Lunar eclipse montage
Montage of a partial lunar eclipse, by Sebastien Gauthier. Taken from Astronomy Picture of the Day.

Here is a list of astronomical observing projects that you can do yourself, either with the naked eye or with the aid of a decent pair of binoculars. The module numbers indicate which of the taught courses is relevant to each project. If you do one of these projects and write it up carefully, as you would a lab assignment, you may be able to claim extra lab credit: take your completed write-up to Dr Littlefair or Dr Cartwright for assessment.

list of projects

naked eye projects
Note: for most of these projects you need to be dark-adapted: spend 10 minutes or so in the dark (either outside or in an unlit room) before starting to observe. Red light does not destroy dark adaption, so use a red torch (tape a piece of red cellophane over an ordinary torch!) to do your drawings.
The Phases of the Moon (PHY115/106)
Observe the Moon as many times as possible during a lunar month, and make careful drawings of the shape of the illuminated portion, trying to ensure that you get the thickness of the crescent quantitatively correct. Work out the geometry of the Earth/Sun/Moon triangle and try to use this to draw a diagram showing the Moon's position in its orbit for each of your observations.
The Magnitude Limit (PHY104)
The Hellenistic Greek astronomer Hipparchos constructed a star catalogue which classified stars in six brightness levels or 'magnitudes' — the origin of the system we still use today. In principle, therefore, the typical naked eye can detect stars down to the sixth magnitude. However, street lighting was not widespread in the second century BC, and Hipparchos' skies were therefore much darker (and probably less polluted) than a typical night sky today. What is the apparent magnitude of the faintest star that you can see?
There are two methods of determining this number. They count as separate exercises.
Method 1: Star Charts
Locate the star Aldebaran (alpha Tauri). One way to do this is to find Orion and look for the first really bright star to the northwest of Orion: it's about as far from the northwest corner star of Orion as that star is from the southeast corner star. Aldebaran is one apex of a fairly conspicuous triangular group of stars forming part of the constellation Taurus (the Bull). Draw this triangular group, including all the stars you can see – the drawing should be as nearly accurate and to scale as you can make it. Then use a planetarium program to draw the same area of sky, showing stars down to magnitude +6.0 with the magnitudes shown on the plot (a simple program which does this is the online star-chart utility Your Sky by John Walker). Identify the stars that are shown on your drawing: your limiting magnitude is somewhere between the magnitude of the faintest stars you saw and the brightest ones you didn't see (for example, if you saw a star of magnitude 4.2, but failed to see one of magnitude 4.6, your limiting magnitude is 4.4 plus or minus 0.2). We choose the region round Aldebaran because it includes stars covering a useful range of magnitudes between 3.6 and 6.0; you might also try the bottom half of Orion itself.
Method 2: Star Counts
Obtain a hollow tube — an old bog roll or kitchen towel roll will do nicely — and measure its length, L, and the radius of the hole, R. Point the tube in a random direction (avoiding street lights!) and count the number of stars you see through it. Repeat at least a dozen times - the more the better - and calculate the average number n of stars seen through the tube. The solid angle subtended by the tube aperture is πr2 where r = R/L, the radius of the tube divided by its length. The solid angle of the whole sky is 4π, so the total number of stars in the whole sky that would be visible to you is N = 4n/r2. A lab demonstrator will supply you with a table showing the number of stars brighter than apparent magnitude m, for values of m between 1 and 9: use this to make a plot of log N against m and hence read off your limiting magnitude.
The Light Curve of Algol (PHY104)
Algol (beta Persei) is the classic eclipsing binary, with a deep eclipse every 69 hours as its hot class B primary passes almost directly behind its cool class G subgiant secondary. The eclipse is easily visible with the naked eye.

Use the online calculator in this Sky and Telescope article to find the times of the next few minima. Check with a planetarium program like Your Sky that Algol is going to be visible from Sheffield on the relevant dates. The eclipse lasts about 10 hours from start to finish, so even eclipses that reach their minimum during daylight may be worth observing to detect the changing brightness as the primary star comes out from behind the secondary. Use nearby stars of comparable brightness as references for your brightness measurement: gamma Andromedae, west of Algol, is about the same brightness as Algol usually is (magnitude 2.1), epsilon Persei, east of Algol, is magnitude 2.9, and rho Persei just south of Algol is about as bright as Algol at absolute minimum (magnitude 3.4). If weather permits, try to obtain the whole eclipse light curve by observing several eclipses at different stages. This article gives detailed advice on how to construct the light curve, either visually or photographically.

