Journal cover Journal topic
Atmospheric Chemistry and Physics An interactive open-access journal of the European Geosciences Union
Atmos. Chem. Phys., 14, 8701-8721, 2014
© Author(s) 2014. This work is distributed
under the Creative Commons Attribution 3.0 License.
Research article
26 Aug 2014
Diagnosing the average spatio-temporal impact of convective systems – Part 2: A model intercomparison using satellite data
M. S. Johnston1,2, S. Eliasson2,3, P. Eriksson1, R. M. Forbes4, A. Gettelman5, P. Räisänen6, and M. D. Zelinka7 1Department of Earth and Space Sciences, Chalmers University of Technology, Gothenburg, Sweden
2Swedish Meteorological and Hydrological Institute, Norrköping, Sweden
3Department of Computer Science, Electrical and Space Engineering, Division of Space Technology, Luleå University of Technology, Kiruna, Sweden
4European Centre for Medium-Range Weather Forecasts, Reading, UK
5National Center for Atmospheric Research, Boulder, Colorado, USA
6Finnish Meteorological Institute, Helsinki, Finland
7Lawrence Livermore National Laboratory, Livermore, California, USA
Abstract. The representation of the effect of tropical deep convective (DC) systems on upper-tropospheric moist processes and outgoing longwave radiation is evaluated in the EC-Earth3, ECHAM6, and CAM5 (Community Atmosphere Model) climate models using satellite-retrieved data. A composite technique is applied to thousands of deep convective systems that are identified using local rain rate maxima in order to focus on the temporal evolution of the deep convective processes in the model and satellite-retrieved data.

The models tend to over-predict the occurrence of rain rates that are less than ≈ 3 mm h−1 compared to Tropical Rainfall Measurement Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA). While the diurnal distribution of oceanic rain rate maxima in the models is similar to the satellite-retrieved data, the land-based maxima are out of phase.

Despite having a larger climatological mean upper-tropospheric relative humidity, models closely capture the satellite-derived moistening of the upper troposphere following the peak rain rate in the deep convective systems. Simulated cloud fractions near the tropopause are larger than in the satellite data, but the ice water contents are smaller compared with the satellite-retrieved ice data. The models capture the evolution of ocean-based deep convective systems fairly well, but the land-based systems show significant discrepancies. Over land, the diurnal cycle of rain is too intense, with deep convective systems occurring at the same position on subsequent days, while the satellite-retrieved data vary more in timing and geographical location.

Finally, simulated outgoing longwave radiation anomalies associated with deep convection are in reasonable agreement with the satellite data, as well as with each other. Given the fact that there are strong disagreements with, for example, cloud ice water content, and cloud fraction, between the models, this study supports the hypothesis that such agreement with satellite-retrieved data is achieved in the three models due to different representations of deep convection processes and compensating errors.

Citation: Johnston, M. S., Eliasson, S., Eriksson, P., Forbes, R. M., Gettelman, A., Räisänen, P., and Zelinka, M. D.: Diagnosing the average spatio-temporal impact of convective systems – Part 2: A model intercomparison using satellite data, Atmos. Chem. Phys., 14, 8701-8721, doi:10.5194/acp-14-8701-2014, 2014.
Publications Copernicus