Inria at COP21
© Inria / AIRSEA - CNRS / LEGI / Photo N. Hairon
Inria researchers constantly work to help improve our understanding of the phenomena that impact on our planet, to predict future change and seek solutions. Medicine, biology, robotics and ecology are among the many fields of application in which digital science now has a key contribution to make. This is also true of the effort to reverse climate change, a fundamental challenge for the whole planet.
Digital science provides support for energy transition, helping to address the issues of primordial concern to society. It is an essential component in assessing climate issues, imagining climate change scenarios and aiding decision-making processes. A number of our teams are thus working on improving our understanding of the phenomena that impact on our planet, predicting future change and seeking solutions to counter the phenomena observed. Their studies contribute to our knowledge and are used as source material for international discussions, but they also lead to collaborative projects with industry and technology transfer to small- and medium-sized enterprises, as well as to the development of spin-offs. Digital science also enables citizens to be part of building this initiative: individual people can add to our data, rendering the phenomena more tangible, providing information for discussion and ensuring that awareness is raised across society.
Antoine Petit, Inria Chairman and CEO
Digital Science and Climate Change: how Inria teams are involved
In addition to research on ecological impact, the current situation regarding climate change raises many questions in terms of changing weather conditions and the resulting vulnerability of many sectors of activity. Digital and computing tools, in the broadest sense (numerical modelling, high performance computing, statistical tools, observation and communications networks), play a central role in addressing these challenges since they help us to understand the physical, economic and social mechanisms inherent in reversing climate change and its consequences.
Ecology involves a huge number of cross-disciplinary challenges, which mobilise every branch of digital science, and also has the advantage of the cross-disciplinary nature of digital science with its myriad applications in other fields of research.
François Sillon, Deputy CEO for Science
Understanding climate change and the stakes involved
Models and simulation, combined with the possibilities for observation and detection, are used to study the complex phenomena involved with increasing precision. In particular, digital science helps develop computerised models, for example, for tracking and predicting future climate change. Such tools are used to bring together many different kinds of observation. For example, we can fine-tune a model to simulate oceanic and atmospheric currents and their impact on temperature, precipitation and wind, etc., and ultimately, on crop yields, the availability of freshwater, biodiversity, etc.
To address the questions raised by climate change and its impact, the IPCC (the Intergovernmental Panel on Climate Change) reports on studies published in recent years, which draw on such models of the climate system together with economic and demographic studies. They draw up scenarios projecting climate change, assumed to cover a broad spectrum of possible changes. Digital science is thus at the service of the IPCC. This implies a major challenge for the mathematicians and digital scientists because of the huge difficulties involved in modelling.
The AIRSEA team is involved, among other things, in improving the two French climate models (IPSL-CM and CNRM-CM) through research in several areas: numerical schemes (how to accurately solve complex mathematical equations relative to ocean and atmospheric physics), coupling sub-models (e.g. how to correctly represent mass and energy transfer between the atmosphere and the ocean), and data assimilation (how to "adjust" a model using available temperature or precipitation observations). Such questions raise many challenges for numerical optimisation and fundamental mathematics. The research carried out by AIRSEA is integrated into ocean and climate models, and made available to the scientific community.
The GEOSTAT team works on obtaining very high resolution spatial mappings of global physical variables, for example, maps of partial pressure of greenhouse gases or flows transferred between the ocean and the atmosphere, using Earth observation acquisition data and methods used in statistical physics and signal processing. Determining these physical variables at high spatial resolution is essential in order to make quantitative assessments of global warming.
The EVA team is working to provide a revolutionary remote real-time monitoring solution involving a cutting-edge technology from Linear Technology and Metronome System. The result is a mesh networking solution where sensor poles are densely deployed in hydrologically-representative locations, with each sensor reporting snow depth every 15 minutes, in real-time. 15 such networks have been deployed, a total of 945 sensors. The wealth of information these networks produce is far denser than what manual measurement campaigns produced in the past, allowing far more accurate hydrological models, and a better understanding of the snow melt process. This technology can be extended for predicting the production of hydroelectric power plants, and is mature and ready to be extended to Europe.
Predicting the consequences of climate change
Rising sea levels, changes in rainfall patterns, more intense extreme weather phenomena, melting sea ice, and the impact on flora and fauna: the consequences of global warming are extremely varied, including effects on crop yields, access to drinking water and soil erosion. In response to these new problems, Inria researchers are developing models and software designed to predict local and regional impact.
The problem of air pollution is also becoming more important at local (air quality), regional (cross-border pollution), and global scales (greenhouse effect). Modelling systems are used to evaluate the scale of ongoing changes: short and mid-term predictions, case studies, studies on the impact of industrial sites, etc.
