ML20151H047
| ML20151H047 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 04/12/1988 |
| From: | Atwater B INTERIOR, DEPT. OF, GEOLOGICAL SURVEY |
| To: | |
| Shared Package | |
| ML20151H012 | List:
|
| References | |
| NUDOCS 8808010118 | |
| Download: ML20151H047 (7) | |
Text
?
\\
PHOBABLE LOCAL PRECEDENT FOR EARTHQUAKES OF MAGNITUDE 8 OR 9 IN THE PACIFIC NORTHWEST by Brian F. Atwater U.S. Geological Survey at Department of Geological Sciences University of Washington AJ-20 Seattle, Washington 98105 Great earthquakes probably can happen in the Pacific Northwest.
Such earthquakes, being of magnitude 8 or 9, would release as least as much energy as did the 1906 San Francisco earthquake.
Their source would be i
the plate-bounding fault that descends gently eastward beneath the continental margin from southern British Columbia to northern California.
This huge fault, the Cascadia subduction zone (fig. la), is not known to have produced great earthquakes in the 200 years since white people arrived in the Pacific Northwest.
But Cascadia has much in common with subduction zones elsewhere on which great earthquakes have occurred historically (Heaton and Hartzell, 1987). Moreover, as reviewed in this report, great earthquakes seem to have occurred on the Cascadia subduction zone itself--
at least twice in the past 1700 years.
COASTAL EVIDENCE FOR THE PAST OCCURRENCE OF GREAT CASCADIA EARTHQUAKES If great earthquakes have occurred on the Cascadia subduction zone during the past 1700 years, evidence of the earthquakes should abound on the Northwest coast.
This is chiefly because a great subduction-zone
(
earthqueke usually causes the adjoining coast to undergo meters of uplift or subsidence.
The uplift can result in the permanent emergence of wave-cut coastal benches; the subsidence can cause the estuarine burial of well-vegetated coastal lowlands that drop to the level of tideflats.
In addition, coastal lowlands may preserve anomalous bodies of sand that result from shaking during the earthquake and from the tsunami that comes ashore minutes later.
L Earth scientists have barely begun to ask whether all these great-earthquake te11 tales are present on the Northwest coast.
But in just the past two years they have found much evidence for rapid subsidence, some evidence for consequent tsunamis, and a little evidence for uplift and shaking.
Subsidence.
That great Cascadia earthquakes probably have occurred is indicated chief ly by evidence of sudden coastal subsidence.
This evidence takes the form of marshes that have been buried by tidal mud.
First recognized in the Northwest in 1986, such buried marshes are now known to range in location from southernmost British Columbia to northern California (Rogers, 1988: Atwater, 1987, 1988; Grant and McLaren, 1987; Darienzo and 4
Peterson, 1987; Nelson, 1987; G.A. Carver, oral commun., 1988). Simple tests can eliminate storms, floods, far-traveled tsunamis, differential compaction, and global sea-level rise as alternative explanations for the marshland burial ( Atwater,1987, p. 943).
i l
f 1
l 4
i I
8308010118 880615 PDR ADOCK 05000344 j
P PDC I
f
f At least in southwestern Washington, Jerky coastal subsidence probably had too much areal extent to be explained by anything other than great Cascadia earthquakes.
A great Cascadia earthquake is likely to entail subsidence of a coastal strip at least 100 km long and tens of kilometers wide (Atwater, 1987).
Subsidence on that scale is suggested by regionality in the sequence and radiocarbon age of buried marshland soils in southwestern Washington (figs. Ib, 2).
This regionality implies that at least two Jerks of subsidence--one about 300 years agot, the other about 1700 years ago--involved a coastal strip no less than 85 km long and no less than 30 km wide.
It is remotely possible that each correlated soil in figure 2 represents a series of moderate earthquakes that successively Jerked adjoining areas during an interval too brief to dissect by conventional radiocarbon dating.
This possibility is now being tested by the tree-ring dating of cedars that died from sudden subsidence into the intertidul zone about 300 years ago (D.K. Yamaguchi, written commun., 1988).
Tsunami.
Tsunamis probably resulted from at least some of the events that jerked the coast downward in Washington and Oregon.
The evidence for tsunamis consists of sand that locally veneers some of the Nurted coastal marshes (Atwater, 1987; Reinhart and Bourgeois, 1987; Grant and McLaren, 1987). This sand is typically coarser than other intertidal deposits in the vicinity.
