Published online 6 February 2009 | Nature | doi:10.1038/news.2009.81


The world's top ten telescopes revealed

The best observatories ranked by their scientific impact.

Whirlpool galaxySDSS image of Messier 51, the Whirlpool Galaxy.Sloan Digital Sky Survey

It doesn't take a big mirror to have a big impact. The Sloan Digital Sky Survey, a project conducted with a modest 2.5-metre-wide telescope in New Mexico, performed the most highly cited science in 2006, according to a new analysis of the top ten 'high impact' astronomical observatories1.

"It measures how hot the science of the telescope is," says Juan Madrid of McMaster University in Hamilton, Canada, of the top-ten table he has released for most years since 1998. "In a way it measures how good the time-allocation committee is and how good the telescope is. I will also say it measures how good the scientists are."

Also in the top five is another modest telescope — Swift, a satellite that looks for γ-ray bursts — followed by three technological giants of the astronomy world: the Hubble Space Telescope, the four 8-metre telescopes of the European Southern Observatory in Paranal, Chile, and the twin 10-metre Keck telescopes in Hawaii.

The table shows that a telescope's technological advantages can push it to the top of the list, but also that the culture and constraints of the operating institution are important. However, some astronomers caution that citations are just one of many metrics that should be used to assess an observatory's value.

"I think a number of them need to be taken together in determining which observatories are important," says Robert Williams, president-elect of the International Astronomical Union and former director of the Space Telescope Science Institute in Baltimore, Maryland, which operates the Hubble telescope.

Masters of the Universe

Madrid began compiling his citation rankings in 2001 while working at the Space Telescope Science Institute. It was a time, he says, when the political will for refurbishing Hubble was not very strong. His method: take the top 200 most-cited papers from a given year, throw out the theory papers, determine the observatories responsible for the remaining papers, and apportion the citations appropriately. His latest rankings, accepted for publication in the Bulletin of the American Astronomical Society, are for highly cited papers from 2006 because it takes a couple of years for citations to build up.

Click here for a larger image.

The rankings have served Hubble well: the telescope has been in the top five every year that Madrid has compiled his table. Other observatories spike in the rankings for brief periods before fading away. The Wilkinson Microwave Anisotropy Probe, for instance, took first place in 2003, following the data release of its seminal map of the cosmic microwave background. It was fourth in 2004. But as its data has been mined, it has since dropped off: it didn't make the top ten in 2006.

The Sloan Digital Sky Survey, by contrast, has been ascendant: it was the most highly cited observatory in both 2004 and 2006, the two most recent years in which Madrid has performed his analysis. The survey has helped unravel the structure of the Milky Way and has indirectly gauged dark energy's effect on the accelerating expansion of the Universe, says the survey's spokesman, David Weinberg of Ohio State University in Columbus. "It's not a huge, honking telescope, but it has very powerful instruments." He points to the richness of the dataset, along with its uniformity and quick public availability, as reasons for why so many scientists, including many outside the Sloan consortium, have done important work with its data.

Winners and losers

Some astronomers have wondered aloud about a few conspicuous absences from the top-ten list, including Gemini, an observatory with two 8-metre telescopes — one in Hawaii and one in Chile. Gemini science committee chair, Timothy Beers of Michigan State University in East Lansing, points to the fact that the Gemini telescopes were designed more for observations in the infrared part of the spectrum than for optical work, so they draw proposals from a smaller scientific community. "These were not instrumented as workhorse telescopes," he says. "It remains a problem." Without a broader array of instruments for a wider spectrum, Beers says scientists "vote with their feet and go to other observatories".

Another issue, he says, is control over the amount of telescope time, which is always limited. Privately owned and operated telescopes, such as Keck, can give large amounts of time to small groups of scientists aiming at the juiciest topics. Gemini, a publicly funded consortium representing seven nations, has seven separate time-allocation committees. They make it more difficult to organize long, unified campaigns — even though Beers recognizes that large datasets created from long campaigns tend to create the high-impact science.

But most of the astronomers agree that a citation metric is just one of many ways to measure an observatory. Other metrics include the overall production rate of scientific papers as well as the over-subscription rate — the ratio of proposed experiments to accepted ones. "My overall view is that citation mania has taken over to an absurd degree," says Weinberg. "But do they make a difference? I think so, especially when you are trying to raise funds."

And so Weinberg was sure to present the rankings based on the 2004 data — and the presence of the Sloan Digital Sky Survey atop it — when applying for a six-year extension to the project. The survey ended up winning US$9 million from the National Science Foundation (NSF) last summer. "How much impact did that have on reviewers and on the NSF programme officers? I don't know. But it certainly had some. And it should." 

  • References

    1. Madrid, J. P. & Macchetto, F. D. High-impact astronomical observatories. Preprint at (2009).


If you find something abusive or inappropriate or which does not otherwise comply with our Terms or Community Guidelines, please select the relevant 'Report this comment' link.

Comments on this thread are vetted after posting.

  • #60728

    How soon can we expect new images after the recent repairs ?

Commenting is now closed.