We present a comprehensive description and benchmark evaluation of the tropospheric chemistry version of the global chemistry transport model TM5 (Tracer Model 5, version TM5-chem-v3.0). A full description is given concerning the photochemical mechanism, the interaction with aerosol, the treatment of the stratosphere, the wet and dry deposition parameterizations, and the applied emissions. We evaluate the model against a suite of ground-based, satellite, and aircraft measurements of components critical for understanding global photochemistry for the year 2006. <br><br> The model exhibits a realistic oxidative capacity at a global scale. The methane lifetime is ~8.9 years with an associated lifetime of methyl chloroform of 5.86 years, which is similar to that derived using an optimized hydroxyl radical field. <br><br> The seasonal cycle in observed carbon monoxide (CO) is well simulated at different regions across the globe. In the Northern Hemisphere CO concentrations are underestimated by about 20 ppbv in spring and 10 ppbv in summer, which is related to missing chemistry and underestimated emissions from higher hydrocarbons, as well as to uncertainties in the seasonal variation of CO emissions. The model also captures the spatial and seasonal variation in formaldehyde tropospheric columns as observed by SCIAMACHY. Positive model biases over the Amazon and eastern United States point to uncertainties in the isoprene emissions as well as its chemical breakdown. <br><br> Simulated tropospheric nitrogen dioxide columns correspond well to observations from the Ozone Monitoring Instrument in terms of its seasonal and spatial variability (with a global spatial correlation coefficient of 0.89), but TM5 fields are lower by 25–40%. This is consistent with earlier studies pointing to a high bias of 0–30% in the OMI retrievals, but uncertainties in the emission inventories have probably also contributed to the discrepancy. <br><br> TM5 tropospheric nitrogen dioxide profiles are in good agreement (within ~0.1 ppbv) with in situ aircraft observations from the INTEX-B campaign over (the Gulf of) Mexico. <br><br> The model reproduces the spatial and seasonal variation in background surface ozone concentrations and tropospheric ozone profiles from the World Ozone and Ultraviolet Radiation Data Centre to within 10 ppbv, but at several tropical stations the model tends to underestimate ozone in the free troposphere. <br><br> The presented model results benchmark the TM5 tropospheric chemistry version, which is currently in use in several international cooperation activities, and upon which future model improvements will take place

Abstract. Airborne measurements of formaldehyde (FA), glycolaldehyde (GA), glyoxal (GL), methylglyoxal (MG), and pyruvic acid (PD) were made on board instrumented aircraft platforms, the Department of Energy G 1 and National Oceanic and Atmospheric Administration P3 (FA only), during the 1995 Nashville/Middle Tennessee Ozone Study. FA data determined on these two aircraft during three intercomparison flights agreed to within-10%. The mean and median (in parentheses) concentrations observed within the boundary layer (< 2000 m) for FA, GA, GL, MG, and PD were 4.2 (3.9), 0.78 (0.70), 0.07 (0.06), 0.14 (0.10), and 0.24 (0.23), respectively (all in parts per billion by volume (ppbv)). At the observed concentrations, FA accounted for a substantial fraction of the OH reactivity, comparable to isoprene and CO. Further, it served as an important free radical precursor, contributing 25-30% of the radical production during midday and even greater during morning and late afternoon periods. Source attribution of FA was assessed using measured concentrations of hydrocarbons, their reaction kinetics with the OH radicals, and FA yields. The results showed that isoprene was the dominant source of FA, responsible for-67%, methane and peroxyacetyl nitrate contributed equally at-13% each, and alkanes and alkenes which are largely of anthropogenic origin accounted for only-7%. This assessment based on reaction kinetics is consistent with the strong correlations (r > 0.8) observed between FA and two other isoprene products, GA and MG. Further, the magnitudes of the nonzero FA intercept exhibited in these correlation plots are found to qualitatively agree with the fraction of precursors that did not concomitantly produce GA and MG. Inspection of specific flights showed direct evidence of the dominance of isoprene as a precursor for FA, appreciable contribution of FA to CO, and negligible decay of FA overnight. Because of the dominant role isoprene plays as a precursor of FA, FA could be used as a proxy of isoprene for assessing the applicability of various versions of biogenic emission inventory.

