Filters








309 Hits in 4.0 sec

Machine Learning Emulation of 3D Cloud Radiative Effects [article]

David Meyer, Robin J. Hogan, Peter D. Dueben, Shannon L. Mason
2021 arXiv   pre-print
Here we propose to correct the European Centre for Medium-Range Weather Forecasts 1D radiation scheme ecRad for 3D cloud effects using computationally cheap neural networks. 3D cloud effects are learned  ...  Thus, rather than emulating the whole of SPARTACUS, we keep Tripleclouds unchanged for cloud-free parts of the atmosphere and 3D-correct it elsewhere.  ...  Here, the hybrid physical machine learning model improves Tripleclouds with an important correction for 3D effects rather than seeking the emulation of the whole SPARTACUS scheme.  ... 
arXiv:2103.11919v2 fatcat:azzin6bb3rf73lqvggnicigpn4

Technical note: Parameterising cloud base updraft velocity of marine stratocumuli

Jaakko Ahola, Tomi Raatikainen, Muzaffer Ege Alper, Jukka-Pekka Keskinen, Harri Kokkola, Antti Kukkurainen, Antti Lipponen, Jia Liu, Kalle Nordling, Antti-Ilari Partanen, Sami Romakkaniemi, Petri Räisänen (+2 others)
2022 Atmospheric Chemistry and Physics  
In addition, we present two different machine learning methods (Gaussian process emulation and random forest) that account for different boundary layer conditions and cloud properties.  ...  Finally, we apply these machine learning methods to find the key parameters affecting cloud base updraft velocities.  ...  This research has been supported by the H2020 European Research Council (ECLAIR (grant no. 646857) and FORCeS (grant no. 821205)) and the Academy of Finland (grant nos. 322532 and 309127).  ... 
doi:10.5194/acp-22-4523-2022 fatcat:etj6cv5225fqrisv3gub32f6pi

Process‐based climate model development harnessing machine learning: III. The Representation of Cumulus Geometry and their 3D Radiative Effects

Najda Villefranque, Stéphane Blanco, Fleur Couvreux, Richard Fournier, Jacques Gautrais, Robin J. Hogan, Frédéric Hourdin, Victoria Volodina, Daniel Williamson
2021 Journal of Advances in Modeling Earth Systems  
The radiative effect of cumulus  ...  Cloud-radiation interactions, through their strong impact on the Earth's global energy balance (Ramanathan et al., 1989) , are key processes in the evolution of the Earth's climate.  ...  Radiatively effective cloud scale.  ... 
doi:10.1029/2020ms002423 fatcat:hlykrhpqbba2dgrrtlpbs736da

Process‐based climate model development harnessing machine learning: I. a calibration tool for parameterization improvement

Fleur Couvreux, Frédéric Hourdin, Daniel Williamson, Romain Roehrig, Victoria Volodina, Najda Villefranque, Catherine Rio, Olivier Audouin1, James Salter, Eric Bazile, Florent Brient, Florence Favot (+5 others)
2020 Journal of Advances in Modeling Earth Systems  
HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not.  ...  By developing fast and accurate radiative tools that account for the full 3D radiative transfer in LES cloud scene, as proposed by Villefranque et al. (2019) , we can compute many types of radiative metrics  ...  In this paper we propose to harness machine learning to improve physical parameterizations.  ... 
doi:10.1029/2020ms002217 fatcat:kjwyta7f4rd47pgxmoto3pw3hi

Process‐based climate model development harnessing machine learning: II. model calibration from single column to global

Frédéric Hourdin, Daniel Williamson, Catherine Rio, Fleur Couvreux, Romain Roehrig, Najda Villefranque, Ionela Musat, Laurent Fairhead, F. Binta Diallo, Victoria Volodina
2020 Journal of Advances in Modeling Earth Systems  
HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not.  ...  A lot can be done for radiative effect of clouds with 1-year long simulations forced by SST, which already means hundreds of simulations.  ...  when exploring very different regimes from those which were explored in the SCM/LES machine learning sequence.  ... 
doi:10.1029/2020ms002225 fatcat:aumbtgqkh5au3mjsnedcqbwlum

Route towards complete 3D hydro-chemical simulations of companion-perturbed AGB outflows [article]

Silke Maes, Lionel Siess, Ward Homan, Jolien Malfait, Frederik De Ceuster, Thomas Ceulemans, Dion Donne, Mats Esseldeurs, Leen Decin
2022 arXiv   pre-print
Currently, a handful of simulations exist, albeit they mainly focus on the hydrodynamics of the outflow.  ...  In order to improve our understanding about these systems, theoretical studies are needed in the form of numerical simulations.  ...  Indeed, machine learning can be used to construct a 'chemistry emulator' that we will be trained to produce the same results as a chemical code, only in a much faster way.  ... 
arXiv:2206.12278v1 fatcat:jkkqs4jta5bnlc7ufzerxhqnc4

Using Machine Learning for Model Physics: an Overview [article]

Vladimir Krasnopolsky, Aleksei A. Belochitski
2022 arXiv   pre-print
Machine learning (ML) tools that can be used to emulate and/or approximate mappings are introduced.  ...  Applications of ML to emulate existing parameterizations, to develop new parameterizations, to ensure physical constraints, and control the accuracy of developed applications are described.  ...  the machine learning problem (Vapnik 2019) .  ... 
arXiv:2002.00416v2 fatcat:xthy77425rhcjjzgwdxxuyhame

