The ESTIMR team has as its main objectives: understanding and modeling climatic and environmental variability at different spatial scales – from very large structures related to atmospheric dynamics to very local phenomena – and at different temporal scales – for the study of past climates, present processes, and future developments. Within this vast framework, one of ESTIMR team’s strengths lies in the use and development of state-of-the-art statistical models tailored to climatic issues, through sustained multidisciplinary interaction between climatology, modeling, physics, statistics, and artificial intelligence.
Statistics and Dynamics of Climate Extreme Events
How can statistics and dynamical systems help study extreme events? Atmospheric extreme events such as storms, heatwaves, cold spells, and intense precipitation have serious effects on human activities and natural ecosystems. These events are difficult to study because they are rare or may never have occurred before, and because their effects can be extremely localized in space and time. Understanding extreme events thus requires special tools to overcome sample size and resolution issues.
In this context, our efforts have focused on three main axes: i) studying the relationship between the occurrence of extreme events and their large-scale dynamical or thermodynamic triggers, ii) providing robust estimates of their return periods in a changing climate, iii) improving the representation of observed extreme events in climate simulations through statistical downscaling.
Extreme Events and Climate Extremes
Extreme Event Attribution is an emerging field in climate science, intersecting statistics, atmospheric dynamics, and social sciences. One of the goals is to describe if and how the probability of an event, like a heatwave, depends on climate change. The development of climate services and interaction with society require scrutiny of numerous scientific and communication aspects. Discussion with professionals outside the scientific community, or with scientists from other disciplines, reveals specific scientific questions and requirements from the scientific community. New questions also arise, to which there are still few answers. This calls for specific activity linking climate science to the challenge of meeting these requirements.
Over the past four years, the ESTIMR team has developed some examples of impact studies involving dialogue with external professionals. This activity complements the development of statistical tools such as bias correction and downscaling, which already develop scientific aspects usable for climate services. This has led to several projects and contracts, and a number of applicative studies have been conducted. We have primarily focused on activities in four sectors: energy, insurance, agriculture, and health.
Combining research on extreme weather events with artificial intelligence (AI) opens up exciting new perspectives for climate risk prediction and management. AI offers powerful tools for analyzing large sets of climate data, identifying complex patterns, and forecasting future trends with increased accuracy. By integrating these emerging technologies, we are able to strengthen our ability to anticipate and mitigate the impacts of extreme weather events on human communities and ecosystems. This convergence of climate science and AI represents a crucial advancement in our understanding and response to the challenges posed by climate change, paving the way for fruitful interdisciplinary collaboration to build a resilient and sustainable future.