|
|
Post by chaos on May 21, 2020 21:50:35 GMT 5.5
Hey everyone, for our next task we are going to attempt to generate a Hertzsprung-Russell Diagram (HR-Diagram). A Hertzsprung-Russell diagram is a method for classifying stars. It has two major axes - luminosity, and effective temperature. Upon generating this diagram, we can classify several stars, and in fact this is what led to the first proper stellar classification and the modern understanding of stellar evolution.
To generate an HR diagram, we will have to analyze data on a much larger scale than what we did for cluster distancing. For this, we will first pull the HYG data from astronexus.com and if possible, try to analyze data from Gaia itself.
This is to be a collaborative project, but it will require significant skill in programming. As always, if you have to learn programming, the best way to learn it is learning on the job. If you're truly interested - we have exams on the horizon, so we'll target to finish at least the HYG data by the third week of June - reply in this thread.
|
|
|
|
Post by ArchDukeBarbatos on May 21, 2020 21:59:15 GMT 5.5
This is a really good idea. What about the statistics involved? Could you make a list of prerequisites?
|
|
|
|
Post by HuLaLaBoO on May 21, 2020 22:03:05 GMT 5.5
Yasss Boii✌✌
|
|
omg
CAC Active Member 
Posts: 2
|
Post by omg on May 22, 2020 17:23:52 GMT 5.5
Yes good idea 🙌
|
|
|
|
Post by Rucha on May 22, 2020 21:29:47 GMT 5.5
OK. Excited!! I have a few queries though : - When you say it will be a collaborative project, does that mean we all are going to work on a common code ?
- And, is the data supposed to be downloaded or will we attempt to do something like web scraping ( quite hard , in my opinion) ?
Also, if we someone gets some good reference for studying the basics involved in HR diagrams. Please post it on this forum |
|
|
|
Post by ArchDukeBarbatos on May 23, 2020 19:12:38 GMT 5.5
The HR Diagram is a plot of Luminosity(Relative magnitude at 10 parsec) vs the effective temperature of the star. The plot is given by, L = 4*pi*R^2*sigma*T^4, where R is stellar radius and T is effective temperature. Following site has methods about effective temperature. www.ucolick.org/~bolte/AY4_00/week4/star_temp.htmlThe HYG data has RA/Dec, Distance, Relative Magnitude, Absolute Magnitude, Spectrum and Colour Index as recorded parameters. All in all our task is finding effective temperature of stars, as no errors are involved, this task is relatively easy I feel. |
|
|
|
Post by chaos on May 24, 2020 21:10:40 GMT 5.5
It seems easy at face value, and it is, but the problem will be scaling it up to Gaia data. Right now, I'd like everyone to put forth their views on two things:
1. Since the HYG data has no errors, we have to generate our own errors for stars that lie on the edge of the main sequence and drift into white dwarf territory.
2. What methods can we use to resolve these fuzzy patches?
Also I'd like one of you to volunteer for preparing the code for generating the basic HR diagram. It's not difficult, and maybe it'll be easier if two of you work on this - one to create the method and one to write the actual code.
|
|
|
|
Post by ArchDukeBarbatos on May 26, 2020 19:41:14 GMT 5.5
It seems easy at face value, and it is, but the problem will be scaling it up to Gaia data. Right now, I'd like everyone to put forth their views on two things: 1. Since the HYG data has no errors, we have to generate our own errors for stars that lie on the edge of the main sequence and drift into white dwarf territory. 2. What methods can we use to resolve these fuzzy patches? Also I'd like one of you to volunteer for preparing the code for generating the basic HR diagram. It's not difficult, and maybe it'll be easier if two of you work on this - one to create the method and one to write the actual code. Could you elaborate points 1 and 2? Also The two man thing seems quite good. |
|
|
|
Post by chaos on May 28, 2020 10:36:51 GMT 5.5
It'll be made more clear when we actually generate our first HR diagram.
What we're looking for is fine structures in the diagram i.e. those small lines that lie on the edge of main sequence and white dwarves, or main sequence and giants. For them we'll have to assume that something's wrong in the measurement. In general all measurements are wrong, but since the HYG data is made of 'interesting stars' by the guy's own admission, we don't know what he's picked for a bast majority of them. As a starting step, we can assume the temperature error to be 10% and see where it leads us. This isn't strictly important for HYG, but 100% for Gaia.
Also, does no one want to volunteer?
|
|
|
|
Post by chaos on Jun 16, 2020 18:57:11 GMT 5.5
Alright, no one responded to this. I ended up doing it on my own, which is something I'd really not like to see again. Anyway, here's the HR diagram for the HYG database, which we will hopefully scale up to Gaia's data, but I'm not confident at all seeing the response. The code that enabled this to happen is the product of messing around a lot in a Jupyter notebook, and I'll post it as soon as I can. The two gifs I generated are attached to this post. drive.google.com/drive/folders/1tfv3x1TbjIwvIVe8noTU-kWN8WyAX1RD?usp=sharingHere's the link to the folder for the two gifs. The method's going to be explained as soon as I finish cleaning up the code. |
|
|
|
Post by ArchDukeBarbatos on Jun 16, 2020 21:11:47 GMT 5.5
I will try writing a robust code based on hygfull.csv that is v1 next week
|
|
|
|
Post by chaos on Jun 17, 2020 15:36:21 GMT 5.5
Here's the code for the H-R diagram generator. Despite my best efforts, I did not manage to run it in the terminal, and had to use a Jupyter notebook. The method is simple - it's a graph of the absolute magnitude vs the color index. Since a brighter star has a lower magnitude, you have to flip the y-axis. The twinkle effect is generated with the help of a lambda function, and the gif is generated with Imagemagick. github.com/astronexus/HYG-Database is the data used. To convert the color index to RGB values, the method found in stackoverflow.com/questions/21977786/star-b-v-color-index-to-apparent-rgb-color was used, converting the code to Python. Next stop, the Gaia data. Attachments:hr.py (3.16 KB)
|
|
|
|
Post by ArchDukeBarbatos on Jun 17, 2020 17:26:46 GMT 5.5
Will Python be sufficient for Gaia?
|
|
