Both these key issues should be resolved prior to final "short list" Chile site selection.At that time, i.e. by end CY 2001, the list of characteristics of the then identified sites will need to be matched with the scientific and other requirements, in order to make a selection of a few sites for detailed, long-term testing.

Requirements Imposed by Science

A critical parameter is the importance attached to IR observations beyond 2.2 microns, which drive the water-vapor requirement and therefore would argue for choosing a very high altitude site.Given that in general the wind strength is correlated with altitude, there may be a trade between 2.1 and 2.2 microns. Despite the existence of infrastructure (MMA) in the region of the likely Chilean very high altitude sites, the construction cost and operational difficulties of choosing to put the NBT at altitudes near 5500m should not be under-estimated. Other important parameters driven by science are the degree of cloud cover, which can restrict the use of laser-guided AO, prevent photometry, and reduce overall efficiency; seeing, sky clarity, and sky brightness. SCIENCE and AO groups will provide input of the relative importance of these parameters.

Requirements Imposed by Structures.

The critical parameter is wind, given the likely approx. 1 Hz fundamental resonance of a 30-50 m telescope structure.Tests and analysis (Nokayama, Gemini) by the STRUCTURES group will further detail the requirements for the site. These requirements are likely to include wind speed and direction measurements, wind flow modeling , and in-site wind spectra with spatial variation.

Cloud Cover Analysis (Chile)

An analysis of Meteosat and Geos satellite (6 and 10 micron channels) data by Dr D.A. Erasmus is in progress, covering 1993-2000, and latitudes 20.5 to 30.5 S. This includes all the major Chile sites.This project has been facilitated by a data-sharing collaboration with ESO, and a data and results sharing collaboration with U. Tokyo,the latter who are proposing to build a 6-m class IR-optimized telescope near Chajnantor.The analysis will provide an 8-year baseline, long enough to cover the most recent El Niño and La Niña phenomena as well as several normal years. The study will identify the best sites based on lack of cloud, and compare them with the existing sites.

Topographical analysis (Chile)

A set of maps of Northern Chile has been purchased in CY1999 from the Instituto Geographica Militar. These maps are at a scale of 1250000, most areas also have 1:50000 available which will be purchased for potential sites, as will digital data from radar imagery. The maps will be studied in order to provide a list of possible mountains, using as criteria: distance from city lights and mining operations, altitude, relief, isolation from the surrounding landscape, and summit area size. Previous site survey information from the AURA, MMA/LSA, and ESO surveys will be gathered. A number of sites on the Coast range, the Domeyko range, and the Andes have been surveyed.Many of these sites have had a small amount of weather data that would be valuable for an initial idea of local wind speeds and diurnal temperature variations.

Ground to Boundary layer wind-modeling (Chile and Mauna Kea)

A NOAO-funded wind-flow modeling analysis of Chajnantor sites for Cornell. by Dr D. De Young (NOAO) will serve to characterize the wind flow patterns for the high altitude sites chosen by Cornell as possible locations for a large IR-optimized telescope. These analyses, based on radar imagery, will be invaluable in our own evaluation of these sites for the NBT, and as a model for analyses of other sites. See De Young & Charles (A.J., 110, 3107, 1995). An analysis of the Gemini site on Mauna Kea is contained in the above reference. A similar analysis should be carried out for the Mauna Kea NBT site, if it can be identified at this stage. Due to the large amounts of human and computer time needed for these analyses, it is feasible to analyze only a few sites in this way.

Initial Survey (Chile)

A ground and/or air visual survey of the final sites needs to be made, to ascertain access, structure, and proximity to any mining activity. 

Mauna Kea atmosphere structure

The NBT will be a telescope critically dependent on Adaptive Optics in order to achieve its science goals. This will require detailed knowledge of the heights and strengths of the turbulent layers in the atmosphere, and their variability on time-scales ranging from hours to months. We will comprehensively evaluate the Mauna Kea atmosphere via two complementary approaches, and then later use the same techniques to characterize the possible NBT sites in Chile.

