Till fabric analysis involves recording the compass orientation and dip of elongated clasts within a till. Generally only clasts with a pronounced long axis relative to the short and intermediate axes are analysed. Suitable clasts are carefully excavated from a cleared face of undisturbed till and the dip or inclination of each particle, along its long axis, is measured using a compass clinometer. The orientation of each particle, in the direction in which the long axis dips, is also recorded with a compass. The data are then plotted in a variety of ways - rose diagrams or equal area stereographic plot - to illustrate the preferred orientation (if present) and dip of the sampled clasts. The statistical distribution of clasts within a sample can be analysed in a variety of ways but it has become common place to do so via an analysis of eigenvectors. This multivariate method defines the mean orientation of the clasts within three-dimensional space and the degree of variability around this mean. It does this via three eigenvectors and three eignen-values, which describe the degree of clustering around the three eigenvectors. Dowdeswell and Sharp (1986) showed how the fabric of different sediment from modern glaciers could be discriminated via plots of eigenvalues S1 and S2 (Diagram A), although Benn (1994) demonstrated how the data could be better illustrated via a ternary plot (Diagram B). This led to many authors attempting to use clast fabrics as a means of assigning a genesis to glacial diamicts. This approach was challenged by Bennett et al. (1999) who argued, on the basis of a large number of fabrics taken from sediments of known origin at modern glacier margins, that there is simply too much overlap between the fabric characteristics of different sediments to allow the genetic fingerprinting of diamicts on the basis of their clast fabric alone. Although not all workers agree with this view, it has led to a re-evaluation of the use of clast fabrics within glacial sedimentology. Although the fabric of glacial sediment can contain valuable information about the pattern or direction of cumulative strain, the key question is how to interpret this in terms of glacial processes.
Sources: Bennett, M.R., Waller, R.I., Glasser, N.F., et al. (1999) Glacigenic clast fabrics: Genetic fingerprint or wishful thinking? Journal of Quaternary Science, 14, 125-35. Dowdeswell, J.A. and Sharp, M.J. (1986). Characterization of pebble fabric in modern terrestrial glacigenic sediments. Sedimentology, 33, 699-710. Benn, D.I. (1994). Fabric shape and interpretation of sedimentary fabric data. Journal of Sedimentary Research, A64,910-5. [Modified from: Bennett et al. (1999) Journal of Quaternary Science, figure 1, p.126]
is normally portrayed on a rose diagram or on a stereograph. A variety of different statistical techniques have been used to describe fabrics, to determine the presence or absence of a preferred particle orientation, and to assess the distribution of particles about the preferred orientation (if present). Lodgement tills are traditionally thought to have a strong fabric in the direction of ice flow and the deviation or scatter of particles about the mean fabric is small. Similarly subglacial meltout and sublimation till should have a strong particle fabric parallel to the ice-flow direction but usually show a greater scatter of particles about the mean orientation. This reflects the fact that individual particles are disturbed as they meltout from the basal ice. Deformation till may possess well-developed particle fabrics orientated in the direction of tectonic transport, usually similar to the ice-flow direction. In contrast, supraglacial meltout and flow tills do not possess consistent particle fabrics, and particle orientation often varies through the deposit. In flow tills, for example, the fabric will reflect the direction of flow, which is normally downslope, and this will change over time as the ice topography changes during ablation. Similarly in supraglacial meltout tills, strong particle orientations may be recorded but these only reflect the orientation of former ice slopes against which the debris accumulated. Consequently, when sampled at several points through these deposits a random or scatter fabric is usually recorded.
Clast shape is also a helpful property in the interpretation of till origin (Table 8.1). As we saw in Chapter 7, debris transported without coming into contact with the glacier bed has very different particle shape and size characteristics from debris transported in contact with the bed. Subglacial tills should therefore contain a high proportion of subglacially transported debris with subglacial characteristics (i.e. rounded, striated, faceted and spherical clasts with a bimodal size distribution; Figure 7.9). In contrast supraglacial tills contain debris transported at high levels within the glacier and possess its characteristics (i.e. angular, non-spherical clasts with a coarse unimodal grain size distribution; Figure 7.6). These tills may also contain subglacial debris due to the upward transfer of basal debris by compression at the glacier margin, although in lower proportions. Particle size sorting within flow till may also be useful in their recognition. In very fluid flows, sediment sorting may occur and dewatering in other flows may give rise to pockets of sorted sediments and occasional layers of silts and sands. In addition, coarse boulder horizons may also be identified beneath some flow packages, formed as a tractive carpet of debris. Clast lithology may also be of some value in distinguishing till genesis, because subglacial tills tend to be dominated by local lithothogies, whereas supraglacial tills often contain a higher proportion of far-travelled lithologies transported on the surface of the glacier or englacially. Subglacial tills often contain evidence of shear, including features such as: clast smudges (brecciated clasts), low-angle shear planes and foliations, sole marks caused by erosion of the substrate and the extrusion of till into the underlying bedrock. In addition, glaciotectonites should contain evidence of folds and faults consistent with shear from overriding ice. In practice, however, there are very few hard and fast rules with which to interpret till genesis on the basis of internal sedimentary properties alone. This problem has become even more evident with the increasing use of micromorphology. Micromorphology involves the extraction of small in situ blocks of orientated sediment, which are then impregnated and thin-sectioned for study using a petro-graphic microscope. A variety of complex and sophisticated recording schemes now exist with which to document the observed structures and interpret these in terms of specific depositional and tectonic processes. This has provided valuable insight into the origin of many glacial deposits, but has also demonstrated that most glacial diamicts are formed by a combination of processes, thereby challenging traditional till classifications.
In many situations, the external setting or the context of a till unit is often more useful. In particular the association with surface landforms can be particularly diagnostic. For example, subglacial till is likely to be found in association with subglacial landforms such as drumlins and flutes (see section 9.2.1). In contrast, supraglacial tills such as flow or supraglacial meltout till are likely to be associated with areas of hummocky moraine (see section 9.1.3). Consequently the geomor-phological context of the upper boundary or contact of a till unit may provide important insight into its origin. More recently the introduction offacies analysis has added further criteria. This approach is based not on the interpretation of individual units, but on the interpretation of the complete depositional sequence. It rests on the premise that most depositional environments can be characterised by distinctive associations or combinations of sediments or facies (lithofacies) and the bounding surfaces between them (Box 8.3). Sedimentary facies are bodies of sediment that are the product of a particular depositional environment or process and the relationship of one facies to another gives us the ability to assemble a picture of the depositional environment in which sedimentation occurred. At a modern glacier one deposit does not continue infinitely in any one direction, but will grade into other deposits. For example, a subglacial till surface may be dissected by a melt-water stream in which glaciofluvial sediments are being deposited. By studying the relationships between one deposit and the next we can reconstruct the depositional environment.
Was this article helpful?