Ship-borne ozone (O 3 ) measurements over the Atlantic Ocean during the period from 1977 to 2002 show that O 3 trends in the northern mid-latitudes are small. In contrast, remarkably large O 3 trends occur at low latitudes and in the Southern Hemisphere, where near-surface O 3 has increased by up to a factor of 2. The likely cause is the substantial increase of anthropogenic emissions of nitrogen oxides (NO x ) associated with energy use in Africa, which has added to NO x from biomass burning and natural sources.

Abstract. An intensive field measurement campaign was performed in July/August 2002 at the Global Atmospheric Watch station Izaña on Tenerife to study the interaction of mineral dust aerosol and tropospheric chemistry (MINATROC). A dense Saharan dust plume, with aerosol masses exceeding 500 µg m-3, persisted for three days. During this dust event strongly reduced mixing ratios of ROx (HO2, CH3O2 and higher organic peroxy radicals), H2O2, NOx (NO and NO2) and O3 were observed. A chemistry boxmodel, constrained by the measurements, has been used to study gas phase and heterogeneous chemistry. It appeared to be difficult to reproduce the observed HCHO mixing ratios with the model, possibly related to the representation of precursor gas concentrations or the absence of dry deposition. The model calculations indicate that the reduced H2O2 mixing ratios in the dust plume can be explained by including the heterogeneous removal reaction of HO2 with an uptake coefficient of 0.2, or by assuming heterogeneous removal of H2O2 with an accommodation coefficient of 5x10-4. However, these heterogeneous reactions cannot explain the low ROx mixing ratios observed during the dust event. Whereas a mean daytime net ozone production rate (NOP) of 1.06 ppbv/hr occurred throughout the campaign, the reduced ROx and NOx mixing ratios in the Saharan dust plume contributed to a reduced NOP of 0.14-0.33 ppbv/hr, which likely explains the relatively low ozone mixing ratios observed during this event.

[1] We present HCHO measurements performed during the Upper Tropospheric Ozone: Processes Involving HO x and NO x (UTOPIHAN) II aircraft campaign over centralwestern Europe and the northwestern Mediterranean region in 2003 and 2004. We compare these with box model and three-dimensional (3-D) chemistry-transport model results for ''background'' and ''convective'' conditions. Generally, we obtain good agreement between the model and measurement data, although the 3-D model tends to underestimate HCHO near the tropopause, whereas the box model tends to slightly overestimate HCHO for the background case. Sensitivity simulations indicate that the most important factors determining HCHO concentrations are the abundances of NO, reactive hydrocarbons (e.g., isoprene), CH 3 OOH, CH 3 OH, and acetone and also the level of irradiance. In the middle and upper troposphere the most important HCHO production term is the reaction of CH 3 O 2 with NO (68 ± 10%, including a contribution of 38 ± 7% from CH 4 oxidation), followed by the destruction of CH 3 OOH by OH and photolysis (14 ± 2%), and the reaction of CH 3 OH with OH (14 ± 8%). HCHO loss is dominated by photolysis rather than by the reaction with OH, except in the upper tropospheric convective case, where the loss by OH contributes about 50%. Our measurements furthermore indicate that convective transport of H 2 O 2 can be very efficient.