Improved weather forecasting using neural network emulation for radiation parameterization

Hwan‐Jin Song, Soonyoung Roh
2021 Journal of Advances in Modeling Earth Systems  
The post-processing of numerical model outputs is the most typical example of AI application to numerical weather and climate forecasting (  ...  Recent advances in artificial intelligence (AI) techniques have provided new challenges in the development of theory-based numerical weather-climate prediction models (Hutson, 2020; Reichstein et al.,  ...  We also acknowledge the dedicated efforts of Vijay Tallapragada and Hyesook Lee for supporting international joint research between the NOAA and the KMA since 2018 to develop physics emulators for numerical  ... 
doi:10.1029/2021ms002609 fatcat:hc66athv3bhodgr2pcbz3rl4qq

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
This has led to a growing interest in replacing subroutines that explicitly compute physical processes with approximate machine learning (ML) methods that are fast at inference time.  ...  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.  ...  to the effects of climate change.  ... 
arXiv:2111.14671v1 fatcat:rte6f5xvfvgifedloly4gjccy4

Assessing the Potential of Deep Learning for Emulating Cloud Superparameterization in Climate Models with Real-Geography Boundary Conditions [article]

Griffin Mooers, Mike Pritchard, Tom Beucler, Jordan Ott, Galen Yacalis, Pierre Baldi, Pierre Gentine
2021 arXiv   pre-print
Our results confirm the parameterizability of superparameterized convection with continents through machine learning and we highlight advantages of casting this problem locally in space and time for accurate  ...  We explore the potential of feed-forward deep neural networks (DNNs) for emulating cloud superparameterization in realistic geography, using offline fits to data from the Super Parameterized Community  ...  The work of JO and PB in part supported by NSF NRT grant 1633631 to PB. PG thanks NSF OAC-1835863 and REC synergy grant USMILE.  ... 
arXiv:2010.12996v3 fatcat:qqp3qih7abe7fonw4uoajzxnx4

Use of machine learning to improve simulations of climate [article]

Janni Yuval, Paul A. O'Gorman
2020 arXiv   pre-print
A promising alternative approach is to use machine learning to build new parameterizations directly from high-resolution model output.  ...  Global climate models represent small-scale processes such as clouds and convection using quasi-empirical models known as parameterizations, and these parameterizations are a leading cause of uncertainty  ...  We stress that unlike standard supervised machine learning tasks, higher accuracy on test data is not our only goal.  ... 
arXiv:2001.03151v1 fatcat:gxjgbxipszgstpbx4yamzvwzpq

Assessing the Potential of Deep Learning for Emulating Cloud Superparameterization in Climate Models With Real‐Geography Boundary Conditions

Griffin Mooers, Michael Pritchard, Tom Beucler, Jordan Ott, Galen Yacalis, Pierre Baldi, Pierre Gentine
2021 Journal of Advances in Modeling Earth Systems  
The hope is that these machine learning emulators can help power the next generation of climate models with similar accuracy but at a fraction of the computational cost.  ...  This design would make using these machine learning emulators with climate models very difficult. This motivates learning convection under realistic geography with a simpler network.  ...  microphysics parameterization and the resulting errors in associated turbulence and cloud-radiative effects produced by superparameterized models.  ... 
doi:10.1029/2020ms002385 fatcat:vpverrmiqjcflgcs25i6ze4xla

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

Vladimir M. Krasnopolsky, Michael S. Fox-Rabinovitz, Dmitry V. Chalikov
2005 Monthly Weather Review  
The approach uses neural networks as a statistical or machine learning technique for an accurate and fast emulation or statistical approximation of model physics parameterizations.  ...  A new approach based on a synergetic combination of statistical/machine learning and deterministic modeling within atmospheric models is presented.  ...  This study is based upon the work supported by the Department of Energy Grant ER63197-0007185.  ... 
doi:10.1175/mwr2923.1 fatcat:yovaougqwbbizjlxsqiegukvzq

Combining 3D Radiative Transfer Model and Convolutional Neural Network to Accurately Estimate Forest Canopy Cover From Very High-Resolution Satellite Images

Decai Jin, Jianbo Qi, Huaguo Huang, Linyuan Li
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
In this study, we proposed a hybrid model, which combines a 3D RTM and transfer learning-based convolutional neural network (T-CNN), to estimate FCC from very high-resolution satellite images (e.g., Chinese  ...  For large-scale applications, FCC is usually estimated from remotely sensed data by inverting radiative transfer models (RTMs) or using data-driven regressions.  ...  Section 3 describes the simulation process of a 3D RTM, the realization of the proposed T-CNN method and the inversion of FCC using other machine learning methods.  ... 
doi:10.1109/jstars.2021.3122509 fatcat:rbbletd6w5anbpvyueartydpwi

ESiWACE Newsletter 4/2021 [article]

ESiWACE2
2021 Zenodo  
:21-11:23 CEST • Presentation on Machine Learning Emulation of 3D Cloud Radiative Effects by David Meyer, Robin J.  ...  Mason Friday, 30 Apr, 13:46-13:48 CEST • Presentation on Machine learning emulation of gravity wave drag in numerical weather forecasting by Matthew Chantry, Sam Hatfield, Peter Dueben, Inna Polichtchouk  ... 
doi:10.5281/zenodo.4674593 fatcat:orkxl2engfawzbecxay6ceol3a
« Previous Showing results 1 — 15 out of 309 results