Filters








1,984 Hits in 4.3 sec

Predicting atmospheric optical properties for radiative transfer computations using neural networks [article]

Menno A. Veerman, Robert Pincus, Robin Stoffer, Caspar van Leeuwen, Damian Podareanu, Chiel C. van Heerwaarden
2020 arXiv   pre-print
To minimize computational costs, we reduce the range of atmospheric conditions for which the neural networks are applicable and use machine-specific optimised BLAS functions to accelerate matrix computations  ...  Predicted optical properties are highly accurate and the resulting radiative fluxes have average errors within 0.5 compared to RRTMGP.  ...  The de Arellano for valuable discussions. RP is grateful to the conference organisers for the invitation to speak and motivation to think through the problem.  ... 
arXiv:2005.02265v3 fatcat:p4mhnxbv6bbjlpzfitylu6qnse

Neural network radiative transfer for imaging spectroscopy

Brian D. Bue, David R. Thompson, Shubhankar Deshpande, Michael Eastwood, Robert O. Green, Vijay Natraj, Terry Mullen, Mario Parente
2019 Atmospheric Measurement Techniques  
This study demonstrates that nonparametric function approximation with neural networks can replicate radiative transfer calculations and generate accurate radiance spectra at multiple wavelengths over  ...  RTMs' computational expense makes them difficult to use with high-volume imaging spectrometers, and forces approximations such as lookup table interpolation and surface–atmosphere decoupling.  ...  Neural networks for radiative transfer modeling Our goal is to construct a model that emulates a firstprinciples RTM using precalculated outputs generated by that RTM for a representative set of atmospheric  ... 
doi:10.5194/amt-12-2567-2019 fatcat:scohxrv4q5chjntmjdbe7avbm4

Visualizing Model Data Using a Fast Approximation of a Radiative Transfer Model

Valliappa Lakshmanan, Robert Rabin, Jason Otkin, John S. Kain, Scott Dembek
2012 Journal of Atmospheric and Oceanic Technology  
The resulting approximation is very close to the result derived from the complex radiative transfer model and has the advantage that it can be computed in a small fraction of the time required by the forward  ...  A forward radiative transfer model is capable of providing such a visible-channel depiction of numerical weather prediction model output, but present-day forward models are too slow to run routinely on  ...  The authors are grateful for support from the NOAA High Performance Computing and Communication (HPCC) program, which made this project possible.  ... 
doi:10.1175/jtech-d-11-00007.1 fatcat:fdtldqgmqbbbfminhhi2lba3fi

Deep-learning enhancement of large scale numerical simulations [article]

Caspar van Leeuwen, Damian Podareanu, Valeriu Codreanu, Maxwell X. Cai, Axel Berg, Simon Portegies Zwart, Robin Stoffer, Menno Veerman, Chiel van Heerwaarden, Sydney Otten, Sascha Caron, Cunliang Geng (+2 others)
2020 arXiv   pre-print
This type of application, deep learning for high-performance computing, is the theme of this whitepaper.  ...  Recently deep learning has been employed to enhance solving problems that traditionally are solved with large-scale numerical simulations using HPC.  ...  Hinton, "ImageNet Classification with Deep Convolutional Neural Networks," NIPS, 2012.  ... 
arXiv:2004.03454v1 fatcat:l4vs2ham6ngdjdvio66uieb5my

ClimART: A Benchmark Dataset for Emulating Atmospheric Radiative Transfer in Weather and Climate Models [article]

Salva Rühling Cachay, Venkatesh Ramesh, Jason N. S. Cole, Howard Barker, David Rolnick
2021 arXiv   pre-print
Within weather and climate models, atmospheric radiative transfer (RT) calculations are especially expensive. This has made them a popular target for neural network-based emulators.  ...  We also present several novel baselines that indicate shortcomings of datasets and network architectures used in prior work.  ...  The solar and thermal radiative transfer is computed using a 2-stream solution [29] .  ... 
arXiv:2111.14671v1 fatcat:rte6f5xvfvgifedloly4gjccy4

RAD-NNET, a neural network based correlation developed for a realistic simulation of the non-gray radiative heat transfer effect in three-dimensional gas-particle mixtures

