Predicting atmospheric optical properties for radiative transfer computations using neural networks

Details

• Journal Title:
  Philosophical Transactions of the Royal Society A
• Personal Author:
  Veerman, Menno A. ; Pincus, Robert ; Stoffer, Robin ; van Leeuwen, Caspar M. ; Podareanu, Damian ; van Heerwaarden, Chiel C.
• NOAA Program & Office:
  OAR (Oceanic and Atmospheric Research) ; PSL (Physical Sciences Laboratory) ; CIRES (Cooperative Institute for Research in Environmental Sciences)
• Description:
  The radiative transfer equations are well known, but radiation parametrizations in atmospheric models are computationally expensive. A promising tool for accelerating parametrizations is the use of machine learning techniques. In this study, we develop a machine learning-based parametrization for the gaseous optical properties by training neural networks to emulate a modern radiation parametrization (RRTMGP). To minimize computa- tional costs, we reduce the range of atmospheric conditions for which the neural networks are applicable and use machine-specific optimized BLAS functions to accelerate matrix computations. To generate training data, we use a set of randomly perturbed atmospheric profiles and calculate optical properties using RRTMGP. Predicted optical properties are highly accurate and the resulting radiative fluxes have average errors within 0.5 W m−2 compared to RRTMGP. Our neural network-based gas optics parametrization is up to four times faster than RRTMGP, depending on the size of the neural networks. We further test the trade-off between speed and accuracy by training neural networks for the narrow range of atmospheric conditions of a single large-eddy simulation, so smaller and therefore faster networks can achieve a desired accuracy. We conclude that our machine learning-based parametrization can speed-up radiative transfer computations while retaining high accuracy.
• Source:
  Phil.Trans. R. Soc. A 379: 20200095.
• DOI:
  https://doi.org/10.1098/rsta.2020.0095
• Document Type:
  Journal Article
• Rights Information:
  CC BY
• Compliance:
  Submitted
• Download URL:
  https://repository.library.noaa.gov/view/noaa/31015/noaa_31015_DS1.pdf
• File Type:
  [PDF - 848.43 KB]
• Collection(s):
  Office of Oceanic and Atmospheric Research (OAR) ; NOAA Cooperative Institutes
• Main Document Checksum:
  urn:sha256:f4a981c1ad6a5623129ac3b49537f6fcf4671492250bd2b433a69108c1627754