Geoscience Reference
In-Depth Information
they are not fed from upslope rills) with localised mud-
flows. Shrink-swell processes lead to the development of
a typical 'popcorn' surface. Interbedding may lead to os-
cillations between these types before leading to a basal
pediment, perhaps with small fans at the break in slope
and distributary channels building up a surface of very
thinly bedded units. At Dinosaur Provincial Park in Al-
berta, Kasanin-Grubin and Bryan (2007) evaluated differ-
ences between rills on mudrock and sandstone bedrocks.
They found that while rill systems on the latter remained
fairly static over a two-year period, those on mudrock
varied significantly, both in terms of morphology and the
extent to which a 'popcorn' layer was present. Structural
controls such as the locations of tectonic joints and bed-
ding planes have been demonstrated as being significant
in controlling the position and magnitude of soil pipes
by both Torri and Bryan (1997) and Farifteh and Soeters
(1999) (Figure 10.8).
These same controls are also important in controlling
the style and extent of mass movements. In areas where
the bedrock is thinly bedded and dipping, shallow trans-
lational slides can be significant (see the discussion in
Howard, 2009), and the orientation of the dip relative to
gullies can lead to distinct asymmetries in morphology, as
bedrock dipping with the side slope is much more likely to
fail than bedrock dipping into the side slope. That gullies
may pick lines of weakness following dipping bedrock
means this pattern is not uncommon. Farifteh and Soeters
(2006) suggested that the size and spacing of joints and
faults in basilicata was responsible for the patterns of bian-
cane (small, isolated) rather than calanchi (large, gullied)
badland systems. Griffiths
et al
. (2005) noted that land-
slides in badlands in the Aguas basin in southern Spain
were more likely to occur on specific lithological bound-
aries (e.g. 39 % of all landslides where multiple litholo-
gies were present occurred where the lower unit was a
calcareous mudstone). Elsewhere in their study area, ma-
jor landslides were found in areas of anomalously rapid
uplift.
Climate has been mentioned above in the ways it
controls aspect-related differences in badland evolution
through both insolation and frost action, as moderated
through the action of vegetation. The ability of differ-
ent types of vegetation to be sustained in different cli-
matic settings also has a significant impact on the de-
velopment of badlands in drylands. Climatic variability
over longer timescales can have a significant impact on
badland formation. Bryan, Campbell and Yair (1986) sug-
gested that in the Dinosaur Provinical Park in Alberta, bad-
land evolution has followed broad phases of incision and
filling relating to relatively wetter and drier phases through
North
ALL LINEAMENTS
(NO RESTRICTIONS)
5° INTERVALS
350
0
10
340
20
330
30
LENGTH IN
KILOMETRES
320
40
310
50
300
60
290
70
280
80
5
West
East
60
48
36
24
12
80
280
70
290
60
300
50
310
FREQUENCY
320
40
30
330
20
340
10
0
350
N
Figure 10.8
Rose diagram and spatial pattern of lineations
passing through soil pipes in badlands in Basilicata, Italy,
demonstrating the occurrence of preferential spatial patterns
related to structural controls at the landscape scale (after
Farifteh and Soeters, 1999).
2004, and similar arguments for the evolution of southern
French badlands in Descroix and Gautier, 2002), although
Evans (2000) has demonstrated that this landscape is also
strongly controlled by inherited features from past glacial
and periglacial processes. Initial formation of the South
Dakota badlands seems to have been triggered by gully-
ing as a response to base-level change (Mather, Stokes and
Griffiths, 2002; Howard, 2009), and base-level change has
also been implicated in Spanish gully systems (Griffiths
et al
., 2005), so that large-scale climate variability can
