Observations of the nucleus of NGC 4151 at 2.2 μm using the two 10-meter Keck telescopes as an interferometer show a marginally resolved source less than or equal to 0.1 pc in diameter. These observations are the first measurement of an extragalactic source with an optical/IR interferometer. These observations represent a ten-fold improvement in angular resolution when compared to previous near-infrared measurements of AGN and make it possible to test the subparsec-scale, near-infrared emission models of NGC 4151.

Dusty tori have been suggested to play a crucial role in determining the physical characteristics of active galactic nuclei (AGN), but investigation of their properties has stalled for lack of high resolution mid-IR imaging. Recently, a long-awaited breakthrough in this field was achieved: NGC 1068, a nearby AGN, was the first extragalactic object to be observed with a mid-IR interferometer, thereby obtaining the needed angular resolution to study the alleged torus. The instrument used was MIDI mounted on the ESO's VLT interferometer. The resulting 8-13 micron interferometric spectra indicated the presence of a thick (3 x 4 parsec) configuration of warm dust surrounding a hot ~1 pc component, marginally elongated in the direction perpendicular to the main orientation of the warm component. The structure of the 10 micron "silicate" absorption feature hinted at the presence of non-typical dust. In this proceeding, first the field of AGN research is briefly reviewed, with an emphasis on models of dusty tori. Second, the general properties of the key object NGC 1068 are discussed. Third, the MIDI data set is presented together with a first attempt to interpret this data in the context of tori models. Fourth, preliminary MIDI interferometric spectra of the nucleus of the nearby starbursting galaxy Circinus are presented. The apparent observed absence of both a hot component as well as a sharp absorption feature suggest that we view the torus more edge-on than is the case for NGC 1068. Finally, we briefly discuss the prospects of ESA's Darwin mission for observing nearby and distant AGN. The required capabilities for Darwin's first goal -- the search for and subsequent characterization of earth-like planets orbiting nearby stars -- are such that for its second goal -- high resolution astrophysical imaging -- the sensitivity will be similar to JWST and the angular resolution 1-2 orders better. This will allow detailed mapping of tori of low luminosity AGN such as NGC 1068 up to redshifts of 1 - 2 and more luminous AGN up to redshift of 10 and beyond.

In April 2004, MIDI, the first of the two first generation instruments of the VLT interferometer, started official operation as a facility instrument on Paranal. It allows interferometric observations covering the full astronomical N band (7.8 μm - 13.5 μm). Initially, only observations with low spectral resolution λ/Δλ ~ 30 were offered. Examples for observations in this observing mode are presented in order to document the performance of the instrument. A number of new instrumental options are in preparation, which should render future observing with this instrument more sensitive, more accurate and more versatile.

The Keck Interferometer is a NASA funded project developed by the Jet Propulsion Laboratory, the William M. Keck Observatory and the Michelson Science Center at the California Institute of Technology. A technical description of the interferometer is given elsewhere in this volume. This paper will discuss the science topics and goals of the Keck Interferometer project, including a brief description of the Key Science projects, the science projects executed to date and the current availability of the interferometer for new projects. The Keck Interferometer Project consists of the Keck-Keck Interferometer, which combines the two Keck 10-meter telescopes on an 85-meter baseline, and the Outrigger Telescopes Project, a proposal to add four to six 1.8-meter telescopes that would work in conjunction with the two Kecks.

We present direct detections of the spatial extent of the circumbinary disk around HR 4049 and its companion. Observations were obtained with the ESO Very Large Telescope Interferometer using the VLT Interferometric Commissioning Instrument (VINCI) at 2 micron and the Mid Infrared Instrument (MIDI) between 8 and 12 micron. A single uniform disk model fit to the VINCI data gives an angular diameter of 27 milli-arcseconds. After taking into account the contribution from an unresolved central star we find that the observed visibilities indicate a second component with a spatial extent of 37 milli-arcseconds (which is identified as the circumbinary disk). The MIDI interferometric spectra show features which are due to PAH emission lines (8.6 and 11.3 micron). The visibilities of the emission lines indicate that the spatial extent in the lines (50 to 60 milli-arcseconds) is larger than in the continuum (35 to 45 milli-arcseconds). This leads us to propose a three emission components model to explain the interferometric observations: a central unresolved star, a 37 milli-arcseconds circumbinary disk and polar lobes emitting in the PAH bands with a size of 50 to 60 milli-arcseconds.

Atmospheric turbulence is a serious problem for ground-based interferometers. It places tight limits on both sensitivity and measurement precision. Phase referencing is a method to overcome these limitations via the use of a bright reference star. The Palomar Testbed Interferometer was designed to use phase referencing and so can provide a pair of phase-stabilized starlight beams to a second (science) beam combiner. We have used this capability for several interesting studies, including very narrow angle astrometry. For close (1-arcsecond) pairs of stars we are able to achieve a differential astrometric precision in the range 20--30 micro-arcseconds.

We present near infrared aperture synthesis maps of the well known binary star Capella (alpha Aur) produced with the upgraded IOTA interferometer on top of Mount Hopkins, Arizona. Michelson interferograms were obtained simultaneously on three interferometer baselines in the H-band between 2002 November 12 and 16. The simultaneous observation of fringes on three baselines permitted a single phase closure to be estimated along with three visibility amplitudes. Hybrid Mapping techniques were then used to reconstruct the source brightness distribution with a beam size of 5.4 x 2.6 mas, which allows for the resolution of the stellar surfaces of the Capella giants. The maps provide the first demonstration of imaging with phase closure on the IOTA instrument.

We review the theory of rotating stars, first developed 80 years ago. Predictions include a specific relation between shape and angular velocity and between surface location and effective temperature and effective gravity. Seen at arbitrary orientation rapidly rotating stars will display ellipsoidal shapes and possibly quite asymmetric intensity distributions. The flattening due to rotation has recently been detected at PTI and VLTI. With the increasing baselines available in the visible and the implementation of closure phase measurements at the NPOI it is now possible to search for the surface brightness effects of rotation. Roche theory predicts only large scale deviations from the usual centro-symmetric limb-darkened models, ideal when the stellar disks are only coarsely imaged as now. We report here observations of Altair and Vega with the NPOI using baselines that detect fringes beyond the first Airy zero in both objects. Asymmetric, non-classical intensity distributions are detected. Both objects appear to be rotating at a large fraction of their breakup velocity. Vega is nearly pole on, accounting for its low apparent rotational velocity. Altair's inclination is intermediate, allowing high S/N detection of all the predicted features of a Roche spheroid. We describe how these objects will test this fundamental theory and how Vega's role as a standard will need reinterpretation.

We present preliminary results from an ongoing survey for multiplicity among the bright stars using the Navy Prototype Optical Interferometer (NPOI). While the NPOI has previously concentrated on producing "visual" orbits of known close speckle and spectroscopic binaries, we have now embarked on a broader survey to detect new binary/multiple systems. We first present a summary of previous NPOI observations of known binary and multiple systems to illustrate the instrument's detection sensitivity for binaries at large magnitude differences over the range of angular separation detectable by the NPOI (currently 3 - 300 mas). We then discuss early results of the survey of bright stars north of declination -20°. This survey, which compliments previous surveys of the bright stars by speckle interferometry, initially emphasizes stars in a proposed Terrestrial Planet Finder (TPF) target list. To date, 29 of the 60 brightest TPF candidate stars (V ≤ 4.3) have each been observed on multiple nights. Preliminary analysis of these data indicates the possible detection of stellar companions to several of these stars.

The Palomar Testbed Interferometer has observed several binary star systems whose separations fall between the interferometric coherence length (a few hundredths of an arcsecond) and the typical atmospheric seeing limit of one arcsecond. Using phase-referencing techniques we measure the relative separations of the systems to precisions of a few tens of micro-arcseconds. We present the first scientific results of these observations, including the astrometric detection of the faint third stellar component of the kappa Pegasi system.

The coronagraphic techniques serving to reject most light from a star, when trying to image a nearby planet, can be pushed with an adaptive holographic element. Located after the coronagraph, it can in principle remove most of the residual star light by adding a phase-shifted holographic reconstruction of it . The scheme is also usable within each sub-aperture of a diluted hypertelescope array, sufficiently large to resolve details of an exo-Earth. A possible panoramic version of the previously mentioned Exo-Earth Imager is shaped as a virtual bubble of 400 km diameter , consisting of thousands of 3-meter mirrors, free-flying and arranged co-spherically. The half-size focal sphere is explored by beam combiners, one for each exo-Earth observed within tens of parsecs. Each beam-combiner includes a kilometer-sized corrector of spherical aberration at F/2, which is also diluted and consisting of small free-flyers. The instrument is expected to provide direct coronagraphic images of exo-Earths, resolved in 50x50 resels, with enough dynamic range obtained in 30mn exposures to search colored features and their seasonal variations, indicative of photosynthetic life .

Extragalactic Astronomy is one of the new domains being successfully addressed by the new generation of optical long baseline interferometers. The recent results on the unresolved Seyfert 1 nucleus NGC4151 by Keck Interferometer team and on the partially resolved Seyfert 2 nucleus NGC1068 by VLTI team confirm our current understanding of the Active Galactic Nuclei phenomenology. Once an extra level in increased angular resolution is reached by new observatories (e.g., Optical Hawaiian Array for Nanoradian Astronomy or Overwhelmingly Large Array), it will become possible to study the Broad Line Region of Seyfert 1 nuclei. In this paper, we present key results that interferometry could obtain on the topic, how they compare to an existing technique - reverberation mapping, and we describe a new observational technique that will lead to a tomographic picture of this region.

As a near-infrared (NIR) wide field interferometric imager offering an angular resolution of about 10 milliarcseconds LINC/NIRVANA at the Large Binocular Telescope will be an ideal instrument for imaging of galactic nuclei including the center of the Milky Way. Recent optical/IR imaging surveys can now quite successfully be used to search for star-galaxy pairs that are suitable for interferometric observation with LINC NIRVANA. These objects can then be used to efficiently investigate galaxy interaction, nuclear activity, and star formation in distant galaxies. In the NIR these investigations will be carried out at scales below 100~pc for z<0.05 and at scales below 500~pc at z<2. The Galactic Center measurements of stellar orbits and strongly variable NIR and X-ray emission from Sagittarius A* at the center of the Milky Way have provided the strongest evidence so far that the dark mass concentrations seen in many galactic nuclei are most likely super massive black holes. Observations with LINC NIRVANA will allow to simultaneously investigate the stellar dynamics of the entire central cluster, the determination of the amount of extended mass within the cusp region, and to monitor the activity of the 3 million solar mass black hole at the position of Sagittarius A* at separations of only about 10 light hours or 15 Schwarzschild radii.

Young Stellar Objects (YSOs) play a central role in the understanding of stellar and planet formation, and progress in spatial resolution and sensitivity of long infrared interferometers made such instruments particularly well suited to probe the inner part of the disk where essential physical processes and interactions are believed to take place. The first astrophysical results obtained on young stars arising from this technique are already giving a handful of informations about the structure of the extended component. However, model-fitting methods used to reduce the data prevent from definitive and unambiguous interpretations. Interferometric observations of Herbig Ae/Be stars is one of the most striking example. Whereas first results seemed to be in good agreement with accretion disk model, latest observations tend to favor the presence of a uniform ring with a inner radius set by dust sublimation temperature. Direct imaging of close environments around YSOs with infrared (IR) interferometers will allow to disentangle between current suggested models and to improve one step further the scenarios of stellar formation. Within this framework, we anticipate observations of YSOs with the VLTI and we investigate the potential of AMBER to recover images. Modelling their circumstellar environment, we simulate realistic observations of Herbig Ae/Be and TTauri stars. By using reconstruction technique specially dedicated to infrared interferometry and to sparse ($u,v$) data coverage, we analyze the quality of the recovered images, and we emphasize the critical points to take into account in the image reconstruction process. We conclude that it requires at least three nights of observations to perform imaging of YSOs with AMBER on the VLTI.

Interferometry with the Very Large Telescope Interferometer (VLTI) will allow imaging of the Galactic Center and the nuclei of extragalactic sources at an angular resolution of a few milliarcseconds. VLTI will be a prime instrument to study the immediate environment of the massive black hole at the center of the Milky Way. With the MID infrared Interferometric instrument (MIDI) for example the enigmatic compact dust embedded MIR-excess sources within the central parsec should be resolvable. Further the observations of external galactic nuclei will allow unprecedented measurements of physical parameters (i.e. structure and luminosity) in these systems. With the exception of a few 'self-referencing' sources these faint-target observations will benefit from the available off-axis wavefront-correction system STRAP, working on suitable guide stars (GS). To fully exploit the use of VLTI within this context, the following questions have to be addressed among others: How feasible is blind-pointing on (faint) science targets? Are VLTI observations still efficiently feasible if these faint science targets exceed the usual angular distance (≤1 arcmin) to a GS candidate, enabling a standard closed-loop tip-tilt correction? How is the fringe-tracking procedure affected in densely populated regions such as the Galactic Center? What preparatory steps have to be performed to successfully observe these non-standard targets with the VLTI? In this contribution, we present aspects for the preparation of VLTI observations, which will be conducted in the near future. Considering these example observations of the Galactic Center region, several details of observing modes are discussed, which are necessary to observe such science targets. The final goal is the definition of observational strategies that are optimized for the discussed classes of targets, which provide properties touching the limits of VLTI observability.

Over the last year, a set of well defined science requirements has been established for the Terrestrial Planet Finder (TPF) mission. They consist of top level specifications, such as the number and characteristics of stars to be observed, the planetary sizes and orbital phase spaces to be searched for, the desired completeness of the search, etc. For each of the concurrent observing techniques considered - thermal infrared nulling interferometry and optical coronagraphy-, dedicated spectroscopy requirements have also been formulated. On the interferometry side, the most promising design studied so far consists of a free flyer assembly of four 4m class telescopes. It basically allows to thoroughly search for planets in the habitable zone of ~ 160 nearby main sequence F,G and K dwarfs in 1 year of continuous integration (~ 2 years of operation). Over 1.5 year of subsequent observation, this design would also enable low resolution (20) spectroscopic characterization of up to 10 exo-planetary atmospheres in the [6.5 - 17] micron range, assuming 260K exo-planets with Earth albedo, and at least half the Earth area are present around the target stars. With only minor additions to the nulling design, and taking advantage of a spatial resolution 10 to 50 times higher than JWST, the free flyer design would also provide fantastic contributions in the fields of comparative planetology, the study of very young stellar objects and high-z galaxies.

Optical and infrared interferometers operating with high spectral resolution can provide velocity-resolved information on the milliarcsecond and sub-milliarcsecond scale. This enables completely new observational approaches to many open problems in stellar and circumstellar astrophysics. Fiber coupling of the output from a relatively simple beam combiner to existing spectrographs offers a cost-effective way to implement interferometric high-resolution spectroscopy at the European Southern Observatory's Very Large Telescope Interferometer.

We briefly summarize the status of dynamic model atmospheres for carbon rich red giants and present some recent results of comparisons between synthetic spectra and observations. We then discuss synthetic intensity profiles and visibilities for such stars. In particular, the effects of atmospheric dynamics on different spectral bands and for different model parameters are studied.

The prime objective of GENIE (Ground-based European Nulling Interferometry Experiment) is to obtain experience with the design, construction and operation of an IR nulling interferometer, as a preparation for the DARWIN / TPF mission. In this context, the detection of a planet orbiting another star would provide an excellent demonstration of nulling interferometry. Doing this through the atmosphere, however, is a formidable task. In this paper we assess the prospects of detecting with nulling interferometry on ESO's VLTI, low-mass companions in orbit around their parent stars. With the GENIE science simulator (GENIEsim) we can model realistic detection scenarios for the GENIE instrument operating in the VLTI environment, and derive detailed requirements on control-loop performance, IR background subtraction and the accuracy of the photometry calibration. We analyse the technical feasibility of several scenarios for the detection of low-mass companions in the L'-band.

The perfect site for astronomical interferometry would be one where the wind speeds throughout the atmosphere were very low, there was little atmospheric turbulence (especially at high altitudes), the seismic activity was negligible and the climate very stable. Perhaps surprisingly, these conditions describe the Antarctic plateau perfectly. At Dome C, for example, the average wind speed at ground level is just 2.7 metres/sec. This, combined with low wind speeds at all altitudes up to the stratosphere, leads to a dramatic increase in coherence times. Despite the extreme cold, the Antarctic plateau is a relatively benign environment. At thermal infrared wavelengths the high elevation (typically over 3000 m), intense cold and exceptional dryness also combine to give greatly increased background-limited sensitivities relative to other sites. This increased sensitivity can be used to further enhance interferometer performance, or can be traded against mirror size to allow for a smaller instrument of the same sensitivity.

The Antarctic Planet Interferometer is a concept for an instrument designed to detect and characterize extrasolar planets by exploiting the unique potential of the best accessible site on earth for thermal infrared interferometry. High-precision interferometric techniques under development for extrasolar planet detection and characterization (differential phase, nulling and astrometry) all benefit substantially from the slow, low-altitude turbulence, low water vapor content, and low temperature found on the Antarctic plateau. At the best of these locations, such as the Concordia base being developed at Dome C, an interferometer with two-meter diameter class apertures has the potential to deliver unique science for a variety of topics, including extrasolar planets, active galactic nuclei, young stellar objects, and protoplanetary disks.

We propose a science demonstrator for the Antarctic Plateau Interferometer. It is a comparatively much simpler system than API but dedicated to the goal of obtaining the first low-resolution spectra in the thermal infrared of a few "hot Jupiter" type exoplanets. It would provide a unique platform to acquire operational experience on antarctic stellar interferometry, and build up an extensive database on the relevant site properties, as a preparation for API.

Turbulence in the earth's atmosphere severely limits the resolution and sensitivity of astronomical observations. The vertical distribution of turbulence in the atmosphere has a profound effect on the residuals after correction by an active instrument such as adaptive optics or a fringe tracking interferometer. It has already been shown that the South Pole has turbulence profiles unlike those at any other site, dominated by ground layer turbulence, with low free air seeing. This paper examines the meteorology, climatology and atmospheric physics that produces these conditions. Combining meterological observations at remote sites with models of atmospheric turbulence allows quantitative extrapolation to the likely conditions at sites now under development and consideration that may provide the ultimate ground based site for near and mid-infrared interferometry. The high plateau sites in Antarctica will likely enable complete sky coverage for adaptive optics and interferometry in the near infrared with natural guide stars.

