Sword of Orion ©Greg Parker. Photographed from the New Forest Observatory, Brockenhurst, England.
Photography has an inseparable connection to science and technology. Camera technology is often used in scientific investigations to document research findings and produce images that illustrate natural phenomena. From time to time – and perhaps more regularly than we might realise – such research yields strange and beautiful pictures. Emeritus Professor of Photonics Greg Parker’s deep-sky images are an inspiring example of this. Since retiring from the University of Southampton in 2010, Parker set up two observatory domes to photograph galaxies and nebulae from his New Forest Observatory in Brockenhurst, England.
Parker’s career has been in research and development since graduating from Sussex University in 1978 with a 1st Class Honours Degree in Physics, Mathematics and Astronomy. A common factor throughout his research has always been light, including lasers and optical instruments and optical devices. For over twenty-three years he researched into optical components called Photonic Crystals at the University of Southampton, together with Optical Biomimetics, looking at how Nature had come up with elegant solutions to optical problems.
“The tracking has to exactly follow the motion of the stars in the region being imaged or we will get unwanted star trailing – perfect tracking will show all stars, right across the frame, as perfectly round balls of light.”
In September 2004 Greg first started taking deep-sky images from his observatory using a new device called a Hyperstar. Since then he has also constructed a second observatory housing his “mini-WASP” array – a multiple refractor, multiple imager, parallel imaging array so that 4-hours worth of data can be downloaded in just one-hour, very useful given the British climate.
Professor Parker explains the technology and theory of his process
Comet Lulin ©Greg Parker. Photographed from the New Forest Observatory, Brockenhurst, England.
In order to take deep-sky images we need to accurately track an astronomical CCD camera at the focus of a telescope for the duration of the sub-exposure time. The tracking has to exactly follow the motion of the stars in the region being imaged or we will get unwanted star trailing – perfect tracking will show all stars, right across the frame, as perfectly round balls of light. Tracking is often carried out using a second telescope, usually a small aperture refractor and a small guide camera. The guide camera sends its image to a computer which controls the motion of the (equatorial) mount which carries the main imaging scope and camera. A star is chosen on the guide camera image and the guiding software makes sure that the chosen star remains in the same position to within a fraction of a pixel on the guide camera by “nudging” the mount to the appropriate degree. In this way, the main imaging camera can take the very long sub-exposures which are required to acquire the faintest objects, without any obtrusive star-trailing.
During an imaging session as many sub-exposures as possible are taken to improve the quality of the final image. In practice the individual sub-exposures are added together, using software packages created specifically for the purpose, so that the final image has an improved signal to noise ratio proportional to the square root of the number of sub-exposures. In plain English this means that nine sub-exposures can be combined using the software to give a resulting image that looks three times better than a single sub-exposure. In addition the software can also remove annoying “hot pixels” produced by the imaging camera during long sub-exposures, even though the image camera itself is cooled to reduce the number of hot pixels present.
Horsehead Nebulae ©Greg Parker. Photographed from the New Forest Observatory, Brockenhurst, England.
Very faint objects require long sub-exposures to capture the faint detail, and lots of sub-exposures are required to give a “clean” noise-free image. So it is clear that the most demanding (of time) deep-sky imaging is the capture of faint deep-sky objects, usually nebulae. Stars being intensely bright point sources of light require much shorter sub-exposure times to create a nice looking image, so star field imaging is much less time-consuming than imaging nebulae. As a rough approximation, a high quality star field image will require sub-exposure times of around five minutes and a total exposure time of three to four hours. A medium brightness nebula may require ten to twenty minute sub-exposures with a total imaging time of at least eight hours. These figures hold reasonably well for both f#2 and f#4.5 imagers for reasons that are too technical to go into in any detail.
The second observatory dome at the New Forest Observatory (the North dome) houses the mini-WASP array. This is an array of four refractors each working at f#4.5 which all image the same object at the same time. Why do this? The main reason is that working at f#4.5 means that we are operating 5 times slower than the Hyperstar system working at f#2, so by using 4 refractors in parallel we are nearly as “fast” as the Hyperstar system. So why bother using refractors at all? There are several reasons, including much better image contrast using refractors rather than reflectors, and the ability to image very bright stars without bad “ghost” flaring – something that cannot be done using the Hyperstar.
How to photograph a galaxy
Galaxies are “island universes” – great masses of stars all grouped together in what we call a galaxy, and what used to be called a nebula before it was realised what galaxies actually were. Our nearest galaxy is M31, the Great Andromeda Galaxy which is a mere 2.5-million light years away. Because M31 is so close to us (galaxy-wise) it actually appears quite large in a telescope, much larger than most people think. The Andromeda galaxy measures over 4-degrees across at its widest part, and the full Moon in the sky is only half a degree across, so the Andromeda galaxy has the width of 8 full Moons! M31 contains around a trillion (a million million) stars, whereas our own galaxy, the Milky Way contains only around 200-400 billion (thousand million) stars. Galaxies generally are very much smaller than M31 so they require long focal length telescopes to image them to give sufficient magnification. However, there are a sufficient number of “large” galaxies (M33, M101, M81/M82) that the Hyperstar and mini-WASP arrays can capture them at a reasonable image scale to give a pleasing result.
Emission nebula IC1396 in Cepheus ©Greg Parker. Photographed from the New Forest Observatory, Brockenhurst, England.
How to photograph nebulae
Most nebulae, such as the Great Nebula in Orion – M42 – are regions of ionised gas that emit light in the red part of the spectrum. If you ionise the gas Neon in a Neon sign by passing an electric current through it you get the usual red colour. In the case of emission nebulae (like M42) the gas is Hydrogen and rather than an electric current it is a nearby very bright star that ionises the gas. Ionised Hydrogen, like Neon, emits light in the red part of the spectrum. There are however other types of nebula. In “dark” nebulae the nebula appears like a smoke cloud cutting out the light from the stars behind it as can be seen in the clouds surrounding the Iris nebula. A dark nebula does not have a nearby bright star to ionise it, so it acts just like a light absorber and absorbs the light from more distant stars.
M45 ©Greg Parker. Photographed from the New Forest Observatory, Brockenhurst, England.