Living on the Edge

Bert Rubash lives in Bellingham and has a view of Mount Baker.

The city of Bellingham and Bellingham Bay lie on the western edge of the North American continent at an interface between land and ocean, between two very different environments. Their character is determined more by interaction between these two environments than it is directly determined by either of them.

The city and the bay are also located at a geological interface, and the geological character of the neighborhood is likewise determined by interaction between environments, by the collision of oceanic and continental crustal plates. Viewed through the perceptions of a geophysicist, Bellingham Bay is located within an active continental margin.

Much of the continental rock here at the margin is old, and much of it formed into rock in other places and under other circumstances, but the mountains, themselves, are young and still growing.

Not far west of Bellingham Bay, four or five degrees of longitude, the young Juan de Fuca ocean plate is forming under the ocean at a spreading zone, a place where oceanic plates move apart. There, molten basaltic rock from deeper in the Earth wells up and solidifies on contact with cold seawater.

The Juan de Fuca Plate, a remnant of the Farallon Plate that disappeared under the North American Continent along the California coast, is moving northeastward toward the continent. The oceanic and continental plates are converging at a speed of about one and one half inches per year (42 mm); their collision forces the thinner, denser ocean crust under the continental crust.

The effects at the surface of the Earth in the zone of collision are spectacular. Some of the sedimentary accumulation from the sea floor is scraped off at the initial point of collision, the subduction trench, and shoved against the continent; further in and deeper in the Earth, water saturated rock in the upper part of the descending plate melts and convects upward.

At the surface of the Earth, the stress of the collision raises up rugged mountains, and volcanoes appear at the tops of the convection plumes of molten rock from the descending plate.

Bellingham Bay is located within this zone, with Mount Baker and Glacier Peak, two of the subduction zone volcanoes to the East, and the subduction trench to the West, just offshore of Vancouver Island. Cross Section: Juan de Fuca Subduction Zone (Figure 1) below left is of the active margin, of our geologic neighborhood.

The Earth’s crust is slowly deforming here at the continental margin, and at times adjusts abruptly, fracturing where it is weakest or yielding along pre-existing fractures. Such adjustments produce earthquakes that are often felt throughout the surrounding region; see above Northwest Washington Earthquakes (Figure 2) top of page.

The fractures are called faults, and those faults where crustal readjustments have occurred in the last few centuries are considered active faults; they are faults that are expected to rupture again. Most of the known faults in the region near Bellingham are inactive.

Those that mark the boundaries between rock formations of different ages and different types have been mapped and studied extensively, but active faults have not been well-located and few have been mapped at the surface of the Earth in this region and northward. A well-mapped fault is the contact between the continental crust and the subducting oceanic plate which passes below Bellingham at a depth of about 50 kilometers; the location and geometry of this large thrust fault1 is well-known and its intersection with Earth’s surface along the subduction trench has been accurately mapped.

Another known active fault is the Devils Mountain fault, which extends 125 kilometers from the Cascade foothills to Vancouver Island, passing just south of the San Juan Islands. The surface trace of the Devils Mountain fault is included in Northwest Washington Earthquakes (Figure 2) top of the page.

One or two unmapped active faults lie under Orcas and San Juan Islands. A grouping of Earthquake source locations, called epicenters, near Friday Harbor is shown in three-dimensional view in Friday Harbor Earthquakes (Figure 3) on the facing page top. The double clustering of epicenters indicates repeated local slippage centered about 15 kilometers down, in the continental plate, and also local slippage centered about 55 kilometers below the islands in the subducting oceanic plate.

The clustering of quakes beneath Friday Harbor indicates active faults—faults that may, one day, slip along their entire extent producing a strong earthquake.

See the facing page bottom Bellingham Bay Earthquakes (Figure 4), a close-up view of the epicenters of Bellingham Bay earthquakes. Unlike those under Friday Harbor, the epicenter locations are scattered rather than clustered, and only one occurred at the depth of the oceanic plate. Local stress in the continental crust under Bellingham is being relieved incrementally through small scale fracturing and slippage. None of the earthquakes in the figure were large enough to be readily felt by people in Bellingham.

Several important characteristics of the dynamic geology of the Pacific Northwest are different from those at other subduction zones on the planet: the oceanic plate at the point of subduction is younger, warmer and thus more buoyant, and the angle of subduction is shallower than at other subduction zones.

Until recently, until ten or fifteen years ago, it was thought that these differences implied that the earthquake hazard in the Pacific Northwest was small. Earthquakes within the crust and within the oceanic plate were well-documented and part of the experience of most people living here; but earthquakes generated by the region-wide, seven hundred and forty-mile long Cascadia Subduction Zone thrust fault extending from Northern California into British Columbia were unknown.

It was thought that there were no subduction fault earthquakes because, although crustal strain was evident, there appeared to be little accumulation of strain; the subducting plate must be sliding smoothly under the continent.

Perhaps it was the warmer temperature of the Juan de Fuca oceanic plate that enabled the fault to yield slowly instead of rupturing. In any case, something was different about this active continental margin because there was no record of any earthquakes of the magnitude that have been experienced in other subduction zones, nor had seismographs recorded activity within the central area of the fault itself—all activity was within the plates. Cross Section: Juan de Fuca Subduction Zone (Figure 1) facing page bottom left shows the locations of the two groups of earthquake epicenters.

