Vibrational Raman scattering (VRS) in the water is an inelastic scattering process that causes a spectral redistribution of the radiation in the ocean. It should be considered for correctly determining the underwater light field and could be exploited for various applications. The VRS signal in hyperspectral satellite radiances has been investigated in the UV and used as a proxy for the effective in-water light path and for total chlorophyll-a estimations (Vountas et al. 2003, Joiner et al. 2006). For determining the light availability directly in the visible range and retrieving a spectrally derived PAR (photosynthetically available radiation) product, the VRS signature in the visible channel from hyperspectral radiances measured by the Scanning Imaging Absorption Spectrometer for Atmospheric Chartography (SCIAMACHY) has been exploited (Dinter et al. 2015). Since SCIAMACHY stopped measuring in 2012, newer hyperspectral sensors such as the recently launched TROPOspheric Monitoring Instrument (TROPOMI) on the Sentinel-5-precurser satellite could be used for light availability investigations from 2012 onwards. Here, we are presenting the VRS retrieval on top of atmosphere radiances measured by three different hyperspectral satellite sensors in comparison, SCIAMACHY, GOME-2 (Global Ozone Monitoring Experiment-2), and OMI (Ozone Monitoring Instrument). GOME-2 and OMI can potentially add additional data to the already existing SCIAMACHY time series and prolong the time series from 2012 onwards. These newer sensors provide earth observations with increased temporal coverage and, in case of OMI and TROPOMI, greatly increased spatial resolution compared to the SCIAMACHY data series. As has been done for SCIAMACHY, diffuse attenuation coefficients in the wavelength region 390 to 426 nm are derived from the VRS signal using a look-up-table built up through radiative transfer modeling. Results are compared to a wavelength-converted OC-CCI (Ocean Colour - Climate Change Initiative) product of the diffuse attenuation coefficient at 490 nm. Temporal and regional deviations between the satellite products are addressed on global scale and in selected Longhurst provinces.

Area D - Board 357

Monte-Carlo radiative transfer solvers are a powerful tool by their flexibility and ability to take into account virtually any physical process, but their applicability is generally limited by their demands in terms of processing power. The SMART-G (Speed-Up Monte-Carlo Advanced Radiative Transfer code with GPU) uses GPGPU (General-purpose processing on graphics processing units) technology through the CUDA framework to implement a fast code that simulates the propagation of light in the atmosphere, through the wavy sea surface or land surfaces and in the ocean. This code accounts for polarization and works in either plane-parallel or spherical shell geometry. For typical simulations, an acceleration factor of several hundreds is obtained compared with CPU calculation. Advanced variance reduction techniques have been implemented. We present this code, its domain of application, cross-comparison with other reference radiative transfer simulations. We describe also the way SMART-G handles spectral computations (band models, k-distribution or high spectral resolution absorption) including inelastic scattering and fluorescence, the way we produce also Jacobians of outgoing radiation relative to atmospheric, oceanic, and surface parameters. Some results concerning the evolution toward 3D are given through the example of the introduction of 3D geometrical objects within the scene, like mirrors and buildings of a large Solar Concentration Plant. We then show examples of fast sensitivity studies like the effect of the atmosphere sphericity on the Rayleigh polarized reflectances that are used for satellite in-flight radiometric calibration activities and the effect of the horizontal heterogeneities of the surface albedo, important for high spatial resolution sensors like Sentinel 2/MSI. Some applications to the simulation and sensitivity of various Sentinels and future sensors are shown.

Area D - Board 358

The RAdiative transfer Model Intercomparison exercise (RAMI) was launched 20 years ago by the community that develops canopy radiative transfer (RT) models (Pinty et. al., 2001). RAMI has served as a collective platform for the evaluation of models simulating bidirectional reflectance factors (BRFs) as well as radiative fluxes in 1-D and 3-D vegetation canopies. Starting with relatively simple abstracted plant environments in 1999 to more complexity realistic representations of plant environments in 2015, the four first phases of RAMI were used by participants to identify coding bugs and to improve some of the RT formulations in their models. This has enabled to define surrogate reference results and the development of a web-based benchmarking facility known as the RAMI On-line Model Checker (ROMC). Since 2006, more than 180 scientists have used the ROMC in order to debug or/and validate 74 models. This presentation will highlight the accomplishment of RAMI and will present the next challenge for RAMI-V phase.

