• Why can't Earth-size planetary transits be observed from the ground?
• Don't the stars vary more than the change caused by a transit?
• What is the typical distance to the stars where Kepler will find Earth-size planets?
• Why not use the Hubble Space Telescope?
• What are the new developments that now make this mission possible?
• How do CCDs (charge coupled devices)work?
• Couldn't the mission be done with a smaller photometer and cut the cost?
• Are there any other photometry missions?
• Can the data from this mission support other research programs?
• How does the Kepler Mission contribute to the Origins missions SIM and TPF?

There are two major reasons why these observations can't be done from the ground:

1. The motions in the atmosphere are constantly bending the rays of light from each star into different directions. This is why stars appear to twinkle. If you can see the change with your eye, you already know that the apparent brightness is changing by more than 50% (one stellar magnitude). With a lot of effort and for a very small region of the sky, astronomers have been able to measure changes as small as one part in 1,000 by comparing each star in a group to the whole group. This precision is still not good enough to find Earth-size planets, but should still be okay for detection of giant planets from the ground.
   
2. To detect a planetary transit as short as 2 hours out of a year requires measuring the brightness of the stars continuously. You can't blink! That means that you would need to set up dedicated telescopes in many places around the globe, so that there would be at least one of them on the night side of the Earth at all times. But, as the Earth orbits the Sun, the available night sky continuously changes. So there is no one part of the sky that can be continuously monitored throughout the year. In addition, the inevitable bad weather and the moon makes the prospects for ground based observing even that more inefficient. This would end up being a very expensive operation, even if the stars didn't twinkle. To detect Earth-size planets, space is the necessary place to be.
   

Yes, the stars do vary in brightness all the time. In fact it is almost impossible to make a perfectly constant source of light. Fortunately, the stars we are most interested in are stars like our Sun. These are the most commonly seen dwarf stars and vary less than the change in brightness caused by an Earth-size planetary transit on the same time scale as a transit.

Our Sun varies over many time scales: There are Maunder minimums, which do not occur for many centuries or longer and have caused "mini ice ages" even as recently as during the 17th century. There is an eleven year "solar cycle" of minimum and maximum activity. The largest short term variations are caused by "sun spots" that appear and fade, and rise and set as the Sun rotates with a period of four weeks.

Planetary transits have durations of a few hours to less than a day. The measured solar variability on this time scale is 1 part 100,000 (10 ppm) as compare to an Earth-size transit of 1 part in 12,000 (80 ppm). Even then most of the variability is in the UV, which is excluded from the measurements by the Kepler Mission .

What is the typical distance to the stars where Kepler will find Earth-size planets?

The distance is determined by the apparent magnitude range of 9th to 15th for the stars that Kepler will observe, the size of the star which determines the absolute magnitude of the various stellar types being observed and the size of the planet being detected. The detection threshold of Kepler is based on observing four transits of a 1.0 diameter Earth-size planet orbiting a 12th magnitude solar-like star. Since the Sun has an absolute magnitude of 4.82 taken at the standard distance of 10 pc, (1 pc=3.26 light years), it would have to be at a distance of 273 pc to have an apparent magnitude of 12. Almost all the stars for which 1.0 Earth-size planets in one year orbits exceed the detection threshold are between 100 and 300 pc.

If a planet has a shorter orbital period producing more transits in four years, then it can be detected at a similar distance but be smaller in diameter or if it is similar in size to the Earth it can be detected orbiting a fainter star at a somewhat greater distance.

For larger planets, the same detection threshold allows for fainter stars at greater distances. For planets 1.26 times the diameter of the Earth (about two Earth masses), the typical distance is about 300-400 pc with about half as many in the 200-300 pc range and about half as many again in the 400-500 pc range. For planets 1.71 times the diameter of the Earth (about five Earth masses), the typical distance is about 400-800 pc. Again this is based on planets in a one-year orbit.

Why not use the Hubble Space Telescope?

There are three basic reasons why the HST could not be used to look for planets in the way described here:

1. The field of view (FOV) of the HST is too small to observe a large number of bright stars. The FOV of the HST is about the size of a grain of salt held at arms length. There is almost never more than one bright star in the HST FOV at any one time. However, the FOV of the Kepler Mission photometer is about the size of both of your open hands held at arms length. Or another way of looking at it is, that it is about equal to the size of two "dips" of the Big Dipper.
   
2. The brightnesses have to be measured continuously, not just once in a while, since one does not know when to expect a transit to happen. The HST is for the use of the entire astronomical community to address thousands of questions and would not be dedicated to just one question requiring continuous use for four years.
   
3. The HST does not have a specially designed photometer on over 100,000 stars simultaneously with the precision required for the measurements needed to detect Earth-size transits.
   

The HST has been used by Ron Gilliland to look for transits of giant planets with periods of only a few days in the globular cluster 47 Tuc, a region of very high star density. No transits were detected.

