Institute of Astronomy


Ask an Astronomer at the IoA

Ever had a question about astronomy you've want answered? Have a look through the previous questions which we've been asked and if you can't find find your answer, ask us!

Ask a question

Programming in Astronomy

Published on 03/01/2013 

What computer programming languages do astronomers use? IDL? Java? C++? C#? Ada?

The programming languages used in astronomy are somewhat down to the preferences of the individual astronomer, however I think it would be fair to say that the most commonly used ones are IDL, C/C++ and Fortran, though Python is becoming more and more popular as well.

Motion of the Galaxy

Published on 03/01/2013 

I'm a teacher in Belgium and get a lot of questions about the universe. I did my research but never found out one thing which is also my question. Is there proof that the whole Galaxy (the earth, sun, moon everything) also travels with a certain amount of speed?

There is indeed evidence that the whole galaxy, and indeed the whole of the Local Group of galaxies (the small cluster of galaxies that the Milky Way is part of) is moving relative to the rest of the universe.  What we use is the Cosmic Microwave Background.  This is radiation left over from the very early universe and is very smooth (the variations are less than 1 part in 100,000).  Since it comes to us from everywhere on the sky equally it makes a good reference to measure our speed against, and since it is very smooth you can see even quite small velocities.  If you look at this image the big gradient that you can see is due to our motion through the Universe, the radiation seems slightly hotter in the direction we are moving towards and slightly cooler in the opposite direction due to the Doppler effect.  Now of course Earth is orbiting the Sun, and the whole Solar system orbits the centre of the galaxy, but we know what these motions are and they don't account for all of the motion in that image, so the Milky Way must also be moving.

Using astronomy to date historical events

Published on 03/01/2013 

I cannot get any clear answers to what should be a simple question. "What percentage accuracy do the ancient astronomers have in fixing dates for the first millennium BCE?"

I am researching the period 1000BCE to 00BCE.
There is a wealth of information about Sumerian and Babylonian astronomical dating.
The "Kings lists" are constantly mentioned.
I simply want to have a reasoned answer to the following:

Allowing for human error, scribal miscopying etc - did the ancients in 1000BCE to 00 accurately observe and plot the planets so that we can retrospectively today recognise that which they recorded and afford it absolutely certain dates (to the nearest five years or so)
Do we know, from the positions that they give a planet in the sky that they called X, that it was that which we call, say, Mars.

Are there examples which clearly establish this e.g.
Carved in stone –
“In the 4th year of King X reign Planet V appeared in the west (whatever) and then moved behind the full moon to disappear in the northeast" - this today our computers tell us is exactly a description of Venus on 4th December 560BCE
Carved in stone –
In the 56th year of the reign of King W planet H did rise in the east and set at midnight one step into the quarter west(whatever) - this is an exact description of Mars on 4th January 500BCE"
Carved in stone –
King X was replaced by King W in his eighth year.

That is - we now know these astronomical events were indeed 60 years apart, they could be seen exactly like that from Babylon and it is clear that this was indeed the 60 years from King X 's 4th year to King W's 56th year.

There is much discussion of accuracy and more of eclipses. The only solid facts must surely come from verifying planet observation and then tracking those dates and comparing them to those given by the ancients.

Or as usual am I being simple minded?

As you have found astronomical observations can indeed be used to help provide support for chronologies of the ancient world, no method is perfect however and there are problems with this.

All astronomical observations one can make related to the Solar system contain many layers of cycles.  Everything of course varies on the cycle of a year as the Earth orbits the Sun, however due to the motion of other bodies and slow periodic changes in Earth's orbit there are also longer period cycles.  For example the times at which Venus rises and sets has a cycle of about 8 years (as well as other longer ones).  In general the shorter cycles are the most prominent while distguishing where an observation falls in a longer term cycle is more difficult.

The end result is that for your example stone carving there might be a match with 560BCE, 552BCE, 544BCE, 536BCE or 528BCE.  Sometimes knowing that it must be one of those dates might be sufficient to combine with other data and pin down an exact match, and sometimes even being able to pin it down within 5 cycles or whatever the case may be might be superior to what was known before, but gaps and inaccuracies in ancient records make it difficult to pin down exact dates from historical astronomical observations.

The most accurate records tend to be those of eclipses (because they are hard to miss and have a short duration), which helps with placing individual eclipses within the longest eclipse cycles and so providing more precise dates.  There are still problems with gaps simply because the further back you go the fewer records have survived whether or not they were originally taken, and scribal errors accumulate over time and are difficult to account for.

An additional problem is that in early history (and even comparatively recently) the calendars were not constant.  Many ancient calendars, such as that used by the Babylonians, were lunisolar with months based on lunar phases plus leap months as required to keep reasonable synchrony with the solar year.  That is fine provided that one knows how the scheme on which the leap months are added, but therein lies the difficulty in that they tend to be adjusted on a more ad-hoc basis or systematically over the medium term but with unknown larger jumps in the long term (think about the break that occurred in England in 1752 with the switch from the Julian to Gregorian calendars).  In fact even today this problem exists in the form of 'leap seconds' added by the international body which supervises global time standards, as they are irregular and unpredictable.

Sorry I can't really give you a simple answer, but I hope this goes some way to at least explaining why it isn't a simple question!

Definition of Meteoroid vs Meteorite

Published on 03/01/2013 

Does a meteoroid have to become a meteor first, before being defined as a meteorite when reaching the surface of a planet? So, in other words a meteoroid that impacts our moon, which does not have an atmosphere, would never be classified as a meteor. So when it impacts the moon's surface, is it still a meteoroid or meteorite?

There is no requirement for a meteorite to have previously been a meteor.  You are quite right that in the case of something impacting the Moon as there is no atmosphere it will never be a meteor, so it will be a meteoroid right up until it hits the surface, at which point it is a meteorite.

Gravitational waves from the Sun and Moon

Published on 17/12/2012 

Having attended an outreach talk on gravitational waves, we had a couple of questions:

1) The moon is a massive orbiting object near to the Earth. Does it produce measurable gravitational waves? If so, do these waves affect the measurements taken by LIGO?

2) The sun is changing mass due to solar wind, flares and nuclear reactions. Is it possible to measure this effect with current (or near future) gravitational wave detectors? We appreciate that it won't be an oscillation but it will be changing the curvature of space-time as the Earth orbits it over time.

Glad that you enjoyed the lecture. In answer to your questions:

  1. Yes, the Earth-Moon system will produce gravitational waves! Because the Earth and Moon aren't that massive (compared to black holes say), they won't be too big; you might think they are still important because we are right there... As it is, the waves are at too low a frequency for detection with LIGO: the Moon orbits about one every 28 days, and the ground based detectors are not sensitive to frequencies low than about 1 Hz (one cycle per second).

    However, more directly, the Moon does affect LIGO though its gravitational pull. This is a direct force that can pull at the mirrors at the end of the arms and move them slightly. Fortunately, since we know where the Moon is, we can take this effect into account. In fact, the detectors are so sensitive, that not only must we take into account the Moon, but also the tides. The tides are the movement of water in response the the gravitational pull of the Moon and the Sun. At different points in the cycle there is a different distribution of water across the Earth, and the pull from having extra water in regions close to detectors can be measured.

    In conclusion, the Moon orbitting will produce gravitational waves, but these are not of the right frequency to be detected. However, the Moon's gravity still needs to be taken into account.

  2. You are quite right that the Sun is changing mass. This change is quite small, at least over human time-scales (the Sun will live for billions of years, so only changes by a tiny amount over a few years, or even centuries). I don't believe that we are currently able to measure the mass of the Sun precisely enough to observe the change; however, this may be possible in the future. Certainly the technology developed for gravitational wave detectors could be adapted to precisely map the Sun's gravitational field. I don't think anyone has come up for a mission design for this yet, although we have recently had missions that have mapped the Earth's and the Moon's gravitational fields (GRACE, GOCE and GRAIL; a GRACE II mission is planned, and will hopefully include technology originally designed for LISA).

    Regarding gravitational wave measurement, you've hit a very important point. There shouldn't be any (significant) oscillation. To excite gravitational waves you must have something like a binary (technically where the mass quadrupole is changing) and not something spherically symmetric. The Sun's mass loss is very nearly perfectly spherical, and so there shouldn't be any real gravitational wave emission.