Area D - Board 359

Earth Observation land products require a continuous validation activity which is generally based on the comparison with reference ground-based observations. Despite the increasing efforts of the monitoring networks to adopt up-to-date observation strategies, a certain degree of uncertainty affects in-situ data as well as the space retrievals. The total uncertainties depend of sensor calibration and design, sampling strategies, up-scaling approaches, and retrieval algorithm assumptions. Therefore, a model based approach for validation, using radiative transfer physics, is proposed to assess the uncertainties budget.

This work presents the activities carried at the Joint Research Centre of the European Commission (JRC) for the evaluation of the land products such as the Fraction of Absorbed Photosynthetically Active Radiation (FAPAR) and Leaf Area index (LAI) over a number of virtual sites with a selection of the most representative vegetation covers, including tropical forests, savanna, shrubs, boreal birch and pine forests, crops.
The validation strategy is applied to different ground-based sensors and satellite products.

Area D - Board 360

Conventional methods for estimating environmental variables are usually parameter-specific and require the input of many surface and atmospheric parameters product. Also the cloudy observations and surface parameters under cloudy-sky conditions are usually not considered. New generation of geostationary satellite sensors have the traditional advantage of high temporal resolution, as well as high spectral and spatial resolution characteristics. In this study, the geostationary satellite Himawari 8/9 AHI TOA observations were explored for estimating a suit of physical-consistent land surface, atmosphere and TOA parameters, based on optimization and a coupled radiation transfer model, which was established under both clear- and cloudy-sky conditions, by coupling the soil, snow, canopy and atmosphere model within the range of 0.24 μm to 14 μm.

Fig. 1 illustrates the overall structure of the algorithm to retrieve multiple land surface, atmosphere, and TOA parameters from AHI TOA observations. This algorithm mainly consists of three steps. The first is cloud detection, 8-day time series clearest TOA reflectances were extracted according to minimum blue band reflectance, then the 8-day surface BRDF information can be retrieved from the clearest observations based on the coupled land-atmosphere RT model and optimization estimation, together with a pre-calculated atmosphere LUT, the observations accumulated in a day can then be classified as clear-sky or cloudy-sky observation. Secondly, the clear-sky TOA reflectances each day were employed to invert the most sensitive set of land and atmosphere parameters using optimization algorithm, and the cloudy-sky atmosphere parameters (cloud optical thickness (COT) and cloud effective radius (CER)) were inverted assuming land surface conditions remain unchanged as under the clear-sky conditions. Finally, the inverted atmosphere and land surface parameters were input into the coupled model to simulate the remaining set of surface parameter (ISR, PAR, FAPAR, surface albedo, reflectance and emissivity) and TOA parameters (TOA albedo).

Current available satellite product (MODIS, GLASS and CERES) and field measurement data from Ozflux network sites were collected to compare with and validate the retrieved parameters. The retrieved vegetation parameters LAI and FAPAR show high agreement with GLASS LAI/FAPAR product (with R2 of both 0.92, RMSE of 0.61 and 0.08 for LAI and FAPAR respectively), direct validation of ISR and surface albedo indicates that the retrieved ISR can achieve higher accuracy than CERES ISR product (with R2 of 0.85 and 0.80, RMSE of 106.2 and 118.8 W/m2, respectively), and comparable accuracy with MODIS daily albedo product (both RMSE are 0.03). Besides, the retrieved TOA albedo correlates better with CERES than the AVHRR TOA albedo product (with R2 of 0.73 and 0.54, RMSE of 0.07 and 0.1, respectively). The validation results indicates that, this unified inversion algorithm, which is based on the coupled RT model, free from atmospheric correction and does not rely on higher-level remote sensing data, can produce a suit of remote sensing products from the geostationary satellite TOA observations simultaneously, with physical and temporal consistency and acceptable accuracies.