Two recent research results have enabled the practicality of the Kepler Mission:

1. The demonstration that charge coupled devices(CCDs) have the needed photometric performance to make the measurements. All sources of noise (photon shot noise, stellar variability(see above), CCD noise and pointing jitter) when combined together must be less than one part in 50,000 (20 ppm); four times less than the effect of an Earth-size transit. The required CCD performance with all the known noise sources has been achieved in recent laboratory measurements along with the detection of Earth-size transit signals (Koch, et al. 2000). Thus, CCDs can be used to simultaneously measure tens of thousands of stars at one time.
   
2. Until recently, no one knew what the variations in stellar brightness were on the time scale of a fraction of a day. This information is now available for one star, our Sun. These data indicate that on the time scale of a transit, the variability is typically ten times less than the effect being measured. Fortunately, our Sun is one of the more common stellar types, and we expect other solar-like stars to behave in a similar fashion.
   

CCDs are at the heart of each digital camera and "HandyCam" TV camera . Special purpose CCDs are what we use for Kepler.

When light strikes a piece of silicon, it releases electrons that are free to move about the silicon material. These electrons form a charge or a current which is measured to determine the amount of light that has fallen on to the silicon.

In a CCD, the silicon region is divided electrically into small individual picture elements or pixels with about four hundred elements per cm in each direction, like a very finely divided sheet of graph paper. The free electrons are kept from moving around by permanent channel stops (the vertical lines in the figure) and externally applied voltages (the horizontal lines in the figure). Each pixel can then be thought of as an individual bucket or well that collects electrons.

As shown in the animation, first the CCD is exposed to light from a telescope or camera lens. Overtime this produces an image made up of electrons in the CCD.

To readout an image that has been captured with the CCD requires shifting the information out of the pixels. First, the columns of pixels are all shifted down one row. The bottom row of pixels is shifted into a readout register. Each pixel in the readout register is shifted out to an amplifier and the number of electrons in each pixel are recorded. This produces a series of 1's and 0's that represent the image. This is repeated over and over until all the pixels have been read. The stream of 1's and 0's is then digitally processed to reproduce the image that is later displayed.

In the Kepler Mission the 1's and 0's are recorded onboard the spacecraft and sent to the ground, where the data are processed to look for changes in the brightness of each star that may be caused by a planetary transit.

A representation of the scientific performance versus project cost is shown in the figure. A well conceived project is at A with maximum possible science per dollar available. Many times, those who fund a program perceive the project to be at B, where costs can be cut without much loss in science; well the science team tries to believe that they are at C, where more science can be achieved at little extra cost. Good clever scientists and engineers might be able to get to point D, but this is unusual. Project managers worth their weight in gold are those who can push toward E, keeping the performance,but saving on cost Any project headed from A to B A to C or A to F is doomed to be canceled or should be canceled.

For the Kepler Mission to work, a 100,000 main-sequence stars must be monitored to a differential photometric precision of 1:50,000 every 6.5 hours. Substantially fewer stars and the results may turn out to be ambiguous. The necessary precision requires recording ten billion photons from each star every 6.5 hours. Thus, a smaller photometer would mean either fewer stars at the required precision or poorer precision for most of the stars and thereby the inability to detect Earth-size planets. Also, the photometer would need to be much smaller before other costs, such as the launch vehicle, would begin to drop significantly.

Our project manager has worked hard at both increasing the performance by increasing the downlink data rate to permit monitoring 100,000 stars (originally we planned to monitored only 5000 stars) and in reducing the cost by changing the orbit to an Earth-trailing heliocentric orbit and thereby eliminating an expensive propulsion stage needed to get to an L2 halo orbit. This also allowed us to use a smaller and less costly launch vehicle. In essence, we have already pushed the cost-performance curve in both the D and E directions.

There are two other photometry mission, one that is currently operating MOST and the other under development COROT. However, they are considerably less capable than the Kepler Mission, since their primary science mission is to measure the properties of stars. COROT has 1/10 the collecting area for photons, 1/20th the field of view of the sky and stares at a given star field for 1/10 the amount of time that the Kepler Mission stares. MOST is an even smaller mission and less capable for planet detection. MOST was launched on June 30, 2003 and has produced spectacular photometric results on microvariability of stars - asteroseismology.

Obtaining astrophysical information about each star is a natural byproduct of detecting planetary transits. The following are some of the potential uses for these data:

Phenomena	Information Obtained
Stellar rotation rates	Variation in rates with stellar type
p-mode oscillations	Window on stellar interior: mass, age, metallicity of stars
Characteristics of solar-type stars	Determine what is a "normal" star
Frequency of Maunder minimum	Earth climate implications
Stellar activity	Star spot cycles, white light flaring, Paleoclimatology
Cataclysmic variables	Pre-outburst activity, mass transfer
Eclipsing binaries	Detection of high-mass ratio binaries
Active Galactic Nuclei variability	"Engine" size in BL Lac, quasars and blazars