Machine learning related to GEOS-Chem: Difference between revisions

From Geos-chem
Jump to navigation Jump to search
No edit summary
 
(19 intermediate revisions by 6 users not shown)
Line 9: Line 9:


|-valign="top"
|-valign="top"
!bgcolor="#CCCCCC"|Oxidants and Chemistry Working Group email list
!bgcolor="#CCCCCC"|GEOS-Chem Machine Learning email list
|<tt>geos-chem-oxidants [at] g.harvard.edu</tt>
|<tt>geos-chem-ml [at] g.harvard.edu</tt>


|-valign="top"
|-valign="top"
!bgcolor="#CCCCCC"|To subscribe to email list
!bgcolor="#CCCCCC"|To subscribe to email list
|Either
|Either
*Send an email to <tt>geos-chem-oxidants+subscribe [at] g.harvard.edu</tt>
*Send an email to <tt>geos-chem-ml+subscribe [at] g.harvard.edu</tt>
Or
Or
*Go to the [https://groups.google.com/a/g.harvard.edu/forum/#!forum/geos-chem-oxidants GEOS-Chem Oxidants and Chemistry]
*Go to the [https://groups.google.com/a/g.harvard.edu/forum/#!forum/geos-chem-ml GEOS-Chem Machine Learning]
*Click on '''Subscribe to this group'''
*Click on '''Subscribe to this group'''


Line 23: Line 23:
!bgcolor="#CCCCCC"|To unsubscribe from email list
!bgcolor="#CCCCCC"|To unsubscribe from email list
|Either
|Either
*Send an email to <tt>geos-chem-oxidants+unsubscribe [at] g.harvard.edu</tt>
*Send an email to <tt>geos-chem-ml+unsubscribe [at] g.harvard.edu</tt>
Or
Or
*Go to the [https://groups.google.com/a/g.harvard.edu/forum/#!forum/geos-chem-oxidants GEOS-Chem Oxidants and Chemistry]
*Go to the [https://groups.google.com/a/g.harvard.edu/forum/#!forum/geos-chem-ml GEOS-Chem Machine Learning]
*Click on the '''My Settings''' button
*Click on the '''My Settings''' button
*Click on '''Leave this group'''
*Click on '''Leave this group'''
Line 32: Line 32:




== GEOS-Chem Machine Publications ==
== GEOS-Chem Machine Learning Publications ==
{| border=1 cellspacing=0 cellpadding=5
{| border=1 cellspacing=0 cellpadding=5
|- bgcolor="#cccccc"
|- bgcolor="#cccccc"
Line 50: Line 50:
|[https://www.earth-syst-sci-data-discuss.net/essd-2019-40/ A machine learning based global sea-surface iodide distribution]
|[https://www.earth-syst-sci-data-discuss.net/essd-2019-40/ A machine learning based global sea-surface iodide distribution]
|14th May 2019
|14th May 2019
|-
|[mailto:kyle.w.dawson@nasa.gov Kyle Dawson] <br> [mailto:nmeskhi@ncsu.edu Nicholas Meskhidze]
|[http://bit.ly/catch-dawson Creating Aerosol Types from CHemistry (CATCH)]
|29th May 2019
|-
|[mailto:mkelp93@gmail.com Makoto Kelp] <br>
|[https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2021MS002926 Online machine learning chemical solver in GEOS-Chem]
|19th May 2022
|}
|}


== Current GEOS-Chem Machine Learning Projects ==
== Current GEOS-Chem Machine Learning Projects ==
Line 68: Line 75:
|Duke University
|Duke University
|Ozone metrics predictor
|Ozone metrics predictor
|[mailto:psk9@duke.edu Prasad Ka Kasibhatla]
|[mailto:psk9@duke.edu Prasad Kasibhatla]
|14th May 2019
|-
|University of York
|Spatial and temporal concentration prediction from sparse observations at the ocean surface
|[mailto:tomas.sherwen@york.ac.uk Tomás Sherwen]
|14th May 2019
|14th May 2019
|-
|NASA GMAO
|GEOS-Chem emulator within the GEOS model
|[mailto:christoph.a.keller@nasa.gov Christoph Keller]
|16th May 2019
|-
|North Carolina State University
|Creating Aerosol Types from CHemistry (CATCH) <br>
A supervised clustering approach to assign aerosol ''types'' to GEOS-Chem output
|[mailto:nmeskhi@ncsu.edu Nicholas Meskhidze] <br> [mailto:kyle.w.dawson@nasa.gov Kyle Dawson]
|29th May 2019
|-
|Boston University
|Machine Learning Parameterization of Stomatal Resistance
|[mailto:ayhwong@bu.edu Anthony Wong]
|26th June 2019
|}
|}


== Useful resources ==
== Useful resources ==
Line 82: Line 109:
|Default Wikipedia ML launch page
|Default Wikipedia ML launch page
|-
|-
|https://www.tensorflow.org Google Tensor Flow]
|[https://www.tensorflow.org Google Tensor Flow]
|Google's Machine Learning package
|Google's Machine Learning package
|-
|[https://pytorch.org/ PyTorch]
|Open-source deep learning platform
|-
|[https://scikit-learn.org/stable/ Scikit-Learn]
|Scikit-Learn package for machine learning in Python
|-
|[https://xgboost.readthedocs.io/en/latest/ XGboost]
| Machine learning algorithms implemented under the Gradient Boosting framework in Python
|-
|[https://lightgbm.readthedocs.io/en/latest/ GBMLight]
| Microsofts' distributed high-performance gradient boosting implementation
|-
|[https://rapids.ai/ RAPIDS]
| NVIDIA's software package for GPU-accelerated data analytics and machine learning
|}
|}


