July 22, 2019
In 1972, after my freshman year at MIT, Prof. Nafi Toksoz was kind enough to hire me to work in his research group that used data from seismometers the Apollo astronauts installed on the moon. I learned a lot from these excellent scientists, and (hopefully) helped them a little. It was exciting to have even a very minor part in a group investigating the moon’s structure and evolution. I figure I was at the bottom of the roughly 400,000 people involved in the Apollo program. This started my science career – I coauthored papers about detecting meteoroid impacts on the moon and about the moon’s internal structure. It also helped my newly forming outdoor interest – the money I made paid for my first backpack, tent, sleeping bag, snowshoes, ice axe, etc. and one of the then-new H-35 scientific calculators, which cost $400 ($2500 in today’s dollars)!
The program was so complicated that the paperwork was legendary. For example, our research group had to send NASA monthly, as compared to NSF’s required annual, reports. There was a joke that all the paperwork would be piled up and someone would step from the top onto the moon. My group was part of the “Passive Seismic Experiment” (PSE), to study the deep structure of the moon. There was also the “Active Seismic Experiment” (ASE) that used a mortar to fire shells a short distance from the landing site, so as to study near-surface structure. Rumor had it that this experiment was planned for Apollo 11, but NASA HQ decided that firing a mortar didn’t fit the spirit of “we came in peace for all mankind.” Hence the ASE was put off to later landings.
The PSE used seismometers in the instrument package installed at each landing site. The Apollo 11 package died because of a solar panel failure, so later missions used a radiothermal generator that converted heat released by the decay of radioactive material into electricity. We had data from seismometers at the Apollo 12, 14, 15, 16, and 17 sites (Apollo 13 didn’t make it to the moon). Our sources were moonquakes (produced by tidal stresses), meteoroid impacts (the moon has no atmosphere to burn them up, so many hit the surface), and the impacts of the last stage of the Saturn V rocket (which gave us sources with known position and impact time).
One of the biggest challenges was that because the moon is so dry, near-surface rocks scatter lots of seismic energy. Hence each arrival is followed by an intensely scattered wave train, and no later arrivals can be identified. There’s an example online at http://levee.wustl.edu/seismology/book/chapter3/chap3_sr/3_7_10_s.jpg
One of my tasks was trying to identify arrivals in the scattered wave train. I tried many methods, none of which worked. The positive side was it got me deep into the then-new field of digital signal processing, which I really liked. Hence anyone who’s been willing to face the math in EARTH 327 (Geophysical Time Series Analysis) benefited from my failure.
The computing available to us was primitive by today’s standards. We typed programs onto paper punch cards, one line per card, using a machine like a big typewriter. We took boxes of cards to the computer center, and gave them to operators who lined them up to be fed to room-sized computers with far less computing power than an iPhone. Since the computers couldn’t store the seismic data, the programs directed the operators to mount big magnetic tapes, about the size of frisbees. If all went well, the programs produced seismograms drawn by plotters that moved an ink pen over a moving paper roll. Typically, when we came back a few hours later to see if anything had worked, there had been an error. We found the error, retyped the offending card, and repeated the process.
Despite all this, the group developed a model of the moon’s internal structure
It will be interesting to see how that looks when (eventually) more modern seismic technology and analysis methods are applied. Hopefully some of today’s students will get that chance!
Seth Stein, William Deering Professor, Department of Earth & Planetary Science and Institute for Policy Research, Northwestern University