Optical Turbulence

The following sections are included:

• PREFACE

• CONFERENCE GROUP PICTURE

• Organizers

• IN MEMORY OF ANDRE ERASMUS

• SCIENTIFIC PROGRAM

• CONTENTS

All astronomical observations are done best from space where the absorption by and turbulence in the Earth atmosphere are absent. One has access to the entire electromagnetic radiation spectrum and the absence of seeing allows unlimited angular resolution. However, the cost of facilities in space and their operation is 3 orders of magnitude of similar sized facilities on Earth. Experimental astrophysicists have therefore in the past decades pursued the development of techniques to overcome the seeing limitations by the atmosphere. So far they have been very successful at this and much more is almost certain to come. Adaptive Optics (AO) will make very large (8 - 10-meters diameter) and extremely large (30 – 42 meters diameter) telescopes diffraction limited first at infrared wavelengths and eventually at visible wavelengths. The development of fast optical turbulence/seeing wavefront sensing techniques using artificial sources (Laser Beacons) will enable doing that over the entire sky. Atmospheric Tomography (AT) needed for Multi-Conjugate Adaptive Optics (MCAO) will give 3D maps of the rapidly variable atmospheric turbulence. Large interferometers with baselines of hundreds of meters will further enhance the angular resolution using fringe tracking for both co-phasing and coherent operation. Ground-based astronomy is therefore entering a new era in which milli-arcsecond observations and better are foreseen even of objects at the edge of the universe. The astronomical techniques will result in information of atmospheric optical turbulence which is likely to be of interest for meteorologists.

Optical turbulence is a key determinant of an astronomical telescope science performance. Forecasting optical turbulence at altitude is thus of paramount importance for astronomy. And so are the understanding and management of ground layer and locally induced turbulence. This paper presents an astronomer's view of both the futility and the utility of various site characterization activities, of the challenges they pose and of dreams one has about understanding and forecasting site qualities. It is pointed out that, for likely suitable sites, the average integrated optical turbulence can be calculated to good accuracy by a simple model and that turbulence profiles differ less between sites that between nights or times at a given site. The challenges therefore stem from the temporal and spatial variability of the turbulence. This variability is illustrated and briefly discussed. Collaboration between astronomers and atmospheric physicists must hold the key to the astronomers' dream of knowing in detail what the optical turbulence at the site will be tomorrow night, or next week, and which science program will then make best use of the facility.

The scintillation of point-like objects is primarily caused by thermal fluctuations in the upper atmosphere. For it the scintillation index (σ_I²) is proportional to the height integral of C_n²(h) weighted by a height dependent function F(h) = h^α where α = +5/3. For extended objects like the Sun or the Moon the height contribution to the (much smaller) scintillation is quite different. Because of their size the effects of the optical turbulence is averaged over an ever increasing area as the distance to the detector increases. Assuming vertical viewing, the area diameter increases like h*Ω where Ω is the angular diameter of the Sun or Moon. For Kolmogorov turbulence the function F(h) then still has the same shape, but with α = -1/3 so that the lower layers contribute more to scintillation. This makes it a good tool for the probing of the lower atmospheric layers. Unlike the σ_I² for stellar scintillation, the σ_I² for the Sun and the Moon is wavelength independent. Using an array of scintillometers one can probe the C_n²(h) distribution of those lower layers in a technique called SHABAR. SHABARs have been used in site testing for lower atmosphere probing for solar and nighttime telescopes. The aim is to establish the height to place telescopes, like the Advanced Technology Solar Telescope (ATST), to minimize boundary layer seeing effects. SHABAR site tests using the Moon are planned both for Arctic sites (Hickson's contribution to this meeting) and Antarctic sites (Storey's contribution to this meeting) where boundary layer heights are very site dependant reaching sometimes very small values. In my contribution I described some of the solar results related to the ATST site testing. The scintillation of planets have an F(h) function different in shape from that of the Sun or Moon. For low heights, where their beams still are narrow, F(h) has an α of +5/3 (as for stars); for large heights it is -1/3 (as for the Sun & Moon). For Mars the height contributions F(h) for seeing and scintillation are similar.

