揭露遠古岩漿的來源

原文網址：http://www.nature.com/ngeo/journal/v10/n6/full/ngeo2955.html

揭露遠古岩漿的來源

目前仍不確定地球最老地殼的成分為何。比較最古老的礦物顆粒跟年代近上許多的同類礦物，顯示最原始的地殼形成時，深受火成岩質基岩的再熔融作用影響。

原作：Elizabeth Bell

企圖重建地球嬰兒期的面貌會面臨一項重大阻礙：目前為止我們仍然尚未定年出可追溯至地球最初數百萬年――冥古宙(Hadean eon)時形成的岩石。澳洲Jack Hills的變質沉積岩擁有地球已知最古老的礦物，年代可達44億年。這些碎屑鋯石顆粒可以用來研究形成冥古宙地殼的岩漿來自何處。儘管科學家已經進行了大量研究，他們對於源岩的化學特性仍未取得共識。Burnham和Berry在6月《自然―地質科學》(Nature Geoscience)的論文中提出，最原始的地殼殘留下來的線索中，稀有元素的變化可以證實曾經發生火成岩質基岩熔融而產生的岩漿活動。

在研究整部地質歷史發生的沉積作用時，碎屑鋯石顆粒佔有的地位越發重要。它們是冥古宙(46億至40億年前)的岩石唯一留存下來的訊息。但是，我們需要某些工具來將碎屑鋯石跟岩漿來源的成分和地體構造做出特定的對應關係。由於鋯石跟特定類型的岩漿來源之間並沒有很清楚的關聯，使得冥古宙鋯石的可能來源有許多說法。有人假設跟月岩類似的富集玄武岩可以解釋鋯石的某些同位素地球化學特性，另一方面則有人援用含水量飽和的岩漿來解釋它們較低的結晶溫度。其中一個令人特別感興趣的問題是：在形成Jack Hills鋯石的時候，沉積物在過程中的參與程度。在冥古宙已有完整沉積循環的證據顯示，水不只會在岩漿中以溶質的身分出現，甚至有可能會在地表形成水圈，這對生命的適居性來說具有重大意涵。某些冥古宙的鋯石呈現出高比例的¹⁸O/¹⁶O(寫作δ¹⁸O)，有些則含有白雲母的包裹體，兩者皆是岩漿形成過程中有大量變質沉積物參與的顯著特徵。

Burnham和Berry提出以兩種不同類型的花崗岩作為類比，鋯石中的稀有元素變化可以用來辨識岩漿來源的類型。這兩種花崗岩分別是火成岩型花崗岩(I型花崗岩)和沉積岩型花崗岩(S型花崗岩)，前者由本來是火成岩的源岩重新熔融而產生，後者的來源則是熔融後的變質沉積物。雖然陸源的鋯石顆粒一般來說都有類似的稀有元素組成，Burnham和Berry卻呈現出由火成岩型岩漿結晶而成的大多數鋯石顆粒，擁有跟源自於沉積岩型岩漿的鋯石不同的化學訊號。在沉積岩型鋯石中，磷跟稀有元素加上釔的總合比例大致為1，但在火成岩型鋯石中卻低了許多。此外，沉積岩型鋯石中磷的總含量一般也比較高。這些差異使得在多數情況下可以用來區分兩種不同來源的鋯石，不過當磷的總含量較低時，磷含量和磷跟稀有元素的比例在兩種鋯石中的差異也比較模糊不清。將此區別方法運用到Jack Hills的冥古宙鋯石顆粒，Burnham和Berry發現跟預期的不同，這些遠古岩石的化學特性顯示它們的來源為火成岩型岩漿，也就是先前存在的基性下部地殼熔融過後產生的岩漿。此意謂沉積岩型岩漿的的地位不是那麼重要，並可能減少冥古宙早期沉積循環就已經建立的證據，使得冥古宙水圈的存在與否變得更不確定。

當時或許不是所有地殼的來源皆為火成岩型，如同某些磷比例低的鋯石所示，其來源可能還是富有沉積物。這可以解釋Jack Hills冥古宙鋯石中其他強烈指出沉積岩型來源的地球化學證據，即使它們在群體中只占一小部份。冥古宙的鋯石顆粒中有75分之1含有原生白雲母、39分之1具有高鋁含量、5分之1呈現出偏高的δ¹⁸O，這些通通都指示了沉積物來源。在Burnham和Berry用來區分兩種來源的圖表中，磷含量低的區域為意義模糊的鋯石。後續對這些鋯石進行更深入的探討，或許有助於釐清在冥古宙鋯石群集中，兩種來源所佔的相對比例為何。

Burnham和Berry借助古老鋯石顆粒中的元素比例，呈現出另一種概念模型來解釋地球最古老地殼的起源。比較地質紀錄中各時期鋯石稀有元素含量的研究，或許可以闡明冥古宙有多少類型的岩漿作用，此外也有助於我們了解在整部地質歷史中，地殼的形成環境是如何演變。

Ancient magma sources revealed

The composition of Earth’s oldest crust is uncertain. Comparison of the most ancient mineral grains with more recent analogues suggests that formation of the earliest crust was heavily influenced by re-melting of igneous basement rocks.