Fig. 1 Land surface, atmosphere, and TOA parameters retrieval algorithm from AHI TOA observations

Area D - Board 361

The focus on the quality of satellite-derived biophysical products has increased in recent years as new perspectives on the level of uncertainty required for specific applications are being defined through the Global Climate Observing System (GCOS). The derived nature of these products makes them difficult to assess fairly and as such there are numerous published studies which show discrepancies between satellite-derived products, while comparison with in situ reference data is limited due to a lack of adequate information and discrepancies between the measurands (e.g. in the case of fAPAR).
The absence of controlled conditions in the field to test the sources of the differences between different products and ground data has lead researchers to pursue avenues where conditions can be controlled. The use of radiative transfer models with an explicit 3D description of plant architecture and spectral characteristics is one of these. The majority of work in this area has focused on realistic-looking scenes, although some attempts have been made to recreate real sites (e.g. using allometry). Ground-based terrestrial laser scanning (TLS) from multiple points provides the means of generating an accurate 3D description of the surrounding area as a point cloud. A secondary step to convert the point cloud into scene objects is required to provide input to the radiative transfer (RT) models. Scenes generated in this way can be more consistent (between different sites) and capture a greater level of complexity in the canopy than is possible using other techniques. In theory, they can also be linked to site measurements providing sufficient ancillary information is known. It is this final point that is arguably the most important since this can provide confidence in extending/applying conclusions drawn from the simulations to the field.
In the context of satellite-derived fAPAR validation, assessment of modelling assumptions from satellite, and measurement assumptions from in situ sensors are needed. This can be achieved by comparing the various configurations with the assumptions/conditions experienced by the others. Creating tables of conversion factors between these scenarios would facilitate fair comparisons between satellite and in situ fAPAR at the designated reference site.
In this presentation, we discuss the use of a large-scale (> 1 ha) realistic 3D model of Wytham Woods, Oxford derived from terrestrial LiDAR scans to compare different fAPAR quantities based on simulations using the model. Our current focus is on the general assumptions made by satellite products as well as other in situ methods (e.g. hemispherical photography), e.g. green vs. total. However, future activities will focus on specific satellite fAPAR products such that validation data will be created directly relating to those. We discuss future improvements to the model and validation methodology, as well as our progress towards traceable fAPAR validation.

Area D - Board 362

In this work, a radiative transfer model, so-called OSOAA, acronym for “Ocean Successive Orders with Atmosphere – Advanced”, which has been developed for the purpose of ocean color remote sensing studies, is presented. OSOAA is a radiative transfer code, based on the successive orders of scattering method [Deuzé et al, 1989] [Lenoble et al., 2007], which simulates the propagation of the polarized solar light in the coupled system « ocean / atmosphere » from 400 nm to 900 nm [Chami et al., 2015]. The radiance and degree of polarization are provided at various viewing angles and at any level from the top of atmosphere to the ocean seabed.
The scattering of light in the water column accounts for interactions of photons with molecules and hydrosols, namely phytoplankton and mineral-like particles, thus providing a complete description of the directional effects of hydrosols on the water leaving radiances. The absorption by colored dissolved organic matter (CDOM) and detritus is taken into account as well. In addition, OSOAA simulates a realistic air / sea interface by considering the roughness of the sea surface based on the wind speed and wave statistics as modelled by Cox & Munk [1954]. This latter point is a major improvement of OSOAA model relatively to its previous version, so-called OSOA model, which was developed considering a flat sea surface [Chami et al., 2001]. Another main improvement brought by OSOAA is to offer a user friendly interface (GUI) and a user manual that describes all the physical models that are implemented in the code as well as how to run the OSOAA model for simulations.
Another originality of the OSOAA model relies on its ability to simulate various realistic atmospheric conditions such as pressure, aerosol optical thickness and, more importantly, aerosol models taken either from the literature (e.g., WMO or Shettle & Fenn [1979] models), or from user-defined conditions using mono-modal or bi-modal size distributions. Note, however, that the current version of OSOAA (v1.5) does not take into account the atmospheric gaseous transmittance.
The validation of OSOAA model has been carried out through comparisons with another operational radiative transfer model [Chowdhary et al., 2006]. Satisfactory agreements for the intensity within 0.5% at sea surface and 0.2% at top of the atmosphere in the blue spectral domain, and within 0.8% for the polarization Stokes parameters, were found. The scientific community working on coastal waters remote sensing or open ocean applications can take benefit of the exact modeling proposed by the OSOAA model, for example to better understand marine ecosystems, or to develop and improve inverse ocean color algorithms.
Note that OSOAA model is publicly available for the scientific or industrial community with a distribution ensured by CNES (https://github.com/CNES/RadiativeTransferCode-OSOAA).

