skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Fast domain-aware neural network emulation of a planetary boundary layer parameterization in a numerical weather forecast model

Abstract

Abstract. Parameterizations for physical processes in weather and climatemodels are computationally expensive. We use model output from the WeatherResearch Forecast (WRF) climate model to train deep neural networks (DNNs)and evaluate whether trained DNNs can provide an accurate alternative to thephysics-based parameterizations. Specifically, we develop an emulator usingDNNs for a planetary boundary layer (PBL) parameterization in the WRF model.PBL parameterizations are used in atmospheric models to represent thediurnal variation in the formation and collapse of the atmospheric boundarylayer – the lowest part of the atmosphere. The dynamics and turbulence, aswell as velocity, temperature, and humidity profiles within the boundarylayer are all critical for determining many of the physical processes in theatmosphere. PBL parameterizations are used to represent these processes thatare usually unresolved in a typical climate model that operates athorizontal spatial scales in the tens of kilometers. We demonstrate that adomain-aware DNN, which takes account of underlying domain structure (e.g.,nonlocal mixing), can successfully simulate the vertical profiles within theboundary layer of velocities, temperature, and water vapor over the entirediurnal cycle. Results also show that a single trained DNN from one locationcan produce predictions of wind speed, temperature, and water vapor profilesover nearby locations with similar terrain conditions with correlationshigher than 0.9more » when compared with the WRF simulations used as the trainingdataset.« less

Authors:
 [1];  [1]; ORCiD logo [1]
  1. Argonne National Lab. (ANL), Argonne, IL (United States)
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21). Scientific Discovery through Advanced Computing (SciDAC)
OSTI Identifier:
1607653
Grant/Contract Number:  
AC02-06CH11357
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Geoscientific Model Development (Online)
Additional Journal Information:
Journal Volume: 12; Journal Issue: 10; Journal ID: ISSN 1991-9603
Publisher:
European Geosciences Union
Country of Publication:
United States
Language:
English
Subject:
54 ENVIRONMENTAL SCIENCES

Citation Formats

Wang, Jiali, Balaprakash, Prasanna, and Kotamarthi, Rao. Fast domain-aware neural network emulation of a planetary boundary layer parameterization in a numerical weather forecast model. United States: N. p., 2019. Web. doi:10.5194/gmd-12-4261-2019.
Wang, Jiali, Balaprakash, Prasanna, & Kotamarthi, Rao. Fast domain-aware neural network emulation of a planetary boundary layer parameterization in a numerical weather forecast model. United States. doi:10.5194/gmd-12-4261-2019.
Wang, Jiali, Balaprakash, Prasanna, and Kotamarthi, Rao. Thu . "Fast domain-aware neural network emulation of a planetary boundary layer parameterization in a numerical weather forecast model". United States. doi:10.5194/gmd-12-4261-2019. https://www.osti.gov/servlets/purl/1607653.
@article{osti_1607653,
title = {Fast domain-aware neural network emulation of a planetary boundary layer parameterization in a numerical weather forecast model},
author = {Wang, Jiali and Balaprakash, Prasanna and Kotamarthi, Rao},
abstractNote = {Abstract. Parameterizations for physical processes in weather and climatemodels are computationally expensive. We use model output from the WeatherResearch Forecast (WRF) climate model to train deep neural networks (DNNs)and evaluate whether trained DNNs can provide an accurate alternative to thephysics-based parameterizations. Specifically, we develop an emulator usingDNNs for a planetary boundary layer (PBL) parameterization in the WRF model.PBL parameterizations are used in atmospheric models to represent thediurnal variation in the formation and collapse of the atmospheric boundarylayer – the lowest part of the atmosphere. The dynamics and turbulence, aswell as velocity, temperature, and humidity profiles within the boundarylayer are all critical for determining many of the physical processes in theatmosphere. PBL parameterizations are used to represent these processes thatare usually unresolved in a typical climate model that operates athorizontal spatial scales in the tens of kilometers. We demonstrate that adomain-aware DNN, which takes account of underlying domain structure (e.g.,nonlocal mixing), can successfully simulate the vertical profiles within theboundary layer of velocities, temperature, and water vapor over the entirediurnal cycle. Results also show that a single trained DNN from one locationcan produce predictions of wind speed, temperature, and water vapor profilesover nearby locations with similar terrain conditions with correlationshigher than 0.9 when compared with the WRF simulations used as the trainingdataset.},
doi = {10.5194/gmd-12-4261-2019},
journal = {Geoscientific Model Development (Online)},
issn = {1991-9603},
number = 10,
volume = 12,
place = {United States},
year = {2019},
month = {10}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Save / Share:

Works referenced in this record:

A Review of Planetary Boundary Layer Parameterization Schemes and Their Sensitivity in Simulating Southeastern U.S. Cold Season Severe Weather Environments
journal, June 2015

