H. G. J. Smit

Measurement of ozone and water vapor by Airbus in-service aircraft: The MOZAIC airborne program, an overview

Tentative estimates, using three-dimensional chemistry and transport models, have suggested small ozone increases in the upper troposphere resulting from current aircraft emissions, but have also concluded to significant deficiencies in todays models and to the need to improve them through comparison with extended data sets. The Measurement of Ozone and Water Vapor by Airbus In-Service Aircraft (MOZAIC) program was initiated in 1993 by European scientists, aircraft manufacturers, and airlines to collect experimental data. Its goal is to help understand the atmosphere and how it is changing under the influence of human activity, with particular interest in the effects of aircraft. MOZAIC consists of automatic and regular measurements of ozone and water vapor by five long range passenger airliners flying all over the world. The aim is not to detect direct effects of aircraft emissions on the ozone budget inside the air traffic corridors but to build a large database of measurements to allow studies of chemical and physical processes in the atmosphere, and hence to validate global chemistry transport models. MOZAIC data provide, in particular, detailed ozone and water vapor climatologies at 9–12 km where subsonic aircraft emit most of their exhaust and which is a very critical domain (e.g., radiatively and stratosphere/troposphere exchanges) still imperfectly described in existing models. This will be valuable to improve knowledge about the processes occuring in the upper troposphere and the lowermost stratosphere, and the model treatment of near tropopause chemistry and transport. During MOZAIC I (January 1993–September 1996), fully automatic devices were developed, installed aboard five commercial Airbus A340s, and flown in normal airline service. A second phase, MOZAIC II, started in October 1996 with the aim of continuing the O3 and H2O measurements and doing a feasibility study of new airborne devices (CO, NOy). Between September 1994 and December 1997, 7500 flights, representing 54,000 flight hours, were made over the continents (Europe, North America, Asia, South America, and Africa) and the Atlantic Ocean. Most of the measurements (90%) correspond to cruise altitudes (9–12 km), the remaining being obtained during ascents and descents near the 50 cities frequented by MOZAIC operations. This paper reports the main characteristics of the program and the flights, with a brief summary of the general content and focus of papers already published and companion papers of this special issue. These deal with the following: description and validation of the ozone and water vapor measurement methods; presentation of an accurate ozone climatology at 9–12 km altitude, over the Northern Hemisphere (130°W–140°E; 0°–80°N), and down to 30°S over South America and Africa; comparison between a 2-year MOZAIC ozone climatology (1994–1996; 0–12 km) and a long series of older measurements made since the 1980s at 8 stations of the Ozone Sounding Network; study of ozone-rich transients, up to 500 ppbv on a horizontal scale of 5–80 km, in the upper tropical troposphere; and comparison between MOZAIC ozone data and output from the global chemistry and transport model (CTM) TOMCAT.

Observations of Near-Zero Ozone Concentrations Over the Convective Pacific: Effects on Air Chemistry

A series of measurements over the equatorial Pacific in March 1993 showed that the volume mixing ratios of ozone were frequently well below 10 nanomoles per mole both in the marine boundary layer (MBL) and between 10 kilometers and the tropopause. These latter unexpected results emphasize the enormous variability of tropical tropospheric ozone and hydroxyl concentrations, which determine the oxidizing efficiency of the troposphere. They also imply a convective short circuit of marine gaseous emissions, such as dimethyl sulfide, between the MBL and the uppermost troposphere, leading, for instance, to sulfate particle formation.

Ubiquity of quasi-horizontal layers in the troposphere

Fine laminar structures in the atmosphere have been described previously, but their characterization has been limited. The modern global coverage of aircraft flights offers an opportunity to provide such a characterization, and examine the ubiquity of such structures, in space and time. Research aircraft measuring vertical profiles of atmospheric chemical constituents frequently discern quasi-horizontal atmospheric layers with mean thicknesses of the order of 1 km and mean altitudes between 5 and 7 km (refs 10,11,12). These layers can be characterized and categorized by various combinations of ozone, water vapour, carbon monoxide and methane deviations from background profiles. Five commercial aircraft have been recently equipped to measure water vapour and ozone concentrations, and automatically collect vertical profile information on landing and take-off (refs 13,14,15). Here we synthesize measurements from both research and commercial flights and demonstrate the ubiquity in space and time of four layer types (as categorized by their chemical signatures). Up to one-fifth of the lowest 12 km of the atmosphere is occupied by such layers. We suggest that this universality reflects basic characteristics of the atmosphere hitherto unexplored, with potential implications for present understanding of a wide variety of dynamic and chemical atmospheric processes.

