Every clear night, a telescope in the Chilean desert records light from distant stars, and in those recordings, like old ship logs, traces of weather remain. The thin dark bands in the spectrum are imprints of water vapor that hung over the observatory at the moment of observation. Over six years, so many such recordings accumulated that they can be read as a whole chronicle of humidity above Cerro Armazones, where the European Extremely Large Telescope will rise. Astronomers usually discard these bands because they obscure the stars themselves, but here they decided to read them anew.
Water vapor for a telescope is like frosted glass for binoculars: it eats infrared light and fogs the view. To choose a site for the future giant, one needs to know how much vapor hangs over the site and how it is distributed across the sky. Previously, weather stations and radiometers were used for this, but they provide only point measurements at the ground. Here, however, an archive of 5879 spectra collected over 918 nights was available, and each spectrum contains information about humidity along the entire line of sight. The molecfit program fitted an atmospheric absorption model to two narrow spectral regions in the near-infrared, and from that fit the amount of vapor was calculated.
This means the Armazones site is drier not only because it is higher, but also because of some local peculiarities in air circulation. Comparison with data from the X-shooter spectrograph on the Very Large Telescope and the LHATPRO microwave radiometer confirmed the reliability of the method: correlation slopes were 0.85 and 0.84, with coefficients of determination around 0.9. The global weather model GDAS, however, systematically overestimated humidity by 52 percent, predicting a wetter sky than actually exists. The distribution of vapor across the sky turned out to be uniform: within 60 degrees of zenith, average values in all directions lie in the range of 3–4 mm with no noticeable asymmetries, including between east and west.
This approach is not just a way to save on equipment; it is a way to obtain a spatially resolved picture of humidity that neither radiometers nor weather balloons provide. For the future Cherenkov Telescope Array (CTA), where flux accuracy is set at the 5–7 percent level, atmospheric monitoring can be carried out with a small telescope equipped with a fiber-fed spectrograph. The method can also be extended to measure ozone, particularly in ultraviolet bands below 400 nm, which current spectra do not cover. Global weather models in this coastal zone of the Atacama proved unsuitable for precise predictions, so local refinements or direct spectroscopic measurements will be necessary.
The chronicle read from water vapor lines has already answered the main question: the Armazones site is indeed dry and uniform, meaning the giant eye of the E-ELT will have something to breathe. It remains only to turn the page and learn to read traces of ozone in the same recordings, so that the sky becomes transparent not only for infrared but also for ultraviolet gaze.
Over six years, the telescopes accumulated spectra from 996 nights, but only 63 percent of them passed strict quality filtering, yet the night coverage remained at 92 percent. This means that even rejecting nearly half of the spectra did not distort the picture of seasonal humidity variations.
In Kim Stanley Robinson's novel 'Aurora', colonists arriving at another star face the problem of atmospheric transparency, on which the operation of their telescopes depends. Here on Earth, the same problem is being solved for a site where sky transparency will be decisive for the giant eye of the E-ELT.