How we designed the Kepler-186f artists concept image

Our Kepler-186f paper came out last week and we got a lot of press attention (lots more than I expected!). Tim Pyle and Robert Hurt at Caltch/JPL did an amazing job of of creating a beautiful image which was shown around the world on TV and in print.

Artists concept of the Kepler-186 system by Tim Pyle and Robert Hurt. Clicking on this will take you to the full image which is fantastic!

I received a couple of questions from people about why the image looks the way it does and whether there is some scientific basis to the depiction. First, I want to say that the magic was all performed by the artists at Caltech/JPL. However, Elisa Quintana and I discussed with them at length about how we wanted the planet to look. There was a lot of very fun back-and-forth where we tweaked things to get them how we wanted. The artists turned our blurry vision into a high resolution reality (if I sound gushing here it’s because I am truly in awe of their work).

The information we have at hand are the size of the planet, the size and temperature of the star and the distance the planet is from the star. This actually provides us with quite a bit of information to work with.

The temperature of the star is roughly 3800 Kelvin. There has been some interesting work on convolving the spectrum of various stellar types with the response function of the human eye. Without an atmosphere, the Sun would appear almost white to our eyes (not yellow as you might imagine). We started with using the correct color for this star. However, rendering the star the true color made it look very much how we might imagine the Sun to be (i.e. a yellow/orange color). Part of the story we wanted to tell was that this star is not Sun-like, so we did make the star slightly more orange than it would really appear (the color show is correct for a star a few hundred Kelvin cooler).

The planet is bathed in orange light from the star, the planet receive the correct level of illumination from the star given its orbital distance. We wanted to show the planet illuminated and also show the star. This meant we had to alter the angle of incoming light from the star in order to see some of the planet’s surface. In essence, the star is shown at the wrong phase angle relative to the planet. If we didn’t do this we would only see the night side of the planet which would be fairly boring and not allow us to show a depiction of surface conditions on the planet.

The planet is shown with yellowy continents and grey/blue oceans. There are ice caps and clouds that appear orangey. Oceans on the Earth appear a deep rich blue color, this is because of the blue light from the Sun Rayleigh scatters off the ocean. This star emits very little blue light which we represented by making the sea a dull grey/blue color. Ice and clouds Mie scatter light which is fairly uniform across all wavelengths hence clouds and ice appear the same color as the star. Then we come to the color of the continents – we had fun with this one. When we were designing the image Elisa Quintana found an article by Nancy Kiang titled The Color of Plants on Other Worlds. Nancy is a scientist based at NASA Ames (she moved to Ames from GISS the week after we talked to her, small world heh!) who works with the Virtual Planetary Laboratory. We called her up and chatted about what colors plants might be on planets orbiting cool stars. While this is a very complex issue involving evolution of photosynthesis, she recommended a dark yellow/green color as a potential color for alien planet life on this world.

We wanted to highlight the differences between this world and Earth. The low illumination level is purely because this planet receives less light from its star than we do from the Sun. Because of this lower illumination, we speculated that this planet may be a little colder (note that a planet’s temperature is determined by its atmosphere, something we know nothing about). So we made the planet have prominent ice caps and also had the plant life (i.e. the yellow color on the surface) cluster at the equator, similarly to how we see lots of green colors at mid-latitudes on Earth in temperate regions – temperate regions in the artists concept are clustered around the equator.

Finally, we wanted to depict the other planets. Three of the other planets are clearly seen nearby the star. They are near to the star because they much closer to the star than Kepler-186f. One of my favorite parts of the image is that we have one of the interior planets transiting the star. We wanted to make the planet just large enough to be detectable. This meant increasing its size a little – in reality it would be able 1/30th the size of the star. The purpose of having the planet there was to show how we detect planets with Kepler.

I hope this post has been informative and demonstrates that a lot of thought goes into making these artists concepts. They are necessarily speculative, for instance we don’t know whether the planet has water, continents, ice, clouds etc. but we do know that Earth has these things. Hopefully this image provides a nice tool to explain what might be the same and what might be different between this planet and Earth.

Giordano Bruno said some amazing things

I finally got around to watching the first episode new Cosmos. Coincidently, I’d been chatting to Jill Tarter from the SETI Institute the previous day so Carl Sagan was on my mind. We were chatting about pointing the Allen Telescope Array at some interesting Kepler planets. More about this another time, but just to say this: it almost certainly won’t see anything interesting but… but… if it does this will be the single most important discovery in history.

