5.4 PALEOCURRENT ANALYSIS
169
tion. Examples of the second group include pebble imbrication (see Section 3.2.3.1.4),
cross-lamination, cross-bedding, slump folds, flute marks, and the asymmetric profiles
of ripples. The measurement of the orientation of sedimentary structures must be done
with care. Ideally some kind of areal sampling grid should be used for regional paleocurrent mapping. In practice, this ideal approach is commonly restricted by limitations
of access, exposure and time.
Each sample station will generally consist of a cliff, quarry, stream section, road cut,
etc. If it is to be worth anything, paleocurrent analysis must be integrated with a full sedimentological study. Thus each sample station will also be the location of a measured
section, or at least some detailed notes on stratigraphy, lithology, facies, and fauna. At
each station it is necessary to record structural dip and strike. If it is excessive (greater
than about 10~ each measurement must be corrected on a stereographic net. The orientation of the structures will be recorded, including both the azimuth and dip of planar
structures that need correction. For linear structures and for planar structures in outcrops of low tectonic dip only the azimuth need be recorded. At the same time, it is necessary to note the type and scale of the structure and the lithology in which it occurs.
Foreset dip directions from cross-bedding should always be measured in plan view. Dip
directions seen in vertical sections should only be recorded as a last resort. There are
two reasons for this. First, as pointed out earlier in this chapter, cross-beds do not always
dip directly downcurrent. In troughs and laterally infilled channels, foresets are deposited obliquely or perpendicular to current flow. Examination of cross-bedding in plan
view gives a clue to the structural arrangement of the foresets. If cross-bedding is measured from vertical facies generally only an apparent dip can be recorded. This may
diverge considerably from the true dip direction, especially if there is well-developed
jointing. The discrepancy will not be too erroneous, however, as foresets appear horizontal when viewed normal to their dip direction (Fig. 5.14).
The number of readings that need to be measured at a sample station is a matter for
debate and may fortunately be dictated by the size of the exposure. Discussions of the
statistics of sampling are given in Miller and Kahn (1962) and Krumbein and Graybill
(1965). There is great scope here for statistical aerobics. As a rule of thumb in unipolar
cross-bed systems, as in alluvium, 25 readings are generally sufficient to determine a
vector mean with an accuracy of +_30 ~ . This is sufficiently accurate for most purposes.
Many more readings may be needed, however, to establish well-defined modes in a section of interbedded facies with different and often polymodal vectors. For example, in
shoreline deposits, fluvial channel sands with unimodal downslope dipping foresets may
be interbedded with marine sands with bipolar dipping foresets due to tidal currents
unrelated to the paleoslope.
5.4.2 Presentation of Paleocurrent Data
Paleocurrent data may be entered in a field notebook and subsequently published in
tabular form. The azimuths are, however, generally manipulated in some way to make
their interpretation easier. The first step involves the removal of tectonic dip on a stereographic net where applicable (Schlumberger Ltd., 1970; Potter and Pettijohn, 1977,
p. 371; Lewis and McConchie, 1994, p. 87). Then the azimuths are divided into class intervals from 0 to 360 ~ Class intervals of between 30 and 45 ~ are about typical. The data
may then be presented on a histogram. More usually, however, a compass rose is used
169
tion. Examples of the second group include pebble imbrication (see Section 3.2.3.1.4),
cross-lamination, cross-bedding, slump folds, flute marks, and the asymmetric profiles
of ripples. The measurement of the orientation of sedimentary structures must be done
with care. Ideally some kind of areal sampling grid should be used for regional paleocurrent mapping. In practice, this ideal approach is commonly restricted by limitations
of access, exposure and time.
Each sample station will generally consist of a cliff, quarry, stream section, road cut,
etc. If it is to be worth anything, paleocurrent analysis must be integrated with a full sedimentological study. Thus each sample station will also be the location of a measured
section, or at least some detailed notes on stratigraphy, lithology, facies, and fauna. At
each station it is necessary to record structural dip and strike. If it is excessive (greater
than about 10~ each measurement must be corrected on a stereographic net. The orientation of the structures will be recorded, including both the azimuth and dip of planar
structures that need correction. For linear structures and for planar structures in outcrops of low tectonic dip only the azimuth need be recorded. At the same time, it is necessary to note the type and scale of the structure and the lithology in which it occurs.
Foreset dip directions from cross-bedding should always be measured in plan view. Dip
directions seen in vertical sections should only be recorded as a last resort. There are
two reasons for this. First, as pointed out earlier in this chapter, cross-beds do not always
dip directly downcurrent. In troughs and laterally infilled channels, foresets are deposited obliquely or perpendicular to current flow. Examination of cross-bedding in plan
view gives a clue to the structural arrangement of the foresets. If cross-bedding is measured from vertical facies generally only an apparent dip can be recorded. This may
diverge considerably from the true dip direction, especially if there is well-developed
jointing. The discrepancy will not be too erroneous, however, as foresets appear horizontal when viewed normal to their dip direction (Fig. 5.14).
The number of readings that need to be measured at a sample station is a matter for
debate and may fortunately be dictated by the size of the exposure. Discussions of the
statistics of sampling are given in Miller and Kahn (1962) and Krumbein and Graybill
(1965). There is great scope here for statistical aerobics. As a rule of thumb in unipolar
cross-bed systems, as in alluvium, 25 readings are generally sufficient to determine a
vector mean with an accuracy of +_30 ~ . This is sufficiently accurate for most purposes.
Many more readings may be needed, however, to establish well-defined modes in a section of interbedded facies with different and often polymodal vectors. For example, in
shoreline deposits, fluvial channel sands with unimodal downslope dipping foresets may
be interbedded with marine sands with bipolar dipping foresets due to tidal currents
unrelated to the paleoslope.
5.4.2 Presentation of Paleocurrent Data
Paleocurrent data may be entered in a field notebook and subsequently published in
tabular form. The azimuths are, however, generally manipulated in some way to make
their interpretation easier. The first step involves the removal of tectonic dip on a stereographic net where applicable (Schlumberger Ltd., 1970; Potter and Pettijohn, 1977,
p. 371; Lewis and McConchie, 1994, p. 87). Then the azimuths are divided into class intervals from 0 to 360 ~ Class intervals of between 30 and 45 ~ are about typical. The data
may then be presented on a histogram. More usually, however, a compass rose is used