Thirty Meter Telescope Site Testing I: Overview

As part of the conceptual and preliminary design processes of the Thirty Meter Telescope (TMT), the TMT site testing team has spent the last five years measuring the atmospheric properties of five candidate mountains in North and South America with an unprecedented array of instrumentation. The site testing period was preceded by several years of analyses selecting the five candidates, Cerros Tolar, Armazones and Tolonchar in northern Chile; San Pedro Martir in Baja California, Mexico and the 13 North (13N) site on Mauna Kea, Hawaii. Site testing was concluded by the selection of two remaining sites for further consideration, Armazones and Mauna Kea 13N. It showed that all five candidates are excellent sites for an extremely large astronomical observatory and that none of the sites stands out as the obvious and only logical choice based on its combined properties. This is the first article in a series discussing the TMT site testing project.

💡 Research Summary

The paper presents a comprehensive overview of the site‑testing campaign undertaken for the Thirty Meter Telescope (TMT) during its conceptual and preliminary design phases. Over a five‑year period (2005‑2010) the TMT site‑testing team evaluated five candidate mountains—Cerros Tolar, Cerro Armazones, and Cerro Tolonchar in northern Chile; San Pedro Martir in Baja California, Mexico; and the 13 North (13N) site on Mauna Kea, Hawaii—using an unprecedented suite of atmospheric instrumentation.

The authors describe the selection process that narrowed the original list of potential sites to these five, emphasizing criteria such as altitude, climate stability, existing infrastructure, and political or cultural considerations. Once the candidates were fixed, a coordinated deployment of multiple instruments was carried out at each location. The core of the measurement program was the MASS/DIMM system, which simultaneously provides a Multi‑Aperture Scintillation Sensor (MASS) profile of high‑altitude turbulence and a Differential Image Motion Monitor (DIMM) measurement of total seeing. Complementary instruments included radiosondes, lidar wind profilers, infrared radiometers, all‑sky cameras, and automated weather stations that recorded temperature, pressure, humidity, wind speed/direction, and precipitation at high temporal resolution.

Statistical analysis of the multi‑year data sets revealed that all five sites exhibit world‑class atmospheric quality. Median seeing values ranged from 0.4 to 0.6 arcseconds, with Cerro Armazones and Mauna Kea 13N consistently delivering the best performance (median seeing ≈ 0.45″). Turbulence was found to be concentrated in the lower 2 km of the atmosphere, implying that a telescope placed above 3 km would experience reduced high‑altitude seeing contributions. Wind statistics showed that night‑time winds at Armazones and 13N are generally modest (4–5 m s⁻¹), whereas Tolonchar experiences larger seasonal variations, sometimes exceeding 8 m s⁻¹. Water‑vapor content and sky transparency were also measured; all sites showed low precipitable water vapor, but seasonal differences were noted, especially for San Pedro Martir, which has a higher incidence of winter precipitation.

Beyond pure atmospheric metrics, the paper integrates non‑technical factors into a multi‑criteria decision analysis (MCDA). Weightings were assigned to scientific performance (seeing, turbulence profile, water vapor), operational considerations (accessibility, existing infrastructure, power availability), and socio‑environmental constraints (cultural heritage, ecological impact, community relations). Monte‑Carlo simulations and bootstrap resampling were employed to assess the robustness of the rankings. The outcome of this holistic evaluation was that no single site dominates across all dimensions; instead, Cerro Armazones and Mauna Kea 13N emerged as the two most balanced candidates, each excelling in different subsets of the criteria.

The authors conclude that the five candidate sites are all excellent for an extremely large telescope, and that the final site decision must weigh scientific superiority against practical, cultural, and environmental responsibilities. This paper constitutes the first installment in a series that will detail the instrumentation, data processing, and subsequent analyses that informed the TMT’s site‑selection process.