[meteorite-list] Astronomers Calculate Comet's Orbit Using Amateur Images From The Web

From: Ron Baalke <baalke_at_meteoritecentral.com>
Date: Tue, 5 Apr 2011 12:53:18 -0700 (PDT)
Message-ID: <201104051953.p35JrIcJ026606_at_zagami.jpl.nasa.gov>

http://www.technologyreview.com/blog/arxiv/26603/

Astronomers Calculate Comet's Orbit Using Amateur Images From The Web
The Physics arXiv Blog
April 4, 2011

Amateur astrophotographs posted online represent a massive untapped
resource. Now astronomers have worked out how to mine it

For a short time back in October 2007, Comet 17P/Holmes became the
largest object in the Solar System as the thin ball of dust and gas that
surrounds it briefly became larger than the Sun. At the same time,
Holmes brightened by a factor of half a million, making it visible to
the naked eye. (All this activity seems to have been caused by a sudden
outburst of gas from the comet's nucleus.)

This sudden brightening triggered a huge wave of interest from
astrophotographers all over the world, many of whom posted their images
on the web. To find out how many, Dustin Lang from Princeton University
in New Jersey and David Hogg at the Max-Planck-Institut fur Astronomie
in Heidelberg, Germany, searched the web. They found 2476 different
shots of Holmes.

That's a significant astronomical database that represents a huge amount
of work. But is it any use?

Today, Lang and Hogg use these images to work out an accurate orbit of
Comet 17P/Holmes, a significant achievement given that the data is taken
from an ordinary web search and its provenance is entirely unknown.

The method is relatively straightforward. These guys fed each image into
the astrometry.net <http://astrometry.net/> website which analyses the
pattern of stars in the shot and then tells you which part of the sky it
shows.

They then created a giant montage of these images, carefully
superimposing the stars. Since the pictures were all taken at different
times, the superimposed images show the comet moving across the sky (see
image above).

They then compared the comet's trajectory with the orbit calculated by
the Jet Propulsion Laboratory, finding a remarkably close match.

That's an impressive piece of crowdsourcing. All the more so because it
differs in one very important way from the various other crowdsourcing
projects on the go, such as GalaxyZoo. None of the astrophotographers
who took these shots knew they were taking part and most still don't.

More impressive still is Lang and Hogg's assertion that this is only the
beginning for this kind of data mining. "We have only scratched this
surface," they say.

The big question that concerns them is how far it is possible to take
this data mining technique.They say there is a similar body of images
for Comet Hyakatuke and have begun an analysis of these. And they point
out that there are more than 3500 images of the Orion Nebula on Flickr
alone.

They conclude by asking whether it might be possible to use the
collected images of the world's astrophotographers to carry out a survey
of the entire night sky. "We expect the answer is yes," they say.

We'll look forward to seeing it.

Ref: arxiv.org/abs/1103.6038 <http://arxiv.org/abs/1103.6038>: Searching
For Comets On The World Wide Web: The Orbit Of 17P/Holmes From The
Behavior Of Photographers
Received on Tue 05 Apr 2011 03:53:18 PM PDT


Help support this free mailing list:



StumbleUpon
del.icio.us
reddit
Yahoo MyWeb