Associate Professor
Department of Climate and Space Sciences and Engineering
University of Michigan
Email: xianglei at
Phone: (734) 936-0491
Fax: (734) 936-0503
1533 Space Research Building
2455 Hayward Street, Ann Arbor, MI 48109-2143
An open project for model-based cloud radiative kernel development

by Xianglei Huang(Univ. of Michigan) and Qing Yue(JPL/Caltech)

    This is a project to build cloud radaitive kernel for each particpating climate model using the method pioneered in Yue et al. (2016) and extended into the climate model applications by Huang et al. (2019). The method utilizes the output from each individual model without engaging any offline radiative transfer calculation. As such, the radaitive kernel is derived from the statistics of each model itself and avoids the issue that different models can have very different mean states and variability in terms of clouds.

    If you are interested in participating in this project, below is a list of the variables needed for building the cloud radiative kernel. Once such set of data is provided, we will use the method described in Yue et al. (2016) to construct a set of cloud radiative kernel from the output.

    Please email us if you are interested in particpation. We will provide ftp instruction for you to upload your data.

    Experiment Set-up and List of varibles

    Slab-ocean or fully-coupled simulation with constant forcing is preferred. If not practical, AMIP-type simulation with climatological SST is recommended.

    Please run the model for 10 years (spin-up time not included), and archive 3-hourly instantaneous fields for following variables:

    If GCM can turn on ISCCP simulator, archive 3-hourly fields for the following variables:
  • 4. CLDTOT(2D)
  • 5. All-sky TOA LW fluxes (2D)
  • 6. All-sky TOA SW fluxes (2D)
  • 7. Clear-sky TOA LW fluxes (2D)
  • 8. Clear-sky TOA SW fluxes (2D)

  • Otherwise, archive regular 3-hourly fields for the following variables:
  • 1. Surface temperature (2D)
  • 2. Surface pressure (2D)
  • 3. Atmospheric temperature profiles (3D)
  • 4. Atmospheric specific humidity profiles (3D)
  • 5. Cloud fraction (3D)
  • 6. Liquid water content (3D)
  • 7. Ice water content (3D)
  • 8. Clear-sky longwave flux at the TOA (2D)
  • 9. Clear-sky shortwave flux at the TOA (2D)
  • 10. All-sky LW flux at the TOA (2D)
  • 11. All-sky SW flux at the TOA (2D)
  • 12. Total cloud fraction (2D)
  • (Optional)
  • 1. The effective size of ice and liquid clouds used in the model radiation scheme (3D)
  • 2. #8-#11 if a band-by-band spectral cloud radiative kernel is needed, in addition to the broadband cloud radiative kernel.
  • Remarks: 1. Instantaneous fields are preferred. But if only 3-hourly averaged fields are available, they can be used as well. 2. If the data is not in netcdf format, please make sure to provide a metadata to describe the units for all variables. 3. To accumulate sufficient statistics, it is desirable to have 10 years of 3-hourly output.

  • Yue, Q., B. H. Kahn, E. J. Fetzer, M. Schreier, S. Wong, X. H. Chen, X. L. Huang, Observation-based Longwave Cloud Radiative Kernels Derived from the A-Train, J. Climate, 29, 2023-2040, 2016.
  • Huang, X.L., X. H. Chen, Q. Yue, Band-by-band contributions to the longwave cloud radiative feedbacks, Geophysical Research Letters, 46,, 2019.

  • Acknowledgements: This project is originated from the research project awarded to Qing Yue and Xianglei Huang by the NASA CloudSat/CALIPSO program.