In-situ measurements of stratospheric ozone depletion rates in the Arctic winter 1991/1992: A Lagrangian approach

A Lagrangian approach has been used to assess the degree of chemically induced ozone loss in the Arctic lower stratosphere in winter 1991/1992. Trajectory calculations are used to identify air parcels probed by two ozonesondes at different points along the trajectories. A statistical analysis of the measured differences in ozone mixing ratio and the time the air parcel spent in sunlight between the measurements provides the chemical ozone loss. Initial results were first described by von der Gathen et al. [1995]. Here we present a more detailed description of the technique and a more comprehensive discussion of the results. Ozone loss rates of up to 10 ppbv per sunlit hour (or 54 ppbv per day) were found inside the polar vortex on the 475 K potential temperature surface (about 19.5 km in altitude) at the end of January. The period of rapid ozone loss coincides and slightly lags a period when temperatures were cold enough for type I polar stratospheric clouds to form. It is shown that the ozone loss occurs exclusively during the sunlit portions of the trajectories. The time evolution and vertical distribution of the ozone loss rates are discussed.

Regional and Global Tropopause Fold Occurrence and Related Ozone Flux Across the Tropopause

This paper gives a synthesis of three algorithms to detect the presenceof tropopause folds from vertical ozone/radio-sounding profiles and frommeteorological analysis. Also an algorithm to identify injection ofstratospheric air into the lower troposphere fromozone/7beryllium time series is presented. Differences in theresults obtained from the algorithms are observed and discussed with respectto the criteria for fold detection and input data used. Spatial gradients inthe obtained folding frequencies are made evident on a global scale from thealgorithm based on meteorological analysis (Q-vector/potential vorticity)and probably also on a regional European scale from algorithms both basedmeteorological analyses and on ozone/PTU soundings. The observed seasonalvariation of folding occurrence is rather flat except during summer whenalso some differences appear between the algorithms. By combining thefolding frequencies with literature estimates of the cross-tropopause ozonetransfer in single folding events, an average stratospheric ozone influxinto the troposphere of 5.7 ± 1.3× 1010 mol.cm-2 s-1 is obtained for the Northern hemisphereand 12± 2.7× 1010 mol. cm-2s-1 for Western Europe. Potential additional contributions dueto other stratosphere-troposphere exchange processes than folds are not yetincluded in these estimates. Finally, the link between statistics fromozone/7beryllium data and folding statistics is brieflydiscussed.

Calibration and performance of automatic compact instrumentation for the measurement of relative humidity from passenger aircraft

Compact airborne humidity sensing devices using capacitive sensors are employed on board in-service aircraft to measure water vapor concentrations in the troposphere up to 13 km altitude. The sensors are individually calibrated before onboard installation. After every 500 flight hours, each sensor is calibrated in an environmental simulation chamber under typical middle/upper tropospheric flight conditions. A Lyman-Alpha fluorescence hygrometer is used as reference instrument. Preflight and postflight calibration of each flown sensor agreed very well and showed good response. Typical overall uncertainties for the 1995 Measurement of Ozone by AIRBUS In-Service Aircraft (MOZAIC) relative humidity (RH) measurements are within ±4% RH in the middle troposphere, increasing to ±7% RH between 9 and 13 km. In-flight comparison of the MOZAIC humidity device with other water vapor measuring techniques showed agreement within ±(5–10)% RH and a time response of better than 10 s in the lower/middle troposphere, increasing to values of 1–3 min at 10–12 km altitude.

Cirrus cloud formation and ice supersaturated regions in a global climate model

At temperatures below 238 K, cirrus clouds can form by homogeneous and heterogeneous ice nucleation mechanisms. ECHAM5 contains a two-moment cloud microphysics scheme and permits cirrus formation by homogeneous freezing of solution droplets and heterogeneous freezing on immersed dust nuclei. On changing the mass accommodation coefficient, α, of water vapor on ice crystals from 0.5 in the standard ECHAM5 simulation to 0.006 as suggested by previous laboratory experiments, the number of ice crystals increases by a factor of 14, as a result of the delayed relaxation of supersaturation. At the same time, the ice water path increases by only 29% in the global annual mean, indicating that the ice crystals are much smaller in the case of low α. As a consequence, the short wave and long wave cloud forcing at the top of the atmosphere increase by 15 and 18 W m−2, respectively. Assuming heterogeneous freezing caused by immersed dust particles instead of homogeneous freezing, the effect is much weaker, decreasing the global annual mean short wave and long wave cloud forcing by 2.7 and 4.7 W m−2. Overall, these results provide little support, if any, for kinetic growth limitation of ice particles (i.e. a very low α).