The Micro Arcsecond X-ray Imaging Mission (MAXIM) has been proposed to NASA in response to the Black Hole Imager line in the newly created "Beyond Einstein" program. Meeting the scientific goals of event horizon imaging has created a new set of technical challenges for the Maxim team. Certainly the most difficult of these is the need for 0.1 micro-arcsecond resolution imaging in the x-ray. We will review the key scientific challenges and show how they flow down into instrument requirements. We will then review our baseline design, discuss the array of technical challenges facing Maxim, and present the current status of our technology.

The Submillimeter Probe of the Evolution of Cosmic Structure (SPECS) is a space-based imaging and spectral ("double Fourier") interferometer with kilometer maximum baseline lengths for imaging. This NASA "vision mission" will provide spatial resolution in the far-IR and submillimeter spectral range comparable to that of the Hubble Space Telescope, enabling astrophysicists to extend the legacy of current and planned far-IR observatories. The astrophysical information uniquely available with SPECS and its pathfinder mission SPIRIT will be briefly described, but that is more the focus of a companion paper in the Proceedings of the Optical, Infrared, and Millimeter Space Telescopes conference. Here we present an updated design concept for SPECS and for the pathfinder interferometer SPIRIT (Space Infrared Interferometric Telescope) and focus on the engineering and technology requirements for far-IR double Fourier interferometry. We compare the SPECS optical system requirements with those of existing ground-based and other planned space-based interferometers, such as SIM and TPF-I/Darwin.

The Darwin and Terrestrial Planet Finder missions, represent the European Space Agency (ESA) and NASA's interest in ultimately searching for and when found studying planets similar to the Earth-like planets in our own Solar System. As such they may be technologically very challenging space missions but recent developments points towards robust solutions. In this talk, we compare the technologies, the available solutions, and the current status in both projects. We put the emphasis on the optical technologies required, and address both main possibilities considered for planet finding, i.e. a nulling interferometer and a coronograph. We outline the strategies for selecting the appropriate technology for each element of the missions. Finally we also address the synergy in the technologies required for other missions, as well as other applications except purely scientific.

The Fourier-Kelvin Stellar Interferometer (FKSI) is a mission concept for a nulling interferometer for the near-to-mid-infrared spectral region (3-8µm). FKSI is conceived as a scientific and technological precursor to TPF. The scientific emphasis of the mission is on the evolution of protostellar systems, from just after the collapse of the precursor molecular cloud core, through the formation of the disk surrounding the protostar, the formation of planets in the disk, and eventual dispersal of the disk material. FKSI will answer key questions about extrasolar planets: Σ What are the characteristics of the known extrasolar giant planets? Σ What are the characteristics of the extrasolar zodiacal clouds around nearby stars? Σ Are there giant planets around classes of stars other than those already studied? We present preliminary results of a detailed design study of the FKSI. Using a nulling interferometer configuration, the optical system consists of two 0.5m telescopes on a 12.5m boom feeding a Mach-Zender beam combiner with a fiber wavefront error reducer to produce a 0.01% null of the central starlight. With this system, planets around nearby stars can be detected and characterized using a combination of spectral and spatial resolution.

The Stellar Imager (SI) is a far-horizon or "Vision" mission in the NASA Sun-Earth Connection (SEC) Roadmap, conceived for the purpose of understanding the effects of stellar magnetic fields, the dynamos that generate them, and the internal structure and dynamics of the stars in which they exist. The ultimate goal is to achieve the best possible forecasting of solar/stellar activity and its impact on life in the Universe. The science goals of SI require an ultra-high angular resolution, at ultraviolet wavelengths, on the order of 0.1 milliarcsec and thus baselines on the order of 500 meters. These requirements call for a large, multi-spacecraft (>20) imaging interferometer, utilizing precision formation flying in a stable environment, such as in a Lissajous orbit around the Sun-Earth L2 point. SI's resolution (several 100 times that of HST) will make it an invaluable resource for many other areas of astrophysics, including studies of AGN's, supernovae, cataclysmic variables, young stellar objects, QSO's, and stellar black holes. In this paper, we present an update on the ongoing mission concept and technology development studies for SI. These studies are designed to refine the mission requirements for the science goals, define a Design Reference Mission, perform trade studies of selected major technical and architectural issues, improve the existing technology roadmap, and explore the details of deployment and operations, as well as the possible roles of astronauts and/or robots in construction and servicing of the facility.

The design and performance of a Fizeau interferometer with long focal length and large field of view are discussed. The optical scheme presented is well suited for very accurate astrometric measurements from space, being optimised, in terms of geometry and aberrations, to observe astronomical targets down to the visual magnitude mV=20, with a measurement accuracy of 10 microarcseconds at mV=15. This study is in the context of the next generation astrometric space missions, in particular for a mission profile similar to that of the Gaia mission of the European Space Agency. Beyond the accuracy goal, the great effort in optical aberrations reduction, particularly distortion, aims at the optimal exploitation of data acquisition done with CCD arrays working in Time Delay Integration mode. The design solution we present reaches the astrometric goals with a field of view of 0.5 square degrees.

This overview paper is a progress report about the system design and technology development of two interferometer concepts studied for the Terrestrial Planet Finder (TPF) project. The two concepts are a structurally-connected interferometer (SCI) intended to fulfill minimum TPF science goals and a formation-flying interferometer (FFI) intended to fulfill full science goals. Described are major trades, analyses, and technology experiments completed. Near term plans are also described. This paper covers progress since August 2003 and serves as an update to a paper presented at that month's SPIE conference, "Techniques and Instrumentation for Detection of Exoplanets."

We investigate whether the design of the DARWIN nulling interferometer can be simplified, while maintaining or even improving its performance, by accepting somewhat higher levels of stellar leakage. We establish detailed requirements on stellar suppression for a nulling interferometer, considering a realistic DARWIN stellar target sample. The dominating noise source, represented by the local zodiacal cloud, is essentially constant for all target stars while the stellar leakage decreases as the inverse of the distance squared. This means that such stellar leakage has an effect on the integration times of near-by target stars, while for more distant targets its influence decreases significantly. We assess the impact of different types of nulling profiles and identify those stars for which the detection sensitivity can be maximised.

A number of stellar systems that can be searched for presence of Earth-like planets in a given mission lifetime is a key figure of merit for planet hunting stellar interferometers. We have developed a method to calculate the number of stellar systems that can be searched and characterized. Using this method we have evaluated the performance of a number of architectures. We conclude that simpler second-order null architectures outperform more complicated fourth-order null architectures. We also quantify the advantages of the variable length formation-flying configurations vs. fixed length structurally connected configurations.

This paper describes the current status of the technical program aiming to demonstrate the viability of a formation-flying mid-infrared nulling interferometer architecture for the Terrestrial Planet Finder (TPF) program. Until recently the TPF project was considering four architectures, with the goal of selecting one in the 2006 timeframe. In April 2004, the project office opted instead to follow a path leading to a small (4x6m) visible-light coronagraph, to launch around 2014, and a formation-flying interferometer (FFI), to launch before 2020. The FFI is proposed to satisfy the full TPF science goal to completely survey 150 stars for evidence of terrestrial planets similar to Earth, while the coronagraph will perform a survey of 30-50 stars at visible wavelength. FFI trade studies conducted since mid-2003 have focused on key factors driving overall flight segment mass and performance, including launch vehicle packaging, deployment approach, thermal design (particularly the thermal shield configuration), structural design, and formation flying approach. This paper summarizes the results of the recent design trades, with discussion of the primary requirements that drive the baseline design concept. Analyses supporting the baseline design are summarized, and areas for future study are discussed.

The MIT Space Systems Laboratory and Payload Systems Inc. has developed the SPHERES testbed for NASA and DARPA as a risk-tolerant medium for the development and maturation of spacecraft formation flight and docking algorithms. The testbed, which is designed to operate both onboard the International Space Station and on the ground, provides researchers with a unique long-term, replenishable, and upgradeable platform for the validation of high-risk control and autonomy technologies critical to the operation of distributed spacecraft missions such as the proposed formation flying interferometer version of Terrestrial Planet Finder (TPF). In November 2003, a subset of the key TPF-like maneuvers has been performed onboard NASA's KC-135 microgravity facility, followed by 2-D demonstrations of two and three spacecraft maneuvers at the Marshall Space Flight Center (MSFC) in June 2004. Due to the short experiment duration, only elements of a TPF lost in space maneuver were implemented and validated. The longer experiment time at the MSFC flat-floor facility allows more elaborate maneuvers such as array spin-up/down, array resizing and array rotation be tested but in a less representative environment. The results obtained from these experiments are presented together with the basic estimator and control building blocks used in these experiments.

Metrology and pointing will be enabling technologies for a new generation of astronomical missions having large and distributed apertures and delivering unprecedented performance. The UV interferometer Stellar Imager would study stellar dynamos by imaging magnetic activity on the disks of stars in our Galaxy. The X-ray interferometer Black Hole Imager would study strong gravity physics and the formation of jets by imaging the event horizons of supermassive black holes. These missions require pointing to microarcseconds or better, and metrology to nm accuracy of optical elements separated by km, for control of optical path difference. This paper describes a metrology and pointing system that meets these requirements for the Stellar Imager. A reference platform uses interferometers to sense alignment with a guide star. Laser gauges determine mirror positions in the frame of the reference platform, and detector position is monitored by laser gauges or observations of an artificial star. Applications to other astronomical instruments are discussed.

This paper describes the broad goals and the current status of the Space Interferometry Mission (SIM). SIM was endorsed in the 1990 decadal report of the Astronomy and Astrophysics survey committee of the National Research Council. The SIM mission would be the first long baseline interferometer in space. The goals of SIM represent not factors of two or three improvement in astronometric accuracy, but two to three orders of magnitude improvement. The current most accurate astrometric measurements are from the Hipparcos satellite launched by ESA in 1990. Hipparchos achieved slightly better than 1 milliarcsec global astrometric accuracy. SIM's goal is 4 microarcsec accuracy for global astrometry (for a nominal 5 yr mission) and 1 microarcsec for single measurement narrow angle accuracy. The narrow angle precision translates to the ability to measure the "wobble" of stars with an error of 0.14 uas, if the target is observed 50 times during the 5 year mission. The paper gives an overview of the type of scientific questions SIM will address, concentrating on the planet detection aspects of SIM.

Optical interferometry will open new vistas for astronomy over the next decade. The Space Interferometry Mission, operating unfettered by the Earth's atmosphere, will offer unprecedented astrometric precision that promises the discovery of Earth-class extra-solar planets as well as a wealth of important astrophysics. Optical interometers also present severe technological challenges: laser metrology systems must perform with sub-nanometer precision; mechanical vibrations must be controlled to nanometers requiring orders of magnitude distrubance rejection; a multitude of actuators and sensors must operate flawlessly and in concert. The Jet Propulsion Laboratory along with its industry partners, Northrop Grumman Space Technology, and Lockheed Martin, are addressing these challenges with a technology development program that is nearing completion. Emphasis is shifting from technology demonstration to technology transfer to the flight team that wil build and launch the space system.

The Space Interferometry Mission (SIM) will perform global astrometry (full sky), local wide-angle (15 degree) and narrow-angle (1 degree) observations to search extra-solar planets, and can calibrate stellar and galactic evolution theories. The astrometric accuracy of the SIM mission depends on spectral characteristics of the optics, detectors and targets. This paper will discuss the photometric throughput of the SIM instrument, and analyze effects of wavefront errors, optical mismatches and control biases as a function of wavelength. The color dependence models of the instrument optics including mirrors, lenses, field-stop and beam-splitter are presented. The performances of different detectors with a variety of coatings are compared. A model of the SIM fringe spectrometer is created. For early and late types of stars, brightness dependency errors are analyzed for different combinations of optics and detectors. Visibility loss due to imperfect optics is investigated in detail. Based on the models of instrument and estimated visibilities, the astrometric accuracies for various kinds of stars are evaluated. It is important to emphasize that not only light sources, mirrors, lenses, field stop and detectors are all wavelength dependent, but also fringe visibility loss, wavefront error, optics control error, etc. are all a function of wavelengths. For the first time the estimate of SIM performance is based on spectral analysis of all factors above, rather than monochromatic approximations of detected fringes, or simply adopted constants. This paper summarizes the astrometric accuracies for a wide range of stars and various combinations of optical design and detector configurations. It has been verified that SIM has astrometric accuracy of about 4 μas for targets with different spectra.

The Space Interferometer Mission (SIM) flight instrument will not undergo a full performance, end-to-end system test on the ground due to a number of constraints. Thus, analysis and physics-based models will play a significant role in providing confidence that SIM will meet its science goals on orbit. The various models themselves are validated against the experimental results of severl "picometer" testbeds. In this paper we describe a set of models that are used to predict the magnitude and functional form of a class of field-dependent systematic errors for the science and guide interferometers. This set of models is validated by comparing predictions with the experimental results obtained from the MicroArcsecond Metrology (MAM) testbed and the Diffraction testbed (DTB). The metric for validation is provided by the SIM astrometric error budget.

We present a study of how the TPF interferometer sensitivity relates to vibration level and control system performance. The error budgets for the TPF interferometer have two parts: the null floor, i.e. the suppression of starlight leakage due to instrument imperfections and performance limits; and null variation, i.e. planet-mimicking signal variations due to instrument errors and performance variations. The first impacts the statistical noise and thus the integration time; the second represents a planet sensitivity floor which limits further improvement with integration time. Oliver Lay (JPL) has developed an extensive analysis tool known as the Interferometer Performance Model (IPM) for managing both budgets. The budgets we use have many of the same features, but are less well-developed; the requirements are similar. Here we develop an example implementation of the TPF interferometer control system, analyze the system performance using an integrated model, and show that it meets the TPF planet sensitivity requirements using our performance budgets. We summarize key requirements and lessons arising from this exercise.

The Large Binocular Telescope (LBT) will be a unique interferometric facility when it is completed in 2005. The telescope incorporates two, 8.4-meter diameter primary mirrors on a single mounting. With 14.4 meter center-to-center spacing, this interferometer provides the equivalent collecting area of a 12-meter telescope, and, depending on the beam combination scheme, the spatial resolution of a 14.4 or 22.8-meter telescope. We report on the status of two initial interferometric instruments planned for the LBT. A group based at the University of Arizona is constructing LBTI, a thermal infrared beam combiner focusing on nulling, but allowing thermal imaging as well. This instrument will search for and measure zodiacal light in candidate stellar systems in preparation for the Terrestrial Planet Finder (TPF) and Darwin missions. There is also a program to search for young Jupiters. A second group, based in Heidelberg, Arcetri, Cologne, and Bonn, is building LINC-NIRVANA, a near-infrared Fizeau-mode beam combiner with multi-conjugated adaptive optics (MCAO). Fizeau interferometry preserves phase information and allows true imagery over a wide field of view. Using state-of-the-art detector arrays, coupled with advanced atmospheric correction strategies, LINC-NIRVANA will enable a broad variety of scientific programs that require the ultimate in sensitivity, field-of-view, and spatial resolution.

The Mauna Kea Observatory offers a unique opportunity to build a large and sensitive interferometer. Seven telescopes have diameters larger than 3 meters and are or may be equipped with adaptive optics systems to correct phase perturbations induced by atmospheric turbulence. The maximum telescope separation of 800 meters can provide an angular resolution as good as 0.25 milli-arcseconds in the J band. The large pupils and long baselines make 'OHANA very complementary to existing large optical interferometers. From an astrophysical point of view, it opens the way to imaging of the central part of faint and compact objects such as active galactic nuclei and young stellar objects. On a technical point of view, it opens the way to kilometric or more arrays by propagating light in single-mode fibers. First instruments have been built and tested successfully at CFHT, Keck I and Gemini to inject light into single-mode fibers thus partly completing Phase I of the project. Phase II is now on-going with the prospects of the first combinations of Keck I - Keck II in 2004 and Gemini - CFHT in 2005.

The Magdalena Ridge Observatory Interferometer will be the first facility-class optical interferometer optimized strictly for an imaging science mission. The array in its final form is envisaged to comprise ten 1.4 m aperture movable telescopes in a Y configuration, baselines extending from 8 to 400 meters, delay lines capable of tracking well-placed sources for 6 continuous hours, fringe-tracking in the near-infrared , and undertake science observations at both near-infrared and optical wavelengths. The science reference mission includes studies of young stellar objects, a full range of stellar astrophysics, and imaging studies of the nearest and brightest 100 active galactic nuclei. We will be staffing up to our full complement of personnel in New Mexico over the next year. Our goal for first fringes on the first baseline is 2007.

According to the "hypertelescope" imaging mode, stellar interferometers could provide direct snapshot images. Whereas the Fizeau imaging mode is useless when the aperture is highly diluted, a "densified-pupil" or "hypertelescope" imaging mode can concentrate most light into the high-resolution central interference peak, allowing direct imaging of compact sources and stellar coronagraphy for exoplanets finding. The current VLTI is able to combine light from 2 to 3 telescopes coherently, but the combination of 4 to 8 beams is foreseen in subsequent phases. In order to exploit the full forthcoming VLTI infrastructure, a next generation instrument has been proposed (VIDA) in 2002 for very high-resolution snapshot imaging with UTs and/or ATs. This paper presents a new attractive design studied for this instrument using single mode optical fibers and allowing a multi-field imaging mode. We also give the expected performances with a coronagraph, computed from numerical simulations including cophasing and adaptive optics residual errors.

The PRIMA facility will implement dual-star astrometry at the VLTI. We have formed a consortium that will build the PRIMA differential delay lines, develop an astrometric operation and calibration plan, and deliver astrometric data reduction software. This will enable astrometric planet surveys with a target precision of 10μas. Our scientific goals include determining orbital inclinations and masses for planets already known from radial-velocity surveys, searches for planets around stars that are not amenable to high-precision radial-velocity observations, and a search for large rocky planets around nearby low-mass stars.

