An Update on the TCE Review: Nearly Half Way There!
I wanted to update y’all on the status of our TCE review that we launched at the beginning of the month.
This side project is to do our own Planet Hunters review of the ~18,000 potential transit events, dubbed Threshold Crossing Events or TCEs, identified by the Kepler team’s automated computer algorithms during a search of the first ~3 years of Kepler data. The majority are false detections, but a few are real transits due to orbiting exoplanets. A subset of the Kepler team examine the TCE list and whittle it down to make the Kepler planet candidate list. These newly released TCEs have yet to fully searched by the Kepler team, meaning there are likely discoveries waiting to be found. We have launched a Planet Hunters review of the Kepler TCEs to identify new planet candidates. For each TCE, you’ll be presented with a light curve that has been zoomed-in and folded so that the repeat transits all line up on top of each other. With the folded light curves we can see smaller planets, the rocky ones that are so hard for most of us to see in the regular light curves we show on the main Planet Hunters website. We think Planet Hunters has an advantage in the ability to review the entire TCE list (with your help) and identify not just the rocky planet transits but also the Jupiter-sized and in between. You can learn more about the TCE list and the TCE Review by reading the launch blog post.
As of today, we’re now at the nearly half way mark towards the 184,060 classifications needed, with 7,669 of the 18,406 TCEs complete with 10 looks. For those who’ve already contributed to the TCE review, thank you for your hard work. If you’d like to join in the TCE review and help get us to the finish line with 10 classifications for each of the 18,406 TCEs, please visit http://tcereview.planethunters.org
Cheers,
~Meg
Live Chat Today
We’re having a live chat today with science team members Tabetha Boyajian (Yale University), Chris Lintott (University of Oxford/Adler Planetarium), and Meg Schwamb (Yale University). starting at 4pm BST/ 11 am EDT /8 am PDT/3pm GMT.
There’s lots to talk about today including the failure of Kepler’s reaction wheel 4 and updates on Planet Hunters science. Join us here to watch the live video feed. You’ll also be able to find us on the Zooniverse Google+ Page. If you can’t watch live, the video is recorded and will be available to view here later.
If you have questions for the Planet Hunters team you can ask them, either by leaving a comment here on the blog or by tweeting us @planethunters.
Thanks for watching. You can find the TCE review at http://tcereview.planethunters.org/ and as always you can classify light curves at http://www.planethunters.org
Save the Date: The Next Planet Hunters Live Chat
Mark your calendars! We’re planning our next live chat for May 20th, 2013 at 4pm BST/ 11 am EDT /8 am PDT/3pm GMT to talk about Planet Hunters science and news. You can find the video feed here or you’ll also be able to find us on the Zooniverse Google+ Page.
We’ll be talking to Tabetha Boyajian (Yale University), Chris Lintott (University of Oxford/Adler Planetarium), and Meg Schwamb (Yale University).
If you have questions for the Planet Hunters team you can ask them, either by leaving a comment here on the blog or by tweeting us @planethunters.
A Planet Hunters Review of the Kepler Transit Candidate List
The Kepler team uses automated routines, specifically the Transiting Planet Search (TPS) algorithm, to search for transit signals in the Kepler light curves. TPS triggers on many repeated transit-like features in the light curves dubbed Threshold Crossing Events or TCEs. TPS generates many many TCEs, much more than the number of real extrasolar planets. The majority are false detections, but a few are real transits due to orbiting exoplanets. A subset of the Kepler team examine the TCE list and whittles it down to make the KOI (Kepler Object of Interest) list. A handful of Kepler scientists review each TCE and data validation report, results from a series of checks and test to help rule out astrophysical false positives that might produce a transit-like signal such as blended background eclipsing binary. It takes many many months for this process. The current Kepler planet candidate list released in January was using Q1-8, but there are many more Quarters of Kepler data available.