Another major challenge is related to determining the impact of future political and economic choices. Deciding whether or not to build a dam, assessing the impact of an urban development project, choosing a waste treatment technology: matters such as these all involve technological choices that will have repercussions in terms of sustainable development. Moreover, local authorities, from the smallest towns to regional councils, seriously lack the tools needed to make such choices.
Analysing changing climate phenomena relies heavily on the results of research in computer science and mathematics. One of the key issues involved for our branches of science entails developing effective models and efficient simulations, particularly large-scale predictive models that incorporate highly diverse descriptive levels. The relevance of such models is also dependent on the most recent advances in high performance computing, and in processing huge volumes of data. In our current strategic plan, one of our priority areas of research is "the Digital Planet".
François Sillon, Deputy CEO for Science
The LEM0N team focuses on modelling natural processes in coastal areas (the erosion of beaches, coastal oceanography, submergence and flooding, pollution, etc.). The team's overall aim is to contribute to designing and improving models used to simulate such phenomena, and to then couple them with each other or with external data) to produce a global projection system that factors in the maximum number of natural phenomena. In particular, the team is interested in real-time simulation of flooding in coastal areas, which, together with intense episodes of rainfall and rising sea and ocean levels, is becoming increasingly frequent.
© Inria / CLIME - Numtech
The CLIME team is working on coupling numerical simulation models with environmental observations, including images such as images acquired via satellites, drones, and ground-based sensors. Part of CLIME's research focuses on studying air quality, from the scale of a continent to the local scale of a city. The researchers are also developing air pollution impact analysis and projection tools at the urban scale.
The STEEP team explores two new types of decision aid tools. The first simulates complex systems in which many factors interact, including, in particular, human factors. The aim here is to predict the impact of such policy decisions on biodiversity and local resources, including forward-planning in light of climate scenarios and global economic developments. The second tool being developed is designed to optimise choices in terms of costs, from the point of view of economics, the environment and society. These tools are being studied with a view to applications in urban planning, farming and in understanding food supply sectors, etc.
The CARDAMOM team develops robust and adaptive tools for the simulations of complex flows many of which are inspired by applications related to the environnement: the study of the impact of ocean waves on the coast, and the conversion of their potential energy in electricity, the study of Organic Rankine Cycles (ORC), etc. The members of team provide numerical models answering to two essential needs :
- being able to control and locally modify the differential equations and the numerical methods used, in order to maximize the physical relevance and the efficiency of the models
- quantify the impact of uncertainties on the physical conditions and include them in the adaptation mechanisms.
Playing a part in developing solutions: the example of energy transition
Scientists are seeking concrete solutions to the problems raised by the challenge currently faced by our planet. Their activities centre around issues such as improving energy efficiency and reducing the impact of energy production and use on the environment. In the sectors of renewable energy and nuclear fusion, we need models and simulations to optimise new energy production technology. This must go hand-in-hand with efforts to reduce energy consumption. In the age of Big Data and server farms, Inria research teams analyse the energy footprints of operations, computations, requests, software functions, and of data storage and processing, with a view to optimising computers, languages and algorithms to improve energy efficiency. Research studies in Digital Science also play an increasingly large part in "smart", energy-saving management for various other systems, including vehicles, building and cities.
Because of the limited availability of fossil fuel resources and, above all, their ecological cost, one current trend is the development of renewable energy. We are seeing rapid growth in solar and wind energy in particular, although they still have the disadvantage of unpredictable output. Inria is also involved in research on nuclear fusion, through the international ITER programme. At the scale of a wind farm, there are diverse sources of uncertainty relative to electricity production which are difficult to mitigate, requiring innovative modelling and computing methods. Inria is also involved in the development of other renewable energy sources, such as biofuels and hydraulic energy.
The TOSCA team is developing a new method for calculating wind resources, which can dialogue with a dynamic weather forecast system combined with detailed modelling of wind turbines and where they are positioned at a site. This makes it possible to predict and simulate electricity output and uncertainty due to atmospheric turbulence.
The BIOCORE team carries out research on designing, modelling, analysing, controlling and optimising artificial ecosystems. It helps protect the environment by developing new energy sources, preventing water pollution and replacing the use of toxic chemicals on crops. Their research has many applications, especially in the production of low carbon footprint biofuels using micro-algae.
The ANGE team is working on modelling, analysing and numerically simulating free-surface flows found in geophysics. The research carried out combines methodological and applied aspects. The main scientific challenge is thus to obtain effective models that must be adapted to the physical phenomenon under study, and that can be efficiently discretised and validated. Among other subjects, the team focuses on phenomena that combine hydrodynamics and biology, together with marine energy (wave power, tidal power and biomass). Regarding this last point, the team aims to optimise the system used, for example, for tidal power, to increase energy output.