Unreel science

Thinking of bad geology movies reminded the Grumpy Geophysicist of how little most folks know about how science is conducted in the field. (In fact, a fair number of scientists have no idea how field work is done.)  So when the Seismo Lab truck rolls up to a spot to be investigated in Volcano and the scientist opens up the sides to reveal a perfectly arranged, fully equipped field support vehicle, GG laughed.  Why? Because the trucks we use usually look as though they were packed by chimps on meth.  Tools are in old army ammo cans or cheap plastic tubs or maybe under the seats; a few empty soda cans are usually kicking around somewhere and the dust is more than what is desirable.

Field truck, Death Valley National Park, 1994. Working on some equipment.

Field truck, Death Valley National Park, 1994. Working on some equipment.

The funny thing is that real field work can be more amusing than the brilliantly orchestrated, hyper-tech vision that shows up in movies from time to time (less so these days; the sort of crappy looking Alien and Star Wars future tech has migrated back to the modern-day tech vision).  So GG would like to share just how he and some colleagues managed to get seismic records from the backcountry of the Sierra.  While the safety of millions was not on the line, this at least is all true…

First, the experiment in 1988 was not a well-funded one.  Hiroo Kanamori redirected some USGS funds to help cover some costs, but this was low budget.  At the time, Caltech had little in the way of portable seismic equipment (the mantra at the time was that field seismology was dead–which turned out to be very wrong), so equipment had to be borrowed.  Most notably, six “LBS” seismometers were borrowed from Kei Aki at USC.  Unlike the movies where really nice looking stuff with state-of-the-art interfaces was used, this equipment looked like it was built in a garage from spare parts (see photo below). These were prototypes made years before at MIT that had never been farther than 100m from a road; we decided to take them into the backcountry.  This required a new power system; fortunately some of the techs at the Seismo Lab had an idea of how to do this and they devised a solar power solution (in 1988, this was a novelty; today it is SOP) that was complicated by the fact that the seismometers needed two independent power sources. The system was tested once and the techs realized that the battery would drain through the solar panel at night, so they revised the power system a couple days before we left to install the equipment.  It wasn’t fully checked out before we left…

LBS seismometer and Tom Fairbanks

Tom Fairbanks wrestles with an “LBS” recorder near Lewis Camp in Sequoia National Park in 1988.

 That was the highly organized part.  How do you get this stuff into the backcountry? In 1988, the Park Service chose mules as the preferred means of transport.  (They also chose mules for a 1993 and 2005 deployment but allowed helicopters in 1997.  Go figure). This is trickier than you might think as there were also lead acid RV batteries going in (we had to get ones not yet filled with acid–we couldn’t afford the sealed batteries used now). This is what it looks like shipping in equipment on mules:

Mules with seismic gear decide to back off the bridge near the Kern Hot Springs, Sequoia National Park, 1988.

Mules with seismic gear decide to back off the bridge near the Kern Hot Springs, Sequoia National Park, 1988.

(the cloud of dust is because the whole line of animals decided to go backwards rather than over the bridge despite the cursing and dragging of the packer).

And yet, we still haven’t reached the truly makeshift part of this operation. A key element of the experiment was to get the arrival time of P-waves from earthquakes around the world–but to do that, we had to have the time on the seismograph within about 10 milliseconds. In previous experiments, the scientists had a field clock that they would synchronize to a radio signal or reference clock in the lab and then drive it around to the seismographs to get them synced.  This was not an option for us, and the interesting thing is that we didn’t really solve the problem until we sat around the first station being installed. How could we do this?

This was brainstorming 101, and if we failed, the experiment would not get off the ground.  One participant (Steve Roecker of RPI, who generously gave up 3 weeks of his time to help with this) had us bring into the backcountry an older analog recorder–a drum recorder with a small sensor that he intended we use to pick the best spot for our sensor (we never used it that way).  This instrument happened to have a WWVB radio receiver (WWVB radio is a binary time signal still in use to set clocks) that required nine volts to be powered.  GG had convinced the Caltech techs to buy a new, portable  digital oscilloscope, which he plucked from their eager hands a day after it arrived. Tom Fairbanks (one of exactly two people on Earth at the time who really understood the LBS instrument) knew that the clock inside the LBS could be monitored from outside the instrument.  On the spot we realized that we could take the output from the WWVB radio, the output from the LBS clock, and connect both into the oscilloscope to compare the signals.  Even better, since this was a new digital oscilloscope, we could freeze the image and move cursors around to measure the exact difference in time between the internal clock and the radio signal. Voila! We had time corrections.  And as these older instruments had some of the most rock-solid clocks ever put into a field instrument, the fact that we could only visit them every few weeks was not a problem.

It was as though we had reversed Murphy’s Law. Even the power systems worked as designed.

There were other misadventures (a Caltech graduate, for instance, connected black to red and red to black wires, frying a fuse on the oscilloscope not once but twice), but for shear improbability, this took the cake.  In the field, after hiking about 40 miles [from the south near Quaking Aspen to the Hot Springs and back south, if some of you know the area], no real plan, and yet there was just enough stuff and skill to make it fly.  It was entirely Rube Goldberg (even the battery was a hack at the time; we replaced our goofy battery setup with a real 9v battery for later visits), but it worked.  And that isn’t as uncommon as you might think.  Even with fairly modern looking gear being shipped by IRIS to scientists, you sometimes have to have some imagination to do things (in 2007, the pins on a connector to a seismometer sheared off as we were removing the instrument; to lock the seismometer for shipment so it wouldn’t be ruined, we had to figure out a way to apply the proper voltage to two pins to get the instrument to lock, which involved some cursing and some jury-rigged wiring).

The usual reality is that scientific field gear is usually stuff that has been handmade with technology that isn’t too expensive (so while the gear might do things other stuff can’t do, it often is made in a far more clumsy way from seemingly outdated pieces).

[Maybe it is unfair to criticize Hollywood on this too much; for instance, Spock had to build a jury-rigged system to access his tricorder in Star trek’s City on the Edge of Forever.  But usually the equipment functions fine and any innovation is in interpreting what is being seen].

(Updated late 6/4 with the photo of the truck full of gear)

Advertisements

Tags: , ,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: