Tides trigger tremors deep inside the Earth*

David Perlman, Chronicle Science Editor

Friday, November 23, 2007

A new mystery faces scientists who study stresses that trigger
earthquakes both large or small: Even the rise and fall of the ocean's
tides are strong enough to trigger pulses of underground tremors that
periodically send seismic faults slowly slipping beneath the northwest
coast, quake researchers have found.

The scientists, led by a young researcher who will soon join the U.S.
Geological Survey in Menlo Park, came upon their puzzling discovery
after installing highly sensitive seismic detectors on the Olympic
Peninsula in Washington and across Vancouver Island in British Columbia.
Readings from the detectors revealed that twice a day when it's high
tide, the strength of the faint tremors increases, and twice a day at
low tide, those underground tremors grow fainter.

And because it's the moon's gravity, as well as the sun's, that causes
the tides to rise and fall, those heavenly bodies are, in fact, exerting
their legendary influence from afar to alter seismic activity deep
beneath the Earth's surface.

But no one knows how or why gravity should reach into the Earth's deep
interior, conceded geophysicist Justin Rubenstein, whose work and that
of his colleagues appeared in Thursday's online version of the journal
Science Express.

Tremors are like tiny earthquakes, but they can never be felt on the
surface. Seismologists have long known that when volcanoes are about to
erupt, the Earth shudders in tremors along cracks deep beneath the
volcanic mountains as the molten magma rises.

But six years ago, Berkeley seismologists performing in-depth earthquake
analysis along the San Andreas Fault near quake-prone Parkfield in
southern Monterey County recorded episodes of similar faint tremors far
beneath the Earth's surface.

No volcanoes were erupting anywhere near Parkfield, and no significant
earthquakes were striking, but the non-volcanic tremors - what
scientists term "slow slip" quakes or sometimes just "chatter" along the
fault - were detected as deep as 15 miles down, well below the level
where true earthquakes have ruptured the ground in that area.

Now tremors like these near the ocean have been linked to tidal changes
for the first time.

Other Earth scientists said they were impressed with the findings of the
team of scientists in Seattle headed by Rubenstein, a research associate
at the University of Washington who is set to join the U.S. Geological
Survey in March.

"The discovery gives us a new window into parts of the Earth that we
haven't seen before, and into events that occur at the deep roots of
earthquake faults. Now we're all thinking about their possible causes,"
said William Ellsworth of the Geological Survey who is one of a dozen
experts on the agency's advisory panel on earthquake prediction,
forecasting and hazard assessment.

"This is an exciting and extremely important development," said
Ellsworth, who was not a part of the study.

Along the Pacific coast from Northern California below Eureka and up
into British Columbia and beyond - a region called the Cascadia
Subduction Zone - these flurries of non-volcanic tremors seem to occur
deep underground roughly every 14 months, and typically last as long as
two or three weeks, according to Rubenstein.

Non-volcanic tremors that are somewhat similar, and possibly also linked
to tidal action, have been reported by Japanese earthquake researchers.

And while a single tremor in Rubenstein's study may be recorded as tiny,
three weeks of them in succession can release as much energy as a large
earthquake, Rubenstein said in an interview.

To observe the tiny changes in the rates and strength of those tremors,
Rubenstein and his colleagues observed the signals from their seismic
monitoring instruments and found three sets of events where the strength
of the tremors either increased or decreased every 12.4 hours, and again
every 24 to 25 hours.

The faint signals from the pulses were hard to analyze, but they were
detected and recorded twice a day in July 2004, again in September 2005,
and once more in January. They all occurred just when the tide was
either high or low.

"What was just as puzzling," Rubenstein said, "is that although we
expected that the weight of the water at high tide would be more likely
to damp down on the tremors and weaken them, instead the tremors were
stronger at high tide and weaker at low tide.

"It's still a big mystery, and the results from our instruments are
counter to what we might have expected. But my guess -although I don't
necessarily believe it - is that a change in the weight of only two or
three meters of water at high and low tide alters what presses down on
the Earth and causes detectable pulses in the periodic tremors that
occur in very weak unknown faults."