A machine learning based global sea-surface iodide distribution

Abstract

Abstract. Iodide in the sea-surface plays an important role in the Earth system. It modulates the oxidising capacity of the troposphere and provides iodine to terrestrial ecosystems. However, our understanding of its distribution is limited due to a paucity of observations. Previous efforts to generate global distributions have generally fitted sea-surface iodide observations to relatively simple functions of sea-surface temperature (Chance et al., 2014; MacDonald et al., 2014). This approach fails to account for coastal influences and variation in the bio-geochemical environment. Here we use a machine learning regression approach (Random Forest Regression) to generate a high resolution (0.125° x 0.125°, ∼ 12.5 km), monthly dataset of present-day global sea-surface iodide. We use a compilation of iodide observations (Chance et al., 2019b) that is 45 % larger than has been used previously (Chance et al., 2014) as the dependent variable and co-located ancillary parameters (temperature, nitrate, phosphate, salinity, shortwave radiation, topographic depth, mixed layer depth, and chlorophyll-a) from global climatologies as the independent variables. We investigate the regression models generated using different combinations of ancillary parameters and select the ten best-performing models to be included in an ensemble prediction. We then use this ensemble of models, combined with global fields of the ancillary parameters, to predict a new high resolution global sea-surface iodide field. Sea-surface temperature is the most important variable in all of the top ten models. We estimate a global average sea-surface iodide concentration of 106 nM (with an uncertainty of ∼ 20 %), which is within the range of previous estimates (60–130 nM). Similar to previous work, higher concentrations are predicted for the tropics than for the extra-tropics. Unlike the previous parameterisations, higher concentrations are also predicted for shallow areas such as coastal regions and the South China Sea. Compared to previous work, the new parameterisation better captures observed variability. The iodide concentrations calculated here are significantly higher (40 % on a global basis) than the commonly used MacDonald et al. (2014) parameterisation, with implications for our understanding of iodine in the atmosphere. The global iodide dataset is made freely available to the community (DOI: https://doi.org/10/gfv5v3) and as new observations are made, we will update the global dataset through a "living data" model.

Ocean Biogeochemistry Control on the Marine Emissions of Brominated Very Short‐Lived Ozone‐Depleting Substances: A Machine‐Learning Approach

“…Machine‐learning and artificial intelligence have proven to be a valuable tool in the Earth system science. To name a few inspiring applications, Sherwen et al () and Roshan and DeVries () leveraged existing satellite or gridded in situ observations and developed new products for model use. Recent years have seen “online” applications as well.…”

“…For example, machine‐learning “emulators,” trained by process‐level models (often computationally expensive) or observations, may be used to replace the (sometimes highly uncertain) parameterizations (e.g., planetary boundary layer schemes and cloud microphysics schemes) in Earth system models (Sobhani et al, ). Although common algorithms (e.g., random forest) may be used in the online (e.g., Sobhani et al, ) and offline applications (e.g., Sherwen et al, ), the online approach has advantages: The machine‐learning emulator incorporated into the Earth system model (online) is coupled to other physical and chemical processes within the Earth system model, therefore will respond to changes in local conditions or external forcing, and hence may reveal insights into the feedback mechanisms. In the aforementioned examples (Roshan & DeVries, ; Sherwen et al, ), climatologies (usually monthly) of gridded observations and satellite products are used; therefore, the temporal variations beyond the climatologies (e.g., decadal, interannual, and daily) or feedback mechanisms cannot be resolved.…”

“…Although common algorithms (e.g., random forest) may be used in the online (e.g., Sobhani et al, ) and offline applications (e.g., Sherwen et al, ), the online approach has advantages: The machine‐learning emulator incorporated into the Earth system model (online) is coupled to other physical and chemical processes within the Earth system model, therefore will respond to changes in local conditions or external forcing, and hence may reveal insights into the feedback mechanisms. In the aforementioned examples (Roshan & DeVries, ; Sherwen et al, ), climatologies (usually monthly) of gridded observations and satellite products are used; therefore, the temporal variations beyond the climatologies (e.g., decadal, interannual, and daily) or feedback mechanisms cannot be resolved. Moreover, the offline approaches, trained by present‐day observations, may be prone to systematic biases when used for future/past projections.…”

Estimation of Reactive Inorganic Iodine Fluxes in the Indian and Southern Ocean Marine Boundary Layer

“…Daily averaged atmospheric and oceanic parameters combined from ISOEof model SSI predictions(Sherwen et al, 2019b), filled circles in dark blue correspond to measured SSI from ISOE-9 for each observation, (e) chlorophyll-a observations from ISOE-8 and ISOE-9 (blue circles) and satellite data for all campaigns (red circles). (f) ozone mixing ratios from campaigns ISOE and IIOE-2.…”

Measurements of Ozone Deposition to a Coastal Sea by Eddy Covariance

“…Ozone deposition is likely to depend both upon physical exchange, facilitated by diffusion and turbulence, and chemical reaction at the water's surface (Chang et al, 2004;Fairall et al, 2007;Luhar et al, 2018). Iodide in sea water has been identified as a key reactant (Garland et al, 1980), and there has been considerable recent progress in understanding its global distribution (Chance et al, 2014;Macdonald et al, 2014;Sherwen et al, 2019). However, there has only been one report of the dependence of the iodideozone rate constant with temperature (Magi et al, 1997), and this remains a considerable uncertainty in global models.…”