There was evidence of large earthquakes, of cataclysmic earthquakes, but no one identified the evidence until recently. In 1983, John Adams, a New Zealand geologist working for the Geological Survey of Canada published evidence from a re-survey of highway survey markers that highways crossing the Coast Range were tilting toward Puget Sound; it was a sign of accumulating strain.

Geologists began to consider the possibility that the subduction zone was not sliding, but stuck, causing the surface of the continental crust to bulge near its western edge. More evidence appeared when in 1986 Brian Atwater, a geologist with the United States Geological Survey, discovered evidence of a forest near Walapa Bay in Oregon that was drowned when the surface of the land suddenly dropped and was flooded with sea water, killing the trees.

Drowned forests and freshwater marshes were subsequently found all along the coasts of Oregon, Washington, and Vancouver Island and in the San Juan Islands. Native American legends, radiocarbon dating, tree ring analysis, and finally correlations made with tsunami records in Japan established that the Pacific Northwest experienced a very large earthquake at 9 p.m. on January 26, 1700.

The rupture extended nearly the entire length of the Cascadia Subduction Zone, from California into British Columbia, releasing crustal strain and thus relieving the bulge at the western edge of the continent.

Strain has been building since then; sea level changes found in records from 1934 of tide levels at Neah Bay and Friday Harbor and recent Global Positioning System studies of crustal movement contribute evidence of accumulating strain. Yearly Average Tide Levels (Figure 5) top of the page compares Neah Bay and Friday Harbor tide records.

Evidence has been found for other very large earthquakes during the period since the retreat of the glaciers of the last ice age. Geologists now generally agree that the Pacific Northwest is a land where some of the largest quakes on Earth periodically alter shorelines, shake slopes loose from hillsides and mountains, and send giant waves onto the shores abutting the Pacific Ocean.

Details of this recent and dramatic change of consensus within the scientific community is narrated by Robert S. Yeats in his book “Living with Earthquakes in the Pacific Northwest.”

Not far east of the San Juan Islands stand two snow- covered mountains that are taller than the peaks and ridges around them: Mount Baker and Glacier Peak. Both are active volcanoes. Their location is evidence that the subduction zone that begins in the Pacific Ocean to the west extends well to the east of Bellingham Bay.

The volcanoes are cones of lava and ash that have formed above the point where the water-saturated sea floor on the upper surface of the subducting plate has descended into a region of the Earth’s interior where pressure and temperature are high enough to cause it to melt. The rock surrounding the melting sea floor is denser and more heat resistant, and remains solid.

The molten sea floor, called magma, is buoyant, and it works its way upwards through fissures and by incorporating rock above it until it reaches the surface where it is called lava. The lava of the Cascade volcanoes is viscous and erupts only when great pressure has accumulated within the volcano.

Both volcanoes share another feature that makes them even more explosive. The cones are formed of layers of water-permeable ash and less-permeable lava that channels and traps water within the cone. Magma working its way upwards into the cone and traveling along fissures turns trapped water into steam producing immense internal pressure. When the eruption eventually occurs it is accompanied by an additional explosive release of steam.

The next eruption of Mount Baker should be interesting, and those of us living in Bellingham will have one of the best views of a mountain transforming itself. We also have the best view of other, equally fascinating, if less immediate geologic processes. We live on the edge of the continent in the midst of mountains and fjords under construction. The geologic dynamism of this land is both fearsome and magnificent.

Parts of this article are adapted from contributions to a technical paper submitted to the Padilla Bay Reserve.

Atwater, B. F. (2000, May). “Coastal Evidence for Great Earthquakes in Western Washington,” U.S. Geological Survey Professional Paper 1560, U.S. Geological Survey, Reston, VA, United States.

Claque, J. J. and P. T. Bobrowsky (1999, March). “The geological signature of great earthquakes off Canada’s west coast,” Geoscience Canada 26(1), 1-15.

Engebretson, D. C., A. Cox, and R. G. Gordon (1985), “Relative Motions between Oceanic and Continental Plates in the Pacific Basin.” Special Paper 206, The Geological Society of America, Boulder, CO.

Johnson, S. Y., S. V. Dadisman, D. C. Mosher, R. J. Blakely, and J. R. Childs (2000), “Late Quaternary tectonics of the Devils Mountain fault and related structures, northern Puget Lowland.” In Geological Society of America Abstracts with Programs, Volume 32, No. 6, Boulder, CO, United States, pp. A-21 and A-22 Geological Society of America.

Melosh, H. J. (1987), “A Finite Element Study of Strain Accumulation and Release in the Pacific Northwest.” In Eos, Transactions, American Geophysical Union, Volume 68, No. 44, Washington, DC, American Geophysical Union.

Waitt, R. B., L. G. Mastin, and J. E. Beget (1995), “Volcanic-hazard zonation for Glacier Peak Volcano, Washington.” U. S. Geological Survey Open-File Report; Map OF 95-0499, U. S. Geological Survey, Reston, VA, United States. Geologic hazards map. Report is available for download at http://vulcan.wr.usgs.gov/Volcanoes/GlacierPeak/Hazards/OFR95-499.

Yeats, R. S. (1998). “Living with Earthquakes in the Pacific Northwest,” Corvalis, OR, Oregon State University Press.

1 The plane of a thrust fault is closer to horizontal than vertical, and the activity of the fault tends to shorten the distance between points on opposite sides of the fault; the activity is compressive.