Comment on amt-2022-104

Abstract

The understanding of the uncertainties in the retrieval of the aerosol and surface properties is very important for an adequate characterization of the processes that occur in the atmosphere. However, the reliable characterization of the error budget of the retrieval products is a very challenging aspect that currently remains not fully resolved in most remote sensing approaches. The level of uncertainties for the majority of the remote sensing products relies mostly on post-processing validations and inter comparisons with other data while the dynamic errors are rarely provided. Therefore, implementations of fundamental approaches for generating dynamic retrieval errors and the evaluation of their practical efficiency remains of high importance. This study describes and analyses the dynamic estimates of uncertainties of aerosol retrieved properties by GRASP (Generalized Retrieval of Atmosphere and Surface Properties) algorithm. GRASP inversion algorithm described by Dubovik et al. (2011, 2014, 2021) is designed on the concept of statistical optimization and provides dynamic error estimates for all retrieved aerosol and surface properties. The approach takes into account the effect of both random and systematic uncertainties propagations. The algorithm provides error estimates both for directly retrieved parameters included in the retrieval state vector and for the characteristics derived from these parameters. For example, in the case of the aerosol properties, GRASP retrieves directly the size distribution and the refractive index that are used afterwards to provide phase function, scattering, extinction, single scattering albedo, etc. Moreover, GRASP algorithm provides full covariance matrices, i.e. not only variances of the retrieval errors and also correlations coefficients of these errors. The analysis of correlation matrix structure can be very useful for identifying unobvious retrieval tendencies that appear to be a useful approach for optimizing observation schemes and retrieval setups. In this study we analyse the efficiency of GRASP error estimation approach for applications to ground-based observations by sun/sky photometer and lidar. Specifically, diverse aspects of the error generations and their evaluations are discussed and illustrated. The studies rely on the series of comprehensive sensitivity tests when simulated sun/sky photometer measurements and lidar data are perturbed by random and systematic errors and inverted. Then, the results of the retrievals and their error estimations are analyzed and evaluated. The tests are conducted for different observations of diverse aerosol types including biomass burning, urban, dust and their mixtures. The study considers observations of AERONET sun/sky photometer measurements at 440, 675, 870 and 1020 nm and multi-wavelength elastic lidar measurements at 355, 532 and 1064 nm. The sun/sky photometer data are inverted alone or together with lidar data. The analysis shows overall successful error retrievals for different aerosol characteristics including aerosol size distribution, complex refractive index, single scattering albedo, lidar ratios, aerosol vertical profiles, etc. Also, the main observed tendencies in the error dynamic agree with known retrieval experience. For example, the main accuracy limitations for retrievals of all aerosol types relate to the situations with low optical depth. Also, in situations with multi-component aerosol mixtures, the reliable characterization of each component is possible only in limited situations, for example from radiometric data obtained for low solar zenith angle observations or from a combination of radiometric and lidar data. At the same time, total optical properties of aerosol mixtures are always retrieved satisfactorily. In addition, the study includes the analysis of the detailed structure of correlation matrices for the retrieval errors of mono- and multi- component aerosols. The conducted analysis of error correlation appears to be a useful approach for optimizing observations schemes and retrieval setups. The application of the approach to real data is provided.