Chemical hydrology of subduction zones: Processes, signals and fluid flow
Zusammenfassung der Projektergebnisse
Fluids are the agent for the coupling of rock mechanical, hydrogeological and geochemical processes in subduction zone forearcs. Therefore, the chemical hydrology plays an important role for our understanding of these processes. The overarching goal of this project was to characterize water-rock interaction, its geochemical fingerprints, and fluid flow up to seismogenic depth. As geologic focus sites the Nankai Trough (SW Japan) and Costa Rica subduction zones have been selected because they represent end members of subduction mode (accretionary vs erosion), subduction rate and lithology of incoming sediments. I employed a twofold approach in which hydrothermal laboratory experiments are conducted in addition to analyzing natural pore waters. The main purpose of the hydrothermal experiments were the investigation of geochemical signals of water-rock interaction and clay-dehydration in sediments of the incoming section for the Nankai Trough and Costa Rica margin. I tested 4 natural specimen sampled from sediment inputs to the Nankai Trough and the Costa Rica margin, an illite-rich reference material and re-measured fluids from a smectite-rich standard material previously tested by Hüpers and Kopf (2012). Each sample was loaded up to 70-100 MPa in a heated odometer at constant rate of strain and at temperatures between 20°C to 100°C. To characterize water-sediment interaction I analyzed the major and minor element geochemistry, and B and Li isotopic composition of expelled fluids. Except for the illite-rich reference sample, all samples show fluid freshening, which likely reflects stress induced clay-dehydration. The results show strong variations of sediment-water interaction that is probably a function of clay content. For example, fluids expelled from sediments with high clay content have low Li/B ratios while clay-starved sediments have high Li/B ratios. Observable trends towards isotopically lighter values of B and Li correlate with increasing temperature and on a second order with clay content. Stress-induced clay dehydration seems to have no to little implications for the isotopic composition. A first order comparison with natural fluids suggest that the results may help to constrain the origin of deep fluid sources in subduction zones. In addition to the laboratory study, I investigated B and B isotopic compositions of natural pore waters sampled from major structural regimes of the Nankai Trough forearc. Measured pore water δ11B values range between 20 and 53‰ without any first order trend while B concentration profiles show depletion for all sites down to 600 mbsf. The continuous B depletion can be best explained by ash alteration. However, B isotopic compositions are smaller than expected for ash alteration. This suggests the release of isotopical light B that is probably related to organic matter degradation. In the forearc basin modeled δ11B values for ash alteration fit well with measurements in the upper 200 mbsf. Below, B concentrations increase from gas hydrates. At the tip of the megasplay fault zone, B concentrations and B isotopic composition suggest that underthrust slope sediments may provide a pathway for fluids out of the accretionary prism. The pore water geochemistry of the sedimentary input seems to be influenced by basement alteration. Next to mineralogical evidence for low temperature alteration (<150°C) of the oceanic crust, isotopically light B at the basement-sediment interface may be the result of a hydrothermal fluid component formed at 200-400°C. Such a deep sourced hydrothermal fluids could migrate along fractures in the oceanic crust and ascend along deep sourced faults into the overlying sediment. A second study focused on ash alteration and its implications for physical properties. Within sediments entering the Nankai Trough subduction zone (SW Japan) an up to 240 m thick anomalous high porosity zone (HPZ) exists that has been related to the alteration of volcanic glass. A detailed analyses of logging-while-drilling data reveals that HPZ mudstones are characterized by a large proportion of dispersed volcanic ash (~0.2 to 0.3). However, consolidation experiments of artificial mixtures containing volcanic material have smaller porosities compared to HPZ samples, which suggests that additional secondary porosity is created by dissolution of the volcanic glass. We estimated a dissolved ash content of ~0.15 to 0.30 g/g from the difference in porosity between HPZ and background sediment, and found this amount to be consistent with values from kinetic reaction modeling of ash dissolution. SEM images confirm that the associated precipitation of amorphous silica cement prevents the collapse of the newly formed pore space and occurs only locally near the individual grains, so the cohesive strength of the bulk sediment does not increase. This two-step process depends critically on the abundance of finegrained dispersed volcanic ash/pumice. A first order investigation of other selected subductions zone suggests that HPZs may not be unique to the Nankai Trough.
Projektbezogene Publikationen (Auswahl)
- (2013). Implications of water-clay interaction on pore water geochemistry constrained by laboratory experiments. Abstract presented at 2013 IODP- ICDP Colloquium, Freiberg, Ger., 25-27 Mar.
Hüpers, A. and Kopf, A.J.
- (2013). Anomalously high porosity in subduction inputs to the Nankai Trough (SW Japan) potentially caused by volcanic ash and pumice. Abstract T31G-2591 presented at 2013 Fall Meeting, AGU, San Francisco, Calif., 9-13 Dec.
Hüpers, A., Ikari, M., Underwood, M., and Kopf, A.J.
- (2013). Coupled deformation and dehydration processes in smectite-rich sediments constrained by laboratory experiments. Abstract EGU2013-3310 presented at 2013 EGU Meeting, Vienna, Aus., 7-12 Apr.
Hüpers, A. and Kopf, A.J.
