What would Galileo think of new revolution?

Afew years ago we borrowed a replica of one of Galileo's telescopes for demonstration at a recent Open House at our observatory.

Galileo was an astronomer, physicist, engineer and mathematician in 16th Century Italy. He heard about a new invention — the telescope — and from a description of how it worked, designed and built his own.

With it he saw mountains and craters on the Moon, spots on the Sun, the rings of Saturn and the moons of Jupiter.

A look through that replica telescope really conveys how patient and determined an observer he was. It only shows a tiny piece of sky and you have to put your eye at exactly the right place to see anything at all. You can definitely only observe one thing at a time.

As any user of a modern telescope knows, since the days of Galileo we have made huge advances in telescope design and engineering.

Following the technical advances made in World War II, radio telescopes have become important astronomical tools.

However, despite these advances, we continued the tradition of observing one thing at a time, with ever greater sensitivity and precision.

You would measure the spectrum of a star, or spot variable stars in a distant galaxy. Radio telescopes acted like photographic exposure meters, measuring the radio brightness of a single patch of sky.

We imaged the radio sky by scanning the radio telescope, building the image bit by bit. When we developed the technique of radio imaging, we still imaged one little bit of sky at a time. Doing any sort of large-scale survey with such instruments was very time-consuming and regarded as the last application of an instrument no longer “cutting edge”.

This was unfortunate in that even though we learned a lot, with many new things being found by accident, other things were missed, and just as importantly, we lacked the context. A particular star or galaxy becomes unusual only when compared with lots of others. The change came when we became able to develop telescopes that could see big chunks of sky in a single operation and to record everything in that field of view.

Not only were we getting the “Big Picture” at last, but also in doing so we were finding lots of new things. Today, large-scale surveys are considered front-line research.

However, this means we are collecting huge amounts of data. The value of this data is so large that we can no longer rely on individual scientists or groups of scientists to look after it.

We now store and archive that data at special facilities, such as the Canadian Astronomy Data Centre, where everything is backed up, looked after and made accessible to scientists when they need it.

OK, you have access to all this data. You might be able to find what you want, but you are more likely to be faced with a needle-in-a-haystack scenario.

Issues like this have driven the development of a new process called “data mining”. Thanks to the power of modern computers, we have been able to develop techniques like “machine learning”, “neural networks” and “artificial intelligence”.

We can now have machines search for what we are after, often with only the vaguest instructions. In one case a catalogue of data on over 20,000 galaxies was given to a computer, which was directed to see if there were different classes of galaxy, and to classify them.

No further information was given to the machine, which minimized any effect of any prejudices on the part of the scientists involved.

The process worked, but took two weeks.

Today the revolution has taken us well into the world of surveys, big data, and of robotic assistants helping us find meaning in it. What would Galileo have thought about this?

Venus shines brilliantly, low in the Southwest after sunset. Mars, redder and much fainter, lies to Venus' left. Jupiter rises in the early hours. The Moon will reach First Quarter on the 5th.

Ken Tapping is an astronomer with the National Research Council's Dominion Radio Astrophysical Observatory, Penticton.