Sorting Out Chemistry and Dynamics: The ASTEX/MAGE Experiment

Contributed by B.J. Huebert, University of Hawaii, USA
Reprinted from IGACtivities Newsletter No. 4, March 1996.
We've all had similar experiences: your daughter received her college diploma, but you can't even identify her in the photo because she was moving just as the shutter opened. Blurring eliminated the critical information from the picture. Exactly the same thing happens when you try to study aerosol dynamics in the atmosphere, since everything is moving. The air you want to take a picture of is advecting downwind, material is being added and removed at the surface, entrainment is mixing in different air from above, dispersion is causing horizontal mixing, and vertical wind-shear is slowing down air at the surface. How can you hope to tease out the oxidation rate of sulfur dioxide or the source strength of ammonia from such a cauldron? That's the challenge ASTEX/MAGE scientists undertook in the marine atmosphere near the Azores in 1992.

The Atlantic Stratocumulus Transition Experiment (ASTEX), one of the second series of FIRE international cloud-climatology experiments, took place in June of 1992 in the stratocumulus-capped marine boundary layer. The primary purpose of ASTEX was to study the factors influencing the formation and dissipation of marine clouds. The chemical experiment within ASTEX was organized by IGAC's Marine Aerosol and Gas Exchange (MAGE) Activity. Its objective was to study air-sea exchange and the formation and transformation of marine aerosols, in part by making Lagrangian observations (moving the measuring systems to stay with the same air). Chemical instrumentation was deployed on two islands, one French and two U.S. ships, and three aircraft from the U.S. and U.K. (Huebert et al., 1996). Although the meteorological situation was complex, chemists benefited from the observations of dozens of other groups who were studying cloud physics, boundary layer dynamics, and radiative transfer (Albrecht et al., 1995). These ASTEX scientists characterized turbulent and large-scale air motions in and above the boundary layer, thus enabling MAGE scientists to quantify the impact dynamics had on chemical concentrations.

To study marine chemical processing, we developed a Lagrangian sampling strategy for repeatedly studying the same air parcel over a two-day period. This approach relied on the release of constant-density balloons which floated downwind with the parcel and radioed their GPS-derived locations to a relay of sampling aircraft. The balloons served as markers for the boundary layer airmass, which was continually being modified by chemical and energy fluxes at the surface, entrainment of free tropospheric air, wind shear within the boundary layer, horizontal dispersion, chemical reactions, and aerosol transformations.

Unfortunately, the balloons also served as excellent drizzle detectors, since a 0.1 mm thick layer of water deposited on their 1 m2 upward facing surfaces (100 grams) was enough to drive them into the ocean. None of the balloons survived more than 7 hours in our first attempt. The second try was more successful: during the second Lagrangian (L2) one balloon (#7) remained aloft for over forty-two hours, seven research flights, and two sets of observations by ships (Businger et al., 1996; Huebert et al., 1996). Incidentally, the popular hypothesis that #7's success was due to the fact that it was the only balloon with a happy face drawn on it was recently disproved in ACE-1 (see following article by T.S. Bates and J.L. Gras), where three faceless "smart" balloons designed to adjust their own buoyancy were followed for 2 days by NCAR's C130 aircraft.

While it is an oversimplification to say we studied "the same air" over a two day period, moving with the wind allowed us to see how dynamics modified a parcel of air. Conditions near the Azores made our job harder: the 2 km deep boundary layer was frequently separated into several sublayers, each of which had to be characterized. Mixing between layers and dilution by entrainment of free-tropospheric air were often the major causes of concentration changes. Using an entrainment velocity of 0.6 cm/sec, Bretherton et al. (Bretherton and Siems, 1995) computed a time scale of about 4 days for replacing the L2 boundary layer with free-tropospheric air. Chemically this usually meant that our wet, aerosol-laden boundary layer air was diluted by drier, cleaner air from above. However, late in L2 a dusty Saharan airmass passed over us, entraining large mineral aerosols into the cloud layer. This addition of large aerosols produced a noticeable modification in the microphysics and dynamics of the cloud field (Martin et al., 1995), creating larger droplets and increasing the likelihood of drizzle. We could not have sorted this out from Eulerian observations alone.

