There are two reasons I became an astronomer : 1) because Star Trek was cool; 2) pretty pictures of nebulae. While Star Trek may indeed be inspirational it doesn't really have any direct lessons for astronomy, but the importance of pretty pictures cannot be overstated. Or more accurately, how we go about turning raw data from telescopes into something we can understand really matters. Especially so for radio astronomy. With telescopes receiving light well beyond the visual wavelengths, we have to apply some quite creative methods to transform the data into something visibly meaningful.
The advantage of this is we have a great deal of freedom in how we choose to process the data. While radio astronomers almost never "listen" to the data (that's usually a bad idea), in some ways it does have more in common with sonar than optical astronomy. We generally receive not just how bright the gas is, but how fast it's moving – the Doppler shift we're more familiar with from sound waves. So we have 3D data sets but where one dimension is velocity, not space, and we have the particular concern that most of the data is generally noise rather than signal.
In this section you can find both descriptions of the methods I use for visualising data and specific projects I've worked on. Data visualisation, in my opinion, is as artistically valid as any other medium : how we choose to display the data can be legitimately as much matter of self-expression as how and what we choose to paint. But as science in general is creative only under the constraint of evidence, so is data visualisation not a matter of pure artistic preference. For these projects I've chosen ones where the main goal is to display the data in some sense as directly as possible. The Art section also includes some other scientific illustrations, but those are ones in which the driving force is much more dominated by subjective choice.
As much of the accompanying code as possible is given in the Code section. For more on the typical data structure itself, see this page.
Techniques : Here I describe the basic methods of data visualisation I use. This mainly revolves around Blender but also includes some raw data processing and cleaning as well.
Projects : My main data visualisation projects, from making the Virgo Cluster look like it was catalogued in the 18th Century to turning it into a large glass brick; from artsy-fartsy pretentious nonsense about dualism and HI to some rather technical stuff about mapping the Milky Way.