Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -59,6 +59,8 @@ one or two plain sentences, 120 to 155 characters, saying what the page is about
News posts can rely on `heroSubHeading`. Team members only appear on the team
lists, so they need no description. Add `thumbnail:` to news posts so links share with an image.

### Google verification
static/google39e3f61aaa8692da.html : Search Console verification, do not delete

### Updating Tags
Member interests are defined in the form of tags which can link to personalized landing pages.
Expand Down
2 changes: 1 addition & 1 deletion archetypes/default.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,4 +4,4 @@ title: "{{ replace .TranslationBaseName "-" " " | title }}"
description: ""
date: {{ .Date }}
draft: true
---
---
2 changes: 1 addition & 1 deletion archetypes/team.md
Original file line number Diff line number Diff line change
Expand Up @@ -15,4 +15,4 @@ draft: false
weight: 100
---

M²LInES (M2LInES) is an international collaboration using interpretable machine learning to improve climate models at the air-sea-ice interface.
M²LInES (M2LInES) is an international collaboration using interpretable machine learning to improve climate models at the air-sea-ice interface.
12 changes: 8 additions & 4 deletions config.toml
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
baseURL = "https://m2lines.github.io/"
relativeURLs = "True"
relativeURLs = false
canonifyURLs = "True"
locale = "en-us"
title = "M²LInES"
Expand All @@ -8,6 +8,10 @@ theme = "hugo-hero-theme"
sectionPagesMenu = "content"
enableEmoji = true

enableRobotsTXT = true
enableGitInfo = true
disableKinds = ["taxonomy", "term"]

[markup.goldmark.renderer]
unsafe = true

Expand All @@ -20,7 +24,7 @@ enableEmoji = true
description = "M²LInES (M2LInES), pronounced M-square-lines, is an international collaborative project with the goal of improving climate projections, using scientific and interpretable Machine Learning to capture unaccounted physical processes at the air-sea-ice interface."
image = "images/m2lines-share.png"
twitter = "@M2LInES"

[params.homepage]
show_call_box = true

Expand Down Expand Up @@ -69,7 +73,7 @@ enableEmoji = true
[[menu.main]]
identifier = "former"
name = "Former members"
url = "/team/former"
url = "/team/former/"
parent = "team"
weight = 3

Expand Down Expand Up @@ -113,7 +117,7 @@ enableEmoji = true

[[menu.main]]
name = "Archive"
url = "/publications/archive"
url = "/publications/archive/"
parent = "publications"
weight = 2

Expand Down
4 changes: 0 additions & 4 deletions content/404/_index.md

This file was deleted.

2 changes: 2 additions & 0 deletions content/blog/_index.md
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,8 @@ title: ' Blog'
heroHeading: 'M²LInES Blog'
heroSubHeading: ''
heroBackground: '/images/retrosupply-jLwVAUtLOAQ-unsplash.jpeg'
build:
render: never
---

Coming soon!
3 changes: 3 additions & 0 deletions content/blog/research2.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,9 @@ heroHeading: 'Developing new physics-aware machine learning tools'
heroSubHeading: ''
heroBackground: '/images/32109555763_eb9bb215ef_k.jpg'
weight: 2
build:
render: never
list: always
---


Expand Down
3 changes: 3 additions & 0 deletions content/blog/research4.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,9 @@ weight: 2
heroHeading: 'High-resolution simulations and observational data'
heroSubHeading: ''
heroBackground: '/images/32109555763_eb9bb215ef_k.jpg'
build:
render: never
list: always
---

We will diagnose small-scale physical processes such as:
Expand Down
4 changes: 2 additions & 2 deletions content/news/2406Falasca.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,8 +4,8 @@ title: "Data-driven dimensionality reduction and causal inference for spatiotemp
heroHeading: ''
heroSubHeading: 'Data-driven dimensionality reduction and causal inference for spatiotemporal climate fields'
heroBackground: ''
thumbnail: 'images/news/2406Falasca.jpeg'
images: ['images/news/2406Falasca.jpeg']
thumbnail: 'images/news/2406Falasca.jpg'
images: ['images/news/2406Falasca.jpg']
link: 'https://doi.org/10.1103/PhysRevE.109.044202'
---

Expand Down
2 changes: 1 addition & 1 deletion content/news/2503Brettin.md
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ heroHeading: ''
heroSubHeading: 'Learning Propagators for Sea Surface Height Forecasts Using Koopman Autoencoders'
heroBackground: ''
thumbnail: 'images/news/Brettin2024.png'
images: ['images/news/2Brettin2024.png']
images: ['images/news/Brettin2024.png']
link: 'https://doi.org/10.1029/2024GL112835'
---
Sea surface height forecasts are influenced by many uncertainties. Traditional statistical methods help make predictions but often rely on assumptions that don’t fully capture the climate system's complexity. In this [paper](https://doi.org/10.1029/2024GL112835), **Andrew Brettin** and co-authors develop a machine learning model that learns a simplified representation of the climate system, **improving sea surface height predictions**. Their approach **outperforms methods that separate data compression and prediction**. Additionally, the model highlights key regions where better sea level representation can enhance regional forecasts.
2 changes: 1 addition & 1 deletion content/news/2503Connolly.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,4 +8,4 @@ thumbnail: 'images/news/2503Connolly.gif'
images: ['images/news/2503Connolly.gif']
link: 'https://doi.org/10.22541/essoar.173869578.80400701/v1'
---
Representing atmospheric turbulence in climate models requires estimating unresolved small-scale forces. This [study]('https://doi.org/10.22541/essoar.173869578.80400701/v1), led by **Alex Connolly**, uses a deep neural network (DNN) to improve turbulence modeling by predicting these forces in large-eddy simulations (LES) of the atmospheric boundary layer. The DNN is trained on high-resolution data and tested across different conditions. Results show that **models using physics-based scaling perform better than those relying only on statistical normalization**. Embedding physical symmetries into the model further improves accuracy. This research **highlights the importance of physics-informed machine learning for improving turbulence representation in Earth system models for more reliable climate projections**.
Representing atmospheric turbulence in climate models requires estimating unresolved small-scale forces. This [study](https://doi.org/10.22541/essoar.173869578.80400701/v1), led by **Alex Connolly**, uses a deep neural network (DNN) to improve turbulence modeling by predicting these forces in large-eddy simulations (LES) of the atmospheric boundary layer. The DNN is trained on high-resolution data and tested across different conditions. Results show that **models using physics-based scaling perform better than those relying only on statistical normalization**. Embedding physical symmetries into the model further improves accuracy. This research **highlights the importance of physics-informed machine learning for improving turbulence representation in Earth system models for more reliable climate projections**.
2 changes: 1 addition & 1 deletion content/news/2610Balwada.md
Original file line number Diff line number Diff line change
Expand Up @@ -11,4 +11,4 @@ link: 'https://doi.org/10.1029/2025MS005245'

Mesoscale eddies, the ocean’s swirling currents, move heat and tracers and drain potential energy from the large-scale circulation. Most climate models can’t fully resolve them, so their effects are approximated with the Gent-McWilliams (GM) scheme. GM captures the overall energy removal but assumes eddy fluxes always run down the gradient, missing the up-gradient and cross-gradient fluxes seen in reality. It can also dampen eddies that models partially resolve.

In new **M²LInES [work](https://doi.org/10.1029/2025MS005245) led by Dhruv Balwada**, the team developed **a machine learning parameterization free of this diffusive constraint**. Using a small, inexpensive neural network, it captures the full structure of eddy thickness fluxes while still extracting energy, adapts to both the flow and model resolution, and is tuned with a single simple parameter. **Implemented in NOAA GFDL’s MOM6 ocean model**, it performed well in idealized channel and gyre simulations at eddy-permitting and coarse resolutions.
In new **M²LInES [work](https://doi.org/10.1029/2025MS005245) led by Dhruv Balwada**, the team developed **a machine learning parameterization free of this diffusive constraint**. Using a small, inexpensive neural network, it captures the full structure of eddy thickness fluxes while still extracting energy, adapts to both the flow and model resolution, and is tuned with a single simple parameter. **Implemented in NOAA GFDL’s MOM6 ocean model**, it performed well in idealized channel and gyre simulations at eddy-permitting and coarse resolutions.
2 changes: 1 addition & 1 deletion content/news/2610Bushuk.md
Original file line number Diff line number Diff line change
Expand Up @@ -9,4 +9,4 @@ images: ['images/news/2610Bushuk.jpg']
link: 'https://doi.org/10.1029/2026GL124043'
---