The Kepler team has made all of their data products publicly available in the extended mission. In December, the Kepler team released the list of 18,406 TCEs found during a search of Quarters(Q) 1-12 data and the resulting reports produced by their data validation pipeline. These Q1-12 TCEs have yet to fully searched by the Kepler team, meaning there are likely discoveries waiting to be found.
For the past few months I’ve been working with Chris and Arfon to set up a Planet Hunters review of the TCE list. Today the review site is live, and we need your help to review these potential transit candidates and identify the ones that are likely due to real planets. We’re using a version of the round 2 review interface, we used before to vet planet candidates for my short period planets paper. For each TCE, you’ll be presented with a light curve (from the data validation report) that has been zoomed-in and folded on the period determined by TPS so that the repeat transits all line up on top of each other.
A good TCE candidate – The black data points all the actual measured Kepler data folded on the period identified by TPS. The blue diamonds represent an average of the folded data. The red line plotted is a model of the transit that TPS identified.
We are asking you to confirm that there is a visible transit in the light curve identified by TPS (“Is there a transit?”) and determine whether the red line matches the light curve (“Does the red line fit the data?”).
With the folded light curves we can see smaller planets, the rocky ones that are so hard for most of us to see in the regular light curves we show on the Planet Hunters website. There are other teams who are using the TCE list in their research and as targets for follow-up observations, but where I think we have an advantage is that we have the ability to review the entire TCE list, not just the rocky planet transits but also the Jupiter-sized and in between.
We’re not in a race with the Kepler team who am I sure are also vetting the current released list, but I believe what is unique to this project and Planet Hunters is the ability to review uniformly all the ~18,000 potential transit signals identified by TPS. The current versions of the Kepler KOI list currently has not gone back and reanalyzed all the previous planet candidates detected in previous TPS runs with the longer observational baseline. So we’ll have the first independent vetting of the Kepler Quarters 1-12 TCE catalog providing a uniform selected sample of planet candidates.
Each TCE will require 10 independent review before being retired. Once we’ve gotten through identifying what looks to be real transit candidates from the non-detections, I’ll apply some additional cuts based on the output from the Kepler validation pipeline (like how the consistent is the depth of the odd and even transits to rule out eclipsing binaries and pixel offsets in and out of transit that might indicate a blended background eclipsing binary is producing the signal) to come up with our very own Planet Hunters planet candidate list from the TCEs.
I think this project will result in a very interesting paper looking at the frequencies of super-Earth to Neptunes to Jupiter-sized planets in the Kepler field, and also serve as an efficiency estimate for the Kepler KOI vetting process. If it goes well, we may consider making this a more permanent fixture on Planet Hunters for future releases of the Kepler TCE list. My goal is to have the first results from the TCE review to show in a poster at Protostars and Planets VI conference in Heidelberg, Germany in July.
If you are interested in participating and helping out with this project, you can go to http://tcereview.planethunters.org/ where you can join in and characterize the TCEs. Please do read through the tutorial on the front page. It will guide you on what you should be doing, as well as show you some examples of non-detections and good TCE transit detections.
Thanks in advance,
~Meg
Quarter 14 Data
Thanks to everyone’s efforts we’ve nearly completed all of Quarter 7 data. Last week, Quarter 14 was uploaded and went live on the site. Quarter 14 is the most recent Kepler observations to be released by NASA during Kepler’s extended mission. It covers observations spanning June 28th – October 3rd 2012 (the last full Quarter during the primary mission).Quarter 14 was processed and made available on the MAST (Mikulski Archive for Space Telescopes) public archive in early February. With Q7 mostly done, we decided to jump ahead and show the data as close to off the telescope as we could get!
As with any new Quarter, there are now new opportunities to find previously unknown planets. Quarter 14 was released to both the Kepler team and to the public at the same time in February. Quarter 14 has yet to fully analyzed by the Kepler team. The Kepler team has released their list of potential transit detections from Quarters 1-12 in December and list of planet candidates from Quarters 1-8 in January. So there may very likely be never before seen transits found in Q14.
