diff --git a/jose.00241/10.21105.jose.00241.crossref.xml b/jose.00241/10.21105.jose.00241.crossref.xml new file mode 100644 index 0000000..c65e2bd --- /dev/null +++ b/jose.00241/10.21105.jose.00241.crossref.xml @@ -0,0 +1,417 @@ + + + + 20241226110213-4ba62b54480eb7dc6d187bd0cbeac7be365a8c26 + 20241226110213 + + JOSS Admin + admin@theoj.org + + The Open Journal + + + + + Journal of Open Source Education + JOSE + 2577-3569 + + 10.21105/jose + https://jose.theoj.org + + + + + 12 + 2024 + + + 7 + + 82 + + + + Learning Machine Learning with Lorenz-96 + + + + Dhruv + Balwada + + Lamont Doherty Earth Observatory, Columbia University + + https://orcid.org/0000-0001-6632-0187 + + + Ryan + Abernathey + + Lamont Doherty Earth Observatory, Columbia University + + https://orcid.org/0000-0001-5999-4917 + + + Shantanu + Acharya + + Courant Institute of Mathematical Sciences, New York University + + https://orcid.org/0000-0002-9652-2991 + + + Alistair + Adcroft + + Program in Atmospheric and Oceanic Sciences, Princeton University + + https://orcid.org/0000-0001-9413-1017 + + + Judith + Brener + + National Center for Atmospheric Research + + https://orcid.org/0000-0003-2168-0431 + + + V + Balaji + + Schmidt Futures + + https://orcid.org/0000-0001-7561-5438 + + + Mohamed Aziz + Bhouri + + Earth and Environmental Engineering, Columbia University + + https://orcid.org/0000-0003-1140-7415 + + + Joan + Bruna + + Courant Institute of Mathematical Sciences, New York University + Center for Data Science, New York University + + https://orcid.org/0000-0002-2847-1512 + + + Mitch + Bushuk + + NOAA Geophysical Fluid Dynamics Laboratory + + https://orcid.org/0000-0002-0063-1465 + + + Will + Chapman + + National Center for Atmospheric Research + + https://orcid.org/0000-0002-0472-7069 + + + Alex + Connolly + + Earth and Environmental Engineering, Columbia University + + https://orcid.org/0000-0002-2310-0480 + + + Julie + Deshayes + + Sorbonne Universités, LOCEAN Laboratory, Paris, France + + https://orcid.org/0000-0002-1462-686X + + + Carlos + Fernandez-Granda + + Courant Institute of Mathematical Sciences, New York University + Center for Data Science, New York University + + https://orcid.org/0000-0001-7039-8606 + + + Pierre + Gentine + + Earth and Environmental Engineering, Columbia University + Columbia Climate School, Columbia University + + https://orcid.org/0000-0002-0845-8345 + + + Anastasiia + Gorbunova + + Univ. Grenoble Alpes, CNRS, IRD, Grenoble INP, INRAE, IGE, 38000 Grenoble, France + + https://orcid.org/0000-0002-3271-2024 + + + Will + Gregory + + Program in Atmospheric and Oceanic Sciences, Princeton University + + https://orcid.org/0000-0001-8176-1642 + + + Arthur + Guillaumin + + Queen Mary University of London + + https://orcid.org/0000-0003-1571-4228 + + + Shubham + Gupta + + Tandon School of Engineering, New York University + + https://orcid.org/0009-0002-6966-588X + + + Marika + Holland + + National Center for Atmospheric Research + + https://orcid.org/0000-0001-5621-8939 + + + J Emmanuel + Johnsson + + Univ. Grenoble Alpes, CNRS, IRD, Grenoble INP, INRAE, IGE, 38000 Grenoble, France + + https://orcid.org/0000-0002-6739-0053 + + + Julien Le + Sommer + + Univ. Grenoble Alpes, CNRS, IRD, Grenoble INP, INRAE, IGE, 38000 Grenoble, France + + https://orcid.org/0000-0002-6882-2938 + + + Ziwei + Li + + Courant Institute of Mathematical Sciences, New York University + + + + Nora + Loose + + Program in Atmospheric and Oceanic Sciences, Princeton University + + https://orcid.org/0000-0002-3684-9634 + + + Feiyu + Lu + + NOAA Geophysical Fluid Dynamics Laboratory + + https://orcid.org/0000-0001-6532-0740 + + + Paul + O’Gorman + + Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology + + https://orcid.org/0000-0001-6532-0740 + + + Pavel + Perezhogin + + Courant Institute of Mathematical Sciences, New York University + + https://orcid.org/0000-0003-2098-3457 + + + Brandon + Reichl + + NOAA Geophysical Fluid Dynamics Laboratory + + https://orcid.org/0000-0001-9047-0767 + + + Andrew + Ross + + Courant Institute of Mathematical Sciences, New York University + + https://orcid.org/0000-0002-2368-6979 + + + Aakash + Sane + + Program in Atmospheric and Oceanic Sciences, Princeton University + + https://orcid.org/0000-0002-9642-008X + + + Sara + Shamekh + + Earth and Environmental Engineering, Columbia University + Courant Institute of Mathematical Sciences, New York University + + https://orcid.org/0000-0001-7441-4116 + + + Tarun + Verma + + Program in Atmospheric and Oceanic Sciences, Princeton University + + https://orcid.org/0000-0001-7730-1483 + + + Janni + Yuval + + Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology + + https://orcid.org/0000-0001-7519-0118 + + + Lorenzo + Zampieri + + Ocean Modeling and Data Assimilation Division, Fondazione Centro Euro-Mediterraneo sui Cambiamenti Climatici - CMCC + + https://orcid.org/0000-0003-1703-4162 + + + Cheng + Zhang + + Program in Atmospheric and Oceanic Sciences, Princeton University + + https://orcid.org/0000-0003-4278-9786 + + + Laure + Zanna + + Courant Institute of Mathematical Sciences, New York University + + https://orcid.org/0000-0002-8472-4828 + + + + 12 + 26 + 2024 + + + 241 + + + 10.21105/jose.00241 + + + http://creativecommons.org/licenses/by/4.0/ + http://creativecommons.org/licenses/by/4.0/ + http://creativecommons.org/licenses/by/4.0/ + + + + Software archive + 10.5281/zenodo.13921550 + + + GitHub review issue + https://github.com/openjournals/jose-reviews/issues/241 + + + + 10.21105/jose.00241 + https://jose.theoj.org/papers/10.21105/jose.00241 + + + https://jose.theoj.org/papers/10.21105/jose.00241.pdf + + + + + + Predictability: A problem partly solved + Lorenz + Seminar on Predictability + 1 + 10.1017/cbo9780511617652.004 + 1995 + Lorenz, E. N. (1995). Predictability: A problem partly solved. Seminar on Predictability, 1, 1–18. https://doi.org/10.1017/cbo9780511617652.004 + + + Parametrization in weather and climate models + Christensen + 10.1093/acrefore/9780190228620.013.826 + 2022 + Christensen, H., & Zanna, L. (2022). Parametrization in weather and climate models. https://doi.org/10.1093/acrefore/9780190228620.013.826 + + + + + + diff --git a/jose.00241/10.21105.jose.00241.pdf b/jose.00241/10.21105.jose.00241.pdf new file mode 100644 index 0000000..f5bcc9d Binary files /dev/null and b/jose.00241/10.21105.jose.00241.pdf differ diff --git a/jose.00241/paper.jats/10.21105.jose.00241.jats b/jose.00241/paper.jats/10.21105.jose.00241.jats new file mode 100644 index 0000000..ec04dc3 --- /dev/null +++ b/jose.00241/paper.jats/10.21105.jose.00241.jats @@ -0,0 +1,577 @@ + + +
+ + + + +Journal of Open Source Education +JOSE + +2577-3569 + +Open Journals + + + +241 +10.21105/jose.00241 + +Learning Machine Learning with Lorenz-96 + + + +https://orcid.org/0000-0001-6632-0187 + +Balwada +Dhruv + + + + +https://orcid.org/0000-0001-5999-4917 + +Abernathey +Ryan + + + + +https://orcid.org/0000-0002-9652-2991 + +Acharya +Shantanu + + + + +https://orcid.org/0000-0001-9413-1017 + +Adcroft +Alistair + + + + +https://orcid.org/0000-0003-2168-0431 + +Brener +Judith + + + + +https://orcid.org/0000-0001-7561-5438 + +Balaji +V + + + + +https://orcid.org/0000-0003-1140-7415 + +Bhouri +Mohamed Aziz + + + + +https://orcid.org/0000-0002-2847-1512 + +Bruna +Joan + + + + + +https://orcid.org/0000-0002-0063-1465 + +Bushuk +Mitch + + + + +https://orcid.org/0000-0002-0472-7069 + +Chapman +Will + + + + +https://orcid.org/0000-0002-2310-0480 + +Connolly +Alex + + + + +https://orcid.org/0000-0002-1462-686X + +Deshayes +Julie + + + + +https://orcid.org/0000-0001-7039-8606 + +Fernandez-Granda +Carlos + + + + + +https://orcid.org/0000-0002-0845-8345 + +Gentine +Pierre + + + + + +https://orcid.org/0000-0002-3271-2024 + +Gorbunova +Anastasiia + + + + +https://orcid.org/0000-0001-8176-1642 + +Gregory +Will + + + + +https://orcid.org/0000-0003-1571-4228 + +Guillaumin +Arthur + + + + +https://orcid.org/0009-0002-6966-588X + +Gupta +Shubham + + + + +https://orcid.org/0000-0001-5621-8939 + +Holland +Marika + + + + +https://orcid.org/0000-0002-6739-0053 + +Johnsson +J Emmanuel + + + + +https://orcid.org/0000-0002-6882-2938 + +Sommer +Julien Le + + + + + +Li +Ziwei + + + + +https://orcid.org/0000-0002-3684-9634 + +Loose +Nora + + + + +https://orcid.org/0000-0001-6532-0740 + +Lu +Feiyu + + + + +https://orcid.org/0000-0001-6532-0740 + +O’Gorman +Paul + + + + +https://orcid.org/0000-0003-2098-3457 + +Perezhogin +Pavel + + + + +https://orcid.org/0000-0001-9047-0767 + +Reichl +Brandon + + + + +https://orcid.org/0000-0002-2368-6979 + +Ross +Andrew + + + + +https://orcid.org/0000-0002-9642-008X + +Sane +Aakash + + + + +https://orcid.org/0000-0001-7441-4116 + +Shamekh +Sara + + + + + +https://orcid.org/0000-0001-7730-1483 + +Verma +Tarun + + + + +https://orcid.org/0000-0001-7519-0118 + +Yuval +Janni + + + + +https://orcid.org/0000-0003-1703-4162 + +Zampieri +Lorenzo + + + + +https://orcid.org/0000-0003-4278-9786 + +Zhang +Cheng + + + + +https://orcid.org/0000-0002-8472-4828 + +Zanna +Laure + + + + + +Lamont Doherty Earth Observatory, Columbia +University + + + + +Courant Institute of Mathematical Sciences, New York +University + + + + +Program in Atmospheric and Oceanic Sciences, Princeton +University + + + + +Earth and Environmental Engineering, Columbia +University + + + + +Queen Mary University of London + + + + +Univ. Grenoble Alpes, CNRS, IRD, Grenoble INP, INRAE, IGE, +38000 Grenoble, France + + + + +Ocean Modeling and Data Assimilation Division, Fondazione +Centro Euro-Mediterraneo sui Cambiamenti Climatici - CMCC + + + + +Tandon School of Engineering, New York +University + + + + +NOAA Geophysical Fluid Dynamics Laboratory + + + + +Sorbonne Universités, LOCEAN Laboratory, Paris, +France + + + + +Department of Earth, Atmospheric, and Planetary Sciences, +Massachusetts Institute of Technology + + + + +National Center for Atmospheric Research + + + + +Center for Data Science, New York University + + + + +Columbia Climate School, Columbia University + + + + +Schmidt Futures + + + + +10 +10 +2023 + +7 +82 +241 + +Authors of papers retain copyright and release the +work under a Creative Commons Attribution 4.0 International License (CC +BY 4.0) +2024 +The article authors + +Authors of papers retain copyright and release the work under +a Creative Commons Attribution 4.0 International License (CC BY +4.0) + + + +Python +Machine Learning +Neural Networks +Dynamical systems + + + + + + Summary +

