Origin of the recrystallisation front in the Ronda peridotite by km-scale pervasive porous melt flow
Origin of the recrystallisation front in the Ronda peridotite by km-scale pervasive porous melt flow
复制标题
DOI：
10.1007/s004100050135
复制
发表时间：
1996-01
影响因子：
3.5
通讯作者：
D. V. der Wal;J. Bodinier
中科院分区：
文献类型：
--
作者：
D. V. der Wal;J. Bodinier
研究表明，上地幔的多孔熔体流动对部分地幔熔体的组成有显著影响。对于多孔流动的长度尺度以及多孔熔体流动是否可能是一个具有重要体积意义的岩浆过程，还没有得到很好的确定。关于普遍多孔熔体流动的长度尺度和性质的唯一观测来源是橄榄岩块体。在这里，我们以西班牙南部朗达地块橄榄岩的结构和主要元素和微量元素数据的形式提出了这样的观察结果。痕量元素浓度具有较高的分析精度（ICP-MS），包括在橄榄岩中很少分析的微量元素，如Rb、Th、Nb和Ta。隆达地块西部可划分为两个构造相。第一个也是最古老的是由变形的斑状碎屑尖晶石橄榄岩组成，第二个是几乎未变形的粒状尖晶石橄榄岩。它们被再结晶前沿分开，所有相的晶粒生长都发生在再结晶前沿。颗粒域可进一步划分为粗颗粒、细颗粒和层状颗粒橄榄岩三个亚相。根据构造相，Mg数[原子Mg/(Mg±Fe)]和不相容元素如稀土元素（REE）、Th和高场强元素（HFSE，包括Ti）在km尺度上的空间变化与Ca和Al丰度无关。这种变化让人想起那些通常归因于地幔交代作用的变化，但从未在千米尺度上有过记录。再结晶锋的成因与千米尺度的普遍熔融渗流有关。晶粒生长和熔体分数之间的反馈过程可能导致熔体在再结晶前沿的重要积累，主要通过熔化/溶解完成。熔体分数的变化解释了玄武岩再结晶程度、mg数、REE分馏和HFSE丰度的空间变化，并可以解释收敛构造和伸展构造背景下玄武岩之间的许多经典差异。粗粒橄榄岩反映了稳态普遍多孔熔体流动阶段，而细粒和层状颗粒相则可能反映了多孔熔体流动的终止阶段。与这两个领域相关的过程是在熔体体积减小时形成辉石岩的冻结反应，以及与橄榄石生成反应相关的熔体流动的渐进通道。这两种工艺在时间和空间上都表现出复杂的套印关系。
It is well established that porous melt flow in the upper mantle may significantly affect partial mantle melt compositions. Less well established are the length-scale of porous flow and whether porous melt flow can be a volumetrically important magmatic process. The only source for observations concerning the length-scale and nature of pervasive porous melt flow are peridotite massifs. Here we present such observations in the form of structural, and major and trace element data from peridotites of the Ronda massif, southern Spain. Trace element concentrations were obtained with high analytical precision (ICP-MS) and include trace elements rarely analysed in peridotites, such as Rb, Th, Nb and Ta. The western portion of the Ronda massif can be divided into two structural facies. The first and oldest is composed of deformed, porphyroclastic spinel peridotites, the second of virtually undeformed granular spinel peridotites. They are separated by a recrystallisation front across which grain growth of all phases occurred. The granular domain can be further subdivided into three subfacies: coarse-granular, fine-granular, and layered-granular peridotites. According to structural facies, km-scale spatial variations unrelated to Ca and Al abundances have been recognised for mg-numbers [atomic Mg/(Mg±Fe)] and incompatible elements such as rare earth elements (REE), Th and high-field-strength elements (HFSE; including Ti). Such variations are reminiscent of those commonly ascribed to mantle metasomatism, but have never been documented on the km-scale. The origin of the recrystallisation front is related to km-scale pervasive melt percolation. Feed-back processes between grain growth and melt fraction could have led to important accumulation of melt at the recrystallisation front, accomplished mainly by melting/dissolution. Variation in melt fraction across the front explains the spatial variation in the degree of recrystallisation, mg-numbers, REE fractionation, and HFSE abundances, and could account for many of the classical differences between basalts from convergent and extensional tectonic settings. Whereas the coarse-granular peridotites reflect a stage of steady-state pervasive porous melt flow, the fine- and layered-granular facies probably reflect the terminate stages of porous melt flow. Processes associated with both domains are pyroxene-forming freezing reactions at decreasing melt volumes, and progressive channelling of melt flow associated with olivine-producing reactions. Both processes show complex overprinting relationships in both time and space.