Session: DO51A: Physics-Guided Machine Learning and Differentiable Programming for Ocean Modeling II eLightning

Time: 08:30 - 10:00

iPoster number: DO51A-05*

Advancing global sea ice prediction capabilities using a fully coupled climate model with integrated machine learning Machine learning models must be exposed to coupled climate processes to generalize in future climate model predictions.

This Friday at #OSM26 i’ll be showing some highlights from our recent Science Advances paper (doi.org/10.1126/scia...) on using ML to improve sea ice predictions in coupled models. Come on down if you fancy chatting about hybrid models, emulators, sea ice, or whatever you’re into!

Every paper I have published over the last 5 years has relied on these sea ice data sets. This is a real blow

Also, if you're interested in how ML can be incorporated into large-scale sea ice model Fortran code, check out the Github repo! github.com/m2lines/SIS2...

We also investigate the importance of exposing ML models to coupled feedbacks before implementation, the. impacts of coupled model biases on ML generalization, and forecasting extreme events

We have a new @m2lines.bsky.social pre-print which shows how machine learning can be used for state-dependent bias correction to improve global coupled climate model sea ice forecasts - take a look!

doi.org/10.48550/arX...

NOAA Halts Maintenance of Key Arctic Data at National Snow and Ice Data Center - Eos

Key snow and ice datasets from @nsidc.bsky.social may no longer be maintained or updated after a loss of support from NOAA.

eos.org/research-and...

So sorry to hear this Zack. I hope you're ok

i just checked scholar yesterday to see if this paper was out yet! Great job to Harry!

Not to mention being part of a very friendly climate research lab and @m2lines.bsky.social group! 😊

Postdoc Job Alert at Princeton University / NOAA GFDL
puwebp.princeton.edu/AcadHire/app...

If you're interested in sea ice modeling and machine learning, then look no further! This job is a great opportunity to be part of the exciting developments happening in the field of hybrid climate modeling

On my way down to DC for #AGU24! If you’re attending, why not check out our very exciting session tomorrow at 08:30 on Machine Learning for subgrid parameterizations and reanalysis! 🌎

Room: Marquis 12-13

agu.confex.com/agu/agu24/me...