JANUARY/FEBRUARY 2020. DISCOVER 67i
Southern California is no
stranger to earthquakes,
but research published last
summer shows it’s home to far more
than previously thought.
Seismologists at Caltech and
Los Alamos National Laboratory
identified 1.81 million tiny tremors
hidden in data from 2008 to 2017 —
roughly one every three minutes.
This newly detected seismic activity
has already helped scientists take
steps toward fully understanding
how earthquakes start.
In the May study, published in
Science, researchers unearthed
these rumblings in a new catalog
of local earthquakes. They used a
technique called template matching,
which requires supercomputers to
comb through the shake-induced
background noise of helicopters,
construction and Los Angeles
traffic. From that sea of squiggles,
the machines can pluck out
vibrations with the same shape as
larger quakes.
This comprehensive compilation
helps resolve a seeming discrepancybetween laboratory models and
what actually happens in Earth’s
crust. During small-scale simulations
with miniature rocks, the primary
earthquake, or mainshock, always
follows smaller shakes, called
foreshocks. In real life, previous
evidence from seismographs showed
foreshocks precede mainshocks only
10 to 50 percent of the time.
But in a July Geophysical Research
Letters paper, researchers using
this earthquake catalog found faint
foreshocks before 72 percent of the
actual mainshocks they examined.
This catalog, and the subsequent
foreshock information, won’t
let researchers predict the next
massive earthquake, but it will help
additional analyses further clarify
their causes. Detecting regional
ramping-up patterns gives clues
about what’s happening miles below
Earth’s surface, beyond the reach of
current tools.
And the more we understand
what’s happening down there, the
better we can predict what might
happen on the surface.Unearthing
Quakes
34 BY SARAH WHITEWeinstein says he hopes his technique,
published in June in the journal Cell, will
become part of a wave of new approaches
that move away from averaging genetic
activity across whole organisms — like in
those blenderized fish — and toward a more
cell-specific view. “A lot of the stuff we need
to pay attention to, to truly understand these
systems,” he says, “needs to be learned by new
methods — new ways of probing the inner
workings of cells.”
One of the advantages of the DNA micro-
scope, Weinstein says, is that it only requires
lab pipettes and a standard DNA sequencer
to yield results.
But Lee and Crocker say it may not be
as cheap and easy as Weinstein thinks. So
far, Weinstein and his coauthors have only
shown that their idea is theoretically possible.
Applying it to larger-scale projects may prove
more challenging, Lee says. “Their paper cap-
tures the imagination. It’s a completely new
way of doing microscopy,” Lee says. “It’s up to
the scientists to decide whether it’s practical.”
And while it might be relatively inex-
pensive for Weinstein to sequence DNA at
the major research institutions where he’s
worked, Crocker says, “the computational
firepower you need would be pretty exten-
sive,” and not easily available, for example,
at hospitals that wanted information on
patients’ tumors.
This fall, Weinstein launched his own lab
at the University of Chicago. Prior to that, he
worked at the Broad Institute and MIT in the
labs of Aviv Regev, who is leading an effort to
map all the types of cells in the body, and of
Feng Zhang, who was among the discoverers
of the CRISPR gene editing tool. Regev and
Zhang are the other two authors on the DNA
microscope paper.
Despite making advances in the field of
biology, Weinstein’s background is in physics.
He still tries to understand the world by boil-
ing it down to its simplest parts, as a physicist
would. Biology, he says, can’t be broken down
as easily or understood as intuitively — but
that’s part of its appeal. “It’s such a great mag-
net for physicists who like being puzzled,” he
says. “[It’s] kind of the perfect kindling for
SEI crazy ideas.”
SM
OG
RA
PH
:^ AN
DR
EY^
VP
/SH
UT
TER
ST
OC
K.^
MA
P:^
TO
NE
LLO
PH
OT
OG
RA
PH
Y/S
HU
TT
ER
STO
CK