diff --git a/content/code/_index.md b/content/code/_index.md index b43d27f9..935be71b 100644 --- a/content/code/_index.md +++ b/content/code/_index.md @@ -4,7 +4,7 @@ heroHeading: 'Code' heroSubHeading: 'Public repository and data access' heroBackground: '/images/photo-1542831371-29b0f74f9713.jpeg' --- - +
@@ -32,7 +32,7 @@ heroBackground: '/images/photo-1542831371-29b0f74f9713.jpeg' -#### MOM6 with M²LInES parametrizations +#### MOM6 with M²LInES parametrizations
@@ -45,20 +45,20 @@ heroBackground: '/images/photo-1542831371-29b0f74f9713.jpeg'
-#### Samudra +#### Samudra

Samudra: An AI Global Ocean Emulator for Climate
- The first 3D global emulator for Climate, 360x faster than traditional models! + The first 3D global emulator for Climate, 360x faster than traditional models!


-#### SamudrACE +#### SamudrACE
@@ -135,4 +135,4 @@ Upcoming codes will be linked to this repository as they become available. #### Datasets -All of the datasets used in M²LInES publications will be made openly available. When possible, they will be hosted on [Pangeo](https://pangeo.io/) --> +All of the datasets used in M²LInES publications will be made openly available. When possible, they will be hosted on [Pangeo](https://pangeo.io/) --> diff --git a/content/news/2504Holland.md b/content/news/2504Holland.md index 51312e45..0432d111 100644 --- a/content/news/2504Holland.md +++ b/content/news/2504Holland.md @@ -8,4 +8,4 @@ thumbnail: 'images/news/2504Holland.png' images: ['images/news/2504Holland.png'] link: 'https://doi.org/10.1175/JCLI-D-24-0258.1' --- -This new [article](https://doi.org/10.1175/JCLI-D-24-0258.1) explores the predictability of Antarctic sea ice, which varies by region and is influenced by ocean and atmospheric conditions. Using climate models, Marika Holland and co-authors, confirm that **sea ice changes can be predicted months in advance, with cycles of predictability linked to ice growth and retreat**: it remains predictable during growth, loses predictability when melting, and regains it as it reforms. Ocean temperature patterns near the ice edge play a key role, with variations across different regions. These ice changes also influence marine ecosystems by affecting light availability. Understanding these patterns can **improve climate predictions and support the management of the Southern Ocean’s biodiversity**. \ No newline at end of file +This new [article](https://doi.org/10.1175/JCLI-D-24-0258.1) explores the predictability of Antarctic sea ice, which varies by region and is influenced by ocean and atmospheric conditions. Using climate models, Marika Holland and co-authors, confirm that **sea ice changes can be predicted months in advance, with cycles of predictability linked to ice growth and retreat**: it remains predictable during growth, loses predictability when melting, and regains it as it reforms. Ocean temperature patterns near the ice edge play a key role, with variations across different regions. These ice changes also influence marine ecosystems by affecting light availability. Understanding these patterns can **improve climate predictions and support the management of the Southern Ocean’s biodiversity**. diff --git a/content/news/2504Pedersen.md b/content/news/2504Pedersen.md index cdf419b4..2be3ec33 100644 --- a/content/news/2504Pedersen.md +++ b/content/news/2504Pedersen.md @@ -8,4 +8,4 @@ thumbnail: 'images/news/2504Pedersen.png' images: ['images/news/2504Pedersen.png'] link: 'https://doi.org/10.48550/arXiv.2503.18731' --- -Autoregressive surrogate models (emulators) make fast predictions for dynamical systems but become unstable over time due to error buildup. This [study](https://doi.org/10.48550/arXiv.2503.18731), led by Chris Pedersen, introduces thermalization, a technique leveraging diffusion models to adaptively correct errors during inference. **By stabilizing predictions, it extends emulated rollouts of chaotic and turbulent systems by several orders of magnitude**. This