Scientific Highlights

An example of the research contributions of ASP postdoctoral fellows is the research of first-year postdoc Matthias Rempel, who used two modeling approaches to investigate the nature of transport in the "overshoot region" at the base of the convection zone in the Sun. In the first, he modifed the MHD code of Mark Rast (HAO) to allow for separation of the thermal conductivity into a turbulent and radiative component. This separation enables adjustment of the convective properties apart from the radiative ones in the lower half of the convection zone. He then employed a thermal relaxation scheme to accelerate the approach to equilibrium in the deep radiative layers. These modifications allowed him to apply the model to studies of the overshoot region at the base of the solar convection zone, where crucial storage of strong toroidal magnetic field is thought to play an important role in the solar dynamo. A 2D model with these modifications was able to reproduce many characteristics of previous overshoot models through different assumptions regarding the energy flux and Prandtl number. The model also shows how sensitively the overshoot depends on the structure of convection, which has to be considered when applying numerical results to solar overshoot. The accompanying figure shows a simulation of single downflows driven by cooling from the top boundary and entering a subadiabatic layer (below z = 2). Shown are the temperature perturbations with respect to the initial stratification. The length scale is made dimensionless with the pressure scale height at z = 2. In the upper part (z < 2) the Prandtl number is 1 in both cases, whereas in the lower part (z > 2) the Prandtl number is 0.25 for the left and 4 for the right case. Even though the Prandtl influences significantly the internal structure of the plume, it has only little influence on the penetration depth.

In a parallel study Rempel (collaborating with Mark Rast, HAO) investigated properties of single downflows driven by cooling at the top boundary (see Fig.1). In contrast to the full convection simulations these studies allow a much higher resolution of a single downflow. It was shown that the Prandtl number has nearly no influence on the penetration depth of a single downflow, whereas the full convection simulations show a significant increase of the overshoot depth with decreasing Prandtl number. Thus the dependence of the overshoot depths on the Prandtl number is caused by the different thermal adjustment of the mean stratification and not by a change of the deceleration of a single downflow. With the help of a semi-analytical convection model based on the assumption that the convection is driven by dominant downflows, fundamental results of the numerical simulations could be verified: the structure of the overshoot region is mainly determined by a) the Mach number of the downflows at the base of the convection zone and b) the efficiency of the mixing between downflow material and the surrounding plasma in the overshoot region.

Mercury released to the atmosphere by biomass burning:

Senior Research Associate Hans Friedli (ASP and ACD), with colleagues from NCAR, U. Washington, Meteorological Services of Canada, Canadian Forest Service, has advanced the understanding of the cycling of mercury during biomass burning, progressing from laboratory to airborne measurements on wildland fuels and expansion to agricultural waste fuels The data indicate that all mercury contained in fuel is released as mostly elemental gaseous mercury with some fuel-dependent portion present in particulate form. The mercury release from fires in temperate US forests is relatively small (2-4 Mg/y), about 22 Mg/y for boreal forests and about 500 Mg/y for all biomass burning. These estimates reflect the carbon release from biomass burning and landscape-specific emission factors. For reference, the total mercury released to the atmosphere is about 6000 Mg/g.

Friedli and Larry Radke also measured elemental gaseous mercury from sea level to the tropopause as part of the ACE-Asia and ICTC2K2 experiments and detected its transport from the western to the eastern pacific. The ACE-Asia data are extremely complex representing dust storms, anthropogenic, volcanic and possibly biomass burning signatures contained in multiple layers. The mercury profiles measured over the eastern pacific challenge the currently accepted lifetime for elemental mercury of one year; they indicate that the lifetime may be closer to 100 days. They also identified a large new sink for mercury in the stratosphere as revealed by a stratospheric intrusion depleted of elemental gaseous mercury. An anthropogenic plume encountered off the California coast had back trajectories out of China and Mongolia passing over Korea and Japan.

With coworkers from ACD, CU and as part of the Wildland Fire R&D collaboratory, Friedli also initiated laboratory experiments to determine the type and quantity of compounds emitted from live vegetation, simulating the conditions of an approaching wildfire. An understanding of the evolution of flammable volatiles is important because they are expected to modify flammability, fire dynamics, influence combustion chemistry and physics and, if not combusted, contribute directly and indirectly to the air pollution associated with wildfires. Roasting green or dry vegetation at increasing temperature to 300°C released methanol, acetaldehyde and acetone, readily measured by proton transfer reaction mass spectroscopy. The source of these compounds is not known at this time. They may be stored in the plant, be decomposition products of wood components or be partial oxidation products.

Nonlinearities in Climate Systems:

Some issues in climate dynamics revolve around whether, on interannual timescales, the climate system can be well approximated by linear dynamics or whether it is fundamentally nonlinear. To date most work aimed at making this distinction has considered probability density functions of the state vector, nonGaussian distributions being an indicator of nonlinearity. J. Berner (ASP graduate fellow from the University of Bonn) and Grant Branstator (GTP member in CGD) investigated the use of trajectories through phase space as an alternative approach. Their results indicate that distinct signatures of nonlinearity can be seen in the distribution of mean phase space tendencies for states from an extended integration of CCM0. Furthermore, using these mean tendencies as drift vectors in stochastic models, they have produced results that indicate that nonGaussian features in the GCM's PDFs can be attributed to nonlinearities in the mean tendencies. Additional work, in conjunction with F. Selten (KNMI), has concerned identifying which circulation features are primarily responsible for the nonlinearities. They have found that much of the GCM low-frequency behavior consists of regularly propagating features whose signatures mask the nonlinearities in the system. This helps explain why the nonlinearities have not been noticed in earlier studies.