A multiwavelength numerical model in support of quantitative retrievals of aerosol properties from automated lidar ceilometers and test applications for AOT and PM10 estimation.

The use of automated lidar ceilometer (ALC) systems for the aerosol vertically resolved characterization has increased in recent years thanks to their low construction and operation costs and their capability of providing continuous unattended measurements. At the same time there is a need to convert the ALC signals into usable geophysical quantities. In fact, the quantitative assessment of the aerosol properties from ALC measurements and the relevant assimilation in meteorological forecast models is amongst the main objectives of the EU COST Action TOPROF ("Towards operational ground-based profiling with ALCs, Doppler lidars and microwave radiometers for improving weather forecasts"). Concurrently, the E-PROFILE program of the European Meteorological Services Network (EUMETNET) focuses on the harmonization of ALC measurements and data provision across Europe. Within these frameworks, we implemented a model-assisted methodology to retrieve key aerosol properties (extinction coefficient, surface area, and volume) from elastic lidar and/or ALC measurements. The method is based on results from a large set of aerosol scattering simulations (Mie theory) performed at UV, visible, and near-IR wavelengths using a Monte Carlo approach to select the input aerosol microphysical properties. An average "continental aerosol type" (i.e., clean to moderately polluted continental aerosol conditions) is addressed in this study. Based on the simulation results, we derive mean functional relationships linking the aerosol backscatter coefficients to the abovementioned variables. Applied in the data inversion of single-wavelength lidars and/or ALCs, these relationships allow quantitative determination of the vertically resolved aerosol backscatter, extinction, volume, and surface area and, in turn, of the extinction-to-backscatter ratios (i.e., the lidar ratios, LRs) and extinction-to-volume conversion factor (cv) at 355, 532, and 1064 nm. These variables provide valuable information for visibility, radiative transfer, and air quality applications. This study also includes (1) validation of the model simulations with real measurements and (2) test applications of the proposed model-based ALC inversion methodology. In particular, our model simulations were compared to backscatter and extinction coefficients independently retrieved by Raman lidar systems operating at different continental sites within the European Aerosol Research Lidar Network (EARLINET). This comparison shows good model–measurement agreement, with LR discrepancies below 20 %. The model-assisted quantitative retrieval of both aerosol extinction and volume was then tested using raw data from three different ALCs systems (CHM 15k Nimbus), operating within the Italian Automated LIdar-CEilometer network (ALICEnet). For this purpose, a 1-year record of the ALC-derived aerosol optical thickness (AOT) at each site was compared to direct AOT measurements performed by colocated sun–sky photometers. This comparison shows an overall AOT agreement within 30 % at all sites. At one site, the model-assisted ALC estimation of the aerosol volume and mass (i.e., PM10) in the lowermost levels was compared to values measured at the surface level by colocated in situ instrumentation. Within this exercise, the ALC-derived daily-mean mass concentration was found to reproduce the corresponding (EU regulated) PM10 values measured by the local air quality agency well in terms of both temporal variability and absolute values. Although limited in space and time, the good performances of the proposed approach suggest it could possibly represent a valid option to extend the capabilities of ALCs to provide quantitative information for operational air quality and meteorological monitoring. [ABSTRACT FROM AUTHOR]