Feiyue Wang

Redox chemistry controls the behavior of mercury during long‐range atmospheric transport. Despite major progress in understanding the oxidation process in the atmosphere, the mechanism of mercury reduction, especially the aqueous‐phase and multi‐phase reactions in cloudwater, remains poorly known. Here we report experimentally determined rates of in‐cloud mercury photoreduction using cloudwater samples collected from a forested area in the Canary Islands, Spain. Mercury concentrations in cloudwater varied greatly from 0.04 to 19.9 nM (7.89 ± 6.11 nM) for total mercury and 0.02 to 4.98 nM (1.16 ± 1.49 nM) for the dissolved fraction, with particulate mercury being the dominant fraction in most of the samples. The mercury concentrations were elevated in comparison with those previously reported for atmospheric waters, reflecting the impact of a large‐scale forest wildfire that occurred shortly before the sampling campaign on a neighboring island. Mercury photoreduction was determined by two independent methods: under natural solar radiation at a nearby mountain site located in the free troposphere and under UV radiation in the laboratory. In addition to aqueous‐phase photoreduction, multi‐phase photoreduction involving particulate mercury also occurred in cloudwater, with the latter becoming dominant once dissolved mercury had been depleted. Overall, the pseudo‐first‐order rate constants for in‐cloud mercury photoreduction varied from 0.07 to 0.21 hr −1 , which are one order of magnitude lower than the values presumed in global mercury transport models. Our results suggest that in‐cloud reactions alone are insufficient to account for mercury reduction in the atmosphere and that other pathways, such as gas‐phase reactions, must exist and need to be properly incorporated into future models.