Tree rings give two important clues about
the occurrence of a great Cascadia earthquake in the Pacific Northwest.
First, they provide evidence that one large quake, instead of many small
quakes, contributed to land-level changes along Washington's coast. Second,
by matching sequnces of tree ring widths of trees that died due to changes in land level
to tree ring widths of living trees, scientists can deduce that a great
Cascadia quake occurred between the growing seasons of 1699 and 1700.
Great
earthquakes are known to cause changes in land levels.
Subsidence, the downdrop of coastal land, typically leaves long-lasting evidence.
When coastal land sinks below sea-level it kills plants and trees that cannot
grow in salty water.
To test for coeval (at the same time) subsidence, which would indicate simultaneous
lowering of land levels in a single
event, nine Sitka spruce (Pica sitchensis) stumps were unearthed
from the uppermost soil layer in two locations, Willapa Bay and the Copalis
river, 55 kilometers apart. If the stumps found at the two separate locations died
at the same time, it is likely the same event caused their demise.
If the dates differ, the trees likely died from separate events.
High-precision radiocarbon dating
was used to date the tree stumps. Although this method of dating is accurate only to
within 20 carbon-14 years, all the samples from each location
were found to be consistent with tree-killing subsidence falling between
1680 and 1720. This date range is very small in geological terms. Although the samples
covered only 55 kilometers of coast, the results
support a single large Cascadia earthquake occurring sometime between
1680 and 1720.
Another dating technique is based on the study of patterns of thick and thin tree-rings.
Good growing seasons produce more growth and thicker rings, while thin rings occur in less favorable seasons.
Spruce roots, used for the carbon-14 dating, lack long series of tabular rings, so dating Spruce roots by ring-width patterns was not possible.
In 1997 scientists' eyes turned to the Western red cedar (Thuja plicata)
for more accurate dating. Tree rings taken from the Western red cedar,
in contrast to the Sitka spruce, are tabular. This means their rings
provide useful data in determining their time of death. To date the
occurrence of tree-killing subsidence, 19 living upland trees with consistent
ring patterns were sampled. The composite ring-width pattern of these
19 trees spanned from 993 to 1986. Next, trunk rings of 75 snags from
the Copalis River, Grays Harbor, Willapa Bay, and the Columbia River,
along 90 kilometers of coast, were correlated with the composite ring-width
pattern of the 19 living trees. From this comparison, tree-ring sequences from
the dead snags
are matched to sequences within the lifespan of trees of a known age.
Most of the ghost trees were shown to have died sometime after the 1680's, though the
the outer rings of the snags had been eroded away due to exposure, and an exact year
of death could not be determined.
After finding Japanese historical records of a 1700 tsunami that arrived from a distant source, a further attempt was made to precisely date the ghost trees. Bark-covered roots
complete to the outer rings were recovered from the correlated snags. The outermost growth ring of the roots showed that the trees
died after the 1699 growing season and before the growing season of
1700.
Printable
Versions | References
