Measurements of peroxy radicals (RO2* = HO2 + ∑RO2, where HO2 is the hydroperoxyl radical and R is an organic group) were made using the chemical amplifier technique (PERCA) during the Air chemistry and Lidar studies of tropospheric and stratospheric species on the Atlantic Ocean (ALBATROSS) campaign on board the German research vessel Polarstern (cruise ANT XIV/1, 1996). The data obtained are compared to previous results from an earlier cruise in 1991 (ANT X/1). Reasonable agreement between the two data sets was observed, indicating the reliability of these measurements. Both data sets take into account the sensitivity of the radical amplifier to the presence of ambient water vapor. Maximum RO2* mixing ratios around noon between 40 and 80 pptv were measured. Nighttime signals were observed on many days. Air masses in different latitude regions in the North and South Atlantic could be characterized using back trajectories. In spite of the fact that the RO2* is relatively short lived, its mixing ratio appears to be influenced by the path traveled by the air mass somewhat higher levels being associated with sources of pollution. A box model based on CH4 and CO oxidation chemistry describes RO2* reasonably well but could not explain the persistent nighttime signals and the HCHO observed. An additional source of HCHO is required, indicating the importance of nonmethane hydrocarbon (NMHC) chemistry in the remote Atlantic boundary layer. Both back trajectories and variations of trace gas concentrations indicate that biomass burning, ship, and natural emissions are likely responsible for the observed deviations from the assumed chemistry.

[1] Selected trace gas mixing ratios (i.e., peroxy radicals (RO* 2 = HO 2 + AERO 2 ), nonmethane hydrocarbons (NMHCs), O 3 , CO, HCHO, and NO) and photolysis rate coefficients of j(NO 2 ) and j(O( 1 D)) were measured in the marine boundary layer (MBL) over the Indian Ocean. The measurements were performed during February, March, and April 1999 as a part of the Indian Ocean Experiment (INDOEX) on board the research vessel R/V Ronald H. Brown. During the campaign, air parcels having different origins and consequently variable compositions were encountered, but all air masses, including those heavily polluted with NMHCs and aerosols, were in the regime of rapid photochemical ozone destruction. The influence of aerosols on the photolysis frequencies was investigated by comparison of measurements and results from the radiative transfer model PHOTOST: the high optical depth (up to 0.6) and low single scattering albedo of the aerosol reduces the UV flux at the surface substantially downwind of India and Arabia causing, for instance, a reduction in j(O( 1 D)) by up to 40%. The diurnal behavior of the trace gases and parameters in the MBL has been investigated by using a timedependent zero-dimensional chemical model. Significant differences between the diurnal behavior of RO* 2 derived from the model and observed in measurements were identified. The measured HCHO concentrations differed from the model results and are best explained by some missing chemistry involving low amounts of Cl. Other possible processes describing these two effects are presented and discussed.

[1] The objective of this study has been to investigate the origin of the water vapor effect on the chain length (CL) of peroxy radical chemical amplifiers (PERCA). Results of the investigation of the water interference in the determination of peroxy radicals by using the PERCA technique are presented. The experimental conditions have been analyzed and modeled. A nonlinear dependence of the CL on the relative humidity (RH) has been accurately determined. The combined analysis of experimental and simulated results rules out wall loses as a single explanation of the CL variation observed and indicates three reactions, which possibly account for this water effect:À À À À À! nonradical products (e.g., HNO 3 ). Assuming a mechanism involving the formation of a HO 2 -nH 2 O complex, the corresponding rate coefficients and their water dependence have been estimated. The quadratic dependence of these rate coefficients upon the RH implies the participation of two H 2 O molecules in the proposed reactions. The study has shown that at a RH of 80% an effective secondorder rate coefficient of 10 À15 cm 3 molecule À1 s À1 for the reaction of CO with HO 2 , or 10À13 cm 3 molecule À1 s À1 for the reaction of HO 2 with NO, explains the observed behavior. Both these complex reactions have potential significance for the chemistry of the marine boundary layer (MBL) and their atmospheric implications are discussed.

No abstract

No abstract