On the whole I enjoyed Cosmos. I was reminded my how prescient Giordano Bruno had been. I’m using this post to just paste a quote of Bruno’s from 1584. It sends shivers down my spine.

There are countless suns and countless earths all rotating round their suns in exactly the same way as the seven planets of our system. We see only the suns because they are the largest bodies and are luminous, but their planets remain invisible to us because they are smaller and non-luminous. The countless worlds in the universe are no worse and no less inhabited than our earth. For it is utterly unreasonable to suppose that those teeming worlds which are as magnificent as our own, perhaps more so, and which enjoy the fructifying rays of a sun just as we do, should be uninhabited and should not bear similar or even more perfect inhabitants than our earth. The unnumbered worlds in the universe are all similar in form and rank and subject to the same forces and the same laws. Impart to us the knowledge of the universality of terrestrial laws throughout all worlds and of the similarity of all substances in the cosmos! Destroy the theories that the earth is the centre of the universe! Crush the supernatural powers said to animate the world, along with the so-called crystalline spheres! Open the door through which we can look out into the limitless, unified firmament composed of similar elements and show us that the other worlds float in an ethereal ocean like our own! Make it plain to us that the motions of all the worlds proceed from inner forces and teach us in the light of such attitudes to go forward with surer tread in the investigation and discovery of nature! Take comfort, the time will come when all men will see as I do.

Quote taken from Wikiquote who reference The Discovery of Nature (1965), by Albert W. Bettex

K2 to observe M35 continuously for 75 days

The target list for K2 Campaign 0 (C0) was released earlier today. C0 is an engineering test primarily to learn about pointing over en entire campaign, but we hope to do some amazing science. The test will begin next week and is designed to last around 75 days. We will target over 7500 stars and galaxies, with targets including cool stars, bright stars, nearby galaxies, a pulsar, red giants, cataclysmic variables, and many more.

However, the most exciting thing to me is that we are going to put a giant pixel mask over the open cluster M35. There are so many stars of interest in this cluster we decided to get the whole thing. What will be learn? Firstly I hope we will find some planets orbiting the (relatively) young stars in this cluster. Do planets form in clusters, what sizes are they? Do they look different from planets in the Kepler field? Additionally, we can learn about how fast every star rotates helping us to understand how to measure a star’s age from its spin rate.

The green regions are the pixels that are going to downlinked to Earth for one of K2′s 76 functional CCD channels. As you can see, there is a giant continuous region – this is the open cluster M35.

Link to C0 information: http://keplerscience.arc.nasa.gov/K2/Fields.shtml

 

 

The first K2 observation of a transiting exoplanet

My life has been consumed recently with planning for the K2 mission (a mission where Kepler observes into the ecliptic). The Senior Review proposal to NASA has just been submitted and we have received over 100 proposals from community scientists to observe targets during our first science campaign.

WASP-28b

K2 managed to catch one transit of WASP-28b during a short test observation in Jan 2014. The data were obtained at 1 min intervals.

We have some data on the ground and the good news is that we can detect planets! We have observed our first exoplanet transit, WASP-28b – A previously known hot Jupiter type planet. I’ve posted an image of the transit, it looks beautiful. The noise may seem a little high but this is because the data were taken at 1-min cadence rather than our more usual 30-min cadence. There is also only one transit so we don’t benefit from overlaying multiple transits.

I’ve also being doing some work on estimating what sizes of planet we are going to be sensitive to. It looks we are going to be able to find Earth-sized planets  orbiting relatively bright G and K type stars and well as around fainter M-dwarfs. This is great because these are the stars we are going to be able to get follow-up ground based observations of. I can’t wait to get some more data on the ground.

A second planet with a comet-like tail

Last year Saul Rappaport and team reported the discover of what appeared to be a slowly disintegrating planet. The evidence pointing to this was a planet transit-like event of a body very close to its star. The size of the body appeared to change dramatically over time and the shape of the transit was not symmetric. They worked out that if they simulated a planet that was slowly losing material and had a comet-like tail they could reproduce the observations. This was a very neat discovery and there has been plenty of follow-up observations and theoretical work performed to understand that system. However, it was not clear whether this was an extremely rare event or a long lived and relatively common occurrence among very hot, small plant

Well, now we have another one. KOI-2700b is the second planet with a tail found in Kepler data. We actually identified this one a couple of years ago but could not explain what was going on (this was before the first disintegrating planet). The transit was very asymmetric in shape so we dubbed it planet Nike (as in a Nike tick).