We are studying an optical concept aiming at recombining four mid-infrared telescope beams, where interference fringes are sampled in the pupil plane. Such a principle is perfectly adapted for reconstructing images by aperture synthesis with teh VLTI. It could be used for building a new generation 10 μm instrument, but instead of doing a totally new instrument, we propose the design of an optical module that can supply the surrent MIDI-VLTI instrument with 4 beams. The combined use of this module together with the MIDI instrument is the project called APreS-MIDI. Such an instrument at the VLTI focus will have an unique and very strong astrophysical potential.

The science objectives of VITRUV is to investigate the morphology of compact astrophysical objects in optical wavelengths like the environment of AGN, star forming regions, stellar surfaces. This instrument will take full advantage of the VLTI site with 4 very large telescopes and 4 auxiliary telescopes. The instrument concept is to built aperture synthesis images like the millimeter-wave radiointerferometer of the IRAM Plateau de Bure. VITRUV coupled to the VLTI will have similar and even better resolution than ALMA. The astrophysical specifications although not yet finalized will be a temporal resolution of the order of 1 day, spectral resolution from 100 to 30,000, image dynamic from 100 to 1,000, a field of view of 1 arcsec for an initial wavlength coverage from 1 to 2.5 microns that could be extended from 0.5 to 5 microns. The technology that is contemplated at this stage is integrated optics.

The Very Large Telescope Interferometer (VLTI) on Cerro Paranal (2635 m) in Northern Chile reached a major milestone in September 2003 when the mid infrared instrument MIDI was offered for scientific observations to the community. This was only nine months after MIDI had recorded first fringes. In the meantime, the near infrared instrument AMBER saw first fringes in March 2004, and it is planned to offer AMBER in September 2004. The large number of subsystems that have been installed in the last two years - amongst them adaptive optics for the 8-m Unit Telescopes (UT), the first 1.8-m Auxiliary Telescope (AT), the fringe tracker FINITO and three more Delay Lines for a total of six, only to name the major ones - will be described in this article. We will also discuss the next steps of the VLTI mainly concerned with the dual feed system PRIMA and we will give an outlook to possible future extensions.

Keck Interferometer is a NASA-funded project to combine the two 10 m Keck telescopes for high sensitivity near-infrared fringe visibility measurements, nulling interferometry at 10 μm to measure the quantity of exozodiacal emission around nearby stars, and differential-phase measurements to detect "hot-Jupiters" by their direct emission. It is being developed by the Jet Propulsion Laboratory, the W. M. Keck Observatory, and the Michelson Science Center. Recent activity has included formal visibility mode commissioning, as well as science observations, and we briefly review some of the significant technical aspects and updates to the system. We have also completed laboratory development of the nuller. The nuller uses two modified Mach-Zehnder input nullers, a Michelson cross combiner, and a 10 μm array camera to produce background-limited null measurements. To provide required temporal stability for the nuller, the system incorporates end-to-end laser metrology with phase referencing from two 2.2 μm fringe trackers. The nuller recently completed its pre-ship review and is being installed on the summit. After nuller integration and test, the differential phase mode will be deployed, which will use a 2-5 μm fringe detector in combination with a precision path length modulator and a vacuum delay line for dispersion control.

We describe recent science projects that the Navy Prototype Optical Interferometer (NPOI) scientific staff and collaborators are pursuing. Recent results from the wide angle astrometric program and imaging programs (rapid rotators, binaries and Be stars) will be summarized. We discuss some of the technology that enables the NPOI to operate routinely as an observatory astronomical instrument.

Georgia State University's Center for High Angular Resolution Astronomy (CHARA) operates a multi-telescope, long-baseline, optical/infrared interferometric array on Mt. Wilson, California. We present an update on the status of this facility along with a sample of preliminary results from current scientific programs.

Closure-phase science and technology are dominant features of the recent activity at IOTA. Our science projects include imaging several spectroscopic binary stars, imaging YSOs including Herbig AeBe stars, detecting asymmetries in a large sample of Mira stars, and measuring water shells around Miras. Many technology projects were pursued in order to make these science observations possible. These include installation of a third-generation integrated-optics 3-beam combiner (IONIC), completion of the real-time control system software, installation of fringe-packet tracking software, use of narrow sub-H band filters, validation of the phase-closure operation, development of CPLD control of the science camera (PICNIC) and star-tracker camera (LLiST), installation of a new star-tracker camera, expansion of the observing facility, and installation of new semi-automated optical alignment tools.

When we last reported the status of the U.C. Berkeley Infrared Spatial Interferometer (ISI) in 2002, we presented simulations, based upon our two-telescope experience, for the expected performance of a three-telescope array that would be capable of measuring three simultaneous visibilities and one closure phase at mid-infrared wavelengths. The ISI is now fully operational as an imaging array and is routinely making fringe visibility and closure phase measurements of late-type stars in the 9 to 12 micron wavelength region. We describe here the technology which is currently in use, along with actual measurements and preliminary 11.15 micron (one-dimensional) image reconstructions.

The Sydney University Stellar Interferometer (SUSI) is a long-baseline optical interferometer operating at an observatory near Narrabri in Australia. SUSI features a 640 m long North-South array with 11 fixed siderostat stations. New science from the Blue (400-500 nm) and from the recently commissioned Red (500-950 nm) fringe detectors will be presented. Recent technological developments, mainly associated with the new Red detection system, encompassing wavefront correction, fringe encoding, wavelength switching and data analysis strategies, are described.

The Palomar Testbed Interferometer (PTI) is a near-infrared, long-baseline interferometer located at the Palomar Observatory. PTI obtained first fringes in 1995, and has been in routine scientific operations since 1998. PTI was primarily designed as a technology demonstration experiment for the Keck Interferometer, and has been successful in demonstrating 100-uas-class differential astrometry and two-combiner phase referencing. In addition to its engineering development accomplishments, PTI has been extraordinarily scientifically productive, producing more than 25 refereed scientific papers to date. This contribution will provide an update on PTI’s operational, technical, and scientific status.

We present a summary of the activity of the Cambridge Optical Aperture Synthesis Telescope (COAST) team and review progress on the astronomical and technical projects we have been working on in the period 2002--2004. Our current focus has now moved from operating COAST as an astronomical instrument towards its use as a test-bed for strategic technical development for future facility arrays. We have continued to develop a collaboration with the Magdalena Ridge Observatory Interferometer, and we summarise the programmes we expect to be working on over the next few years for that ambitious project. In parallel, we are investigating a number of areas for the European Very Large Telescope Interferometer and these are outlined briefly.

After the first fringe detection with MIRA-I.2 30m baseline for Vega in June 2002, fringes for Vega and Deneb has been confirmed and then construction continued. Fast delay line has been evacuated and extended from 4m to 8m long. Coarse delay line has been extended from 2m to 4m. Baseline vector has been determined with 0.1mm accuracy. Aluminized mirrors have been changed to gold-plated ones, and the total throughput has become four times larger than before at the 600nm-1000nm band. The photon rate is 150 per ms for a 2 mag (I-band) star and the present limiting magnitude is better than 3mag. A delay modulation PZT has been set to push a cat's eye retro-reflector. Observations have been made for 6 stars with successful fringe packet detections. Visibility stability has been being studied with artificial light sources and Vega, which preliminary results are better than a few percent. A three-color system between 600-1000nm is now on the half way of installation. Gregorian cat's eye retro-reflectors with fine delay line PZTs and fringe tracking control software is planned to be installed.

FINITO is the first generation VLTI fringe sensor, optimised for three beam observations, recently installed at Paranal and currently used for VLTI optimisation. The PRIMA FSU is the second generation, optimised for astrometry in dual-feed mode, currently in construction. We discuss the constraints of fringe tracking at VLTI, the basic functions required for stabilised interferometric observations, and their different implementation in the two instruments, with remarks on the most critical technical aspects. We provide an estimate of the expected performance and describe some of their possible observing and calibration modes, with reference to the current scientific combiners.

We describe the fringe-packet tracking software installed at the infrared optical telescope array (IOTA). Three independently developed fringe-packet tracking algorithms can be used to equalise the optical path lengths at the interferometer. We compare the performance of these three algorithms and show results obtained tracking fringes for three independent baselines on the sky.

Differential phase observations with a near-IR interferometer offer a way to obtain spectra of extrasolar planets. The method makes use of the wavelength dependence of the interferometer phase of the planet/star system, which depends both on the interferometer geometry and on the brightness ratio between the planet and the star. The differential phase is strongly affected by instrumental and atmospheric dispersion effects. Difficulties in calibrating these effects might prevent the application of the differential phase method to systems with a very high contrast, such as extrasolar planets. A promising alternative is the use of spectrally resolved closure phases, which are immune to many of the systematic and random errors affecting the single-baseline phases. We have modeled the response of the AMBER instrument at the VLTI to realistic models of known extrasolar planetary systems, taking into account their theoretical spectra as well as the geometry of the VLTI. We present a strategy to determine the geometry of the planetary system and the spectrum of the extrasolar planet from closure phase observations in two steps. We show that there is a close relation between the nulls in the closure phase and the nulls in the corresponding single-baseline phases: every second null of a single-baseline phase is also a null in the closure phase. In particular, the nulls in the closure phase do not depend on the spectrum but only on the geometry. Therefore the geometry of the system can be determined by measuring the nulls in the closure phase, and braking the remaining ambiguity due to the unknown system orientation by means of observations at different hour angles. Based on the known geometry, the planet spectrum can then be directly synthesized from the closure phases.

With ESO's Phase Referenced Imaging and Micro-arcsecond Astrometry (PRIMA) facility well into its procurement phase expectations are made about the astrometric performance. It appears that in almost all respects the instrumentally induced errors are expected to have Power Spectral densities well below those due to the atmosphere. However for the target performance to be achieved, some effects must reduce by averaging by some 4 orders of magnitude. The most serious worry foreseen is lack of thermal stability in the air-filled Delay Line tunnel, and it is recommended that outside wind influence be impeded.

We summarize the implementation of the Differential Phase (DP) mode at the Keck Interferometer. Multicolor phase measurements are potentially a powerful astrophysical probe -- and can allow direct detection of Roaster-type exoplanets from the ground. The Keck Interferometer will measure differential phases between H-K, and K-L bands to levels of 3 mrad or better. At JPL, we are engaged in development and testing of instrumentation that will enable these extremely sensitive measurements. First on-sky observations are expected to start in middle 2005. In this article we describe DP and other related techniques, provide an outline of the instrument and present results from preliminary laboratory experiments.

The near-infrared instrument AMBER at the VLTI allows, among other interferometric observables, the simultaneous measurement of the phase between various spectral channels. Color-differential phase thus yields spatial and spectral information on unresolved sources, and could lead to such ambitious goals as the spectroscopy of nearby hot, giant exoplanets. This will require, though, an extreme stability on the measurement, which is likely to be affected by chromatic effects at the various stages of the light path. We present how AMBER has been designed to minimize and to calibrate such effects. We give estimates of their contributions from different origins, and present recent measurements of the instrumental stability. We discuss the possibility to supress the residual chromatic effects in post-data treatment in order to reach a precision limited by the photon noise on the differential phase.

This investigation focuses on observational measurements of the differential interferometric phase between spectral channels in the VLTI/MIDI instrument. Measurements of target stars are compared with theoretical predictions in order to investigate the effects of dispersion in humid air on differential phase measurements at N band (~10 micron wavelength). An accuracy of 1 degree RMS phase error is achieved after calibration during stable environmental conditions, but this accuracy is degraded if there are fluctuations in humidity between observations. Stabilisation and/or monitoring of the environmental conditions in the VLTI ducts and tunnels will be required in order to achieve the best differential phase performance with MIDI. The measured differential phases are found to be consistent with a model for the refractive index of air based on the HITRAN database.

The Very Large Telescope Interferometer (VLTI) that currently combines the four VLT 8.2-m Unit Telescopes (UTs) is now being equipped with its dedicated array of Auxiliary Telescopes (ATs). This array includes four 1.8-m telescopes which can be relocated on thirty observing stations distributed on the top of the Paranal Observatory. This array, albeit less sensitive than the array of UTs, is a key element for the scientific operation of the VLTI. After more than five years of design, development, manufacturing and extensive testing in Europe by the company AMOS (Belgium), the first AT arrived on Paranal in October 2003 where it was re-assembled in two months. This was followed by the final testing on the sky, the so-called 'commissioning', that took place in January and February 2004. This paper describes the recent activities from the delivery of AT1 in Europe up to its commissioning at Paranal. It presents a few results from the commissioning and reports the achieved performance. The status of the other ATs is also briefly described.

We present ASPRO-VLTI, the newly developed JMMC software that will allow astronomers to prepare observations with both first-generation VLTI instruments, AMBER and MIDI. This software has been specifically designed to hide as much as possible instrument complexity and should permit to focus primarily on the science objectives. The targeted users are all astronomers interested in observing with VLTI but with limited interferometric background. It should be particularly well suited for the preparation of ESO-VLTI Phase I proposals. The user can define its target model, choose observational parameters and check if the observations will match its scientific goals. Output quantities and plots can be used for observation proposals.

The ESO Very Large Telescope Interferometer (VLTI) is the first general-user interferometer that offers near- and mid-infrared long-baseline interferometric observations in service mode as well as visitor mode to the whole astronomical community. Regular VLTI observations with the first scientific instrument, the mid-infrared instrument MIDI, have started in ESO observing period P73, for observations between April and September 2004. The efficient use of the VLTI as a general-user facility implies the need for a well-defined operations scheme. The VLTI follows the established general operations scheme of the other VLT instruments. Here, we present from a users' point of view the VLTI specific aspects of this scheme beginning from the preparation of the proposal until the delivery of the data.

We report results from development of a prototype extremely coherent single mode fiber optic array for the TPF visible nulling interferometer. This device is used to negate the effects of residual stellar leakage (scattering) due to imperfections in the TPF telescope optics and the visible nulling interferometer optical train. This prototype consists of 10 X 10 single mode fibers, V-groove arrays, two lenslet arrays (one at the front end and the other at the back end) and auxiliary mechanical components. This first development will pave a clear path for making a final coherent fiber array with ~ 1000 fibers. The final array, once fully functional, should be able to improve image contrast by another ~ 3 orders of magnitude after 6-7 orders of magnitude star light subtraction using the visible nulling interferometer to allow detection of earth-like planets as close as 0.1 arcseconds around stars at ~ 10 pc in space with a 4m size TPF. This concept combines all the advantages of a nulling interferometer with the simplicity of a modest size and modest optical quality single aperture telescope, which allows tremendous reduction of the total cost and simplicity of the operation of a visible TPF over other TPF approaches. This fiber array can also be used in the visible coronagraph for rejecting scattered light.

Modal filtering is mandatory in nulling interferometers dedicated to direct detection of extrasolar terrestrial planets. However, up to date no appropriate waveguides to act as wavefront filter were available for the mid-infrared wavelengths in question. We present the development of silver-halide fibers and chalcogenide fibers to be used for modal filtering within the European DARWIN mission. We give a trade-off of suitable waveguides geometries, possible materials, and fabrication technologies and present measurements of the beam profiles, the insertion loss, and of the modal filtering capability of the developed fiber samples.

Feedhorns like those commonly used in radio-telescope and radio communication equipment couple very efficiently (>98%) to the fundamental Gaussian mode (TEM00). High order modes are not propagated through a single-mode hollow metallic waveguides. It follows that a back to back feedhorn design joined with a small length of single-mode waveguide can be used as a very high throughput spatial filter. Laser micro machining provides a mean of scaling successful waveguide and quasi-optical components to far and mid infrared wavelengths. A laser micro machining system optimized for THz and far IR applications has been in operation at Steward Observatory for several years and produced devices designed to operate at λ=60μm. A new laser micromachining system capable of producing mid-infrared devices will soon be operational. These proceedings review metallic hollow waveguide spatial filtering theory, feedhorn designs as well as laser chemical etching and the design of a new high-NA UV laser etcher capable of sub-micron resolution to fabricate spatial filters for use in the mid-infrared.

During the last years, LETI has developed integrated optics components for stellar interferometry using silica on silicon technology. Recent astrophysical results obtained with the three-telescope combiner IONIC3 on IOTA confirm the interest of this kind of integrated devices in terms of overall performance and stability. New combiners based on symmetric three-waveguide couplers have been theoretically studied, manufactured, and tested in laboratory in collaboration with LAOG over the H-band. Simulation et experimental preliminary results are presented and compared to asymmetric two-waveguide couplers. Applications to three- and four-telescope combination are discussed.

Electron multiplying CCDs, e.g. as delivered by E2V Technologies and Texas Instruments, have the potential to become the detectors of choice for all future optical interferometers, replacing expensive fibre-fed arrays of APDs. We report here on the development of a new 500-channel spectroscopic back end for the COAST interferometer that exploits such a device. An E2V Technologies CCD97 (back illuminated) EMCCD with sub-electron readout noise is used as the detector, and can be read out at pixel rates of up to 30~MHz. We present software and hardware approach used to integrate a new CCD controller with the COAST as well as results from lab tests of the detector and controller.

The Darwin/TPF mission aims at detecting directly extra solar planets. It is based on the nulling interferometry, concept proposed by Bracewell in 1978, and developed since 1995 in several European and American laboratories. One of the key optical devices for this technique is the achromatic phase shifter (APS). This optical component is designed to produce a π phase shift over the whole Darwin spectral range (i.e. 6-18 μm), and will be experimentally tested on the NULLTIMATE consortium nulling test bench (Labèque et al). Three different concepts of APS are being simulated: dispersive plates focus crossing and field reversal. In this paper, we show how thermal, mechanical and optical models are merged into a single robust model, allowing a global numerical simulation of the optical component performances. We show how these simulations help us to optimizing the design and present results of the numerical model.

In this paper, we present results on a test-bed for the use of adaptive optics (AO) in optical interferometry. The test-bed is based on two deformable mirrors made by OKO technologies. The two mirrors are simultaneously controlled by the same computer and control software. The experimental set is based on our portable adaptive optics system. The goal of this test-bed is to study and characterize the effects of aberrations on the fringe contrast and the effects and characterization of the use of AO for improving fringe contrast. In this paper we will report some field test of our portable AO system. We will also describe the test-bed and some of the experimental results obtained so far.

We applied an algorithm for the coherent integration of visibility data of the Navy Prototype Optical Interferometer in the reduction of observations of Altair. This algorithm was first presented at the SPIE meeting in Kona in 2002 and is based on the principle of phase bootstrapping a long baseline using the fringe delays and phases measured on the two shorter baselines with which it forms a triangle in a three-station array. We show that the SNR of the visibility amplitudes and closure phases is significantly increased compared to the standard incoherent integration, also enabling us to use all 28 wavelength channels (instead of 20) afforded by the NPOI spectrometers. The recovery of the data at the blue end is important for constraining any models of this star.

We describe and illustrate the techniques used on spectrally dispersed data from the MIDI interferometer to estimate and remove variations in the atmospheric delay and dispersion. The resulting interferograms can then be added coherently yielding estimates of spectral correlated flux and differential phase. The primary advantage over incoherent estimation of correlated flux is that the system noise need not be accurately known.

We discuss methods of interferometric data reduction using coherent integration of fringe visibility. Unlike incoherent estimation techniques which discard the phase of interference, coherent integration retains as a complex quantity the contribution from each frame (or scan). In order to integrate these coherently, one must apply an OPD correction (or "phase reference") to compensate for random atmospheric pathlength fluctuations. In an instrument with substantial bandwidth, it is also necessary to correct for fixed and random dispersion. The integrity of phase functions obtained is dependent on correct modelling of fixed optical phase functions (obtained from a calibrator observation), dispersion from air filled delay-lines (calibratable in principle), and averaging over time to reduce the effect of random atmospheric water vapor dispersion. To achieve the best performance, it is necessary to include a dispersion tracker as well as tracking achromatic OPD, applying each as a phase correction as a function of time and of optical frequency. Using MIDI, the N band instrument of the VLTI, which has a wide bandwidth, it is often possible to uninterruptedly track random dispersion fluctuations over an observation. Plots of dispersion fluctuations due to water vapor above the VLTI are shown, which are used (along with OPD tracking) to coherently integrate raw frames from that instrument. The resulting complex visibility includes a unique phase delay signature reflecting the source structure. A residual "water-vapor-like" phase may be present due to unmonitored humidity in the delay line paths, and to incomplete averaging of (nominally zero-mean) atmospheric water vapor fluctuations. Nevertheless, the use of visibility phase results corrupted by random dispersion is possible.

Though interferometric techniques are now used routinely around the world, the processing of interferometric data is still the subject of active research. In particular, the corruption of the interferometric fringes by the turbulent atmosphere is currently the most critical limitation to the precision of the ground-based interferometric measurements. In this paper, we discuss the data acquisition and processing procedures of the VINCI/VLTI instrument. Optimal data acquisition parameters and wavelets based processing allow us to remove a posteriori part of the data corrupted by atmospheric turbulence. A relative precision better than 0.1% on the instrumental visibility (for a 5 minutes observation) was already achieved on bright stars without fringe stabilization. Using a dedicated fringe tracker, an even better precision, of the order of a few 10^-4, appears to be within reach. However, we show in this paper that the calibration of the instrumental visibility measurements can easily be the source of significant systematic errors beyond this statistical precision.

In this report we explore replacing the widely used optimal V² estimator with a model-fitting approach. We show that it is possible to fit the fringe power spectra with a physically reasonable model. This approach eliminates the biggest problem with the standard squared visibility estimator - determining the additive, dector-noise bias. We examine the dependence of the bias on count rate for consistency betwee on- and off-fringe measurements. The change of bias with fringe frequency provides additional information about the performance of the detectors. We have also applied a similar approach to the bias correction for the triple product, with comparable results.

This paper describes a method of beam-combination in the so-called hypertelescope imaging technique recently introduced by Labeyrie in optical interferometry. The method we propose is an alternative to the Michelson pupil reconfiguration that suffers from the loss of the classical object-image convolution relation. From elementary theory of Fourier optics we demonstrate that this problem can be solved by observing in a combined pupil plane instead of an image plane. The point-source intensity distribution (PSID) of this interferometric "image" tends towards a psuedo Airy disc (similar to that of a giant monolithic telescope) for a sufficiently large number of telescopes. Our method is applicable to snap-shot imaging of extended sources with a field comparable to the Airy pattern of single telescopes operated in a co-phased multi-aperture interferometric array. It thus allows to apply conveniently pupil plane coronagraphy. Our technique called Interferometric Remapped Array Nulling (IRAN) is particularly suitable for high dynamic imaging of extra-solar planetary companions, circumstellar nebulosities or extra-galactic objects where long baseline interferometry would closely probe the central regions of AGNs for instance.

Darwin is one of the most challenging space projects ever considered by the European Space Agency (ESA). Its principal objectives are to detect Earth-like planets around nearby stars and to characterise their atmospheres. Darwin is conceived as a space "nulling interferometer" which makes use of on-axis destructive interferences to extinguish the stellar light while keeping the off-axis signal of the orbiting planet. Within the frame of the Darwin program, the European Space Agency (ESA) and the European Southern Observatory (ESO) intend to build a ground-based technology demonstrator called GENIE (Ground based European Nulling Interferometry Experiment). Such a ground-based demonstrator built around the Very Large Telescope Interferometer (VLTI) in Paranal will test some of the key technologies required for the Darwin Infrared Space Interferometer. It will demonstrate that nulling interferometry can be achieved in a broad mid-IR band as a precursor to the next phase of the Darwin program. The instrument will operate in the L' band around 3.8 μm, where the thermal emission from the telescopes and the atmosphere is reduced. GENIE will be able to operate in two different configurations, i.e. either as a single Bracewell nulling interferometer or as a double-Bracewell nulling interferometer with an internal modulation scheme.

The Large Binocular Telescope with its single mount design and adaptive optics integrated into the secondary mirrors, provides a unique platform for mid-infrared interferometry. The Large Binocular Telescope Interferometer is designed to take advantage of this platform, specifically for extrasolar planet detection in preparation for the Terrestrial Planet Finder mission. The instrument consists of three components: a general purpose or Universal Beam Combiner (UBC) which preserves the sine condition of the array, a nulling interferometer for the LBT (NIL) to overlap the two beams and sense phase variations, and a nulling-optimized mid-infrared camera (NOMIC) for detection of the final images. Here we focus on the design and tolerancing of the UBC. The components of the system are currently being fabricated and the instrument is planned to be integrated with the LBT in 2006.

The first high-dynamic-range interferometric mode planned to come on line at the Keck Observatory is mid-infrared nulling. This observational mode, which is based on the cancellation of the on-axis starlight arriving at the twin Keck telescopes, will be used to examine nearby stellar systems for the presence of circumstellar exozodiacal emission. This paper describes the system level layout of the Keck Interferometer Nuller (KIN), as well as the final performance levels demonstrated in the laboratory integration and test phase at the Jet Propulsion Laboratory prior to shipment of the nuller hardware to the Keck Observatory in mid-June 2004. On-sky testing and observation with the mid-infrared nuller are slated to begin in August 2004.

We present results of experiments obtained using a new nulling technique that enables deep nulling without the use of achromatic phase shifters. The experimental set-up consists of a three-beam interferometer that should provide a nulling depth of several thousands over a wavelength range of 500 to 650 nm. The intended depth of null was not achieved and further experiments on determining the spectrum of each beam revealed why. We describe a method of obtaining accurate beam spectra in a multi-beam interferometer. The insights on the need of spectral shape control were tested with our nulling theory and proved the sensitivity of this nulling approach with respect to spectral mismatches.

3D common-path interferometer is proposed to obtain achromatic nulling for star coronagraphy. Common-path scheme compensates optical path difference (OPD) effectively and is stable to mechanical vibrations. 3D ray geometry involves polarization rotations ±90° in each interferometer arm and results in achromatic 180° phase shift for destructive interference for on-axial source. The interferometer throughput is obtained at nearly 100% for entire polarized light and nearly 50/50 ratio of light energy is split between Bright and Nulled ports for off-axial source. Theory, simulations and preliminary breadboard experiments are shown in reasonable agreement.

Nulling interferometry of exo-solar planets requires as a minimum two telescopes, of which one is phase shifted by 180 degrees, such that the on-axis stellar object is cancelled, while the light from the off-axis planet interferes constructively. Improvement of the nulling performance and the introduction of chopping leads to space interferometers of four or more telescopes and a separate spacecraft dedicated to beam recombination, as currently baselined for Darwin and TPF. It has recently been demonstrated that the stellar leaks mainly affects the integration times for near-by target stars [o,c]. Considering that there are only a few near-by targets and that the integrations times for each of these is short compared to that of distant stars, it appears advantageous to simplify the interferometer, by accepting higher levels of stellar leaks for near-by targets. A simple, chopping nulling interferometer can be obtained by adding one equal size telescope to the basic two telescope nulling interferometer. Modulation is obtained by applying time-varying phase-shifts to the beams before recombination, i.e. inherent modulation [d]. The recombination of 3 multi-axial beams is achieved by coupling into a single mode waveguide, leading to high modulation and coupling efficiencies, and a single focal plane [i]. Linear and circular telescope configurations are proposed and investigated, including a discussion on the need of a separate spacecraft for beam recombination. The associated transmission and modulation maps and efficiencies are calculated and discussed.

The Multi-Aperture Imaging Interferometer (MAI²), which Alcatel Space has been developing for ESA for deep nulling demonstration in preparation of the Darwin project, is based on an innovative layout, where both beam combination and modal wave-front filtering functions are achieved by means of an Integrated Optics (IO) component. Two different components, based on different designs and technologies, have been developed and characterised by LAOG with detailed design and manufacturing performed by IMEP/GeeO/LETI. SAGEIS-CSO (optical path control) and Alcatel Space have developed the other breadboard functions. The MAI² interferometer achieved stable Darwin-class nulling (10-5) of a simulated star in monochromatic light, and with a relative bandwidth of several percent (10-4). Operation in non-polarised light, with unchanged nulling performances, was also demonstrated. Preliminary characterisation of the relationship between nulling and bandwidth is also provided.

Future nulling space interferometers, such as Darwin and TPF, under study by the European Space Agency and NASA respectively, will rely on fast internal modulation techniques in order to extract the planet signal from the much larger background noise. In this modulation scheme, the outputs of a number of sub-arrays are combined with a variable, achromatic phase shift. In this paper, we discuss the use of well-known OPD modulation techniques in nulling interferometry. The main attractiveness of this approach is that a small OPD modulation at frequency f will modulate the stellar leakage at frequency 2f, since leakage does not depend from the sign of the OPD. In turn, a planet transiting a quasi-linear portion of the transmission map will induce a signal at frequency f at the nulled output, which can be extracted by coherent detection techniques. The properties of this modulation scheme are analyzed, using the Bracewell configuration as a test case. The significance of this technique for ESA's Darwin mission, and its ground-based technology precursor GENIE, are discussed.

By the middle of 2006, the Interferometry Technology development program for NASA's Terrestrial Planet Finder (TPF) Mission has the goal of demonstrating deep and stable interferometric nulling of broadband Mid-IR thermal radiation under conditions that are traceable to the expected on-orbit conditions. Specifically, the task is to demonstrate null levels of 10^-6, with a 50% bandwidth centered at 10 μm, with null stabilities of 10^-7 all at cryogenic temperatures for observational periods of a couple of hours. The Achromatic Nulling activity at JPL addresses this concern in two testbeds: the warm nulling testbed and the cryonulling testbed. The warm nulling testbed will demonstrate the physics of nulling broadband thermal sources in an environment that is conducive to efficient research. We'll explore nulling techniques, optical-mechanical alignment methods, motion control, and path-length metrology for a single beam interferometer, as well as preliminary planet detection techniques. Ultimate nulling capabilities under conditions that are more flight-like will be demonstrated in the cryogenic nulling testbed. Knowledge gained from operation at room temperature will be applied to the cryogenic experiment where we face the additional challenges of extreme temperatures, cryogenic actuators, component survivability and fluxes that are within an order of magnitude of expected flux levels on orbit. Concurrently, we will develop a low flux mid-IR camera that will allow us to measure the nulls at these faint photon fluxes. This talk will review this development activity and will include recent nulling experimental results and plans for future work.

The nulling interferometers proposed for planet detection are arrays of collector telescopes whose amplitudes and phases are carefully controlled to generate a null response at the star. Perturbations in the amplitude and phase response of the instrument lead to time-dependent fluctuations in the stellar leakage that can mimic a planet signal. Understanding these non-linear systematic errors is important, since they drive most of the instrument requirements for missions such as the Terrestrial Planet Finder and Darwin. We show that 'amplitude-phase' errors are the dominant source of instrument noise. They are unaffected by the technique of phase chopping, increase rapidly at short wavelengths, are largely independent of the size and transmission efficiency of the collector optics, and depend only weakly on the nulling configuration and distance to the target system. Detection of an Earth around a G-type star like the sun requires ~1.5 nm of path control and ~0.1% control of the amplitude, integrated over all frequencies, including DC. This paper also introduces the X-Array - a new nulling configuration with 4 collectors and a central combiner arranged in an X pattern. This has a number of advantages over the standard dual Bracewell layout, and over other configurations that have been proposed.

We present a formal comparison of the performance of algorithms used for synthesis imaging with optical/infrared long-baseline interferometers. Five different algorithms are evaluated based on their performance with simulated test data. Each set of test data is formatted in the OI-FITS format. The data are calibrated power spectra and bispectra measured with an array intended to be typical of existing imaging interferometers. The strengths and limitations of each algorithm are discussed.

Wide field interferometry has become a subject of increasing interest in the recent years. New methods have been suggested in order to avoid the drawbacks of the standard wide field method (homothetic mapping) which is not applicable when the aperture is highly diluted; for this reason imaging with non-homothetic arrays is being extensively studied1,2. The field of view of a pupil plane interferometer or a densified array consists only of a few resolution elements; in order to improve these systems, we developed a new method consisting of a Michelson pupil-plane combination scheme where a wide field of view can be achieved in one shot. This technique, called "staircase mirror" approach, has been described in a previous paper3 and uses a stair-shaped mirror in the intermediate image plane of each telescope in the array, allowing for simultaneous correction of the differential delay for the on and off-axis image positions. Experimental results have been obtained recovering the fringes of off-axis stars with an angular separation of approximately 1 arcmin simultaneously, and with a contrast similar to that of the on-axis reference star. With this example, we demonstrate an increase of the field of view by a factor of five, with no need of extra observation time. An algorithm to recover the visibilities from the stars focused on the edge of the steps is described and experimental results are shown that prove that a continuous wide field of view can be acquired in one shot.

Current optical interferometers are affected by unknown turbulent phases on each telescope. The complex Fourier samples measured by the instrument are thus multiplied by unknown phasers corresponding to the turbulent differential pistons between each couple of telescopes. So, the only unaffected phase information is the closure phase of each coherent sub-array. Following the radio-interferometry paradigm, we account for the lack of phase information by introducing system aberration parameters, which are structurally analogous to the turbulent differential pistons. Then, we reconstruct the object by minimizing an original criterion in the object and these aberrations. We have recently designed a metric such that the minimization problem is convex for given aberrations while modeling accurately the noise statistics. The joint criterion is obtained by taking into account the aberrations in this metric. Here, we show how to compute the global minimum of the joint criterion for the aberration step, in spite of the fact that the latter is dramatically non unimodal. This is achieved by exploiting the separable structure of the aberration estimation problem for a known object. Then, we minimize the by optimizing alternatively the object for the current aberrations and the aberrations for the current object. We are currently testing our technique on experimental data.

We present recent results from the Wide-Field Imaging Interferometry Testbed (WIIT). Using a multi-pixel detector for spatial multiplexing, WIIT has demonstrated the ability to acquire wide-field imaging interferometry data. Specifically, these are "double Fourier" data that cover a field of view much larger than the subaperture diffraction spot size. This ability is of great import for a number of proposed missions, including the Space Infrared Interferometric Telescope (SPIRIT), the Submillimeter Probe of the Evolution of Cosmic Structure (SPECS), and the Terrestrial Planet Finder (TPF-I)/DARWIN. The recent results are discussed and analyzed, and future study directions are described.

LINC/NIRVANA (LN) is the German-Italian beam combiner for the Large Binocular Telescope (LBT). It is a Fizeau interferometer and it will provide multiple images of the same target corresponding to different orientations of the baseline. For each one of these images the resolution is not uniform over the field since it is the resolution of a 22.8m mirror in the direction of the baseline and that of a 8.4m mirror in the orthogonal one. Therefore a unique high-resolution image can only be obtained by means of deconvolution methods. Four-years ongoing work of our group on this problem has already clarified the effects of partial adaptive optics corrections and partial coverage of the u,v plane and has produced the Software Package AIRY, a set of modules IDL-based and CAOS-compatible, which can be used for simulation and/or deconvolution of multiple images from the LBT instrument LN. In this paper we present a general approach to the design of methods for the simultaneous deconvolution of multiple images of the same object. These can include both quick-look methods, to be used for routinely process LN images, and ad-hoc methods for specific classes of astronomical objects. We describe several examples of these methods whose implementation and validation is in progress. Finally we present the last version of the Software Package AIRY.

In the last two years the Very Large Telescope Interferometer (VLTI) has been operated with a wavefront controlled down to the Coude focus of each 8m Unit Telescope. From this focus, the stellar beam is passively relayed by more than 10 mirrors distributed along a 100m subterranean path before to be coherently superimposed in the VLTI laboratory. Experience has proven that the observation efficiency would be largely improved by controlling the tilt of the beam directly inside the VLTI laboratory. In this article, we present the justification and basic features of the InfraRed Image Sensor (IRIS) as well as its implementation within the already packed VLTI laboratory. The forthcoming milestones of the project are presented.

A laboratory testbed to demonstrate and characterize control systems needed for interferometric nulling has been under development for two years at Ball Aerospace. Our testbed uses visible light and ambient temperature operation in air, unlike the mid-IR, cryogenic vacuum operation that will be used for Terrestrial Planet Finder. We have developed automated, closed-loop control in delay and tip-tilt.

