ice and air mass back trajectories

Satellite chlorophyll-a data in the surface seawater were obtained by Moderate Resolution Imaging Spectroradiometer (MODIS) from NASA satellites (http://oceancolor.gsfc.nasa.gov). Air mass back trajectories (BTs) were calculated for the samples using HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory) transport and dispersion model from the NOAA Air Resources Laboratory (http://www.arl.noaa.gov/ready/hysplit4.html). 7-day BTs for each sampling were traced with 6 h steps at 50 m above sea level.

Additional Information

How to cite this article: Hu, Q. et al. Methyl iodine over oceans from the Arctic Ocean to the maritime Antarctic. Sci. Rep. 6, 26007; doi: 10.1038/srep26007 (2016).