The POESY project aims at improving the probabilistic prediction of high-impact weather (HIW) events with an innovative combination of standard physical modelling approaches and computationally-efficient Artificial Intelligence (AI) methods. Probabilistic prediction currently takes the form of small ensembles of perturbed forecasts (O(10)) with a kilometre-scale resolution. In this project, AI and in particular deep generative models will be applied to increase both the probabilistic and spatial resolutions of ensemble forecasts. The ultimate goal is to provide ensembles of several hundreds of weather forecasts at sub-kilometre scales, which are practically unfeasible using standard modelling with the available computational resources. As a consequence, this project could lead to a strong methodological break in the design of ensemble forecasts. It shall also lift scientific locks for the real-time monitoring of HIW and in many domains where HIW predictive information is critical.

Snow cover is a key component of our weather and climate system, in particular through its high ability to reflect sunlight (albedo). Light absorbing impurities in snow such as soot and mineral dust decrease the albedo causing an acceleration of snow metamorphism and melt. Currently, the effect of light absorbing impurities in snow is only partially understood and investigated, as most of the past studies were focused on the direct radiative effect of impurity and only of a few of them were dedicated to the study of the non-direct radiative impact such as direct impact on snow metamorphism. Consequently, the deposition, evolution and impact of impurities in snow are not accurately taken into account in any snow model. The EBONI project aims at better understanding these processes through intensive field and laboratory measurement campaigns, focusing on soot and dust. This crucial knowledge will be transferred to a detailed snow model in order to better quantify and forecast the effect of light absorbing impurities in snow. Thus the EBONI project aims to implement a detailed snow model able to accurately quantify the radiative and non-radiative impacts of light absorbing impurities in snow enabling the representation of the complex feedbacks between snow, climate and impurities. EBONI is divided into three research actions. Action 1 aims to better understand the effect of light absorbing impurities in snow through intensive measurement campaigns and controlled experiments. Action 2 is dedicated to the implementation of the observed processes in the detailed snow model Crocus and action 3 will use the improved snow model to quantify the impact of impurities on snow melt timing, snowpack stability, run-off, glacier surface mass balance in the French Alps. Lastly, this model will benefit to the simulations of the Greenland Ice Sheet surface mass balance and of its potential contribution to global sea level rise. The EBONI project consequently enables to transfer the knowledge acquired at the micro-scale to the regional-scale effects of impurities in snow and to operational forecasting systems. The project will cover 48 months and will be mainly conducted at the Centre d’Etudes de la Neige (CNRM, Grenoble, France). It is based on the state-of-the art snow model Crocus and key developers of the model based at CEN are involved in the project. It offers a great opportunity to strengthen interactions between snow researchers at CEN and LGGE (Laboratoire de Glaciologie et de Géophysique de l’Environnement, Grenoble, France) especially within the framework of OSUG (Observatoire des Sciences de l’Univers de Grenoble). The budget for this proposal is 300 keuros. This includes a 3-year PhD fellowship for actions 1 and 2, and a 18-month research engineer position to facilitate actions 2 and 3. The PI M. Dumont will be 75% involved (36 months) and 7 other permanent researchers from CEN will be involved for a total of 37 more months. The PI has recently taken responsibility as research team leader and the EBONI project will thus crucially contribute to the emergence and growth of the research axes she proposed. This project addresses questions of direct interest for the society and the media. Our results will therefore be communicated to the public, including through dedicated web pages but also through innovative communication means via Météo-France and CNRS channels.

Our aim is to set up advanced assimilation techniques for improving the prediction of high impact weather, such as fog and thunderstorms. The ultimate goal is that the assimilation scheme is able to infer the balanced three dimensional, thermodynamic structure of the atmosphere from the available observations. We intend to blend the best features of variational methods with the ones of ensemble methods, combining the advantages of these two approaches and avoiding their drawbacks. The benchmark algorithm for data assimilation in numerical weather prediction is the four-dimensional variational scheme (4DVar). The 4DVar minimizes a cost function over a time window. Non-linear effects are taken into account using advanced formulations of 4DVar. However, the 4DVar algorithm requires the development of tangent linear and adjoint models which is particularly difficult at the mesoscale. The sequential nature of 4DVar raises concerns about the scalability on future computer architectures. Alternatively, (mostly linear) ensemble methods have become popular in the community, in various algorithmic forms, including the popular Ensemble Kalman Filter (EnKF). Such schemes allow for time and flow-dependent evolution of the analysis and forecast error covariances contrary to standard variational methods. However there are drawbacks to the current algorithmic implementation of EnKF variants, including a lack of scalability and suboptimal use of non-local observations. Despite recent progress, the question of a four-dimensional, non-linear, ensemble-based assimilation algorithm for applications in high dimensions remains open. The first objective of the SuNDAE project is to evaluate the scientific and computational feasibility of developing such an algorithm. We further plan to evaluate this algorithm for numerical weather prediction with the AROME model. AROME is the flagship, high resolution numerical model of Météo-France. AROME gathers contribution from the European and in particular the French research community. The development of the AROME towards more advanced data assimilation is possible through the use of the Object-Oriented Prediction System (OOPS), a flexible framework for data assimilation that separates the algorithm from its model-dependent implementation and that is shared by various institutes and universities. The SuNDAE project will develop a prototype of a non-linear assimilation scheme for the AROME model under OOPS. We plan it to be fully functional in order to allow for extensive comparison at the same spatial resolution. A progress of the AROME performance through a better assimilation scheme will go with the future investment on the Météo-France supercomputer and may be regarded as highly beneficial to the society. This development would valorize AROME, a tool that gather two European Numerical Weather Prediction consortium as well as the french research community. The choice of OOPS will allow the application of the algorithm to other communities such as the ocean and air quality, with reinforced collaborations between institutes. It is noteworthy that within our ocean-atmosphere community, similar question arise from the continuous increase in scale both in the observation and in the modelization areas. SuNDAE achievements will also benefit this wider community. Contacts are strong between scientists, being either focused on atmosphere, land or ocean, and we will took the various opportunities we have to interact and share the results of this work.

Wildfires are an ever-present and increasing threat in Europe. Due to climate change, the length of the fire season, as well as the size, intensity and severity of wildfires have increased in Europe, and this trend is set to continue. Wildfires are a massive, highly variable, and uncontrollable source of emission of aerosols to the atmosphere, which can affect the climate at the regional and global scales, along with impacts on air quality and weather. Despite the importance of wildfires in the context of climate change and their potential impact on the earth system, observations of biomass burning aerosols are surprisingly scarce in Europe and their impacts are highly uncertain. The EUBURN project proposes to conduct the first-ever multi-scale in-situ and remote-sensing observations to better understand the effects of increasing forest fires in southern Europe on the local and regional and global atmosphere. A wide range of airborne, satellite and surface-based measurements will allow to gather a unique data set on the properties of biomass burning aerosols throughout their atmospheric lifetime and their distribution in the tropospheric column to: (1) yield in new insights into the influence of fire emissions on the atmospheric composition with continuity from initial emissions to evolved impacts far from the source, (2) improve understanding of the physical and chemical processes that govern aerosol formation and evolution in biomass burning aerosols and (3) quantify key aerosol variables relevant for meteorological and climate modeling and satellite retrievals.