What have we found?

Summary of Results to Date

To date, there have been 52 confirmed extrasolar planets discovered, with another 12-15 ìpossiblesî waiting on confirmation (Schneider, 2000) .  Rather than provide a table with the characteristics of all the extrasolar planets found to date, I am including a link to the Extrasolar Planets Encyclopaedia, which is updated frequently to keep up with new discoveries and announcements.

Figure 11 contains a partial listing of discovered extrasolar planets, sorted by each planetís distance from its star.  The estimated mass of each planet is also included.

Figure 11.  Extrasolar planets summary (partial) (Marcy, 2000)

As you can see, most of the planets found have been quite close to their star.  It should be noted, however, that this sample is biased by the Doppler technique, which favors planets with small orbital diameters (Croswell, 1997) . 

Another factor that needs to be taken into account is that these latest search programs have been operating for less than ten years.  Jupiter is 5.2 AU from the Sun and has an orbital period of about 12 years.  In order for the RV studies to find planets that far from their stars, astronomers will have to collect data on these systems for a number of years before significant changes in their Doppler signatures can be found. 

Challenges to ìConventional Wisdomî

While the discovery of extrasolar planets has captured the attention and wonder of the media and interested individuals, the astronomical community has been confronted with a number of challenges to our theories of planetary system dynamics and formation.

ìJupitersî Found Inside 3 AU (the ìSnow Lineî)

One of the first puzzles provided by the discovery of extrasolar planets is the fact that there have been a large number of Jupiter-massed planets found in orbits less than one-fifth the size of Mercuryís.  One of the most puzzling aspects of this discovery is that these planets exist at all (Jayawardhana, 2000) .  For years, the theories of planetary formation have asserted that it is impossible for gas giants such as Jupiter to form within 3 AU of a sun-like star.  This limit is sometimes referred to as the ìsnow lineî (Brown, 2000) . 

This presents us with a number of choices.  They either

1. are not gas giants,
2. were formed in a process different to our currently-held models, or
3. were formed outside the snow line and drifted inward.

Given the mass estimates provided by RV studies and the diameter measurements of transits, it is quite easy to estimate the density of these planets, and they seem to be quite similar to Jupiterówhich rules out option 1.

Accepting option 2 would entail invalidating the model of gravitational collapse, collisional accretion, and gravitational accretion that is widely held.  While this is still a possibility, astronomers are hesitant to abandon it, since there is a great deal of evidence to support it. 

This leave us with option 3óthese planets formed outside the snow line and somehow drifted inward.  While the exact process by which this would occur has not been agreed upon, these planets do exist.  These so-called ìhot Jupitersî share another propertyómost of them have near-circular orbits.  This could be explained by tidal effects as their orbits decreased to the distances at which we have observed them (Croswell, 1997) . 

Regardless of the explanation, the existence of these planets has opened the astronomersí eyes to the possibility of finding planets where they arenít expected.  As mentioned earlier, in order to detect a planet in a Jupiter-like orbit (~5 AU), it would take yearsí worth of data collection before seeing any results.  This is exactly what Marcyís team at San Francisco State believed when they started collecting data.  They had been collecting data for over a year when the planet orbiting 51 Pegasi was announced.  When they realized that this planet had a period of just 4.2 days, they started to go through their data and discovered that they had already captured the RV signature of extrasolar planets in their unprocessed data (Brown, 2000) !

Highly Eccentric Orbits, Why?

In our solar system, the average eccentricity of the nine major planets is 0.0812, and ranges from 0.007 (Venus) to 0.25 (Pluto).  The average eccentricity of the extrasolar planets discovered to date is 0.2434, ranging from perfectly circular orbits of e=0.0 (51 Pegasi) to e=0.71 (HD 222582).  Figure 12 compares the eccentricity values to each planetís minimum mass and orbital diameter as published in September of this year (Jayawardhana, 2000) .

Figure 12.  Orbital eccentricity comparisons

I have been unable to find any explanation for the difference between the average Solar System eccentricity and that of the aggregated extrasolar planet data.  Perhaps our solar system is unusual in that most of our planets enjoy more circular orbits, but I doubt it.  I find it likely that this could be nothing more than selection effectsóafter all, weíve yet to find any planets significantly smaller than Saturn.  It is possible that these planetary systems contain smaller planets that are below our detection thresholds.  These hypothetical planets could possibly ìeven outî the eccentricity values to be more in line with our own solar system.

This is definitely a topic that deserves to be re-visited as new techniques are developed and we gain the ability to detect smaller planets.

Can We Detect Solar System ìLook Alikes?î

Once it was realized that astronomers could detect planetary systems around other stars, it was only a matter of time until they asked, ìCould we detect our solar system from similar distances?î  The answer turns out to be somewhat of a surprise.  Take a look at Figure 13.  The upper left panel shows the RV curve of the star Upsilon Andromedae, caused by the three planets known to be orbiting it. As you can see, Ups. Andromedae has a very complex curve that varies by approximately 400 m/s over the 3.5-year period of its outer planet.  Detecting a radial velocity of ±200 m/s is well within the detection range of all extrasolar planet search programs.

Figure 13.  Planetary system comparison (Marcy et al., 1999)

Now look at the upper right panel.  This shows the RV curve of the Sun, from a simulated viewpoint in deep space.  As you can see, the RV curve is much more subtle.  The Sunís curve only varies by 24 m/s across the 12-year period of Jupiterís orbit.  The inner planets have virtually no impact on the RV curve, and would be impossible to detect by any of todayís search methods.

If we were to find an exact duplicate of our solar system, we would only be able to determine that it had one Jupiter-sized planet orbiting it once every 12 years, with a maximum RV of ±12 m/s.  Clearly, if we have any hopes of finding Earth-like planets, we must develop techniques for detecting low-mass planets.