• Question: Our textbooks state “observations tell us…” when explaining what we currently know about the Universe but what the observations are feel quite abstract to me and my students. What do astrophysicists actually “observe”? What sort of data do you collect and how?

    Asked by Mrs M to Susan, Meirin, Anne on 7 Nov 2019.
    • Photo: Susan Cartwright

      Susan Cartwright answered on 7 Nov 2019:

      This is way too general a question! What you observe depends on what you want to know, and how you collect the data depends on what you are observing. I will give a few examples, but if you want a useful answer you really need to be much more specific in your question.

      Example 1: stars.
      What do we observe?
      Primarily, electromagnetic radiation; mostly in optical and infra-red wavelengths.
      The data consist of (1) the apparent brightness of the star; (2) its colour; (3) its spectrum; (4) its position as a function of time. What you would probably want to know is (A) its mass; (B) its chemical composition; (C) its physical properties (size, surface temperature); (D) its age. How you get from one to the other, for a relatively nearby star in our Galaxy, is:
      – the pattern of spectral lines in the star’s spectrum tells you its surface temperature and surface chemical composition
      – the appearance of the lines tells you about its surface gravity, which relates to its size and density
      – its motion over the course of a year (its parallax) tells you its distance, and when you combine this with its apparent brightness you get its luminosity (i.e. the amount of energy emitted per second)
      – its luminosity and surface temperature together give you its surface area, and therefore its size
      – its mass can only be determined directly from observation if it is part of a binary system with another star (fortunately this is quite common). You then use its motion to determine the parameters of its orbit, and then apply Newton’s law of gravity to determine the masses involved.

      Example 2: cosmology
      If you want to know what happened early in the history of our Universe, there are a number of things that you can observe. The most important is the cosmic microwave background (CMB), which is radiation emitted when the Universe first became transparent to light, about 380000 years after the Big Bang. The CMB is best observed from space, because it is brightest in the far infra-red and submillimetre wavelengths, which don’t get through the atmosphere. The most recent spacecraft to do this was Planck.
      The key feature of the CMB is that, although it is extremely uniform over the whole sky (this in itself is an interesting fact), it has tiny temperature variations: slightly hotter (we are talking microkelvin here) and slightly cooler regions. The characteristic size of such regions (about one degree) and the extent to which they are hotter or cooler than their neighbours, depends on various properties of the Universe: how much matter (of different types) there is, what its geometry is (is it flat or curved?), how fast it is expanding, etc. By carefully studying the temperature variations, it is possible to extract an enormous amount of data from them, but the actual extraction process is very complicated (I can tell you that it’s a multidimensional Bayesian likelihood analysis, but somehow I don’t think that’s going to make you feel much wiser).

      Example 3: exoplanets
      Detecting planets around other stars is challenging because you are looking for a very small faint thing right next to a much larger brighter thing. If you want to observe fireflies, you don’t look for them next to searchlights. There are two main methods:
      (1) Transits: if a planet passes between us an its star, it will dim the star’s light ever so slightly (Jupiter passing in front of the Sun would block 1% of the Sun’s light; the Earth would block only 0.01%). This can be observed. It’s easiest to do this from space as the star’s light is steadier if it is not being distorted by atmospheric turbulence. The information you get is the size of the planet and its orbital period.
      (2) Radial velocity: a planet does not really orbit its star: rather, star and planet orbit their common centre of mass. You can detect the star’s orbit by observing very small Doppler shifts in its spectral lines. The information you get is a lower limit on the mass of the planet (only a lower limit, because you do not know the angle at which you are viewing the orbit, and the Doppler shift only gives you the velocity component along the line of sight) and the orbital period.

    • Photo: Anne Green

      Anne Green answered on 8 Nov 2019:

      As Susan’s explained in detail, “observations tell us…” is short-hand for “we collect data and when we analyze it and compare the results to theoretical models we find…”. In astronomy and cosmology the data we collect is often in the form of photons from stars, galaxies or the cosmic microwave background radiation. However with new sensitive detectors we now have other ways of “observing” the universe. For instance gravitational waves and neutrinos.


Leave a comment

Log in to comment.