American Society of Civil Engineers.

Transactions of the American Society of Civil Engineers (Volume 81) online

. (page 27 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 27 of 167)
Font size
QR-code for this ebook


solution of the problem of control must be the restoration of the river
to its old channel. Objections to that plan, which have been raised,
are political, and do not apply to the soundness of the plan con-
sidered solely from an engineering standpoint. The inexpediency of
doing any work of that magnitude in Mexico, until the rights and
privileges of citizens of both countries in the matter of use of the
river have been readjusted and defined, may be conceded.

There is no longer the imminent menace of disaster to the Imperial
Valley canal system that made the closure of the Bee River breach in
1910 apparently an emergency job of vital importance. Construction
and adequate maintenance of the line of levees from the old river
bank to the west end of Volcano Lake, together with the provision of
the interior lines of defense described in this paper, will safeguard
the Valley for some time to come.

Therefore, the discussion of methods of closing Bee River and
returning the Colorado to its channel close to the Sonora Mesa does
not connote advocacy of immediate application of these methods. It
is the writer's purpose to show how, in his opinion, such work may
be done more readily and cheaply than similar work has been done
heretofore, and that the costly processes proposed by some engineers
are not essential to successful achievement of the end in view.

Diversion of virtually the entire low-stage flow of the river into
the Imperial Valley canals, by the simple and cheap method herein
described, removes the first difficulty otherwise to be met in making
the closure. It is entirely feasible to make that diversion more nearly
complete, and to turn a larger volume of water from the river into
the intake and through the gates.

Having taken care of the low-stage flow and dried the river bed
at the Bee River break, a levee or earth dam could be constructed



COXTROL OF THE COLORADO RIVER 323

across the break with scrapers, in the same way as levees have been
built along the river. It would be advisable, in the writer's opinion,
to locate this dam nearer the old channel than the site of the dam
with which the Government engineers attempted to close the break
in 1910-11. The Government levee, from the C. D. levee down to
where it was deflected westward near the break, is in fairly good con-
dition, and should be repaired and utilized. A very important factor
in the maintenance of the levee is a railroad on the crest connecting
with the track on the C. D. levee. That is the vital factor in all levee
protection on the Lower Colorado, a fact now recognized by the United
States Reclamation Service engineers, who were unable to maintain
the levee of the Yuma Project on the Arizona shore until they followed
the examiDle of the C. J), levee and placed on it a track connecting
with the rock quarry at Yuma.

It is of primary importance that the Government levee be repaired
and the track laid in time to extend it over the Bee River dam as
soon as the dam has been completed by the scrapers, in order that
rock from the Pilot Knob quarry may be used immediately for the
revetment of the levee and dam.

Changes in the location of the dam, and of the connecting levee
for short distances from both ends, seem advisable. An important
reason for the changes is the resulting reduction of head against the
dam and levees during overflow from the old channel.

If the work was to be done as originally projected, doubtless a
better location than that of the Government levee could be made,
nearer the Sonora Mesa, for the reason that the less room the river
is allowed for meandering, the greater its velocity and scouring
action, and the easier it is to keep it under control. The advantages,
in the case of the Colorado River, of a narrow, confined bed, are
to be seen plainly in the stretch of river from Yuma down to a point
4 miles below Hanlon Heading, and, in the writer's opinion, they
completely refute the theory that the river should be permitted to
meander erratically through the delta. The nearer the river course
can be made to an efiicient hydraulic channel, the simpler becomes
the problem of control.

However, the advantages of a new location may be considered as
more than offset by the advantages of the utilization of work already
done, one of which is the settling, consolidation, and "seasoning" of



324 CONTROL OF THE COLORADO RIVER

the sections of the Government levee remaining. The original dis-
advantage of river side borrow-pits has been minimized to a great
extent by the growth of brush and willows that has choked them. As
time elapses, that obstruction to the flow of water along the foot of
the levee increases. It is essential that, in the restoration of the levee
south of the break, the river face shall be blanketed with gravel and
the track extended to the south end.

The old channel, from the Bee River diversion some 5 or 6 miles
southward, would have to be cleared, and it would be advisable to
shorten the course by cutting across some of the bends, thereby aug-
menting the velocity and the scouring action. In opening the old
channel, it would be advisable to use a dipper-dredge, making a double
cut for about 6 miles. The writer made a reconnaissance of that region
recently, and found outside of the old channel a thick jungle of brush,
but no big growth and very few trees. The work of clearing off the
lines ahead of a dredge would be inconsiderable, as the dredge could
handle the brush. In the old river bed, however, the growth of brush
and trees has been very rapid and heavy, and it would be necessary
to do thorough clearing, and perhaps it would be advisable to break
up root growth and loosen the ground with dynamite.

