地質學家利用放射性時鐘來考究最長的地震紀錄

原文網址：www.sciencedaily.com/releases/2017/05/170502121249.htm

地質學家利用放射性時鐘來考究最長的地震紀錄

地質學家利用美國新墨西哥州，乳白色結晶礦脈中含有的放射性元素，回顧了超過400000年以闡釋格蘭德河張裂帶的布蘭卡山斷層中地震發生的證據。

這是刻畫在斷層中最長的地震紀錄，顯示出13起獨立的地震事件――其中九次的間隔呈現出40000至50000年的規律，另外四次則聚集在一起，僅相差5至11000年。

刊於上周《美國國家科學院院報》(Proceedings of the National Academy of Sciences)的研究中描述了這項成果。此研究由威斯康辛大學麥迪遜分校的博士後研究員Randy Williams與他的指導教授，地質科學系的Laurel Goodwin主持。

「我們並沒有期待會有如此成果。」Goodwin說，「對我們來說這確實是個漫長的探索之旅。」

布蘭卡山斷層位於美國新墨西哥州阿爾伯克基的南部，全長為14英里。研究人員最初的計畫是要更加瞭解沿布蘭卡山斷層發生的地震背景值，也就是固定發生的地震。像這樣的板塊內斷層相較於位處板塊邊界的斷層，比方說加州聖安地列斯斷層，地震發生的頻率通常比較低，地質學家對它們的瞭解也比較少。然而，近年來某些板塊內斷層的地震活動卻變得更加頻繁，可能原因為處置汙水用的深層注水井。

Williams解釋像德州、奧克拉荷馬州和俄亥俄州等地區，科學家認為地震活動跟注入地底深處的高壓汙水有關。研究人員部分是為了更深入瞭解這些人為導致的事件，他們想要掌握是什麼樣的力學因子控制了自然地震的發生，以及要讓斷層滑動而產生地震需要多少時間。

「我們無法預測出地震發生的確切日期，而且未來也不一定能辦到。」Goodwin解釋，「但我們想要瞭解驅動它們發生的機制，如此我們才能做出更好的準備。」

為了尋找答案，Williams開始研究斷層周圍的岩層中，由方解石礦物形成的條狀礦脈。某些地震發生時，會有受壓液體流經斷層附近的岩石並沉澱出方解石，它們會一層一層地堆積下來，就像樹木的年輪。當之後又有地震發生，方解石會破裂然後癒合，而留下像是曾經斷裂的骨頭上擁有的特殊痕跡。

Williams觀察了留存在這些方解石中的放射性元素――鈾和釷。根據鈾衰變成釷的速率，Williams可以將它們當作一種時鐘來使用。他和研究團隊中的其餘成員，包括柏克萊地質年代學中心的Warren Sharp和新墨西哥礦業理工學院的Peter Mozley，藉此測量出礦脈中每一「世代」的方解石年代為何，而定出地震彼此之間的相對發生時間。對比至較近期的地震事件之後，研究人員估計發生在布蘭卡山斷層的地震規模為6.2至6.9。

團隊顯示該斷層的地震受控於兩種不同的機制。每40000年左右就會發生，間隔規律的地震並非由流體導致，而是應力不斷累積最終破壞斷層的結果。然而，發生在將近430000年以前的不尋常群震，卻是地表深處液壓增加導致的結果。地球內部的液體壓力增加會使得斷層兩側之間的磨擦力下降，造成斷層更容易滑動――就像空氣曲棍球檯面上加壓空氣的作用一樣。

團隊也證明群震中，跟兩起事件相關的方解石指出曾發生相當迅速的二氧化碳脫氣作用(degassing)。此作用會發生於高壓下的流體突然釋放出來的時候――就像打開搖晃過後的汽水瓶瓶蓋一樣。

Goodwin解釋：「當孔隙水壓(pore pressure)的增加幅度遠遠超出背景值，斷層就會破壞並形成裂縫，而釋出盆地中的流體。」Williams表示注入廢水使壓力增加的速率，可能超出大部分斷層在過往地質年代中曾經歷過的速率。

此發現也有助於科學家解開地質學中歷時已久的問題：板塊內斷層的地震形成機制為何？它們是否有特定週期或者是隨機發生？

現下Williams正在研究如何改良團隊使用的方法，他說：「我們想瞭解方解石定年法要如何應用在其他斷層中所刻劃的地震紀錄。」

Geologists use radioactive clock to document longest earthquake record

Using radioactive elements trapped in crystallized, cream-colored "veins" in New Mexican rock, geologists have peered back in time more than 400,000 years to illuminate a record of earthquakes along the Loma Blanca fault in the Rio Grande rift.

