cam5-kernels

Demos and tools for radiative feedback kernels from CESM-CAM5

---

Citation

Download (and cite) the data associated with these scripts: Angeline G. Pendergrass (2017) , and this software

Scientific reference for the data associated with these scripts:

Pendergrass, A.G., Andrew Conley and Francis Vitt: Surface and top-of-atmosphere radiative feedback kernels for CESM-CAM5. Earth System Science Data. doi:10.5194/essd-2017-108.

---

Quick Start

You'll need to have NCL and Matlab.

1. Download the kernels, forcing, and demo data here: https://zenodo.org/record/997902
2. Unzip them.
   tar -xvf cam5-kernels.tar Organize the files as follows:

mv *.kernel.nc kernels/
mv PS.nc kernels/
mkdir forcing
mv *.forcing.nc forcing/
mkdir demodata
mv *fields.nc demodata/
mkdir scripts
mv *.ncl scripts/

3. Get the code here into the cam5-kernels directory.
   a. Download the latest release as a .tar.gz and then:
   tar -xvvzf cam5-kernels-0.0.tar.gz -C cam5-kernels/ --strip-components=1
   b. An alternative if you have git installed:
   cd cam5-kernels/
rm -fr scripts/
git init
git remote add origin https://github.com/apendergrass/cam5-kernels.git
git pull origin master
   Either way, you should have tools/ sitting alongside kernels/, forcing/, and demodata/; and scripts/ will be replaced with the up-to-date version.
4. Then you are ready for a test run! Try the things below - they use the included demo data. Expected results are included; if you get a different result, something might have gone wrong with your installation.
5. After the test run, you can go through and replace everything from demodata/ with your own data.

Basic package

Calculate the necessary components and calculate the temperature, water vapor, and surface albedo feedbacks on CESM hybrid sigma-pressure levels.
ncl scripts/calcp.ncl
ncl scripts/calcdq1k.ncl
matlab -nosplash -nodesktop -r "addpath ./scripts; kernel_demo"

Expected result:

Temperature feedback: -3.7261 W m^-2 K^-1
Surface albedo feedback: 0.53264 W m^-2 K^-1
Water vapor feedback: Water vapor feedback: 2.0881 W m^-2 K^-1

Conversion from hybrid sigma-pressure to standard pressure vertical coordinate

Standard CMIP tropospheric pressure levels are the script's default. Stratospheric extension is also included.
ncl scripts/calcp.ncl
ncl tools/convert_base_to_plevs.ncl
ncl tools/calcdp_plev.ncl
ncl tools/convert_change_to_plevs.ncl
ncl tools/t_kernel_to_plev.ncl
ncl tools/q_kernel_to_plev.ncl
matlab -nosplash -nodesktop -r "addpath ./tools; kernel_demo_pressure"

Expected result:

Temperature feedback: -3.7424 W m^-2 K^-1
Surface albedo feedback: 0.53264 W m^-2 K^-1
Water vapor feedback: 1.5897 W m^-2 K^-1

Logarithmic moisture

For the water vapor kernel, use the logarithm of mositure instead of moisture itself as the independent variable. This is more accurate because the radiative effects of water vapor scale with its logarithm.

ncl scripts/calcp.ncl
matlab -nosplash -nodesktop -r "addpath ./tools; logq_demo"

Expected result:

Water vapor feedback: 1.6158 W m^-2 K^-1

Planck and lapse rate feedback decomposition

Calculate the Planck and lapse rate feedbacks
matlab -nosplash -nodesktop -r "addpath ./tools; planck_lapserate_demo"

Expected result:

Planck feedback: -3.1844 W m^-2 K^-1
Lapse rate feedback: -0.54166 W m^-2 K^-1
Temperature feedback: -3.7261 W m^-2 K^-1
Planck+lapse rate components: -3.7261 W m^-2 K^-1

Cloud feedback

Calculate LW and SW cloud feedbacks
matlab -nosplash -nodesktop -r "addpath ./tools; cloud_feedback_demo"

Expected result:

LW Cloud Feedback: -0.055163 W m^-2 K^-1
SW Cloud Feedback: 0.57637 W m^-2 K^-1

---

Acknowledgements

Andrew Conley and Francis Vitt (NCAR): help setting up the PORT calculation to make the kernels.
Thanks to Ryan Kramer (University of Miami), William Frey (University of Colorado), Paulo Ceppi (University of Reading), and Hanjun Kim (Pusan University) for feedback, testing, and bug identification.

---

Questions/comments? Post in the issues tab, or email me at apgrass at ucar dot edu