Geoscience Reference
In-Depth Information
Bunker,
1985
; Teller,
1995
). Glacial Lake Agassiz was larger in area than all the
present Great Lakes combined, and it drained at intervals in the late Pleistocene and
early Holocene (Teller,
1995
).
Late Pleistocene Lake Bonneville overflowed near Red Rock Pass in Idaho about
14,500 years ago. O'Connor (
1993
) estimated that peak discharge amounted to
1.0
10
6
m
3
/second at the Lake Bonneville outlet near the pass. Below the outlet,
stream power (see
Chapter 10
) ranged from 10
1
watts m
−
2
in ponded reaches to
10
5
watts m
−
2
in constricted reaches. The magnitude of this flood is shown in the size
of the cobbles and boulders that were transported and deposited, which ranged from
10 cm to more than 10 m in diameter.
The arroyos and ephemeral stream channels of the American Southwest have been
the focus of nearly a century of detailed investigation (Bryan,
1925a
;Bryan,
1925b
;
Leopold and Miller,
1956
; Schumm and Hadley,
1957
; Bull,
1964a
; Bull,
1964b
;
Lamarche,
1966
; Leopold et al.,
1966
; Tuan,
1966
; Haynes,
1968
; Cooke and Reeves,
1976
; Graf,
1979
; Balling and Wells, 1980; Graf,
1982
a; Graf,
1983a
; Graf,
1983b
;
Graf,
1987a
; Bull,
1991
; Schumm,
1991
, pp. 108-119; Bull,
1997
;Tuckeretal.,
2006
). Bull (
1997
) distinguished between arroyos and gullies, with the latter being
small and ephemeral and the former being up to 200 km long and up to a century
old. In the case of both gullies and arroyos, one of the major issues is why they
sometimes cut down and sometimes deposit sediment along their beds. Many factors
are involved (Cooke and Reeves,
1976
), so that no one explanation will cover all
cases. Bull (
1997
) investigated changes in the balance between stream power and
resistance to erosion (see
Chapter 10
). Because both stream power and sediment
transport rate are roughly proportional to stream velocity cubed (Schumm,
1977
), any
factor that reduces stream velocity will ultimately promote channel aggradation. Bull
emphasised the importance of plant cover in minimising erosion and in promoting
sedimentation within the ephemeral stream network, a conclusion that was confirmed
by the work of Tucker et al. (
2006
) in the semi-arid rangelands of the Colorado High
Plains. If the plant cover began to die, as in time of drought, run-off would become less
diffuse and would become concentrated around the headwalls of the arroyo, resulting
in incision, further drying out of the soil, continued plant death and continued channel
incision. Arroyo bank sections revealed multiple episodes of Holocene incision and
sedimentation, with less than a century needed for complete incision along the arroyo
but more than five times that long needed for complete aggradation (Bull,
1997
). Local
hydrologic factors appear to be at least as important as regional climatic changes in
controlling Holocene arroyo erosion and sedimentation.
Although they might appear to be non-controversial and intuitively valid, the con-
clusions of Bull (
1997
) and Tucker at al. (
2006
) in regard to the influence of changes
in plant cover on sedimentation in arid areas have not gone unchallenged. Antinao
and McDonald (
2013
) investigated four localities at different elevations in the Mojave
and northern Sonoran deserts, and found that the onset of sedimentation on alluvial
×