Environmental context. Oceans represent 70% of the blue planet, and surprisingly, ocean emission in term of volatile organic compounds is poorly understood. The potential climate impacts on a global scale of various trace organic gases have been established, and the terrestrial inputs are well studied, but little is known about which of these can be emitted from oceanic sources. In the present study, atmospheric samples were taken over the Southern Indian Ocean, while crossing some oceanic fronts and different phytoplankton species. Such a study should aid in understanding oceanic emission, especially from phytoplankton, and will help modellers to determine concentrations of organic traces in the remote marine troposphere. Abstract. Considering its size and potential importance, the ocean is poorly characterised in terms of volatile organic compounds (VOC) that play important roles in global atmospheric chemistry. In order to better understand their potential sources and sinks over the Southern Indian Austral Ocean, shipborne measurements of selected species were made during the MANCHOT campaign during December 2004, on board the research vessel Marion Dufresne. Along the transect La Réunion to Kerguelen Island, air measurements of selected VOC (including dimethylsulfide (DMS) isoprene, carbonyls and organohalogens), carbon monoxide and ozone were performed, crossing subtropical, temperate and sub-Antarctic waters as well as pronounced subtropical and sub-Antarctic oceanic fronts. The remote marine boundary layer was characterised at latitudes 45–50°S. Oceanic fronts were associated with enhanced chlorophyll and biological activity in the seawater and elevated DMS and organohalogens in the atmosphere. These were compared with a satellite-derived phytoplankton distribution (PHYSAT). Diurnal variation for isoprene, terpenes, acetone and acetaldehyde was observed, analogously to recent results observed in mesocosm experiments.

Environmental context Volatile organic compounds are key compounds in atmospheric chemistry as precursors of ozone and secondary organic aerosols. To determine their impact at a megacity scale, a first important step is to characterise their sources. We present an estimate of volatile organic compound sources in Paris based on a combination of measurements and model results. The data suggest that the current emission inventory strongly overestimates the volatile organic compounds emitted from solvent industries, and thus needs to be corrected. Abstract A positive matrix factorisation model has been used for the determination of volatile organic compound (VOC) source contributions in Paris during an intensive campaign (May–June 2007). The major sources were traffic-related emissions (vehicle exhaust, 22% of the total mixing ratio of the measured VOCs, and fuel evaporation, 17%), with the remaining emissions from remote industrial sources (35%), natural gas and background (13%), local sources (7%), biogenic and fuel evaporation (5%) and wood-burning (2%). It was noted that the remote industrial contribution was highly dependent on the air-mass origin. During the period of oceanic influences (when only local and regional pollution was observed), this source made a relatively low contribution (<15%), whereas the source contribution linked to traffic was high (54%). During the period of continental influences (when additional continental pollution was observed), remote industrial sources played a dominant role, contributing up to 50% of measured VOCs. Finally, the positive matrix factorisation results obtained during the oceanic air mass-influenced period were compared with the local emission inventory. This comparison suggests that the VOC emission from solvent industries might be overestimated in the inventory, consistent with findings in other European cities.

Abstract. Formaldehyde (HCHO) is an important intermediate product in the photochemical degradation of methane and non-methane volatile organic compounds. In August 2001, airborne formaldehyde measurements based on the Hantzsch reaction technique were performed during the Mediterranean INtensive Oxidant Study, MINOS. The detection limit of the instrument was 42 pptv (1σ ) at a time resolution of 180 s (10-90%). The overall uncertainty of the HCHO measurements was 30% at a mixing ratio of 300 pptv. In the marine boundary layer over the eastern Mediterranean Sea average HCHO concentrations were of the order of 1500 pptv, in reasonable agreement with results from a three-dimensional global chemical transport model of the lower atmosphere including non-methane volatile organic compound (NMVOC) chemistry. Above the boundary layer HCHO mixing ratios decreased with increasing altitude to a minimum level of 250 pptv at about 7 km. At higher altitudes (above 7 km) HCHO levels showed a strong dependency on the airmass origin. In airmasses from the North Atlantic/North American area HCHO levels were of the order of 300 pptv, a factor of 6 higher than values predicted by the model. Even higher HCHO levels, increasing to values of the order of 600 pptv at 11 km altitude, were observed in easterlies transporting air affected by the Indian monsoon outflow towards the Mediterranean basin. Only a small part (∼30 pptv) of the large discrepancy between the model results and the measurements of HCHO in the free troposphere could be explained by a strong underestimation of the upper tropospheric acetone concentration by up to a factor of ten by the 3D-model. Therefore, the measurement-model difference in the upper troposphere remains unresolved, while the observed dependency of HCHO on airmass origin might indicate that unknown, relatively long-lived NMVOCs -or their reaction intermediCorrespondence to: H. Fischer (hofi@mpch-mainz.mpg.de) ates -associated with biomass burning are at least partially responsible for the observed discrepancies.