header image
Link to Whats the Difference Link to New Worlds Atlas

SCIENCE & TECHNOLOGY

What's the difference between a candidate and a confirmed exoplanet?

Confirmed - validated by multiple observations

Confirmed exoplanets are planets that astronomers have proven to a high degree of confidence, using multiple observations and, sometimes, two or more different instruments. Sometimes, further data will cause a confirmed planet to be retracted, but this is a fairly uncommon occurrence.

Candidate - likely discovery, but still needs to be verified

PlanetQuest's list of "candidate exoplanets" comes from the Kepler mission. Kepler has made thousands of likely exoplanet discoveries, but in order for each planet to be considered "confirmed", its existence has to be verified using other instruments, a process that takes time.

As the Kepler team and other researchers across the world sift through the Kepler mission's immense collection of data, it's possible that some candidates will prove to be "false positives" - findings that appear to be exoplanets at first, but later turn out to be the result of other cosmic phenomena.

However, it's very likely that the vast majority of the Kepler exoplanet candidates, especially those in mult-planet systems, are real exoplanets.

How confident is "confident"?

Because exoplanets are so hard to spot, astronomers need to pay careful attention to any sources of errors that could creep up in their observations. They also need to calculate how likely it is that their observations might be inaccurate.

Usually, their level of confidence in a particular result is expressed in a probability figure. For example, a group of astronomers might find a possible exoplanet and calculate that the chance of that discovery being in error is 5%. In other words, they are 95% confident that their discovery is indeed accurate. This level of accuracy (known as "two-sigma") typically isn't enough for an exoplanet to be considered "confirmed".

Usually, a new discovery needs to have a minimum confidence level of 99.9999% to be considered a "confirmed" discovery. In other words, a one in a million chance that the result is in error. This level of confidence is referred to as "five-sigma".

Remember that confidence levels are based a all known sources of error. An astronomer may later find a flaw in their instrument or calculation method, and that would make their previous confidence level inaccurate.