Viewing the Result

00:00 Instead of showing you the command line and the small graph that pops up, I’ve copied the results into my slide here and expanded it a bit. This is our 650 some stars showing their colors and sizes plotting temperature against luminosity.

00:15 You can see from here why I chose to modify the edge color. All those bigger orange points in the top right would just be a big old blob without the black edge color.

00:25 The chart actually has quite a variety of star colors, temperatures, and sizes. Note that the data for the chart is based on the temperature and luminosity, while the B-V color axis and the magnitude axis actually weren’t plot based on the data at all.

00:39 They’re just line ranges that I told Matplotlib to stick in there. Because the luminosity axis is graphed in log scale and the absolute magnitude is a log number, it actually kind of works out relatively well as an approximation.

00:54 Did you hear the but coming? But the B-V color axis is rather misleading, although the temperature is graphed as log, the relationship to B-V color and temperature is discontinuous.

01:07 The formula is actually two formulas, one for when the log of the temperature is less than four and a different one for when it’s bigger than four.

01:14 Now, both formulas are log-based, so it sort of works, but the B-V index value for 10,000 Kelvin is closer to zero, and our axes is currently approximating it at 1.5.

01:27 There is a mechanism in Matplotlib where you can specify the math relationship between two axes. So for example, if you wanted to plot degrees in radians, you can use a special call to do that, but using it with B-V color is messy, and I wanted to keep the code simpler, even if it is a little less accurate. To compare, this is the HR Diagram from the Wikipedia page on that same topic.

01:51 This one doesn’t change the size of the dots and the temperature is on top while the B-V color is on the bottom, and it also has these pretty purple lines that divide the groupings of stars into size-based categories, white dwarfs, giants, and supergiants for example. One of the things that HR Diagrams reveal is that there tends to be gaps.

02:11 The different classifications of stars tend to clump together. There really aren’t any 6,000 Kelvin stars with luminosity of 0.1. In fact, almost all the 6,000 Kelvin stars are in the one to 10 luminosity range. As this is real-world data, that’s not perfect, there are exceptions, but it kind of tells scientists that there’s patterns about how stars behave and how they exist.

02:36 That’s the core part of the course. Last up, I’ll summarize and point you at other places for information.