Walter W. Yuen
2009 International Journal of Heat and Mass Transfer  
A neural network correlation, RAD-NNET, is developed to simulate the realistic effect of non-gray radiative absorption by a homogeneous mixture of combustion gases (CO 2 and H 2 O) and soot using numerical  ...  RAD-NNET is then applied to assess the accuracy of some commonly accepted approximate approaches to evaluate radiative heat transfer in three-dimensional non-gray media.  ...  In heat transfer, the concept of neural network has also been used extensively in areas such as retrieval of optical properties [13] [14] [15] , inverse heat transfer problems [16] [17] [18] and heat  ... 
doi:10.1016/j.ijheatmasstransfer.2009.01.041 fatcat:fge7a673zngvrdjtrcznllrfwq

A feasibility study to use machine learning as an inversion algorithm for aerosol profile and property retrieval from multi-axis differential absorption spectroscopy measurements

Yun Dong, Elena Spinei, Anuj Karpatne
2020 Atmospheric Measurement Techniques  
Our method relies on a multi-output sequence-to-sequence model combining convolutional neural networks (CNNs) for feature extraction and long short-term memory networks (LSTMs) for profile prediction.  ...  The model was trained and evaluated using data simulated by Vector Linearized Discrete Ordinate Radiative Transfer (VLIDORT) v2.7, which contains 1 459 200 unique mappings.  ...  The model describes atmospheric radiative transfer processes connecting the atmospheric states and the measurements.  ... 
doi:10.5194/amt-13-5537-2020 fatcat:fk3aapmrqfhw3ay4bblbq2ly4e

Zero-Shot Learning of Aerosol Optical Properties with Graph Neural Networks [article]

Kara D. Lamb, Pierre Gentine
2021 arXiv   pre-print
Accurate and fast calculations of BC optical properties are needed for remote sensing inversions and for radiative forcing calculations in atmospheric models, but current methods to accurately calculate  ...  Recent advances in machine learning approaches have shown the potential of graph neural networks (GNN's) for various physical science applications, demonstrating skill in generalizing beyond initial training  ...  Supplementary Information: Zero-Shot Learning of Aerosol Optical Properties with Graph Neural Networks Kara D.  ... 
arXiv:2107.10197v1 fatcat:xlwopwx4crf3zc7mqhlf7xjq6e

Microwave radiometric technique to retrieve vapor, liquid and ice. I. Development of a neural network-based inversion method

L. Li, J. Vivekanandan, C.H. Chan, Leung Tsang
1997 IEEE Transactions on Geoscience and Remote Sensing  
A parameterized radiative transfer model is used to quantify radiation emanating from the atmosphere.  ...  Success of a neural network-based approach is demonstrated using a simulated time series of vapor, liquid, and ice.  ...  Hwang of the University of Washington for developing the neural network models used in this research and NOAA's Environment Technology Laboratory which operates radiometer and K-band radar.  ... 
doi:10.1109/36.563260 fatcat:ntoinrftu5cipoxv5c7zp5hpnu

A neural network radiative transfer model approach applied to TROPOMI's aerosol height algorithm

Swadhin Nanda, Martin de Graaf, J. Pepijn Veefkind, Mark ter Linden, Maarten Sneep, Johan de Haan, Pieternel F. Levelt
2019 Atmospheric Measurement Techniques Discussions  
</strong> To retrieve aerosol properties from satellite measurements of the oxygen A-band in the near infrared, a line-by-line radiative transfer model implementation requires a large number of calculations  ...  This paper proposes a forward modeling approach using artificial neural networks to speed up the retrieval algorithm.  ...  Chimot et al. (2017) describe an artificial neural network approach using the same radiative transfer model as for TROPOMI to generate training data, in combination with the NASA MODIS aerosol optical  ... 
doi:10.5194/amt-2019-143 fatcat:juyzu3ha6vcwzgk4nvr3muhgui

