Aerosol particles contribute directly and indirectly to several atmospheric chemical and physical processes of crucial importance to meteorology, environment and climate. To better understand their role, we need to better quantify their sources, optical and microphysical properties and interaction processes. We address these questions through a combination of observation and modelling.

Our first scientific motivation is to analyze the information provided by the A-Train and prepare for Earth-CARE observations on aerosol sources (natural and anthropogenic). We aim to characterize long-range transport, and the evolution of aerosol optical properties during transport. Space observations will help better understand and quantify the relative impact of long-range transport on regional budgets. The long record of CALIPSO is also a good opportunity to look at the interannual variability of the aerosol distribution. 2020 data will be special due to the significant reduction of transport emission during the covid lockdown in Asia, Europe and America.

Joint analysis of satellite observations and ground-based network data (see T3.2 and T5.1) plays a specific role, as ground-based data give access to detailed information on aerosol composition, optical and microphysical properties at the local scale. Ground-based lidar (combined with passive remote sensing) can establish a link between the in-situ observations and large scale spaceborne measurements. The good calibration of the CALIOP lidar is also a good opportunity to discuss the intercomparison of the ground based lidar calibration.

Preparation of the Earth-CARE aerosol data product retrieval and validation, as well as intercomparison of present and future spaceborne lidar data will also benefit from a detailed analysis of the multispectral aerosol optical properties for the different aerosol types and aerosol aging. Ground-based lidar observations including multispectral observations are complementary of satellite observations in the UV (Earth-CARE, ADM) or in the visible and the IR (CALIOP).

CALIOP aerosol products are elaborated after the separation of aerosol and clouds and identification of their type. The quality of the data depends on these two steps. In CALIOP algorithms, it may be difficult to identify aerosols of low optical depth or when clouds are embedded in aerosol layers. The retrieval of extinction relies also on aerosol type identification in order to provide a lidar ratio. Therefore, new algorithms are being developed using IIR information and lidar surface reflectance analysis (T1.1 and T3.1). Aerosol optical depths retrieved with these algorithms are compared with ground-based and airborne observations during various field campaigns in Europe and Arctic (T3.1), and against in situ and ground-based lidar observations at Puy de Dome (T3.2). Regarding aerosol identification, depolarization is a key parameter. Accurate laboratory measurements of the particle depolarization ratio of mineral dust are developed for all wavelengths (UV, VIS, IR), using a time-resolved polarimeter at exact backscattering (T3.4). Such laboratory measurements may help to improve CALIOP products and future space-based lidar instruments.

Recent events have led to place a particular emphasis on the analysis of emissions from large fires and volcanic eruptions. In the troposphere, these emissions are an extreme source of pollution still subject to strong uncertainties both on the number of aerosols emitted, their injection heights, transport/mixing processes and interactions with clouds (T3.1). Volcanic eruptions, like the intense fires from Australia and California in 2019, can lead to injections into the stratosphere, where they can self-organize spatially in localized vortices, modify the long-term stratospheric composition, and influence the radiative transfer. These impacts, whose importance might rise with climate change, are the object of the new T3.5 devoted to stratospheric aerosols.

Finally, modelling is necessary to understand the processes leading to the observed features, and perform impact analyses. It is the goal of task T3.3, involving the development of methodologies and software, to:

• Maximize the information provided by the observations on the regions analyzed but also accounting for instrumental limitations (signal/noise in particular)
• Simulate observations from model outputs to allow quantitative comparisons

Observations are analyzed using transport models for source attributions and chemistry transport models for more precise analysis of composition and properties of the transported layers (MOCAGE and CHIMERE). The analysis of the role of transport emission during the lockdown period will benefit from model simulations of the 2020 spring period.