  • Cohen, Ariel E.; Cavallo, Steven M.; Coniglio, Michael C.
  • Weather and Forecasting, Vol. 30, Issue 3
  • DOI: 10.1175/WAF-D-14-00105.1

Future atmospheric methane concentrations in the context of the stabilization of greenhouse gas concentrations
journal, August 1999

  • Kheshgi, Haroon S.; Jain, Atul K.; Kotamarthi, V. R.
  • Journal of Geophysical Research: Atmospheres, Vol. 104, Issue D16
  • DOI: 10.1029/1999JD900367

A New Vertical Diffusion Package with an Explicit Treatment of Entrainment Processes
journal, September 2006

  • Hong, Song-You; Noh, Yign; Dudhia, Jimy
  • Monthly Weather Review, Vol. 134, Issue 9
  • DOI: 10.1175/MWR3199.1

Complex hybrid models combining deterministic and machine learning components for numerical climate modeling and weather prediction
journal, March 2006


Approximations of Functions by a Multilayer Perceptron: a New Approach
journal, August 1997


Use of a neural-network-based long-wave radiative-transfer scheme in the ECMWF atmospheric model
journal, January 2000

  • Chevallier, F.; Morcrette, J. -J.; Chéruy, F.
  • Quarterly Journal of the Royal Meteorological Society, Vol. 126, Issue 563
  • DOI: 10.1002/qj.49712656318

Using Ensemble of Neural Networks to Learn Stochastic Convection Parameterizations for Climate and Numerical Weather Prediction Models from Data Simulated by a Cloud Resolving Model
journal, January 2013

  • Krasnopolsky, Vladimir M.; Fox-Rabinovitz, Michael S.; Belochitski, Alexei A.
  • Advances in Artificial Neural Systems, Vol. 2013
  • DOI: 10.1155/2013/485913

Approximation capabilities of multilayer feedforward networks
journal, January 1991


A Deep Learning Algorithm of Neural Network for the Parameterization of Typhoon-Ocean Feedback in Typhoon Forecast Models
journal, April 2018

  • Jiang, Guo-Qing; Xu, Jing; Wei, Jun
  • Geophysical Research Letters, Vol. 45, Issue 8
  • DOI: 10.1002/2018GL077004

New Approach to Calculation of Atmospheric Model Physics: Accurate and Fast Neural Network Emulation of Longwave Radiation in a Climate Model
journal, May 2005

  • Krasnopolsky, Vladimir M.; Fox-Rabinovitz, Michael S.; Chalikov, Dmitry V.
  • Monthly Weather Review, Vol. 133, Issue 5
  • DOI: 10.1175/MWR2923.1

Neural Networks Technique for Filling Gaps in Satellite Measurements: Application to Ocean Color Observations
journal, January 2016

  • Krasnopolsky, Vladimir; Nadiga, Sudhir; Mehra, Avichal
  • Computational Intelligence and Neuroscience, Vol. 2016
  • DOI: 10.1155/2016/6156513

Emulation of a complex global aerosol model to quantify sensitivity to uncertain parameters
journal, January 2011

  • Lee, L. A.; Carslaw, K. S.; Pringle, K. J.
  • Atmospheric Chemistry and Physics, Vol. 11, Issue 23
  • DOI: 10.5194/acp-11-12253-2011

A Neural Network Approach for a Fast and Accurate Computation of a Longwave Radiative Budget
journal, November 1998


Parameterizations: representing key processes in climate models without resolving them: Parameterizations
journal, May 2011

  • McFarlane, Norman
  • Wiley Interdisciplinary Reviews: Climate Change, Vol. 2, Issue 4
  • DOI: 10.1002/wcc.122

Downscaling with a nested regional climate model in near-surface fields over the contiguous United States: WRF dynamical downscaling
journal, July 2014

  • Wang, Jiali; Kotamarthi, Veerabhadra R.
  • Journal of Geophysical Research: Atmospheres, Vol. 119, Issue 14
  • DOI: 10.1002/2014JD021696

Modelling climate change: the role of unresolved processes
journal, October 2005

  • Williams, Paul D.
  • Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 363, Issue 1837
  • DOI: 10.1098/rsta.2005.1676

An artificial neural network model to predict thunderstorms within 400 km 2 South Texas domains : Thunderstorm artificial neural network
journal, April 2015

  • Collins, Waylon; Tissot, Philippe
  • Meteorological Applications, Vol. 22, Issue 3
  • DOI: 10.1002/met.1499

Approximation by superpositions of a sigmoidal function
journal, December 1989

  • Cybenko, G.
  • Mathematics of Control, Signals, and Systems, Vol. 2, Issue 4
  • DOI: 10.1007/BF02551274