Led by **Mitch Bushuk**, this **[article](https://doi.org/10.1029/2026GL124043)** investigates the hypothesis that eddy-resolving climate models may improve the simulation of Antarctic sea ice trends. Using high-resolution model hierarchies from GFDL and NCAR alongside three generations of CMIP models, the study finds **no clear relationship between ocean or atmospheric resolution and Antarctic sea ice trends**, either over the historical period or the remainder of the 21st century. Instead, models that successfully reproduce the observed trend tend to exhibit large-amplitude multidecadal variability in the Southern Ocean, suggesting that **these models may be getting the “right answer for the wrong reasons.”**
Led by **Mitch Bushuk**, this **[article](https://doi.org/10.1029/2026GL124043)** investigates the hypothesis that eddy-resolving climate models may improve the simulation of Antarctic sea ice trends. Using high-resolution model hierarchies from GFDL and NCAR alongside three generations of CMIP models, the study finds **no clear relationship between ocean or atmospheric resolution and Antarctic sea ice trends**, either over the historical period or the remainder of the 21st century. Instead, models that successfully reproduce the observed trend tend to exhibit large-amplitude multidecadal variability in the Southern Ocean, suggesting that **these models may be getting the “right answer for the wrong reasons.”**
2 changes: 1 addition & 1 deletion content/news/2610ESI.md
Original file line number Diff line number Diff line change
Expand Up @@ -12,4 +12,4 @@ _Left to right: Laure Zanna, Timon McPhearson, Donald Chahbazpour, Dharik Mallap

On September 21, NYU’s new **[Earth Systems Institute (ESI)](https://wp.nyu.edu/earthsys/)** made its public debut during Climate Week NYC. Launched by NYU's Office of Global Science and Technology, and **directed by Laure Zanna**, ESI brings together more than 40 scientists across Courant, Tandon, and Arts & Science to build data-driven models of the Earth system and put them in the hands of decision-makers. The afternoon featured talks on Samudra, our AI ocean emulator that can simulate 4,000 years of ocean in a single day on one GPU, the surprising greening of the Sahel, and how home electrification is reshaping peak power demand. A fellowship, a Distinguished Visitor Program, and a spring summit are planned for the year ahead.

**[Read the full story](https://www.nyu.edu/about/news-publications/news/2026/october/nyu-earth-systems-institute-launches-at-climate-week-nyc.html?utm_source=nyuogst-amplify&utm_medium=email&utm_campaign=2026-10-esi-climate-week-recap&utm_content=headline-earth-systems)**
**[Read the full story](https://www.nyu.edu/about/news-publications/news/2026/october/nyu-earth-systems-institute-launches-at-climate-week-nyc.html?utm_source=nyuogst-amplify&utm_medium=email&utm_campaign=2026-10-esi-climate-week-recap&utm_content=headline-earth-systems)**
2 changes: 1 addition & 1 deletion content/news/2610Holland.md
Original file line number Diff line number Diff line change
Expand Up @@ -17,4 +17,4 @@ Marika’s research explores **how polar sea ice both responds to and shapes glo

This recognition is richly deserved. Congratulations, Marika!

Read more on the **[MacArthur Foundation website](https://www.macfound.org/fellows/class-of-2026/marika-holland)**.
Read more on the **[MacArthur Foundation website](https://www.macfound.org/fellows/class-of-2026/marika-holland)**.
4 changes: 2 additions & 2 deletions content/news/2610Simonsjob.md
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,6 @@ link: 'https://wp.nyu.edu/sccgf/'
We are excited to share a new opportunity from our partners at the **[Simons Center for Computational Geophysical Flows](https://wp.nyu.edu/sccgf/)**! They are hiring Simons Center Postdoctoral Fellows with a background in machine learning, computer science, engineering, physical science, applied mathematics, atmosphere/ocean science, or related technical discipline.


You can **[apply here](https://apply.interfolio.com/193211)** to solve important climate problems.
You can **[apply here](https://apply.interfolio.com/193211)** to solve important climate problems.