Area D - Board 363

The analysis of shallow waters seabed using remote sensing data is of great interest for many environmental applications in coastal areas such as benthic habitat mapping, monitoring of seabed aquatic plants and the subsequent management of littoral zones. The inversion of water column by radiative transfer model is performed to estimate the seabed, so the need of model refinements to improve the estimation quality appeared these last years. In this work, we consider the adjacency effects caused by the neighborhood of a given target in shallow waters. They are quantified for three water turbidity conditions, namely clear, moderately turbid and turbid waters and using measured bottom reflectance spectra that are representative of natural waters.
Firstly, an original analytical formulation of the upward subsurface radiance, based on radiative transfer theory and modelling, is proposed to make explicit the radiative terms which contribute to the adjacency effects induced by the light reflection on the bottom and by the light scattering within the water column itself. Such a formulation could thus be used either to better understand and quantify the role of the adjacency effects in shallow waters or to develop inverse remote sensing algorithms for retrieving bottom features such as its composition or its bathymetry.
Secondly, a sensitivity study is performed using radiative transfer modeling to quantify the influence of the adjacency effects on the upward subsurface radiance with respect to various parameters such as the bathymetry or the bottom brightness. In shallow waters (5 m depth), for observations at a high spatial resolution (typically, for a pixel size lower than 0.5 m), the results show that the highest contributions to the upward subsurface radiance of the adjacency effects induced by the neighborhood of a bottom target could reach up to 35%, 26% and 12% respectively for clear, moderately turbid and turbid water conditions. This means that the adjacency effects can significantly bias the detection of a target from highly resolved remote sensing measurements if they are ignored in the inversion process. The analytical formulation of the subsurface radiance proposed in this study could be further used to improve the performance of inverse algorithms dedicated to the retrieval of bottom composition, water optical properties and/or bathymetry from satellite sensors.

Area D - Board 364

The Successive Orders of Scattering (SOS) code is a radiative transfer code which simulates the scattering and polarization of the solar light in the atmosphere, under clear sky conditions, including the surface reflections. The code was improved to take account gaseous absorptions, leading to a new version, so called SOS-ABS.
Based on the Successive Orders of Scattering method [Deuzé et al, 1989] [Lenoble et al., 2007], the SOS-ABS code provides the polarized radiance field from the surface to the top of atmosphere, for the reflective part of the solar spectrum (from 0.2 to 4.0 microns). Its large variety of aerosol models offers the possibility to simulate aerosol phase functions either from mono-modal or bi-modal size distributions [Dubovik et al, 2002], from aerosol models taken from the literature (e.g., WMO or Shettle & Fenn [1979] models), or from a user-definition (which allows the simulation of non-spherical particles). The surface reflection is simulated over lands by a lambertian model or by bidirectional reflectance models such as the Roujean model [Roujean et al, 2002], the Rondeaux polarized model [Rondeaux et al, 1991] over vegetations, the Bréon polarized model [Bréon et al, 1995] for bare soils and vegetation. Over sea, the sun-glint can be simulated by considering the roughness of the sea surface based on wave statistics depending on the wind speed as modelled by Cox & Munk [1954]. The gaseous absorption takes into account the absorption lines for water vapour (H2O), ozone (O3), carbon dioxide (CO2), oxygen (O2,), carbon monoxide (CO), methane (CH4) and dinitrogen monoxide (N2O). The continuum absorptions are also introduced for H2O, CO2, O3, N2, O2, NO2 and SO2. The amount of water vapour and ozone can be defined by the user to weight the atmospheric profile of species (predefined profiles or user profile).
The light propagation through the atmosphere is described by the Equation of Radiative Transfer (ETR). At each level of the atmospheric profile, the code starts by computing the radiance field caused by the single-scattering of the direct solar beam. From these first radiation fields, the code derives the global field of single-scattering: by integrating the ETR for the upward field (including the surface reflection of the direct solar beam) or for the downward field, the SOS code computes iteratively (layer per layer) the single scattering radiance emerging at TOA or illuminating the ground. The polarized radiance is computed as well.
This radiance field of order 1 induces the 2nd surface reflection and the source function of the 2nd scattering at each level of the atmospheric profile. The global radiation field of order 2 is then obtained by the same way as for the single-scattering. By computing iteratively the successive terms of interaction and by adding them, the code obtains the global field of polarized radiance for each layer.
The modelling of gaseous transmissions is based on the correlated K distribution (CKD) method [Lacis et al, 1991], associated to a spectral database for each gas and to predefined atmospheric profiles. For each gas, the transmission is then calculated by a summation of exponential terms weighted by coefficients. A calculation based on the successive orders of scattering method is performed for each component of the exponential expansion. Finally, the code agglomerates elementary calculated radiances, for all the expansion terms and for all gas species, to obtain the full polarized radiance field.
The SOS-ABS has been validated by comparison to simulations performed by radiative transfer codes from ARTDECO (Atmospheric Radiative Transfer Database for Earth and Climate Observation) [Compiègne et al., 2013] with a satisfactory agreement.