We report initial results from a program of observations of boundary-layer turbulence at Cerro Tololo, Chile, using a 12-element lunar scintillometer. Observations were conducted on 75 clear nights over a period from July 2007 to June 2008, from a location near the northern end of the summit area. The median seeing FWHM at a height of 2.5 m above ground was found to be 0.60 arcsec. In median conditions, 50% of the contribution of the ground layer to the turbulence integral originates in the first 400 m of the atmosphere.

Since a few years measurements of the optical turbulence vertical distribution have been done at Mt. Graham with a Generalized Scidar (GS) located at the focus of the 1.75 m Vatican Advance Technological Telescope (VATT). Such a telescope is placed on the summit of Mt. Graham (Arizona) at around 250 m far away from the Large Binocular Telescope (LBT). Thanks to a new technique¹ that we recently proposed based on the GS observations of wide-binaries (30-35 arcsec), measurements of profiles characterized by a vertical resolution as high as 20-30 m in the first 600 m have been also collected. The statistic sample of measurements consists, at present, of 43 nights distributed in different periods of the year. In this contribution we present the main scientific motivations as well as the analysis of this extended survey and new insights into the turbulence characterization achieved so far by this on-going activity.

The principle of a differential Optical Turbulence Profiler based on Angle-of-Arrival statistics as the SLODAR is presented. This instrument is well-adapted to study the terrestrial atmosphere boundary-layer in daytime and nighttime conditions.

SLODAR turbulence monitors have been installed and operated at the Cerro Paranal, Mauna Kea and SAAO Sutherland observatories. The instruments, developed at Durham University, provide real-time measurements of the atmospheric turbulence strength, altitude and velocity, for site characterization and for real-time support of adaptive optics for astronomy. We present sample results and compare contemporaneous data obtained with SLODAR, MASS and DIMM monitors at the ESO Paranal observatory.

We have made high order (32×32 subaperture) Shack-Hartmann wavefront sensor observations of binary stars with separations of approximately 20 arcseconds using the University of Hawaii 2.2 m telescope. We present preliminary results of a Slope Detection and Ranging (SLODAR) analysis of the data yielding measurements of turbulence strength, wind velocity and velocity dispersion as a function of altitude, with approximately 500 m vertical resolution. The aim of the investigation is to explore the validity of the Taylor frozen flow approximation and the implications for layer-oriented predictive AO reconstruction algorithms.

We present the statistical results of the monitoring of the optical turbulence profiles at the Roque de los Muchachos (ORM) and Teide observatories (OT), Spain. The data has been obtained using the Generalized-SCIDAR technique at the 1m Jacobous Kaptein telescope at the ORM and 1.5m Carlos Sánchez telescope at the OT. Statistical profiles are based on 150 nights of measurements recorded at the ORM from February 2004 to July 2008 and 140 nights at the OT obtained from November 2002 to July 2008. The monthly statistical profiles for different years present a similar structure for different years, suggesting a stable seasonal evolution of the turbulence at the Canary Islands sites. Statistically, the turbulence structure at ORM and OT is similar and present a similar evolution along the seasons. The consistence of the behaviour of the turbulence in different time scales is crucial for AO/MCAO systems implementation.

profile monitoring usually make use of wavefront slope correlations or of scintillation pattern correlations. Scintillation is rather sensitive to high turbulence layers whereas wavefront slope correlations are mainly due to layers close to the receiving plane. Wavefront slope and scintillation correlations are therefore complementary. Slopes and scintillation being recorded simultaneously with a SHWFS, we propose here to exploit their correlations to retrieve the profile. The measurement method named COupled SLodar scIDAR (CO-SLIDAR¹) uses correlations of SHWFS data from two separated stars. CO-SLIDAR method is presented with results of estimation from simulated SHWFS. Preliminary results with CO-SLIDAR are presented using real data from a binary star.

