1 posizione di post-doc in “Machine Learning for climate models spinup”
Si segnala una posizione di post-doc in “Machine Learning for climate models spinup” nell’ambito del progetto MARTINI (MAchine leaRning Techniques for spINup acceleratIon) presso la sede CNR-ISAC di Torino con scadenza il 5 Gennaio 2023. Maggiori informazioni di seguito.
The increase of the resolution in the Earth System Models is strongly limited by the cost of the
preparatory simulations called “spinup”.
Such simulations, whose length can exceed hundreds of years, are necessary as all the components of the climate model must be brought to thermal and radiative equilibrium. Running hundreds of years at high resolution (finer than 25 km mesh) is currently prohibitive even on most advanced HPC, thus severely limiting the effective feasibility of high-definition climate simulations. Given the large amount of climate model data currently available it could be possible to exploit machine learning approaches to develop data-driven “spinup” techniques aiming at reducing the computational cost of spinup integrations by an order of magnitude. Developing such methodology could save hundreds of thousands of core hours, marking a breakthrough for climate integrations. This will be investigated in the project MARTINI (MAchine leaRning Techniques for spINup acceleratIon).
Institute of Atmospheric Sciences and Climate of the Italian National Research Council (CNR-ISAC)
will be opening a 2-year position to work on developing machine-learning approaches for
accelerating Earth System Models spinup. The position is the result of a joint initiative supported by
the OGS and CINECA.
● Goal: Design and develop, from both the scientific and the technical side, a data-driven novel
approach to accelerate the spin-up simulations of Global Climate Models.
● Degree: physics, engineering, informatics. PhD required.
● Experience: Climate data analysis, Machine learning. Experience in Earth-system modeling is
welcome but not fundamental.
● Coding: Python (possibly some background in Tensorflow + Keras), other languages welcome
(R, Julia, etc.), version control (git), experience in parallel environments (MPI and/or
OpenMP) and GPU.
● Salary level: 1650 per month (net)
● Deadline for application: 5th of January 2023
● Starting date: Starting date February 2023, 24 months.
● Location: CNR-ISAC, Torino, Italy (flexibility can be agreed).
If you are interested interested in joining our team for this exciting project and for any further
information please contact:
Paolo Davini (firstname.lastname@example.org)
Susanna Corti (email@example.com)