Skip to Content

CARBO France: Impact of Climate Extremes on Carbon Flows at the Level of France

The CARBO-FRANCE project is aimed at understanding extreme climate events and quantifying their impact on carbon flows, based on the data gained from the 2003 and 2005 droughts, across all of France. We propose to combine a variety of complementary approaches: 1) fine-tuned characterisation of the processes, based on flow measurements supplemented by ecophysiological metrics where the French workshop sites are concerned, 2) high-resolution digital modelling of the response of CO2 flows to climate extremes, 3) remote detection measures indicating physiological and hydric stress on plants, and 4) large-scale carbon balances deducted from atmospheric inversion results. We will look at the 1995-2005 period, marked by the two exceptional droughts in 2003 and 2005.

In order to successfully conduct this study, we will first draw upon finely-detailed analysis of turbulent flow measurements of matter (CO2 and water) and energy at 14 French workshop sites, as well as ecophysiological measurements, from soil carbon stocks and flows and plant growth. This analysis will help establish quantitative balances for the effects of recent droughts undergone by the various types of plants, forests and agrosystems, in terms of carbon and water flow anomalies, as well as loss of productivity in dry matter, in response to changes in climate parameters (soil temperature and humidity).

We will use two spatialised models of the carbon cycle in vegetation and soils: ISBA-A-gs and ORCHIDEE (including the STICS crop-growing model and the PaSIM prairies model). The simulated carbon and water flows for each of the workshop sites will first be compared with observations from workshop sites and the parameters set for the hydric and thermal control of CO2 and water flows will be adjusted as necessary. Thanks to climate data from Meteo France, we will then simulate, on each of the two national-level carbon flow models and across an 8-km mesh (10 times greater than that of common simulations of European carbon flows), over the 1995-2005 period. This “finer” scale will make it possible to take into account the heterogeneity of the surface and becomes compatible with the measurement scale (remote detection and sites). We will carry out critical analysis of differences between models, as well as a detailed validation of each model, with remote detection measures on the activity of plants (fAPAR, LAI, hydric stress) and agricultural yield. To round out the study, a lower-resolution simulation on Europe since 1900 will be used to put the findings about France into perspective as regards the rest of Europe, as well as the 1995-2005 decade as regards the 20th century as a whole.

A final innovative component of this project consists of comparing the variability of simulated carbon balances with the inversion of atmospheric concentrations. We have the atmospheric measurement network ORE-RAMCES, with 3 continuous stations, to which 2 towers were added in 2005; it is part of a European network jointly funded by the CARBOEUROPE Programme, which has some 30 stations. This network, unique in its coverage level, will make it possible to perform an inversion of daily flows on each gridpoint of the IPSL’s LMDz model, using the simulated flows and their spatio-temporal error structure as additional information. The findings, computed as averages across broad regions, will be compared to the flows estimated using ISBA–A-gs and ORCHIDEE.

Our objective is thus to quantify the impact of climate variability and the influence of climate extremes on carbon balances, incorporating knowledge gained, from the ecosystem level to the regional level. The approach we suggest offers the advantage of incorporating climate and edaphic variability into explicit flow computation, with high spatial resolution across mainland France (8 km) and a time step (1 hour) compatible with the variability of turbulent flows. This issue of the “separability” of processes and their contribution to biospheric flow variability on different time scales is central to the current debate on instituting a greenhouse gas emissions reduction policy.


Philippe Ciais, D. R. au CEA, responsable thème Cycle Biogéochimiques, LSCE, CEA Saclay
Philippe Peylin, C. R. au CNRS, laboratoire BioEMCO, INRA-INAPG


LSCE Laboratoire des Sciences du Climat et de l’Environnment
BioEMCO UMR BioEMCo, Université Paris 6, INRA-CNRS
LMD Laboratoire de Météorologie Dynamique
CESBIO Centre d'Etudes Spatiales de la Biosphère
FGEP-INRA Equipe Fonctionnement et Gestion de l'Ecosystème Prairial
EEF UMR INRA-UHP 1137 Ecologie et Ecophysiologie Forestières,
ESE Département d’Ecophysiologie Végétale (Unité ESE, UMR 8079),
EGC UMR 1091 INRA/INAPG Environnement et Grandes Cultures
CSE Unité INRA, Climat, Sol et Environnement
UEPF Unité d'Ecophysiologie des Plantes Fourragères

  • Impacts