You might have also noticed a change in the naming of the light curves, and that the APH ids now significantly differ from the SPH ID. We basically ran out of namespace to do the SPH-APH mapping well, so we’ve switched over to this new naming system. You can tell what SPH star and Kepler ID correspond to the light curve you are looking at by going to the Talk page for the light curve and clicking on Examine Star (that will take you to the source page). The APH ids still can tell you what quarter of Kepler data you’re looking at. The quarter identifier comes right after “APH” . Before Q14, it was a number (1-7). Now for Q14, we are using “E” to denote 14. The next number following the quarter identifier, still tells you which chunk of Q14 you’re looking at.
Making Way for Q13 Data
The Kepler field will be high in the sky starting in the next month or so and continuing over the Summer months. Thanks to all of your hard work and classifications, the science team has been writing observing proposals to ask for telescope time on the the Keck telescopes in Hawaii to follow up on our best planet candidates. We’ll learn in a few months whether we have been granted the nights. So stay tuned!
In the meantime, the team is continuing to search for new planet candidates with your classifications and Talk comments. You have been analyzing light curves from Quarter 7 released by NASA during Kepler’s primary mission. To begin searching the first data release of Kepler’s extended mission, Quarter 13, we need to finish Quarter 7. We need your help to make room for the new light curves.
As with any new data, there are now new chances to find even more planets. Let’s make the final push so that by April we could be looking at Q13 data. Please get clicking today at http://planethunters.org
Happy Hunting,
~Meg
A slightly unusual look at PH1
One of the many varied things I get to do with my time is act as an advisor to the Oxford Sparks project. As part of our mission to inform the world about the wonderful science this place is involved in, we get to produce animations like this :
The eagle-eyed will have spotted that there’s a world in there familiar to Planet Hunters volunteers, as our slightly intrepid hero is whizzed past PH1. There’s another interesting link, I think, between the two projects; both planet hunters and the animation take us to the cutting (some would say bleeding edge) of science.
Pretty much everything in the animation is open to question (and you can read more background over on the main Sparks site) – we have only weak evidence that there was a fifth giant planet in the Solar System, and many question the portrayal of the sudden bombardment of the Moon as shown here – it’s difficult to tell whether the evidence we have points to a true sudden bombardment or the mere end of a longer period of increased impact probability. On the broader questions too, there is disagreement – what sort of world would really be suitable for life? Do Earth-like planets such as the one we end up with really exist out there? (Probably – but we’re not sure yet).
All of this is ok. My aim – our aim – was to present science with the ink still wet rather than wait for the final draft. After all, it’s most inspiring when we can still make discoveries, and hopefully the video will make people think – and maybe even make a few discoveries of their own on Planet Hunters.
Chris
Planet Hunters: An Online Community
The team at Possible including Elena Moffet and Tyler Brain made a short documentary exploring online communities from the perspective of the individuals who compose them. This included talking to people involved with Wikipedia and open source software developers. Our own Katy Maloney reflects on being a citizen scientist and part of the Planet Hunters community. You can watch the final product titled COLLECTIVE: An exploration of online community below:
New Features on the Source Pages and Other Updates
With the Kepler extended mission in full swing, we have added some new data to the source pages to help your investigations. If you don’t know what a source page is, it’s where we show all the light curve data for a given Kepler target that we have uploaded on the Planet Hunters website as you would see it in the main classification interface. We show roughly 30 day light curve segments in the main Planet Hunters classification interface, so on the source pages you can peruse the other light curve sections. In addition the source page is the place where you can get the Kepler ID of your favorite star and look up other information about the Kepler source.
You can get to the source pages by clicking on the “Examine Star” link (see below) on any Talk light curve page.
The source pages now include some new links to reflect some of the new data on Kepler stars now available. At the top of each source page you’ll see several links: Add to Favorites – Download Data – Kepler Archive – Kepler TPS – UKIRT DB
Download Data: The Download Data link will download to your harddrive a csv file of all the light curve data we have uploaded for that star.