We were also able to derive the sea-to-air flux of ammonia vapor from studying the ammonia budget during L2 (Zhuang and Huebert, 1996). Biologically, this ammonia emission represents an unexpected loss of a limiting nutrient from the surface ocean. It is also important from an atmospheric standpoint, because the climatic and visibility impact of sulfuric acid droplets change dramatically when they are converted into ammonium salts. The ammonium/non-seasalt sulfate (NSS) ratio in this polluted European airmass increased with time, while the total NSS decreased. Several terms (all in units of micromoles (umol) of ammonia per square meter per day) dominated the budget (Figure 1): wet and dry deposition could have removed -4, while dilution by entrainment corresponded to a loss of -29, for a net removal flux of -33. Since the observed mixed layer concentration change (the net effect of all fluxes) was only -7, sea-to-air exchange must have provided a flux of +26 umol/m^2/day. Although substantial uncertainties remain, particularly in the estimation of the entrainment term, the Lagrangian strategy allowed us to measure enough of the air-motion terms to sort out a surface flux.

Figure 1. Schemtic of the ammonia budget for the marine boundary layer
during the second ASTEX/MAGE Lagrangian experiment.

Chemical reaction rates were also determined by this approach. A group from the University of California at Irvine used repeated Lagrangian observations from a single flight (during which the plane was advecting with the balloons) to derive information on oxidation by free radicals (Blake et al., 1996). They used measurements of hydrocarbons and halocarbons versus altitude to characterize the impact that dilution by entrainment would have on concentrations. Then they employed the differing reactivity of several species to attack by hydroxyl and chlorine radicals and the observed concentration changes with time to solve for the concentrations of each radical. Significant levels of chlorine radicals could explain why several species disappear faster than expected from the marine atmosphere, but the attempts to directly measure their concentration have been controversial. Thus, this indirect observation of their importance is a big step forward. Here again, the Irvine group's ability to quantify the impact of mixing was crucial for separating out the changes caused by these oxidants.

Another group reversed the process and used aerosol measurements to derive exchange rates between the various layers of the decoupled boundary layer (Clarke et al., 1996). They confirmed that the dynamically-derived entrainment velocity of 0.6 cm/sec for the main inversion was consistent with a simple aerosol mixing model, and concluded that the surface mixed layer entrained cloud-layer air with an effective entrainment velocity of 0.45 cm/sec. They also identified a method whereby the ratio of volatile to nonvolatile nuclei can be used to characterize mixing between air masses with different histories. This is one of many examples in which chemical and aerosol measurements were able to constrain dynamics and provide the meteorological investigators with information they could not have derived from their usual suite of observations. Clarke's group also used the Lagrangian observations to demonstrate that no aerosol nucleation had occurred in the marine boundary layer during the course of L2.

The various platforms played complementary roles. Eleven-hour impactor samples from Santa Maria Island proved to be important for estimating particle removal rates. The aircraft were able to gather vertical profiles and keep up with the tagged air masses. The two ships made unique contributions because of their ability to move to locations of interest (like the starting and ending points of the Lagrangians), stay on station around the clock, make measurements very close to the surface, and support instruments with long sampling times. Jodwalis and Benner (1996) demonstrated that a new variance method can be used to measure air/sea sulfur fluxes, based on a fast total gaseous sulfur detector. In view of the need for ways to test the wind speed-based parameterizations of dimethylsulfide (DMS) emissions from which most submicrometer marine aerosols are derived, this is a valuable addition to our arsenal. The variance method generally found larger fluxes than estimates derived from simultaneous measurements of DMS in the water and air (Blomquist et al., 1996). From the other ship, Putaud and Nguyen (1996) used measurements of DMS concentration gradients to estimate fluxes. These complementary approaches improve our ability to derive a consensus among flux estimation techniques.

