DOAS-observation of halogen radical-catalysed arctic boundary layer ozone destruction during the ARCTOC-campaigns 1995 and 1996 in Ny-Ålesund, Spitsbergen

Abstract

During two field campaigns in spring 1995 and 1996, boundary layer concentration levels of BrO, ClO, IO, SO2, NO2, HNO2, CH2O and O3, were observed at a time resolution of better than 1.5 h by Differential Optical Absorption Spectroscopy (DOAS). Up to 30 ppt of both, BrO and ClO, were found to coincide with low ozone events (LOE). During periods of normal ozone, average levels of BrO and CIO were close to zero. Average CIO levels during LOE were considerably higher in 1995 (21 ppt) than in 1996 (3.3 ppt). Generally, only upper limits for NO2 and SO2 of 50 ppt can be given, however, a series of concentration spikes (up to a few ppb) observed are in the case of NO2 most likely due to local pollution, while the origin of the SO2 spikes remains unclear. Analysis of the data suggests that the ozone loss is most likely caused by a combination of BrOx (= Br + BrO)- and BrOx+ ClOx- catalysed ozone destruction (with the BrO self reaction and ClO + BrO reaction being the rate limiting steps, respectively). The presence of IO at levels exceeding 1–2 ppt and thus its contribution to ozone destruction is unlikely, but cannot be completely excluded. The observed levels of BrO and ClO are sufficiently high to destroy 1–2 ppb O3 per hour, thus completely eliminating 40 ppb of initial O3 within 1–2 days. However airmass dynamics is also likely to be important, in particular observed O3 decrease rates of up to 7 ppb/h can only be attributed to advection of air already depleted in ozone. The Cl-atom and Br-atom concentrations deduced from ClO and BrO measurements and estimates of ClO/CL and BrO/Br ratios, are compared with data derived from “hydrocarbon clock” observations in the arctic. While both sets of data are compatible for BrO and 1996 ClO data, in 1995 the DOAS ClO data indicate much higher Cl-atom levels than seen by hydrocarbon clock observations.