Area D - Board 365

The a priori knowledge of the Earth surface reflectance is required for the operational exploitation of space-borne missions (both active and passive) measuring the electromagnetic radiation reflected by the surface in the solar domain, as well as for the preparation of future missions and processing chains. In this prospect, a monthly climatology, ADAM (A Surface Reflectance Database for ESA’s Earth Observation Missions), has been designed in order to simulate realistic spectro-directional variations of the Earth surface reflectance at the global scale with a 0.1°x0.1° resolution. A calculation toolkit associated to the ADAM database allows simulating the spectral BRDF over the 240-4000 nm spectral domain (with a spectral resolution of 1 nm) and in any observation geometry. ADAM is freely available online at http://adam.noveltis.fr.

For land surfaces (soil/vegetation and snow), the ADAM database is made of a climatology of normalized reflectances (corrected from the directional effects) in seven wavebands, processed by the FondsDeSol processing chain from MODIS MOD09 products (representative of year 2005). For ocean surfaces, the database consists in mean climatologies of ocean colour (chlorophyll content derived from SeaWiFs) and wind speed (from QuickScat) over the 1999-2009 time period.
For the soil/vegetation pixels, the spectral interpolation/extrapolation of the MODIS normalized reflectances between 240 and 4000 nm is performed using EOF (Empirical Orthogonal Functions) derived from spectral reflectance databases of soil/vegetation/leaf optical properties. The Ross-Li-HS kernel based BRDF model is then used to calculate the reflectance spectrum in any observation geometry. A specific processing is applied for snow covered surfaces: It relies on the model of Kokhanovsky, fitted to the MODIS normalized reflectances, to simulate the spectro-directional variations of snow reflectance. Moreover, it is possible to calculate the uncertainty attached to land surface reflectance: the calculation relies on the variance covariance matrix of the reflectance values between the seven MODIS bands for each 0.1°x0.1° pixel. Over ocean, the surface reflectance is generated by the combination of three components: i) the water column that mainly shapes the spectral variation of reflectance depending on the chlorophyll content, ii) the specular component and iii) foam that essentially drive the directional variation of ocean reflectance depending on the wind speed.

Recent updates of the ADAM database and calculation tools have been performed. They consist essentially in an improved gap filling over sea ice, oceans and lakes and over snow covered areas in winter for polar regions. We will present the evaluation of ADAM products over land (vegetation, soil and snow, surfaces) relative to two independent remotely sensed datasets: 1) PARASOL multi-spectral/multi-directional observations and 2) LIDAR observations performed by CALIPSO at 532 nm.

Area D- Board 366

Radiative transfer model simulations are an important component of atmospheric correction and trace gas retrieval algorithms. During the last few years at the German Aerospace Center (DLR) we developed a software package PYDOME for the simulation of satellite-based measurements of reflected and scattered solar radiation in the ultraviolet and visible spectral ranges. The tool is based on a one-dimensional scalar radiative transfer model, which incorporates the exact solver, namely, the discrete ordinate method with matrix exponential (DOME) (Doicu&Trautmann, 2009, Efremenko et al. 2017), as well as several approximate solvers (the single scattering solver, the two-stream solver and others). The model is implemented for coupled atmosphere-ocean systems. It takes into account the change in the refraction index across the interface between two media and the wind-induced roughness of the water surface described by the Cox-Munk model. In addition, PYDOME can also treat a specular reflection from the surface (a “mirror” case).
PYDOME is equipped with tools for modeling the atmospheric state. The interfaces to OPAC database (Hess et al, 1998) of aerosol models and molecular spectroscopic database HITRAN are provided. The computations of optical properties of the atmosphere are implemented via object-oriented programming paradigm. The atmospheric state can be constructed in an interactive mode by using the corresponding methods. For instance, the class Map contains a table of scattering, extinction and absorption coefficients together with expansion coefficients of phase functions for each atmospheric level. Molecular scattering, molecular absorption, aerosol and cloud models provide maps, which are then mixed into a new map corresponding to all atmospheric constituents. The optical parameters are derived from the resulting map.
Besides the radiance field, the PYDOME tool provides the additional quantities, such as the path radiance, the spherical albedo, the direct transmittance, the diffuse transmittance, as well as the irradiance at the given altitude. These parameters can be stored in the look-up tables (either in ASCII or hdf5 formats) and then used in the atmospheric correction algorithms. PYDOME is implemented in python3 and relies on the NumPy library. The code is vectorized by means of the matrix algebra. Thus, most of the time consuming for-loops are eliminated. Consequently, the performance of PYDOME is comparable to the models implemented in FORTRAN or C. For the retrieval of atmospheric properties from remote sensing measurements by using the Gauss-Newton algorithm, PYDOME can be used for computing the Jacobian matrix comprising the partial derivatives of the radiances with respect to atmospheric parameters. The code is differentiated automatically by overloading NumPy functions.
In the talk, we summarize the implementation issues and present comparisons of PYDOME results in several spectral ranges with those of the radiative transfer codes used worldwide by the scientific community (DISORT, LIDORT, LibRadTran, MODTRAN).

