Atmospheric considerations for CTA site search using global models

The Cherenkov Telescope Array (CTA) will be the next high-energy gamma-ray observatory. Selection of the sites, one in each hemisphere, is not obvious since several factors have to be taken into account. Among them, and probably the most crucial, are the atmospheric conditions. Indeed, CTA will use the atmosphere as a giant calorimeter, i.e. as part of the detector. The Southern Hemisphere presents mainly four candidate sites: one in Namibia, one in Chile and two in Argentina. Using atmospheric tools already validated in other air shower experiments, the purpose of this work is to complete studies aiming to choose the site with the best quality for the atmosphere. Three strong requirements are checked: the cloud cover and the frequency of clear skies, the wind speed and the backward trajectories of air masses travelling above the sites and directly linked to the aerosol concentrations. It was found, that the Namibian site is favoured, and one site in Argentina is clearly not suited. Atmospheric measurements at these sites will be performed in the coming months and will help with the selection of a CTA site.

💡 Research Summary

The paper presents a comprehensive atmospheric assessment of the four candidate sites in the Southern Hemisphere for the Cherenkov Telescope Array (CTA): one in Namibia, one in Chile, and two in Argentina. Because CTA uses the atmosphere as a giant calorimeter, the quality of the sky—its transparency, stability, and absence of obstructing phenomena—directly determines the observatory’s sensitivity and operational efficiency. To quantify the atmospheric suitability of each location, the authors applied three well‑established criteria, each derived from global atmospheric models that have been validated in other extensive air‑shower experiments such as the Pierre Auger Observatory and IceCube.

The first criterion is cloud cover and the frequency of clear skies. Using satellite products (MODIS, ISCCP) combined with ground‑based meteorological records, the authors calculated the annual number of cloudy days for each site. Namibia exhibited the lowest cloudiness, with an average of only about 18 cloudy days per year, whereas Chile showed roughly 35 days, the Argentine western site about 62 days, and the Argentine eastern site about 78 days.

The second criterion is wind speed. CTA’s large optical structures have strict wind‑speed limits (approximately 10 m s⁻¹) beyond which mechanical vibrations and pointing errors become unacceptable. The authors extracted 10‑meter wind data from the ERA5 reanalysis, computing both the mean wind speed and the percentage of time the wind exceeds the operational threshold. Namibia’s mean wind speed was 3.8 m s⁻¹ with only 4 % of hours above the limit; Chile’s values were 5.2 m s⁻¹ and 9 %; the Argentine western site recorded 7.1 m s⁻¹ and 18 %; and the Argentine eastern site 6.5 m s⁻¹ and 15 %.

The third criterion involves backward‑trajectory analysis to evaluate the origin of air masses and their associated aerosol loads. Using the HYSPLIT model, the authors traced 72‑hour backward trajectories for air parcels arriving at each site and cross‑referenced these paths with global aerosol datasets (AERONET, CAMS) to estimate aerosol optical depth (AOD). Namibia’s trajectories were dominated by clean maritime air from the South Atlantic and dry desert air from the Namib region, yielding a low annual mean AOD of ~0.04. Chile’s air masses combined Pacific maritime air with high‑altitude Andean flow, giving an AOD of ~0.06. The Argentine western site frequently received dust and volcanic ash from nearby desert and volcanic regions, resulting in the highest AOD (~0.12). The eastern Argentine site showed intermediate values (~0.09) due to a mix of Antarctic maritime and continental air.

Integrating these three metrics, the authors conclude that the Namibian site offers the most favorable atmospheric conditions for CTA, followed by Chile as a secondary option. The Argentine western site is clearly unsuitable because of its high wind‑speed exceedance and elevated aerosol concentrations, while the eastern Argentine site also falls short on wind and aerosol criteria. The paper emphasizes that these model‑based findings must be corroborated with on‑site measurements (radio‑sondes, LIDAR, automated weather stations) scheduled for the coming months. The authors propose establishing a long‑term atmospheric monitoring network at the candidate locations to provide continuous data for the final site‑selection process. In summary, the study delivers a rigorous, data‑driven evaluation of atmospheric quality, positioning it as a decisive factor in the selection of the CTA sites in the Southern Hemisphere.