Multi-term LSM as methodological platform for advanced multi-sensor remote sensor synergy

The presentation discusses the Multi-term Least Square Method (LSM) as a methodological platform for realizing of the multi-instrument synergy. As discussed  by Dubovik et al. (2021), the Multi-term LSM has been used to develop complex inversion algorithms for a number of years and have been successfully applied to aerosol retrievals from diverse  satellite, ground-based and laboratory measurements. Theoretically, the approach unites the advantages of a variety of approaches and to provide transparency and flexibility in development of efficient retrievals. It provides a methodology for using multiple a priori constraints to atmospheric problems. One of the most important  practical features of the approach is that it allows for synergy processing of observations that are not fully coincident nor fully co-located. Specifically, synergy of such observation can be realized following the multi-pixel approach (Dubovik et al., 2011), when the large groups of satellite observations (pixels) are inverted simultaneously. By processing observations from multiple pixels together, the retrieval efficiently incorporates prior knowledge about the temporal and spatial variability of the retrieved parameters.

Indeed, while the  synergy of  not coincident or not co-located observation is less intuitive, it is very promising.  Whereas the fusion of co-incident multi-angular polarimeter and lidar observations is considered as efficient approach, in practice the coincidence of such observations can be limited. For example, the trajectories of currently operating EarthCARE and PACE have very limited overlaps, therefore the possible synergy product of these two satellites can only be very sparse. In contrast, the synergy of not fully coincident or co-located observations can be applied always for combining any observations from operating satellites with different trajectories. Based on our current experience such synergy, realized using multi-pixel approach, allows for substantial improvement of aerosol characterization due to two phenomena: (i) propagation of superior information about aerosol details from more sensitive observations to less sensitive, and (ii) overall increase of observations volume of the same aerosol event in different times and locations. The benefits of such synergy of non-coincident observations have been demonstrated in the framework of ESA SYREMIS project (https://www.grasp-earth.com/portfolio/syremis/), where the synergetic multi-instrument retrieval approach was developed for characterizing aerosol and surface properties using different combinations of S-3A, S-3B, S-5p, polar and HIMAWARI  geo observations. It was shown that realized methodology helped the information  from polar satellite to propagate geo retrieval and made possible the retrieval of AE and SSA for the pixel with HIMAWARI observations reasonable accuracy, while processing these observations separately does not provide these parameters. In these regards, the combined processing of PACE, EarthCARE and HIMAWARI could also be used to provide enhanced aerosol global product.

Dubovik, O., M. Herman, A. Holdak, et al., “Statistically optimized inversion algorithm for enhanced retrieval of aerosol properties from spectral multi-angle polarimetric satellite observations”, Atmos. Meas. Tech., 4, 975-1018, https://doi.org/10.5194/amt-4-975-201, 2011.

Dubovik, O., D. Fuertes, P. Litvinov, et al. , “A Comprehensive Description of Multi-Term LSM for Applying Multiple a Priori Constraints in Problems of Atmospheric Remote Sensing: GRASP Algorithm, Concept, and Ap-plications”, Front. Remote Sens. 2:706851. doi: 10.3389/frsen.2021.706851, 2021