The Antarctic Plateau offers many benefits to astronomers, including dark and transparent infrared skies, long periods of uninterrupted observations, and very low levels of atmospheric turbulence. Efforts to quantify these benefits are ongoing. Characterizing the turbulence is particularly challenging, and requires a different approach to that commonly used at temperate sites. First, the atmosphere has two quite distinct regimes: a free atmosphere that is largely devoid of turbulence, and a thin but highly turbulent stable boundary layer. Second, if heat is used to avoid frost formation on optical surfaces, local turbulence might inadvertently be created by the instrument trying to measure it. In this paper we review the work that has been performed to date, and discuss what is required to advance our understanding of the Antarctic atmosphere.

The Thirty Meter Telescope (TMT) project has undertaken a large program to select the final site for the telescope. Five candidate sites were selected in Chile, Mexico and Hawaii for testing, and a site testing station was erected on each mountain. The site testing station consists of an extensive suite of instruments that operate robotically on each of the candidate sites. The systems are all monitored and the data archived to a central server system. Testing began in April, 2003, and each site testing station was installed for at least three years. The project is now nearing completion, with the final site selection decision scheduled for mid-2009.

The Thirty Meter Telescope Project has been testing several mountains in Chile, Mexico and Hawaii to determine the eventual placement of the telescope. The site testing project has operated for several years, and employed a variety of methods to measure the suitability of each of the sites. Here, we discuss the calibration of instruments and methods and some top level results from the candidate sites.

In the frame of the EU funded ELT Design Study, a wavefront characterization campaign has been conducted at Paranal VLT Observatory in December 2007 involving concurrent measurements with DIMM, MASS, GSM, MOSP, LuSci and SCIDAR. The campaign goals and instrumentation are described and preliminary results are presented.

Atmospheric turbulence outer scale is a relevant parameter for High Angular Resolution techniques as Adaptive Optics and Interferometry. This paper reviews in a non-exhaustif way different techniques and instruments for the measurement of the wavefront coherence outer scale which should be distinguished from the in-situ geophysical parameter.

In the framework of the Extremely Large Telescope design study, the Work Package (WP) 12000 is studying the Site Characterization for an European Extremely Large Telescope. In particular, INAF is in the WP 12300 group for the Large scale atmospheric properties study. Previous studies done in many astronomical sites have been optimized on spatial scales comparable with 3m-4m to 10m class telescopes. The strong interest of the Astronomical Community in giant telescopes imposes a different site characterization opportune for 30-40m class telescopes. One of the central point in the Adaptive Optics for Extremely Large Telescopes is given from the achievable sky coverage. Generally speaking, sky coverage is dominated by the high altitude layers correction. In other words ground layer adaptive optics has a sky coverage much larger than other kind of corrections. That means that ways to improve the sky coverage in the sensing of high altitude layers can be very effective in terms of overall performances. Moreover, there are good reasons to translate high coherence time of flowing layers, in a generalized Taylor assumption, into larger sky coverages. This paper presents the optical design of TOE, The Onduline Experiment, a WaveFront Sensor for sensing a Very Large Field of View on-board the VLT and possibly other telescopes as GranTeCan in Canary islands. Such a WFS is to be intended as a tool to probe the atmospheric parameters in the free atmosphere (i.e. far from the ground layer) on a linear scale of the same order of magnitude of the diameter of the ELTs currently in the design phase.

The wind pattern over Paranal changed suddenly in 1998 when winds from the NNE-NE became much more frequent than before. At the same time the seeing measured by the Paranal DIMM became significantly worse than hitherto. The seeing change, however, was not seen in 1999 when the first VLT Unit Telescope (UT) was commissioned, and while the DIMM seeing continued to deteriorate in subsequent years, the image quality delivered by the UT's remained constant, or even improved with time. In this contribution we describe our attempts to understand the origin of the discrepancy between DIMM and UT's, and the reasons for the dependency of wind direction on DIMM seeing.

No abstract received.

Data assimilation is the process by which observations are inserted into a forecast model to prevent it from drifting away from the true state of the atmosphere. It is not a trivial process and involves specific observations treatment as well as errors statistics definition for both model and observations. In the context of fine scale forecast, useful for predicting parameters such as optical turbulence, appropriate data assimilation system can be designed, with high resolution observations such as radar data and synoptic observations. That kind of system is used for a few years in Meteo-France with an operational status, and it proved us that the maximum improvement is mostly due to the high density of small scale observations.

