遠方地震能夠引發海底山崩

原文網址：http://www.washington.edu/news/2017/06/27/distant-earthquakes-can-cause-underwater-landslides/

遠方地震能夠引發海底山崩

新研究顯示大地震可以在發生後的數周或數月之間引發數千公里遠處的海底山崩。

研究人員分析了華盛頓至奧勒岡州外海的海底地震儀，而將發生在北美西北太平洋外海80至161公里處，卡斯卡迪亞隱沒帶上的一系列海底山崩，跟2012年發生在距其13500公里遠處的印度洋規模8.6地震連結起來。在四月的地震之後，這些山崩斷斷續續地發生將近四個月之久。

過往的研究顯示地震可以誘發全球各處的其他斷層產生別場地震，但這篇新研究顯示地震同樣可以在距其相當遠處引起海底山崩。

「基本假設是……海底山崩是由當地的地震產生。」主要作者，華盛頓大學的海洋學家Paul Johnson表示。此篇新研究發表於美國地球物理聯盟發行的期刊《地球物理研究期刊：固體地球》之中。

「但我們的論文表示：『不，海底山崩可能是由地球任何一處的地震引發。』」

這項新發現讓利用沉積物紀錄來評估地震風險的方法變得更加複雜。根據研究作者所述，如果海底山崩會由遠方的地震引發，而不僅限於附近的地震，那科學家利用海底山崩產生的沉積物來定年區域地震事件或是評估地震風險之前，可能要先考慮這些沉積物到底是由當地的地震產生，抑或是遠方的地震。

根據作者表示，在此研究中觀察到的海底山崩雖然規模較小且發生在局部地區，不像由卡斯卡迪亞板塊邊緣直接形成的大地震所產生的山崩分布如此廣泛，但由遠方地震形成的海底山崩仍有可能產生區域性海嘯，並損壞海底通訊纜線。

令人驚喜的意外

Johnson表示發現卡斯卡迪亞的山崩是由遠方的地震引發是場意外。

科學家在華盛頓至奧勒岡州外海的海底假設地震儀來偵測微震，並測量同一地點的海水溫度和壓力變化。當Johnson在一場學術會議上發現有這些地震儀時，他決定要分析儀器採集到的數據，以觀察是否有證據可以顯示海底的溫度會被某些熱力學作用影響，像是甲烷水合物的形成作用。

Johnson和他的團隊結合了2011至2015年的海底溫度、壓力和地震儀的數據，另外還有儀器被沉積物覆蓋的影片截圖。他們發現溫度在數個月的微幅變動之後，突然出現持續2到10天的高峰。他們總結這種溫度變化只可能代表了發生好幾道將沉積物捲入水裡的海底山崩，它們使溫暖的淺層海水密度變高，而沿著卡斯卡迪亞板塊邊緣往下方流動。此事件發生在4月11日規模8.6的印度洋地震之後，造成海底溫度突然出現高峰。

卡斯卡迪亞板塊邊緣在北美西北的太平洋外海地區從南到北綿延了超過1100公里，其包含的地區下方為一個板塊滑到另一個板塊之下的隱沒帶。

高達數百公尺的海底陡坡標記出卡斯卡迪亞板塊邊緣的所在位置，在這些陡坡的頂端會有沉積物堆積於此。當印度洋地震產生的震波到達此處，便會晃動累積於坡頂的厚層沉積物，使得大面積的沉積物脫落並往下坡滑動，然後在整個坡段產生一連串的山崩。根據作者表示，這些沉積物並非在地震之後一口氣全數崩落，而是在長達四個月的期間內斷斷續續發生。

在華盛頓至奧勒岡州的外海分布著一些坡度高於平均的海底斜坡，比方說Quinault峽谷急遽下降1420公尺，坡度達40度的峭壁，使得此處特別容易發生海底山崩。此外，堆積得相當厚實的沉積物也會放大來自遠方的震波。微小的沉積物顆粒可以像懸浮在水中的漣漪般運動，因此能放大震波。

Johnson表示：「這些東西已經蓄勢待發準備好要崩塌了，只等著某處發生一場地震。」

攪亂沉積物紀錄

根據研究作者所述，這項新發現可以對此區的海嘯研究有所啟發，但可能也會讓地震風險的評估變得更加複雜。

像卡斯卡迪亞板塊邊緣的隱沒帶具有發生海嘯的風險。當一座板塊滑入另一座之下，它們兩個會鎖在一起而蓄積能量。當板塊終於開始滑動，就會將其中的能量釋放出來而產生地震。突如其來發生的移動不只會將斷層上方的所有海水往上推動一大段距離，也會讓上覆板塊變平而讓附近海岸地區的高度下降，使海岸線更容易受到海水位移產生的巨浪侵襲。

Johnson表示海底山崩會增加海嘯災害的風險。當海底山崩發生時會將海水從原本的位置推開，而在局部海岸地區引起海嘯。

科學家也會利用海底沉積物紀錄來評估地震的風險。透過鑽探外海的沉積物岩芯，計算其中含有的山崩沉積物彼此之間的年代相距多遠，科學家可以建立出過去地震的發生年代表，並以此來預測一個地區未來地震可能發生的頻率以及達到的強度。

在北美西北太平洋外海產生的地震，會沿著英屬哥倫比亞至加州的海岸產生海底山崩。然而此篇研究發現遠方地震造成的山崩最大寬度或許可以到達20到30公里。這意謂著當科學家利用沉積物岩芯來判斷當地地震發生的頻率時，他們可能無法分辨這片到達海底的沉積物是由遠方還是當地的地震產生。

Johnson表示在此板塊邊緣科學家需要於更大的範圍中採集更多岩芯，才能提高地質紀錄解讀的精確度，並更新地震風險的評估結果。

Distant earthquakes can cause underwater landslides

New research finds large earthquakes can trigger underwater landslides thousands of miles away, weeks or months after the quake occurs.

