Atmos. Chem. Phys., 19, 14149–14171, 2019 https://doi.org/10.5194/acp-19-14149-2019 © Author(s) 2019. This work is distributed under the Creative Commons Attribution 4.0 License. Different strategies to retrieve aerosol properties at night-time with the GRASP algorithm Jose Antonio Benavent-Oltra1,2, Roberto Román3, Juan Andrés Casquero-Vera1,2, Daniel Pérez-Ramírez1,2, Hassan Lyamani1,2, Pablo Ortiz-Amezcua1,2, Andrés Esteban Bedoya-Velásquez2,4, Gregori de Arruda Moreira2,5, África Barreto3,6,7, Anton Lopatin8, David Fuertes8, Milagros Herrera9, Benjamin Torres9, Oleg Dubovik9, Juan Luis Guerrero-Rascado1,2, Philippe Goloub9, Francisco Jose Olmo-Reyes1,2, and Lucas Alados-Arboledas1,2 1Department of Applied Physics, Universidad de Granada, 18071, Granada, Spain 2Andalusian Institute for Earth System Research, IISTA-CEAMA, Granada, Spain 3Grupo de Óptica Atmosférica (GOA), Universidad de Valladolid, Valladolid, Spain 4Sciences Faculty, Department of Physics, Universidad Nacional de Colombia, Medellín, Colombia 5Institute of Research and Nuclear Energy (IPEN), São Paulo, Brazil 6Cimel Electronique, Paris, France 7Izaña Atmospheric Research Center, Meteorological State Agency of Spain (AEMET), Izaña, Spain 8GRASP-SAS, Remote Sensing Developments, Université de Lille, 59655, Villeneuve D’ASCQ, France 9Laboratoire d’Optique Atmosphérique (LOA), UMR8518 CNRS, Université de Lille, 59655 Villeneuve D’ASCQ, France Correspondence: Jose Antonio Benavent-Oltra ([email protected]) Received: 28 July 2019 – Discussion started: 5 September 2019 Revised: 4 November 2019 – Accepted: 6 November 2019 – Published: 22 November 2019 Abstract. This study evaluates the potential of the GRASP Evaluations of the columnar aerosol properties retrieved algorithm (Generalized Retrieval of Aerosol and Surface by GRASP are done versus standard AERONET retrievals. Properties) to retrieve continuous day-to-night aerosol prop- The coherence of day-to-night evolutions of the different erties, both column-integrated and vertically resolved. The aerosol properties retrieved by GRASP is also studied. The study is focused on the evaluation of GRASP retrievals extinction coefficient vertical profiles retrieved by GRASP during an intense Saharan dust event that occurred dur- are compared with the profiles calculated by the Raman ing the Sierra Nevada Lidar aerOsol Profiling Experiment I technique at night-time with differences below 30 % for all (SLOPE I) field campaign. For daytime aerosol retrievals, schemes at 355, 532 and 1064 nm. Finally, the volume con- we combined the measurements of the ground-based li- centration and scattering coefficient retrieved by GRASP at dar from EARLINET (European Aerosol Research Lidar 2500 m a.s.l. are evaluated by in situ measurements at this Network) station and sun–sky photometer from AERONET height at Sierra Nevada Station. The differences between (Aerosol Robotic Network), both instruments co-located in GRASP and in situ measurements are similar for the differ- Granada (Spain). However, for night-time retrievals three ent schemes, with differences below 30 % for both volume different combinations of active and passive remote-sensing concentration and scattering coefficient. In general, for the measurements are proposed. The first scheme (N0) uses li- scattering coefficient, the GRASP N0 and N1 show better re- dar night-time measurements in combination with the inter- sults than the GRASP N2 schemes, while for volume con- polation of sun–sky daytime measurements. The other two centration, GRASP N2 shows the lowest differences against schemes combine lidar night-time measurements with night- in situ measurements (around 10 %) for high aerosol optical time aerosol optical depth obtained by lunar photometry ei- depth values. ther using intensive properties of the aerosol retrieved during sun–sky daytime measurements (N1) or using the Moon au- reole radiance obtained by sky camera images (N2). Published by Copernicus Publications on behalf of the European Geosciences Union. 14150 J. A. Benavent-Oltra et al.: Different strategies to retrieve aerosol properties at night-time 1 Introduction extinction (α) coefficient by assuming a constant aerosol extinction-to-backscattering ratio, which is called the lidar Knowledge of the atmospheric aerosol optical and micro- ratio (LR). On the other hand, more advanced lidar systems physical properties is important due to their different effects implement the Raman (e.g. Ansmann et al., 1992; Whiteman on the Earth–atmosphere radiative budget (IPCC, 2013). et al., 1992) technique for independent retrievals of aerosol The aerosol particles can scatter and absorb solar and ter- backscatter and extinction measurements. These multiwave- restrial radiation. The Earth–atmosphere radiative forcing