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 climate models are computationally expensive. We use model output from the Weather Research Forecast (WRF) climate model to train deep neural networks (DNNs) and evaluate whether trained DNNs can provide an accurate alternative to the physics-based parameterizations. Specifically, we develop an emulator using DNNs for a planetary boundary layer (PBL) parameterization in the WRF model. PBL parameterizations are used in atmospheric models to represent the diurnal variation in the formation and collapse of the atmospheric boundary layer – the lowest part of the atmosphere. The dynamics and turbulence, as well as velocity, temperature, and humidity profiles within the boundary layer are all critical for determining many of the physical processes in the atmosphere. PBL parameterizations are used to represent these processes that are usually unresolved in a typical climate model that operates at horizontal spatial scales in the tens of kilometers. We demonstrate that a domain-aware DNN, which takes account of underlying domain structure (e.g., nonlocal mixing), can successfully simulate the vertical profiles within the boundary layer of velocities, temperature, and water vapor over the entire diurnal cycle. Results also show that a single trained DNN from one location can produce predictionsmore » of wind speed, temperature, and water vapor profiles over nearby locations with similar terrain conditions with correlations higher than 0.9 when compared with the WRF simulations used as the training dataset.« less

Authors:
; ; ORCiD logo
Publication Date:
Research Org.:
Argonne National Laboratory (ANL), Argonne, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR). Scientific Discovery through Advanced Computing (SciDAC)
OSTI Identifier:
1570018
Alternate Identifier(s):
OSTI ID: 1607653
Grant/Contract Number:  
AC02-06CH11357
Resource Type:
Journal Article: Published Article
Journal Name:
Geoscientific Model Development (Online)
Additional Journal Information:
Journal Name: Geoscientific Model Development (Online) Journal Volume: 12 Journal Issue: 10; Journal ID: ISSN 1991-9603
Publisher:
Copernicus Publications, EGU
Country of Publication:
Germany
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. Germany: 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. Germany. https://doi.org/10.5194/gmd-12-4261-2019
Wang, Jiali, Balaprakash, Prasanna, and Kotamarthi, Rao. 2019. "Fast domain-aware neural network emulation of a planetary boundary layer parameterization in a numerical weather forecast model". Germany. https://doi.org/10.5194/gmd-12-4261-2019.
@article{osti_1570018,
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 climate models are computationally expensive. We use model output from the Weather Research Forecast (WRF) climate model to train deep neural networks (DNNs) and evaluate whether trained DNNs can provide an accurate alternative to the physics-based parameterizations. Specifically, we develop an emulator using DNNs for a planetary boundary layer (PBL) parameterization in the WRF model. PBL parameterizations are used in atmospheric models to represent the diurnal variation in the formation and collapse of the atmospheric boundary layer – the lowest part of the atmosphere. The dynamics and turbulence, as well as velocity, temperature, and humidity profiles within the boundary layer are all critical for determining many of the physical processes in the atmosphere. PBL parameterizations are used to represent these processes that are usually unresolved in a typical climate model that operates at horizontal spatial scales in the tens of kilometers. We demonstrate that a domain-aware DNN, which takes account of underlying domain structure (e.g., nonlocal mixing), can successfully simulate the vertical profiles within the boundary layer of velocities, temperature, and water vapor over the entire diurnal cycle. Results also show that a single trained DNN from one location can produce predictions of wind speed, temperature, and water vapor profiles over nearby locations with similar terrain conditions with correlations higher than 0.9 when compared with the WRF simulations used as the training dataset.},
doi = {10.5194/gmd-12-4261-2019},
url = {https://www.osti.gov/biblio/1570018}, journal = {Geoscientific Model Development (Online)},
issn = {1991-9603},
number = 10,
volume = 12,
place = {Germany},
year = {Thu Oct 10 00:00:00 EDT 2019},
month = {Thu Oct 10 00:00:00 EDT 2019}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record at https://doi.org/10.5194/gmd-12-4261-2019

Citation Metrics:
Cited by: 21 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

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


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


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


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


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


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


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


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


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


Approximation by superpositions of a sigmoidal function
journal, December 1989