利用碳十四定年改善北極海沉積物錳循環作為地層對比之工具

Abstract

北極海深海沉積物中的錳循環在過去十幾年間一直被做為指示冰期和間冰期循環的對比工具 (Löwemark et al., 2014)，錳含量高的地方可以代表較暖的間冰期，反之，量少處代表冰期。這主要是因為冰期和間冰期之間，進入北極海中的錳受到海冰和陸上冰棚分布的影響，而有很大來源上的差異。我們的海洋岩心主要來自於LOMROG(07-12)研究計畫和AO96研究計畫所採樣回來之海洋沉積物。為了讓沉積物中錳循環可以做為更精確的對比和年代建立工具，我們採取深棕色且錳含量高的間冰期沉積物中的浮游有孔蟲Neogloboquadrina pachyderma (sinistral)，利用碳十四定年，提供絕對定年，建立前25公分沉積物正確的年代。過去研究認為，沉積物岩心最上端的第一個錳峰值代表全新世，其下為MIS 3和MIS 5等間冰期之紀錄，而這些間冰期沉積物由低錳含量且缺乏鈣質微體化石的冰期沉積物所區隔，形成一個由錳值高低可以區別的冰期與間冰期循環。 然而，只單使用錳循環作為對比工具仍有些缺點和限制，例如上部沉積物在鑽取過程中的缺失或是沉積後受氧化還原環境變化而再次位移的氧化錳。我們共對18根海洋岩心的前25公分進行碳十四定年，結果顯示有岩心的上部沉積物(全新世的沉積物)缺失，以及許多沉積物在MIS 2也存有缺失之現象。MIS 2沉積物的缺失是由於上次冰河最盛期(LGM)海冰幾乎覆蓋整個北極海，導致極低的沉積速率或是完全沒有沉積紀錄所造成。而記錄到MIS 2的海洋岩心可能是由於區域性的冰間湖提供少許沉積物來源，或是特定的海流循環造成海冰覆蓋區域的改變，而造成開放性水域，得以沉積。沉積物中的錳循環雖可以做為初步且快速的對比工具，指示出其為冰期或間冰期的沉積物，但上部和MIS 2的缺失可能會造成往下對比的精確性。因此，利用錳循環作為冰期和間冰期對比工具必須佐以碳十四定年，方能成為一個更精確且可信的對比工具。對於MIS 2沉積物缺失的研究是未來關於此研究的重要方向，以提供更多對於沉積物中錳循環實用性的參數和適用範圍。 對於具有MIS 2缺失的岩心，能指示出全新世和MIS 3沉積物的交界可以更進一步地研究這兩個時期間古海洋上的變化。由於在有MIS 2缺失的沉積物中，錳循環顯示為連續的峰值(全新世和晚MIS3的錳訊號)，難以界定其交界為何處。而利用IRD變化可以看出這兩個時期的交界，因為全新世的IRD較MIS 3的IRD含量低許多，因此在此兩時其交界處有明顯的IRD變化。而記錄到MIS 2的沉積物中，也因為MIS 2海冰覆蓋嚴峻，其IRD含量遠少於全新世與MIS 3，因此，IRD含量亦可作為指示出MIS 2沉積物的指標。此外，浮游及底棲有孔蟲之絕對豐度以及浮游有孔蟲之組成也顯示在格陵蘭北部陸棚區域，在全新世和MIS 3間，海冰狀況沒有太大的變化。然而在北極海中央，全新世沉積物中有較高的有孔蟲豐度，且浮游有孔蟲組成中有較高的T. quinqueloba比例，顯示全新世的海冰狀況沒有MIS 3般嚴峻，而可能有區域性的冰間湖出現，形成較適合浮游有孔蟲T. quinqueloba生長的環境。

The distinctive pattern of Mn content in Arctic deep marine sediment has been used as a proxy to indicate glacial-interglacial cyclicity (Löwemark et al., 2014). As it has been observed in many sediment cores, Mn peaks correspond to warm interglacial periods. In order to improve the preciseness of Mn pattern as a proxy, we collected the foraminifera Neogloboquadrina pachyderma (sinistral) from brownish, Mn-rich layers, and performed radiocarbon dating on selected cores collected during the LOMROG07, LOMROG09 and LOMROG12 expeditions. Additional cores form the AO96 expeditions are also included. Based on our general understanding of the Mn system in the Arctic Ocean, we predicted a Mn pattern with a high peak in the uppermost core top, representing the Holocene. This Holocene peak in Mn is separated from the underlying warm period MIS 3 by a Mn-poor interval also characterized by a drop in Ca. This low Mn and Ca interval represents MIS 2 and the LGM. Older warm periods, such as MIS 3, 5, 7 etc will display a similar pattern with distinct Mn peaks, separated by Mn minima representing cold periods. For example, the MIS 5 sometimes shown a distinct pattern with three Mn peaks representing MIS 5a, 5c and 5e. However, there are still some limitations in the applicability of Mn stratigraphy, such as the remobilization of the Mn layer and the core-top loss during coring. We performed AMS carbon 14 dating on 18 cores. The result revealed several cases of core top loss, leading to depletion or complete loss of the Holocene interval. In several cores, our AMS dating revealed absence of in the MIS 2 interval. The complete lack of MIS 2 sediment likely is the result of extremely slow sedimentation rate due to severe sea ice conditions, while places with records of LGM may be the result of polynyas within the sea ice, or certain circulation pattern. Consequently, although Mn pattern can be used as a preliminary tool to identify glacial-interglacial cycles, the loss core tops and glacial hiatuses limits the usage and accuracy of the correlation of Mn stratigraphies. Therefore, radiocarbon dating can refine our understanding of the Mn patterns in Arctic marine sediment and help to make it a better proxy for both paleo-environmental reconstructions and for the age models. Further study on the cause of hiatus often encountered in the LGM interval is necessary to ensure the usefulness of Mn stratigraphy. In sediments with MIS 2 hiatus, it is important to identify the intersection of the Holocene and MIS 3, which hardly can be distinguished simply by the Mn variation, since the continuous Mn peaks represent the combination of both warm periods. The IRD content/index can show an apparent difference between these two stages, with low amount of IRD in Holocene and high amount in MIS 3, additionally, MIS 2 was also characterized with extremely low IRD due to the severe sea ice coverage. Furthermore, the foraminifera abundance and planktonic foraminifera assemblage performed both in the interval of Holocene and MIS 3, revealed that in the southern Lomonosov Ridge off Greenland and Moris esup Rise, sea ice condition was relatively consistent throughout these two stages. Whereas in the central Arctic and the Siberian side of the Lomonosov Ride, higher foraminifera abundance and higher planktonic foraminifera T. quinqueloba assemblage in the record of Holocene indicates a more open water condition in this area during the Holocene than late MIS 3. And this may be a result of a regional polynyas-setting.