The pervasive nature of plastic pollution has expanded beyond the oceans to include the atmosphere. While research on atmospheric microplastics (MPs) is still in its early stages, a significant body of work has focused on quantifying deposition in urban and rural environments. Studies in Paris, London, and Dongguan, China, have reported significant MP fallout, prompting increased monitoring efforts. Further research in Hamburg and the French Pyrenees Mountains have confirmed atmospheric deposition, with the latter showing transport distances exceeding 95 km. Studies in the USA identified a correlation between dry deposition, regional dust deposition, and large-scale atmospheric patterns, suggesting the potential influence of the free troposphere (FT) on MP transport. Marine studies have also highlighted the presence of MPs in marine aerosols, with back-trajectory modeling suggesting transport from distant sources such as Japan, mainland China, and Korea. The discovery of MPs in Arctic snow further underscores the potential for long-range transport. A global analysis of tire and brake wear MPs has shown that these particles are transported long distances, with a significant portion deposited in the oceans. These findings demonstrate MP presence in the planetary boundary layer (PBL), but the extent of transport into the FT and its implications for global distribution remain largely unexplored. The FT, with its high-speed winds and lack of surface friction, acts as a global transport vector for various pollutants. The detection of MPs in the FT would signify their potential to reach even the most remote areas of the globe, altering our understanding of the global extent of microplastic pollution. This study focuses on the Pic du Midi Observatory (PDM), a high-altitude, clean station in the French Pyrenees, offering an ideal location to investigate FT MP transport due to its minimal local environmental influence and infrequent PBL intrusion.

Existing literature extensively documents oceanic plastic debris, dating back to Carpenter & Smith's 1972 Sargasso Sea study. In contrast, atmospheric microplastic research is a relatively recent field, with a majority of studies concentrating on quantifying deposition rather than transport. Megacity studies in Paris, London, and Dongguan revealed substantial MP fallout (175–1008 MP particles/m²/day). Similar deposition rates were observed in urban and rural Hamburg. However, the investigation into atmospheric MP transport has been limited. Studies in the French Pyrenees identified daily MP deposition of ~365 MP/m²/day at 1425 m altitude, showing transport exceeding 95 km. London’s atmospheric MP analysis revealed local transport of 12–60 km, with a long-range influence area of up to 8700 km². A broad spatial study across US wilderness areas revealed comparable deposition rates (48–435 MP/m²/day), noting that larger particles tended to be regionally sourced, while smaller particles exhibited longer-distance transport via dry deposition. Importantly, this US study highlighted a correlation between dry deposition, regional dust deposition, and large-scale atmospheric patterns (southerly jet stream), suggesting free troposphere involvement. Brahney et al. modeled potential US MP sources, identifying roads, agricultural soil, and oceans as key contributors to atmospheric deposition in remote wilderness areas. Marine studies, from samples taken across vast distances (Shanghai to Mariana Islands, Pearl River to Indian Ocean), found significant MP concentrations (up to 1.37 MP/m³) in marine aerosols, up to 550 km offshore. HYSPLIT back-trajectory modeling linked these particles to potential sources in Japan, mainland China, Korea, and the Philippines. Bergmann et al. documented high MP concentrations in snow samples ranging from the French Alps to Greenland and the Arctic, suggesting long-range atmospheric transport similar to mercury’s transport pathways. The evidence points to widespread MP presence in the PBL, but the question of FT ubiquity and transport distances remains.

This study utilized the Pic du Midi Observatory (PDM) in the French Pyrenees, a high-altitude clean station with infrequent PBL influence, for aerosol sampling. A TISCH high-volume PM10 sampler, equipped with a quartz fiber filter membrane, collected samples over 15 seven-day periods between June 23 and October 23, 2017. The total pumped air volume per sample averaged 7880 m³. Three 30-mm diameter areas from each filter were analyzed for MPs using µRaman spectroscopy (Horiba XploraPlus), identifying polymer types (LD/HDPE, PS, PVC, PET, PP). Particle size and shape (fibers vs. fragments) were determined using Raman spectroscopy and image analysis (FIJI). Aerodynamic diameter was calculated using a standard equation incorporating particle density, aspect ratio, and cylindrical diameter. Meteorological data (wind speed, direction, temperature, humidity, precipitation) were obtained from the P2OA database. HYSPLIT model version 4 (April 2018) was used for 7-day back-trajectory analysis of air masses, providing hourly latitude, longitude, and elevation data. FLEXPART version 9.02 modeled particle dispersion, considering PBL/FT mixing and turbulence. The model was run in backward mode for each hour of sampling, releasing particles at the PDM location (3000 m ASL) and using ERA-Interim meteorological data. Results were presented as potential emission sensitivities (PES), indicating source areas. Statistical analysis (Pearson and Spearman tests) assessed correlations between MP characteristics (count, size, polymer type) and meteorological parameters. Aerodynamic diameter calculations were performed using Equation 1, accounting for particle density, aspect ratio, and cylindrical diameter. The analysis included assessing correlations between MP characteristics and meteorological data, using appropriate parametric or non-parametric methods as needed. Log10 transformation was used for non-parametric data to enhance analysis. HYSPLIT back-trajectory elevations were processed to determine the 25th, 50th, 75th, maximum, and minimum elevations, with ArcGIS used for distance calculations.

