In the numerical modeling and experimental studies conducted to investigate the development of magnetic fabric during progressive deformation, the underlying theme during the last quadrennium was continuation of a quest from previous quadrennia, the accurate quantification of rock strain with magnetic fabric intensity. Throughout this work, though, it became apparent that magnetic mineral composition of rocks and sediments was an important control on the intensity of magnetic fabric.
Borradaile and colleagues conducted a series of laboratory deformation experiments during the past four years designed to understand the effects of rock deformation on the remanence and magnetic fabric of magnetite-bearing sandstones [ Borradaile and Mothersill, 1991] and limestones [ Borradaile, 1991]. He found rotation of the pre-deformation remanence vector away from the shortening axis, but more significantly an increase in magnetic coercivity with deformation, possibly due to fracturing of magnetite grains since no pore fluid had been used in these experiments. Borradaile continued this type of rock deformation laboratory experiment with U.S. workers [ Jackson et al., 1993], but both dry and wet synthetic, magnetite-bearing calcite sandstones were deformed. This important and thorough set of experiments indicated that magnetic remanence reorientation and AMS fabric development are controlled by bulk strain due to reorientation of particles, whereas AAR fabric development and coercivity changes during deformation are caused by intragranular stress between particles. The most significant result of this set of experiments for magnetic fabric studies is the potential importance of magnetic mineral grain deformation for controlling magnetic fabric development. The effects of deformation on the coercivity, and hence the magnetic grain size distribution of rocks, were further demonstrated by Borradaile and Jackson's [1993] laboratory experiments in which dry granular rocks were compacted to pressures equivalent to 8 km of burial. Two different grain sizes of magnetite were used in these experiments and intragranular stresses were found to demagnetize the larger, lower coercivity magnetic grains. This result may suggest that in regions of low geothermal gradient the low coercivity components of NRM could be demagnetized naturally by rapid sedimentary burial.
Richter et al. [1991] also investigated the effects of pure shear on the development of magnetic fabric, but they used clay-water dispersions which either did or did not contain magnetite. In the samples which didn't contain magnetite they found that the mineralogy of the clay, the concentration of paramagnetic minerals, and the initial fabric controlled the final intensity of magnetic fabric, rather than the amount of strain. In clay-water dispersions that contained magnetite, the magnetite was the dominant carrier of the magnetic fabric, but significantly, the concentration of magnetite did not affect the intensity of the magnetic fabric that developed. Johns and Jackson's [1991] experimental deformation work also showed that in paramagnetic and ferrimagnetic mixtures the ferrimagnetic mineral will dominate the resulting magnetic fabric, but that the grain shape (individual grain anisotropy) of the ferrimagnetic mineral will be an important control on the final magnetic fabric.
The experiments reviewed above have concentrated on the effects of pure shear on remanence rotation and magnetic fabric development. Kodama and Goldstein [1991] examined the effects of simple shear strain on remanence in silicone putty-magnetite or kaolinite-magnetite mixtures and found experimental evidence for individual magnetic particles rotating actively [ Jeffery, 1923], rather than passively [ March, 1932] in simple shear since the remanence can pass through the shear plane. These results suggest that the magnetic fabric developed during simple shear strain should be interpeted differently than fabric developed during pure shear (see Ramsay and Huber [1983] for definitions of pure and simple shear).
Numerical modeling of magnetic fabric development reinforces the notion that the magnetic strain gauge can, theoretically, be achieved, but that the strong effect of magnetic mineral composition can be paramount in all but the simplest cases. Richter [1992] developed numerical models to investigate the quantitative correlations between magnetic fabric and preferred mineral orientation for single magnetic mineralogies undergoing coaxial deformation. He finds a simple log-linear correlation for strains less than 200%. For higher strains the AMS fabric will approach the individual particle anisotropy. Housen et al. [1993] investigated multiple mineralogies, and hence composite fabrics, numerically and found that to fully unravel composite fabrics in deformed rocks the anisotropy degree and relative susceptibility of each component must be accurately determined. If a rock contains only two planar fabrics, then the maximum AMS axis will be the intersection of those fabrics and the shape of the susceptibility ellipsoid will depend on the relative intensity and angle between the two fabrics.