Kepler mission – Anticipating First Light – the answering post
As I write these words Kepler has still its sleep mask on and has been, in the last days, dreaming with the future view, with the wonders waiting in the distance, in some sort of REM as calibration tests take place.
A good timing to publish the answers provided by Jon Jenkins, Kepler Co-Investigator to the questions sent by BtC readers about the mission.
Jon was thankful for the opportunity to answer these since, according to his words, they’re all great questions indicating that BtC readers are “very much on the ball”.
Here they are:
The Press kit mentioned a possible prolongation of the mission beyond 3.5 years. I guess (technically = excluding budget reasons) this will depend on the amount of coolant or the life time of the coolant radiator? Any other points that could influence a possible extension?
Kepler launched with enough hydrazine fuel for our thrusters (used mostly to manage the spacecraft attitude by removing excess momentum from the reaction wheels) for a 6 year mission. We don’t carry coolant, as our CCDs are passively cooled by heat pipes that conduct heat generated by the detector electronics in the focal plane to the radiator. The hydrazine fuel is our only consumable on orbit. Electronics and spacecraft components do age in space. Kepler was designed with a high reliability for the original primary mission duration of 4 years. If it is operating well at that point, we would expect good performance for the extended mission as well.
Great to see the launch and checkout seem to be proceeding well, many thanks for finding some time for us during this intensely busy period.
I have a question about the mission duration. The primary mission’s 3.5 years and I’ve seen mention of the possibility of a secondary mission of the same length. With the experience of the MERs, Cassini’s latest extended mission and even the Ulysses spacecraft’s improbable survival months after frozen hydrazine lines were expected to end it, I wonder what the hard limits are on Kepler’s long-term survival? How long will consumables for position-holding last, for instance, and is there any prospect of getting useful science data beyond that point? Thanks!
We have enough fuel to last for 6 years. After the hydrazine is exhausted we don’t expect the reaction wheels to last very long as they will spin up to rates that exceed their design capacity. Excess momentum needs to be removed from the reaction wheels periodically through the firing of the thrusters, so our long term mission requires hydrazine fuel.
First of all thanks for taking some of your precious time to answer some of our questions, recently I was at NASA´s site watching the pages of the SIM lite mission and at some place (on SIM´s book) I read: “The candidate Earth-like planets revealed by Kepler will raise the question: Are they really Earths? Instead, they could be grazing-incidence eclipsing binaries with a brighter third star that dilutes the photometric dimming. SIM Lite offers a valuable way to check some of these potential false-positives.”
Is there a way to eliminate this false positives with the current instruments,so that we can be truly sure that we have a planet?
Are you sure you weren’t on one of our review panels, Jose, this is a rather important question! The answer is that in almost all cases, a background eclipsing binary whose light is diluted by a brighter target star to the point that the eclipses mimic planetary transits reveals itself by exhibiting motion of its photocenter (or centroid) that is correlated with the change in brightness due to the transit-like events. Based on the statistics we’ve estimated for our Field of View, we expect a small handful of cases for which the offset between the dim eclipsing binary and the target star is too small for centroid discrimination to work well. We have a large team of Follow-up Observation Program (FOP) team members to catch these few cases by conducting a combination of ground-based and space-based observations. Our plan for Kepler is to reduce the fraction of false positives to a very small percentage of the planetary announcements through our FOP. One very effective way to do this is to observe the transits in a much redder filter than Kepler’s because the background system will appear brighter compared to the target star and hence, the “transit” depth will change significantly if the source is a background eclipsing binary. SIM-Lite is one resource capable of helping out with eliminating false positives from Kepler’s Results Catalog. Unfortunately, we can’t count on having SIM Lite help out with this task as its launch has yet to be scheduled.
Will any of the raw data be made available for enthusiastic amateurs to play with? Several aspects of the problem of sorting out small planets from large ones from binaries from random cosmic rays etc. seem tailor-made for various machine-learning algorithms, and it would be interesting to see how well those performed in practice.
Along those things, what sort of algorithms are you planning to use for this?
We have a staggered data release plan that calls for incremental release of both raw and calibrated science data products to the public starting at the one year mark. The Kepler team is very keen to establish a valuable and highly credible planetary results catalog and needs to have time to vet the results before we make any announcements. However, we will release data for all planetary discoveries when they are announced. The entire data set, both raw and calibrated, will be released to the public one year after the end of flight operations. I’m sure that enthusiastic amateurs and professionals will comb through the data to make new discoveries, planetary and otherwise, once the data are available.
Our baseline algorithm for detecting transiting planets is a wavelet-based, adaptive matched filter. This algorithm performs a joint noise characterization/signal detection task as stars like the Sun exhibit changing noise characteristics (such as Sun spots) throughout the stellar activity cycle. Once transit-like features are identified for a particular target star, we subject it to a comprehensive set of statistical tests to ensure that it isn’t simply a noise fluctuation or a background eclipsing binary masquerading as a planet. We also have a large team of astronomers dedicated to conducting follow-up observations to exclude any remaining false positives and to learn as much as we can about the host star. We are interested in learning about planetary systems and planetary formation from the results.
While a “Kepler at Home” Project would be very interesting, the technology was sufficiently immature at the time we were developing the Kepler mission system and also raised security issues (with premature data release). We simply didn’t have the resources to effectively deal with all the issues raised by this idea, although it was considered. We do have a Participating Scientist, Dr. Laurance Doyle, who is plannning to use a “Kepler at Home” approach to detecting transiting planets orbiting short period eclipsing binaries in the Kepler Field of View. Stay tuned to http://www.kepler.nasa.gov/ to see how you might participate in this exciting activity!
I’m curious to know if Kepler might be repointed at some future time. Imagine if Kepler is very successful its first 3 years and finds 100 Earth-size planets. Then during year 4 Kepler finds 2 such planets, leading us to believe that year 5 would probably be even less fruitful. During years 5 and 6 (or perhaps earlier) might Kepler be pointed at some other star cluster?
It is unlikely that we will point Kepler at multiple Fields of View (FOV). Our analysis suggests that Kepler will detect more Earth-like planets by remaining pointed at the nominal FOV for the entire mission, rather than by pointing at different FOVs. Even if we don’t find one Earth-Sun analog system, we will have stronger constraints on the frequency of such systems with one FOV as opposed to multiple FOVs. Our baseline plan is to keep Kepler pointed at the same FOV for the entire mission, however long it lasts.
This is it for now, stay tuned for more news as the First Light Day approaches!