breakthrough application of diffusion models enhances the utility of emulators, and can be applied to autoregressive models across science and engineering. \ No newline at end of file +Autoregressive surrogate models (emulators) make fast predictions for dynamical systems but become unstable over time due to error buildup. This [study](https://doi.org/10.48550/arXiv.2503.18731), led by Chris Pedersen, introduces thermalization, a technique leveraging diffusion models to adaptively correct errors during inference. **By stabilizing predictions, it extends emulated rollouts of chaotic and turbulent systems by several orders of magnitude**. This breakthrough application of diffusion models enhances the utility of emulators, and can be applied to autoregressive models across science and engineering. diff --git a/content/news/2504Zanna.md b/content/news/2504Zanna.md index fb79ec80..b2037fca 100644 --- a/content/news/2504Zanna.md +++ b/content/news/2504Zanna.md @@ -9,4 +9,4 @@ images: ['images/news/2504Zanna.jpeg'] link: 'https://doi.org/10.1038/s41612-025-00955-8' --- -New tools like kilometer-scale modeling, parameter tuning, and AI/machine learning are reshaping the climate modeling field, sparking debate on the best path forward. Five internationally renowned female climate scientists, including Laure Zanna, sign this [perspective piece](https://doi.org/10.1038/s41612-025-00955-8) in Nature climate and atmospheric science. Their analysis draws lessons from past research to guide the future of climate modeling. The conclusion: **the future of climate modeling depends on embracing diverse tools and methodologies to drive meaningful progress!** \ No newline at end of file +New tools like kilometer-scale modeling, parameter tuning, and AI/machine learning are reshaping the climate modeling field, sparking debate on the best path forward. Five internationally renowned female climate scientists, including Laure Zanna, sign this [perspective piece](https://doi.org/10.1038/s41612-025-00955-8) in Nature climate and atmospheric science. Their analysis draws lessons from past research to guide the future of climate modeling. The conclusion: **the future of climate modeling depends on embracing diverse tools and methodologies to drive meaningful progress!** diff --git a/content/news/2507CREDIT.md b/content/news/2507CREDIT.md index e738694c..8ff6fd67 100644 --- a/content/news/2507CREDIT.md +++ b/content/news/2507CREDIT.md @@ -9,4 +9,4 @@ images: ['images/news/2507CREDIT.png'] link: 'https://doi.org/10.1038/s41612-025-01125-6' --- -A new framework from NCAR, called CREDIT (Community Research Earth Digital Intelligence Twin), is making it easier for researchers to **develop and test AI-based weather prediction models**. Introduced in this [paper](https://doi.org/10.1038/s41612-025-01125-6), CREDIT supports flexible model design and training, helping **address key challenges in AI-based forecasting**. Using this platform, researchers introduced WXFormer, a novel model that outperforms traditional forecasting systems like ECMWF’s IFS on 10-day forecasts, while being much more computationally efficient. CREDIT aims to accelerate innovation and collaboration in AI-driven weather prediction. **William Chapman** and **Judith Berner** contributed to this research. \ No newline at end of file +A new framework from NCAR, called CREDIT (Community Research Earth Digital Intelligence Twin), is making it easier for researchers to **develop and test AI-based weather prediction models**. Introduced in this [paper](https://doi.org/10.1038/s41612-025-01125-6), CREDIT supports flexible model design and training, helping **address key challenges in AI-based forecasting**. Using this platform, researchers introduced WXFormer, a novel model that outperforms traditional forecasting systems like ECMWF’s IFS on 10-day forecasts, while being much more computationally efficient. CREDIT aims to accelerate innovation and collaboration in AI-driven weather