Landward thinning of the sand indicates deposition by surge from a bay or from the sea.
About 300 years ago in southwestern Washington, the sand from such a surge entombed the rooted stems and leaves
(
of grass that had been living on the marsh when the marsh was Jerked downward.
This relation indicates that the surge took place within a few years of the Jerk.
Such coincidence, though unlikely for storms or far-traveled tsunamis, would be expected of tsunamis from great Cascudia earthquakes.
It remains conceivable that storms or seiches caused the surges from which the sand was deposited.
These alternatives are now being tested through inference of the duration, depth, and veiocity of the surges (M. A.
Reinhart and Joanne Bourgeois, written commun., 1988).
Uplift.
Analogies with uplift at other subduction zones imply that great Cascadia earthquakes would produce elevated shorelines on the Northwest coast (West and McCrumb, 1988), particularly where a subduction-zone rupture extends beneath the coast.
Two candidates have been identified thus far (fig. 1)--one a beach gravel containing 3000-year-old wood in southern Oregon (Xelsey and others, 1988), the other a wave-cut bench perhaps 1000 years old in northern California (G.A. Carver, oral commun., 1988).
These poorly understood features may record Cascadia earthquakes whose ruptures splayed upward into subsidiary faults.
As shown by Kelsey and Carver (1988), faults probably rooted in the Cascadia 1
All ages in text are in sidereal years; calibrution of radiocarbon ages follows Stuiver and Pearson (1986) and Pesrson and Stuiver (1986).
(
2
subduction zone come ashore in northe-o Califorria.
The most recent
(-
thrusting on at least one of these faults occurred about 300 years ago (Carver and Burke, 1987).
Kelsey and Carver (1988) liken northern California to the Gulf of Alaska, where thrusting on subsidiary faults produced a complex pattern of uplift, and may have also caused local subsidence, during the great (magnitude 9.2) Alaskan earthquake of 1964 (Plafker and Ruben, 1978, p. 706, 721).
Youthful uplifted terraces are scarce or absent on the coast of northern Oregon and Washington (West and McCrumb, 1988), probably because these areas undergo only subsidence during great Cascadia earthquakes.
By analogy with the 1960 chile earthquake (magnitude 9.5) and the 1979 Tumaco, Colombia earthquake (magnitude 8.2), uplift might be chiefly confined to areas within 105 km (Chilean analogy) or 90 km (Colombian analogy)- of the base of the continental slope (fig. 3a, b).
At such distances, little or no coseismic uplift would occur along the northern quarter (Chilean analogy) or northern half (Colombian analogy) of the Oregon coast (fig.
3c). Even Cape Blanco (fig. 3c, d) could escape coseismic uplift if the Colombien analogy applies and if, as seems likely (llerd and others, 1979),
the Colombian subsidence extended tens of kilometers offshore (fig. 3b).
Shakinst.
The published case for great Cascadia earthquakes includes no compelling evidence that shaking accompanied the jerks of coastal subsidence.
The strongest known hint is vented sand, containing clasts of the mud through which it rose, that buried part of a spruce woodland in coastal southwestern Washington about 1000 years ago (sand blow at site Co-u, fig. 2).
The venting, indicative of strong shaking, does not seem to 1
have accompanied subsidence of the woodland, or of coastal southwestern Washington regionally.
But rapid subsidence did occur about 1000 years ago in northwesternmost Washington and, perhaps, near the mouth of the Columbia River (site De-1, fig. 2).
Only by this kind of permissive correlation does shaking seem have accompanied a Jerk of coastal subsidence in the Pacific Northwest.
The hypothesis of shaking during subsidence needs to be tested wherever easily vented sand underlies subsided wetlands that are well exposed in cross section.
Few such places exist in coastal southwestern Washington.
Additional evidence of shaking could take the form of landslides, provided that wet weather and non-Cnocadia carthquukes can be excluded as triggers.
CONC 1,USIONS Earth scientists have recently begun to study ancient subsidence, uplift, tsunamis, and shaking as elues to the seismic potential of the Pacific Northwest.
The little work done so far shows that great earthquakes probably have occurred on the Cascadia subduction zone in the recent pre-white-man past.
Particularly suggestive is the widesprend evidence for sudden coastal subsidence and accompanying tsunamis.
