Heterogeneous BrONO2 hydrolysis: Effect on NO2 columns and ozone at high latitudes in summer


  • L. K. Randeniya,

  • P. F. Vohralik,

  • I. C. Plumb,

  • K. R. Ryan


The heterogeneous reaction, N2O5 + H2Oaerosol → → 2HNO3, is responsible for increasing [HOx] and repartitioning of active NOx into HNO3. Throughout much of the atmosphere, N2O5 is formed predominantly at night owing to the rapid photolysis of its precursor, NO3, in sunlit hours. Laboratory measurements have shown that BrONO2 + H2Oaerosol → HOBr + HNO3 (reaction (3)) also has the potential to cause repartitioning of ozone-depleting species, although better determination of γ, the reaction probability, is still required for some stratospheric conditions. The diurnal behavior of N2O5 and BrONO2 are entirely different. In contrast to N2O5, BrONO2 is formed predominantly during the daytime. The result of (3) is to increase [HOBr] at the expense of [BrONO2]. Photolysis of HOBr then leads to increased [OH] and increased O3 loss. In this work two-dimensional calculations show clearly that the impact of (3) is greatest for high aerosol levels and for high latitudes in summer. The calculations have been used to determine the effects of increased aerosol loading on calculated NO2 columns in the Antarctic during summer and autumn of 1990, 1991, 1992 and 1993. It is shown that (3) could be responsible for reductions in NO2 columns during polar day comparable to those measured in 1992 and 1993 following the eruption of Mount Pinatubo. Reaction (3) results in only marginal changes to ozone catalytic loss cycles in 1990. However, for the high aerosol levels of 1992, the inclusion of this reaction results in up to 50% higher ozone loss rates in the 12 to 20 km range. This is caused predominantly by a large increase in [HOx] tempered by a reduction in loss due to NOx. Calculations in which transport terms were switched off showed that, between 12 and 20 km at 77.5°N, local chemistry removes about 30% of the ozone between April and September compared with 20% when the effects of (3) are not included.