References

Doicu A., Trautmann T. Discrete ordinate method with matrix exponential for a pseudo-spherical atmosphere: Scalar case. Journal of quantitative spectroscopy and radiative transfer. 110(1), 146, 2009

Efremenko D.S., Molina Garcia V., Gimeno Garcia S., Doicu A. A review of the matrix-exponential formalism in radiative transfer. Journal of quantitative spectroscopy and radiative transfer. 196, 17, 2017.

Hess M., Koepke P., Schult I. Optical Properties of Aerosols and Clouds: The Software Package OPAC. Bulletin of the American Meteorological Society. 76(5), 831, 1998.

Area D - Board 367

The Sentinel-2 mission is fully operating since June 2017 with a constellation of two polar orbiting satellite units. Both Sentinel-2A and Sentinel-2B are equipped with an optical imaging sensor MSI (Multi-Spectral Instrument) which acquires high spatial resolution optical data products. The Sentinel-2 mission is dedicated to land monitoring, emergency management and security. It serves for monitoring of land-cover change and biophysical variables related to agriculture and forestry, monitors coastal and inland waters and is useful for risk and disaster mapping.
Accurate atmospheric correction of satellite observations is a precondition for the development and delivery of high quality applications. Therefore the atmospheric correction processor Sen2Cor was developed on behalf of ESA. Sen2Cor is designed for mono-temporal processing of Sentinel-2 L1C data products providing Level-2A surface (Bottom-of-Atmosphere) reflectance product together with Aerosol Optical Thickness (AOT), Water Vapor (WV) and Scene Classification (SCL) maps. Sen2Cor can be downloaded from ESA website for individual processing by the users. In parallel, Sen2Cor is used by ESA for systematic L2A-processing of Sentinel-2 acquisitions over Europe since June 2017. ESA is preparing an operational global L2A-processing with the integration of Sen2Cor in Sentinel-2 ground segment (PDGS).
A key-parameter for accurate atmospheric correction is the knowledge of the spatially and temporally very variable aerosol content of the atmosphere. Sen2Cor estimates the aerosol optical thickness (AOT) at 550 nm using dark reference areas, most preferably dense dark vegetation (DDV) pixels. In case there are not enough DDV-pixels available in the granule, then a fallback solution is required. The former fallback solution was to generate an AOT map based on the start visibility set in the configuration file. This fallback solution is one of the limitations of Sen2Cor aerosol retrieval, because it can introduce large errors into the atmospheric correction.
An evolution of Sen2cor atmospheric module is foreseen for introducing the use of the data from the Copernicus Atmosphere Monitoring Service (CAMS) of ECMWF. CAMS delivers several operational services, including daily production of near-real-time analyses and forecasts of global atmospheric composition.
Since February 2017, a prototype which implements the direct use of the CAMS AOT at 550nm for the fallback solution has been developed and is being validated over a set of sites. The poster will remind the method of integration and assimilation, and present the current results of the validation. In addition the progress on an enhanced version of the prototype will be reported, which aims to introduce aerosol components in Sen2Cor in place of fixed aerosol types, and deduce their mixing from CAMS data. In a last part will be discussed the integration of such a prototype into an operational environment.

Area D - Board 368