This evidence implies local precedent for the future occurrence of great earthquakes in the Pacific Northwest.
(
3
REFERENCES CITED C
Atwater, B.F.,1987, Evidence for great Holocene earthquakes along the outer coast of Washington state:
Science, v. 236, p. 942-944.
Atwater, B.F.,1988, Geologic studies for seismic zonation of the Puget lowland, _i_D National Eerthquake Hazards Reduction Program, Summaries of Technical Reports, v. 25:
U.S. Geological Survey Open-File Report 88-16, p.
120-133.
- Carver, G.A., and Burke, R.M., 1987. Late Holocene paleoseismicity of the southern end of the Cascadia subduction zone:
Eos (Transactions.
American Geophysical Union), v. 68, p. 1240.
Darienzo, Mark, and Peterson, C.D., 1987 Episodic tectonic subsidence recorded in late-Holocene salt-marshes, northwest Oregon:
Cos (Transactions, American Geophysical Union), v. 68, p.
1469.
Grant, W.C., and McLaren. D.D.,1987, Evidence for Holocene subduction earthquakes along the northern Oregon coast:
Eos (Transactions, American Geophysical Union), v. 68, p.
1239.
- Heaton, T.H.,
and Hartzell, S.H.,
1987, Earthquake hazards on the Cascadia subduction zone:
Science, v. 236, p.
162-168.
Herd, D.G.,
- Youd, T.L., Meyer, H.,
Arango C.,
J.L.,
- Person, W.J.,
and
- Mendoza, C., 1979, The great Tumaco, Colombia earthquake of 12 l
December 1979:
Science, v. 211, p. 441-445.
Kanamori Hiroo, 1977, The energy release in great earthquakes:
Journal t
of Geophysical Research v. 82, p. 2981-2987.
Kanamori, Hiroo, and McNally, K.C., 1982, Variable rupture mode of the subduction zone along the Ecuador-Colombia coast:
Bulletin of the Seismological Society of America, v. 72, p. 1241-1253.
Kelsey H.M., and Carver, G.A.,
1988, Late Neogene and Quaternary tectonics associated with the northward growth of the San Andreas transform fault, northern California:
Journal of Geophysical Research I
(in press).
Kelsey, H.M., McInclly, Galan, Abelli, Terese, Burke, R.M., and Carver, G.
A., 1988, Investigations of recent crustal deformation in south I
coastal Oregon, Lrl National Earthquake Hazards Heduction Program.
Summaries of Technical Reports, v. 25:
U.S. Geological Survey Open-file Report 88-16, p.
157-160.
j l
4 i
4
r National Ocean Su.vey, 1974, North Pacific Ocean:
NOS Seamap Series, NOS 12042-12B, scale 1:1,000,000.
- Nelson, A.R.,
1987, Apparent gradual rise in relative sea level on the south-central Oregon coast during the late Holocene--implications for i
the great Cascadia earthquake hypothesis:
Eos (Transactions, American Geophysical Union), v. 68, p. 1240.
O Pearson, G.W.,
and Stuiver, Minze, 1986, High precision calibration of the I
radiocarbon time scale, 500-2500 BC:
Radiocarbon, v. 28, p. 839-862.
l Plafker, George, and Rubin, Meyer,1978, Uplift history and earthquake recurrence as deduced from marine terraces on Middleton Island, Alaska, in Proceedings of Conference VI, Methodology for Identifying l
Seismic Gaps and Soon-to be Gaps:
U.S. Geological Survey Open-File i
Report 78-943, p. 687-721.
l' Plafker, George, and Savage, J.C., 1970, Mechanism of the Chilean l
earthquakes of May 21 and 22,1960:
Geological Society of America Bulletin, v. 81, p. 1001-1030.
I
- Prince, R.A., and others, 1980, Part I, Bathymetry of the Peru-Chile continental margin and trench:
Geological Society of Amercia Map and l
Chart Series MC-34, scale 1:1,095,706 at equator.
i i
Reinhart, M. A., and Bourgeois, M. A.,1987, Distribution of anomalous sand at Willapa Bay, Washington: evidence for large-scale landward-directed l
5
(
processes:
Eos (Transactions, American Geophysical Union), v. 68, p.
(
1469.
l Rogers, G.C.,1988, Seismic potential of the Casendia subduction zone:
l Nature, v. 332, p.
17.