The first approach is to build upon the expertise of. Dr R. Knabb, a research meteorologist employed by IfA to provide weather (including seeing) forecasts for Mauna Kea by modeling meteorological conditions with the MM5 analysis program.The delineation of the turbulent structures is restricted by the limited vertical resolution of the model grids, a limit imposed by available computer power. This is only an issue when rapid forecasting is the goal. We will therefore run model predictions using a much finer vertical grid, and compare the model to reality via a series of balloon launches, either from the summit of Mauna Kea or from Hale Pohaku. The anticipated improvements to the model resulting from this direct comparison should then give confidence that it can predict the atmospheric structure realistically. The model, in the low vertical-resolution mode used for forecasting, provides an estimate for the integrated vertical turbulence, converted to a fwhm of a star image. The best way to compare this prediction to that actually observed is by use of a DIMM (Differential Image Motion Monitor, see http://www.astro.washington.edu/rest/dimm/). Mauna Kea does not have a DIMM, and plans by IfA to install one have not been assigned much priority. We will provide a DIMM similar to the portable DIMMs we have assembled for use in Chile, this will also allow direct comparison between possible Chile sites for the NBT and Mauna Kea for this parameter.

The second approach will be to measure the vertical structure of the atmosphere directly using SCIDAR (Vernin et al. 1990, Radio Sci. 25, 953).Generalized SCIDAR (gSCIDAR, Fuchs et al. 1994, Proc. SPIE Vol. 2222, 682) has extended the technique to also provide structure information below 1000-m. SCIDAR characterization is normally run in campaign mode, generally by the U. Nice group, since it requires use of a special camera on an already existing telescope of at least 1-m aperture. This is inappropriate for an extensive site-testing campaign, and so we will investigate whether it is possible to construct a simple portable system that does not require use of an existing facility. Such an instrument would also be of great value to any existing AO system, over and above our use as a site characterization tool. If the SCIDAR plus telescope technique proves difficult or impossible to be made portable, we propose to make SCIDAR measurements at Mauna Kea in the conventional way, using an existing telescope in campaign mode.

Complementary approaches to the use of SCIDAR to completely probe the atmosphere are possible. High frequency "mini-SODAR" (see http://www.a-research.com.au/) can probe the atmosphere from a height of a few meters to a few hundred meters, and the conventional low-frequency SODAR covers 50-1000m. Radio frequency sounding can probe to a height of several km. For most sites, perhaps all, the most important turbulent layer lies within 1000m above the site, and SODAR can effectively characterize this. In addition, the high-frequency SODAR is capable of measuring turbulence between 5-200 m, and thus can determine the height of the ground layer, the region of turbulence due the interaction between the wind and the local topography. All telescope optics will need to be placed above this layer.

Knowledge of the outer scale of turbulence, and its range of variation, is a critical parameter in the design of an active and adaptive optics system. This can be measured by an array of telescopes with a range of (small) apertures, all simultaneously recording the scintillation. A Shack-Hartmann camera can also provide the outer scale, and we will investigate whether this is possible to incorporate such a camera along with the portable SCIDAR camera described above.

Site Test equipment

Equipment purchase began in CY1999. Three stand-alone weather stations with wind, humidity, solar flux and temperature sensors have been assembled and are presently being calibrated and tested. Three portable DIMMs, based on that developed by ESO, with software from U. Washington, are about to undergo testing on Cerro Tololo. One DIMM will remain on Tololo. We will participate in late CY1999, using this equipment, with the Cornell testing campaigns of peaks in the Chajnantor region.

During CY2000 we plan to purchase SODAR equipment, which will be invaluable for initial site characterization. Present indications are that the high-frequency SODAR may well be able to replace a vertical tower and micro-thermal sensors, the use of which imposes severe logistical challenges, at least for the initial testing of sites. Also in CY2000 we will investigate the feasibility of building a fully portable SCIDAR camera and telescope.

Design studies for the portable SCIDAR monitor mentioned in sec 3.6 begin in CY2000 and continue through CY2001. Such a portable telescope might be equipped with a SCIDAR, mesospheric sodium monitor, and an outerscale monitor and would provide not only critical design parameters but also could be used during the operational phases for queue scheduling and optimization of the AOS.

Deliverables at end of CY2000

By the end of CY2000 we will have completed the following milestones:

a)Begin the detailed analysis of the Mauna Kea site including measuring the vertical temperature and turbulence profiles.
b)Identify candidate sites in Chile based on cloud cover, remoteness, and climatology
c)Purchased the necessary primary monitoring equipment for sites in Chile. This includes purchase and testing of the DIMMs and SODAR. Development of the portable SCIDAR telescope will begin.

Mauna Kea characterization

In order to provide as long a timeline as possible, we will continue to evaluate the conditions on Mauna Kea as outlined above.We will make every effort to use equivalent techniques and instrumentation so that direct comparison can be made with Chilean sites, including those already developed.