The Space Interferometry Mission (SIM) requires fringe measurements to the level of picometers in order to produce astrometric data at the micro-arc-second level. To be more specific, it is necessary to measure both the position of the starlight central fringe and the change in the internal optical path of the interferometer to tens of picometers. The internal path is measured with a small heterodyne metrology beam, whereas the starlight fringe position is estimated with a CCD sampling a large concentric annular beam. One major challenge for SIM is to align the metrology beam with the starlight beam to keep the consistency between these two sensors at the system level while articulating the instrument optics over the field of regard. The Micro-Arcsecond Metrology testbed (MAM), developed at the Jet Propulsion Laboratory, features an optical interferometer with a white light source, all major optical components of a stellar interferometer and heterodyne metrology sensors. The setup is installed inside a large vacuum chamber in order to mitigate the atmospheric and thermal disturbances. Astrometric observations are simulated by articulating the optics over the 15 degrees field of regard to generate multiple artificial stars. Recent data show agreement between the metrology and starlight paths to 20pm in the narrow angle field and to 350pm in the full wide angle field of regard of SIM. This paper describes the MAM optical setup, the observation process, the current data and how the performance relates to SIM.

The Space Interferometry Mission (SIM) System Testbed-3 has been integrated in JPL's new Optical & Interferometry Development Laboratory. The testbed consists of a three baseline stellar interferometer whose optical layout is functionally equal to SIM's current flight layout. The main testbed objective is to demonstrate nanometer class stability of fringes in the dim star, or science, interferometer while using path length & angle feed-forward control, and while the instrument is integrated atop a flight-like flexible structure. This work marks the first time an astrometric 3-baseline interferometer is tested in air and on a flight-like structure rather than on rigid optical tables. This paper discusses the system architecture, dim star fringe tracking, and the testbed's latest experimental results.

The Darwin mission is a project of the European Space Agency that should allow around 2015 the search for extrasolar planets and a spectral analysis of their potential atmospheres in order to detect gases and particularly tracers of life. The basic concept of the instrument is a Bracewell nulling interferometer. It allows high angular resolution and high dynamic range. However, this concept, proposed 25 years ago, is very difficult to implement with high rejection factor and has to be demonstrated in laboratory before being applied in space. Theoretical and numerical approaches of the question highlight strong difficulties : - The need for very clean and homogeneous wavefronts, in terms of intensity, phase and polarisation distribution ; - The need for achromatic optical devices working in a wide spectral range (typically 6 to 18 microns for the space mission). A solution to the first point is the optical filtering which has been successfully experimentally demonstrated at 10 microns using a single mode laser. We focus now on the second point and operate a test bench working in the near infrared, where the background constraints are reduced. We present this test bench and the first encouraging results in the 2-4 microns spectral range. We particularly focus on recent optical developments concerning achromatic component, and particularly the beam combiners and the phase shifter, which are keypoints of the nulling interferometry principle. Finally, we present the future of this experimental demonstration, in the thermal infrared, covering the real and whole spectral range of Darwin.

In the context of the Darwin mission, aiming to detect terrestrial extrasolar planets, European Space Administration (ESA) has an R&D program trying to solve the crucial problems, like flotilla spacecraft control, optical spatial filtering, etc... One of the key optical devices of this mission will be Achromatic Phase Shifter (APS) able to accurately provide a 180° phase shift in the IR 6 - 18 microns range. The Institut d'Astrophysique Spatiale (IAS) is leading, in the frame of an ESA granted contract, an European consortium of 9 universities and companies, named Nulltimate, aiming to develop and test three different APS. IAS itself is in charge of the cryogenic test bench facility which is presented here.

A special case of optical aperture synthesis, homothetic mapping, is the topic of this paper. It allows for a wide field of view for interferometric instruments, interesting for astrometric measurments of wide objects. This paper describes a testbed constructed and tested in TNO-TPD in Delft (the Netherlands). This testbed is intended as a tool to investigate the ins and outs of homothetic mapping. The homothetic mapping approach is explained, the whole setup is specified and results are shown.

Kite is a system level testbed for the External Metrology System of the Space Interferometry Mission (SIM). The External Metrology System is used to track the fiducials that are located at the centers of the interferometer's siderostats. The relative changes in their positions needs to be tracked to an accuracy of tens of picometers in order to correct for thermal deformations and attitude changes of the spacecraft. Because of the need for such high precision measurements, the Kite testbed was build to test both the metrology gauges and our ability to optically model the system at these levels. The Kite testbed is a redundant metrology truss, in which 6 lengths are measured, but only 5 are needed to define the system. The RMS error between the redundant measurements needs to be less than 140pm for the SIM Wide-Angle observing scenario and less than 8 pm for the Narrow-Angle observing scenario. With our current testbed layout, we have achieved an RMS of 85 pm in the Wide-Angle case, meeting the goal. For the Narrow-Angle case, we have reached 5.8 pm, but only for on-axis observations. We describe the testbed improvements that have been made since our initial results, and outline the future Kite changes that will add further effects that SIM faces in order to make the testbed more representative of SIM.

The Fizeau Interferometer Testbed (FIT) is a ground-based laboratory experiment at Goddard Space Flight Center (GSFC) designed to develop and test technologies that will be needed for future interferometric spacecraft missions. Specifically, the research from this experiment is a proof-of-concept for optical accuracy and stability, closed-loop control algorithms, optimal sampling methodology of the Fourier UV-plane, computational models for system performance, and image synthesis techniques for a sparse array of 7 to 30 mirrors. It will assess and refine the technical requirements on hardware, control, and imaging algorithms for the Stellar Imager (SI), its pathfinder mission, and other sparse aperture and interferometric imaging mission concepts. This ground-based optical system is a collaborative effort between NASA's GSFC, Sigma Space Corporation, the Naval Research Laboratory, and the University of Maryland. We present an overview of the FIT design goals and explain their associated validation methods. We further document the design requirements and provide a status on their completion. Next, we show the overall FIT design, including the optics and data acquisition process. We discuss the technologies needed to insure success of the testbed as well as for an entire class of future mission concepts. Finally, we compare the expected performance to the actual performance of the testbed using the initial array of seven spherical mirrors. Currently, we have aligned and phased all seven mirrors, demonstrated excellent system stability for extended periods of time, and begun open-loop operations using "pinhole" light sources. Extended scenes and calibration masks are being fabricated and will shortly be installed in the source module. Installation of all the different phase retrieval/diversity algorithms and control software is well on the way to completion, in preparation for future tests of closed-loop operations.

We present the results of designing and fabricating distorted diffraction gratings which allow us to map multiple objects, placed along the optical axis of such grating based systems, onto a single image plane. We plan to use the gratings in the alignment of a long optical train, with the intention of simulating part of the beam transport system of a long baseline optical interferometer. The objective is to produce a simple system that will allow for the automatic alignment of the long optical trains within complex systems, such as the Magdalena Ridge Observatory Interferometer.

At the Navy Prototype Optical Interferometer (NPOI) we have developed a two-stage method for preparation and installation of the optical feed relay stations (elevators). This method reduces contamination, increases consistency, and allows greater management in testing and upgrades. In stage one, we prepare a pre-alignment facility in a laboratory. Using this facility we accurately position the feed stations, internal optics and detector optics relative to the NPOI array line-of-sight. The feed station is cleaned, assembled, internally aligned, tested and placed in its vacuum canister. It is stored under vacuum until transported to the array. In stage two, we align the station on the array by global five-axis adjustments of the vacuum canister. No further independent internal alignments are necessary. The canister is continuously under vacuum during global alignments. We describe the methodology and techniques for installing the optical feed stations.

At the Navy Prototype Optical Interferometer (NPOI), during stellar fringe acquisition and tracking, optical stations along the NPOI vacuum line array remain in passive mode. Optical drift amplitude and rate must remain below certain limits lest stellar acquisition and fringe tracking become unachievable. Subsequent to each observation, relay mirrors are reconfigured within the long delay line stations to provide appropriate constant delays. The placement of these mirrors must be reliable and repeatable within certain tolerances. We describe the results of drift tests conducted on the current long delay line stations.

The Keck Interferometer includes an autoalignment system consisting of pop-up targets located at strategic locations along the beam trains of each arm of the instrument along with a sensor and control system. We briefly describe the hardware of the system and then proceed to a description of the two operational modes of the system. These are: 1) to provide an initial alignment of the coude paths in each arm, and 2) to recover coude alignments between changes of the static delay sled positions. For the initial alignment mode, we review the system performance requirements along with the software used for image acquisition and centroiding. For coudé alignment recovery, we describe beam-train surveys through the static delay (Long Delay Line) and criteria for a successful recovery of a coudé alignment. Finally, we describe the results of testing of the autoalignment system.

Laser metrology systems are a key component of stellar interferometers, used to monitor path lengths and dimensions internal to the instrument. Most interferometers use 'relative' metrology, in which the integer number of wavelengths along the path is unknown, and the measurement of length is ambiguous. Changes in the path length can be measured relative to an initial calibration point, but interruption of the metrology beam at any time requires a re-calibration of the system. The MSTAR sensor (Modulation Sideband Technology for Absolute Ranging) is a new system for measuring absolute distance, capable of resolving the integer cycle ambiguity of standard interferometers, and making it possible to measure distance with sub-nanometer accuracy. We describe the design of the system, show results for target distances up to 1 meter, and demonstrate how the system can be scaled to kilometer-scale distances. In recent experiments, we have used white light interferometry to augment the 'truth' measurements and validate the zero-point of the system. MSTAR is a general-purpose tool for conveniently measuring length with much greater accuracy than was previously possible, and has a wide range of possible applications.

The ARAL system of the VLTI is a multipurpose facility that helps to have the interferometric instruments ready for night observations. It consists of an artificial source (allowing a Mach-Zehnder mode of the interferometric instruments for autotest), an alignment unit (verifying the position of the celestial target in the VLTI field-of-view), and an optical path router (controlling the optical switchyard and the instrument feeding-optics in the VLTI laboratory). With the multiplication of VLTI instruments and their specific features (wavelength coverage, number of beams), an upgrade of ARAL (from its November 2002 version) had to be carried out: the alignment unit has been redesigned, as well as the artificial source. This source will provide a point in the visible and in J, H, K and N infrared bands, split into four beams (with a zero optical path difference at the reference position). After a description of the optomechanics and of the computer architecture of ARAL, we detail the difficulties of building an interferometric artificial source with a wide spectral range.

AMBER, Astronomical Multi BEam combineR, is the near-infrared focal instrument dedicated to the VLTI. It is designed to combine three of the VLTI Telescopes and to work simultaneously in the J, H and K spectral bands (1.0 to 2.4 μm). The project successfully passed the Preliminary Acceptance in Europe in November 2003, resulting in the validation of the instrument laboratory performance1, of the compliance with the initial scientific specifications, and of the acceptance of ESO for AMBER to be part of the VLTI. After the transportation of the instrument to Paranal, Chile in January 2004, the Assembly Integration and Verification phase occurred mid-March to succeed with the first fringes observing bright stars with the VLTI siderostats. This paper describes the different steps of the AIV and the first results in terms of instrumental stability, estimated visibility and differential phase.

Since mid 2002, the complete optical trains of the four 8m Unit Telescopes (UT) are installed and aligned to provide the Very Large Telescope Interferometer (VLTI) with a unique choice of beam combination possibilities [1]. A description of the optical alignment method used and of the final image quality has been given in [2]. We describe in this document the analytical approach used to quantify the geometrical alignment errors not only at the end of the optical train but also at each optical subsystem level.

In this paper we briefly discuss the possibility to use Adaptive Optics in long baseline interferometric gravitational wave detectors. Analisys is carried out to demonstrate the usefulness of Adaptive Optics as a method to integrate double-mode-cleaner systems, presently used or foreseen in the next generation detectors as systems for the reduction of geometrical fluctuations of input laser beam. Finally a prototype of (AO) system for the control of geometrical fluctuations in a laser beam, based on the interferometric detection of phase front, is presented. By comparison with the usual Shack-Hartmann based AO system, we show that this technique is of particular interest when high sensitivity and high band-pass are required for correction of small perturbations like, for instance, the control of the input beam of gravitational waves interferometric detectors.

We argue that adding large apertures with adaptive optics to an optical interferometer improves the performance of the interferometer in two ways: it improves the signal to noise of bright, low-visibility fringes and it also improves the sensitivity of the interferometer. A simple model is presented to support this conclusion.

We present a comparative study of aperture-masking on the Keck-I telescope and adaptive optics with the Keck-II telescope. Recent results from an aperture-masking program at the Keck Observatory in the near-infrared amply demonstrate that this method occupies an important niche in achieving diffraction-limited images despite the many advances in adaptive optics technology. Examples of the efficacy of aperture-masking are the images of the persistent dust-producing Wolf-Rayet star WR 104 and the massive young star with IR excess, MWC 349A. Both these objects were resolved, providing fundamental new insights into their nature. Here we present images of these objects made using adaptive optics in the same wavelength band. These provide a unique opportunity for a direct comparison of two important and competing techniques for ground-based high-resolution imaging. From the AO images, we are unable to recover the gross morphology or detail seen in the aperture-masking results. We note that the AO program might have been hindered by less than ideal observing conditions.

In this paper we address the problem of designing a controller to compensate for phase errors in an array of ground based telescopes, subject to atmospheric distortion. Since the observations from the telescopes are coherently combined to improve resolution, this problem calls for two classes of control loops: a) local control on each telescope to compensate for random phase distortions, b) global control to align all phases coherently so that the array would act as a virtual telescope with a large aperture. By proper use of frequency decomposition, we show that a globally stable system can be designed where the local and global control systems are decoupled. This is obtained by assigning the control of high frequency disturbances to the local controller, and the overall control of low frequency disturbances to the global control system. These two controllers work at different sampling frequencies and can be decoupled using well known results from Multi Rate Processing techniques used as the basis of Quadrature Mirror Filters and Wavelets decompositions. The feasibility of this approach will be tested using data from the Naval Prototype Optical Interferometer (NPOI) in Flagstaff, AZ (USA).

The tip-tilt correction system at the Infrared Optical Telescope Array (IOTA) has been upgraded with a new star tracker camera. The camera features a backside-illuminated CCD chip offering doubled overall quantum efficiency and a four times higher system gain compared to the previous system. Tests carried out to characterize the new system showed a higher system gain with a lower read-out noise electron level. Shorter read-out cycle times now allow to compensate tip-tilt fluctuations so that their error imposed on visibility measurements becomes comparable to, and even smaller than, that of higher-order aberrations.

Long baseline stellar interferometers have been considered an essential tool in studying astrophysics; however, fringe visibilities for stellar interferometers with large apertures are often corrupted by atmospheric turbulence. To reduce the atmospheric turbulence effect, adaptive optics may be used to enhance fringe visibility for stellar interferometers with aperture sizes larger than the atmospheric coherence length. Fringe visibility performance evaluation for long baseline stellar interferometers with and without adaptive optics is presented in this paper. The methodologies used in this paper are described as follows: the optical transfer function for stellar interferometers with large apertures is derived first; then, performance metrics, coherence loss factor and Strehl ratio, are defined. Finally, fringe visibility performance with and without adaptive optics for different turbulent strengths is evaluated using computer simulation results. We show that Noll's mean square residual phase error can be used to compare the coherence loss factor of an interferometer with the Strehl ratio of a single telescope.

In stellar interferometry, the raw fringe visibilities must be calibrated to obtain the intrinsic object visibilities and then object parameters which can be interpreted in term of astrophysical parameters. The selection of suitable calibration stars is crucial to reach the ultimate precision of the interferometric instruments like VLTI. So, we have developed a user-dedicated software to create an evolutive catalog of such calibration stars. This gives useful information for the selection of calibrators with respect to the requirements of the astrophysical program and of the instrumental configuration. A list of potential calibrators is obtained from a set of catalogs available at the Centre de Donnees Astronomiques de Strasbourg (CDS). The CDS request is based on some selection criteria like the maximum angular distance and the range of magnitude around the scientific target. This calibrator selection tool is integrated into ASPRO the interferometric observing preparation software developed by the Jean-Marie Marriotti Center (JMMC), and which is accessible at http://mariotti.ujf-grenoble.fr/~aspro/

ESO's PRIMA (Phase-Referenced Imaging and Micro-arcsecond Astrometry) facility at the VLT Interferometer on Cerro Paranal in Chile is expected to be fully operational in only a few years from now. With PRIMA/VLTI, it will then be possible to perform relative astrometry with an accuracy of the order of 10 microarcseconds over angles of about 10 arcseconds. The main science driver for this astrometric capability is a systematic search for extrasolar planets around nearby stars. Target stars as well as reference stars for this astrometric planet search have to be very carefully chosen in order to make the measurements robust and effective. Most importantly, reference stars have to be astrometrically stable to only a few microarcseconds in order to provide a suitable reference for the astrometric measurements. Target stars should be located at small distances so that a possible planet would cause a detectable astrometric signal. Moreover, a suitable target star and a suitable reference star have to be found within about 10 arcseconds of each other to ensure the highest accuracy and effectiveness, which obviously requires some trade-off in the final target list. Possible strategies and preparatory observations for the assembly of a suitable target list for the astrometric planet search with PRIMA/VLTI will be discussed.

We present in this paper new and accurate calibrations of the surface brightness-color relations that can be used to predict accurately the angular diameter of dwarf stars and subgiants. These stars present significant advantages as calibrators for interferometric observations. In many cases, they are more stable than giants and supergiants, as they are steadily burning their hydrogen. They are present in large numbers in the solar neighborhood, and offer a broad variety of colors. Their proximity allows to minimize the problems related to interstellar extinction in estimating their true magnitudes and colors. Excluding multiple stars, fast rotators and highly variable stars, it is possible to select reliable calibrators from spectro-photometric observations. Moreover, as opposed to the giant and supergiant stars, the photospheric diameter of dwarf stars is well defined and less sensitive to the assumed atmosphere models. In particular, the limb darkening related problems are alleviated at infrared wavelengths.

We present a catalog of reference stars suitable for calibrate long baseline interferometric observations in the infrared. This work includes and extends the previous catalog by Bord\'e et al. (2002): in the K band, a precision of 1~\% or better can be achieved on the visibility with interferometric baselines up to $\sim 200$ meters instead of $\sim 100$ in the preceding version. Angular diameters are computed by way of the absolute spectro-photometric calibration method proposed by Cohen et al. (1999), using IRAS and 2MASS photometric measurements. Our updated catalog contains G8--M0 stars with angular diameters 0.6--1.8~mas (median 1.1~mas) and a median error of 2.1~\%. The median magnitudes are 3.5 in K and 6.5 in V. Our grid is dense enough that one would find a calibrator star closer than 10 degrees to any direction on the sky.

The ESO Data Flow Operations group (also called Quality Control group) is dedicated to look into the performance of the different VLT instruments, to verify the quality of the calibration and scientific data, to control and monitor them on different time scales. At ESO headquarters in Garching, Germany, one QC scientist is dedicated to these tasks for the VLTI instruments: VINCI, MIDI, AMBER, and (eventually) PRIMA. In this paper, we focus on MIDI. In this presentation, we define the tasks of the Quality Control scientist and describe the lessons learned on quality control and instrument trending with the commissioning instrument VINCI. We then illustrate the different aspects of the MIDI Data Flow Operations supported by the QC scientist such as data management issues (data volume, distribution to the community), processing of the data, and data quality control.

A search for extrasolar planets using the ESO VLTI PRIMA facility will become feasible in 2007. An astrometric accuracy of 10 micro-arcseconds will allow us to detect sub-Uranus mass planets around the most nearby stars, as well as to conduct a planet search around stars of different ages. Most of the PRIMA hardware subsystems are currently being developed by industry. At the same time a scientific Consortium has formed that will deliver the differential delay lines and astrometric software for PRIMA to ESO. In this paper we describe the planned efforts by the Consortium related to the "PRIMA astrometry operations and software". These activities include an overall "PRIMA astrometry error budget", a "PRIMA astrometry calibration and observation strategy", the "PRIMA astrometry observation preparation tools" and the "PRIMA astrometry data reduction tools". We describe how all these components fit together in an overall approach to the flow of knowledge within the project. First by quantifying the fundamental limits of the VLTI infrastructure and the astronomical sources under study. Followed by elimination or suppression of the errors through either a hardware change to the system, software control of the system, or a proper calibration and observation strategy. The ultimate goal is being able to calibrate all PRIMA astrometric data acquired over the full lifetime of PRIMA (5 to 10 years) to a uniform accuracy of 10 micro-arcseconds. This will allow identification of long-term trends in the astrometric parameters due to planetary companions around nearby stars and to determine the distances and proper motions for the selected sources.

One of the goals of the VLTI PRIMA (Phase Referenced Imaging and Micro-arcsecond Astrometry) facility will be to obtain high accuracy astrometry (of the order of 10 micro-arcsec) for the measurement of the reflex motion due to planets. In order to achieve this an offline astrometric Data Analysis Facility (DAF) is planned to perform a homogeneous and iterative analysis over several years of observations. This system will be part of the PRIMA Data Reduction Library (DRL), which also contains the online pipeline. The most important module of the DAF will be the Trend Analysis to identify and fit the systematic errors and feed them back into the data reduction. This requires an infrastructure which allows for comprehensive access to all raw and derived data and enough flexibility to easily introduce new algorithms in the system. We plan to realize this with a database, sophisticated middleware and Application Programmers' Interfaces (APIs) for the algorithms and user interface plug-ins. We present in this paper the requirements and preliminary design of the DAF, as well as the implementation issues concerning the integration with other modules of the DRL and ESO compliance.

We present here the general formalism and data processing steps used in the data reduction pipeline of the AMBER instrument. AMBER is a three-telescope interferometric beam combiner in J, H and K bands installed at ESO's Very Large Telescope Interferometer. The fringes obtained on the 3 pairs of telescopes are spatially coded and spectrally dispersed. These are monitored on a 512x512 infrared camera at frame rates up to 100 frames per second, and this paper presents the algorithm used to retrieve the complex coherent visibility of the science target and the subsequent squared visibility, differential phase and phase closure on the 3 bases and in the 3 spectral bands available in AMBER.

This paper describes a standard for exchanging calibrated data from optical (visible/infrared) stellar interferometers. The standard is based on the Flexible Image Transport System (FITS). The formal definition of the standard is contained in a separate document, the Format Specification. The latest version of the Format Specification is available from the website http://www.mrao.cam.ac.uk/~jsy1001/exchange/. This document gives an overview of the format, and explains some of the decisions taken in designing it.

Numerical simulations of interferometers with spatial filters and large apertures are presented. These simulations are particularly aimed at measuring the fringe (or OPD) motion caused by atmospherically induced wavefront corrugations (high order Zernike modes) across the telescope apertures. This component of the fringe motion results from the coupling in a spatial filter of high-order Zernike modes to the fringe motion. The Piston Mode component is found to be an inadequate approximation to the OPD in an interferometer with large apertures, spatial filters and tip-tilt correction. The fringe motion at high temporal frequencies is dominated by the effects of wavefront corrugations across the telescope apertures.

Aperture Masking Interferometry (AMI) is one of the high-resolution astronomical image observation technologies. It is also an important research way to the Optical Aperture Synthesis (OAS). The theory of OAS is simply introduced and AMI simulation method is raised. The mathematics model is built and the interferogram fringes are got. The aperture mask u-v coverage is discussed and one image reconstruction method is done. The reconstructed image result is got with CLEAN method. Shortcoming of this work is also referred and the future research work is mentioned at last.

Two competitive design studies for the Ground-based European Nulling Interferometer Experiment (GENIE) have been initiated by the European Space Agency and the European Southern Observatory in November 2003. The GENIE instrument will most probably consist of a two-telescope Bracewell interferometer, using the 8-m Unit Telescopes and/or the 1.8-m Auxiliary Telescopes of the VLTI, and working in the infrared L' band (3.5 - 4.1 microns). A critical issue affecting the overall performance of the instrument is its capability to compensate for the phase and intensity fluctuations produced by the atmospheric turbulence. In this paper, we present the basic principles of phase and intensity control by means of real-time servo loops in the context of GENIE. We then propose a preliminary design for these servo loops and estimate their performance using GENIEsim, the science simulation software for the GENIE instrument.

Knowledge of the dispersion due to (humid) air in the light path of the Very Large Telescope Interferometer (VLTI) is crucial to obtaining good science data from MIDI, PRIMA and GENIE. To calculate the refraction due to air at infra red wavelengths in the ducts and delay line tunnel, the temperature and humidity has to be monitored during observations. To accomplish these measurements an easy to use and reliable system was assembled, based on commercially available components. In-house calibration of four humidity and temperature sensors of the system was done in Leiden. A test and calibration program was carried out to make sure that they work reliably and accurately and to determine the sensor characteristics. For this purpose a calibration box was designed which isolates the sensors from the environment so that there is no exchange of air with the outside environment. Using constant humidity salt solutions, the humidity in the box can be controlled. This allows the calibrations to be carried out for typical values of relative humidity and temperature at Cerro Paranal. Calibration of the sensors includes: 1. Reducing the systematic relative humidity differences between the sensors to less than 0.1 % and 2. Reducing the systematic temperature differences between the sensors to less than 0.01 K. In this paper we will present the outcome of the calibrations and the future of the sensors at Paranal.

We report on the current status of a horizontal path length laser propagation campaign that is being performed over the sea, just off the coast of Puerto Rico. The effects of atmospheric turbulence in a tranquil marine environment have been measured with a video rate Shack--Hartmann wavefront sensor. The small perturbations in the wavefront phase and the degree of scintillation are presently being determined in a single pass from source to receiver, over a trial distance of 110 meters. Additional sites have been identified which allow for single pass measurements up to approximately 1 kilometer. Over 70 hours of data have been sampled to date, between December 2003 and April 2004.

Observatories are affected by dust particles in the air and those that settle on the mirror surfaces. Here we analyze seven years of weather and dust particle data from the Apache Point Observatory (APO) in Sunspot, New Mexico, to find correlations between dust and other weather factors, and also find correlations between the APO data and the Magdalena Ridge Observatory (MRO) weather data. Exposure of optics to strong dust events should be avoided, so we considered different possible observatory clousre criteria based on these data.

The Magdalena Ridge Observatory is a congressionally funded project to deliver a state-of-the-art observatory on the Magdalena Ridge in New Mexico to provide astronomical research, educational and outreach programs to the state. In this paper we report results from one of our undergraduate projects being run at New Mexico Tech. This project focuses on the design and characterization of a novel instrument for sensing the atmospheric flow instabilities related to seeing at the observatory site. The instrument attempts to find the power of turbulence on millisecond time scales by measuring a voltage difference between two active microphones. The principles behind the instrument are explored here and a description of the limitations of the current experimental implementation is given. Initial results from the experiment are presented and compared with simultaneous measurements from a co-located Differential Image Motion Monitor. The instrument is shown to be a valuable and robust tool for monitoring the atmospheric conditions during site testing campaigns, but further data will be needed to confirm the precise nature of the correlation between measurements made with this system and more conventional seeing metrics.

The astronomical site parameters for the Magdalena Ridge Observatory (MRO) are being studied from numerous aspects including meteorological, environmental, seismic and sky quality (e.g. "seeing", cloud cover). Results to date indicate that MRO is an excellent site for astronomical observing. Seeing measurements of less that 1 arc second in the optical are routinely obtained. Seismic conditions on the mountain ridge are below levels that will cause any major problems for construction and operation of an optical interferometer. Nighttime "allsky" camera imagery indicates a large percentage of clear nights.

The DIMMWIT (Differential Image Motion Monitor, Which Is Transportable) is a portable DIMM that can measure the Fried parameter r₀ and the average wind speed of the turbulent layers. Analysing DIMM images to calculate r₀ is a standard procedure, but wind speeds have rarely been calculated from differential image motion before. Here, we describe how wind speeds can be derived from either differential image motion power spectra or differential image velocities. The DIMMWIT wind speeds are then compared with a wind speed derived from the coherence times, t₀, of interferometric fringes recorded simultaneously at COAST (Cambridge Optical Aperture Synthesis Telescope). Although t₀, and hence the wind speed, is routinely measured by the interferometer at the COAST site, the Fried parameter had not been studied. The results of seeing campaigns at COAST and MROI (Magdalena Ridge Observatory Interferometer) are presented, along with a comparison of DIMMWIT r₀ measurements with the FWHM of long exposure images recorded at the same time.

We present an experiment to measure the thermal background level and its fluctuations with the European Southern Observatory (ESO) Very Large Telescope Interferometer (VLTI). The Mid Infrared Instrument (MIDI) operating between 8 and 12 micron was used in both dispersed and non-dispersed modes. By using an interferometric instrument, in non-interferometric mode, we probe the same optical path as can be expected for other infrared interferometric instruments, e.g. GENIE and MIDI itself. Most of the infrared thermal background detected with MIDI originates from the VLTI infrastructure. This can be attributed to the absence of a pupil re-imaging mirror. Only for a small region around the optical axis of the system the signal from the VLTI infrastructure can be considered small and the atmospheric background fluctuations can be characterized. The fluctuations of the thermal emission are described in terms of their power spectral densities (PSD). We have identified two regions in the PSD. For the low frequency range (0-10 Hz) the fluctuations are dominated by the Earth atmosphere. The slope of the log-log PSD is close to -1. For the high frequency (larger than 10 Hz) range the fluctuations are due to photon noise and the PSD flattens off. Many narrow peaks are present in the PSD. Peaks at 1 and 50 Hz occur in almost all data sets and are identified as the effects of the MIDI closed cycle cooler and the power lines respectively. Other peaks at 10 and 30 Hz, as well as peaks above 50 Hz, are assumed to be VLTI or MIDI-specific frequencies.

The 'OHANA aims at demonstrating the potential of single-mode fibers to make large telescope arrays with kilometric baselines. In this respect, propagation in single-mode fibers is critical. Transmission needs to be excellent, differential dispersion needs to be cancelled and polarizations have to be matched. 2x300 m of fibers have been produced for the first 'OHANA baselines. Their transmission has proved to be excellent and their differential dispersion has been almost perfectly cancelled thus bringing the fundamental components of the experiment to reality. The work reported in this paper is part of a global work on fibers in the 'OHANA project. Vergnole et al. describe the results achieved on silica fibers in the astronomical J and H bands.

This article deals with one of the important instrumentation challenges of the stellar interferometry mission IRSI-Darwin of the European Space Agency: the necessity to have a reliable and performant system for beam combination has enlightened the advantages of an integrated optics solution, which is already in use for ground-base interferometry in the near infrared. Integrated optics provides also interesting features in terms of filtering, which is a main issue for the deep null to be reached by Darwin. However, Darwin will operate in the mid infrared range from 4 microns to 20 microns where no integrated optics functions are available on-the-shelf. This requires extending the integrated optics concept and the undergoing technology in this spectral range. This work has started with the IODA project (Integrated Optics for Darwin) under ESA contract and aims to provide a first component for interferometry. In this paper are presented the guidelines of the characterization work that is implemented to test and validate the performances of a component at each step of the development phase. We present also an example of characterization experiment used within the frame of this work, is theoretical approach and some results.

Since the pioneering work of Haniff et al. (1987), aperture-masking interferometry has been demonstrated on large class telescopes. The usual implementation lays in the avoidance of redundancies in the pupil plane, which, in presence of aberrations and turbulence, depress the transfer function of the telescope. In a recent experiment on Keck I, a non-redundant pupil geometry allowed diffraction-limited imaging, with dynamic range in excess of 200:1 (Tuthill et al., 2000). Yet, the final image quality is still limited by the optical defects induced by turbulence in sub-pupils. We propose to overcome this issue by using the same technique of spatial filtering by single-mode fibers that we have used in long-baseline interferometry. Each sub-pupil element is focused in a single-mode fiber thus eliminating spatial phase fluctuations and trading these against instantaneous intensity fluctuations which can be directly measured. Therefore, each sub-pupil becomes spatially coherent. Simulations show that the dynamic range would be dramatically increased. Moreover, the idea of using fibers in the pupil plane could lead to outstanding prospects, like filtering the whole aperture, sub-divided into a filled array of sub-apertures.

We present a laboratotry interferometer simulator specifically designed to characterize prototype integrated optics combiners. In the current configuration it allows to simulate a complex object made of luminous points and observe it with a reconfigurable array made of three telescopes. We used a three-way integrated optics similar to the one used at IOTA to combine the beams. We describe it in detail and present first validation measurements and the first measurements of a binary star. This work paves the way for an up-to eight telescope simulator capable of simulating the VLTI and the VITRUV focal instrument.

Second generation VLTI instruments will be able to use of the array full imaging capability with up to 8 telescopes. Such an instrument will allow astronomers to measure 28 visibilities and 21 independent closure-phases at the same time, providing therefore rapidly imaging abilities with a spatial resolution of one milliarcsecond in the near infrared range. The VITRUV project is a proposition to achieve the VLTI interferometric combination thanks to single-mode planar optics (the so-called integrated optics, IO). IO technologies allow to design integrated combiners with remarkable stability and self allignement properties. In addition, modal filtering associated with photometric calibration will lead to accurate visibility and closure-phase measurements. In this paper we present a detailed analysis of beam combination concepts that takes into account several constraints: throughput, signal to noise ratio, interferometric efficiency, integrated optics circuit design constrains and astrophysical requirements for imaging mode.

We present the design of the Michigan Infra-Red Combiner (MIRC). MIRC is planned for deployment at the Georgia State University CHARA array to simultaneously combine all six telescope beams in an image-plane combiner. The novel design incorporates spatial-filtering with single-mode fiber optics, a synthetic (densified) pupil, and a low-resolution spectrometer to allow good calibration and efficient aperture synthesis imaging in the near-infrared. In addition, the focalization and spectrometer optics can accommodate an integrated optics component with minimal re-alignment. The MIRC concept can be scaled-up for interferometer arrays with more telescopes.

This paper describes the principle function and the possible applications of a new micropositioning device for the optical fiber, which aligns it precisely to a light source, with a resolution better than 100 nm. One end of an optical fiber is fixed to one end of a piezoelectric tube. The electrical voltage applied to the 5 external electrodes around the piezoelectric tube will create transverse motion (up till +/- 20 μm) and longitudinal motion (of 1 to 2 μm) and the optical fiber fixed to this tube will make the same motion. The other end of the optical fiber passing through the tube fixed to a support is connected to a photometer, which measures the light intensity. The measure allows determining the best voltage for the command of the 5 electrodes with a help of programmed algorithms. Small dimension and very short time response of this device would allow multiple applications for the light injection in a wave-guide. The first application is related to the guide to guide light coupling, for the automatic centering of two optical fibers, and a fiber to the input of an integrated optics beam combiner. The second application concerns pupil's fragmentation and second generation VLTI instruments. The alignment of height optical fibers with an object of the sky, coming from height telescopes or height sub-pupils of one telescope, could be controlled independently and in real time. The light coupling into every fiber and the optical length path are micro-adjusted in an optimal way, in spite of atmospheric turbulence effects.

We are working towards imaging the surfaces and circumstellar envelopes of Mira stars in the near-infrared, using the IOTA interferometer and the IONIC integrated-optics 3-beam combiner. In order to study atmospheric structures of these stars, we installed 3 narrow-band filters that subdivide H-band into 3 roughly equal-width sub-bands - a central one for continuum, and 2 adjacent ones to sample Mira star's (mostly water) absorption-bands. We present here our characterization of the IOTA 3-Telescope interferometer for closure-phase measurements with broad and narrow-band filters in the H atmospheric window. This includes characterizing the stability, chromaticity, and polarization effects of the present IOTA optics with the IONIC beam-combiner, and characterizing the accuracy of our closure phase measurements.

AMBER is the focal near-infrared instrument of the VLTI combining 2 or 3 telescopes in the J, H and K bands with 3 spectral resolution modes. It uses single-mode fibers to ensure modal filtering and high measurement accuracies. AMBER has been integrated and tested in Grenoble during 2003. We report in this paper the lab performances of the instrument in terms of instrumental contrast, measurement accuracy and stability, and throughput.

