Global Azure Science Lab 2017

Sharing is caring   Email this to someoneShare on FacebookTweet about this on TwitterShare on Google+Share on LinkedInPin on PinterestShare on RedditShare on StumbleUponDigg this

One of the missions of Global Azure Bootcamp is to provide a lab that gives researchers access to large scale computing resources to help process vast quantities of data. In previous years we have tried to bring you a science lab that is both interesting and represents great research. This year is no exception, this year we bring you the stars themselves!

Galaxies are the largest single structures in the universe, which harbor almost all the known types of astronomical objects. Among these, the stars represent the building blocks since they process the gas, create new planetary systems, form black holes, and their light makes the galaxies visible to ourselves. Many galaxies form stars throughout their lifetimes, and understanding how they do so is essential to understand the processes driving galaxy formation and evolution, which effectively affect the whole universe.

Dwarf galaxy LGS-3 from Hubble

Figure 1: Dwarf galaxy LGS-3 observed with the Hubble Space Telescope

The light from the stars of distant galaxies, like that in Figure 1, reaches the telescopes where their brightness (also called magnitude) and color are obtained. Astrophysicists gather this information using color-magnitude diagrams (CMDs) like that in Figure 2, left panel. In these diagrams, for a specific metal content (also know as metallicity), stars are distributed in particular sequences where their ages can be identified by comparing them with predictions from models of stellar evolution.

In these CMDs, a very important location is the so-called “turn-off”. This is the point at which a star begin to change its magnitude and color, becoming brighter and redder, and this is caused by the exhaustion of the star’s hydrogen fuel in its interior. At this point, the ages and metallicities of the stars can be calculated more precisely. With this information, the age and metallicity distribution of the stars formed in a galaxy can be obtained. This is called the Star Formation History (SFH), and it shows how many stars of a specific metallicity have formed in a galaxy during its lifetime.

One of the most efficient ways of testing galaxy formation processes is to study in detail the properties of the oldest stars and in particular, their ages, chemical compositions, and kinematics. An accurate determination of the SFH is a key aspect here and it depends on how precisely we can determine the exact moment when the stars reach the “turn-off” in the CMD.

Color-magnitude diagram unaffected by blurring Color-magnitude diagram with observational blurring

Figure 2: Distribution of the luminosity (magnitude) as a function of the temperature (color) of stars. For a given content of heavy elements (often called “metals” in Astrophysics), this color-magnitude diagram shows particular sequences where the ages of the stars can be identified (vertical axis, the redder the color, the older the star). On the left a color-magnitude diagram not affected by the blurring of observations, where the ages of stars can be determined with satisfactory precision. On the right, the observational blurring makes it difficult to determine stellar ages. Credit: Sebastián Hidalgo. (IAC)

However, it is not easy to obtain this information from the light that reaches us from the galaxies, and research groups find themselves with certain difficulties in analyzing it. For example, the uncertainties associated with the observations blur the CMDs (Figure 2, right panel), making it more difficult to determine the ages of the stars. Besides this, the limited accuracy of the models and the limits when running computer codes affect the results, showing SFHs in which some details are missing. The consequence is that some characteristics of the star formation remain “hidden” and hamper the interpretation of the results.

Star formation history

Figure 3: It shows in red the star formation history (SFH) affected by the observations. If these effects could be corrected three peaks could be obtained and distinguished in more detail (in blue) instead of a single peak. Credit: Sebastián Hidalgo (IAC).

For example, Figure 3 shows in red a SFH of a galaxy. The vertical axis shows how many stars have been formed as a function of age and metallicity. This is what is obtained when all the blurring effects described previously affect to the observation, a single extended peak of star formation. However, if we were able to correct of such effects, we would obtain the “true” SFH of the galaxy, that in blue, which shows three different peaks separated by periods of low activity in star formation. This type of “hidden” information is what is expected to obtain by using the Seliga (SEcret LIfe of GAlaxies) algorithm developed by Sebastian L. Hidalgo from the Instituto de Astrofísica de Canarias (IAC).

The objective of the Seliga algorithm is to limit the impact of all these effects so we can compare the predictions of the models more directly with the observations. This task needs a huge number of tests that can be performed successfully only by using distributed computing, like the one deployed in the Global Azure Bootcamp Science Lab.

With the science lab, you will be helping to contribute to the body of knowledge in this important field that allows researchers to understand the very beginnings of the universe itself. We hope you choose to deploy the packages that will run the Seliga algorithm, and deliver real results to the researchers.

Thank you to David Rodriguez and Adonai Suarez from Intelequia for helping with the Azure parts of the solution and giving their time freely, Martin Abbott and Wesley Cabus from the Admin team for wiring up and building the dashboard, and thank you to Sebastian and the team at IAC for giving us this opportunity to uncover the history of the stars themselves!

Instituto de Astrofísica de Canarias


Instituto de Astrofísica de Canarias – IAC

Microsoft Spain Press Release (Spanish).