Microplastic fragments and fibers were detected in all samples, with concentrations ranging from 0.09–0.66 MP/m³ (average 0.23 MP/m³). Most particles were <20 µm (aerodynamic diameter), with 51% ≤ 10 µm. The most abundant polymers were polyethylene (LD/HDPE), polystyrene (PS), polyvinylchloride (PVC), polyethylene terephthalate (PET), and polypropylene (PP). There was a tentative association between the proportion of MPs <10 µm and fragment content. Wind velocity ranged from 1.4–22.6 m/s (mean 7.9 m/s), with stronger winds predominantly from the west-southwest. Larger MPs (>10 µm) showed a correlation with maximum wind velocity from the north. No significant correlation was found between meteorological conditions and total MP counts. Back-trajectory and dispersion modeling showed that air masses had traveled at least 275 km before reaching PDM, maintaining elevations >2000 m ASL. Samples with higher MP concentrations (>0.33 MP/m³) exhibited lower average trajectory elevations (2747 ± 373 m ASL) and more frequent PBL/FT mixing (9% ± 6%) compared to samples with lower concentrations (3276 ± 425 m ASL and 2% ± 1%, respectively). The duration of time the overall trajectories (air mass/particle) spend within the PBL relative to each sample is not significantly correlated to the PDM atmospheric MP concentration, despite a visual trend that suggests a link between greater average duration of back-trajectory occurrence within the PBL and an elevated MP particle count in the PDM samples. Three samples showed backward trajectories reaching surface level within 168 h. A positive correlation was observed between the frequency of PBL/FT mixing and MP concentration (r = 0.69, P < 0.05 for all MPs; r = 0.78, P < 0.05 for MPs > 10 µm). Average air mass movement from PDM was 4550 km over 168 h (range 2047–6631 km). Higher MP samples showed a greater proportion of trajectories over the Mediterranean Sea and North Africa, while lower MP samples had a greater proportion over the Atlantic Ocean and North America. A positive trend existed between larger MPs (>10 µm) and the number of trajectories over North Africa. Potential emission sensitivity (PES) analysis showed extensive areas over the Atlantic, North Africa, Europe, and North America as potential source regions.

Direct comparison with other studies is challenging due to variations in minimum particle size analyzed and sampling methods. However, when focusing on the comparable particle size ranges, the findings are within a similar range to studies conducted in Beijing, USA wilderness areas, and the French Atlantic coast, even though the concentrations were generally lower than those found in urban environments and coastal areas. The differences in concentrations are likely attributed to the location-specific factors and the sampling methods. This study's focus on the inhalable <PM10 fraction might also explain this difference. Marine offshore air samples did not include the PM10 or smaller particle fraction focus within the sampling regime and therefore show lower MP concentrations. Our results demonstrate that long-distance transport of MPs can occur, consistent with findings from other studies. The significant influence of PBL mixing and the spatial trajectory of air masses (over land vs. sea) on MP concentration and composition further highlights the complexity of atmospheric MP transport. The identified long-range transport has significant implications for remote areas, posing potential risks to both environmental and human health due to the possibility of long-distance transport of adsorbed chemicals and pathogens.

This study provides compelling evidence for the presence of microplastics in the free troposphere and their long-range transport, with potential sources identified across the Atlantic Ocean, Mediterranean Sea, Europe, and North Africa. The results highlight the global extent of atmospheric microplastic pollution, emphasizing the need for further research to fully understand the sources, transport mechanisms, and environmental and health impacts of these pervasive pollutants. Future research could focus on refining sampling methodologies, expanding spatial coverage, and investigating seasonal variations in MP transport. A greater understanding of the interaction between MPs and atmospheric processes is crucial for evaluating the associated ecological and human health risks.

The study's findings are based on a limited sampling period (four months) and location (PDM Observatory). The TISCH high-volume sampling system's efficacy in collecting MPs across all size ranges needs further investigation. The back-trajectory modeling relies on assumptions regarding MP particle behavior and atmospheric dynamics, which could introduce uncertainties. Furthermore, the study’s focus on the inhalable fraction might not be entirely representative of the total atmospheric MP load. While the study uses advanced modeling techniques, the complexities of atmospheric transport mean there's always some uncertainty about the precise source areas.