Wobbly Planet Throws Astronomers For A Loop

Exoplanet Kepler 413-b orbits a close pair of orange and red dwarf stars once every 66 days. Not only does the planet’s axis wobble like crazy, its orbit about the stars wobbles too. Credit: NASA, ESA, and A. Feild (STScI)

Can you imagine what Earth would be like if its axis wobbled about during your lifetime? The axis’ steady tilt in one direction causes the amount of sunlight each hemisphere receives to vary in a predicable way, giving us the familiar four seasons.

Enter Kepler-413b, a giant gas planet located 2,300 light years away in the constellation Cygnus the Swan. Its axis is tiled 30 degrees from vertical and wobbles like a top over a remarkably short period of just 11 years. Seasons would change erratically and rapidly, so much you’d hardly know what to wear month to month.

Animation of the Earth’s 26,000-year wobble called precession. The axis remains titled at 23.5 degrees but slowly wobbles over time describing a circle in the sky. Whatever star lies at the “end” of the axis becomes the pole star. Animation repeats every 10 seconds. Credit: Wikipedia

Earth’s axis wobbles too but so slowly we don’t notice it at all during our lifetime. It’s called precession and amounts to 23.5 degrees over 26,000 years.

Keep in mind that the axis doesn’t flip-flop around – it maintains a tilt of 23.5 degrees just like Kepler 423-b keeps its axis tipped at 30 degrees. Instead, the axis describes a repeating circle in the sky as the planet wobbles.

Another view of Earth’s precession or wobble

Kepler 423-b’s orbit around the dwarf star pair appears to wobble, too, because the plane of its orbit is tilted 2.5 degrees with respect to the plane of the star pair’s orbit. As seen from Earth, the wobbling orbit moves up and down continuously.

The Kepler telescope, when it was still active, discovered planets by watching a star’s light repeatedly dim when a planet transited across its face one orbit after another. But something unexpected happened with Kepler 413b. The transits disappeared for a while and then returned, implying a weirdly oscillating orbit.

“Looking at the Kepler data over the course of 1,500 days, we saw three transits in the first 180 days — one transit every 66 days — then we had 800 days with no transits at all. After that, we saw five more transits in a row,” said Veselin Kostov, the principal investigator on the observation.

Kepler 413-b is a giant gas planet 65 times the mass of Earth, what astronomers call a super Neptune. It orbits so close to its host stars that it’s too hot for liquid water and most likely uninhabitable. TheCredit: NASA

No one’s sure yet what causing the orbit to continuously change its tilt.

It’s possible Kepler 413-b feels the tug of yet-to-be-discovered planets in nearby orbits. A neighboring third star may be gravitationally bound to the stellar pair and exert an influence too.

Kepler 413-b, even if it has seasons, is probably too darn hot for life anyway. It orbits so close its parent suns liquid water’s not an option. So I guess no one will ever have to switch from shorts on Sunday to snow pants on Monday there.


19 Responses

  1. RC

    If the planet’s axis rotates once every 11 years, that wouldn’t that basically equate to 12 winter solstices in 11 years? Or once every 11 months? That would make seasons 2.75 months long instead of 3. Right about now I’d be happy if winter was a week shorter!

    If this were the case on Earth, I wonder if people would consider a “year” to be the time it takes to make a complete revolution around the sun, or if it would be the time it takes for the axis to get oriented the same way relative to the sun (aka winter solstice to winter solstice). If a year was still 1 revolution around the sun, that would make for some interesting holiday memories. The great 4th of July snowstorm! Or maybe the Christmas it was 103 degrees and sunny! Six years later 4th of July could be 98and sunny, while Christmas would have 3 ft of snow.

    1. Giorgio Rizzarelli

      Hi RC, well put considerations about the exoplanet. About year definition, at least for Earth, there’s indeed the difference between sidereal year (revolution period respect fixed stars) and solar year (interval between two vernal equinoxes). The difference is indeed because of the precession of the equinoxes (for Earth once in 26,000 years) so just a 1/26000 of a year. That is the sidereal year is about 20 minutes longer than solar year. Both actually also have a difference of about 6 hours respect the calendar year, defined in such way to have an integer number of days. Bob correct me if I’m wrong 🙂 Source: http://en.wikipedia.org/wiki/Sidereal_year

      1. RC

        Interesting point Giorgio. This got me thinking – If a sidereal and solar year are off by 20 minutes, they’ll be off by a day every 72 years. So across the span of an average lifetime, the seasons come one day earlier.