Of course, these examples are just a small part of what was learned about aerosols and their source materials during the ASTEX/MAGE program. A collection of MAGE papers has been published in the February, 1996 issue of the Journal of Geophysical Research - Atmospheres and is available as a compilation from this author (huebert@soest.hawaii.edu). Most of the more dynamically-oriented ASTEX papers, many containing analyses based on the Lagrangian observational strategy, are contained in the August 15, 1995 issue of the Journal of the Atmospheric Sciences. As with all field programs, there is still much to be learned from further analysis of the ASTEX/MAGE data set (publicly available from a database maintained by John Seinfeld at the California Institute of Technology; contact Lynn Russell, lynn@aeolus.che.caltech.edu).

Policymakers rely on models of dynamics and aerosol chemistry to make informed decisions about the costs and benefits of various control strategies. Experiments like ASTEX/MAGE are essential both for properly describing the physics and chemistry of individual processes in these models and for seeing whether the models accurately predict nature's response to our emissions. Society's investment in ships, airplanes, and scientists ultimately returns savings from healthy fisheries, forests, agriculture, and industry.

References:

Albrecht, B.A., C.S. Bretherton, D. Johnson, W.H. Schubert, and A.S. Frisch, The Atlantic Stratocumulus Experiment - ASTEX, J. Atmos. Sci., 52 (16), 1995.

Blake, D.R., O.W. Wingenter, M.K. Kubo, N.J. Blake, T.W. Smith, Jr., and F.S. Rowland, Hydrocarbon and halocarbon measurements in the marine boundary layer as part of the ASTEX/MAGE Lagrangian experiments: Photochemical and dynamical indicators of atmospheric hydroxyl, atomic chlorine, and vertical mixing, J. Geophys. Res., 101 (D2), 1996.

Blomquist, B.W., A.R. Bandy, and D.C. Thornton, Sulfur gas measurements in the eastern North Atlantic Ocean during ASTEX/MAGE, J. Geophys. Res., 101 (D2), 1996.

Bretherton, C.S., and S.T. Siems, Cloudiness and marine boundary layer dynamics in the ASTEX Lagrangian experiments. Part II. Cloudiness, drizzle, surface fluxes and entrainment, J. Atmos. Sci., 52 (16), 2724-2735, 1995.

Businger, S., S. Chiswell, W.C. Ulmer, and R. Johnson, Balloons as a Lagrangian measurement platform for atmospheric research, J. Geophys. Res., 101 (D2), 1996.

Clarke, A.D., T. Uehara, and J.N. Porter, Lagrangian evolution of an aerosol column during the Atlantic Stratocumulus Transition Experiment, J. Geophys. Res., 101 (D2), 1996.

Huebert, B.J., A.A.P. Pszenny, and B. Blomquist, The ASTEX/MAGE Experiment, J. Geophys. Res., 101 (D2), 1996.

Jodwalis, C.M., and R.L. Benner, Sulfur gas fluxes and horizontal inhomogeneities in the marine boundary layer, J. Geophys. Res., 101 (D2), 1996.

Martin, G.M., D.W. Johnson, D.P. Rogers, P.R. Jonas, P. Minnis, and D.A. Hegg, Observation of the interaction between cumulus clouds and warm stratocumulus clouds in the marine boundary layer during ASTEX, J. Atmos. Sci., 52, 2902-2922, 1995.

Putaud, J.P., and B.C. Nguyen, Assessment of dimethylsulfide sea-air exchange rate, J. Geophys. Res., 101 (D2), 1996..

Zhuang, L., and B.J. Huebert, A Langrangian analysis of the total ammonia budget during ASTEX/MAGE, J. Geophys. Res., 101 (D2), 1996.

Back to IGAC Highlights Page.