It is the longest record of earthquakes ever documented on a fault, showing 13 distinct seismic events -- nine of which occurred at regular intervals averaging 40,000-to-50,000 years and four that clustered together just five-to-11,000 years apart.

The work, described in a study published last week in the Proceedings of the National Academy of Sciences, was led by University of Wisconsin-Madison postdoctoral researcher Randy Williams and his advisor, Laurel Goodwin, a professor in the UW-Madison geoscience department.

"We weren't expecting any of this," Goodwin says. "It's been quite the odyssey for us."

The researchers initially set out to better understand the background, or default, earthquake activity along the 14-mile-long Loma Blanca fault, south of Albuquerque. An intraplate fault like this generally produces earthquakes much less frequently than those at the boundaries of tectonic plates, like California's San Andreas Fault, and tends to be less well understood by geologists. However, some intraplate faults have experienced increased seismicity in recent years, likely due to deep injection wells used for wastewater disposal.

In places like Texas, Oklahoma and Ohio, Williams explains, earthquake activity has been linked to high-pressure wastewater injected far beneath Earth's surface. In part to better understand these human-driven events, the researchers wanted to get a handle on what mechanical factors control natural earthquakes and how often a given fault slips to cause one.

"We can't predict an exact date for when they will occur, and it's unlikely that we ever will," Goodwin explains, "but we want to understand what is driving them so we can better prepare."

To look for answers, Williams began to examine "veins" made of the mineral calcite that streak segments of rock along the fault. Calcite precipitates out of pressurized fluids that travel through rock near faults during some earthquakes and gets deposited in layers, like rings of a tree. During subsequent earthquakes, the calcite fractures and heals, leaving a distinct signature like old broken bones.

Williams looked at the radioactive elements uranium and thorium trapped in these calcite crystals, using them as a kind of clock based on the rate at which uranium decays into thorium. He and the rest of the research team, which includes Warren Sharp from the Berkeley Geochronology Center and Peter Mozley of the New Mexico Institute of Mining and Technology, could measure the age of each "generation" of calcite found in the veins and determine when earthquakes occurred relative to one another. The magnitude of these earthquakes at Loma Blanca has been estimated to be between 6.2 and 6.9, by analogy with more recent events.

The team showed that earthquakes on the fault were controlled by two different processes. Earthquakes that occurred at regular intervals were the result of accumulated stress that eventually caused the fault to fail every 40,000 years or so but were not driven by fluids. However, the unusual cluster of earthquakes that occurred roughly 430,000 years ago was the result of an increase in fluid pressure deep beneath the surface. Increases in fluid pressure in Earth can decrease the friction between the two sides of a fault, leading to easier sliding -- like the pressurized air on an air hockey table.

The team also showed that calcite associated with two seismic events in the earthquake cluster indicates very rapid carbon dioxide degassing, which can occur when fluid under high pressure is released -- like opening the top of a shaken bottle of soda.

"When pore pressure increases far enough over the background level, the fault fails and cracks form, releasing fluid from the basin," Goodwin explains. Williams says that injected wastewater is likely to increase pressure at a faster rate than most faults have experienced in the geologic past.

The findings also help contribute to a longstanding question in geology regarding the mechanics of earthquakes in intraplate faults and whether they occur periodically or randomly in time.

Today, Williams is working to improve the methods the team used. "We want to understand how the calcite-dating method can be used to contribute to documenting the seismic record of other faults," he says.

原始論文：Randolph T. Williams, Laurel B. Goodwin, Warren D. Sharp, Peter S. Mozley. Reading a 400,000-year record of earthquake frequency for an intraplate fault. Proceedings of the National Academy of Sciences, 2017; 201617945 DOI: 10.1073/pnas.1617945114

引用自：University of Wisconsin-Madison. "Geologists use radioactive clock to document longest earthquake record." ScienceDaily. ScienceDaily, 2 May 2017.