Given a good, clear start down the old channel, the river in flood
would cut an adequate course for itself.

Having a track on the levee from Hanlon Heading to the south
end, and the quarry at Pilot Knob to furnish unlimited quantities
of rock, loaded directly by steam shovel, revetment could be placed
readily wherever danger of bank-cutting under the levee might develop.
It would not be necessary to revet the entire levee at once. That
work should be done as needed, as part of the maintenance, and
spread over a long term of years.

Conclusions.

In discussing the control of the Colorado River, as related to the
protection of Imperial Valley, most space has been devoted to describ-
ing the hydraulic-fill dam, recently constructed across the Colorado
at Hanlon Heading, and the possibility of its further use in future
river control.

The writer does not mean to state that such a structure could have
been used in stopping the breaks of 1905-06, because of the emergency



CONTROL OF THE COLORADO RIVER 385

nature of such work. It is not presumed that such methods can be
applied in cases similar to those of the past, where the breaks were
stopped while the Colorado Eiver was practically in flood, and where
the resulting closing heads were so great as to prohibit the adoption
of such means as were used in the recent dam.

Necessity for further emergency work to safeguard Imperial Valley
from the flood-waters of the Colorado River no longer exists, and there-
fore it is possible in the future for the engineer to choose his con-
struction period to conform with the low stages of the river. All the
hydrographs of the Colorado River indicate a low period during
August, September, December, and January of each year, with very
few exceptions, and surely the work can be planned in anticipation
of these low stages.

If the engineer will look about him and apply the means at hand
in his work, and will study the characteristics of the stream more
closely, no such expenditures as have been made in the past will ever
again be necessary. The hydraulic-fill dam supplants the trestle and
rock-fill dam previously used, which, of course, represents the greatest
item of expense in the closure. Not only here is opportunity for
great saving of both time and money, but in almost every item of
construction is the same opportunity, if the engineer will profit by
past experiences and study more closely the examples Nature has
provided for him.



336 DISCUSSION ON CONTROL OF THE COLOEADO EIVER

DISCUSSION



Mr. A. L. SoNDEREGGER,* M. Am. Soc. C. E. (by letter). — There are

Sonderegger. ^^ree engineering problems which to-day confront the Imperial Valley
Irrigation District and are more or less related to each other: First,
the protection from overflow of the Colorado River; second, the silt
problem; third, the problem of diverting the low- water flow of the
river during the latter part of the summer, when the greater portion,
or all, of the flow of the Colorado River is required for irrigation
in the valley.

Mr. Allison presents a simple and practicable solution of the first
problem. The efficiency of the proposed system of defense can hardly
be questioned, as it has already had its practical test, at least in
part. The plan includes three main lines of defense.

First line: The diversion of the Colorado River into its old
channel along the Arizona and Sonora Mesa; the closure of the breach
of the Ockerson levee at the head of the Bee River, together with
the restoration of the Ockerson levee, or a new levee similarly situated.

Second line: The westerly portion of the C. D. levee and its pro-
longation, called the Paredones or Holabird levee, and to the west
thereof the Volcano Lake levee.

Third line: The embankment of the Inter-California Railroad as
far west as the xllamo-Mocho Canal, and a new proposed levee along
this canal and westerly thereof to the West Side Main Canal just
south of Calexico.

Of the first line of defense, the upper part of the old C. D. levee,
from Hanlon Heading to its junction with the Ockerson levee, has
withstood successfully the undermining of the river, wherever rip-
rapped with rock. The break in the levee at "House 7", which occurred
in 1914, was due to lack of maintenance. Nevertheless, this first
section of levee forms the weakest portion of the whole system of
defense.

Mr. Allison has referred to the xmdermining of the foundations
of the old concrete piers built on the granite ledge at the Hanlon
head-gate. The writer is quite familiar with these foundations, as
the new Stoney gate was put in, during the winter of 1913-14, under
his direction as Consulting Engineer for Mr. W. H. Holabird as
Receiver. It developed that the granite ledge was a shattered mass
of soft rock with many seams of disintegrated granite and strata
resembling talcum, sometimes several feet in thickness. There was
no need of any blasting, all excavation being made with pick and
shovel. The foundations uncovered uncjer the old piers within the
temporary coffer-dam were badly undermined. In one instance there

* Los Angeles, Cal.