In the frame of 'OHANA project (phase II), IRCOM institute is in charge of the 300-meters-long silica fibres devoted to link the Canada France Hawaii Telescope and the GEMINI telescope over J and H-band. In this paper, we report a method to compensate the differential chromatic dispersion between two 300m-long silica polarization maintaining fibres. It consists in adding short fibre length (from 0 to 3 m) on the less dispersive arm. We experimentally demonstrate that the chromatic differential dispersion is different for the two neutral axis of the polarization-maintaining fibre and that a tradeoff between order 2 and 3 of the chromatic dispersion is necessary to reach the best results. Moreover, thanks to a channeled spectrum analysis and using a Taylor series expansion, the spectral phase is analyzed around a mean frequency. Using these data, it is possible to foresee the evolution of the fringes contrast as a function of the additional fibre length. These simulations fit properly the measurements of the fringes contrasts demonstrating the good modeling. At least, we simulate the variation of the contrast as a function of wavelength and spectral resolution over a given spectral window. Consequently, we are able to adjust the fibre lengths in order to perform a coherent linkage maximizing the contrast over J and H-band. The work reported in this paper is part of a global work on fibers in the 'OHANA project. In another paper, Kotani et al. describe the results achieved on fluoride glass fibers devoted the astronomical K-band.

The Photonic Crystal Fibres (PCF) are microstructured waveguides currently developed in the frame of fibre telecommunications. This study is mainly focused on the improvement of dispersion property and wide spectral single-mode operating domain. Moreover, these fibres are highly birefringent and thus are able to maintain polarization. Consequently, in the astronomical context, this kind of fibre is a good candidate to design a fibre linked version of stellar interferometer. In this paper, we experimentally study the potential of these fibres taking advantage of the wide spectral single-mode operation. We propose an experimental setup acting as a two-beam interferometer using PCF. In a first time, the aim of this experiment is to measure fringes contrast at four different wavelengths (670 nm, 980nm, 1310 nm and 1543 nm) corresponding to four spectral band (R, I, J and H-band) with the same couple of PCF samples.

To achieve high rejection ratios with a nulling interferometer, the beams to be combined have to be as equal as possible, especially concerning the transverse field distribution. This requires highly symmetric beam recombination stages as well as modal wavefront filtering. Usually co-axial beam recombination employing beam splitters is used, ideally in highly symmetrizing double-pass configurations. Considering that a single-mode waveguide is required anyway for spatial wavefront filtering, it seems reasonable to incorporate the functionalities of beam recombination and modal filtering in a single device. If the beams to be combined are injected multi-axially into the single-mode waveguide, co-axial beam recombination is obtained within the waveguide. We show that such beam recombination stages may outperform conventional symmetric co-axial realizations with respect to optical throughput and system complexity.

We present the final results of the injection tests of adaptive optics corrected light into single mode fibers, conducted at Canada-France-Hawaii Telescope, W.M. Keck Observatory and Gemini North Observatory, in the prospect of 'OHANA Phase I (preparatory phase). We emphasize on the impact of the results on both the 'OHANA Phase II (interferometric demonstration phase) sensitivity and observational protocol and the behavior of the different adaptive optics.

LINC-NIRVANA is a near-­infrared (1-­2.4 micron) beam-­combiner instrument for the Large Binocular Telescope (LBT). LINC-NIRVANA is being built by a consortium of groups at the Max-­Planck-­Institut fur Astronomie in Heidelberg, the Osservatorio Astrofisico di Arcetri in Florence, the Universitat zu Koln, and the Max-­Planck-­Institut fur Radioastronomie in Bonn. The MPI fur Radioastronomie is responsible for the near­-infrared detector for the fringe and flexure tracking system (FFTS). We describe the design and construction of the detector control electronics as well as the first laboratory measurements of performance parameters of the NIR detector for the fringe and flexure tracking system of the LBT LINC-NIRVANA instrument. This detector has to record LBT interferograms of suitable reference stars in the FOV at a frame rate of the order of 200 frames per second using, for example, 32 x 32-­pixel subframes. Moreover, special noise reduction techniques have to be applied. The fringe-­tracker interferograms are required for monitoring and closed-­loop correction of the atmospheric optical path difference of the two LBT wavefronts (see C. Straubmeier et al., "A fringe and flexure tracking system for LINC-NIRVANA: basic design and principle of operation"). We will describe our laboratory measurements of maximum frame rate, readout noise, photometric stability, and other important parameters together with first measurements of laboratory simulations of LBT interferograms.

The correction of atmospherical differential piston and instrumental flexure effects is mandatory for full interferometric performance of the LBT NIR interferometric imaging camera LINC-NIRVANA. This is the task of the Fringe and Flexure Tracking System (FFTS), which is part of the contribution of the I. Physikalische Institut of the University of Cologne to the project. Differential piston and flexure effects will be detected and corrected in a real-time closed loop by analyzing the PSF of a guide star at a frequency of up to several hundred Hz. Numerous critical design parameters for both FFTS hardware and control loop have to be derived from simulations. Detailed knowledge of the special shape of the LBT interferometric PSF as a function of a variety of parameters is required to design the fringe tracking control loop. In this paper we will show the results of our software that allows us to generate polychromatic interferometric PSFs for a number of different scenarios. Our fringe detection algorithm is based on an analytic model which is fitted to the acquired PSF. We present the results of the evaluation of the algorithm in terms of speed and residual piston, as well as the first successful implementation of the algorithm in a closed loop system. Simulations of the time evolution of differential piston have been performed in order to investigate necessary correction frequencies and the variation of differential piston across the usable field of view. These simulations are based on the Layer Oriented Adaptive Optics performance simulator "LOST" of the Osservatorio Astriofisico di Arcetri.

The VLTI system foresees two generations of fringe sensor: FINITO and PRIMA FSUs. The former is dedicated to H band; it controls the internal OPD with a temporal modulation with an external reference OPD. The latter, working with the ABCD model and in K band, is based on the introduction of known phase offsets for the interferometric signal (spatial phase modulation) and on the measurement of the corresponding combined power. Simulation models for both FSUs are developed with Matlab. Instrumental parameters, i.e. phase, transmission, visibility, are tabulated for ease of maintenance and to speed execution time. For the use of siderostats, due to fast turbulence, the need for intensity calibration arises. Assuming slow intensity variations with respect to phase variations, different algorithms can apply, yielding to numerical control of perturbations as a function of model parameters.

We have developed a method for performing long coherent integrations with the Navy Prototype Optical Interferometer (NPOI), which is based on fitting a model fringe pattern to the NPOI data frames. The procedure is quite computationally intensive, but gives a better estimation of the phase than the conventional method of location the peak of the group delay power. We mention briefly some of the most important past work on coherent integration, and then describe our method. We conclude that the fitting approach produces a phase with fewer outliers than the Fourier-transform group delay approach. We show how the instrumental squared visibility varies as a function of the fringe model used, and show that it provides a better SNR than the FT method. The phase determination will always be imperfect, and thus cause a reduction in the visibility amplitude relative to the true instrumental visibility. We illustrate a method for calibrating the visibility amplitude. With long coherent integrations the phase of the visibility can be extracted. We show examples of visibility phases and how to correct them for phase variations in the instrument. Finally, we illustrate a method for measuring stellar diameters very precisely, to one part in at least several hundred.

LINC-NIRVANA is the interferometric near-infrared imaging camera for the Large Binocular Telescope (LBT). Operating at JHK bands LINC-NIRVANA will provide an unique and unprecedented combination of high angular resolution (~9 milliarcseconds at 1.25 µm), wide field of view (~100 arcseconds² at 1.25 µm), and large collecting area (~100 m²). One of the major contributions of the I. Physikalische Institut of the University of Cologne to this project is the development of the Fringe and Flexure Tracking System (FFTS). In close cooperation with the Adaptive Optics systems of LINC-NIRVANA the FFTS is a fundamental component to ensure a complete and time-stable wavefront correction at the position of the science detector in order to allow for long integration times at interferometric angular resolutions. Using a dedicated near-infrared detector array at a combined focus close to the science detector, the Fringe and Flexure Tracking System analyses the interferometric point spread function (PSF) of a suitably bright reference source at frame rates of several hundred Hertz up to 1 kHz. By fitting a parameterized theoretical model PSF to the preprocessed image-data the FFTS determines the amount of pistonic phase difference and the amount of an angular misalignment between the wavefronts of the two optical paths of LINC-NIRVANA. For every exposure the correcting parameters are derived in real-time and transmitted to the respective control electronics, or the Adaptive Optics systems of the single-eye telescopes, which will adjust their optical elements accordingly. In this paper we present the opto-mechanical hardware design, the principle of operation of the software control algorithms, and the results of first numerical simulations and laboratory experiments of the performance of this Fringe and Flexure Tracking System.

Interferometric fringes are traditionally decomposed using the Discrete Fourier Transform (DFT). However, the application of the DFT is only correct in cases where the fringes are sampled evenly with delay and over integer number of fringe periods. This ideal case is often not achieved in Optical Interferometry. Fringe spectrography, non-linear fringe sweeps and image-plane beam combiners are typical cases of where the DFT approach fails to make most efficient use of the data. The authors assert that in many cases alternative and more efficient fringe decompositions exist but which may exhibit considerably different noise behaviour to the DFT. The authors present the mathematical results important for correcting for statistical bias in the powerspectrum and bispectrum constructs of a completely general fringe decomposition. An estimator for the noise in the powerspectrum has also been derived. The authors believe this to be the first analytical derivation of statistical bias and noise in interferometry that treats both photon counting noise as well as Gaussian read out noise.

An automatic fringe tracking system has been developed and implemented at the Infrared Optical Telescope Array (IOTA). In testing during May 2002, the system successfully minimized the optical path differences (OPDs) for all three baselines at IOTA. Based on sliding window discrete Fourier transform (DFT) calculations that were optimized for computational efficiency and robustness to atmospheric disturbances, the algorithm has also been tested extensively on off-line data. Implemented in ANSI C on the 266 MHz PowerPC processor running the VxWorks real-time operating system, the algorithm runs in approximately 2.0 milliseconds per scan (including all three interferograms), using the science camera and piezo scanners to measure and correct the OPDs. Preliminary analysis on an extension of this algorithm indicates a potential for predictive tracking, although at present, real-time implementation of this extension would require significantly more computational capacity.

Recent works on interferometric beam recombination techniques for direct imaging have triggered studies for using them as exo-planet hunting instruments. This work describes these techniques and study their specific aspects, such as image formation (on- and off-axis), field contamination by nearby sources and their suitability for exo-planet detection in combination with coronagraphic/nulling techniques. Ongoing plans to implement them in a mid-term future are outlined.

In the framework of the Phase-Referenced Imaging and Micro-arcsecond Astrometry facility (PRIMA) developed for the Very Large Telescope Interferometer (VLTI), a sophisticated opto-mechanical system has been developed by TNO-TPD. It will be placed at the Coudé focus of the telescopes and will allow picking up two stars anywhere in a 2 arcmin field-of-view and collimating their light into two beams that will propagate through the rest of the interferometer toward the instrument. These Star Separator systems have a very high optical quality, fast and accurate pointing and chopping, independent high speed remote control of the beam tip-tilt and of the pupil position. They are very rigid, accurate mechanical systems non-sensitive to temperature variations The Star Separator systems are described in this paper.

APreS-MIDI (APerture Synthesis in the MID-Infrared) instrument function is to recombine 4 telescope beams of the VLTI. Interference fringes are sampled in the pupil plane. The optical principle uses "image densification". It is perfectly adapted for reconstructing images by aperture synthesis at 10mm. This principle could be used for building a new generation 10mm instrument, but instead of making a totally new instrument, we propose the design of an optical module that can supply the current MIDI-VLTI instrument with 4 beams.

The Very Large Telescope Interferometer (VLTI) makes in its final configuration use of four 1.8m Auxiliary Telescopes, which can be located on 30 different stations. These four telescopes can theoretically be arranged in more than 25,000 different configurations. Of course, operational constraints will allow only some dozens of these configurations to be realized over the entire lifetime of the interferometer. Furthermore, there are restrictions on sky accessibility posed by both physical limits of the delay lines and vignetting by the 8.2m telescope enclosures. We describe criteria for an optimum selection of configurations and propose a subset of AT stations to be offered for science operations with the VLTI.

We have enhanced the spectral resolution of the Navy Prototype Optical Interferometer (NPOI) at the H-alpha line to 3 nm (FWHM). We use customized filters that suppresses light in the ~600-725 nm window except for light at the H-alpha wavelength (656.3 nm). The bands shortward of 600 nm and longward of 725 nm are used for fringe tracking and for calibrating the system fringe visibility. We have used these filters to observe H-alpha emission from circumstellar material around Be stars. Closure phases from our initial observations of the Be star zeta Tau with three array elements suggest that the H-alpha emission is not centered on the star. We will show these three-element results, as well as recently-acquired data from the NPOI using 4, 5, and 6 stations.

Previous work by Hindsley and Mozurkewich (2001) showed that analysis of the Modulation Transfer Function (MTF) demonstrated the proportionality of signal-to-noise in a sparse aperture to the fill factor of the aperture. Analysis of the MTR also could enumerate the noise amplification characteristics of particular sparse apertures. However, such image quality metrics as the General Image Quality Equation (GIQE) also include edge effects, basically due to ringing and reduction in the edge sharpness. Here we report on our analysis of the MTF in order to quantify the relationship between the other terms in the GIQE and the structure of the MTF. We find that, for a fixed amount of optical surface, the image quality will improve with decreasing fill fraction due to an increase in resolution. Apodization of the Wiener Filter used to restore the image, as advocated by Hindsley and Mozurkewich, does not result in an improved image quality; use of the traditional unapodized Wiener Filter is highly favored. While the GIQE does not appear very sensitive to input signal-to-noise ratio (SNR), the input SNR does limit the ability to successfully reconstruct the image and is the ultimate limiting constraint on the fill fraction.

An Extremely Large Synthesis Array (ELSA) with 27 ten-meter telescopes and baseline lengths up to 10 km would provide completely new insight into many astrophysics phenomena. It could be used to obtain resolved images of nearby brown dwarfs which would reveal weather phenomena in their atmospheres, to give detailed pictures of stellar surfaces, interacting binaries, and circumstellar material, to study general-relativistic effects on the orbits of stars near the center of our Galaxy, to obtain "movies" of expanding supernovae, to image the broad-line regions of active galaxies, and to measure the geometry of the fireballs producing the afterglow of gamma-ray bursts. Observations of faint objects will be possible by using an external reference star (within the isoplanatic angle) to co-phase the array. Telescopes with large diameters are essential to provide good sky coverage in this observing mode. The use of optical fibers for beam transport and delay compensation is highly desirable, as this eliminates the need for an expensive beam train with meter-sized optical elements, and a very large vacuum system. The most challenging aspect of fiber-coupled interferometry is dispersion in the fibers, which has to be eliminated or compensated precisely. Advances in telescope technology and fiber optics expected for the next decade may bring the cost of a facility similar to the ELSA concept discussed here into a range that would be affordable as an international project.

The goal of future optical aperture synthesis telescopes is to achieve model independent imaging of complex sky structure with the success demonstrated in very long baseline radio interferometry. Tomorrow's optical interferometers must therefore measure both the powerspectrum and the bispectrum of a source and with greater baseline coverage than has so far been achieved. In contrast to VLBI the bispectrum in optical interferometry is not as readily obtained as the powerspectrum components. Although it is clear that image reconstruction in general cannot do without the bispectrum measurements, very little intuition exists on how many and which bispectrum components in particular are most important to record. Such knowledge has implications in the design of the beam combiner and beam handling optics of future interferometers. The authors present results of image reconstruction from simulated optical interferometry data containing a fixed amount of powerspectrum but a varying selection of bispectrum components.

Recent site seeing testing campaigns conducted by our team from University of Nice1 show that Dome C represents the best site on Earth for astronomical high angular resolution (HAR) observations at optical and IR wavelengths. The dramatic gain over relevant HAR parameters r₀, L₀, θ₀ and τ₀, added to very low temperatures during the polar winter nights (-70°C), the dry atmosphere and the possibility of continuous observations during several nights make Dome C the ideal site for deploying a kilometric optical interferometer before the 2015 horizon. Here we describe the concept of Kiloparsec Explorer for Optical Planet Search (KEOPS) that is studied by our group at LUAN. KEOPS is an interferometric array of 36 off-axis telescopes, each 1.5m in diameter. Its kilometric baselines open sub-mas snap-shot imaging possibilities to detect and characterize extra-solar planetary systems, especially exo-Earths out to 300 parsecs from the visible to the thermal IR. KEOPS can be considered as a DARWIN/TPF challenger but at a much lower cost.

Homothetic mapping is an aperture synthesis technique that allows interferometric imaging over a wide field-of-view. A laboratory experiment was set up to demonstrate the feasibility of this technique. Here, we present the first static experiments on homothetic mapping that have been done on the Delft Testbed for Interferometry (DTI). Before a changeable telescope configuration is provided, we first took a fixed telescope configuration and tested the algorithms for their ability to provide an exit pupil configuration before beam combination, that was an exact copy of this telescope configuration. By doing so, we created a homothetic imaging system. This is an imaging system that acts as a masked aperture monolithic telescope, but consists of (in our case) three telescopes of which the light follow their own optical trains.

The Delft Testbed Interferometer (DTI) will be presented. The basics of homothetic mapping will be explained together with the method of fulfilling the requirements as chosen in the DTI setup. The optical layout incorporates a novel tracking concept enabling the use of homothetic mapping in real telescope systems for observations on the sky. The requirements for homothetic mapping and the choices made in the DTI setup will be discussed. Finally the first results and the planned experiments will be presented.

This paper presents progress made regarding the field to resolution ratio for aperture synthesis interferometers. In order to overcome a limit established for the field to resolution ratio of interferometric arrays, we propose an interferometer configuration which allows a better coverage of the spatial frequency plane. This setup requires large sub-apertures, which can be built more easily with a diffractive Fresnel plates than with large mirrors. We compare a dense array of 9 Fresnel sub-apertures, which gives a snapshot field-resolution ratio of 400, versus a sparse array of 150 small apertures, which yields a field-resolution ratio of 150.