This article describes how the simple ISBA^2,3 soil vegetation atmosphere transfer scheme has been adapted to polar conditions of Antarctica to retrieve surface and deep temperatures of polar cap. The main goal of the study is to improve the initialization of surface temperature of the French mesoscale model meso-NH.^4,5

The Antarctic Mesoscale Prediction System (AMPS) effort is one by the United States to provide real-time, high-resolution numerical weather prediction (NWP) over Antarctica. AMPS is a mesoscale modeling system run at the U.S. National Center for Atmospheric Research (NCAR) that generates twice-daily forecasts for the U.S. Antarctic Program. As AMPS and its products have been adapted over the years to provide NWP guidance over Antarctica, supporting ground-based astronomy may be a possible extension of the system. Presented are an introduction to AMPS, its capabilities, and its potential for support of weather forecasting for ground-based astronomy in Antarctica. Limitations are also discussed.

The characterization of the optical turbulence (OT) done with meso-scale models for astronomical applications is an alternative approach to this science that intrinsically presents some interesting and complementary features/advantages with respect to the characterization done with measurements.

The most important advantages are namely: (1) the possibility to describe a 3D map of the in a region around a telescope, (2) the possibility to forecast the optical turbulence i.e. to know with some hours in advance the state of the turbulence conditions above an astronomical site and (3) the possibility to perform a climatology of the optical turbulence extended over decades. No other tool of investigation with comparable potentialities can be figured out at present to achieve these 3 scientific goals.

The forecast of the optical turbulence is a fundamental requirement for the optimization of the management of the scientific programs to be carried out at ground-based telescopes foci. Ground-based astronomy will remain an appealing option for astronomers with respect to the spaced-based one only if the telescopes management will be performed taking advantage of the best turbulence conditions. The future of new ground-based telescopes generation relies therefore upon the success of these studies.

ForOT is a scientific project but even more, it identifies a philosophic approach to studies related to the characterization of the optical turbulence in an astronomical context. In this contribution we will deal about the main success obtained so far in this discipline in the past, the new goals predetermined by ForOT and the main results we obtained so far. To conclude, we will trace a perspective at long time scale indicating where our research is addressed to and how the scientific community and the managers who lead the ground-based astronomical facilities can support our researchers to progress in these studies.

This paper briefly describes the current methods employed at the Mauna Kea Weather Center in the development of optical turbulence and seeing forecasts for the observatories at the summit of Mauna Kea. This paper reports the content of two talks (Businger et al. and Cherubini et al. – see 'Scientific Program, Session 5') given to the International Conference "Optical Turbulence – Astronomy meets Meteorology".

Mesoscale model such as Meso-NH have proven to be highly reliable in reproducing 3D maps of optical turbulence (OT).^1–3 These last years ground-based astronomy has been looking towards Antarctica, especially its summits and the continental plateau where the OT appears to be confined in a shallow layer close to the surface. However some uncertainties remain. That's why our group is focusing on a detailed study of the atmospheric flow and turbulence in the internal Antarctic Plateau. Our intention in this study is to use the Meso-NH model to do predictions of the atmospheric flow in the internal plateau. The use of this model permits us to have access to informations inside an entire 3D volume, which is not the case with observations only. Two different configurations of the model have been used: one with a low horizontal resolution (ΔX = 100 km) and another one with higher horizontal resolution with the help of the grid-nesting interactive technique (ΔX = 1 km in the innermost domain). The impact of the configuration on the meteorological parameters has already been studied.⁴ We present here the results obtained with Meso-Nh of forecasted profiles, surface layer thickness (SLT) and seeing values at Dome C for the 16 winter nights, whose profiles have been measured by Ref.5.

In the north of Chile three of the five pre-selected candidate sites to host TMT are located; namely, these were Cerro Armazones, Cerro Tolar and Cerro Tolonchar. The TMT site testing program conducted measurements of the meteorological conditions and the optical turbulence strength throughout the atmosphere (the turbulence profiles) above these sites. A peculiar case of turbulence occurrence above these sites happened during the night of August 13, 2006. We will show a study on the meteorological conditions above these sites adressing the origin of this particular turbulence behaviour.