SunnyNet: A neural network approach to 3D non-LTE radiative transfer

Bruce A. Chappell, Tiago M. D. Pereira
2021 Astronomy and Astrophysics  
The network was then used to predict non-LTE populations for other 3D simulations, and synthetic spectra were computed from its predicted non-LTE populations.  ...  We develop a machine learning based method to speed up 3D non-LTE radiative transfer calculations in optically thick stellar atmospheres. Methods.  ...  Computational resources have been provided by UNINETT Sigma2 -the National Infrastructure for High Performance Com-puting and Data Storage in Norway and by the High End Computing (HEC) division of NASA  ... 
doi:10.1051/0004-6361/202142625 fatcat:a36naoqlfffstikppwh3lsa32m

Neural Network Reflectance Prediction Model for Both Open Ocean and Coastal Waters

Lipi Mukherjee, Peng-Wang Zhai, Meng Gao, Yongxiang Hu, Bryan A. Franz, Jeremy Werdell
2020 Remote Sensing  
This supervised model is trained using a large volume of data derived from radiative transfer simulations for coupled atmosphere and ocean systems using the successive order of scattering technique (SOS-CAOS  ...  In this work, we report a fast model based on machine learning techniques, called Neural Network Reflectance Prediction Model (NNRPM), which can be used to predict ocean bidirectional polarized reflectance  ...  Acknowledgments: The hardware used in the computational studies is part of the UMBC High Performance Computing Facility (HPCF). The facility is supported by the U.S.  ... 
doi:10.3390/rs12091421 fatcat:rqxy5tyu3jbzpmciicvn3jma2e

A neural network radiative transfer model approach applied to the Tropospheric Monitoring Instrument aerosol height algorithm

Swadhin Nanda, Martin de Graaf, J. Pepijn Veefkind, Mark ter Linden, Maarten Sneep, Johan de Haan, Pieternel F. Levelt
2019 Atmospheric Measurement Techniques  
To retrieve aerosol properties from satellite measurements of the oxygen A-band in the near-infrared, a line-by-line radiative transfer model implementation requires a large number of calculations.  ...  This paper proposes a forward modelling approach using artificial neural networks to speed up the retrieval algorithm.  ...  The synthetic data were produced using the DISAMAR radiative transfer model; therefore, we expect the online radiative transfer retrievals to be generally better than the NN-based retrievals.  ... 
doi:10.5194/amt-12-6619-2019 fatcat:dxn5gdjw6bheldbjzgjgbulajm

Neural Network Radiative Transfer for Imaging Spectroscopy

Brian D. Bue, David R. Thompson, Shubhankar Deshpande, Michael Eastwood, Robert O. Green, Terry Mullen, Vijay Natraj, Mario Parente
2019 Atmospheric Measurement Techniques Discussions  
This study demonstrates that nonparametric function approximation with neural networks can replicate Radiative Transfer calculations over a relevant range of surface/atmosphere parameters.  ...  RTMs' computational expense makes them difficult to use with high volume imaging spectrometers, and forces approximations such as lookup table interpolation and surface/atmosphere decoupling.  ...  Neural Networks for Radiative Transfer Modeling The objective of our function approximation is to map a surface/atmosphere state vector x ∈ R m to a vector of observed at-20 sensor radiances y ∈ R n measured  ... 
doi:10.5194/amt-2018-436 fatcat:ijn52p2utbfbdceybst4ndqcoi

Retrieval of Optical Constant and Particle Size Distribution of Particulate Media Using the PSO-Based Neural Network Algorithm [chapter]

Hong Qi, Ya-Tao Ren, Jun-You Zhang, Li-Ming Ruan, He-Ping Tan
2016 New Applications of Artificial Intelligence  
An improved neural network algorithm was proposed and applied to the inverse radiative problems.  ...  A multi-strategy particle swarm optimization was applied to improve the performance of the back propagation multi-layer feed-forward neural network algorithm.  ...  National Natural Science Foundation of China (Nos. 51476043 and 51576053), the Major National Scientific Instruments and Equipment Development Special Foundation of China (No. 51327803), and the Foundation for  ... 
doi:10.5772/62446 fatcat:iliijd3dwfdodpg4c7fu4wv5wy
« Previous Showing results 1 — 15 out of 1,984 results