== Upcoming conferences / workshops  ==
== Upcoming conferences / workshops  ==
Line 91: Line 132:
|- bgcolor="#cccccc"
|- bgcolor="#cccccc"
!width="300px"|Conference
!width="300px"|Conference
!width="600px"|Date
!width="600px"|Date / Location
|-
|[https://www.eventbrite.co.uk/e/workshop-machine-learning-for-environmental-sciences-tickets-60865256621 Machine Learning for Environmental Sciences 2019]
|17-18th June 2019 / Cambridge, UK
|-
|[https://www.rmets.org/event/atmospheric-science-conference-2019 Machine Learning in Earth Systems workshop at NCAS'/RMetS' Atmospheric Science Conference 2019]
|2-3rd July 2019 / Birmingham, UK
|-
|-
|[http://users.ox.ac.uk/~phys0895/mlwc2019/index.html Machine Learning for Weather and Climate Modelling 2019]
|[http://users.ox.ac.uk/~phys0895/mlwc2019/index.html Machine Learning for Weather and Climate Modelling 2019]
|2-5th Sept 2019
|2-5th Sept 2019 / Oxford, UK
|-
|[https://www2.acom.ucar.edu/workshop/fascinate-2019 Frontiers of Atmospheric Science and Chemistry: Integration of Novel Applications and Technological Endeavors (FASCINATE)]
|9-11th Sept 2019 / NCAR Center Green Campus in Boulder, Colorado, USA
|-
|[https://sites.google.com/view/climateinformatics2019/ 9th International Workshop on Climate Informatics]
|October 3-4, 2019 / École Normale Supérieure, Paris, France
|-
|-
|[https://meetings.agu.org/fall-meeting-2019/ AGU Fall 2019]
|[https://meetings.agu.org/fall-meeting-2019/ AGU Fall 2019]
|9-13th Dec 2019
|9-13th Dec 2019 /  San Francisco, CA, USA
|}
|}

Latest revision as of 19:21, 21 August 2023

All users interested in the use of Machine Learning within the GEOS-Chem community are encouraged to subscribe to the machine learning email list (click on the link in the contact information section below).

Machine Learning in GEOS-Chem support
GEOS-Chem Machine Learning email list geos-chem-ml [at] g.harvard.edu
To subscribe to email list Either
  • Send an email to geos-chem-ml+subscribe [at] g.harvard.edu

Or

To unsubscribe from email list Either
  • Send an email to geos-chem-ml+unsubscribe [at] g.harvard.edu

Or


GEOS-Chem Machine Learning Publications

Contact Persons Paper Date Added
Christoph Keller
Mat Evans 		Application of random forest regression to the calculation of gas-phase chemistry within the GEOS-Chem chemistry model v10 14th May 2019
Sam Silva
Colette Heald 		A Deep Learning Parameterization for Ozone Dry Deposition Velocities 14th May 2019
Tomas Sherwen
Mat Evans 		A machine learning based global sea-surface iodide distribution 14th May 2019
Kyle Dawson
Nicholas Meskhidze 		Creating Aerosol Types from CHemistry (CATCH) 29th May 2019
Makoto Kelp
Online machine learning chemical solver in GEOS-Chem 19th May 2022

Current GEOS-Chem Machine Learning Projects

User Group Description Contact Person Date Added
University of York Post processing bias corrector Peter Ivatt
Mat Evans 		14th May 2019
Duke University Ozone metrics predictor Prasad Kasibhatla 14th May 2019
University of York Spatial and temporal concentration prediction from sparse observations at the ocean surface Tomás Sherwen 14th May 2019
NASA GMAO GEOS-Chem emulator within the GEOS model Christoph Keller 16th May 2019
North Carolina State University Creating Aerosol Types from CHemistry (CATCH)

A supervised clustering approach to assign aerosol types to GEOS-Chem output

Nicholas Meskhidze
Kyle Dawson 		29th May 2019
Boston University Machine Learning Parameterization of Stomatal Resistance Anthony Wong 26th June 2019

Useful resources

Resource Description
Wikipedia Page Default Wikipedia ML launch page
Google Tensor Flow Google's Machine Learning package
PyTorch Open-source deep learning platform
Scikit-Learn Scikit-Learn package for machine learning in Python
XGboost Machine learning algorithms implemented under the Gradient Boosting framework in Python
GBMLight Microsofts' distributed high-performance gradient boosting implementation
RAPIDS NVIDIA's software package for GPU-accelerated data analytics and machine learning

Upcoming conferences / workshops

Conference Date / Location
Machine Learning for Environmental Sciences 2019 17-18th June 2019 / Cambridge, UK
Machine Learning in Earth Systems workshop at NCAS'/RMetS' Atmospheric Science Conference 2019 2-3rd July 2019 / Birmingham, UK
Machine Learning for Weather and Climate Modelling 2019 2-5th Sept 2019 / Oxford, UK
Frontiers of Atmospheric Science and Chemistry: Integration of Novel Applications and Technological Endeavors (FASCINATE) 9-11th Sept 2019 / NCAR Center Green Campus in Boulder, Colorado, USA
9th International Workshop on Climate Informatics October 3-4, 2019 / École Normale Supérieure, Paris, France
AGU Fall 2019 9-13th Dec 2019 / San Francisco, CA, USA
Retrieved from "https://wiki.seas.harvard.edu/geos-chem/index.php?title=Machine_learning_related_to_GEOS-Chem&oldid=52598"