A synoptic analysis for this night shows anticyclonic predominance (AP) and Jet Stream (JS) activity at the high levels of the atmosphere. Simulations using the MM5 mesoscale model are performed and compared to the measured meteorological site conditions. Turbulence above Tolonchar seems be associated to the Jet Stream, but the turbulence above Tolar can not clearly be associated to the synoptic conditions. Also trajectories were calculated for each site at altitude levels of the turbulence profiles. The low level trajectories for Tolar show the origin of air during this night being South-East of the central valley, following the orography. For Armazones the low level trajectories originate from North-East. The flow which affects Tolonchar is strongly influenced by the western circulation which is present at these intermediate altitudes at around 500 hPa. Simulations based on MM5 show clear signs of a local circulation for each site, but to improve the results it will be necessary to increase the spatial resolution of the models in order to trace turbulent processes, follow their evolution and investigate their relation with good and bad seeing.

A study of the ground layer turbulence as seen by the combined measurements by MASS and DIMM during the TMT site survey is presented. The TMT site survey data include also the necessary meteorological parameters to infer important micrometeorological quantities. Several cases are being presented demonstrating the regime in which small scale local effects dominate the seeing within the first few hundred meters of the atmosphere.

This contribution shows some results obtained analysing a long time series of meteorological data taken at two of the most important astronomical sites in the world: the site of Observatorio del Roque de Los Muchachos (ORM, Canary Islands) and the site of Paranal Observatory (Chile). The aim of this analysis is to derive correlations between synoptic parameters and the physical properties of the atmosphere in both short and long time baseline in order to have the best scientific result from the astronomical observations.

In this contribution we briefly summarize the results recently published in Ref.1.

The atmosphere above three sites on the Internal Antarctic Plateau (Dome A, Dome C and the South Pole) is investigated for astronomical applications using the analysis-data from the ECMWF (European Centre for Medium-Range Weather Forecasts) for an entire year (2005). The monthly median of several meteorological parameters, such as wind speed and potential temperature, are calculated. Radiosoundings from Dome C and the South Pole are used to verify the reliability of the analyses and to study the wind speed in the first 100 m as the analysis-data are not optimized for this altitude-range. The wind speed in the free atmosphere is obtained from the ECMWF-analyses for all three sites. In this context a fourth site, Dome F, is also discussed. In the free atmosphere the stability is studied using the Richardson number, which is an indicator of the probability to trigger thermodynamical instabilities. We find that the free atmosphere over the Internal Antarctic Plateau is more stable than at mid-latitude sites. Dome C shows worse thermodynamic instability conditions than those predicted above the South Pole and Dome A in the same vertical slab. Finally we provide a ranking of the three sites with respect to wind speed, in the free atmosphere (ECMWF analyses) as well as in the surface layer (radiosoundings)

The TMT site testing campaign includes the measurement of the ground layer turbulence profiles using SODARs. In this part of the atmosphere, turbulence is caused by the thermal exchange between the ground and the air above it. A complete knowledge of the energy balance can therefore be related to the turbulence profiles. With the measurement of the different heat fluxes, we were able to find a correlation between the kinetic heat flux and the integrated turbulence measured by the SODAR. In this paper we present these results and discuss their limitations.

The Roque de los Muchachos Observatory (ORM) at La Palma (Canary Islands) is one of the two top pre-selected sites for hosting the future European Extremely Large Telescope (E-ELT), the other ones are Ventarrones (Chile), Macon (Argentine) and Aklim (Maroc). Meteorological and seeing conditions are crucial both for the site selection and for telescope design and feasibility studies for adaptive optics. The ELTs shall be very sensitive to wind behavior when operating in open air, therefore ground level wind velocity and wind gust are also required for the feasibility of the telescope construction. Here we analyze the wind speed and wind direction, the air temperature, the relative humidity and the barometric pressure statistical results obtained from data recorded at different sites at the ORM by several Automatic Weather Stations (AWS) since 1985, day and night time separately. Ground wind speed regimes (775mbar) are compared with those provided by satellites from 200 to 700mbar. There exists also observational evidence of the correlation between the seeing and the wind speed and wind direction that will be discussed in this work.

