Researchers' Night in Annecy

A negative body image—or a distorted perception of one’s body—can lead some people to try to reshape their bodies in order to feel less dissatisfied with their weight and/or shape (through dieting, exercise, medication, etc.). When body hatred is too intense or one’s perception of the body is too distorted, the behaviors adopted to reshape the body can become pathological, even leading to eating disorders. These disorders can take various forms: anorexia nervosa, bulimia nervosa, and binge eating disorder.

Morgane Metral has been a faculty member and researcher at the University of Savoie Mont Blanc since 2018. Her research focuses on the origins and impact of body image distortions in psychopathology and eating disorders, with a particular emphasis on their treatment and prevention among the general population.

In Savoie, smart cameras monitor avalanches, but at night, the artificial intelligence seems to be daydreaming as well. These errors reveal the limitations of automated systems in the dark. A local and vivid way to illustrate the challenges AI faces in monitoring natural hazards.

Christophe Lin-Kwong-Chon’s research focuses on developing deep learning models capable of analyzing complex data (satellite images, radar data, or videos) to solve real-world problems: measuring glacier thickness, monitoring avalanches, mapping land use, and optimizing hydrogen production. He is particularly focused on making these models “intelligible,” so that AI is no longer a black box but a reliable and transparent decision-making tool.

In the mountains, as glaciers retreat and the permafrost in rock faces and glacial rock formations degrades, the resulting debris forms sediments that can potentially be carried away by a flooding torrent. Torrential floods are a well-known and feared hazard in Alpine communities. The intensity of sediment transport makes them particularly destructive. While the retreat of ice creates new sources of sediment, having more sediment alone does not necessarily lead to more flooding. What other factors come into play?

Anaïs Fichot is currently in her second year of doctoral studies at the EDYTEM laboratory. She holds a master’s degree in geography and cartography from the University of Savoie Mont Blanc.

Earthquakes are unpredictable and devastating natural phenomena caused by the sudden release of energy accumulated in the Earth’s crust. Their geological complexity and the lack of reliable precursor signals limit any deterministic prediction (date, time, location). Thus, despite technological advances, accurately predicting when, where, and with what intensity an earthquake will occur remains a major challenge for science. This presentation explores the current limitations of earthquake prediction, the methods used, and ongoing research avenues.

Blandine Gardonio is a seismologist. She studies the weak signals that precede earthquakes and combines field observations, sensor installations in the Alps, and global-scale analyses to better anticipate natural hazards.

Crystal-clear water, snow-capped peaks, silence broken only by the wind: Alpine lakes seem untouched. Yet even at high altitudes, they receive pollutants carried by the atmosphere. Highly sensitive to nitrogen inputs, their ecological balance can be disrupted, leading to algal blooms. By studying 26 lakes in the French Alps, we have shown that more than half of the dissolved nitrogen comes from human activities, particularly agriculture and the burning of fossil fuels. The soils and vegetation of the watersheds play a key role in transferring this invisible pollution to the lakes.

Maria Page is currently in her third year of doctoral studies at CARRTEL. She holds a master’s degree in chemistry from the University of Bristol (United Kingdom).