Authors: |
Vladimir M. Krasnopolsky, Stephen J. Lord, Shrinivas Moorthi, and Todd Spindler
EMC/NCEP/NOAA
vladimir.krasnopolsky@noaa.gov
|
Title: | Dealing with Inhomogeneous Outputs and High Dimensionality of Neural Network Emulations of Model Physics in Numerical Climate and Weather Prediction Models |
Additional Bibliographic Information: | Conference paper, The 2009 International Joint Conference on Neural Networks, Atlanta, Georgia, June 14-19, 2009 |
Year: | 2009 |
MMAB Contribution Number: | 274 |
Keywords: | NN emulations, atmospheric physics |
Status: | submitted |
Abstract: | In this paper we discuss our pilot study where
the NN emulation technique developed previously for
emulating model radiation parameterizations was applied to
the part of the NCEP GFS model physics, GBPHYS, that is
complimentary to the radiation parameterization. The results
of the study showed that not all outputs of GBPHYS are
emulated uniformly well with the original emulation approach.
Significant differences between the radiation parameterizations
and GBPHYS block and challenges for the NN emulation
approach due to these differences are demonstrated and
discussed. Several approaches that allowed us to deal with the
challenges and that can be used to compliment the NN
emulation approach for dealing with entire model physics are
also introduced. |
Link: | /mmab/papers/tn274/V.Krasnopolsky.pdf |