        1. Sean

          ah, the effects of precession. multiply that 20 minutes by about 2000 years and u get the month or so that the astrological zodiac is off from the reality of when the sun is ACTUALLY in front of a particular zodiacal constellation. now, of course, even back then the zodiac was sort of idealized, with the signs being of equal length regardless of the actual size of the constellation, but at least back then if you were a Virgo (for example) it was highly probable or perhaps even certain that the sun actually was in front of Virgo when u were born (of course this depends on how u define the boundaries of the constellations). Nowadays, my bday being 9/21, i am one of the few people whose zodiacal sign matches the constellation that the sun was in front of when i was born. Most people are born when the sun is in the constellation corresponding to the zodiac sign PREVIOUS to the one shown in newspapers and whatnot for their birthday. Not that i think any of this matters when it comes to personality traits or whatever; i just find it interesting. if anything, this could count as yet another reason to find astrology silly. sorry to be a hater.

    2. astrobob

      Cool points about how residents of such a planet would choose to define their year.
      As for season length, Kepler 413-b’s axis precesses (not rotates) one full revolution or wobble every 11 years. Let’s assume no or minimal precession for the moment. Given that, it’s year of 66 days works about to a little more than 16 days per season as the axis points toward, sideways and away from its suns. Add in precession and every year (66 days) that passes would see the axis shift by 1.6 degrees, quickly shifting the seasons around the calendar year from our perspective.

      1. RC

        I see, I was mixing and matching Earth days/years with Kepler 413-b days/years. If Earth had the 11 year wobble, our seasons would be 2.75 months. On Kepler 413-b, the seasons are much shorter because the year is also much shorter (66 earth days). I suppose if we wanted to complicate things more, we could discuss how many Kepler 413-b days are in a Kepler 413-b year, since it’s unlikely that an Earth day and a Kepler 413-b day are the same length…

        1. astrobob

          Hard to say how many Kepler 413-b days there are in that planet’s year since we don’t know the rotation period. Since it’s very close to its host suns, the days might be very long like Mercury’s due to tidal locking.

  2. caralex

    Bob, I’m not sure if you’re saying that the axial tilt of 30 degrees increases and decreases, or if it’s the orbit itself that wobbles, or if its period of precession is extrememly fast, or indeed a combination of all. Can you clarify?

    1. astrobob

      Hi Carol,
      Sorry for the confusion Carol. I tried to address it in the blog but I see there’s some ambiguity in an earlier paragraph. I’ve since cleared that up. The axial tilt remains the same throughout (just like Earth’s precession) but the swinging back and forth over such a short period of time causes rapid seasonal changes.

  3. Sean

    for some reason, to me, the sidereal year seems to be a more “pure” form of year. it seems like when i celebrate a year around the sun i want to celebrate a revolution around the sun, not “almost” a revolution. in recent years i’ve realized how our calendar system only approximates the latter in any case, since obviously 365.4 or thereabouts is a tough number to divide anything cleanly into. Basically, on average people should be celebrating their birthday on about one of every four years on a calendar day adjacent to their normal birthday, if by celebrating a year more of life they mean to celebrate one more revolution around the sun, marking the spot around the sun the earth was at the moment of their birth, by either definition of year.

    1. Giorgio Rizzarelli

      and an amateur astronomer sometimes considers the sidereal day more pure than the solar day. A star will always be in same position in sky at a given sidereal time, all days in the year – it’s the Sun which is not…

      1. Sean

        Hm, Giorgio. Not sure i had ever thought of it exactly like that. I had a breakthrough of realization about a year and a half ago (started really getting into astronomy in 2011) when i realized that the background stars are the same every day of the year: in essence they rise and set in the same place, take the same track thru the sky, etc. The only thing that changes are things like planets and the moon, and when the other stars are obscured by the sun being in the sky (what we call time of day). For some reason, before that, i had a feeling that, like the sun, the track of the other stars must change with the seasons. In any case, while i now realize the constancy of those stars, i have not yet switched to picking one time to consider “midnight”, “noon”, etc. and thereby think of the stars in relation to time in a fixed manner. I guess if i had to do that i would select the sky at solar midnight at the turn of the new year (december31-january1) since by coincidence that’s when Sirius has its midnight culmination for us in the N hemisphere. So with that position being midnight, these days midnight happens around perhaps 7PM local time for me. Interesting way to think, having a sidereal day.

        1. Giorgio Rizzarelli

          It’s done exactly as you say, only the conventional date of the year when sidereal and solar midnight coincide is at autumn equinox.

          Here’s a sidereal time clock (automatically gets your longitude):

          And here’s a nice simulator: http://astro.unl.edu/classaction/animations/coordsmotion/siderealSolarTime.html.

          Yes, of course one thing is knowing about sidereal time in theory, another is having the practice to apply it easily (I wish I did!). One trick is, sidereal time = R.A. at meridian. So learning one helps learing the other.

          1. Sean

            Thx for the links, Giorgio. It makes sense that there would be a convention for sidereal time to make sure people who use it are all referring to the same thing. As far as right ascension, i pretty much don’t think of things in those terms as i prefer altitude and azimuth. perhaps if i owned a telescope (well, a goto scope anyway) i would be more inclined to really learn RA.

Comments are closed.