prediction. **William Chapman** and **Judith Berner** contributed to this research. diff --git a/content/news/2507Carlos.md b/content/news/2507Carlos.md index 513e3963..38cd839b 100644 --- a/content/news/2507Carlos.md +++ b/content/news/2507Carlos.md @@ -10,4 +10,3 @@ link: 'https://www.cambridge.org/core/books/probability-and-statistics-for-data- --- **Carlos Fernandez-Granda** new book is out! Published by [Cambridge University Press](https://www.cambridge.org/core/books/probability-and-statistics-for-data-science/CC7DC7E53ED92074008803C96A67620B), the book is a self-contained **guide to the two pillars of data science, probability theory, and statistics**. The materials, which include 200 exercises with solutions, 102 Python notebooks using 23 real-world datasets, 115 YouTube videos with slides, and a free preprint, can be found at this **[website](https://www.ps4ds.net/)**. - diff --git a/content/news/2507Gregory.md b/content/news/2507Gregory.md index 3ca64a9b..600297f6 100644 --- a/content/news/2507Gregory.md +++ b/content/news/2507Gregory.md @@ -9,4 +9,4 @@ images: ['images/news/2507Gregory.png'] link: 'https://doi.org/10.48550/arXiv.2505.18328' --- -**Gregory** et al. present in this [preprint](https://doi.org/10.48550/arXiv.2505.18328), a hybrid modeling approach that **integrates machine learning (ML) into the GFDL SPEAR climate model to correct sea ice biases in real time**. Two versions are tested: one that includes coupled feedbacks (HybridCPL) and one that does not (HybridIO). HybridCPL **significantly improves Arctic and Antarctic sea ice forecasts on seasonal and subseasonal timescales**. In contrast, HybridIO performs poorly due to unanticipated feedbacks. These results highlight the **importance of training ML models within coupled climate systems** for reliable predictions. \ No newline at end of file +**Gregory** et al. present in this [preprint](https://doi.org/10.48550/arXiv.2505.18328), a hybrid modeling approach that **integrates machine learning (ML) into the GFDL SPEAR climate model to correct sea ice biases in real time**. Two versions are tested: one that includes coupled feedbacks (HybridCPL) and one that does not (HybridIO). HybridCPL **significantly improves Arctic and Antarctic sea ice forecasts on seasonal and subseasonal timescales**. In contrast, HybridIO performs poorly due to unanticipated feedbacks. These results highlight the **importance of training ML models within coupled climate systems** for reliable predictions. diff --git a/content/news/2507Perezhogin.md b/content/news/2507Perezhogin.md index be4c9315..8f799c58 100644 --- a/content/news/2507Perezhogin.md +++ b/content/news/2507Perezhogin.md @@ -9,4 +9,4 @@ images: ['images/news/2507Perezhogin.png'] link: 'https://doi.org/10.48550/arXiv.2505.08900' --- -**Perezhogin** et al. propose in this [preprint](https://doi.org/10.48550/arXiv.2505.08900), a **physics-informed neural network to improve the generalization of data-driven mesoscale eddy parameterizations** in ocean models. By applying local input-output scaling based on dimensional analysis, **their method adapts to different grid resolutions and depths**. This approach enhances energy representation and affects biases in both idealized and global ocean simulations. The scaling framework is broadly applicable and robust across configurations. Results show **competitive performance compared to traditional parameterizations**. \ No newline at end of file +**Perezhogin** et al. propose in this [preprint](https://doi.org/10.48550/arXiv.2505.08900), a **physics-informed neural network to improve the generalization of data-driven mesoscale eddy parameterizations** in ocean models. By applying local input-output scaling based on dimensional analysis, **their method adapts to different grid resolutions and depths**. This approach enhances energy representation and affects biases in both idealized and global ocean simulations. The scaling framework is broadly applicable