Machine learning (ML) is a rapidly growing field that is starting + to touch all aspects of our lives, and science is not immune to this. + In fact, recent work in the field of scientific ML, i.e. combining ML + and with conventional scientific problems, is leading to new + breakthroughs in notoriously hard problems, which might have seemed + too distant till a few years ago. One such age-old problem is that of + turbulence closures in fluid flows. This closure or parameterization + problem is particularly relevant for environmental fluids, which span + a large range of scales from the size of the planet down to + millimeters, and remains a big challenge in the way of improving + forecasts of weather and projections of climate.

+

The climate system is composed of many interacting components + (e.g., ocean, atmosphere, ice) and is described by complex nonlinear + equations. To simulate, understand, and predict climate, these + equations are solved numerically under a number of simplifications, + therefore leading to errors. The errors result from numerics used to + solve the equations and the lack of appropriate representations of + processes occurring below the resolution of the climate model grid + (i.e., sub-grid processes).

+

This book aims to conceptualize the problems associated with + climate models within a simple and computationally accessible + framework, and show how some basic ML methods can be used to approach + these problems. We will introduce the readers to climate modeling + using a simple tool, the Lorenz + (1995) + (L96) two-timescale model. We discuss the numerical aspects of the L96 + model, the approximate representation of sub-grid processes (known as + parameterizations or closures), and simple data assimilation problems + (a data-model fusion method). We then use the L96 results to + demonstrate how to learn sub-grid parameterizations from data with ML, + and then test the parameterizations offline (apriori) and online + (aposteriori), with a focus on the interpretability of the results. + This book is written primarily for climate scientists and physicists, + who are looking for a gentle introduction to how they can incorporate + ML into their work. However, it may also help ML scientists learn + about the parameterization problem in a framework that is relatively + simple to understand and use.

