Comparison of model and ground observations finds snowpack and blowing snow both contribute to Arctic tropospheric reactive bromine

Abstract. Reactive halogens play a prominent role in the atmospheric chemistry of the Arctic during springtime. Field measurements and models studies suggest that halogens are emitted to the atmosphere from snowpack and reactions on wind-blown snow. The relative importance of snowpack and blowing snow sources is still debated, both at local scales and regionally throughout the Arctic. To understand implications of these halogen sources on a pan-Arctic scale, we simulate Arctic reactive bromine chemistry in the atmospheric chemical transport model GEOS-Chem. Two mechanisms are included: 1) a blowing snow sea salt aerosol formation mechanism and 2) a snowpack mechanism assuming uniform molecular bromine production from all snow surfaces. We compare simulations including neither mechanism, each mechanism individually, and both mechanisms to examine conditions where one process may dominate or the mechanisms may interact. We compare the models using these mechanisms to observations of bromine monoxide (BrO) derived from multiple-axis differential optical absorption spectroscopy (MAX-DOAS) instruments on O-Buoy platforms on the sea ice and at a coastal site in Utqiaġvik, Alaska during spring 2015. Model estimations of hourly and monthly average BrO are improved by assuming a constant yield of 0.1 % molecular bromine from all snowpack surfaces on ozone deposition. The blowing snow mechanism increases BrO by providing more surface area for reactive bromine recycling. The snowpack mechanism led to increased BrO across the Arctic Ocean with maximum production in coastal regions, whereas the blowing snow mechanism increases BrO in specific areas due to high surface windspeeds. Our uniform snowpack source has a greater impact on BrO mixing ratios than the blowing snow source. Model results best replicate several features of BrO observations during spring 2015 when using both mechanisms in conjunction, adding evidence that these mechanisms are both active during the Arctic Spring. Extending our transport model throughout the entire year leads to predictions of enhanced fall BrO that are not supported by observations.