Geoscience Reference
In-Depth Information
Another threat to this semi-arid landscape emerged in the early part of the twentieth century. Unlike the arroyo
'problem', around which controversy raged as to its cause - overgrazing or climate change or both - this other threat
involved the spread of invasive species such as tamarisk, introduced from Europe in the late nineteenth century
(Graf, 1978). The dense woodlands of phreatophytes that developed on bottomlands and river terraces drew heavily
on shallow groundwater and, hence, were a threat to the water resources of the increasing human population of the
region.
Burkham (1972), reporting as part of the US Geological Survey Gila River Pheatophyte Project, reviewed
channel changes along a 70 km reach that runs through Safford Valley, southeast Arizona, examining archives of
photographs, documents and surveys that dated back as far as 1846 (Figure 13.31). The Gila catchment above
Safford Valley is c. 20 000 km
2
, but the valley also receives water from immediately adjacent ephemeral channels
with a combined catchment area of c. 9000 km
2
. Large floods are associated either with frontal rain on snow in
winter or convectively enhanced frontal storms in summer. Burkham notes that, through the end nineteenth century,
channel width of the Gila in Safford Valley averaged c. 45 m, the banks being well defined and clothed with
cottonwoods, willow and mesquite. Flood-induced widening occurred in 1891 but this was followed by natural
restoration. However, a series of eight winter floods during the period 1905-1917 led to dramatic widening of the
channel to an average of 488 m, with considerable areas of riparian vegetation uprooted and farmland destroyed.
The highest flood peak of this wet period was estimated to have been c. 4255 m
3
/s and flows such as this caused
dramatic straightening of the channel. Sinuosities of 1.2, typical of the nineteenth century channel, were reduced
to unity. Because the flood occurred in winter, Burkham suggests that the sediment load entering Safford Valley
was low and speculates that this was a factor in maximising local erosion. Cottonwood trees, uprooted from the
river's banks, contributed because their root bowls 'took large chunks of alluvium with them'.
The straightening of the Gila channel meant an increase in gradient of c. 20 % and a commensurate increase in
stream power. The floods of 1905-1917 are thought to have set in train sediment waves that progressed downstream
within tributaries and, subsequently, within the trunk channel over the ensuing decades. This provided the material
that, in the absence of major floods in the period through to 1967, accreted as bars, which grew into islands that
were colonised by trees and were eventually attached to the remnant floodplain. Recovery of channel width to
nineteenth century levels took c. 50 years (Figure 13.31), though this was accelerated in places by artificial means
of channel containment and tree planting. Restoration through sedimentation led to increasing channel sinuosity,
values eventually reaching those prior to the 1905 flood. Burkham uses gauge data for the latter part of his study
period to indicate the impact of increasing flow resistance on the celerity of the flood wave, showing this to fall to
about one-third of the value of the flood-straightened channel. The benefits of groundwater recharge that ensued
from greater transmission losses from the slower passage of floods would have been offset in part by the increased
probability of flood-plain inundation.
The history of the Gila River serves to highlight both the fragility of ephemeral channels subject to high magnitude
floods and the long recovery period characteristic of dryland environments, where flood magnitude frequency is
highly skewed (Wolman and Gerson, 1978; Knighton and Nanson, 1997).
References
Alexandrov, Y., Laronne, J.B. and Reid, I. (2007) Intra-event
and inter-seasonal behaviour of suspended sediment in flash
floods of the semi-arid northern Negev, Israel.
Geomorphol-
ogy
,
85
, 85-97.
Alexandrov, Y., Cohen, H., Laronne, J.B. and Reid, I.
(2009) Suspended sediment load, bed load, and dissolved
load yields from a semiarid drainage basin: a 15-year
study.
Ackers, P. and Thompson, G. (1987) Reservoir sedimenta-
tion and influence of flushing, in
Sediment Transport in
Gravel-Bed Rivers
(eds C.R. Thorne, J.C. Bathurst and
R.D. Hey), John Wiley & Sons, Ltd, Chichester, pp. 845-
868.
Alexandrov, Y., Laronne, J.B. and Reid, I. (2003) Suspended
sediment concentration and its variation with water dis-
charge in a dryland ephemeral channel, northern Negev,
Israel.
Journal of Arid Environments
,
53
, 73-84, DOI:
Water
Resources
Research
,
45
,
W08408,
DOI:
10.1029/2008WR007314.
Allen, J.R.L. (1964) Studies in fluviatile sedimentation: 6. Cy-
clothems from the lower Old Red Sandstone, Anglo-Welsh