+

The material in this Jupyter book is presented over five sections. + The first section, Lorenz 96 and General Circulations Models, + describes the Lorenz-96 model and how it can work as a simple analog + to much more complex general circulation models used for simulating + ocean and atmosphere dynamics. This section also introduces the + essence of the parameterization or closure problem. In the second + section, Neural Networks with Lorenz-96, we introduce the basics of + ML, how fully connected neural networks can be used to approach the + parameterization task, and how these neural networks can be optimized + and interpreted. No model, even the well parameterized ones, is + perfect, and the way we keep computer models close to reality is by + guiding them with the help of observational data. This task is + referred to as data assimilation, and is introduced in the third + section, Data Assimilation with Lorenz-96. Here, we use the L96 model + to quickly introduce the concepts from data assimilation, and show how + ML can be used to learn data assimilation increments to help reduce + model biases. While neural networks can be great functional + approximators, they are usually quite opaque, and it is hard to figure + out exactly what they have learnt. Equation discovery is a class of ML + techniques that tries to estimate the function in terms of an equation + rather than as a set of weights for a neural network. This approach + produces a result that is far more interpretable, and can potentially + even help discover novel physics. These techniques are presented in + the fourth section, Equation Discovery with Lorenz-96. Finally, we + describe a few more ML in section five, Other ML approaches for + Lorenz-96, with the acknowledgment that there are many more techniques + in the fast-growing ML and scientific ML literature and we have no + intention of providing a comprehensive summary of the field.

+

The book was created by and as part of M2LInES, an international + collaboration supported by Schmidt Futures, to improve climate models + with scientific ML. The original goal for these notebooks in this + Jupyter book was for our team to work together and learn from each + other; in particular, to get up to speed on the key scientific aspects + of our collaboration (parameterizations, ML, data assimilation, + uncertainty quantification) and to develop new ideas. This was done as + a series of tutorials, each of which was led by a few team members and + occurred with a frequency of roughly once every 2 weeks for about 6-7 + months. This Jupyter book is a collection of the notebooks used during + these tutorials, which have only slightly been edited for continuity + and clarity. Ultimately, we are happy to share these resources with + the scientific community to introduce our research ideas and foster + the use of ML techniques for tackling climate science problems.

+
+ + Statement of Need +

Parameterization of sub-grid processes is a major challenge in + climate modeling. The details of this problem may often be very + context dependent (Christensen & Zanna + (2022)), + but much can be learned by addressing the issue in a general and + simpler sense. Also, a general approach allows non-domain experts, + e.g. ML researchers, to engage and contribute more meaningfully. This + JupyterBook aims to achieve this target with the help of a simple + dynamical system model - Lorenz 96, such that the reader is introduced + to the basic concepts with minimal superfluous complexity. It is + possible to extend the concepts that are presented here to other + dynamical systems, and even to more complex parameterization tasks + (some examples can be found at + https://m2lines.github.io/publications/), and we hope that researchers + and learners aiming to do this find the concepts presented here as a + useful stepping stone in this pursuit.

+

As described above, these notebooks were originally created to + introduce non-domain experts to ideas from the parameterization + aspects of climate modeling and how ML could be used to potentially + address these. Now they have been adapted to act as a pedagogical tool + for self-learning, be used as a reference manual, or for teaching some + modules in an introductory class on ML. The book is organized in + sections that are relatively independent; with the exception that the + first section provides a general overview to the parameterization + problem in climate models. Each notebook covers material that can be + discussed in roughly an hour-long lecture, and sections can be mixed + and matched or ordered as needed depending on the overall learning + objectives.

+
+ + Acknowledgements +

This work is supported by the generosity of Eric and Wendy Schmidt + by recommendation of Schmidt Futures, as part of its Virtual Earth + System Research Institute (VESRI). MAB acknowledges support from + National Science Foundation’s AGS-PRF Fellowship Award + (AGS2218197).

+
+ + + + + + + + LorenzE. N. + + Predictability: A problem partly solved + Seminar on Predictability + ECMWF; ECMWF + Shinfield Park, Reading + 1995 + 1 + https://www.ecmwf.int/node/10829 + 10.1017/cbo9780511617652.004 + 1 + 18 + + + + + + ChristensenHannah + ZannaLaure + + Parametrization in weather and climate models + Oxford University Press + 2022 + 10.1093/acrefore/9780190228620.013.826 + + + + +