The star writes a pattern into its own light
Picture a piece of material heated until it glows. It emits a broad range of colors, so its spectrum forms a smooth bright band. Dense gas deeper in a star supplies a similar background. That light passes through an atmosphere whose temperature and density change on its way to our telescope.
An electron in an atom can occupy only certain energy states. A photon with the right energy can lift it into a higher state. The corresponding wavelength is especially easy to absorb. Many atoms acting together remove some of that light from our direction, producing a dark notch called an absorption line.
The absorbed energy has not vanished. Atoms can emit again, and light travels in other directions too. How much radiation the layer puts back along our particular view depends on its own state. We must compare the layer’s brightness with the background to understand whether the resulting line is dark or bright.
Reading: Krumholz, ASTR 4003/8003 (2019), Class 1, Radiative transfer I · Condon & Ransom, Essential Radio Astronomy, Appendix E
Heating a layer changes two things at once
Put a cool layer in front of a bright background and absorption can dominate. As the layer warms, its own emission fills part of the notch. If its thermal source brightness matches the background, the line disappears in the simplest uniform-layer model. A still hotter layer produces an emission line: a narrow feature brighter than its surroundings.
Temperature also changes how many atoms occupy each energy state. Some lines require the absorbing atom to begin in an excited state. Heating can put more atoms there and strengthen absorption. Further heating can ionize atoms, changing the available population again. The observed depth can reflect all these processes together.
That is why astronomers compare several lines. One may respond strongly to temperature, another more directly to the number of available absorbers. Their different responses help separate the causes. A single deep notch leaves much more room for ambiguity.
Reading: Condon & Ransom, Essential Radio Astronomy, Appendix E
An opaque center leaves room for wider wings
In a sufficiently thin gas, doubling the absorbers roughly doubles the missing light. Eventually the center becomes opaque. Almost none of the background was getting through there already, so extra atoms cannot make it much darker. Farther from the center, the gas was more transparent. Those wavelengths now absorb more, widening the line.
To compare depth and width together, add up the missing light across the whole notch. Replace that area with a rectangle of completely missing light. The rectangle’s width is the equivalent width. A shallow broad line and a deep narrow line can have the same equivalent width.
An instrument spreads light across nearby detector pixels as well. A spectrograph with lower resolving power can turn a deep line into a shallower, wider one. A correctly normalized instrument response preserves the integrated missing area when the wavelength window contains the full line and the continuum is correctly placed. Comparing area and shape helps separate an instrument change from a change in the gas.
A pattern that moves together reveals motion
An approaching siren delivers more closely spaced wave crests; a receding siren delivers more widely spaced ones. Light has a related Doppler effect. Gas moving away shifts its lines to longer wavelengths. Gas approaching shifts them to shorter wavelengths. Astronomers call the component along our view the radial velocity and conventionally take recession as positive.
One line can be blended with a neighbor or misidentified. A whole pattern is more useful. If several lines all move by the same fractional wavelength, they share a clear motion signal. Random atomic speeds instead create many slightly different shifts within the gas, which broaden the combined profile.
Turning these clues into a temperature or elemental abundance requires a model for how light travels through the atmosphere. A single uniform layer is a useful first calculation because absorption, emission and motion can be varied separately. The more demanding question is how well that picture describes a real stellar surface.
Reading: Krumholz, ASTR 4003/8003 (2019), Class 1, Radiative transfer I · Amarsi, Liljegren & Nissen (2022), 3D non-LTE iron abundances in FG-type dwarfs, arXiv:2209.13449v3