The height of the atmospheric boundary layer on the Antarctic plateau is of particular interest to meteorologists and designers of optical telescopes for Antarctica. "Snodar" was developed at the University of New South Wales to measure the height of the boundary layer at Dome A and Dome C on the Antarctic plateau. Snodar operates between 3 kHz and 14 kHz and has a vertical resolution of 1 m or better. Snodar was deployed to Dome A in 2008 and was found to have sufficient sensitivity to be able to measure the height of the atmospheric boundary layer with a single pulse. We present here the technical details of a second-generation Snodar, which will be deployed to Dome A in 2009.

This paper provides an overview of an invited presentation on the astronomical capabilities provided by existing and future adaptive optics (AO) and interferometer facilities, and the limitations and design constraints imposed by atmospheric turbulence.

We present a new concept for an optical differentiation wavefront sensor, featuring a high number of phase measurements across the pupil, with a linear response versus the phase gradient. We present measurements obtained in the lab with this sensor though turbulence on a longitudinally-elongated laser spot.

The CFH Telescope has a long history of wide-field imaging, and this is reflected in its users community. Many projects are competing in the A · Ω race but so far, all are limited by atmospheric turbulence. Wisely choosing a site with a very thin ground layer and excellent free-atmosphere seeing would allow very wide fields to be corrected by a GLAO system.^1,2 The characteristics of the turbulence of Mauna Kea are perfectly matched to such an instrument making the site an integral part of its design. The goal of this project is to achieve exquisite image quality over the largest possible field of view, with a goal of a FWHM of not more than 0.3" over a square degree field in the optical domain. This goal depends crucially on the thickness of the ground layer and its prevalence. But over such large fields the probability of finding sufficiently numerous and bright natural guide sources is high, although a constellation of laser beacons could be considered to ensure homogeneous and uniform image quality. The image is then limited by the free atmosphere seeing (0.2" to 0.4"). This can be further improved by an Orthogonal Transfer CCD camera (Tonry et al.³) which can correct local image motion on isokinetic scales from residual high altitude tip-tilt. Knowing the vertical turbulence profile, the outer scale and the isokinetic patch plays a critical role at the design level.

Multi Conjugated Adaptive Optics, thanks to the Multi Conjugated Adaptive optics Demonstrator, MAD, of European Southern Observatory, entered in the realm of producing astronomical science after more than a decade. I review our experience with the Layer Oriented Wavefront Sensor paying special attention to the practical influence of the encountered atmospheric turbulence on the achieved science. An outcome, maybe very personal, of the perspective in the near and not so near future are outlined.

Extremely Large Telescopes (ELTs) will have aperture diameters up to 42 meters. Adaptive Optics (AO) at short wavelengths (< 1 micron) will be very hard to implement at these wavelengths because of the limited number of actuators on state-of-the-art deformable mirrors and because of the limited brightness of Laser Guide Stars (LGSs). For 1 arcsec seeing at 500 nm wavelength deformable mirrors (DMs) with about 150000 actuators will be needed and LGSs of a brightness of V = 8 to 9. That exceeds our present capabilities by a factor of about 100 and 2 magnitudes respectively. One might expect both to improve with time. We propose to combine the techniques of "pupil slicing" and AO to sharpen the telescope images at short wavelengths to the size of Airy disk of the pupil slices. I refer to this technique as "Pupil Slicing Adaptive Optics" or PSAO. At 500 nm wavelength that would correspond to the Airy disk of an approximately 5 meter diameter aperture, or a FWHM of 0.02 arcsec. As DMs increase in their number of actuators, the size of the pupil slices increases thus improving the angular resolution. Ultimately the full angular resolution of, for example, a 42 meter aperture would be reached (0.0024 arcsec at 500 nm). Of course, this does not resolve the issue of the limited brightness of LGSs. For it one has to wait for more powerful lasers and the development of perspective elongation correction techniques. Alternatively one would accept limited sky coverage (0.1%) when using natural guide stars (NGSs). Particularly interesting is the PSAO technique for high resolution spectroscopy where the smaller image sizes even for many slices results in a significant decrease in spectrograph dimensions.

