Senior researcher-MachineLearning in operational oceanography-Two 50% contracts RBINS and UClouvain

Offre

L’UCLouvain recherche un·e Senior researcher-MachineLearning in operational oceanography-Two 50% contracts RBINS and UClouvain

Référence du poste : SF  35282
Publication interne et externe jusqu'au 18 juin inclus

Pour l'Institut Earth and Life Institute du Secteur des sciences et technologies
Site principal :  Louvain-la-Neuve
Contrat à temps plein (100%) pour une durée déterminée
Entrée en fonction : dès que possible et au plus tard au 1 mars 2024

Contexte

The Operational Directorate ‘Natural Environment’ (OD Nature) at the Royal Belgian Institute of Natural Sciences (RBINS) and the Earth and Life Institute (ELI) at the Université catholique de Louvain (UCLouvain) jointly offer a full-time (100%) research position to develop the next generation of marine forecasting and nowcasting techniques, products, and services for the Belgian part of the North Sea. This vacancy is part of the BELSPO FED-tWIN research profile Prf-2022-021 ‘Next-TIDE’.  

The FED-tWIN programme is a federal research programme to support sustainable cooperation between the ten Federal Scientific Institutions (FSI) which fall under Belgian Science Policy (BELSPO) and the eleven Belgian universities:  https://www.belspo.be/belspo/research/FEDtWIN_en.stm.

Next-TIDE research profile

Next-TIDE is an innovative, long-term collaboration between the Marine Forecasting Centre (MFC) at RBINS/ODNature and the Earth and Life Institute (ELI) at UCLouvain, aimed at advancing and implementing cutting-edge machine learning techniques for marine forecasting and nowcasting applications.

As the blue growth economy drives the expansion of activities in the Belgian Part of the North Sea (BPNS), there is a growing demand for new end-user products provided by RBINS. The current MFC models are insufficient to meet these needs, as they have limited resolution, speed, data integration capabilities, and lack the ability to quantify uncertainties. Machine learning has already made significant progress in atmospheric sciences, dramatically improving weather forecasting speed and accuracy.

Applying machine learning, and more advanced deep learning techniques, to oceanic applications presents unique challenges. To address the need for new end-user products, RBINS and UCLouvain have developed a strategic roadmap for the methodical integration of machine learning techniques into the marine forecasting process.

The immediate objective is to create a nowcasting system capable of providing detailed and accurate predictions up to 12 hours ahead. Long-term goals of this research profile include using machine learning techniques to develop: 1) computationally efficient data assimilation techniques for operational oceanography systems, 2) hybrid models for sub-grid scale processes, 3) automated quality control and bias estimation systems for observations, 4) model accuracy assessment systems, 5) emulators for sub-grid turbulence and non-hydrostatic processes, and 6) end-user applications.

Description du poste

Research

The FED-tWIN Researcher will strive to become a leading expert in applying machine learning techniques to marine forecasting and nowcasting products and services. They will establish their own research group by designing innovative research projects, securing competitive funding, coordinating projects, and mentoring junior researchers, including PhD candidates.The researcher will foster collaborations with national and international partners, including those within the digital twin of the ocean initiative. They will actively disseminate research findings within the global academic community through publications and conference presentations.Working at the intersection of the RBINS Marine Forecasting Centre and UCLouvain Earth and Life Institute, the FED-tWIN Researcher will collaborate closely with colleagues from both institutions, further solidifying the existing partnerships between them.

Teaching: The FED-tWIN Researcher will supervise Master's students and research internships at both institutions. The ideal candidate must thrive in a multicultural and multilingual environment, working effectively with students, researchers, professors, and technical/administrative colleagues.

Services : The FED-tWIN researcher will directly enhance the products and services offered by the RBINS Marine Forecasting Centre. They will actively participate in the operational and administrative aspects of both departments, as well as contribute to fundraising efforts for projects related to the research profile. This includes integrating their research activities within the broader research and outreach strategies of both institutions. With a 50% appointment at each institution, the FED-tWIN researcher will have limited administrative responsibilities at the department or faculty level.We offer a full-time researcher position that will take place as two 50% contracts at the RBINS and UCLouvain.
 

RBINS offers 

• A part-time 50% open-ended employment contract as a “work leader” (SW2 salary scale for senior researcher).
• You will do 50% of your work in a dynamic, challenging, varied, and stimulating research environment at the heart of Brussels, next to the European parliament.
• A full reimbursement of public transport for commuting, or a bicycle allowance if you commute by bike.
   

UCLouvain offers  

• A part-time commitment (50%) as a postdoctoral researcher for the first five years. The initial five-year contract will be an undetermined contract with no intention of termination. An evaluation will be established after the initial 5-year period. After the first 10 years, for which we have a firm commitment, the continuation of the contract will depend on the results and the availability of suitable funding or an academic position for which we cannot make any commitment nowadays.
• A dynamic and stimulating work environment at UCLouvain's Louvain-la-Neuve campus.
• Full reimbursement for commuting via public transport and/or a bicycle allowance.

The start date will be set by mutual agreement with the candidate, but this date shall not be later than 1st March 2024.

Qualifications et aptitudes requises

Diploma

The candidate holds a PhD degree in Sciences (Physics, Mathematics, Oceanography, Meteorology, Marine Ecosystems, etc.) or in Applied Sciences (Mechanical Engineer, Applied Mathematics, Bioengineer, etc.). The PhD degree was obtained at most 12 years prior to the job application submission date. The 12-year period is extended by one year for each maternity, parental & adoption leave of the candidate and for each long-term sick leave of the candidate or immediate family members.

Specific expertise

The candidate will have a solid research expertise in Earth Sciences (preferably in oceanography, meteorology, hydrology, geology and/or marine ecosystem) and in Data Sciences (preferably in machine learning, data mining, etc.).The candidate must have a good background in statistics and probability.

Experience(s)

The candidate is expected to have a strong research curriculum within the domain of oceanography or machine learning. The research quality should be apparent from publications in leading international journals.Previous work experience in international projects is desirable.Previous experience in writing research proposals would be a significant asset.

Technical skills

The candidate will have outstanding programming skills, with experience in programming languages such as Python, C/C++ or Fortran 2008. Experience with neural network frameworks (such as PyTorch, Tensorflow, JAX, etc.) will be an advantage.

Generic skills

The ideal candidate must possess the following skills:

• A passion for applied marine research in a multidisciplinary context.
• Proactive approach to achieving objectives within the given timeframe.
• Flexibility to collaborate effectively in a team on multiple projects.
• Strong communication skills, with the ability to engage with scientists from diverse backgrounds.
• Excellent English communication skills, both oral and written. Knowledge of French and/or Dutch is an advantage.

Processus de sélection

To apply, please email a single PDF file containing: (i) a concise motivation statement for your application, including an outline of your future research plans (maximum 2 A4 pages, in English); (ii) your curriculum vitae accompanied by a publication list and contact information for 3 professional references; (iii) the academic dossier containing all pertinent elements (degree and diploma copies with associated grades, etc.). The application should be sent to job-next-tide@naturalsciences.be (subject of email: “job application”) before June 18th 2023, 23:59 (CET).

Only applications that correspond to the profile and sent within the prescribed deadline will be accepted and responded to.

The selection committee will review all of the applications as soon as possible after the application deadline. As soon as a decision has been made, we will inform you about the next steps in the selection procedure.

For further information regarding the job itself or the terms and conditions of employment, please contact Dr. Sébastien Legrand (slegrand@naturalsciences.be) and Prof. Emmanuel Hanert (emmanuel.hanert@uclouvain.be).  

General information about RBINS and UCLouvain research can be found on their respective websites: https://www.naturalsciences.be/en/science/home and   https://uclouvain.be/en/research/welcome-desk.