Ice‐supersaturated regions and subvisible cirrus in the northern midlatitude upper troposphere

Humidity and temperature data from the Measurement of Ozone by Airbus in-service Aircraft (MOZAIC) project have been used to produce maps of probability for ice supersaturation in two 50 hPa thick layers centered around 200 and 250 hPa. As the MOZAIC data cover only international air routes, the resulting maps cover mainly the northern midlatitudes. The data of ice supersaturation have then been correlated with data of frequency of occurrence of subvisible cirrus from the Stratospheric Aerosol and Gas Experiment (SAGE II) satellite instrument. The correlation analysis provided strong indications that subvisible cirrus (SVC) is associated to ice-supersaturated regions (ISSRs), although processes are possible that can decouple SVC from ISSRs. A first trial to derive a global picture of ice supersaturation near the tropopause was performed using a measure of cirrus fractional coverage constructed from meteorological analyses of European Centre for Medium-Range Weather Forecasts and to correlate this with the supersaturation data. The correlation was only moderate (although significant), leading to the tentative conjecture that regions of frequent ice supersaturation are to be expected over the Indonesian archipelago, over the Amazonas basin, and over the northern Pacific between Japan and Canada. A final correlation analysis between the meteorological analysis data and the SVC data indicated that the formation of SVC is generally thermodynamically controlled, with the exception of the northern midlatitude SVC. The composition of the aerosol at the northern midlatitude tropopause is probably variable due to industrial emissions and air traffic. Hence the freezing properties of these particles may become important, which results in a weaker thermodynamic control of SVC formation in the northern midlatitudes.

Air pollution during the 2003 European heat wave as seen by MOZAIC airliners

This study presents an analysis of both MOZAIC profiles above Frankfurt and Lagrangian dispersion model simulations for the 2003 European heat wave. The comparison of MOZAIC measurements in summer 2003 with the 11-year MOZAIC climatology reflects strong temperature anomalies (exceeding 4 C) throughout the lower troposphere. Higher positive anomalies of temperature and negative anomalies of both wind speed and relative humidity are found for the period defined here as the heat wave (2–14 August 2003), compared to the periods before (16–31 July 2003) and after (16–31 August 2003) the heat wave. In addition, Lagrangian model simulations in backward mode indicate the suppressed long-range transport in the midto lower troposphere and the enhanced southern origin of air masses for all tropospheric levels during the heat wave. Ozone and carbon monoxide also present strong anomalies (both ∼+40 ppbv) during the heat wave, with a maximum vertical extension reaching 6 km altitude around 11 August 2003. Pollution in the planetary boundary layer (PBL) is enhanced during the day, with ozone mixing ratios two times higher than climatological values. This is due to a combination of factors, such as high temperature and radiation, stagnation of air masses and weak dry deposition, which favour the accumulation of ozone precursors and the build-up of ozone. A negligible role of a stratospheric-origin ozone tracer has been found for the lower troposphere in this study. From 29 July to 15 August 2003 forest fires burnt around 0.3×106 ha in Portugal and added to atmospheric pollution in Europe. Layers with enhanced CO and NO y mixing ratios, advected from Portugal, were crossed by the MOZAIC aircraft in the free troposphere over Frankfurt. A series of forward and backward Lagrangian model simulations have been performed to investigate the origin of anomalies durCorrespondence to: M. Tressol (marc.tressol@aero.obs-mip.fr) ing the whole heat wave. European anthropogenic emissions present the strongest contribution to the measured CO levels in the lower troposphere (near 30%). This source is followed by Portuguese forest fires which affect the lower troposphere after 6 August 2003 and even the PBL around 10 August 2003. The averaged biomass burning contribution reaches 35% during the affected period. Anthropogenic CO of North American origin only marginally influences CO levels over Europe during that period.

Intercomparison campaign of vertical ozone profiles including electrochemical sondes of ECC and Brewer-Mast type and a ground based UV-differential absorption lidar

An intercomparison campaign was conducted at the Observatoire de Haute Provence (OHP) in Southern France in September 1989 in order to compare the three instruments used for vertical tropospheric ozone profiling in the European TOR (Tropospheric Ozone Research Project) network: balloon borne ECC and Brewer-Mast sondes and a ground based UV-DIAL (DifferentialAbsorptionLidar). Additionally, a stratospheric lidar system and the Dobson spectrophotometer of the OHP were operated. Seven simultaneously measured vertical ozone profiles gave evidence for systematic differences of 15% between both types of electrochemical sondes in the troposphere, the Brewer-Mast sondes reading the smaller ozone values. These differences might be explained on the one hand by a possible contamination of the ozone sensor with reducing substances, causing a negative bias mainly for Brewer-Mast sondes and, on the other hand, by the evolution of the sonde background current during the flight, causing a positive bias for ECC sondes and a negative bias for Brewer-Mast sondes. The tropospheric lidar system, measuring the vertical ozone distribution between 6 and 12–15 km, showed ozone concentrations intermediate between the sonde results. This is in good agreement with its estimated systematic error of better than 7% in the upper troposphere. In the stratosphere, the differences between electrochemical sondes and the lidar are between 5 and 10% before the normalisation with the total ozone values measured by the Dobson spectrophotometer, and always below 5% after. While the Dobson normalisation thus corrects rather well the stratospheric part of the sonde profile, it only partially reduces errors occurring in the troposphere.