Attempts to reconstruct the conditions during Earth’s infancy face a significant roadblock: we have yet to identify rocks dating from the planet’s first several hundred million years, the Hadean eon. The meta-sedimentary rocks of the Jack Hills in Australia contain the oldest known minerals on earth, with ages of up to 4.4 Gyr . The geochemistry of these detrital zircon grains has been used to investigate where magma that formed Hadean crust originated from. Despite extensive study, no consensus on the chemical nature of the source rocks has emerged. Writing in Nature Geoscience, Burnham and Berry suggest that trace element variations in these remnants of the earliest crust provide evidence of magmatism resulting from melting of igneous basement rocks.

Detrital zircon grains play an increasingly important role in sedimentary investigations throughout the geologic record. They are the only known remnants of the rocks of the Hadean (4.6–4.0 billion years ago). However, there is a need for tools to specifically link detrital zircons to their source magma compositions and tectonic settings. As a result of the ambiguity in linking zircons to a specific magmatic source, a wide range of potential origins for the Hadean zircons have been proposed. Enriched lunar-like basalts have been posited to explain some aspects of their isotope geochemistry, whereas water-saturated magmas have been invoked to explain low crystallization temperatures. One question of particular interest is the extent to which sediments were involved in the formation of the Jack Hills zircons: evidence for mature sedimentary cycling on Hadean Earth would show that water was present not only as a dissolved component of magmas, but potentially also in a hydrosphere at Earth’s surface, with enormous implications for habitability. Some Hadean zircons exhibit elevated ratios of ¹⁸O/¹⁶O (denoted as δ¹⁸O) and some contain inclusions of muscovite, both telltale signs of magmas incorporating significant amounts of meta-sediments.

Burnham and Berry suggest that trace element variations in zircon can be used to characterize the type of magma source by analogy to two characteristic types of granite: I-type granites, which are re-melted from originally igneous source rocks, and S-type granites, which form from melted metamorphosed sediments. Although continental zircon grains generally have similar trace element compositions, Burnham and Berry have shown that the majority of zircon grains crystallizing from I-type magmas exhibit a distinct signature from those derived from S-type magmas. The ratio of phosphorus to the sum of rare-earth elements and yttrium is close to unity in S-type zircons, but significantly lower in those from I-types. S-type zircons also generally have higher overall phosphorus contents. These differences allow discrimination between zircons from these two origins in most cases, although the combination of phosphorus content and the ratio with rare-earth elements is more ambiguous where total phosphorus contents are lower. Applying this discriminant to Hadean zircons from Jack Hills, Burnham and Berry find that contrary to expectations, the chemistry of these ancient rocks suggests an I-type origin, consistent with a source from melted pre-existing mafic lower crust. This suggests a lesser role for S-type magmas, and potentially less evidence for the establishment of an early Hadean sedimentary cycle, making the existence of a Hadean hydrosphere less certain.

The I-type origins may not be universal, as some zircons with a lower proportion of phosphorus could still derive from a sediment-rich source. This could account for other geochemical evidence from the Hadean Jack Hills that are strongly suggestive of an S-type origin, albeit potentially for a minority of the population. 1 in 75 Hadean zircon grains contain primary muscovite, 1 in 39 yield high aluminium contents, and 1 in 5 show8 elevated δ¹⁸O, all suggesting sedimentary origins. Further investigations of zircon falling into the ambiguous, lowphosphorus region of Burnham and Berry’s discriminant diagram may help clarify the two origins and their relative abundances in the Hadean zircon population.

Burnham and Berry offer an alternative conceptual model for the origin of Earth’s oldest crust with the help of elemental ratios in ancient zircon grains. Comparative studies of zircon trace element contents throughout the geologic record may shed light on the diversity of Hadean igneous processes, and additionally help to understand changing crustal conditions over Earth’s history.

Published online: 8 May 2017 Nature Geoscience

原始論文：Burnham, A. D., and A. J. Berry. Formation of Hadean granites by melting of igneous crust. Nature Geoscience 10, 457-461 (2017). doi:10.1038/ngeo2942

引用自：Elizabeth Bell. Petrology: Ancient magma sources revealed. Nature Geoscience 10, 397–398 (2017). doi:10.1038/ngeo2955