and robust across configurations. Results show **competitive performance compared to traditional parameterizations**. diff --git a/content/news/2507Samudra.md b/content/news/2507Samudra.md index e78b3495..c2cebcf5 100644 --- a/content/news/2507Samudra.md +++ b/content/news/2507Samudra.md @@ -9,4 +9,4 @@ images: ['images/news/Samudra.gif'] link: 'https://www.aps.org/apsnews/2025/06/ai-could-shape-climate-science?utm_source=jun-amn&utm_medium=email&utm_campaign=jun-amn' --- -This **[APS news article](https://www.aps.org/apsnews/2025/06/ai-could-shape-climate-science?utm_source=jun-amn&utm_medium=email&utm_campaign=jun-amn)** highlights **how AI is reshaping climate science**, offering faster, smarter ways to model Earth’s complex systems. It features highlights from the Global Physics Summit, where researchers presented how machine learning is accelerating simulations, uncovering new physics, and helping build more precise climate models, all while keeping physics at the core. In particular, it spotlights interview exerts with Laure Zanna presenting **[Samudra](https://doi.org/10.1029/2024GL114318), the M²LInES created AI emulator**. \ No newline at end of file +This **[APS news article](https://www.aps.org/apsnews/2025/06/ai-could-shape-climate-science?utm_source=jun-amn&utm_medium=email&utm_campaign=jun-amn)** highlights **how AI is reshaping climate science**, offering faster, smarter ways to model Earth’s complex systems. It features highlights from the Global Physics Summit, where researchers presented how machine learning is accelerating simulations, uncovering new physics, and helping build more precise climate models, all while keeping physics at the core. In particular, it spotlights interview exerts with Laure Zanna presenting **[Samudra](https://doi.org/10.1029/2024GL114318), the M²LInES created AI emulator**. diff --git a/content/news/2508Ai2Samudra.md b/content/news/2508Ai2Samudra.md index e9c8e012..70154a4e 100644 --- a/content/news/2508Ai2Samudra.md +++ b/content/news/2508Ai2Samudra.md @@ -9,4 +9,4 @@ images: ['images/news/Samudra.gif'] link: 'https://github.com/ai2cm/ace/releases/tag/2025.7.0' --- -The **[Ai2 Climate Modeling](https://allenai.org/climate-modeling)** team has released a new version of their climate emulator. In this **[latest release](https://github.com/ai2cm/ace/releases/tag/2025.7.0), Samudra**, the AI global ocean emulator developed by M²LInES, is now **integrated into Ai2's full model framework**. We encourage you to check out the **[Ai2 codebase](https://github.com/ai2cm/ace)** and Samudra's **[original code](https://github.com/m2lines/Samudra)** to create your own ocean simulations. \ No newline at end of file +The **[Ai2 Climate Modeling](https://allenai.org/climate-modeling)** team has released a new version of their climate emulator. In this **[latest release](https://github.com/ai2cm/ace/releases/tag/2025.7.0), Samudra**, the AI global ocean emulator developed by M²LInES, is now **integrated into Ai2's full model framework**. We encourage you to check out the **[Ai2 codebase](https://github.com/ai2cm/ace)** and Samudra's **[original code](https://github.com/m2lines/Samudra)** to create your own ocean simulations. diff --git a/content/news/2508Balwada.md b/content/news/2508Balwada.md index 3fb7e72c..1d0fb374 100644 --- a/content/news/2508Balwada.md +++ b/content/news/2508Balwada.md @@ -9,4 +9,4 @@ images: ['images/news/2508Balwada.jpg'] link: 'https://doi.org/10.1029/2025GL114951' --- -A new [paper](https://doi.org/10.1029/2025GL114951), led by **Julius Busecke** and **Dhruv Balwada**, highlights how small-scale air-sea interactions, often unresolved in climate models, can **significantly influence large-scale heat exchange between the ocean and atmosphere**. Using high-resolution coupled simulations, researchers found that this small-scale variability leads to a **systematic global ocean cooling of about 4 W/m², with regional effects up to 100 W/m²**. These findings