Every day, about 33 to 38 earthquakes occur around our planet. Most of these earthquakes are magnitude 2 or less, and are generally not felt. Major earthquakes, greater than magnitude 7, happen once or twice per month ,while “great earthquakes”, magnitude 8 and higher, occur about once a year. The most powerful earthquake ever recorded was in Chile in 1960, at 9.5. The extreme magnitude of this event was initially an enigma to geophysicists, as they had no idea how such an extreme amount of force could be released.
Four years later, the second-largest earthquake ever recorded occurred in Alaska, registered at M 9.2. The earthquake, which shook the Alaskan coast for more than four minutes, caused about $2.3 billion in damage in 2018 dollars. Even though the epicenter of the earthquake was close to Anchorage, Alaska, effects were observed as far away as Texas and Louisiana. The team of USGS scientists that were dispatched to the site immediately after the earthquake observed that 184,000 square miles of land and water surface had been disrupted by the earthquake, an area larger than the size of California. While some areas sank by eight feet, others rose by more than 38 feet within this area.
It turns out that these two events were what are now considered “megathrust” earthquakes. These earthquakes happen when tectonic plates – parts of the earth’s crust – slide under one another. As the plates slide past one another, friction causes an enormous amount of strain to accumulate over hundreds of years. When the plates eventually slip, they release a few hundred years’ worth of built-up pressure.
In Alaska, the denser Pacific plate slides under the Alaskan plate. Friction causes the Pacific plate to pull the edge of the Alaskan plate downward, which forces the overriding plate to bend slightly, causing a few meters of uplift over time. Over hundreds of years, strain builds up, inevitably being released as a massive earthquake. In a matter of seconds, the edge of the Alaskan plate will shove forward (toward the ocean) and upward, usually generating a tsunami. The relaxation of the Alaskan plate results in the toe of the plate being slightly higher, and the area that was being uplifted suddenly sinking back down (see the diagram below). This cycle repeats itself every few hundred years and is often recorded in the sediment of salt marshes located along the fault, which can be very sensitive to these land-level changes, as they are always found within a few meters of sea level along the coast.
A) Represents the build-up of pressure as the oceanic Pacific plate slides under the continental Alaskan plate. Note that areas along the coast are uplifted. B) Shows the relaxation of the Alaskan plate following an earthquake, where coastal areas sink (subside). Figure reproduced from Nelson et. al., 1996.A 1996 paper by Alan Nelson of the United States Geological Survey (USGS) serves as an excellent guide for interpreting the information found in marshes as it relates to earthquakes, as well as tsunamis and storms. For example, layers of peat overlain by mud and/or sand layers are indicative of a sudden drop in land level, as sediment devoid of plant matter is deposited when the marsh plants become inundated by saltwater and can no longer survive at that elevation. As friction between the plates begins to re-accumulate strain following an earthquake, the land slowly rises back up, This allows marsh grasses to grow again, resulting in the deposit of more peat. The cycle of deposition repeats itself every few hundred years, resulting in repeating layers of peat, mud, and more peat.
Using Nelson’s 1996 paper as a guide for initial interpretation, scientist Robert C. Witter added to the marsh scientist’s toolbox by focusing on the types of organisms found at different elevations within a marsh. When looking at salt marsh cores, the distribution of microorganisms such as diatoms and foraminifera can provide us with an idea of the past sea level. Different species of these microorganisms thrive at very specific elevations near sea level, and can be used to infer past sea level at centimeter-scale accuracy. Sharp contacts of less than 3mm between sediment layers indicate rapid change in the environment. On the other hand, gradual contacts of more than 3mm indicate the lack of a rapid change. When environmental change is rapid (as in a sudden land-level change due to an earthquake), there is less time for microorganisms to colonize the marsh. Therefore, Witter inferred that muddy layers lacking diatoms and foraminifera are usually indicative of earthquake-induced subsidence.
Based on the knowledge presented by Nelson and Witter, a team of scientists led by Richard W. Briggs, also from the USGS, went to Sitkinak Island, Alaska to obtain salt marsh cores in order to retrace past earthquakes. They compared the written historical records of earthquakes to the cores they obtained, as well as extended the Alaskan earthquake record beyond written accounts of the past few hundred years. Sediment cores showed that five earthquakes occurred in the last 2000 years: 290 – 0, 520 – 300, 1050 – 790, 640 – 510 years before present, and of course the 1964 earthquake.
Briggs and his team found that the oldest three earthquakes caused rapid uplift while the two most recent earthquakes caused subsidence in the area around the marsh. This is likely due to the fact that each earthquake occurred in a slightly different location along the fault, thereby influencing the marsh a bit differently each time. In other words, the Pacific plate is not submerging in one specific location: it’s submerging in different “segments” along the fault.
Prior to Briggs’ study, scientists believed that earthquakes generally occurred along well-defined segments of a fault. This would mean that a single location’s land level should be affected roughly the same way every time the segment slips, either always sinking or always rising. However, when Briggs’ team showed that the marsh on Sitkinak could either uplift or subside depending on the earthquake, a new eye-opening natural process was uncovered. He suggests that the area of subsidence is not as simple as once thought, and may exist on either side of the uplifted area as well as inland of it.
Seismologists would obviously like earthquakes to behave predictably in clean segments, but nature is far more complex than any person or computer can understand, at least for now.
Zane Grissett is in the final year of his undergraduate degree (B.S. Geology & Geological Oceanography) at the University of Rhode Island. He is currently studying the environmental effects of a 9-million-year-old meteorite impact in Argentina, but when he has some free time, you can usually find him surfing or spearfishing somewhere along the coast of Rhode Island.
Logan Thomas is a senior undergraduate at the University of Rhode Island pursuing a degree in geology with interests in the GIS field.
Omar Fahmy is a Masters of Environmental Science Management Student at the University of Rhode Island. Omar is also a professional stone mason. When not working Omar enjoys surfing and spending time with his family and friends on the southern coast of Rhode Island.
Feature image: Damage from the 1964 earthquake in Alaska. USGS.
Nelson, A. R., Shennan, I., & Long, A. J. (1996). Identifying coseismic subsidence in tidal‐wetland stratigraphic sequences at the Cascadia subduction zone of western North America. Journal of Geophysical Research: Solid Earth, 101(B3), 6115-6135.
Briggs, R. W., Engelhart, S. E., Nelson, A. R., Dura, T., Kemp, A. C., Haeussler, P. J., … & Bradley, L. A. (2014). Uplift and subsidence reveal a nonpersistent megathrust rupture boundary (Sitkinak Island, Alaska). Geophysical Research Letters, 41(7), 2289-2296.
Witter, R. C., Kelsey, H. M., & Hemphill-Haley, E. (2003). Great Cascadia earthquakes and tsunamis of the past 6700 years, Coquille River estuary, southern coastal Oregon. Geological Society of America Bulletin, 115(10), 1289-1306.