Atmos. Chem. Phys., 16, 11581–11600, 2016 www.atmos-chem-phys.net/16/11581/2016/ doi:10.5194/acp-16-11581-2016 © Author(s) 2016. CC Attribution 3.0 License. Saharan dust long-range transport across the Atlantic studied by an airborne Doppler wind lidar and the MACC model Fernando Chouza1, Oliver Reitebuch1, Angela Benedetti2, and Bernadett Weinzierl3 1Institut für Physik der Atmosphäre, Deutsches Zentrum für Luft- und Raumfahrt (DLR), Oberpfaffenhofen, Germany 2European Centre for Medium-Range Weather Forecasts, Reading, UK 3Faculty of Physics, University of Vienna, Boltzmanngasse 5, 1090 Vienna, Austria Correspondence to: Fernando Chouza ([email protected]) Received: 4 May 2016 – Published in Atmos. Chem. Phys. Discuss.: 19 May 2016 Revised: 2 September 2016 – Accepted: 8 September 2016 – Published: 20 September 2016 Abstract. A huge amount of dust is transported every year 1 Introduction from north Africa into the Caribbean region. This paper presents an investigation of this long-range transport pro- cess based on airborne Doppler wind lidar (DWL) measure- Every year, huge amounts of Saharan dust originating from ments conducted during the SALTRACE campaign (June– north Africa is transported across the Atlantic into the July 2013), as well as an evaluation of the ability of the Caribbean region and the north of South America. The trans- MACC (Monitoring Atmospheric Composition and Climate) port, mainly occurring during the summer season, starts with global aerosol model to reproduce it and its associated fea- the uplifting of dust by turbulent convection and low-level tures. Although both the modeled winds from MACC and the winds with high speed (Bou Karam et al., 2008), is the measurements from the DWL show a generally good agree- amount of emitted dust regulated by different factors like the ment, some differences, particularly in the African easterly soil humidity and vegetation. Once lofted, the dust is dis- jet (AEJ) intensity, were noted. The observed differences be- persed into a deep mixed layer, reaching altitudes of up to tween modeled and measured wind jet speeds are between 6 km during summer (Messager et al., 2010). The dominating 5 and 10 m s−1. The vertical aerosol distribution within the easterly winds west advect the dust-laden air masses, which Saharan dust plume and the marine boundary layer is investi- are undercut by the cooler and moister air from the marine gated during the June–July 2013 period based on the MACC boundary layer (MBL) as they reach the west African coast, aerosol model results and the CALIOP satellite lidar mea- forming an elevated layer of relatively warm, dry and dust- surements. While the modeled Saharan dust plume extent laden air called Saharan air layer (SAL). As the SAL leaves shows a good agreement with the measurements, a system- the African continent, its lower and upper bounds are defined atic underestimation of the marine boundary layer extinction by a strong inversion at approximately 1.5 km and a relatively is observed. weaker inversion at around 5 to 6 km, respectively (Prospero Additionally, three selected case studies covering differ- and Carlson, 1972; Karyampudi et al., 1999). ent aspects of the Saharan dust long-range transport along Along its life cycle, the airborne dust interacts with the en- the west African coast, over the North Atlantic Ocean and vironment in different ways. During its long-range transport the Caribbean are presented. For the first time, DWL mea- phase, the dust modifies the radiative budget, acts as cloud surements are used to investigate the Saharan dust long-range and ice nuclei and is observed to modify the cloud glacia- transport. Simultaneous wind and backscatter measurements tion process (e.g., Seifert et al., 2010). Various mechanisms from the DWL are used, in combination with the MACC have been proposed to explain the controversial influence of model, to analyze different features associated with the long- the SAL on the evolution of African easterly waves (AEWs) range transport, including an African easterly wave trough, into tropical storms (e.g., Dunion et al., 2004; Evan et al., the AEJ and the intertropical convergence zone. 2006; Lau and Kim, 2007). Regularly mineral dust is im- pacting aviation, in particular in regions near dust sources, Published by Copernicus Publications on behalf of the European Geosciences Union. 