A Hundred Times Sharper Than Hubble: Stellar imaging with intensity interferometry

Abstract: Imaging of stellar surfaces in visible wavelengths is one of the current frontiers in astronomy. With a few exceptions, stars can not be seen in visible light as anything but point objects with current technology. Being able to properly image stars would open up the door to a vast field of new discoveries, permitting direct studies of phenomena such as rotationally deformed stars, circumstellar disks and clouds, stellar winds and wind collision zones, mass accretion structures, pulsating stars etc. Ground based phase interferometers have been able to produce images of a few large stars in infrared light by connecting several telescopes over distances of ∼100 m, but are limited by the need to keep the optical path constant down to a fraction of the wavelength of the light over very large distances. Intensity interferometry works around these problems by measuring correlations in light intensity fluctuations, essentially removing the need for high-quality optics and making the system virtually immune to atmospheric disturbances, permitting the connection of a great number telescopes to form one giant kilometer-scale hypertelescope. This comes at the cost of higher light intensity requirements and the loss of the phase of the Fourier transform of the stellar surface brightness. Invented in the 1940’s by Robert Hanbury Brown, intensity interferometry has not been used in astronomy since the 1970’s (though the physical principle behind it has become very important in particle physics), due to the requirement for very large flux collectors and fast photo-detectors. However, recent technological advances and new mathematical algorithms for image reconstruction have sparked a renewed interest in this technique. A “digital revival” of intensity interferometry would enable visible-light imaging of stellar objects at resolutions that are orders of magnitude better than what is possible today or in the foreseeable future. An interesting possibility is to use air Cherenkov telescopes – which happen to share many requirements with intensity interferometry. In this thesis, the theory and history behind intensity interferometry are laid out and a number of astrophysically interesting targets of study are identified. A new method for simulating intensity interferometry measurements, valid for modern photon-counting detectors, is derived and implemented as a simulation software package. It is also shown how this method can be extended to higher-order correlations. The simulation software is applied to a number of astronomical objects with extra attention given to the use of the upcoming European mega-project CTA (Cherenkov Telescope Array) as an intensity interferometer. Finally, the results from laboratory experiments at Lund Observatory are presented.