underscore the critical role of atmospheric wind and ocean temperature heterogeneity, offering new insights for improving climate model accuracy. \ No newline at end of file +A new [paper](https://doi.org/10.1029/2025GL114951), led by **Julius Busecke** and **Dhruv Balwada**, highlights how small-scale air-sea interactions, often unresolved in climate models, can **significantly influence large-scale heat exchange between the ocean and atmosphere**. Using high-resolution coupled simulations, researchers found that this small-scale variability leads to a **systematic global ocean cooling of about 4 W/m², with regional effects up to 100 W/m²**. These findings underscore the critical role of atmospheric wind and ocean temperature heterogeneity, offering new insights for improving climate model accuracy. diff --git a/content/news/2508Falasca.md b/content/news/2508Falasca.md index 9b2af9f5..c0021daf 100644 --- a/content/news/2508Falasca.md +++ b/content/news/2508Falasca.md @@ -9,4 +9,4 @@ images: ['images/news/2508Falasca.png'] link: 'https://doi.org/10.48550/arXiv.2506.22552' --- -This [study](https://doi.org/10.48550/arXiv.2506.22552) examines fundamental challenges in using data-driven models, especially **neural networks, for simulating complex climate dynamics**. While these models can often reproduce average climate behavior, they struggle to capture responses to external changes. The author, **Fabrizio Falasca**, shows that this limitation becomes especially pronounced when only partial observations are available, a common scenario in real-world climate systems. His **findings highlight the importance of incorporating physically informed methods**, like coarse-graining and stochastic parameterizations, **to improve the accuracy and interpretability of neural climate emulators**. \ No newline at end of file +This [study](https://doi.org/10.48550/arXiv.2506.22552) examines fundamental challenges in using data-driven models, especially **neural networks, for simulating complex climate dynamics**. While these models can often reproduce average climate behavior, they struggle to capture responses to external changes. The author, **Fabrizio Falasca**, shows that this limitation becomes especially pronounced when only partial observations are available, a common scenario in real-world climate systems. His **findings highlight the importance of incorporating physically informed methods**, like coarse-graining and stochastic parameterizations, **to improve the accuracy and interpretability of neural climate emulators**. diff --git a/content/news/2509Balwada.md b/content/news/2509Balwada.md index 6af431fa..3caf2dbc 100644 --- a/content/news/2509Balwada.md +++ b/content/news/2509Balwada.md @@ -9,4 +9,4 @@ images: ['images/news/2509Balwada.png'] link: 'https://doi.org/10.22541/essoar.174835313.30541637/v1' --- -This [study](https://doi.org/10.22541/essoar.174835313.30541637/v1), led by Dhruv Balwada, introduces a **new data-driven parameterization to better represent how mesoscale eddies remove potential energy from the ocean in climate models**. Unlike the widely used Gent-McWilliams (GM) scheme, which can hinder resolved eddies and lacks a robust basis for tuning, this approach is both **flow-aware and scale-aware, minimizing negative impacts on resolved dynamics**. Built with a lightweight neural network, the method is efficient, easy to implement, and successfully tested in NOAA’s MOM6 model. The results highlight a **promising path to reduce structural errors and improve the realism of climate simulations**. \ No newline at end of file +This [study](https://doi.org/10.22541/essoar.174835313.30541637/v1), led by Dhruv Balwada, introduces a **new data-driven parameterization to better represent how mesoscale eddies remove potential energy from the ocean in climate models**. Unlike the widely used Gent-McWilliams (GM) scheme, which can hinder resolved eddies and lacks a robust basis for tuning, this approach is