TalonD said:
... maybe I don't understand what a parsec is. Can someone explain?
Heh... you might get several explanations. Here's wikipedia
http://en.wikipedia.org/wiki/Parsec
A parsec is 3.26 lightyears. Conceptually it goes back to 1838 the first determination of the distance to a star by a guy named Bessel, who used trig with the diameter of the Earth's orbit as a baseline for a long skinny triangle. Or maybe it was HALF the diameter---something like that.
You know the surveyors trig technique where you measure out a baseline and then you sight a distant tree or marker from each end of the baseline, and the angle of bearing is just slightly different. The difference angle is called parallax.
You experience a parallax angle when you stretch your arm straight out and look at your thumb from now your left eye and now your right eye. In that case the baseline distance is the distance between your two eyeballs.
So you asked what is a parsec? It is the distance from which the radius* of the Earth's orbit would subtend an angle equal to one arcsecond (that is 1/3600 of a degree)
*more precisely not the radius, because it's an oval, but half the major axis of the ellipse. It's also called an AU or astronomical unit.
Wikipedia has a diagram that makes it clear, so I won't belabor it. Parsec is short for PARallax angle equals one arcSECond.
If the distance between your eyes was one AU, and your arm was one parsec long, and you squinted at your thumb with one eye and then the other, your thumb would appear to jump by an angle of 1/3600 of a degree---that is by an angle of one arcsecond.
It is just an accident that a parsec is about 3.26 lightyears, and a meaningless accident that 13.7 times 3.26 equals 44.7 which is nearly equal to the 45 or 46 we give for the current distance to the matter that emitted the light we are now getting as CMB.
If the Earth orbit had a different radius, or if the babylonians had divided the circle up into 490 degrees instead of 360 degrees. Or if Bessel in 1838 had used an angle measure different from seconds of arc (1/3600 of a degree), then the parsec would have turned out to be an altogether different distance