Researchers analyzing data from ocean bottom seismometers off the Washington-Oregon coast tied a series of underwater landslides on the Cascadia Subduction Zone, 80 to 161 kilometers (50 to 100 miles) off the Pacific Northwest coast, to a 2012 magnitude-8.6 earthquake in the Indian Ocean – more than 13,500 kilometers (8,390 miles) away. These underwater landslides occurred intermittently for nearly four months after the April earthquake.

Previous research has shown earthquakes can trigger additional earthquakes on other faults across the globe, but the new study shows earthquakes can also initiate submarine landslides far away from the quake.

“The basic assumption … is that these marine landslides are generated by the local earthquakes,” said Paul Johnson, an oceanographer at the University of Washington and lead author of the new study published in the Journal of Geophysical Research: Solid Earth, a journal of the American Geophysical Union.

“But what our paper said is, ‘No, you can generate them from earthquakes anywhere on the globe.’”

The new findings could complicate sediment records used to estimate earthquake risk. If underwater landslides could be triggered by earthquakes far away, not just ones close by, scientists may have to consider whether a local or a distant earthquake generated the deposits before using them to date local events and estimate earthquake risk, according to the study’s authors.

The submarine landslides observed in the study are smaller and more localized than widespread landslides generated by a great earthquake directly on the Cascadia margin itself, but these underwater landslides generated by distant earthquakes may still be capable of generating local tsunamis and damaging underwater communications cables, according to the study authors.

A happy accident

The discovery that the Cascadia landslides were caused by a distant earthquake was an accident, Johnson said.

Scientists had placed ocean bottom seismometers off the Washington-Oregon coast to detect tiny earthquakes, and also to measure ocean temperature and pressure at the same locations. When Johnson found out about the seismometers at a scientific meeting, he decided to analyze the data the instruments had collected to see if he could detect evidence of thermal processes affecting seafloor temperatures, such as methane hydrate formation.

Johnson and his team combined the seafloor temperature data with pressure and seismometer data and video stills of sediment-covered instruments from 2011-2015. Small variations in temperature occurred for several months, followed by large spikes in temperature over a period of two to 10 days. They concluded these changes in temperature could only be signs of multiple underwater landslides that shed sediments into the water. These landslides caused warm, shallow water to become denser and flow downhill along the Cascadia margin following the 8.6-magnitude Indian Ocean earthquake on April 11, 2012, causing the temperature spikes.

The Cascadia margin runs for more than 1,100 kilometers (684 miles) off the Pacific Northwest coastline from north to south, encompassing the area above the underlying subduction zone, where one tectonic plate slides beneath another.

Steep underwater slopes hundreds of feet high line the margin. Sediment accumulates on top of these steep slopes. When the seismic waves from the Indian Ocean earthquake reached these steep underwater slopes, they jostled the thick sediments piled on top of the slopes. This shaking caused areas of sediment to break off and slide down the slope, creating a cascade of landslides all along the slope. The sediment did not fall all at once so the landslides occurred for up to four months after the earthquake, according to the authors.

The steeper-than-average slopes off the Washington-Oregon coast, such as those of Quinault Canyon, which descends 1,420 meters (4,660 feet) at up to 40-degree angles, make the area particularly susceptible to submarine landslides. The thick sediment deposits also amplify seismic waves from distant earthquakes. Small sediment particles move like ripples suspended in fluid, amplifying the waves.

 “So these things are all primed, ready to collapse, if there is an earthquake somewhere,” Johnson said.

Disrupting the sediment record

The new finding could have implications for tsunamis in the region and may complicate estimations of earthquake risk, according to the study’s authors.

Subduction zones like the Cascadia margin are at risk for tsunamis. As one tectonic plate slides under the other, they become locked together, storing energy. When the plates finally slip, they release that energy and cause an earthquake. Not only does this sudden motion give any water above the fault a huge shove upward, it also lowers the coastal land next to it as the overlying plate flattens out, making the shoreline more vulnerable to the waves of displaced water.

Submarine landslides increase this risk. They also push ocean water out of the way when they occur, which could spark a tsunami on the local coast, Johnson said.

Scientists also use underwater sediment records to estimate earthquake risk. By drilling sediment cores offshore and calculating the age between landslide deposits, scientists can create a timeline of past earthquakes used to predict how often an earthquake might occur in the region in the future and how intense it could be.

An earthquake off the Pacific Northwest would create submarine landslides all along the coast from British Columbia to California. But the new study found that a distant earthquake might only result in landslides up to 20 or 30 kilometers (12 to 19 miles) wide. That means when scientists take sediment cores to determine how frequent local earthquakes occur, they may not be able to tell if the sediment layers arrived on the seafloor as a result of a distant or local earthquake.

Johnson says more core sampling over a wider range of the margin would be needed to determine a more accurate reading of the geologic record and to update estimates of earthquake risk.

原始論文：H. Paul Johnson, Joan S. Gomberg, Susan L. Hautala, Marie S. Salmi. Sediment gravity flows triggered by remotely generated earthquake waves. Journal of Geophysical Research: Solid Earth, 2017; DOI: 10.1002/2016JB013689

引用自：University of Washington. "Distant earthquakes can cause underwater landslides."