Orza, J. A. G., Gómez Cascales, P. J., Ferro García, M. A., M Expósito Suárez, V., Milena Pérez, A., Martínez Martínez, B. R., Chham, E., Piñero García, F., Cazorla, A., Alados-Arboledas, L., Olmo, F. J., Erika Brattich, Laura Tositti, Camacho, A., Hernández-Ceballos, M. A., Martiny, N., and J. A. G. Orza, P.J. Gómez Cascales, M.A. Ferro García, V.M Expósito Suárez, A. Milena Pérez, B.R.Martínez Martínez, E. Chham, F. Piñero García, A.Cazorla, L. Alados-Arboledas, F.J. Olmo, E. Brattich, L.Tositti, A. Camacho, M.A.Hernández-Ceballos, N. Martiny
The simultaneous aerosol sampling at two heights in southern Spain may provide valuable information on the vertical structure of the dust transport from North Africa to the Iberian Peninsula. It also allows the characterization of the ambient air at two sites with distinct anthropogenic influence. This work presents the results obtained from the first field campaign of the FRESA project (Impact of dust-laden African air masses and of stratospheric air masses in the Iberian Peninsula. Role of the Atlas Mountains), performed in the period July-November 2017 at El Albergue Universitario in Sierra Nevada (2550 m a.s.l.) and the city of Granada (650 m a.s.l.). The two sites were instrumented with a low-volume sampler with PM10 inlet for daily sampling and mass and chemical composition characterization, a high-volume sampler for total suspended particles (TSP) for weekly sampling and radionuclide activity determination, and with a GRIMM 365 optical particle counter that provides continuously the aerosol size distribution. In Sierra Nevada, daily PM10 concentrations ranged from 0 to 104 µg m-3 depending mainly on the origin and features of the air masses that reach this high-elevation site. Levels overpassed 50 µg m-3 during 5 days and were lower than 20 µg m-3 for 90 days over a total of 124 sampling days. The impact of dust-laden African airflows at this site is particularly intense as dust is transported in most of the cases within welldefined low mid-tropospheric layers. The associated episodic concentrations exceeded 40 µg m-3 . In turn, clean subsiding airflows associated to the polar jet strongly reduced concentrations. In Granada, daily PM10 concentrations are moderately high with values generally between 20 and 40 µg m3 before the first snows fall over Sierra Nevada. Levels were over 50 µg m-3 on 2 days and below 20 µg m-3 on 32 days over a total of 103 sampling days. The highest concentrations do not reach the peaks found at Sierra Nevada. The impact of anthropogenic aerosols of local origin, which are accumulated during highstability conditions, and also of regional origin (both from the Mediterranean and from the SW Iberian Peninsula) is significant. The contribution of African dust outbreaks superimposes to background ambient air concentrations and in some events the increase is observed one day after the African dust impacts over Sierra Nevada. After the first snowfall of the year in Sierra Nevada, there’s a change in wind regime, PM10 levels drop and only accumulation periods or African dust events are able to increase concentrations at levels similar to those detected before. The diurnal pattern of PM10 and particle number concentrations (both of the fine and the coarse fractions) at Granada presents the two typical peaks in the morning and evening; there’s one peak at 16 Local Time - LT in Sierra Nevada. Particle concentrations are smaller in the colder period than in the warmer one, primarily due to the much higher concentrations in the coarse fraction (> 1 µm) in summertime. However, the fine fraction (submicronic) presents a stronger morning peak (centered about 10 LT) in the colder period and also concentrations are slightly higher at about 20-21 LT. Trajectories (calculated 96-hour backward in time starting at 00, 06, 12 and 18 UTC with HYSPLIT using ERAInterim data of 0.5 degree resolution) show that air flows are quite often decoupled at the two altitudes. Dust-laden African flows reached Sierra Nevada on 33 days; of these, dust was advected poleward over the Atlas near the Algerian-Moroccan border on 19 days and on the remaining 14 it followed a pathway over the Atlantic close to the Moroccan coast. However, only on 3 days large-scale flows of African origin reached Granada. In turn, airflows reaching Granada passed previously over the western Mediterranean Sea on 32 days while only on 3 days Mediterranean flows reached Sierra Nevada. Moreover, the total residence time over North Africa of the air parcels reaching Sierra Nevada amounts 9474 hours, and during the corresponding days the total residence time over the Mediterranean of the parcels reaching Granada is 10,055 hours. In contrast, the residence time of the air masses arriving to Granada after residing over Africa is only of 56 hours. This implies that during African dust outbreaks the air masses reaching the study area at the lowest levels do not have African origin but have resided over the Mediterranean, and it is in accordance with Cabello et al. (2017) for Málaga and Orza et al. (submitted) for the whole Spanish Mediterranean. Chemical analysis of the PM10 samples (in progress) will provide the first direct experimental evidence of this fact. The identification of episodes was done by analysing the fine and coarse fractions registered with the Grimm OPCs at the two sites and by the aerosol optical depth (AOD) and Angström exponent (AE) from the AERONET station of IISTA-CEAMA in Granada, in combination with back-trajectories. The low availability of satellite information (Dust RGB product and MODIS aerosol properties) is a major limitation at the study area. A number of distinct episodes can be identified: strong African dust impact at Sierra Nevada while Granada is heavily polluted by anthropogenic aerosols (1 August 2017); strong African dust impact at Sierra Nevada which is observed the following day in Granada (15-16 August 2017); African dust outbreaks impacting simultaneously both sites (e.g., 25 November 2017); episode of accumulation of pollutants at Granada whilst very low concentrations are registered at Sierra Nevada (19-24 November 2017); impact of remote fires at both sites (27 July and 9 September 2017) and fair air quality at both sites (e.g., 16 September 2017). Differences in the onset and duration of the African dust episodes are found between the two sites as well as between the surface and the columnar measurements.