M²LInES homepage

Code

Public repository and data access

Machine learning tutorials and tools

Equation Discovery

Code

Geospatial ML prediction workflow

Code

Machine Learning tutorial for Lorenz 96

Website Code

Software packages

GCM-Filters

Website Code

GPSat

Website Code

XGCM

Website Code

Models and implementations of parameterizations

Photo of Differentiable QG model in JAX

Differentiable QG model in JAX

Website Code

Differentiable QG model in pytorch

Code

Implementation of data-driven parameterizations in MOM6

Code

Implementation of parameterizations in pyqg

Code

Implementation of stochastic parameterization in MOM6

Code

Stacked shallow water model with stochastic subgrid momentum parameterization

Code

Ocean parameterizations

Photo of Data-driven dimensionality reduction and causal inference for spatiotemporal climate fields

Data-driven dimensionality reduction and causal inference for spatiotemporal climate fields

Code
Photo of Generative data-driven approaches for stochastic subgrid parameterizations

Generative data-driven approaches for stochastic subgrid parameterizations

Code
Photo of Neural network parameterization for vertical mixing

Neural network parameterization for vertical mixing

Code

Stochastic parameterization of subgrid momentum forcing

Code Model

Atmosphere parameterizations

Photo of Climate-invariant machine learning

Climate-invariant machine learning

Code
Photo of Climate-invariant machine learning

Climate-invariant machine learning

Code

Neural networks for parameterization of subgrid atmospheric processes

Code Data

Neural-Network Parameterization of Subgrid Momentum Transport in the Atmosphere

Code Data

Random forest to learn atmospheric parameterization

Code

Sea Ice parameterizations

Photo of Deep learning of systematic sea ice model errors from data assimilation increments

Deep learning of systematic sea ice model errors from data assimilation increments

Code
Photo of Machine Learning for Online Sea Ice Bias Correction Within Global Ice-Ocean Simulations

Machine Learning for Online Sea Ice Bias Correction Within Global Ice-Ocean Simulations

Code
Photo of Modeling the Winter Heat Conduction Through the Sea Ice System During MOSAiC

Modeling the Winter Heat Conduction Through the Sea Ice System During MOSAiC

Code

Datasets

Benchmark

Bias corrected ERA5 skin temperature over Arctic sea ice

Data

Pyqg parameterization benchmarks

Website Code

Usage example:

import xarray as xr
import fsspec
# Datasets are hosted on globus as zarr files
def get_dataset(
    path,
    base_url="https://g-402b74.00888.8540.data.globus.org"
    ):
    mapper = fsspec.get_mapper(f"{base_url}/{path}.zarr")
    return xr.open_zarr(mapper, consolidated=True)

In the cloud

CM2.6

Usage example:

import xarray as xr
import gcsfs
fs = gcsfs.GCSFileSystem()
fs.ls("gs://leap-persistent-ro/groundpepper/GFDL_cm2.6")
ds = xr.open_dataset(
    'gs://leap-persistent-ro/groundpepper/GFDL_cm2.6/
    GFDL_CM2_6_CONTROL_DAILY_SURF.zarr',
     engine='zarr')
  • This project is supported by Schmidt Sciences, LLC.