Versions Compared

Old Version 22

changes.mady.by.user Shaocheng Xie

Saved on

compared with

New Version 23

changes.mady.by.user Yuying Zhang

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

The Design Document page provides a description of the algorithms, implementation and planned testing including unit, verification, validation and performance testing. Please read  Step 1.3 Performance Expectations that explains feature documentation requirements from the performance group point of view.

...

The purpose of this task is to improve the capability of the satellite simulators to facilitate an "apple-to-apple" comparison of modeled clouds with satellite observations. COSP  implement the new COSP code to the ACME model. This work includes: 1) upgrade the COSP code for improving its computational efficiency and adding Calipso cloud phase diagnostics and 2) add aerosol diagnostics in the lidar simulator.

 

Requirements


Date last modified: Sep 11, 2015 
Contributors: Yuying Zhang

  1. replace COSP version1.3 using COSP version1.4 in the ACME model
  2. modify the interface code
  3. add radiatively active snow in all simulators with fixed lidar snow treatment error

Algorithmic Formulations

Design solution: reference

Date last modified: Sep 11, 2015
Contributors: Yuying Zhang


A. Bodas-Salcedo, M. J. Webb, S. Bony, H. Chepfer, J.-L. Dufresne, S. A. Klein, Y. Zhang, R. Marchand, J. M. Haynes, R. Pincus, V. O. John, 2011: COSP: satellite simulation software for model assessment. BAMS.  http://journals.ametsoc.org/doi/pdf/10.1175/2011BAMS2856.1

J. E. Kay, B. R. Hillman, S. A. Klein, Y. Zhang, B. Medeiros, R. Pincus, A. Gettelman, B. Eaton, J. Boyle, R. Marchand, and T. P. Ackerman, 2012: Exposing Global Cloud Biases in the Community Atmosphere Model (CAM) Using Satellite Observations and Their Corresponding Instrument Simulators. J. Climate, 25, 5190–5207.
doi: http://dx.doi.org/10.1175/JCLI-D-11-00469.1

Design and Implementation

Date last modified: Sep 11, 2015
Contributors: Yuying Zhang

 

  1. Start from the ACME model in the "master" branch
  2. Made the changes according to the interface code for the COSP1.4 implementation in CAM (bug identified version cam5_3_88)
  3. Fixed the bug identified by Po-Lun Ma for the assignment of convective ice mixing ratio
  4. Replaced the old COSP code with the new version and added the radiatively active snow in all simulators
  5. After the test, push the COSP branch to the GitHub server. Set the pull request and ask for the integrator to merge the changes to the master.

 

Planned Verification and Unit Testing 

Date last modified:

Sep 11

Oct 5, 2015
Contributors: Yuying Zhang

 

COSP v1.4 will be compared to COSP v1.3 for all simulators. For those simulators that are not changed, they should yield the same results. For those simulators that are upgraded, the yielded differences will be analyzed to make sure that they are from the expected changes in these simulator codes.

Planned Validation Testing 

Date last modified:

05

Oct 5, 2015

 


Contributors: Yuying Zhang

 

The output from the original model with COSP v1.3 and the modified model with the COSP1.4 upgrade are examined.

  • The output from the ISCCP/MISR/Calipso simulators are exactly the same as before.
  • The output from the Radar/MODIS simulators have slight differences that come from the changes in these simulator codes from COSP1.3 to COSP1.4.

Some selected comparisons are shown in COSPinACME.pptx

 

Planned Performance Testing 

Date last modified:

Sep 11

Oct 5, 2015
Contributors: Yuying Zhang

 

The new COSP code has been restructured to greatly improve computational efficiency (particularly for the CloudSat simulator) and Calipso cloud phase diagnostics will be output. An offline run of COSP shows ~30% reduction in computational costs. The performance test still needs to be done with inline running COSP with ACME by working on with the performance team.

 

...