both **flow-aware and scale-aware, minimizing negative impacts on resolved dynamics**. Built with a lightweight neural network, the method is efficient, easy to implement, and successfully tested in NOAA’s MOM6 model. The results highlight a **promising path to reduce structural errors and improve the realism of climate simulations**. diff --git a/content/news/2509Yongquan.md b/content/news/2509Yongquan.md index 14ac0a82..552b6284 100644 --- a/content/news/2509Yongquan.md +++ b/content/news/2509Yongquan.md @@ -9,4 +9,4 @@ images: ['images/news/2509Yongquan.png'] link: 'https://doi.org/10.48550/arXiv.2508.00325' --- -Led by **Yongquan Qu**, this [study](https://doi.org/10.48550/arXiv.2508.00325) presents **PnP-DA** (Plug-and-Play Data Assimilation), a new method to improve forecasts in Earth system models. As part of the **LEAP project**, the team combines lightweight analysis updates with a pretrained generative prior to overcome the limitations of traditional approaches that assume overly simple error statistics. Tests on chaotic systems show that it consistently reduces forecast errors across a range of conditions, outperforming classical data assimilation methods. This approach points toward **more reliable and efficient forecasting tools for complex Earth system dynamics**. \ No newline at end of file +Led by **Yongquan Qu**, this [study](https://doi.org/10.48550/arXiv.2508.00325) presents **PnP-DA** (Plug-and-Play Data Assimilation), a new method to improve forecasts in Earth system models. As part of the **LEAP project**, the team combines lightweight analysis updates with a pretrained generative prior to overcome the limitations of traditional approaches that assume overly simple error statistics. Tests on chaotic systems show that it consistently reduces forecast errors across a range of conditions, outperforming classical data assimilation methods. This approach points toward **more reliable and efficient forecasting tools for complex Earth system dynamics**. diff --git a/content/news/2510Chapman.md b/content/news/2510Chapman.md new file mode 100644 index 00000000..19fe0318 --- /dev/null +++ b/content/news/2510Chapman.md @@ -0,0 +1,12 @@ +--- +date: 2025-10-02T09:29:16+10:00 +title: "Atmosphere connected supermodel linking CAM5 and CAM6" +heroHeading: '' +heroSubHeading: 'Implementation and validation of a supermodeling framework into Community Earth System Model version 2.1.5' +heroBackground: '' +thumbnail: 'images/news/2510Chapman.png' +images: ['images/news/2510Chapman.png'] +link: 'https://doi.org/10.5194/gmd-18-5451-2025' +--- + +Led by **Will Chapman**, this [paper](https://doi.org/10.5194/gmd-18-5451-2025) presents the **first atmosphere connected supermodel linking CAM5 and CAM6**, which exchange information interactively to form a new dynamical system. In tests of a single untrained supermodel, the **models synchronize well across timescales** and preserve storm track variability even for variables without direct information exchange. Synchronization is weaker in the tropics, but key large scale patterns such as the North Atlantic Oscillation and Pacific North American Pattern remain intact, and for **some variables mean biases are reduced compared to the individual models.** diff --git a/content/news/2510Samudrace.md b/content/news/2510Samudrace.md new file mode 100644 index 00000000..72e7cc9d --- /dev/null +++ b/content/news/2510Samudrace.md @@ -0,0 +1,12 @@ +--- +date: 2025-10-02T09:29:16+10:00 +title: "SamudrACE" +heroHeading: '' +heroSubHeading: 'Fast and Accurate Coupled Climate Modeling with 3D Ocean and Atmosphere Emulators' +heroBackground: '' +thumbnail: 'images/news/2510SamudrACE.png' +images: ['images/news/2510SamudrACE.png'] +link: 'https://doi.org/10.48550/arXiv.2509.12490' +--- + +In a collaboration between M²LInES and AI2, researchers present **[SamudrACE](https://doi.org/10.48550/arXiv.2509.12490), a machine learning based global climate model emulator**. SamudrACE