Forecasting the weather beyond a few days

SCTODAY

Climate prediction is his forte . At 32 years old, François Massonnet has just been appointed FNRS research associate. While he has been working on the subject for years, he looks forward to digging deeper.

His education: a civil engineering degree, enriched by courses in meteorology, climatology and fluid dynamics as part of his master’s degree completed in 2009, and a PhD, completed in 2014, which focused on the polar regions. ‘Their rapid climate changes provide a taste of what may soon happen elsewhere in the world’, he says. He delved further during two years in Barcelona. ‘I specialised in producing and analysing climate forecasts of a few months to a few years using a supercomputer. In particular, I tried to learn how to extend Arctic and Antarctic pack ice forecasts by several months. This information is crucial because it makes it possible, for example, to more effectively plan shipping routes in these remote and very hostile areas.’

Back at UCLouvain since 2016, he has continued his work at the Georges Lemaître Centre for Earth and Climate Research (Earth and Life Institute). ‘Doing research abroad makes you realise that there’s no single way to work. Working in another research group is also about seeking expertise elsewhere and coming back with new tools.’ 

As an FNRS postdoctoral researcher, Dr Massonnet used two complementary sources of information to estimate the climate’s current and future state: climate models, or computer codes that operate on the basis of known physical laws (these models are often imprecise but offer excellent spatial and temporal coverage), and climate observations via satellite data and field observations (these observations are more accurate but poorly distributed in space and time). ‘This approach allowed me, for example, to propose an unprecedented reconstruction of the Antarctic pack ice over the last 30 years. I used scattered information combined with model outputs to propose a history of the evolution of pack ice thickness. The goal is to understand how and why it evolves.’

He also considered a study of Arctic pack ice evolution. ‘This work is in the latest report by the IPCC, the Intergovernmental Panel on Climate Change. The conclusion is that if greenhouse gas emissions continue at today’s rate, Arctic summer ice will likely be gone by mid-century – in 30 years!’

In the future, Dr Massonnet would like to take advantage of his permanent FNRS position to direct his research towards the development of a seasonal forecasting system. ‘Broadcast, print or online news typically provides two types of information: a forecast for the coming days and “climate change” information, which projects into the very long term. I’ll work between these two time frames.’ The question he intends to answer: Can we extend the limits of our regional weather forecast beyond a few days? ‘This is a fairly new subject, because it requires phenomenal means of data processing, computation – in short IT resources. And from a scientific point of view, it’s complicated. We have to bring together physicists, mathematicians, statisticians, computer programmers…which explains why the sector took a long time to take off. Its implementation is a response to an increasingly pressing demand from governments, businesses and sectors such as agriculture for reliable climate information concerning several months to several years (on which to base infrastructure investment decisions, for example).’

One region will be at the centre of his research: the North Atlantic. And for good reason: as the winds come from the west, the region’s climate has great influence on life in Europe.

As an FNRS research associate, Dr Massonnet will be able to set up a team. ‘In particular, it’ll be necessary to optimise the use of climate data. We’re starting to use artificial intelligence style algorithms to extract useful forecasting information. The amount of climate data has exploded in 30 years, so the idea is to get a handle on a volume of information that’s completely overwhelming.’
 
To do that, there’s no question of confining himself to the office: he intends to meet other researchers, from the same sector and working on other issues. ‘I intend to launch an internal process at the university concerning my research subject. In any case, I’m eager to start, this permanent position allows me to project into the long term. I’m more motivated than ever.’

Published on August 29, 2018