Note from Publisher: This article contains the abstract only.

The operations model of the ESO Very Large Telescope (VLT) heavily relies on a full-scale implementation of Service Mode observing. In this contribution we review the main features of ESO's approach to Service Mode at the VLT, we outline the advantages offered by this mode, and the challenges faced when implementing it given the wide diversity of instrumentation and instrument modes currently available at the VLT and the VLT Interferometer (VLTI). We give special emphasis to the part of this challenge directly derived from the evolution of the atmospheric conditions, which drive the short-term scheduling of the different scientific programmes competing for the available time.

The reconfigurable optics of LBT provides considerable versatility for observers to choose instruments and modes in near real time. Current values and predictions about both cloud cover and turbulence will become a critical factor in those choices, as the suite of available instruments expands with time. Two modes of operation are envisioned, both of which involve changes of focal station in response to conditions. One is partner-observer queue mode for facility instruments, which includes natural guide star adaptive optics. In that case, queue planning will use predictions and actual seeing to accommodate those programs requiring the best conditions. The other is a block-scheduling mode for new or complex capabilities, such as interferometry and early days of laser guide star operations. In that mode, site information will be essential for shifting to backup programs on nights that prove to be unsuitable for stable fringe tracking or reliable laser projection.

The advent of the giant telescopes such as the 42-m European Extremely Large Telescope (E-ELT), expected before the end of the next decade, is bound to produce new paradigms in the operations of astronomical observatories. It will also produce further evolution of many novel concepts that have been successfully introduced in the operation of the current generation of 8/10m-class telescopes, mainly those related to flexible scheduling. The integration of adaptive optics capabilities in the design of the E-ELT and the high performance made possible by new technologies in astronomical instrumentation have enabled very ambitious science cases that push the capabilities of the telescope and its instruments to the limits. However, the exploitation of such capabilities will require a careful planning of science operations aiming to fully optimize the use of the available observing time and the way in which scientists will interact with the facility. The characterization and short-term forecast of optical turbulence are anticipated to play a very important role in optimizing E-ELT operations. We review the most important features of science operations foreseen at the E-ELT, some requirements linked to the variety of adaptive optics modalities that will be available at the telescope and its instruments, and how the current design of the E-ELT facility and its operations model intends to respond to these needs.

Paranal is located in the north zone of Chile, which is considered a privileged zones for astronomical observation for his desert conditions, not much cloudiness and low light contamination. These same condition are present in the Pachón summit where is located the Gemini South observatory. Similar features have the zone of Macón in Salta region in Argentina. This place is one of the sites pre-selected for the construction of the E-ELT. The behavior of the atmospheric conditions is important within the astronomical observation because knowing his future state it would be possible program the instruments for the night of observation. To do this we use meteorological forecasts models to simulate the behavior of the atmosphere. The period in study was year 2005, where the atmosphere conditions were simulated with GFS and MM5 models. In order to perform the analysis of the simulation in those three sites, the meteorological variables data was interpolated using geographic coordinates of each weather station and compare them with the measured meteorological variables. In this report we evaluate the temperature and in the future we will evaluate: the relative humidity, atmospheric pressure, direction and intensity of the wind. In the first part of the analysis we found that the forecast of MM5 is better than GFS model, as expected. In the second part of the analysis a Kalman filter has been implemented to minimize the systematic errors of the simulation of GFS and MM5.

Seismicity induces ground vertical and horizontal displacements that could affect the image quality obtained by telescopes in a similar fashion than atmospheric turbulence. In this work, we study the effect of local seismicity relative to atmospheric turbulence upon the image quality of astronomical observations at El Teide observatory and Roque de los Muchachos observatory, both in the Canary Islands, Spain. Three different aspects of seismicity are studied, namely regional volcanism and seismicity (that is compared with other astronomical sites), seismic noise and possible resonances between seismic noise and the structure of telescopes.

