Methyl iodide production in the open ocean

Production pathways of the prominent volatile organic halogen compound methyl iodide (CH3I) are not fully understood. Based on observations, production of CH3I via photochemical degradation of organic material or via phytoplankton production has been proposed. Additional insights could not be gained from correlations between observed biological and environmental variables or from biogeochemical modeling to identify unambiguously the source of methyl iodide. In this study, we aim to address this question of source mechanisms with a three-dimensional global ocean general circulation model including biogeochemistry (MPIOM-HAMOCC (MPIOM - Max Planck Institute Ocean Model HAMOCC - HAMburg Ocean Carbon Cycle model)) by carrying out a series of sensitivity experiments. The simulated fields are compared with a newly available global data set. Simulated distribution patterns and emissions of CH3I differ largely for the two different production pathways. The evaluation of our model results with observations shows that, on the global scale, observed surface concentrations of CH3I can be best explained by the photochemical production pathway. Our results further emphasize that correlations between CH3I and abiotic or biotic factors do not necessarily provide meaningful insights concerning the source of origin. Overall, we find a net global annual CH3I air-sea flux that ranges between 70 and 260 Gg yr(-1). On the global scale, the ocean acts as a net source of methyl iodide for the atmosphere, though in some regions in boreal winter, fluxes are of the opposite direction (from the atmosphere to the ocean).

1 Simulated and observed methyl iodide concentrations for individual ship cruises Figure S1 Observations are from the Polarstern cruise BLAST 2, 18.10.-21.11.1994(Butler et al., 2007 as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S2
Observations are from a RSS Discovery cruise in May 2007 (EXPOCODE: 74DI20070520) (Jones et al., 2010) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S3
Observations are from the Poseidoin cruise DRIVE in June 2010 as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m. Figure S4 Observations are from the RV Maria S. Merian cruise MSM18/3 in June 2011 as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S5
Observations are from the Polarstern cruise ANT X/1 in November 1994 (Schall et al., 1997) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.   (Chuck et al., 2005) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S8
Observations are from the Discoverer cruise BLAST 1, 18.01.-17.02.1994(Butler et al., 2007 as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S9
Observations are from the NOAA ship Ronald H. Brown cruise Gas Ex 98, 07.05.-27.07.1998 (Butler et al., 2007) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.    -23.10.1999-23.10. (Butler et al., 2007 as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S13
Observations are from the R/V Sonne cruise TransBrom Sonne in Oktober 2009 as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S14
Observations are from the SA Agulhas cruise SWEDARP 97/98 in December 1997 (Abrahamsson et al., 2004) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m. Figure S15 Observations are from the Nathaniel B. Palmer cruise BLAST 3, 22.02.-07.04.1996 (Butler et al., 2007) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m. Figure S16 Observations are from the Aurora Australis cruise CLIVAR 01, 29.10.-13.12. 2001 (Butler et al., 2007) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.   (Chuck et al., 2005) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S19
Observations are from the Polarstern cruise ANT XVIII/2 (Chuck et al., 2005) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m. Figure S20 Observations are from the Polarstern cruise ANT VI/2 in October 1987 as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S21
Observations are from the CCGS Martha L. Black cruise C-SOLAS SABINA in April 2003 (Wang et al., 2009) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m. Figure S22 Observations are from the CCGS Martha L. Black cruise C-SOLAS SABINA in July 2003 (Wang et al., 2009) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m. Figure S23 Observations are from the CCGS Martha L. Black cruise C-SOLAS SABINA in October 2003 (Wang et al., 2009) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m. Figure S24 Observations are from the CCGS Tully C-SOLAS cruise in July 2003 (Moore and Wang, 2006) as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.

Figure S25
Observations are from the James Clark Ross cruise SAMS northen seas program (JR75) in June 2002 as listed in the SI of (Ziska et al., 2013). Values are shown only where the model ocean is deeper than 200m.
2 Sensitivity of CH 3 I concentrations towards atmospheric forcing and production rates Figure S26 Box-Whisker plot of simulated and observed surface ocean CH 3 I concentrations [pmol L-1]. Box widths are determined by the 25% and 75% percentile of data within each 10 degree latitude box, outliers (gray) are located outside 1.5 times the differences of the percentiles, the middle line of each box shows the median. Simulated concentrations are averaged over 1 degree boxes around the location of observations.

Figure S27
Box-Whisker plot of simulated and observed surface ocean CH 3 I concentrations [pmol L-1]. Box widths are determined by the 25% and 75% percentile of data within each 10 degree latitude box, outliers (gray) are located outside 1.5 times the differences of the percentiles, the middle line of each box shows the median. Simulated concentrations are averaged over 1 degree boxes around the location of observations.