I have been looking for some kind of collection of all known stars and their individual respective mass/radius/temperature/and luminosity, can you link me to one if it exists?

My goal is to write an algorithm that creates polynomial functions that calculate a stars temp/lum/radius based on mass. I know there are existing equations, however the ones I have found are more troublesome and imprecise compared to if I could use polynomial functions I create myself.

The Hertzsprung-Russel diagram ( http://upload.wikimedia.org/wikipedia/commons/1/17/Hertzsprung-Russel_StarData.png ) is a great example of representation of what data I need, however it's axes are not constant, so I can not visually gather accurate sample data to deduce any functions.

Edit: Essentially even a large quantity of estimates will suffice. I intend to use this data to generate a semi-realistic galaxy for an MMO I am developing, so it doesn't by any means need to be perfect.

  • $\begingroup$ What do you mean by "known stars"? The answer is basically no, there isn't any complete database. Masses in general cannot be found for anything but binary stars. Radii can only be measured for objects in eclipsing binaries or nearby objects. Luminosities rely on distances and these are not known accurately (better than 10%) for more than about 10,000 stars. $\endgroup$
    – ProfRob
    Commented Dec 18, 2014 at 20:29
  • $\begingroup$ Ok, I would easily settle for a smaller sample. Several thousand or more would suffice. There are functions capable of deriving a specific property from others, for example there are functions that return an estimated radius based on only mass as an input variable. If there is a sample of 10,000 stars with my requested data estimated, that will easily be to satisfaction. $\endgroup$ Commented Dec 18, 2014 at 20:37
  • $\begingroup$ There are probably no more than a couple of hundred stars with reasonably known masses. $\endgroup$
    – ProfRob
    Commented Dec 18, 2014 at 20:39
  • $\begingroup$ If I only need masses estimated to the nearest tenth of a solar mass, would that increase the number of results? By the way thanks for your responses. $\endgroup$ Commented Dec 18, 2014 at 20:44
  • $\begingroup$ No, because those masses will be estimated using the very relationships you are hoping to construct! Or the masses are estimated by using theoretical mass-luminosity relationships. If you are happy with those then why bother to make your own empirical relationships? $\endgroup$
    – ProfRob
    Commented Dec 18, 2014 at 20:49

2 Answers 2


There really isn't anything as comprehensive as you want. And, we don't have accurate information for all of these variables for a large number of stars.

The two main problems are masses and radii.

Really the only set of stars that have both measured masses and radii are those in eclipsing binary systems. If these binary stars also have distances, measured by parallax, then the components could also have luminosity estimates and estimates of effective temperature. The best catalogue I can recommend is that compiled by John Southworth, containing perhaps a 100 systems.


If you just want a mass, but can cope without a radius, then you can also include components of astrometric binary systems. There is a well-known paper by Delfosse et al. (2000) containing stellar masses and luminosities.

  • $\begingroup$ Good answer, and great link. If I had the reputation I would up-vote Link contains 344 stars with very high detail $\endgroup$ Commented Dec 18, 2014 at 21:38

I did something similar with the HRD about a week ago. I found that VizieR has a very large database with observations. You can download the tables in different formats (like csv or plain text), but you should first check if the Temperature (mostly log.Tegg) and luminosity (logL) are available. A very detailed one is table V/19. It has those values for 68 clusters, enough to keep you busy for a while!

There are also masses in that list, but you can also create an isochrone yourself, and determine the age of each cluster. Well that was my assignment, and we got pretty close to the expected values for some clusters!

  • $\begingroup$ Not sure how much information I can use, but this is definitely the best bank of information I could hope for. Lol enough to keep me busy? I think not :P Im writing a program that will analyze the data for me, designed to create dynamic functions I can use to generate my own galaxy. Basically my program analyzes ratios between different properties throughout the entire range of stellar masses for main sequence stars. $\endgroup$ Commented Dec 18, 2014 at 21:04
  • $\begingroup$ SUre, but you will first need to get the data in the right form, fix bugs, make sure the program works, fix bugs. Et cetera :). If you would draw this by hand you would be out of your mind! $\endgroup$
    – Mathias711
    Commented Dec 18, 2014 at 21:05
  • $\begingroup$ Of course you can search for a different table, with more/better results. It was sufficient for our own little research, and we were happy with the outcome, so no need to do further investigation in searching of different tables $\endgroup$
    – Mathias711
    Commented Dec 18, 2014 at 21:06
  • $\begingroup$ Parsing data is no problem for me. I've even got my own general framework to help with this. $\endgroup$ Commented Dec 18, 2014 at 21:08
  • $\begingroup$ Good luck then! $\endgroup$
    – Mathias711
    Commented Dec 18, 2014 at 21:10

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .