New research suggests the size of exoplanets may be routinely overestimated, and clouds, scientists say, are to blame.

Astronomers and their observatories have become rather adept at spotting exoplanets, but gathering details -- the planet's size, mass, composition -- remains a challenge.

To better understand the accuracy of their observations, researchers recently took a closer look at two seemingly similar low-mass exoplanets. The two planets orbit their host star in 5 and 12 days and are roughly the same size, at 4 and 5 times the diameter of Earth, respectively. The inner planet is less than six times the mass of Earth and the outer planet is 28 times as massive as Earth, so their similar size suggests the inner planet has a very low density.

Computer simulations of the two planets' evolution suggest the inner, less massive planet, with its close proximity to its star, should have had its atmosphere burned away within the first 100 million years of its existence. But the star and planet are billions of years old.

The only explanation for the anomaly, astronomers argue in a newly published paper, is that the planet is considerably smaller than originally estimated.

Lammer says a smaller planet with an extended atmosphere that has high-altitude features could confuse observations.

"The radius is based on what we see when the planet makes its transit," astronomer Helmut Lammer, a scientist at the Austrian Academy of Sciences' Space Research Institute, said in a news release. "This is probably distorted by clouds and haze high in the atmosphere, in a region where atmospheric pressure is otherwise very low."

Scientists have identified more than 3,000 exoplanets, and that number will continue to go up as observatories become even more efficient at finding candidates.

In light of the latest research, detailed in the Monthly Notices of the Royal Astronomical Society, Lammer suggests astronomers should review the results of previous exoplanet surveys, like those undertaken by NASA's Kepler observatory.

"Our results show that CHEOPS scientists need to be cautious about their first measurements," study co-author Luca Fossati said of ESA's forthcoming CHaracterising ExOPlanets Satellite mission. "Since Kepler has also discovered several similar low-density and low-mass planets, it is very likely that the size measured for many of them also differ from the true value, so there could be a bias in the results."