Other algorithm studies
Read our other algorithm studies for current, future and multi-missions.
27 January 2023
25 May 2018
The 3MI design and heritage comes from the POLDER/PARASOL missions. A number of mature algorithms exist in the community for retrieval of aerosol optical properties from POLDER/PARASOL, which could, in principle, be applied to 3MI data. However, 3MI presents a particular challenge due to the need to provide near real-time data products (~120 minutes after sensing) combined with an increase in spatial resolution which increases the data rate and, therefore, computational cost of product generation. Plus, 3MI has additional spectral channels (specifically in the ultraviolet and the short-wave infrared) and more polarised channels than its predecessors. 3MI also has stringent requirements on its final product quality.
The most sophisticated POLDER/PARASOL algorithms, which would be capable of producing high quality products, and which make use of not only co-registered multi-spectral, multi-angular, multi-polarisation data, but also spatial and temporal accumulations, are typically too computationally expensive to be appropriate for Near Real-Time (NRT) operational product production. Conversely, look-up-table methods, which are sufficiently fast for NRT product production, are unlikely to reach the accuracy required for 3MI data products.
The aim of this study was to provide support to the development of aerosol optical and microphysical properties retrieval algorithms from the 3MI instrument on EUMETSAT Polar System – Second Generation (EPS-SG).
The main objective of this study was to provide support for the definition of an algorithm baseline for 3MI, which will be sufficiently fast to be able to be run in near real-time, and sufficiently accurate to satisfy the product quality requirements. A secondary objective was to evaluate the impact of the choice of Level 1b to 1c methodology on the final Level 2 products, by comparing results generated from the EUMETSAT 3MI Level 1c product and an independent 3MI level 1c product, provided by the contractor.
This study provided support for the definition of an algorithm baseline for 3MI, which will be sufficiently fast to be able to be run in near real-time and sufficiently accurate to satisfy the product quality requirements. The support comprised provision of a review of aerosol models to be used in the retrieval, provision of proxy test data for algorithm evaluation, and provision of a test environment, capable of processing multiple orbits of 3MI test data in a reasonable timeframe, for evaluation of algorithm performance.
The contractor had the capability for Level 1b to Level 1c data processing and a state-of-the art aerosol microphysical and optical properties retrieval algorithm (supported by peer-reviewed publications) to serve as a reference processor for algorithm testing and performance evaluation (both in terms of product accuracy and processor speed). The contractor was required to base its aerosol optical and microphysical properties retrieval on radiances using multi-angle acquisitions, up to 14 acquisitions, 'views' (see Figure 1). These were co-registered to a regular, equal area, fixed geo-physical grid with configurable grid spacing, close to the instrument footprint spatial resolution (Level 1c data); were parallax corrected using a digital elevation model of sufficient resolution, and covered the full instrument swath. The co-registration was done separately for SWIR and VNIR channels with a maximum of 14 acquisitions for both.
Since the along-track dimension of the SWIR detector is half the size of the VNIR its viewing angle range along-track is restricted to half of the VNIR detector and, therefore, the detectors share only seven out of 14 viewing angles which are identical to the VNIR angles. Specific METimage cloud products, namely the cloud mask and cloud top height information, were available as an input to the 3MI NRT aerosol product processing chain.