Kepler Archive: The Kepler Archive link will take you to a Kepler Catalog Search which outputs the Kepler id of the star, it’s position on the sky (right ascension and declination), magnitude, colors, etc.
***New Features***
The Kepler TPS: In the Kepler Extended Mission, the Kepler team is now releasing their list of Transit Crossing Events (TCEs) from their main Transiting Planet Search (TPS) code. A TCE from TPS is not a planet candidate. It’s a possible series of linked transits, identified by TPS that has hit the criteria for being considered a detection. There’s a lot of work needed to go from TCE to planet candidate. TPS detects may thousands upon thousands of TCEs and most of these are false positives. So lots of other checks have to be made to validate the possible transit detection to a planet candidate. For the Quarter 1-12 run of TPS there are over 18,000 TCEs. The TCE list has not been vetted by the Kepler team. Probably only a few thousand are actually real. Provided by the Kepler team is the data validation report for each TCE. This is the output from their data validation pipeline of some tests to assess whether the TCE may be a real planet candidate. (You can learn more about the TCEs and other data products provided in the Kepler Extended Mission here.) The Kepler TCE link will search the TCE list and report back if there was a hit on the TCE list with some info about the properties of the detection, which you can then further investigate on the NExSci website. If there are columns names and nothing else when you click the Kepler TPS link that means the source is not on the TCE list.
UKIRT DB Link: The Kepler pixels are rather large. Each pixel in a Kepler image is 4 arcseconds. The typical photometric aperture radius for a Kepler target star is 6 arcseconds, and all the photons collected by the CCD within that radius are assumed to come from the target star and summed to make the Kepler light curve you see. With such a wide aperture, stellar contamination and photometric blends are a concern. Adding linearly, the contribution of extra light decreases from other stars within the Kepler aperture causes the observed transit depths to be shallower than they really are. Accurately estimating the size of the transiting planet requires knowledge of the additional stars contributing to the Kepler light curve to correct for this effect. Also knowing there are additional stars in the Kepler photometric aperture is important because it can help identify that a Kepler source;s light curve is being contaminating by a nearby star that is an eclipsing binary. By clicking on the UKIRT DB link, you get a search reporting back UKIRT J band (near infrared) images of the Kepler star from the WFCAM Science Archive. The typical pixel scale is 0.8-0.9 arcseconds per pixel, which allows us to zoom in within a few arcseconds of the Kepler target. (Thanks to Mike Read from the WFCAM Science Archive for helping set this up and to Phil Lucas for generously letting us link to the data in this way).
—-
Other updates to the site include updating the lists of known eclipsing binaries, false positives, and Kepler planet candidates. The Kepler team last week released their list of planet candidates and false positives from analysis and vetting using data from Q1-Q8. The Talk labels have been updated to reflect this. The eclipsing binary list comes from the July 2012, since that’s the last static list to be released.
Planet Hunter Kian awarded the Chambliss Prize
Congratulations to Planet Hunter Kian Jek, who just received the Chambliss Amateur Achievement Award! This is the premiere award for amateur astronomers presented by the American Astronomical Society – and we think it is highly deserved. Many of you know Kian, who works tirelessly on the site and to a large part is responsible for the dramatic success of Planet Hunters. He not only plays a leading role in hunting down planets, but has been responsible as well for much of the work on variable stars that the project has produced. This award is a suitable acknowledgement of his contributions, and we hope it also serves as recognition for all of the Planet Hunters by the American Astronomical Society. When solicited for a comment on the 42 new candidates in the latest Planet Hunters paper, co-author Kian (his third paper!) wrote:
“As someone who grew up with the Apollo moon landings, whose childhood imagination was fired by Kubrick’s 2001 and the original Star Trek, I never had any doubt that planets around other stars existed and that one day we would discover them. But I never dreamed that we would find them in my lifetime, let alone being involved in their discovery. Although there are over 700 discovered since 1996, each new planet opens another door to a strange alien world, some of them we could not even imagine could exist. Planethunters is an exciting project that allows citizens and scientists participate in pushing the final frontier ever slightly further.”
