Who am I ?
I work on a variety of things, from the reconstruction of the climates of the past to seasonal forecasting. I have a strong interest in methods (statistics, machine learning particularly), in scientific computing and open science.
Where do I come from ?
I hail from Burgundy, France, and I obtained my Ph.D (in 2004) in Dijon. After about a year at the LSCE, near Paris, I decided to expand a bit my horizons. In 2006 I was awarded a post-doctoral fellowship from the University of Cape-Town, and I spent about 3 years working at the oceanography department (where I am still a research associate), notably with Prof. Mathieu Rouault, another french import in South Africa and one of the most versatile scientists I have known. In 2009 I joined the CSIR as a senior researcher in the Southern Ocean Carbon and Climate Observatory (SOCCO) where I had the great privilege of working with a bunch of awesome people.
In 2012 we decided to relocate our little family to New Zealand, and I obtained a position at the National Institute for Water and Atmospheric research (NIWA Ltd.).
Since joining NIWA, I have been working on a variety of projects, and been involved in both basic research (e.g. paleoclimate reconstructions) and very applied science (e.g. forecasting climate at sub-seasonal to seasonal time-scales). A common theme though: all this involves processing and analyzing data - oftentimes a lot of data - and using statistical methods to try and make sense of this data, extract patterns, and make predictions. Along the years I used various tools, from Matlab to R, but over the past few years I have transitioned completely to Python: the Python Scientific Ecosystem has grown exponentially both in quality and quantity, and now there's almost nothing that you can do in R or Matlab that you cannot do in Python, and on top of being open-source, Python is just fun to code.
The tools of the trade
I use Python and of course the Python Scientific ecosystem, in particular, this is what I use (almost) daily:
and of course the IPython (now Jupyter) notebook.
Besides Python, I occasionally use:
CDO [The Climate Data Operators], for operations on netcdf files and
file format conversions (i.e.
- NCO [the NetCDF Operators], also handy for netcdf extractions / concatenations