Infrared spectroscopy of cool giants

Abstract

Purpose: To get a clearer picture of the structure and evolution of our Galaxy, astronomers are trying to reach further and further beyond the Solar neighborhood with their observations. The dust obscuration towards the center of the Milky Way necessitates the use of infrared wavelengths for this, and cold giants provide a particularly bright and thus useful target. However, any methods for analyzing NIR spectra of cool giants still require more testing and validation. In this project, we test several methods for determining the fundamental stellar parameters (Teff, [Fe/H], and log g) from H- and K-band spectra, with special focus on the surface gravity. This enables future studies to then target these types of stars for detailed chemical abundance studies.

Method: We use a machine learning code (The Cannon) to evaluate which spectral lines are most sensitive to each fundamental stellar parameter. Based on this, we decide on three different approaches to determine the surface gravity of stars: i) Inferring it from isochrones, or finding it spectroscopically from ii) the molecular equilibirum or iii) the wings of strong lines. We then use a spectral synthesis code (SME) to determine the stellar parameters for our sample of cold giants with each of the three methods.

Results: We find that OH lines can be used for reliable temperature determination in stars below ∼ 4400 K. In terms of log g, relying on CN and CO lines introduces strong degeneracies between the surface gravity and the relevant abundances, making the molecular equilibrium method unreliable. Using the wings of strong lines was a promising approach that worked well in many cases; however, we recommend to additionally use CN and CO lines and to select a larger number of wings for the analysis. The isochrone method was the most reliable in our tests, recovering good estimates of the surface gravity for most of our cool stars. We confirmed these findings by determining the abundances of several s-process elements based on our fundamental parameters – again, the isochrone method’s results most consistently recovered the expected trends.

Conclusion: We have shown that relying on isochrones to determine the surface gravity of cool giants is a robust and reliable method. We further recommend fitting the pressure-broadened wings of strong lines for a spectroscopic determination of log g. Both methods (with possible improvements suggested in our discussion) can be used when studying the Milky Way bulge in the context of Galactic archeology.

Popular Abstract

It is many astronomers’ goal to understand exactly how galaxies like the Milky Way – our home – form and develop. Figuring this out is very difficult, however. For one, we do not get a neat top-down view of the Galaxy as a whole, like it is commonly depicted. We sit inside the galactic disk, so if we want to observe the center of the Milky Way, we need to look through a lot of dust and gas. Luckily for us, certain types of light can simply pass through dust without getting absorbed. This includes near-infrared wavelengths; the light just beyond what we humans see as the color red on the spectrum. Focusing on this kind of light allows us to probe deeper into the Galaxy with specialized telescopes. There is another problem, however: We cannot turn back time to see how the Milky Way looked billions of years ago. As it turns out, though, we can deduce a lot of information about the past from looking at the stars as they are today – this is what the field of Galacic archeology is all about.

Galactic archeology works with two main categories of information: Firstly, where the stars are and how they move, and secondly what they are made of. The stars in the Milky Way generally orbit the supermassive black hole at its center, in a more or less orderly fashion. If we find a group of stars that stick together and move in a different direction, it can be a sign that they used to belong to a separate (smaller) galaxy that collided with the Milky Way. Similarly, what chemical elements we can find in a star can tell us a lot about where it came from, as this chemical signature preserves information from the time when the star was formed. We can estimate how long ago stars formed, if they come from a region with a high or low density of stars, and much more.

Analyzing what chemical elements are present in stars is not trivial, however. We rely on a method called "spectroscopy", in which we split up a star’s light with a prism to examine how bright or dark it looks at specific wavelengths. We can then find different elements’ signature in that light: They each absorb particular wavelengths of light, creating narrow "spectral lines" where the star appears darker. In general, the more of an element there is, the darker the associated spectral lines will be. However, several other factors play into how dark or bright each spectral line is, including things like the temperature of the star. Thus, to accurately measure how much of the element is present, we need to also model several other characteristics of the star.

In this project, we focus on how to determine some of the most fundamental stellar characteristics with spectroscopy. These are the surface temperature, the metallicity (essentially an estimate of how many heavy elements are present in the star), and the surface gravity (a measure of how dense the star is). All of these factors play into how the stellar spectrum looks, including just how dark different spectral lines appear, but the individual effects are very difficult to disentangle. Particularly the surface gravity is tricky to determine with spectroscopy. We test several approaches for finding the surface gravity, including looking at a variety of spectral lines or considering theoretical models that connect the surface gravity to the other – more easily determined – fundamental characteristics. That way, if we can find the temperature and metallicity from the spectra, we can estimate what the corresponding surface gravity should be.

The aim is to establish a reliable method for determining the fundamental characteristics of cool, bright stars based on spectroscopy at infrared wavelengths. This lays the basis for future studies of the chemical composition of far away stars, which in turn will contribute to our understanding of how the Milky Way has formed and evolved.

• author: Kocher, Jessica Laura Erika
• supervisor: Govind Nandakumar; Nils Ryde
• organization: Astrophysics; Department of Physics
• course: ASTM32 20242
• year: 2024
• type: H2 - Master's Degree (Two Years)
• subject: Physics and Astronomy
• report number: 2024-EXA238
• language: English
• id: 9177274
• date added to LUP: 2024-11-22 12:43:38
• date last changed: 2024-11-22 12:43:38