Pauline Bonnet

Logo

Passionate about climate physics, glaciology and science outreach. Life's a dream, go for it!

View My GitHub Profile

Hey, welcome on my personal page !

Feel free to leave comments and contact me for any sort of reason :

drawing drawing drawing

SCIENTIFIC INTERESTS

I am passionate about phenomena related to atmospheric physics, polar glaciers and powerful tools to study these physical processes such as numerical modelling and machine learning.

CURRENTLY

Looking for a new work opportunity in Marseille as a Projet Manager or Science Communicator in an ambitious projet related to Climate Sciences.

WORK EXPERIENCES

Postdoc in Atmospheric modelling and Machine Learning

February 2022 - April 2026 Affiliation: Prof. Veronika Eyring’s group https://www.pa.op.dlr.de/~/VeronikaEyring/index.html, Institut of Atmospheric Physics, DLR, Wessling (close to Munich), Germany.

This position is funded by the ERC Synergy Grant USMILE, and the EU Horizon Europe Projet AI4PEX,

Abstract

In climate model development, “tuning” refers to the important process of adjusting uncertain free parameters of subgrid-scale parameterizations to best match a set of Earth observations, such as the global radiation balance or global cloud cover. This is traditionally a computationally expensive step as it requires a large number of climate model simulations. This step also becomes more challenging with increasing spatial resolution and complexity of climate models. In addition, the manual tuning relies strongly on expert knowledge and is thus not independently reproducible. To reduce subjectivity and computational demands, tuning methods based on machine learning (ML) have become an active research subject. Here, we build on these developments and apply ML-based tuning to the atmospheric component of the Icosahedral Nonhydrostatic Weather and Climate Model (ICON) at 80 km resolution. Our approach follows a workflow similar to other proposed ML-based tuning methods: (1) creating a perturbed parameter ensemble (PPE) of limited size with randomly selected parameters, (2) fitting an ML-based emulator to the PPE to generate a large emulated ensemble with the emulator, and (3) shrinking the parameter space to regions compatible with observations using a method inspired by history matching. However, in contrast to previous works, we apply a sequential approach: the selected set of tuning parameters is updated in successive phases depending on the results of a sensitivity analysis with Sobol indices. We tune for global radiative properties, cloud properties, zonal wind velocities, and wind stresses on the ocean surface. With one iteration of this method, we achieve a model configuration yielding a global top-of-atmosphere net radiation budget in the range of [0, 1] W m−2, and global radiation metrics and water vapour path consistent with the reference observations. Furthermore, the resulting ML-based emulator allows us to identify the parameters that most impact the outputs that we target with tuning. The parameters that we identified to be mostly influential for the physics output metrics are the critical relative humidity in the upper troposphere and the conversion coefficient from cloud water to rain, influencing the radiation metrics and global cloud cover, together with the coefficient of sedimentation velocity of cloud ice, having a strong non-linear influence on all the physics metrics. The existence of non-linear effects further motivates the use of ML-based approaches for parameter tuning in climate models.

PhD in Mechanical Modelling of the Source of Glacial Earthquakes in Polar Region

Octobre 2017 - June 2021
Affiliations : Seismology team of IPGP, Centre des Matériaux of Mines ParisTech, PIMM laboratory in ENSAM, Paris, France

Evaluating glacier mass loss is a current concern to understand the rapid evolution of ice caps related to climate change. Iceberg calving is responsible for an important part of mass loss occurring at the front of marine-terminating glaciers in Greenland. Some just-calved thin icebergs are unstable : during capsize, a force is applied to the terminus and is transmitted to the solid earth. Therefore, a long-period (20s- 100s) seism is generated and seismic signal can be recorded by local and global stations. The emitted seismicity contains precious information about the source mechanisms involved during the calving process.

A versatile mechanical modelling of iceberg capsize has been developed by A. Sergeant et al. . Seismic signals provides constraints on the model and enables the calculation of the iceberg volume and its source dynamics. This model has been first tested on two particular events at the Helheim glacier, Greenland (25 July 2013 at 03:13 UTC and 31 July 2013 at 19:31 UTC) and the results are consistent with the iceberg volume measured separately using images available for this event. Following the work of Amandine Sergeant, a first objective of this PhD is the development and validation of the mechanical model (1) with the validation of the modelling of interaction between the iceberg and the surrounding ocean and (2) with the extension of the modelling to the whole system which account for frictional force between the glacier and the bedrock. A second objective of the PhD is to constrain the mechanical model with seismic signals from events of the last twenty years to evaluate Greenland glacier’s mass loss by icebergs capsize and calving.

PhD Advisors : Anne Mangeney http://www.ipgp.fr/~mangeney/, Olivier Castelnau https://pimm.artsetmetiers.fr/user/54, Vladislav Yastrebov http://www.yastrebov.fr/.

Research internship

Numerical stability analysis of a flexible beam in a confined fluid March - August 2017 | Master year 2
Meudon, France, Onera Aerospace lab, Aerodynamics, Aeroelasticity, Acoustics Department

Engineering internship

Analysis of marine vessels stability
April - August 2016 | Gap year
Leiden, Netherlands, Herema Marine Contractor, Marine Engineering Department

Research internship

Sensitivity analysis on a database of catenary geometry and current quality
October 2015 - March 2016 | Gap year
Paris, France, SNCF Railway company, Direction Innovation Research, Mechanical Systems and Interaction

Research internship

Experimental study of thermo-acoustic instabilities
May - July 2015 | Master year 1
Bombay, India, Aerospace Engineering Department, India Institute of Technology Bombay

PUBLICATIONS

EDUCATION

Master of Science

Modelling and Simulation in Structural Mechanics and Coupled Systems
September 2016 - March 2017, with Honours
ENS Paris-Saclay, ENSTA Paris, Centrale Supelec, top 5 French Uni

Master of Science

Applied Mathematics, Fluid and Solid mechanics
September 2013 - March 2017
ENSTA Paris, top 10 French Uni

Bachelor of Science

Intensive Science program, pre-requisite for French Uni
September 2010 - July 2013
Mathematics and Physics (PCSI-PSI*) in Lycée Condorcet, Paris

High School Diploma

Major in mathematics and life and earth sciences
Sep. 2007 - June 2010
with Honours
Lycée Louis le Grand, Paris, Prestigious high school with competitive admission

DOCTORAL TRAINING

Lectures

-Earthquakes dynamics, IPG Paris 2017 -Pedagogical training, IPG Paris 2017 -Ocean, Atmosphere, Climate ENS Paris 2018 -Contact mechanics, Mines Paris 2018 -Time-dependant seismology, TIDES-cost Prague 2018 -Sun, Northern Lights, UNIS Svalbard 2019

Glaciology Field Course

UNIS, Svalbard, Norway
5 weeks in February - March 2019
Doctoral course AG-825 in UNIS, Longyearbyen
Lectures: Glaciology, Thermalregimes, Remote sensing, QGIS Tutorial
8 days of field work: GPR, Ice Coring

Living-Faults Field Course

Greece
8 days in May 2018
Doctoral field course, IPGP
Analysis of the Corinth region tectonic activity, lithospheric deformation, and seismic history

Oceanographic Campaign

Mediterranean Sea
10 days in September 2017
MOOSE-GE campaign on IFREMER Atalante ship, LOCEAN
Monitoring the impact of climate change on hydrology and biogeochemical cycles in the northwestern Mediterranean Sea

CONFERENCES AND CONGRESS

On our work on modelling of iceberg capsize and the source of glacial earthquakes :

On our work on automatic tuning of parameters in Climate models:

AWARDS & GRANTS

SCIENTIFIC TOOLS

TEACHING

Graduate-students Mechanics tutorial

Material Resistance and Fatigue Failure
January 2018 - March 2019
80 hours of teaching in ENSAM Engineering school, Paris

Undergraduate Students oral examinations

Theoretical and applied physics
September 2014 - June 2015
44 hours of teaching in intensive science program in Condorcet School, Paris

High school students private lessons

Mathematics and Physics
October 2013 - June 2014
40 hours of tutoring to high school students in the region of Paris

SCIENTIFIC EVENTS AND OUTREACH

LANGUAGES

VOLUNTEERING AND HOBBIES