length lidar measurements allow the use of different inver- sign (warming or cooling) is sensitive to aerosol optical sion algorithms based on the regularization technique to re- and microphysical properties and their vertical distribution trieve vertical profiles of aerosol microphysical properties us- (e.g. Boucher et al., 2013). In addition, aerosol particles can ing a 3β C 2α configuration, that is, multiwavelength lidar act as cloud condensation and ice nuclei and, thus, can mod- measurements of three backscatter and two extinction coeffi- ify the development, microphysical properties and lifetime cients (e.g. Müller et al., 1999; Böckmann, 2001; Veselovskii of clouds (e.g. Andreae et al., 2004; Boucher et al., 2013). et al., 2002). Nevertheless, the amount of advanced lidar sys- Recent developments in remote sensing have allowed ad- tems is considerably lower when compared with basic li- vancing the understanding aerosol globally, but the charac- dar systems; therefore, the independent α and β measure- teristics of each system do not allow a complete day-to-night ments are sparse and mostly limited to night-time. In this characterization, especially in aerosol microphysical proper- context there are a lot of passive and active remote-sensing ties (e.g. Pérez-Ramírez et al., 2012). Understanding day-to- measurements that alone do not provide enough informa- night aerosol properties from remote-sensing measurements tion to retrieve advanced aerosol microphysical properties. is essential to advances in aerosol dynamics and changes, However, integrating all these measurements in an appro- which eventually will serve to advance our knowledge on priate inversion scheme allows such retrievals and can even aerosol impact on air-quality and climate. Therefore, current complete the number of unknown aerosol optical proper- efforts are in integrating different measurements that require ties. Such integration is critical for retrieving vertical pro- advancing in the development of retrieval techniques. files where the information content for the retrievals is con- During the last 2 decades, global and regional networks siderably low when compared with classical sun photometer have been established to get a comprehensive, quantitative inversion (e.g. Veselovskii et al., 2005). In the framework and statistically significant database of atmospheric aerosols. of EARLINET, different inversion algorithms were devel- The Aerosol Robotic Network (AERONET; Holben et al., oped, such as the LIdar-Radiometer Inversion Code (LIRIC; 1998) and East Asian SKYNET (Nakajima et al., 2007) use Chaikovsky et al., 2008, 2016), which uses AERONET re- sun–sky photometers to provide aerosol column-integrated trievals and backscatter elastic signals as input, and the Gen- properties with high temporal resolution. These networks eralized Aerosol Retrieval from Radiometer and Lidar Com- use retrieval techniques that allow the characterization of bined (GARRLiC; Lopatin et al., 2013) code, which uses aerosol microphysical properties (e.g. Nakajima et al., 1996; sun–sky radiance and backscatter lidar measurements as in- Dubovik and King, 2000). These networks were focused on puts that make the inversion more consistent (Lopatin et al., daytime measurements, but nowadays they are trying to add 2013). night-time aerosol measurements derived from lunar pho- Of these algorithms, in this study, we use the recently de- tometry. The developments in moon (Berkoff et al., 2011; veloped Generalized Retrieval of Aerosol and Surface Prop- Barreto et al., 2013, 2016) and star photometry (e.g. Pérez- erties algorithm (GRASP; Dubovik et al., 2011, 2014), which Ramírez et al., 2011, 2012; Baibakov et al., 2015) allow the includes the GARRLiC code. GRASP is a versatile and open- acquisition of night-time measurements; however, these mea- source algorithm (https://www.grasp-open.com/, last access: surements are limited in the inversion algorithms to retrieve 1 July 2019) based in the concept of the Dubovik and the aerosol microphysical properties (Pérez-Ramírez et al., King (2000) algorithm which has been used successfully 2015; Torres et al., 2017). by AERONET during the last decades. The GRASP algo- Lidar networks such as EARLINET (European Aerosol rithm is divided into two main independent modules: the Research LIdar NETwork; Pappalardo et al., 2014), LA- forward model and numerical inversion modules. The for- LINET (Latin American LIdar NETwork; Guerrero-Rascado ward model is based on radiative transfer and aerosol mod- et al., 2016; Antuña-Marrero et al., 2017) and MPLNET els, and it is a convenient tool for sensitivity and tuning stud- (Micro-Pulse Lidar Network) (Welton et al., 2002) pro- ies (Dubovik et al., 2014; Torres et al., 2017). The numeri- vide information