Deadline to apply: **December 1, 2026.**
2 changes: 1 addition & 1 deletion content/news/AGUpresentation.md
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ heroHeading: 'AGU presentations by M²LInES team members'
heroSubHeading: ''
heroBackground: ''
thumbnail: 'images/news/JanniYuval.png'
images: ['images/news/JanniYuval.png', 'images/news/LaureZanna.png', 'images/news/PierreGentine.png', 'images/news/ZiweiLi.png']
images: ['images/news/JanniYuval.png', 'images/news/LaureZanna.png', 'images/news/PierreGentine.png', 'images/news/ZiweiLi21.png']
link: ''
---
## Schedule of AGU presentations from team members
Expand Down
2 changes: 1 addition & 1 deletion content/news/Chapmanetal.md
Original file line number Diff line number Diff line change
Expand Up @@ -9,5 +9,5 @@ images: ['images/news/Chapmanetal.png']
link: 'https://doi.org/10.1175/MWR-D-21-0106.1'
---

In this [paper]('https://doi.org/10.1175/MWR-D-21-0106.1'), Will Chapman uses a deep learning (DL) based post-processing framework to generate probabilistic forecasts of atmospheric rivers from single member numerical weather prediction (NWP) forecasts.
In this [paper](https://doi.org/10.1175/MWR-D-21-0106.1), Will Chapman uses a deep learning (DL) based post-processing framework to generate probabilistic forecasts of atmospheric rivers from single member numerical weather prediction (NWP) forecasts.
They show that the DL forecasts are reliable and sharp, and compete well with dynamically generated NWP ensembles.
8 changes: 8 additions & 0 deletions content/pages/_index.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,8 @@
---
title: ' About'
heroHeading: 'About'
heroSubHeading: ''
heroBackground: '/images/retrosupply-jLwVAUtLOAQ-unsplash.jpeg'
build:
render: never
---
2 changes: 1 addition & 1 deletion content/services/_index.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,4 +2,4 @@
title: 'Research axes'
build:
render: never
---
---
2 changes: 1 addition & 1 deletion content/team/SaraShamekh.md
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ image: "/images/team/sara.jpg"
jobtitle: "Lead Scientist"
promoted: true
weight: 9
Website: sshamekh.com
Website: https://sshamekh.com
position: Atmospheric Boundary Layer
tags: [Atmosphere, Machine Learning]
---
Expand Down
1 change: 0 additions & 1 deletion content/team/_index.md
Original file line number Diff line number Diff line change
Expand Up @@ -12,4 +12,3 @@ cascade:
target:
kind: page
---

16 changes: 16 additions & 0 deletions layouts/404/404.html
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
{{ define "main" }}
<div class="container pt-10 pb-10">
<div class="row justify-content-center">
<div class="col-12 col-md-9 text-center">
<h1 class="title-1 black mb-4">Page not found</h1>
<p class="mb-4">The page you were looking for has moved or no longer exists.</p>
<a href="{{ "/" | relURL }}" class="btn btn-primary mr-1">Home</a>
<a href="{{ "research/" | relURL }}" class="btn btn-primary mr-1">Research</a>
<a href="{{ "publications/" | relURL }}" class="btn btn-primary">Publications</a>
<a href="{{ "team/" | relURL }}" class="btn btn-primary mr-1">Team</a>
<a href="{{ "news/" | relURL }}" class="btn btn-primary mr-1">News</a>

</div>
</div>
</div>
{{ end }}
2 changes: 1 addition & 1 deletion layouts/partials/seo-title.html
Original file line number Diff line number Diff line change
Expand Up @@ -5,4 +5,4 @@
{{- $page := .Params.seoTitle | default (trim .Title " ") | default .Params.heroHeading -}}
{{- $title = printf "%s | %s" $page site.Title -}}
{{- end -}}
<title>{{ $title }}</title>
<title>{{ $title }}</title>
4 changes: 2 additions & 2 deletions layouts/partials/seo.html
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,6 @@
<meta name="twitter:title" content="{{ $title }}">
<meta name="twitter:description" content="{{ $desc }}">
<meta name="twitter:image" content="{{ $img }}">
{{- if .Params.noindex }}
{{- if or .Params.noindex (eq .Kind "404") }}
<meta name="robots" content="noindex">
{{- end }}
{{- end }}
4 changes: 4 additions & 0 deletions layouts/robots.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1,4 @@
User-agent: *
Disallow:

Sitemap: {{ "sitemap.xml" | absURL }}
2 changes: 1 addition & 1 deletion static/google39e3f61aaa8692da.html
Original file line number Diff line number Diff line change
@@ -1 +1 @@
google-site-verification: google39e3f61aaa8692da.html
google-site-verification: google39e3f61aaa8692da.html
1 change: 0 additions & 1 deletion static/images/machinelearning.md.html

This file was deleted.

Loading
Loading