Deep, stable nulling of starlight requires careful control of the amplitudes and phases of the beams that are being combined. The detection of earth-like planets using the interferometer architectures currently being considered require that the electric field amplitudes are balanced at the level of ~ 0.1% and the phases are controlled at the level of 1 mrad (corresponding to ~ 1.5 nm for a wavelength of 10 microns). These conditions must be met simultaneously at all wavelengths across the science band and for both polarization states, imposing unrealistic tolerances on the symmetry between the optical beamtrains. Lay et. al. proposed the concept of a compensator that is inserted into the beamtrain, which can adaptively correct for the mismatches across the spectrum enabling deep nulls with realistic, imperfect optics. This proposed design uses a deformable mirror to adjust the amplitude and phase of the electric field that couples into the single-mode spatial filter. We have demonstrated amplitude and phase control at a single wavelength in the near-IR. We are preparing to demonstrate control with our deformable mirror actuator in the near-IR and in parallel are preparing a demonstration in the mid-IR where the compensator will be required to operate.

Suppression of starlight by nulling interferometry has the potential to enable the direct detection of planets around nearby stars, and several interferometer configurations have been proposed which meet, to varying degrees, the numerous associated experimental goals, such as high sensitivity, high angular resolution, high stellar rejection, and extraneous signal rejection. However, the more capable approaches involve fairly complex systems, so that minimization of complexity is vital if a space-based nulling interferometry mission is to become a reality. The simplest case, the two-aperture nulling interferometer, possesses fairly limited capabilities, but most other configurations being considered for nulling interferometry missions involve four or more apertures. Another relatively simple case which has not received much attention to date is that of the three-aperture nulling interferometer. In this paper, the three-telescope case is discussed in detail, from the necessary triple-beam nulling beamcombiners, to the properties of the pupil plane configurations. It is concluded that three-telescope nullers share many of the desirable capabilities of four-telescope nullers, such as a phase modulation capability, and so in fact represent a very capable "simplest case" nulling interferometer configuration.

One of the science goals of NASA's Navigator program is ground-based narrow-angle astrometry for extra-solar planet detection, which could be done as part of the proposed Outrigger Telescopes Project. The narrow-angle measurement process, which would use the outrigger telescopes, starts with the determination of the conventional interferometer astrometric baseline, determined from wide-angle astrometry of Hipparcos stars. A baseline monitor system would be employed at each outrigger telescope. This system monitors the pivot point of each telescope - the end point of the astrometric baseline - to measure telescope imperfections that would cause the baseline to vary with telescope rotation. The baseline monitor includes azimuth and elevation cameras that monitor runout along the azimuth and elevation axes of the telescopes. In conjunction with the baseline monitor system, a pivot monitor camera in the dual-star module is used to register the laser metrology corner-cube reflector to the telescope pivot, tying the narrow-angle baseline, which applies to the narrow-angle astrometric measurement, to the wide-angle baseline. In this paper we present the proposed designs for the baseline monitor and pivot-point camera.

Two years ago, the FLUOR interferometric beam combiner moved from IOTA (Infrared Optical Telescopes Array, Mount Hopkins, AZ) to the Center for High Angular Resolution Astronomy (CHARA) Array (Mount Wilson, CA). Apart from offering the largest baselines in the northern hemisphere, this array can be fully operated remotely to allow observations from a distant place. We present here the automations added to the FLUOR hardware, as well as software modifications made in order to allow us to observe from Paris Observatory. We required the remote service to be as reactive as local observations, implying frequent communications between the instrument and the remote observer. We took particular attention to the available bandwidth and reactivity imposed by the secured connection (Virtual Private Network). The first tests are presented.

MIDI (MID-infrared Interferometric instrument) gave its first N-band (8 to 13 micron) stellar interference fringes on the VLTI (Very Large Telescope Interferometer) at Cerro Paranal Observatory (Chile) in December 2002. An lot of work had to be done to transform it, from a successful physics experiment, into a premium science instrument which is offered to the worldwide community of astronomers since September 2003. The process of "paranalization", carried out by the European Southern Observatory (ESO) in collaboration with the MIDI consortium, has aimed to make MIDI simpler to use, more reliable, and more efficient. We describe in this paper these different aspects of paranalization (detailing the improvement brought to the observation software) and the lessons we have learnt. Some general rules, for bringing an interferometric instrument into routine operation in an observatory, can be drawn from the experience with MIDI. We also report our experience of the first "service mode" run of an interferometer (VLTI + MIDI) that took place in April 2004.

The visibility science mode of the Keck Interferometer fully transitioned into operations with the successful completion of its operational readiness review in April 2004. The goal of this paper is to describe this science mode and the operations structure that supports it.

The VLTI now has performed three years of science operations using the VINCI instrument since the first fringes on a star were obtained on March 17, 2001. Since December 5th, 2001, shared risk science observations have been performed with VINCI. In April 2004 (period 73) we have started science operations with the MIDI instrument. Subsequently both the AMBER instrument and the Auxiliary Telescopes (ATs) will be also running under the science Operations at Paranal and offered to the astronomical community. We will present how the VLTI Science operations currently are performed and integrated into the general Paranal Science Operations scheme, using the extensive experience of Service Mode operations performed by the Paranal Science operations group. We focus on the execution of the Service mode operations, how they are planned, performed, evaluated, and processed and the data finally sent to ESO Garching. The near future developments are also presented and how the new instruments and telescopes will be integrated into the Paranal Science Operations.

The Navy Prototype Optical Interferometry (NPOI) group has started an astrometric search for planets in binary star systems based on the idea of using the binary components as position references for one another and looking for deviations from Keplerian motion. Our search will complement the radial velocity (v_R) searches in three ways. We will observe stars of all spectral types; v_R searches are limited to the FGKM range, where stars exhibit narrow spectral lines. We will search for planets in relatively large orbits (more than about 4 AU) where our method is most sensitive; v_R searches are most sensitive to close-in planets. Finally, we will examine binary star systems, which with a few exceptions have been excluded from v_R surveys. Our targets are binaries with both components in the interferometric field of view, producing a periodic variation in the fringe visibility (V²) across the (u,v) plane. Past NPOI results from closer binaries (separations in the tens of mas) show residuals of tens of microarcseconds about the best-fit orbits. The larger separations we are observing produce more V² oscillations across the (u,v) plane, offering the possibility of higher precision. We discuss the level of precision in test observations and the steps that will be needed to convert precision into accuracy.

We present K-band commissioning observations of the Mira star prototype o Cet obtained at the ESO Very Large Telescope Interferometer (VLTI) with the VINCI instrument and two siderostats. The observations were carried out between 2001 October and December, in 2002 January and December, and in 2003 January. Rosseland angular radii are derived from the measured visibilities by fitting theoretical visibility functions obtained from center-to-limb intensity variations (CLVs) of Mira star models. Using the derived Rosseland angular radii and the spectral energy distributions (SEDs) reconstructed from available photometric and spectrophotometric data, we find effective temperatures ranging from T_eff=3192 +/- 200 K at phase 0.13 to 2918 +/- 183 K at phase 0.26. Comparison of these Rosseland radii, effective temperatures, and the shape of the observed visibility functions with model predictions suggests that o Cet is a fundamental mode pulsator. Furthermore, we investigated the variation of visibility function and diameter with phase. The Rosseland angular diameter of o Cet increased from 28.9 +/- 0.3 mas (corresponding to a Rosseland radius of 332 +/- 38 R_sun for a distance of D=107 +/- 12 pc) at phase 0.13 to 34.9 +/- 0.4 mas (402 +/- 46 R_sun) at phase 0.4. The observational error of the Rosseland linear radius almost entirely results from the error of the parallax, since the error of the angular diameter is only approximately 1%.

We present near-infrared speckle interferometry of the OH/IR star OH 104.9+2.4 in the $K'$ band obtained with the 6m telescope of the Special Astrophysical Observatory (SAO). At a wavelength of λ = 2.12 micron the diffraction-limited resolution of 74 mas was attained. The reconstructed visibility reveals a spherically symmetric, circumstellar dust shell (CDS) surrounding the central star. The visibility function shows that the stellar contribution to the total flux at λ = 2.12 micron is less than ~50%, indicating a rather large optical depth of the CDS. The azimuthally averaged 1-dimensional Gaussian visibility fit yields a diameter of 47 +/- 3 mas (FHWM), which corresponds to 112 +/- 13 AU for an adopted distance of D = 2.38 +\- 0.24 kpc. To determine the structure and the properties of the CDS of OH 104.9+2.4, radiative transfer calculations using the code DUSTY were performed to simultaneously model its visibility and the spectral energy distribution (SED). We found that both the ISO spectrum and the visibility of OH 104.9+2.4 can be well reproduced by a radiative transfer model with an effective temperature T_eff = 2500 +/- 500 K of the central source, a dust temperature T_in = 1000 +/- 200 K at the inner shell boundary R_in = 9.1 R_star = 25.4 AU, an optical depth tau = 6.5 +/- 0.3 at 2.2 micron, and dust grain radii ranging from a_min = 0.005 +/- 0.003 micron to a_max = 0.2 +/- 0.02 micron with a power law with index -3.5. It was found that even minor changes in a_max have a major impact on both the slope and the curvature of the visibility function, while the SED shows only minor changes. Our detailed analysis demonstrates the potential of dust shell modeling constrained by both the SED and visibilities.

AMBER had first light in March 2004. The guaranteed time observations of the AMBER consortium (LAOG, MPIfR, OAA, OCA, UNSA) consists of 87 proposals ranging from cosmology, extragalactic studies, star formation, planetary system, late stages of stellar evolution to physical properties of stars. Some examples, AGN, evolved stars and hot stars are discussed in this paper.

This document shows the first results of the study of the environment of the S star T Sagittarii. Observational constraints are obtained through 10 μm long baseline interferometry with MIDI at the VLTI. Models of the dust envelope are simulated with a monte-carlo radiative transfer code.

Since several years, long baseline infrared interferometry succeeds in providing handful of astrophysical results from model fitting of the visibility measurements alone. Continuing on these encouraging results, and thanks to the development of elaborated recombination scheme which allow to gather stellar light coming from 3 telescopes or more, recent (IOTA/IONIC, NPOI) and new (VLTI/AMBER) interferometers have also access to closure phase measurements as well as to a better (u,v) coverage. When the (u,v) coverage is still insufficient to perform image reconstruction. a least square fit approach is required, taking benefit of the closure phases together with visibility informations. Within this framework, and in the light of the AMBER experiment, we simulate realistic observations of star-planet systems. Computing the statistics of the observables, and then characterizing the performances of this instrument, we investigate the potential of AMBER to detect Jovian planets around sun-like stars by computing Signal to Noise Ratio on the constrained parameters, i.e. the flux ratio and the separation. We focus here on the specific system sun-planet, knowing that the general case will be treated in a forthcoming paper. We particularly study how important is the contribution of the closure phase in the model-fitting process, relatively to the visibility.

AMBER is a 3 beam combiner for the Very Large Telescope Interferometer (VLTI). It will soon add to VLTI tremendous angular resolution, sensitivity and spectral resolution (λ/Δλ) up to 10,000. This combination opens important new opportunities for the study of the close environment of pre-main-sequence stars. In order to understand star formation and its evolution, one needs to solve the problem of ejection and collimation mechanisms in jets from young stars. The importance of jets in pre-main-sequence stars relies on the fact that they regulate its angular momentum. By measuring the jet opening angle at the ejection region we can test models for jet origin. In particular, AMBER will provide crucial information on the mechanisms of mass loss and collimation observed in the most active objects. It will allow, for the first time, the differentiation of competing models for jet origin and collimation, namely the X-wind model of Shu and the disk-wind model of Blandford & Payne. In this paper we compare different jet models presented in the literature.

The Space Interferometry mission's nano-meter class System Testbed has implemented an external metrology system to monitor changes in the length & orientation of the science interferometer baseline vector, which cannot be monitored directly. The output of the system is used in real time fringe tracking of dim stars. This paper describes the external metrology system, its mathematical representation, limitations, and method for estimating the length & orientation of the science baseline vector. Simulations and current system performance are presented and discussed.

We present a 1:3 scale model of the LINC-NIRVANA interferometer. This laboratory Fizeau, or image plane, interferometer allows us to test many aspects of LINC-NIRVANA before the final instrument is integrated. We have used this testbed interferometer to practice alignment procedures, verify the optical design, show that point spread functions with low (10\%) Strehl ratio can maintain high fringe contrast, and test the fringe tracking algorithm by running the interferometer in a closed piston loop.

We describe a laboratory optical prototype being designed and built to conduct, at the Laboratoire Universtaire d'Astrophysique de Nice, France, some patrol experiments to prepare the construction of the interferometric array KEOPS (see these proceedings) at Dome C of Antarctica. Our testbed intends to gain experience on the Interferometric Remapped Array Nulling (IRAN) technique for imaging/nulling optical interferometry in IR (2.5μ-10μ) and its use with the KEOPS project. The optical design, beam-combination and data analysis are described in details. Critical points concerning off-axis, optical delay compensation and F.O. combination are outlined. The recent results obtained in our laboratory on our testbed of IRAN using F.O. bundles are described showing that massively integrated optics are the obvious way to be used in a many-telescope interferometric optical array. Future steps to drive our R&D program are discussed which can also apply to other foreseen optical synthesis arrays like the VLTI or space-borne interferometers.

The Fizeau Interferometer Testbed (FIT) is a ground-based system that will be used for the development and testing of technologies relevant to Stellar Imager (SI) and other sparse aperture/Fizeau imaging interferometer mission concepts. The testbed will utilize image-based wavefront sensing and control to co-phase and maintain closed-loop control over a Sparse Aperture Array (SAA) consisting of spherical mirror elements. The SAA is a re-configurable assembly baselined to incorporate between seven (initially) and thirty 12.5mm diameter (R = 4000mm) mirror elements. In this paper we describe the fabrication, alignment, and initial calibration of the phase I (7 primary elements) FIT hardware and discuss various factors impacting the performance and stability of the testbed.

We discuss the procedure used to characterize the Wide-Field Imaging Interferometry Testbed (WIIT) components and system, including spectral transmission, throughput, wavefront quality, mechanical and thermal stability, and susceptibility to turbulence. The sources of uncertainty and visibility loss are identified and evaluated, and we briefly discuss measures taken to mitigate these effects. We further discuss calibration techniques which can be used to compensate for visibility loss factors, and describe the applicability of these calibration techniques to the future space-based far-IR interferometry missions SPIRIT (Space Infrared Interferometric Telescope) and SPECS (Submillimeter Probe of the Evolution of Cosmic Structure).

Delay lines provide the pathlength compensation that makes the measurement of interference fringes possible. When used for nulling interferometry, the delay line must control pathlengths so that the null is stable and controlled throughout the measurement. We report on a low noise, low disturbance, high bandwidth optical delay line capable of meeting the TPF interferometer optical path length control requirements at cryogenic temperatures.

The Space Interferometry Mission (SIM), scheduled for launch in early 2010, is an optical interferometer that will perform narrow angle and global wide angle astrometry with unprecedented accuracy, providing differential position accuracies of 1 uas, and 4 uas global accuracies in position, proper motion and parallax. SIM astrometric measurements are sythesized from pathlength delay measurements provided by three Michelson-type, white light interferometers. Two of the interferometers are used for making precise measurements of variations in the spacecraft attitude, while the third interferometer performs the science measurement. The ultimate performance of SIM relies on a combination of precise fringe measurements of the interfered starlight with picometer class relative distance measurements made between a set of fiducials that define the interferometer baseline vectors. The focus of the present paper is on the development and analysis of algorithms for accurate white light estimation, and on the preliminary validation of these aglorithms on the MicroArcsecond Testbed.

In order to achieve micro-arcsecond astrometry, SIM must make measurements of various optical pathlengths at the picometer level. In this regime of precision, nearly every simplifying assumption in optics must be reexamined as a potential source of systematic error. SIM makes extensive use of physics-based models to predict the form and level of systematic errors affecting instrument performance. Since many of the modeling areas represent new frontiers in optical modeling, the validation of these physical models is a significant challenge that SIM must meet. In the case of the external metrology truss, the model must account for the imperfections in the corner cubes as well as the distance measuring interferometers ("beam launchers"). This model is being validated using the Kite testbed, a 2-D metrology truss with picometer-level accuracy in displacement measurements. We present the model, and the results of the model validation tests on the Kite testbed.

We describe a test bench designed to study the performances of interferometric imaging systems. The main goal is to study the densified pupil concept in the framework of the VLTI. This work is linked to the proposition of a second generation instrument called VIDA (VLTI Imaging with a Densified Array). This bench aims at comparing the imaging performances of the aperture synthesis, Fizeau and densified pupils beam combination schemes and at specifying the technical requirements like cophasing and tip-tilt correction. A Fizeau assembly, using a multi-apertures mask and associated with a wavefront sensor, has been designed. It allows to measure the differential piston between sub-apertures and to link them to the characteristics of the image recovered. A densified assembly is under study by using reflective surfaces or optical fibers to carry the beams and to densify the pupils before the combination.

We report on a novel approach for implementing a dual Bracewell nulling interferometric beam combiner using miniature conductive waveguides contained in a single monolithic structure. We present modeling results for these devices at mid-infrared wavelengths. Potential applications for these devices in the Terrestrial Planet Finder mission are discussed.

The interference pattern of many beams includes multiple fringe sets, each for a corresponding pair of beams. These fringes must have a limited spectral band, so that they can extend far from the central, white-light fringe. It is possible instead to use a chromatic corrector in order to match the fringe spacing at different wave lengths. As a result, they will overlap over a much larger area, and thus improve the selectivity of the beam pairs.

Adaptive optics systems on single big telescopes correct many modes, allowing imaging in the infra red. At the same time, visible photons can be used as well, especially when infra red light is also employed for wave front sensing. It is argued that pupil-plane interferometry is the most useful application for high-resolution imaging. This is because the isoplanatic patch area and the integration time are larger after correction, and they afford enhanced signal collection in the aperture plane. In contrast, speckle imaging methods only gain indirectly from this enhancement.