11582 F. Chouza et al.: Saharan dust long-range transport across the Atlantic by inducing poor visibility (Weinzierl et al., 2012). As it de- craft Falcon similar to the measurements done during SA- posits, the Saharan dust can affect the air quality (Prospero, MUM (Weinzierl et al., 2009, 2011). For the first time, an 1999) and serves as source of nutrients for plankton and the airborne Doppler wind lidar (DWL) was deployed to study Amazon basin (Yu et al., 2015). the dust transport across the Atlantic Ocean, including its in- Several studies were conducted during the last years to teraction with island-induced gravity waves (Chouza et al., provide further insight in the previously mentioned pro- 2016). cesses, including field campaigns and long-term studies In this study, this unique set of DWL wind and extinction based on models, satellite, airborne and ground-based mea- measurements along the main dust transport acquired during surements. Among others, we can mention the African SALTRACE are used, in combination with dropsondes (DSs) Monsoon Multidisciplinary Analysis (AMMA) and NASA- and CALIPSO (Cloud–Aerosol Lidar and Infrared Pathfinder AMMA (NAMMA) campaigns (Zipser et al., 2009) con- Satellite Observations) satellite measurements, to analyze the ducted in 2006, focused on the analysis of the AEW and Saharan dust long-range transport and to evaluate the per- its evolution into tropical cyclones, and the Saharan Min- formance of the MACC (Monitoring Atmospheric Compo- eral Dust Experiments 1 and 2 (SAMUM-1 and SAMUM- sition and Climate) aerosol global model to reproduce dif- 2) (Heintzenberg, 2009; Ansmann et al., 2011) conducted in ferent associated atmospheric features, like the African east- 2006 and 2008, respectively, which were designed to investi- erly jet (AEJ) and its interaction with the SAL, the AEWs gate the Saharan dust size distribution and morphology, and and the ITCZ (intertropical convergence zone), among oth- the relation with the optical and radiative properties in the ers. Such a comparison provides not only an insight about west African region. Additionally, during April–May 2013, the current model capabilities, which is of great relevance a shipborne lidar onboard the research vessel Meteor con- for model-based studies, but the opportunity to identify the ducted a transatlantic cruise between the Caribbean and the model weaknesses and provide a starting point for future west coast of Africa in order to characterize the mixtures of improvements. This type of evaluation has never been per- the Saharan dust with biomass burning aerosols and evaluate formed with a dataset that includes both meteorological fields the change in its optical properties as result of the long-range as well as atmospheric composition fields. Although there is transport process (Kanitz et al., 2014). While field campaigns no direct feedback from the atmospheric composition fields provide an intensive set of observations in relatively small re- onto the meteorological fields (temperature and winds) as gions and time intervals, satellite observations provide regu- the run was not coupled through interactive radiative pro- lar observations with a limited set of parameters. On the other cesses, the aerosols are transported and advected by the hand, global and regional models are frequently used in dust winds. Hence, an evaluation of both winds and aerosols in long-range transport studies and forecasting (e.g., Schepan- the SAL is a very effective way to understand how well the ski et al., 2009; Kim et al., 2014; Gläser et al., 2015), as they model represents this natural phenomenon. provide valuable information to interpret the observational The paper is organized as follows. Section 2 presents a data collected by campaigns and satellite measurements. The brief description of the datasets used for this study, includ- performance evaluation of these models, like during the in- ing an evaluation of the DWL accuracy based on collocated tercomparison initiative AeroCom (Aerosol Comparison be- dropsonde measurements. Section 3 provides a comparison tween Observations and Models) (Kinne et al., 2003), is then between the DWL and CALIPSO measurements with the not only vital to improve the models but also to know the lim- backscatter and winds from the MACC model in the west its and accuracy of the conclusions extracted based on these African and Caribbean regions for the time period of the models. SALTRACE campaign. Section 4 presents three case studies The Saharan Aerosol Long-range Transport and Aerosol- with relevant features of the Saharan dust long-range trans- Cloud-Interaction Experiment (SALTRACE; Weinzierl et port. Finally, the summary and relevant conclusions are pre- al., 2016) performed in June–July 2013 was framed in sented in Sect. 5. this context. SALTRACE was planned as a closure exper- iment to investigate the Saharan dust long-range transport between Africa and the Caribbean, with a focus on the 2 Observations and model data dust aging and deposition processes and the characteriza- tion of its optical properties. The campaign dataset includes During the SALTRACE campaign, the DLR Falcon re- a set of measurements from the ground-based aerosol li- search aircraft conducted 31 research flights between 10 June dars BERTHA (Backscatter Extinction lidar-Ratio Temper- and 15 July 2013, with most of the flights concentrated ature Humidity profiling Apparatus) (Haarig et al., 2015) close to the west African coast and the Caribbean (Fig. 1). and POLIS (portable lidar system) (Groß et al., 2015), in The DLR Falcon was equipped with a set of instruments situ and sun photometer instruments deployed on Barbados for in situ particle measurements, a DWL and dropson- (main SALTRACE super-site), Cabo Verde and Puerto Rico des.