The planet goes around its parent star every 22 hour and the tail seems to take persist for about about 5 hours. This means that the tail extends behind the planet for about a quarter of the orbital circumference.

One final cool thing we found is that the depth (and indeed shape) of the transit it changing over times. We see the transit become about half as deep over the four years we observed this planet. Seems like we caught this thing shortly before it looses its tail and becomes too small to detect.

koi2700nike

The brightness of the star KOI-2700 decreases when a planet with a large tail passes in front of it. The unusual transit shape caused us to dub this Planet Nike.

 

 

The Two-Wheel Kepler Mission will be awesome

I was in a seminar last week where exoplanet science was described as being in the post-Kepler era. This phrase has much merit given how dramatically our understanding of exoplanet populations has changed since Kepler launched. For example in 2007 the state of the art simulations predicted 10-100 Earth mass planets do not form. Kepler data has shown this predicted to be spectacularly wrong. Turns out most planets all into this range.

However, in the talk post-Kepler era was used to describe the era after Kepler stops taking data. Which is now because Kepler is dead, right? Erm, actually no. Kepler is alive and, if not well, doing pretty good. What is true is that Kepler cannot point accurately at its original field, but there are parts of the sky where it can really kick ass – in the ecliptic. Most of the current information is subject to change but our best guess is that we will be able to observe some 10′s of thousands of targets continuously for about 2.5 months. Now here is the real exciting part. Its very possible that we will be able to obtain better than 100 parts per million photometry. That means we could easily detect rocky Earth-sized bodies around Sun-like stars, albeit pretty hot rocks. This capability is unique, nothing else can achieve precision as good as that. This is why I’m super excited by two wheel Kepler. We just need a good name for this refactored mission.

Robust exoplanet masses from transit timing variations

I’m going to try to post on a semi-regular basis about whatever astronomy thing has interested me on a given day. it’s probably be a bit random and not well explained.

Today I was chatting with some by Kepler buddies about a paper by Ofir where they measure the mass of two planets orbiting KOI-1574 using transit timing variations. The part I find interesting is that Ofir finds one of the planets has a density of 0.1 g/cc – much lower than any planet in our own Solar System. We were concerned with the uniqueness of the solution given there are only 4 transits of the planets of each planet in the Kepler data.

Dan Foreman-Mackey has been developing an n-body/transit model code that can be used to used to model systems like KOI-1574. I hope/plan to run an MCMC simulation that should reveal whether the ultra-low density found by Ofir is significant and robust. Not that I’m trying to pick on Ofir, his analysis seems perfectly reasonable to me. It does however provide a convenient example to test the code upon.

 

We’re back!

We’re back, and by that I mean the US government is back. For the last 16 days I’ve been locked out of my office at NASA Ames. As a contractor I’ve been allowed to work, indeed I’ve had quite a productive few weeks.

The first week of the shutdown the SETI Institute in Mountain View kindly hosted me and provided free coffee! With all the Kepler science office in the same room we were able to chat about a few projects we were all working on and it looks like a few papers that have been a long time coming are nearly finished (Kepler planet catalog papers).

Last week I took the opportunity to visit New York University as a guest of David Hogg. We have a few projects together that have been left on the backburner. Hopefully we/I will get these going this time. One of the more straight forward ideas we have is to measure limb darkening using transit of multi-planet systems.

Anyway, I’m going to try and use this blog as a bit of a research/work journal. It may not work, let’s see.

Announcing Kepler-69c – a super-Earth-sized planet in or near the habitable zone of a Sun-like star

I’m about to do the press conference to announce this system but I wanted to make the paper available. The paper is available from here.

The press conference should be on NASA TV now.

Edit:

A recording of the press conference can be found on the NASA UStream page. I’m not sure how long it will last there though.

I did a live interview with the BBC Radio 4 program The World Tonight. That interview can be found via this link - Radio 4 interview.