Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Lithium Elemental Abundance Determination for 4MOST

Solly, Emma LU (2025) FYSK04 20251
Department of Physics
Astrophysics
Abstract
Stellar lithium abundances can help answer multiple unanswered questions within modern astrophysics. Lithium is fragile in comparison to other elements found in stellar atmospheres. Consequently, its abundance within the stellar atmosphere is depleted by different processes. Studying these lithium abundances gives insight into the evolution of the stars and the processes within their interiors. The main aim of this thesis was to develop a method for calculating lithium abundances, given spectra measured by the 4-metre Multi-Object Spectroscopic Telescope (4MOST); a fibre-fed spectrograph which is to begin its operation in 2026. Our method was developed on high resolution synthetic 4MOST spectra with a signal to noise ratio of 1000, and... (More)
Stellar lithium abundances can help answer multiple unanswered questions within modern astrophysics. Lithium is fragile in comparison to other elements found in stellar atmospheres. Consequently, its abundance within the stellar atmosphere is depleted by different processes. Studying these lithium abundances gives insight into the evolution of the stars and the processes within their interiors. The main aim of this thesis was to develop a method for calculating lithium abundances, given spectra measured by the 4-metre Multi-Object Spectroscopic Telescope (4MOST); a fibre-fed spectrograph which is to begin its operation in 2026. Our method was developed on high resolution synthetic 4MOST spectra with a signal to noise ratio of 1000, and then applied to spectra with a lower signal to noise ratio of 100. In order to calculate the abundance of lithium, the equivalent width of the Li I resonance doublet at 6707.814 Å was first calculated. This line is heavily blended with absorption lines from other elements, including iron, in spectra with sufficient metallicity. In order to calculate the equivalent width of the Li I line in these cases, we assumed that the width of the spectral lines is constant in a wavelength region of 6695 - 6720 Å. Applying a Gaussian fit to unblended lines in this region allowed for measuring this width. A second fit, made to a narrower region across the blended line, was used to obtain the amplitude of the Li I line. The width and amplitude were then used to calculate the equivalent width.

For cases where fewer than three unblended lines were detected in the region around the Li I line, a Gaussian fit was made directly to it. Implemented conditions were used to identify successful, as well as unsuccessful, fits. Determining appropriate conditions for filtering the fits was non-trivial. However, multiple sources of unsuccessful fits were identified.

The curve of growth was used to obtain a lithium abundance from an equivalent width and effective temperature. The dependence on the effective temperature was observed to be significant and so the lithium abundance needed to be calculated as a function of it. The method was applied to 12704 spectra with signal to noise ratio of 100. Out of these, 2775 spectra yielded fits that passed all steps of the filtering process. 4MOST requires the recovered lithium abundance to not vary with more than 0.2 dex from the input lithium abundance. This was achieved for 2000 of the spectra to which successful fits were considered to have been made. A secondary aim of the thesis was to compare our results to those given by the 4MOST pipeline. While a direct comparison could not be made, it seems that the results of our method are comparable to the ones given by the pipeline. Our method has the advantage that it was able to identify cases where a reliable determination of the lithium abundance could not be made. (Less)
Popular Abstract
Twinkle twinkle little stars, how I wonder what your lithium abundances are... That is of course not how the song goes. However, this version might be more relatable to an astrophysicist who already knows quite a bit about what the star on the night sky is, but not as much about its evolution. Astronomers have studied the night sky for centuries and have gained insight about the objects adorning it. This has been done by observing the objects' movement and light. Planets were distinguished from stars by their relatively high velocities across the sky and their steady, non-twinkling light. Modern astrophysicists also rely upon the light and dynamics to draw conclusions about the observed objects. Today, techniques such as spectroscopy and... (More)
Twinkle twinkle little stars, how I wonder what your lithium abundances are... That is of course not how the song goes. However, this version might be more relatable to an astrophysicist who already knows quite a bit about what the star on the night sky is, but not as much about its evolution. Astronomers have studied the night sky for centuries and have gained insight about the objects adorning it. This has been done by observing the objects' movement and light. Planets were distinguished from stars by their relatively high velocities across the sky and their steady, non-twinkling light. Modern astrophysicists also rely upon the light and dynamics to draw conclusions about the observed objects. Today, techniques such as spectroscopy and photometry are used to study the light in detail.

Radiation produced in the core of a star is absorbed by atoms and molecules in its atmosphere, and then emitted again in random directions. This absorption gives absorption lines at specific wavelengths in the light spectrum that is observed. The wavelengths correspond to a certain energy of the photons, given by a specific energy level separation within an atom or molecule. This gives astrophysicists information about what elements are present in the stellar atmosphere. The strength of the absorption lines is affected by how much absorption has taken place, which in turn is dependent on the amount of the corresponding absorbers. The abundance of a certain element in the stellar atmosphere can therefore be derived from studying the strength of an absorption line that has been caused by atoms of that element.

This process is unfortunately not so straight forward, seeing as one cannot directly read the abundance from the spectrum. The strength of the absorption lines is affected by multiple stellar parameters, not only the abundance of the absorbers. They can also be heavily blended, meaning that the lines are too close in wavelength to distinguish between them. They are instead observed as one line. In this project the element of interest is lithium, which has an absorption line at the wavelength 670.781 nm. Some heavier elements, including iron, give absorption lines close to this wavelength. This results in the lithium line being blended in the spectra from most stars with heavier elements present in their atmospheres. The method developed in this project addresses this issue by assuming that the width of the absorption lines is constant in a region around the lithium line. This width is derived from non-blended lines in the region and the calculation of the strength of the lithium line is simplified, seeing as one now only needs to determine the depth of it via a fit to the blended line.

The aim of this project is to determine the lithium abundance of stars observed with 4MOST, which is an instrument that will observe around 2400 objects on the sky at one time and is estimated to collect tens of millions of spectra. The reason why the lithium abundance is of interest is that lithium is fragile. It is destroyed at temperatures above 2.5 MK, which is low in the context of stellar interiors. Certain events in a star's life have been found to cause a decrease in the abundance of lithium. The abundance can therefore give insight into the evolution of the star. So keep on twinkling little stars, while I try to figure out what your lithium abundances are. (Less)
Please use this url to cite or link to this publication:
author
Solly, Emma LU
supervisor
organization
course
FYSK04 20251
year
type
M2 - Bachelor Degree
subject
keywords
4MOST, lithium abundance
report number
2025–EXA244
other publication id
2025–EXA244
language
English
id
9198623
date added to LUP
2025-06-18 12:00:22
date last changed
2025-06-18 12:00:22
@misc{9198623,
  abstract     = {{Stellar lithium abundances can help answer multiple unanswered questions within modern astrophysics. Lithium is fragile in comparison to other elements found in stellar atmospheres. Consequently, its abundance within the stellar atmosphere is depleted by different processes. Studying these lithium abundances gives insight into the evolution of the stars and the processes within their interiors. The main aim of this thesis was to develop a method for calculating lithium abundances, given spectra measured by the 4-metre Multi-Object Spectroscopic Telescope (4MOST); a fibre-fed spectrograph which is to begin its operation in 2026. Our method was developed on high resolution synthetic 4MOST spectra with a signal to noise ratio of 1000, and then applied to spectra with a lower signal to noise ratio of 100. In order to calculate the abundance of lithium, the equivalent width of the Li I resonance doublet at 6707.814 Å was first calculated. This line is heavily blended with absorption lines from other elements, including iron, in spectra with sufficient metallicity. In order to calculate the equivalent width of the Li I line in these cases, we assumed that the width of the spectral lines is constant in a wavelength region of 6695 - 6720 Å. Applying a Gaussian fit to unblended lines in this region allowed for measuring this width. A second fit, made to a narrower region across the blended line, was used to obtain the amplitude of the Li I line. The width and amplitude were then used to calculate the equivalent width.

For cases where fewer than three unblended lines were detected in the region around the Li I line, a Gaussian fit was made directly to it. Implemented conditions were used to identify successful, as well as unsuccessful, fits. Determining appropriate conditions for filtering the fits was non-trivial. However, multiple sources of unsuccessful fits were identified.

The curve of growth was used to obtain a lithium abundance from an equivalent width and effective temperature. The dependence on the effective temperature was observed to be significant and so the lithium abundance needed to be calculated as a function of it. The method was applied to 12704 spectra with signal to noise ratio of 100. Out of these, 2775 spectra yielded fits that passed all steps of the filtering process. 4MOST requires the recovered lithium abundance to not vary with more than 0.2 dex from the input lithium abundance. This was achieved for 2000 of the spectra to which successful fits were considered to have been made. A secondary aim of the thesis was to compare our results to those given by the 4MOST pipeline. While a direct comparison could not be made, it seems that the results of our method are comparable to the ones given by the pipeline. Our method has the advantage that it was able to identify cases where a reliable determination of the lithium abundance could not be made.}},
  author       = {{Solly, Emma}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Lithium Elemental Abundance Determination for 4MOST}},
  year         = {{2025}},
}