Alberto Carrassi

Job Position: 
Researcher
Research Group: 
E-mail: 
Phone: 
+47 467 47 817
Alberto Carrassi

Research interests

My research interests focus on the development of efficient methods to assimilate and combine data with numerical models of environmental systems. I work at the bridge between applied mathematics and geosciences and my main expertise is in dynamical systems and stochastic filtering applications to geophysics. I am interested in fundamental issues in the theory of data assimilation and in efficient methods to quantify uncertainty in high-dimensional problems such those encountered in climate science. 

Some specific areas of research include:

  • Assimilation in the unstable subspace
  • Treatment of model error in data assimilation
  • Ensemble Kalman filtering
  • Adaptive observations
  • New applications of the data assimilation: model selection, attribution of climate change 

Together with Laurent BertinoI lead the Data Assimilation group at NERSC. I currently coordinate three international projects on efficient reduced order methods for data assimilation in high-dimensional chaotic system (REDDA from the Norwegian Research Council), on Bayesian data assimilation (EmblAUS from the Nordic Countries Research Council) and on data assimilation for Lagrangian sea-ice models (DASIM I and DASIM II from the US Office of Naval Research). I also work in the framework of the Nordic Countries Centre of Excellence on ensemble-based data assimilation EmblA.

Recent publications 

The complete list of my published papers is below. 

 The papers currently under revision are:

  1. Grudzien, C., A. Carrassi and M.Bocquet, 2018. Asymptotic forecast uncertainty and the unstable subspace in the presence of additive model error. J. Uncertainty Quantification (Available here). 
  2. Metref, S., A. Hannart, J. Ruiz, M. Bocquet, A. Carrassi and M. Ghil, 2018. Estimating model evidence using ensemble-based data assimilation with localization - The model selection problem. Q. J. Roy. Meteorol. Soc. (Available here
  3. Grudzien, C., A. Carrassi and M. Bocquet, 2018. Chaotic dynamics and the role of covariance inflation for reduced rank Kalman filters with model error. Nonlin. Processes Geophys. Discuss. (Available here

Supervision 

I am currently supervising two postdocs: Ali Aydogdu (postdoc, 2017- ) and Colin Grudzien (postdoc, 2016 - ) and co-supervising one PhD student, Francine Janneke Schevenhoven, at the University of Bergen, Gephysical Department and one MsC student Simon Clement from University of Grenoble (France).   

Past supervision: Patrick Nema Raanes (postdoc, 2016-2018 ), Matthias Rabatel (postdoc, 2016-2017), Maxime Tondeur (MsC student, 2016-2017, Ecole des Mines, France) and Robin Weber (MsC student, 2013-2014, Un. of Heidelberg, Germany).

News ... 

  • Together with Marc Bocquet, Laurent Bertino and Geir Evensen, we just published the review paper: Data assimilation in the geosciences - An overview on methods, issues and perspectives. See preprint here
  • My colleagues Olivier Talagrand, Marc Bocquet and I, have organised a short course entitled Data assimilation in the geosciences - An Overview, during the EGU general assembly 2018. It has been a huge success, at least in terms of affluence: the room (65 seats) was full 15 minutes prior to the start of the course! The slides of the course can be found hereSee also a link to the review paper on which this short course is partially based on. 
  • Following the success of its last year 1st edition, we have lunched this year as well the 4-days summer school Crash Course on Data Assimilation at NERSC, on 22-25 May, 2018
  • The BsC and MsC course on Data Assimilation at Dept. of Mathematics of the University of Bergen started on September 2017. 
  • The Springer book "Mathematical paradigms of Climate Science" has been published and it contains a chapter about model error treatment in DA I co-authored with Stephane Vannitsem.

Teaching

-- BsC/MsC Course on Data Assimilation at Dept. of Mathematics (University of Bergen).

   September-November 2017. Co-lectured with Geir Evensen, Laurent Bertino and Patrick Raanes

   The full description of the course along with its schedule can be found here, or via the Math department webpage here.

   Exams date: 30th November and 1st December.   

-- Short course "Data assimilation in the geosciences - an overview". SC1.13/NP8.5, EGU 2018. Vienna, Austria, 08-13 April. 

   Co-lectured with Olivier Talagrand and Marc Bocquet. 

Editorial and committee appointments 

Editor for Nonlinear Processes in Geophysics 

Member of scientific committee: EnKF annual workshop, EGU Session NP5.1 on Inverse problems and data assimilation in geophysics, SIAM Conference on applications of dynamical systems 

Recent and upcoming meetings, seminar and lectures  

  1. EGU 2018, 08-13 April 2018, Vienna, Austria
  2. SIAM conference on uncertainty quantification, 16-19 April 2018, Garden Grove, USA
  3. Seminar at the Atmospheric and Oceanic Science Dept. of University of Maryland, 23 April 2018, College Park, USA  
  4. Workshop: Prediction and data assimilation for non-local diffusion. International center for mathematical science. 04-05 June 2018, Edinburgh, UK
  5. 32nd IUGG Conference on mathematical geophysics, 23-28 June 2018, Nizhny Novgorod, Russia

Recent lectures at international schools   

  1. Crash-Course on data assimilation: Theoretical foundation and advanced applications with focus on ensemble methods "06-09 June 2017, Bergen, Norway.
  2. "Summer School on Data Assimilation and its application in oceanography, hydrology, risk & safety and reservoir engineering", 17-28 July 2017, Sibiu, Romania. 

Vacancies

Postdoc position open on Data Assimilation for sea-ice Lagrangian model. See vacancy description

If you are interested in working with me in the DA group at NERSC as PhD or Postdoc, here are some lines of research I am currently prioritising:

  • Reduced order ensemble-based methods
  • Efficient particle filters
  • Coupled data assimilation  

Some of them have already attached funding. For others we can explore possible sources.

Not-Science related...

 I am a certified Iyengar Yoga teacher (level: Base 2) and currently teach at yogarommet in Bergen. It would be great to meet you there !

Peer Review Publications and Books

37 total publications.

2013

  1. Palatella L, Carrassi A, Trevisan A. Lyapunov vectors and assimilation in the unstable subspace: theory and applications. Journal of Physics A: Mathematical and Theoretical. 2013;46(25):254020.

2012

  1. Carrassi A, Vannitsem S. Accounting for Model Error in Data Assimilation. In: Oberwolfach Reports. Mathematical and Algorithmic Aspects of Atmosphere-Ocean Data Assimilation. 2012. p. 3417 - 3471.