Development of ensemble neural network convection parameterizations for climate models using ARM data

 

Author

Vladimir Krasnopolsky — National Oceanic and Atmospheric Administration

Category

Modeling

Description

The neural network (NN) approach is formulated and used for development of a NN ensemble stochastic convection parameterization for climate models. This fast parameterization is built based on data from Cloud Resolving Model (CRM) simulations initialized with and forced by TOGA-COARE data. The SAM (System for Atmospheric Modeling), developed by D. Randall, M. Khairoutdinov, and their collaborators, has been used for CRM simulations. The observational data are also used for validation of model simulations. CRM-simulated data have been averaged and projected onto the GCM space of atmospheric states to implicitly define a stochastic convection parameterization. This parameterization is emulated using an ensemble of NNs. An ensemble of NNs with different NN parameters has been trained and tested. The inherent uncertainty of the stochastic convection parameterization derived in such a way is estimated. Due to these inherent uncertainties, NN ensemble is used to constitute a stochastic NN convection parameterization.

The developed NN convection parameterization has been tested in a diagnostic CAM (CAM-NN) run versus the control CAM run. Total precipitation (P) and cloudiness (CLD) time series, diurnal cycles, and distributions for the tropical Pacific Ocean for the parallel CAM-NN and CAM runs show similarity and consistency. The P and CLD distributions for the tropical area for the parallel runs have been analyzed first for the TOGA-COARE boreal winter season (November 1992–February 1993) and then for the winter seasons of the follow-up parallel decadal simulations. The obtained results are encouraging.

The final step and future plans of the project will include (a) preparation of a NN-convection paper in addition to our NCEP Office Note (No. 469) and (b) initiation of a new collaborative effort (that goes well beyond the current project and will require an adequate funding) with the DOE PNNL group dedicated to development of NN emulations for MMF and testing it in the PNNL CAM-MMF.

Acknowledgments: The investigators would like to thank Prof. Marat Khairoutdinov (SUNY) for providing SAM and consultations and Dr. Peter Blossey (UWA) for consultations on SAM.

Publication: Krasnopolsky, V, M Fox-Rabinovitz, A Belochitski, P Rasch, P Blossey, and Y Kogan. 2011. Development of neural network convection parameterizations for climate and NWP models using Cloud Resolving Model simulations. NCEP Office Note No. 469: http://www.emc.ncep.noaa.gov/officenotes/newernotes/on469.pdf