Stuiver, Minze, and Pearson, G.W.,
1986, High-precision calibration of l
the radiocarbon time scale, AD 1960 - 500 BC:
Radiocarbon, v. 28, p.
805-838.
a i
West, D.O.,
and McCrumb, D.R., 1988 Coastline uplift in Oregosi and Washington and the nature of Cascadia subduction-zone tectonics:
l Geology, v. 16, p.
169-172.
t J
1 l
I f
i 6
1 4
4 5
i l
j j
r
alo* '
ris a
Cajah> R.
b sau c.s,.u Co Gr.ys y.a.,
s
~*
got 97*-
N %
L_
B1
'.f
// ape i
u
- Detail o
I.
, sn B T
N'. M5 v
u E *b est C G
s
{Ta N}* ;
o,3..
De 5
f
,_. _ b.,.,_. m Fig. 1.
Index maps.
(a), Cascadia subduction zone.
(b), coastal southwestern Washington.
(
SOUTM NO RTH 0
D. -l T.
W;.J wg y,**
Bl C. m u W vu f:-
n.
~
MC 14 a8 c
,94
? M r j',---.j ' a Q:.. -Q g
eiw-- i.,.. sh d,_,g s
na-3
.o
- 2
.d..u 4,.J, taq:.,
..a--
- au n
e..
.u l,z M
..e, 5
.u
.{g,,d
'h5g '
W f..,== Y Y'ug m
M' h P r h d *-
t 6
w a? I ~ C' y C
- U m
,a.e
=
meu,u ou e.m w
s
.nu u
em au e as L'J.J r,,,. j, # 4 Tig. 2.
Organic horizons and radiocarbon ages of buried-wetland soils in coastal Washington (Atwater, 1988).
Solid lines between columns denote correlations among radiocarbon-dated soils.
Localities (letter symbols at top) keyed to figure 1 and to tables of Atwater (1988). Material dated, i
with strat.igraphic ponition shown at lower right:
( e ) root of tree, chiefly Sitka spruce; ( V), (-) rhizome [below-ground stem) of Triclochin i
p_aritims, a grass-like tidal-marsh plant; (Q ), (X ) stick (s) or cones or both; (O ) uppermost 0.5 or 1.0 cm of organic horizon.
('
6 1
50 0
50 10 0 150 200
(
f^^,vbsv#v 0
IT'O Chil" 8.
y.2 A
(My 9.5) g+
l g
4
..i..
.i,...i...
.m Nw st/
aO nun.. m
.n.
- c. ~i E -2 A
197f Tum.co, l
Colombia (MwS2).
O
.g i....i....i....i....i....-
.i....,....
l000 g-e
!
- V 800 3* ~.
Y w-
- ~ 'v 7 i
tl j 600
'~~-
,- c h-4 c'f00 E
3
- Cape Blanco 6 -
- 4200 C-gg
.........7 19 0
. i.... i.
.. i..
.u. -
1~W Cape Blanco J~
'1 I
--_M
(
.:s 0 i
4
\\
m.1 C
2 m
59 i..'..i....i....i....i....-
50 0
50 100 150 200 Dist'ance (kad normal to base of f.ontinental slope.
Fig. 3.
Bathymetry and coseismic deformation versus distance from base of continental slope.
Triangles point in direction of coseismic vertical movement; open triangles denote offshore movement, queried where doubtful.
Moment magnitudes (N ) from Konamori (1977) and Kanamori and McNally (1982).
A, axis of trough or channel.
(a) Profile S82E through Valdivia, Chile.
Bathymetry from Prince (1980).
Coseismic deformation from Plafker and Savage (1970); arrow shows projected location of coseismienlly subsided coast 17 km SSW of profile.
(b) Profile SS3E through San Juan, Colombia (Herd and others, 1981).
(c) Shoreline location between central Vancouver Island (top) and Cape Mendocino (bottom).
Triangles show sites evincing coseismic aubsidence or uplift of late Holocene age (Darienzo and Peterson, 1987; Grant and McLaren, 1987; Nelson, 1987; Atwater, 1988; Kelsey and others, 1988; G.A. Carver, oral commun., 1988).
Square denotes site with evidence for only gradual coastal submergence (Nelson, 1987).
(d) East-west profile through Cape Blanco, Bathymetry from National Ocean i
Survey (1974).
,l l
7 i
i i
i i
I