Tales from the Ghost Forests

Ghost forest on Washington state's Copalis River during a very high fair-weather tide in December 1997. Photo courtesy of the USGS public domain

black and white image of dead upright trees and water
Ghost forest on Washington state's Copalis River during a very high fair-weather tide in December 1997. Photo courtesy of the USGS public domain

If a megathrust earthquake struck the modern world, most of us would know about it within seconds. Bill Steele, Director of Information Services at the Pacific Northwest Seismic Network, would get a detailed account from an extensive array of seismometers located around the globe. Regular citizens would probably get an email, tweet, or Facebook note from a friend.

But what if the quake hit hundreds of years ago? How were scientists able to determine that a Cascadia Subduction Zone megathrust trembler hit sometime between 9:00 and 10:00 p.m. on January 26, 1700?

The story starts in the 1980s when geologist Brian Atwater, dendrochronologist David Yamaguchi, and others began to investigate the Pacific coast of Oregon and Washington. Traveling by foot and boat through the region’s many bays and river mouths, the scientists examined thousands of dead western red cedar and Sitka spruce. The trees had not fallen to the ground but stuck upright out of the sand in great groves, known as ghost forests.

Intriguingly, James Graham Cooper, a naturalist living on the Washington coast in 1853-1854, had noticed the dead trees, too. He thought they had died slowly, sinking into quicksand. Atwater, however, knew differently. When a magnitude 9.5 earthquake hit Chile in 1960, coastal marshes had dropped several feet, allowing sea water to flow in and quickly kill the trees.

Because the spruce and cedar still stood in place, Atwater and Yamaguchi realized they had a clock that would tell precisely when the trees died. All they had to do was look at the trees’ growth rings. Those rings would reveal how the trees had lived and when they had died.


Tree ring analysis, or dendrochonology, is based on the fact that trees add new material each year: early, lighter-colored cells followed by darker cells. Each set is known as a tree, annual, or growth ring. Width depends on the climatic conditions with wider rings in a good growing season.

Yamaguchi found that the trees had wide rings right up until the time they died. This indicated that disease had not killed the trees. He and other researchers then compared the rings of the dead trees with those living nearby. These trees spanned nearly a thousand years of time, from 993 to 1986. Using this data, they tried to match the growth rings of the living trees with the growth ring of the dead trees. A match would tell when the ghost forest trees died.

They had a problem, though. Erosion of the tree’s trunks had obliterated the final rings. Fortunately, in 1996, they dug up eight stumps whose roots had not decayed. In all but one case the rings showed the trees had died between August 1699 and May 1700. Narrowing the timing down to January 26 required another line of evidence, and the assistance of colleagues and documents from Japan.