The Position of Sunset (or Sunrise) (PHY115)
Find an observing location with a reasonably clear view of the western (or eastern) horizon. A horizon with some helpful landmarks that you can locate on a map would be useful. Over a period of several weeks, note the time and the position of the Sun on the horizon as it sets (or rises). If you have a camera, take photographs; otherwise, make careful drawings. A reference photograph series or panoramic drawing showing the whole of your horizon with its landmarks is a good idea.

Use a compass, or a detailed map showing your landmarks, to calibrate your horizon so that you can work out the absolute directions corresponding to each of your sunsets (or sunrises). Make a plot of direction against date, and also a plot of sunset time against date. Compare your observations with calculations of the sunset/sunrise time (use the Astronomical Almanac or a planetarium program to find the RA and declination of the Sun on the relevant dates).

binocular projects
Note that it is, at least in my experience, very difficult to hold binoculars steady when looking at stars. You will almost certainly want to stabilise them somehow: use a tripod if you have one, lean your elbows on a handy fence, lie back in a deckchair or lounger — use your ingenuity!
The Galilean satellites of Jupiter (PHY104/106)
The four Galilean satellites of Jupiter are fairly easy to see with decent binoculars (I can do it with my little 10x24s, despite less than perfect eyesight, and my big 12x50s show them very clearly). Draw sketches of the Jovian system, showing all the moons that you can see (one or more may be invisible against or behind Jupiter itself), with the distances and orientations as nearly to scale as possible. Record the relative brightnesses of the four moons. (You may see more than four small objects close to Jupiter, in which case the extra ones will be background stars — don't worry about these, as Jupiter's motion will ensure that a star which is close to Jupiter on one night will be out of the field of view in subsequent observations.) Repeat the observations as many times as practicable over the course of a couple of weeks. Then redraw your diagrams in the form of a graph of moon position against time, similar to the sinusoidal plots found in, for example, Sky and Telescope magazine. See if you can work out which moon is which on the basis of your series of diagrams. If you can't, try comparing your drawings with images from TheSky or a similar planetarium program.
The Phases of Venus (PHY106)
One of Galileo's other great discoveries was that Venus shows phases like the Moon. This helped to demonstrate that the Earth-centred solar system of Ptolemy was inferior to the Sun-centred model of Copernicus (and Aristarchos of Samos). It should be possible to see these phases using a good pair of binoculars: alternatively, this project can be done using a telescope.

Observe Venus several times over a period of several weeks (the orbital period of Venus is 224 days, but a couple of months is long enough to see a real change). Draw the image of Venus, paying attention to shape, size and orientation. How does the size of the image change as the phase changes, and why? Use trigonometry to work out the distance of Venus at each of your observations.

To give you an idea of what you might see, here is an animation of telescopic images of Venus made by students at Calvin Observatory (probably using a 16-inch telescope, although this is not quite clear from the website).

Mapping the Moon (PHY106)
On a night with full or nearly full Moon, make a careful drawing of the Moon's surface features. Include as much detail as you can. (Note: if you are interested in photography, you can also try doing this practical by taking a photograph of the Moon either with a conventional camera (you'll need a long lens, and very fast film) or with a digital camera using a combination of optical and digital zoom.) Compare with a map of the Moon and identify the features you have drawn. How accurate were you? What is your resolution (i.e. how big are the smallest features you could consistently draw)?

Repeat this procedure at least twice, and look for variations in the positions of the features, caused by slight changes in the orientation of the Moon.

Deep Sky Objects (PHY104)
How far can you see with binoculars? Ben Crowell's Binosky is a catalogue of interesting deep-sky objects to look for. In each case, make a careful drawing, and describe the visual appearance of the object in your log. Include colour information if you detect any colours. If you are viewing an open cluster, count the number of bright stars.

Binary stars are also good binocular objects. Luis Arguelles has a list of binocular doubles, in many cases with comments from observers. You should draw the configuration of the pair and note any colour that you see.


Hicks Building, Hounsfield Road, Sheffield S3 7RH, UK