Atmospheric test of the J(BrONO2)/k(BrO+NO2) ratio:implications for total stratospheric Br-y and bromine-mediated ozone loss

We report on time-dependent O-3, NO2 and BrO profiles measured by limb observations of scattered skylight in the stratosphere over Kiruna (67.9 degrees N, 22.1 degrees E) on 7 and 8 September 2009 during the autumn circulation turn-over. The observations are complemented by simultaneous direct solar occultation measurements around sunset and sunrise performed aboard the same stratospheric balloon payload. Supporting radiative transfer and photochemical modelling indicate that the measurements can be used to constrain the ratio J(BrONO2)/k(BrO+NO2), for which at T = 220 +/- 5 K an overall 1.7(+0.4 - 0.2) larger ratio is found than recommended by the most recent Jet Propulsion Laboratory (JPL) compilation (Sander et al., 2011). Sensitivity studies reveal the major reasons are likely to be (1) a larger BrONO2 absorption cross-section sigma(BrONO2), primarily for wavelengths larger than 300 nm, and (2) a smaller k(BrO+NO2) at 220 K than given by Sander et al. (2011). Other factors, e.g. the actinic flux and quantum yield for the dissociation of BrONO2, can be ruled out. The observations also have consequences for total inorganic stratospheric bromine (Br-y) estimated from stratospheric BrO measurements at high NOx loadings, since the ratio J(BrONO2)/kBrO+NO2 largely determines the stratospheric BrO/Br-y ratio during daylight. Using the revised J(BrONO2)/kBrO+NO2 ratio, total stratospheric Bry is likely to be 1.4 ppt smaller than previously estimated from BrO profile measurements at high NOx loadings. This would bring estimates of Br-y inferred from organic source gas measurements (e.g. CH3Br, the halons, CH2Br2, CHBr3, etc.) into closer agreement with estimates based on BrO observations (inorganic method). The consequences for stratospheric ozone due to the revised J(BrONO2)/k(BrO+NO2) ratio are small (maximum -0.8%), since at high NOx (for which most Br-y assessments are made) the enhanced ozone loss by overestimating Br-y is compensated for by the suppressed ozone los