Several direct detection techniques have been conceived and tested in exploratory observing runs. This paragraph lists some of the programs envisaged for the RTG. Each observational program is complemented by the full development and assessment of the technique as described in "Observational Techniques, Telescopes, Instrumentation" below. These observational programs will particularly rely on and benefit from cross-disciplinary proceedings. Interaction with the theory group will encompass target selection, spectrum simulations for the correlated detection, and the spectrum synthesis for the analysis of integrated or resolved spectra. Observational results will be fed back into the (still photon starved) theoretical modeling. Interaction with the stellar programs will be required for target selection, i.e. recognition and interpretation of stellar activity, metallicity and rotation.
Planet detection and characterization via RV modulated planet spectra
The large-amplitude velocity modulation of close-in giant extrasolar planets allows to separate planetary signatures from the spatially superimposed much stronger but quasi-static stellar flux. A measured velocity amplitude yields the planet/star mass ratio (without the sin i ambiguity), and spectral information on the planet. The detection principle is similar to the stellar RV technique, yet with the handicap of a much smaller intensity amplitude and the advantage of a much larger velocity modulation of the expected signature. Simulations based on our advanced planet atmosphere models predict observable signatures in the detection range of existing IR spectrographs. IR direct and indirect RV searches are part of the science justification for the VLT high-res IR spectrograph CRIRES, which is expected to go on-line in 2005. We expect excellent access to observing time for this portion of our program. We have conducted exploratory observations in the past, have sufficient high-quality data available for analysis and technique development in the early phase of the RTG program. The plan for the IR direct searches foresees: further development and assessment of the technique; analysis of the existing VLT/ISAAC and IRTF data on HD75289, Boo and And; assessment of extension to the optical regime and observations of bright stars with our own 1 m class telescopes; new observations using CRIRES/VLT, PHOENIX/Gemini S and NIRSPEC/Keck.
Direct detection through spectro-astrometry and -photometry
The spectral separation technique 'spectro-astrometry' uses the small spatial planet-star offset (typically in the mas range) in addition to the spectral information. This spatial offset is wavelength-dependent and proportional to the intensity contrast in the spectrum. By cross-correlation with a template spectrum it can be measured differentially with very high resolution and accuracy. In addition to the direct detection and spectral information, this technique yields orbital parameters. The spectroscopy and therefore the interaction with theory, is similar to the direct RV detection program described above. The spectrophotometric method uses the orbital modulation of the planetary contribution to a system's total brightness. It is conceptually similar to the spectral separation techniques. The plan for the IR spectro-astrometry foresees: further development and assessment of the technique; analysis of the existing VLT/NACO data; new observations using CRIRES/AO/VLT; spectro-astrometric evaluation of high-quality optical spectra of candidate planetary systems. The spectrophotometric method can be applied to the same data-sets, but we will apply for observing time on the HST/NICMOS (if still available) and target the brighter planet host stars with the instruments at our own 1 m telescopes.
Photometric surveys for extrasolar planets
While the RV method is currently the most successful way to discover extrasolar planets, the photometric surveys have the potential to significantly improve our knowledge of planet formation and evolution. HD209458b, the first transiting planet had been detected previously by RV measurements. From the other six transiting planets five originate from the OGLE survey (Konacki et al. 2004). Nevertheless, surveys for transiting planets are only at the beginning. With typically 1000 observations per field, OGLE is mainly sensitive to planets with an orbital period below 3 days. The lack of such short-periodic planets in RV surveys,however, suggests that OGLE only finds the tail of the planetary distribution. With the MONET/Wise observatory and OmegaCam at the VST on we have telescope access for much longer observing campaigns. While the former project has started with the construction of a large field CCD camera for Wise observatory and a definition for an equivalent camera for the MONET telescopes, the latter project will obtain 40 guaranteed nights per year as a combined project of the OmegaCam instrument consortium. Novel analysis tools will allow the detection of transits also in the vicinity of active stars, currently impossible for OGLE, and to efficiently reduce systematic effects allowing the detection of very shallow transits down to Neptune-size objects around solar-like stars. With own photometric surveys we intend to expand our investigation of transiting planets (Dreizler et al. 2003, 2005) to the full procedure from the survey to the final confirmation. A significantly improved number of transiting planets will allow to confront formation and evolution scenarios for planets with a large data set. High precision photometric surveys allow an alternative approach to planet detection via microlensing. In this method, the light of a background star is focused by the gravitational field of a foreground star. If the line of sight is nearly identical this effect leads to a detectable increase in brightness. A very dense follow up of such magnification events allows to detect planets around the foreground star due small to distortions of the magnification curves. This method is of special interest since - in contrast to other methods - a large fraction of the Milky Way can be explored. With MONET and its fast reaction ability we will participate in coordinated follow-up observations of several magnification events per year in cooperation with MOA and other microlensing consortia.