The proper characterisation of the turbulence structure in an astronomical site requires an statistical study of the refractive-index structure constant . Our team is monitoring the profiles since 2002 at the Teide and Roque de los Muchachos observatories (Canary Islands, Spain) with the g-SCIDAR technique. We have compared the turbulence profiles obtained with the radiosonde simultaneous profiles and with the NCEP I reanalysis maps at different levels. We show that turbulence measured at both observatories (being at ~ 160km distant) correlate with the radiosonde data and may be explained through the atmospheric conditions in a synoptical scale. Hence, the statistical predominance of synoptical scaled phenomena may explain the similarity found in the monthly average profiles obtained at both observatories.

The requirements for excellent image quality of current large and future very large telescopes demand a proper knowledge of atmospheric turbulence. Thus several projects are already pursuing this aim. The precise characterization of the turbulence above a particular site requires long-term monitoring. In this sense, due to the lack of long-term information on turbulence, high-altitude winds (in particular winds at the 200-mbar pressure level, V₂₀₀) have been proposed (Sarazin & Tokovinin 2002,¹ S&T02 hereafter) as a parameter for estimating the total turbulence at a particular site, because records of this parameter exist from several sources. This choice is based on the idea that the greatest source for turbulence generation is related to the highest peak in the vertical wind profile, which is located at the 200-mbar pressure level globally. Moreover, S&T02 found a good correlation between the average velocity of the turbulence, V₀, and V₂₀₀ of the form: V₀= 0.4*V₂₀₀ at the Cerro Pachón and Paranal Observatories in Chile. A linear relationship between V₀ and V₂₀₀ was also found in San Pedro Mártir (Mexico) (Masciadri & Egner 2006,² M&E06 hereafter). We study here the possible connection between V₀ and V₂₀₀ at the Teide Observatory (Spain).

During winter and springtime, the flow above Antarctica at high altitude (upper troposphere and stratosphere) is dominated by the presence of a vortex centered above the continent. It lasts typically from August to November. This vortex is characterized by a strong cyclonic jet centered above the polar high. In a recent study¹ of four different sites in the Antarctic internal plateau (South Pole, Dome C, Dome A and Dome F), it was made the hypothesis that the wind speed strength in the upper atmosphere should be related to the distance of the site to the center of the Antarctic polar vortex. This high altitude wind is very important from an astronomical point of view since it might trigger the onset of the optical turbulence and strongly affect other optical turbulence parameters. What we are interested in here is to localize the position of the minimum value of the wind speed at different heights in the troposphere, particularly above 10 km. For that we studied the analyses from the ECMWF for winter 2005 at different levels. We deduced a preferential position of this minimum, tilted with altitude, in a zone between South Pole and Dome A, for the year 2005. This extensive study over one entire winter confirms the "position space" of the polar high deduced by Ref. 1.

A prototype of a new SLODAR instrument has been developed at Durham University CfAI and tested at the Paranal observatory. The instrument targets very wide double star targets, with separations of several arc-minutes, to achieve profiling of the surface layer of turbulence with very high resolution in altitude (10m or less). We describe the instrument and the results of preliminary observations made at Paranal.

Las Campanas Observatory has been designated as the location for the Giant Magellan Telescope (GMT). We report results obtained since the commencement, in 2005, of a systematic site testing campaign at LCO. Measurements of the turbulence profile of the free-atmosphere above LCO have been collected with a MASS/DIMM. We examine the contribution to the seeing arising from turbulence in the ground layer (defined here as below an altitude of 500 m) through the difference between the turbulence integrals in the full atmosphere (as measured by DIMM) and in the free atmosphere (as measured by MASS).

In preparation to characterize the Giant Magellan Telescope site and guide the development of its adaptive optics system, two campaigns to systematically compare the turbulence profiles obtained independently with three different instruments were conducted at Las Campanas Observatory in September, 2007 and January 2008. Slope detection and ranging (SLODAR) was used on the 2.5-m duPont telescope. SLODAR measures the profile as a function of altitude through observations of double stars. The separation of the observed double star sets the maximum altitude and height resolution. Ground layer (altitudes < 1 km) and free atmosphere turbulence profiles are compared with those obtained with a lunar scintillometer (LuSci) and a multi-aperture scintillation sensor (MASS), respectively. In addition, the total atmospheric seeing was measured by both SLODAR and a differential image motion monitor (DIMM).

AUTHOR INDEX