Selecting the site for the advanced technology solar telescope
To measure magnetic fields in the sun's outer atmosphere (the corona), and to provide the highest spatial resolution solar images ever achieved, the Advanced Technology Solar Telescope (ATST) must be located at a site with excellent daytime seeing and low sky brightness. Seeing is affected by turbulence in the Earth's atmosphere that randomly refracts sunlight and causes image degradation. While the ATST will have a high-order adaptive optics system that can greatly reduce these effects, the quality of the observations will always be better by avoiding them in the first place. For coronal magnetic fields, the problem is one of signal-to-noise, since the corona is very faint and is easily masked by a bright sky.
Most astronomical-site testing campaigns are carried out at night, when the ground and air are at nearly the same temperature and the seeing is dominated by turbulence at the troposphere-stratosphere at a height of 20km. During the day, the ground is hotter than the air, causing convective plumes that degrade seeing near the ground. This surface seeing layer has a finite thickness, so it is advantageous to raise a solar telescope as high as practical. Hence, we need to measure the seeing as a function of height to determine the design of the telescope. Night-time seeing instruments typically use stellar images as point sources to estimate seeing, whereas the sun is an extended object. We also need to measure the brightness of the daytime sky close to the edge of the solar disk.
To meet these challenges, we developed three new instruments: two for seeing1 and one for sky brightness.2 The seeing instruments are a solar differential image motion monitor (S-DIMM) and a non-redundant array of six scintillometers, known as SHABAR (SHAdow BAnd Ranging). We also developed a sky brightness monitor (SBM). The combination of the S-DIMM and the SHABAR provided data that allowed us to estimate the seeing as a function of height up to 100m. The SBM measured the relative sky brightness as a function of distance from the solar disk in three wavelengths. We developed an inversion algorithm to provide seeing as a function of height3 and used the daily variation of the SBM data to estimate extinction and sky brightness.4
The S-DIMM imaged the edge of the solar disk through two apertures, a prism, and a slit. The relative motion of the solar limb in the two images can be directly related to a measurement of the Fried parameter, r0, a commonly used seeing measure that is the diameter of a telescope aperture that would give the spatial resolution allowed by the turbulence. The S-DIMM gives r0 integrated over the entire atmosphere. The SHABAR measures the scintillation of sunlight, and the cross-correlations of the measurements over the 15 possible detector separations can be inverted to estimate r0 as a function of height. The SBM is a small coronagraph that blocks most of the direct sunlight with neutral density filters and images the sky adjacent to the solar disk. The S-DIMM and SHABAR were mounted on a tower 6m high, while the SBM was mounted at ground level.
A set of six sites was selected for testing from an initial list of 72 candidates using a combination of geographical and feasibility analyses. The six tested sites were: Sacramento Peak, NM; Big Bear Lake, CA; Panguitch Lake, UT; Haleakala, HI; La Palma, Spain; and San Pedro Martir, Mexico. This set provided six different orographic situations: continental mountain, coastal lake, continental lake, isolated Pacific ocean island, isolated Atlantic ocean island, and peninsula mountain. All six sites were tested initially for 18 months, and then three finalists (Big Bear, Haleakala, and La Palma) were tested for an additional year.
The test results clearly indicated that Haleakala was the best site for the ATST. After correcting for observing schedules, we estimated that, at a height of 28m above the ground, Haleakala would provide approximately 400 hours of seeing annually with r0 greater than 12cm: twice as large as the next best site. The median sky brightness at Haleakala was estimated to be about 5.8 millionths of the solar disk intensity, a factor of six lower than the next best site.
Our work is one of the few solar astronomical site surveys to apply consistent measurement and analysis methods to a variety of sites. We developed new instruments and analysis methods for deriving daytime seeing as a function of height and sky brightness and used these tools to select Haleakala as the site for the ATST. The tools we developed are useful for testing daytime astronomical conditions anywhere in the world.
