sep news

Author: IamShubhamGupto

content/news/2409Gentine.md

@@ -0,0 +1,12 @@
+---
+date: 2024-09-02T09:29:16+10:00
+title: "Pushing the frontiers in climate modelling and analysis with machine learning"
+heroHeading: ''
+heroSubHeading: 'Pushing the frontiers in climate modelling and analysis with machine learning'
+heroBackground: ''
+thumbnail:  'images/news/2409Gentine.png'
+images: ['images/news/2409Gentine.png']
+link: 'https://urldefense.proofpoint.com/v2/url?u=https-3A__rdcu.be_dRMdh&d=DwMFAg&c=slrrB7dE8n7gBJbeO0g-IQ&r=QDJB-3bkJY-UGfMxhPb20cpbFjA-cyx7hX9fyxzJd2g&m=CE4WhYHD96qQB3ADVDneaIexEQMkwinRKqRfx3oQzbT9LVx6pq4drDptvNNfRUed&s=NIoO5G0_IEqzAnHSWiLv96kX3H08VdHUo4TjBhYk3FI&e='
+---
+
+This Nature Climate Change review, explores how machine learning can revolutionize climate modeling and analysis to meet growing demands for better projections and actionable climate information. The authors argue that now is the time to advance beyond current techniques by developing more accurate machine-learning-based Earth system models and creating new tools for predicting extreme events, improving detection methods, and enhancing climate model analysis. The review emphasizes the importance of collaboration between machine learning experts and climate scientists, along with the involvement of the private sector, to drive innovation and progress in climate science. **Pierre Gentine** and **Laure Zanna** are co-authors, with works from M²LInES and LEAP heavily featured in the article!

content/news/2409Gregory.md

@@ -0,0 +1,12 @@
+---
+date: 2024-09-01T09:29:16+10:00
+title: "Scalable interpolation of satellite altimetry data with probabilistic machine learning"
+heroHeading: ''
+heroSubHeading: 'Scalable interpolation of satellite altimetry data with probabilistic machine learning'
+heroBackground: ''
+thumbnail:  'images/news/2409Gregory.png'
+images: ['images/news/2409Gregory.png']
+link: 'https://doi.org/10.1038/s41467-024-51900-x'
+---
+
+In this [study](https://doi.org/10.1038/s41467-024-51900-x), **Will Gregory** and co-authors **developed an open-source Python library for interpolating sparse altimetry data**, using local Gaussian process models. The library allows them to generate full images of daily sea ice fields at high spatial resolution (5 km), which they hypothesize could be used to train ML models for sub-grid scale parameterizations.