test
Search publications, data, projects and authors

Free full text available

Other

French

ID: <

10670/1.98p18s

>

Where these data come from
Remote sensing of environmental dynamics: from the carbon balance of ecosystems to the assessment of water and agricultural resources

Abstract

This scientific volume of Habilitation in Diriger des Recherche summarises the work carried out in the four research laboratories where I have carried out my research since the end of the thesis: the Frontier Research Center for Global Change (FRCGC, Yokohama, Japan), the Centre d’Etudes Spatiales de la Biosphere (CESBIO, Toulouse), the Laboratoire des Sciences du Climat et de l’Environnement (LSCE, Saclay) and the Research Centre for the Organisation of Geographic Information (PRODIG, Paris). They also aim to understand the impact of this change on the carbon balance and on the availability of resources. The effects of climate variability on the phenology of boreal ecosystems, i.e. on their seasonality, have been spatialised thanks to the methodological developments in remote sensing carried out during the thesis. The longer growth season is a direct indicator of global warming during the 20st century, particularly marked in these regions. Remote sensing revealed a breakthrough in the spring in large regions of toundra and taïga in the northern hemisphere. Other similar periods of progress were identified earlier in the century thanks to modelling and in-situ observations but for smaller regions. Comparison with naturalist data from citizen science has shown that these variations are representative of those of a large cohort of species, at least in Canada.Phenology also rises in carbon exchanges between vegetation and the atmosphere, which have an impact on the climate. During the thesis, remote sensing was used to calibrate a vegetation operating model with the aim of simulating the carbon balance of a large region of Central Siberia. Since then, the ability of a nearby model (ORCHIDEE) to model carbon stocks has been assessed, revealing in particular that space gradients of Amazon forest biomass are not adequately replicated due to the representation of vegetation mortality in the model. As this work concerns two of the world’s major terrestrial carbon sinks, it is essential to have the best possible representation of the processes involved (phenological for boreal ecosystems, balance between mortality and productivity in the Amazon) in the models in order to obtain the best future projections of the state of ecosystems and climate. My integration in a laboratory and a department of geography as a researcher teacher led me to diversify my research topics and remote sensing practices. Through the mentoring of master’s students and the coaching of doctoral students, I addressed the issue of water availability in semi-arid environments (Argentina, Senegal) via remote sensing, providing a complementary view to that of my geographical colleagues: in particular, a direct relationship between the winter surface of the snow coat in the Andes and the flow of the rivers irrigating the Argentinian oasis was confirmed. The effects of this inter-annual variability of water resources on agriculture in oasis have been the subject of preliminary work based on the land cover classification and the temporal evolution of spectral indices. Further work on land use and its dynamics has been carried out or is being carried out in France and the Suds, with the objectives of studying human land use, assessing agricultural or forestry resources, and diagnosing vulnerability to hazards. In summary, the work presented relates to the temporal dynamics of physical environments, at seasonal, inter-annual and multi-ten-year time scales, with challenges to the carbon balance and the availability of resources. They include a large part of the methodological development of remote sensing, covering seasonal functioning of vegetation and hydrosystems, or land cover. They use data from high and medium spatial resolution optical remote sensing sensors, but also data from other types of sensors such as radars, gravimeters or microwave radiometers, allowing me to multiply my eyes on the spaces studied and to teach remote sensing in its technical and thematic diversity.

Your Feedback

Please give us your feedback and help us make GoTriple better.
Fill in our satisfaction questionnaire and tell us what you like about GoTriple!