couples 3D atmosphere, 3D ocean, sea ice and land surface emulators to simulate centuries of climate at 1-degree horizontal resolution with 6-hourly atmospheric and 5-daily oceanic output. It produces 145 2D fields across multiple vertical levels with realistic variability in key climate phenomena such as ENSO. The **model remains highly stable and shows low climate biases** comparable to its individual components while capturing coupled climate behavior that uncoupled models cannot reproduce. diff --git a/content/news/2510Shuchang.md b/content/news/2510Shuchang.md new file mode 100644 index 00000000..39e006a4 --- /dev/null +++ b/content/news/2510Shuchang.md @@ -0,0 +1,12 @@ +--- +date: 2025-10-02T09:29:16+10:00 +title: "CERA: A Framework for Improved Generalization of Machine Learning Models to Changed Climates" +heroHeading: '' +heroSubHeading: 'CERA: A Framework for Improved Generalization of Machine Learning Models to Changed Climates' +heroBackground: '' +thumbnail: 'images/news/Shuchang2510.png' +images: ['images/news/Shuchang2510.png'] +link: 'https://doi.org/10.48550/arXiv.2509.00010' +--- + +**Shuchang Liu** and **Paul O’Gorman** introduce **[CERA](https://doi.org/10.48550/arXiv.2509.00010), a machine learning framework designed to improve the generalizability of ML-models under climate change**. CERA uses an autoencoder with latent space alignment followed by a predictor to estimate moist physics processes. Trained only on control climate data with additional unlabeled warmer climate inputs, it **improves predictions in a +4 K climate and outperforms both raw input and physics informed baselines**. CERA captures shifts in precipitation extremes and the structure of moisture tendencies while reducing the need for manual climate invariant feature engineering, **offering promise for hybrid ML physics systems and applications such as statistical downscaling**. diff --git a/content/news/Newsletters/_index.md b/content/news/Newsletters/_index.md index 514c2834..0d02610a 100644 --- a/content/news/Newsletters/_index.md +++ b/content/news/Newsletters/_index.md @@ -11,6 +11,8 @@ tags: Links to our past newsletters are below. ### 2025 +* 10/01/2025 - [M²LInES newsletter - October 2025](https://mailchi.mp/0608f769fe88/m2lines-oct2025) + * 09/02/2025 - [M²LInES newsletter - September 2025](https://mailchi.mp/68907e6e11ec/m2lines-sep2025) * 08/01/2025 - [M²LInES newsletter - August 2025](https://mailchi.mp/24335d82d580/m2lines-aug2025) diff --git a/content/publications/_index.md b/content/publications/_index.md index d9e0b5f3..7bef421b 100644 --- a/content/publications/_index.md +++ b/content/publications/_index.md @@ -115,7 +115,7 @@ You can also check all our publications on our **[Google Scholar profile](https:

- DOI icon + DOI icon Moein Darman, Pedram Hassanzadeh, Laure Zanna, Ashesh Chattopadhyay
Fourier analysis of the physics of transfer learning for data-driven subgrid-scale models of ocean turbulence
Arxiv DOI:10.48550/arXiv.2504.15487 diff --git a/content/team/MartaMrozowska.md b/content/team/MartaMrozowska.md index 695b27ca..577e3080 100644 --- a/content/team/MartaMrozowska.md +++ b/content/team/MartaMrozowska.md @@ -5,7 +5,7 @@ image: "/images/team/MartaMrozowska.jpg" jobtitle: "Postdoc" promoted: true weight: 20 -position: +position: tags: [Ocean, Machine Learning, Climate Model Development] --- diff --git a/static/images/news/2510Chapman.png b/static/images/news/2510Chapman.png new file mode 100644 index 00000000..26b7a7ae Binary files /dev/null and b/static/images/news/2510Chapman.png differ diff --git a/static/images/news/2510SamudrACE.png b/static/images/news/2510SamudrACE.png new file mode 100644 index 00000000..d31bef45 Binary files /dev/null and b/static/images/news/2510SamudrACE.png differ diff --git a/static/images/news/Shuchang2510.png b/static/images/news/Shuchang2510.png new file mode 100644 index 00000000..b9cab039 Binary files /dev/null and b/static/images/news/Shuchang2510.png differ