Basic meteorological characterization (Chile)

Weather stations will be installed at the sites for a few weeks or months. Some sites will be accessible by 4WD vehicle, but others will necessitate hiring of a local contractor plus mules. The weather stations are stand-alone, with data downloaded every few weeks.. The data will be analyzed for basic weather quality.

Climatology analysis(Chile and Mauna Kea)

An analysis should be made high altitude wind (jet stream) behavior and other remote sensing products to determine general trends as function of latitude and longitude. The jet stream can play an important role by setting the sampling frequency of the AOS. This information is presently being archived for Chile and Mauna Kea..

Site quality and ownership (Chile)

A preliminary geological survey will be carried out, in order to eliminate sites with unacceptable faults or rock quality. An initial title search for ownership and mining rights will be made, together with an evaluation of any mining activity in the vicinity of the site.

Modeling(Chile)

Digital maps for each site will be obtained and wind model studies undertaken. These will be compared and calibrated with actual measurements. The same maps will be used to calculate sky brightness from the nearest population centers, with extrapolation into the future using growth models. In addition, detailed meteorological modeling of potential sites will be used to identify sites of low cloud cover, low water vapor, low winds, and potentially good seeing.

Deliverables at end of CY2001

By the end of CY2001 the following milestones will be completed:

a)Complete the first year of a detailed analysis of the Mauna Kea.
b)Obtain in-situ weather measurements at candidate sites in Chile.
c)Complete an initial geological survey of candidate Chilean sites.
d)Complete and begin procurement of portable SCIDAR instrument.
e)Refine candidate site list based on meteorological and climatological modeling of sites.

With the results of the CY2001 studies, together with the scientific, structural, operational and financial constraints, the number of possible NBT sites will be pared down to 1-3. These will be tested extensively, and compared with Mauna Kea and the already established Chilean sites. Ideally these tests should extend beyond the nominal CY2005 used here, and for the prime site should extend until construction of the facility commences.

Nighttime campaign of seeing measurements (Chile and Mauna Kea)

This will use the DIMMs. Results will be compared directly with identical or near-identical DIMMs on Cerro Tololo, Cerro Pachon, Las Campanas, La Silla, Cerro Paranal, and Mauna Kea. Simultaneous meteorological monitoring will continue. Given the likely remoteness of the Chile NBT site, a power container will be purchased to support the testing effort.

Detailed analysis of the atmosphere (Chile and Mauna Kea)

We will characterize the vertical structure of the atmosphere in detail, in order to evaluate sites for AO. This has been described in some detail in 3.6 above, and we plan that by the time we are testing the prime Chile site(s) we have available equivalent measurements for Mauna Kea so that critical comparisons can be made. Possible techniques include balloon soundings, SCIDAR, and SODAR.

The outer scale of the turbulence is anticipated to be of order the diameter of the telescope. This will play an important role in the design of the active and adaptive optics systems. For the primary site in Chile we will monitor the outer scale of the turbulence either using a wavefront sensor on the end of the portable telescope, by coupling two DIMMs or by contracting with a group with a Generalized Seeing Monitor (Martin et al. 1994, A&AS, 108, 1).

Modeling of site wind-flow characteristics in detail (Chile and Mauna Kea)

A detailed analysis will be made of the sites with particular attention on the choice(s) of building structure and specific location.See De Young, AJ 112, 2896, 1996.

Property and Environmental issues (Chile and Mauna Kea)

A detailed and discreet study of the property and mining rights will need to be made of these final Chilean sites. A full geological survey of the final site will be conducted. Site purchase may have to be made. The major issue on Mauna Kea is the environmental impact of an NBT. 

Sodium layer monitoring

For the primary site(s), the variation of sodium in the mesosphere will be monitored. The latitude dependence of the sodium column density is well known so we do not anticipate that sodium monitoring is required for many sites. However, a statistical data set of the seasonal and most importantly the minima of sodium column densities can set a strong constraint on the laser guide star requirements. Depending on the outcome of the sodium monitor design study in CY2001, we will monitor the sodium column density throughout the nighttime seeing campaign.

Deliverables at end of CY2002

By the end of CY2002 the following milestones will be completed:

a)One year of night-time seeing campaign at Chilean sites.
b)Begin monitoring of detailed vertical structure using SCIDAR, SODAR, and/or balloon soundings at prime Chilean sites.
c)Begin outerscale measurements of primary site.
d)Detailed wind flow modeling for building placement at sites. e)Begin sodium monitoring campaign.