The main objective of this proposal is to study aerosol effects on warm ice clouds. These are mixed-phase clouds that contain both ice particles and supercooled liquid droplets, and can occur between 0C and -40C. Whether a cloud is composed of ice or liquid is crucial for determining how much of the incoming radiation from the sun it reflects back to space and therefore crucial for climate prediction. The PI is interested in how dust aerosols and smoke from biomass burning or other anthropogenic aerosols acting as ice nuclei influence the microphysical and radiative properties of these clouds as well as the extent to which precipitation formation processes are altered. To achieve this goal an aerosol model has been coupled with the National Center for Atmospheric Research (NCAR) General Circulation Model (GCM), Community Atmospheric Model (CAM). As part of the PI's present research, she has implemented ice crystal mixing ratio and number concentration prognostic equations in the NCAR CAM in order to study aerosol-cirrus-climate effects, based on a new parameterization for the effects of dust, soot, and sulfur aerosols on these cold clouds. This study will extend this parameterization to treat deposition nucleation and contact nucleation in warm ice clouds (T > -40C). Data developed during the forthcoming Ice in Clouds Experiments (ICE) will be used to guide the development of the parameterization. A cloud parcel model with size-resolved microphysics will be used to both simulate conditions during ICE and to develop the ice nucleation parameterization. This parameterization will then be included in the ice phase microphysical processes in the CAM model. CAM coupled to the IMPACT aerosol model will be used for simulations for the present-day and pre-industrial climate to assess the effects of anthropogenic aerosols. The PI will also examine the possibility of trends observed in dust concentration and their effect on ice clouds and climate. Available data from satellite and lidar observations will be used to evaluate her results.
In addition to the project outlined above, the PI plans to continue her collaboration in the European-based Quantify Project. Quantify is a European-funded project under the Sixth Framework Programme with 40 partners and 8 separate work packages. The main goal of QUANTIFY is to quantify the climate impact of global and European transport systems for the present situation and for several scenarios of future development. The PI's main effort in Quantify will be to compare the predictions of her coupled aerosol/climate model (the IMPACT model coupled to the NCAR CAM model) with ice microphysics to the predictions of a cloud-resolving model from the Danish Meteorological Institute.
The intellectual merit of this project lays in its contribution to an understanding of the possible role of both dust and anthropogenic ice nuclei in indirect effects on climate.
The broader merits of the project lie in its contribution to the science that is useful for policy decisions. Moreover, graduate students will participate in the research, so that this project will contribute to graduate training. Additionally, during March 2007, Penner will be teaching the "clouds and precipitation" course at Michigan. Real time data from the ICE experiments planned for this time frame will be used in an end-of-term project to acquaint students with the role of field experiments in understanding ice nucleation phenomena.
The main objective of this proposal was to study aerosol effects in warm ice clouds. These are mixed-phase clouds that contain both ice particles and supercooled liquid droplets, and can occur between 0°C and -40°C. We were interested in how dust aerosols and smoke from biomass burning or other anthropogenic aerosols acting as ice nuclei influence the microphysical and radiative properties of these clouds as well as the extent to which precipitation formation processes are altered. We studied the formation of ice and precipitation using 4 different ice nucleation parameterizations: 2 for deposition/condensation/immersion (DCI) freezing and 2 for contact freezing. The Phillips et al. (2008) parameterization of DCI ice nucleation caused ice concentrations in the model that reflected the variations in aerosol concentrations. This causes larger ice water paths in the Northern Hemisphere associated with the larger aerosol concentrations there than was the case with the older Meyers et al. (1992) parameterization. In addition, the Phillips et al. (2008) contact ice nuclei parameterization predicts three orders of magnitude less contact IN than the older Young (1974) parameterization. In the cases using the Young (1974) parameterization, the effect of contact IN dominates in mixed-phase clouds, which makes the changes predicted from using different parameterizations for the DCI IN unimportant. The larger ice crystal number concentrations produced using the Young (1974) parameterization depletes water droplets and leads to smaller LWPs. In addition, more cloud ice water and smaller ice crystal effective radius are predicted as a result. The global average effective droplet radius, droplet number concentration and cloud liquid mixing ratios and LWPs decrease because of the more efficient Bergeron-Findeisen process. The influence on the cloud water field is more pronounced than that from changing the DCI parameterization. Therefore, the shortwave cloud forcing (SWCF) and longwave cloud forcing (LWCF) are smaller in magnitude for the cases that use the Young (1974) parameterization. All cases except the case using both the Phillips et al. (2008) DCI freezing and contact freezing parameterization are reasonable compared with observations of IWP in the SH. The IWP from the cases using the Phillips et al. (2008) contact freezing parameterization compare well with observations in the NH, while the cases using the Young (1974) parameterization over-predict IWP in the NH. The fact that the case using Phillips et al. (2008) DCI freezing parameterization agree fairly well with satellite observations in the NH shows that the Phillips et al. (2008) DCI freezing parameterization is able to correctly predict ice water contents using an in-line calculation of aerosol fields. The under prediction of IWP in the case using both the Phillips et al. (2008) DCI freezing and contact freezing parameterization SH, however, suggests the possibility that there are some missing sources of ice nuclei in the SH. These results are not intended to tell which of the parameterizations is best, because of the uncertainties associated with both the satellite data and the parameterizations, especially that for the contact freezing ice nuclei. The anthropogenic effects of BC/OM on mixed phase clouds were estimated. A net cloud forcing of 0.15 W/m2 is calculated for the simulations using the Phillips et al. (2008) contact freezing parameterization and 0.83 W/m2 for the simulations using the Young (1974) parameterization. The total forcing (including the effects of changes in water vapor) is 0.28 W/m2 for the cases using the Phillips et al. (2008) contact freezing parameterization, and 1.17 W/m2 for the cases using the Young (1974) parameterization. This range of forcing change reflects the uncertainty caused by different treatments of contact freezing in mixed-phase clouds. These positive forcings are expected to lead to increasing temperatures if these schemes are implemented into a climate model. Further research is needed to decrease this uncertainty. Since this project helps to define the climate forcing effects of aerosol particles they help to ascertain the uncertainty in climate modeling. Our results on mixed phase clouds are still in the process of being summarized and will be submitted to a journal for publication. We also used the codes developed for this project in a graduate level classroom project.
