Pages for this activity: General directions, surface processes experiment, moons, impact experiment.

Overview for teachers (student directions are below)

Teacher classroom instructions: Use identical bins of flour (same amount of flour as well as the same container type) for the student setups. Have on hand multiple samples of each kind of impactor. I found bags of wooden balls of assorted sizes in a craft store and marked the sizes on the balls with paint pens, added marbles and styrofoam balls of various sizes, and even collected an assortment of steel balls from a science store. I have not tried classroom use of plastic eggs with fishing weights because the weights are lead - but I hope to coat them and then try (I'd stuff the eggs with cotton to avoid bouncing). Before the students arrive, set up a bin and make 3-4 craters using identical parameters (object and height) for the set. Do multiple craters because there will be some variation -- it's important that students understand this is part of science. Be prepared to tell the students whether the surface is fluffy or packed, but don't tell them until they need it. Do not tell them the impactor or height, ever, if you want to simulate real-world science - most craters on most surfaces were made long before humans were around, so we never get an "answer" in that sense, only confidence. Insist that students work in their separate groups before compiling their results, if necessary, and help them compile their results after the experiment, if necessary. Retain one bin and the impactors for later experimentation.

One good strategy is to have each student (or possibly each team) examine the effects of one variable in detail. Pose the questions: "How does the height dropped affect the craters?"; "How does impactor mass affect the craters?"; "How does impactor size affect the craters?"; "How does firmness of the flour affect the craters?" etc. See the list at the end of the text (before the images) for more variables and some things to look for or measure about the craters.

Encourage your students to organize tables logically; it makes a big difference in how they see the results. (You may wish to set up standard formats so different teams can compare their results more easily - or, if you're willing to take the time, you may let the students find for themselves how important it is to use comparable formats.) There are many parameters, and you may wish your students to make graphs. Help them choose appropriate kinds of graphs; bar graphs are generally a poor choice. For most of the experiment they should be looking at two parameters at once - an independent (x) and dependent (y) variable. The results are typically not straight lines, but that's okay. There may be too much variation to see the pattern right away - scientists often use "error bars" to show how much variation there might be, or even color in areas if there are more possibilities than we can record on a graph. Science is messy, and data and graphs will rarely be neat mathematical functions.

Teacher solo learning experiment: You need two identical bins and a range of impactors. Do this part after you've done your general experimentation - that way, the objects already have some flour on them. Assemble the range of objects. Ask someone to assist you. Show them how to make craters and how to recover the object with minimal disturbance to the crater. If necessary, have them push it to one side before removing, so that you have 3/4 of the crater relatively undisturbed. Ask them to choose an object and make 3-4 craters by dropping the object from one measured height. Ask them to stay within reasonable ranges and not do anything intentionally odd; the task can be surprisingly difficult even with ordinary parameters. Have them record the height and the object and seal the results in an envelope. Try to figure out the details as below. Set ranges on your parameters.

It's useful to acquire some good experimental habits if you haven't yet. Use tables to record your systematic results (not the initial "what makes a good crater" but the fine-tuning). Include a column for comments, and references to diagrams if they don't fit in the table. You'll reorganize the tables later when you look at the results - resign yourself to doing it twice. If you're trying to match the craters, or if something changes dramatically (examples follow), vary the parameters in large steps at first, then in smaller steps in the range that counts, then maybe smaller steps yet. Alternatively, go back and forth, halving or quartering the range each time, but only for the important parts.

Example 1: The samples' ejecta blankets are 5.5 cm across, with about 0.3 cm variation either way. I drop a ball from about a foot above the surface and get a reasonable approximation of the sample crater. I try 10, 20, 30, and 40 cm drops and get ejecta blankets of 5.0, 5.5, 5.4 and 6.2 cm. I need to look more closely around 20-30 cm, but also slightly outside that range. I try 15, 25, and 35 cm, and get ejecta of 5.2, 5.6, and 5.9 cm. It's looking like the range could be 15-30 cm, so I try 1-cm changes around those two numbers: 13, 14, 15, 16 cm, (results: 5.0, 5.4, 5.3, 5.5 cm) and 29, 30, 31, and 32 cm (results:5.3, 5.7, 5.6, 5.8 cm). I've reached the limit of my experimental ability - the variation due to chance is greater than the variation due to my controlled variable (height of drop). I might decide to repeat the last set of tests several times to get a better measure of the variation, and I might go an extra cm or two in either direction, but for most purposes I could reasonably stop here.

Example 2: I drop a styrofoam ball from a few inches, waist height, and reaching-up height. I get crater bowl diameters of approximately 1 cm, 1-2 cm, and 4 cm. The first two are just dents in the surface, the third is close to the size of the ball and the crater looks more like craters from other impactors. What makes the difference? I make some more carefully measured drops, starting at 40 cm from the ground and going up in 20-cm increments, then look at my results and decide that it looks like the crater bowl grows with height up to maybe 120 cm, then starts looking like a normal crater. I decide to go back and measure 20 cm from the ground (I'm going to have to subtract the height of the flour or mark it on the graph, later) to get a longer line. Then I do 10-cm changes: 110 cm, 120 cm, 130 cm - and get that the change is between 110 cm and 120 cm. Then I do 115 cm, and it looks like the change is lower. I might try 112 or 113 cm, but the crater is getting harder to measure, so I decide that a 5-cm range is close enough - somewhere btween 110 cm and 115 cm, the foam ball stops just denting the surface and starts to make a full crater.

Below are the key parts of a handout I've given to college students. Students are essentially designing their own experiments, so I would prefer to keep the information minimal. However, my college students needed more in the handout for reference when they wrote up the experiment later. On the next class day, I normally collected each group's estimate of (a) the sample crater measurements, and (b) the parameters determined. They typically agreed on the measurements but not the parameters used to match the samples, and this often surprised them. I think they're not used to a large number of variables and the idea that several very different combinations can give similar results, but science is messy like that.

Experimenting With Impacts:

Use systematic experimentation (vary just one thing at a time, and take measurements as well as qualitative notes) to figure out how a sample crater was produced. (Object used - size, mass, type; height from which it was dropped, and any other significant aspects.) Some notes or limits to the possibilities will be on the board. You may NOT use destructive measurements or testing of the sample craters unless you have permission of the rest of the class! Be considerate.

  1. Do some general experimentation to rule out what you can. For this step, make approximations rather than careful measurements. Remember, however, that sometimes very different situations might produce very similar results - a foam ball dropped from the ceiling into a fluffy surface might give a crater that looks like a the results of a metal ball dropped from barely above a packed surface. How will you tell the differences? You may wish to note that the energy of the impacts you're making can be calculated: E = mgh, where m=mass of impactor and h=height above surface. The other factor, g, is a constant for the surface of a world, and depends on the units. This is the relative potential energy of the impactor before you drop it, which, if you remember, is equal to its kinetic energy right before it hits the surface.)
  2. Within the limits you've determined, experiment systematically. Choose the most likely combination of parameters, and change one at a time, through several possible values. (Height is the most obvious and easiest to vary, for a start.) You may find a rough plot (graph) will help you narrow down the possibilities.
  3. How careful do you need to be? How much variation do you get from random chance? Try doing the same crater 5 or 6 times (same object, height, etc.) and measure the resulting craters carefully. How different are they? How do the differences you get just from different trials compare to the differences you get from varying something (such as height or object, or a re-made surface)? *See below for an example.
  4. When you have one or a few reasonable combinations of impactor and height, fine-tune your measurements. Start with one impactor and the height you think is reasonable to simulate the sample. Drop an impactor from this height, and then from greater and greater heights, measuring each time, until the crater is different from the sample craters. Then go back and check a height between the highest one that worked and the first one that didn't. Repeat this, so you have more measurements where it counts most. Start again at the initial height, and drop from lesser and lesser heights until the crater is different from the sample craters. If you have more than one possible combination, repeat this proceedure for the other combinations.
  5. Organize your information in graphs and tables. When you make tables, list the things you vary on purpose first, and the results you measure further on. In general , you should have a column for comments. Arrange the table so the things you vary are in a sensible order - it's much easier to see the patterns than if you list them in the order you actually did them. It's also easier to see patterns if you use quote marks to indicate that a parameter is unchanged, and if you're careful to list everything in the same format.
  6. When you think you've figured out how the sample craters were made, compare your analysis to what other groups have determined. How similar are they? Did you all try the same things, or different? Go back to re-check anything that needs more attention in your experiments.

* Example: If you make 4 identical trials (same object, same height) and get very similar results - say, the ejecta blanket is 6.0-6.5 cm in diameter in all cases, and the rims, rays, and depths are similarly consistent - then how much can you can vary the height before you get DIFFERENT results? If you later drop the same object from half the height, and you get an ejecta blanket that's 4.0 cm in diameter, you're doing okay. If it's 6.0 or even 5.5 cm in diameter, then that's just the sort of variation you were getting just from different trials - 0.5 cm in this example is not enough difference in the ejecta diameter to be sure the difference is really from the varying height, and not just from random chance. What can you do to make sure the differences you measure are really due to the changes you're making, and not just from random chance?

Things to consider for the craters:

 

A useful crater site: http://cass.jsc.nasa.gov/expmoon/science/craterstructure.html

Some more details about craters:

Parts of a crater:

Small, medium, and large craters, respectively,based on shape. If all else is equivalent, small craters are generally bowl-shaped, medium craters start to have flat bottoms, and large craters may have a central peak - see also the cross-sections, below, for simple and complex craters. Very large craters might also become filled with oozing lava or may affect the appearance of the world as a whole (asteroids and small moons may have significant dents or other effects).

Anatomy of craters: