Geochemical And Petrophysical Responses To Fluid Processes Within Seismogenic Fault Zone

Fault zones have been demonstrated as areas favoring crustal fluid migration. These fluids may influence both mechanical and chemical characters of fault zones. Fluids within fault zones can react with the brecciated fault rocks and produce abundant low strength clay minerals. Furthermore, fluids may be sealed and can build up high pore pressure, consequently leading to weakening of the fault. Earthquake fault rocks preserve a great deal of information regarding the nature of infiltrating fluids and fluid-rock interactions. In order to investigate the fluid-rock reactions, origin and enrichment of clay minerals, as well as the effect of fluids on the fault zone’s evolution and faulting processes, this work performed a systematic study on the internal structures and mineralogical and geochemical compositions as well as transport properties of fault rocks sampled from a surface outcrop and shallow boreholes on the Yingxiu-Beichuan fault nearby the Golden River (GR) mine, that hosted the 2008 Wenchuan Mw 7.9 earthquake.Fault rocks at the GR outcrop consist of incohesive breccias and gouges which formed under relatively shallow depths, and cohesive protocataclasite and cataclasite which are representative of fault rocks at deeper portion of the fault. Fault rock distributions revealed from the boreholes are consistent with those seen at the surface outcrop. Strips of gouges with varied colors (yellow-greenish, grey-greenish and blackish) were found in the fault zone. Fault rocks formed at diverse depths and times were stacked together in the fault zone by multiple seismic slips and creep in long term inter-seismic periods as well as on-going uplift and erosion. The co-seismic slip of the Wenchuan earthquake was inferred to cut through the yellow-greenish gouge. Toward the fault core, fault rocks show increasingly-enhanced fracturing, grain size reduction and fluid alteration trends. Mafic minerals in the granitic cataclasite and mafic rocks were generally altered to chlorite. Microstructure, mineral composition and element correlation analyses, as well as REE patterns of the fault rocks consistently indicate that the material sources of these fault rocks are complicated, and the results show that the blackish gouge was mostly derived from granite; the yellow-greenish and grey-greenish gouges were from a mixture of granitic and mafic rocks.Compared to host rocks, mineralogical and geochemical compositions of the fault rocks were distinctly modified. Pervasive fluid-rock interactions resulted in the transformation of felsic minerals to phyllosilicates. On the basis of the alteration textures and the typical hydrothermal mineral phases identified, we infer that the fault fluid is rich in Mg and Fe cations, and under an intermediate-acidic, reductive condition. The main alteration reactions include 1) alteration of mafic minerals into chlorite and 2) alteration of feldspars (K-feldspar and plagioclase) into clay minerals. Isocon analyses further reveal large volume loss, which is attributed to the removal of mobile components by means of fluid infiltration and compaction. Dissolution-precipitation and pressure solution are the main mechanisms responsible for the mass loss. Furthermore, due to the finer grain size, the fault core shows a larger degree of mass loss and a bigger fluid/rock ratio than the damage zone. Therefore, co-seismic mechanical cataclasis and the intensive fluid-rock interactions during inter-seismic periods modified and controlled the structural and compositional evolution of the fault zone.Permeability is a main parameter that controls fluid migration and fluid-rock interactions, and is essentially related to the generation of high pore fluid pressure.In this study, we performed a comparison of the effect of different pore fluids (gas and water) on the permeability of fault rocks. Results show that gas permeability of fault rocks are high due to both the Klinkenberg effect and water adsorption effect of clay minerals, and therefore the liquid permeability of fault rocks cannot be accounted by gas permeability after Klinkenberg correction. We also measured transport properties of fault rock samples collected from the GR fault zone, with water as pore fluid. The results show that the GR fault zone is characterized by a low permeable fault core and host rocks, and a relatively higher permeable damage zone. Under 165 MPa effective pressure, the permeabilities of host rocks and fault core samples range from 10-22 m2to 10-20 m2, and those from the damage zone range from 10-20 m2 to 10-17 m2. Across the fault zone, permeability shows a typical "conduit/barrier" binary structure. Fluids are constrained in the fault core and damaged zone, and tend to migrate parallel to fault plane. Furthermore, permeability structure of the fault zone is highly related to the composition and particle size distributions of fault rocks.High porosity and low permeability of the fault core provide fundamental basis for sealing high pore pressure within the fault zone. Based on the transport properties, we discuss the thermal pressurization processes associated with the Wenchuan earthquake. Results show that at depth of greater than 1.8 km, the principal slip zone of the Wenchuan earthquake is impermeable enough to help build up excess pore pressure. Clay-clast aggregates and injected gouge vein textures observed in the yellow-greenish gouge and blackish gouge serve as evidence. On this basis, we infer that thermal pressurization has played an important role in the dynamic process during the Wenchuan earthquake.