DISCUSSION" ON CONTROL OF THE COLORADO RIVER 327

was a cavity under the do-wn-stream side of a pier large enough to Mr.
contain an office desk. The 8-in. concrete floor-slab was broken °° eregg-er.
off and eroded in many places. Soimdings along the remaining piers
showed that they also were badly undermined, although calculations
regarding their stability revealed a small margin of safety. Lack
of funds prevented the Receiver from repairing the foundations of
these piers, and they still constitute a danger which should be elimi-
nated. A granite quarry is within 1 000 ft. of the heading, and, in
case of a break, rock could be dumped without delay.

From the heading down to the junction of the Ockerson levee
there is only a single line of defense. This section of the levee has
already been revetted for some distance. If the levee is watched, no
danger need develop, as it carries a standard track, and rock can be
dumped at short notice wherever the current approaches the levee.
Some of the revetments were lost in January, 1916, owing to absence
of maintenance.

Regarding the location of the Ockerson levee, the writer agrees
that, as far as river control is concerned, it is of vital importance to
concentrate the flow of the river, and to prevent meandering between
levees.

As regards the proposed method of closure of the Bee River by
a temporary diversion of the Colorado River at Hanlon Heading during
the low- water stage, two facts seem to stand out very clearly: first,
that the construction of a hydraulic-fill dam can be accomplished suc-
cessfully ; and second, that it can be done at very low expense.

The difficult part of this work— the closure of the last gap in the
dam — was effected successfully, not only once, but several times,
when it became necessary to cut the dam for the passage of temporary
floods; therefore, there can be no doubt as to the success of any future
operation of this kind. Both the method and the low cost bfear witness
to the ingenuity and resourcefulness of the directing engineer.

Considerable dredging will be necessary in the old channel below
the Bee River Heading before the river can be diverted permanently.
It is probable that the work will extend over a period of 2 or 3 years
before a channel can be secured capable of accommodating the winter
flow and occasional freshets as they occur; but, unless there are
high waters of an extreme nature, and out of season, the writer believes
that the diversion can be effected without any great risk.

The second line of defense — the C. D. levee, together with the
Holabird and Volcano Lake levees — ^had proved its efficiency for several
years, until in 1914, in the absence of maintenance, a break occurred.
The difficulties encountered in the attempt to close the breach indicated
clearly the necessity of establishing permanent connection by rail
to the Hanlon Quarry. This was done in 1916. This Volcano Lake



328 DISCUSSION ON CONTKOL OF THE COLORADO EIVER

Mr levee to-day presents the only protection for the valley. It is partly

Sonderegger. j-.g^et^ed with rock.

The third line of defense, especially the proposed levee along the
Alamo-Mocho Canal, and westward therefrom, also had a partial test
during the break in the Volcano Lake levee of 1914. At that time
the bank of the Alamo-Mocho Canal intercepted the floods issuing
from Volcano Lake, causing them to spread out in a thin sheet over
the country to the south and west thereof. This canal crosses the
New River channel on a solid embankment, and although it ends about
5 miles to the west thereof, it effectively prevented a concentration
of flood waters for several days. The water flnally found its way around
the end of the canal bank into New River. The levee proposed by
Mr. Allison would perfect the existing system of protection.

During the summer of 1916 a new canal was built from the west
end of Volcano Lake levee, in a northerly and northwesterly direction,
to the intersection of the Wistaria Canal with the West Side Main
Canal. This point is about 3^ miles south of Calexico. This canal
forms the first side of a triangle, of which the Volcano Lake levee
and its extension to the Inter-Califomia Railroad forms the second
side, and the railroad levee and the Alamo-Mocho Canal, with its
proposed westerly prolongation, the third side. The area over which
a flood from Volcano Lake might be spread has been reduced by
the construction of this new canal, and a deflection of flood waters
to the west is made impossible, unless the new canal is cut. The
lowest point of the triangle is at Wistaria Heading. The levee pro-
posed by Mr. Allison, and running from the Alamo-Mocho Canal
westward, is still a necessity, in order to close the third side of this
triangle and thereby make the third line of defense fully effective.

The question naturally arises as to the effect, on the grade
of the river, of the proposed diversion of the Colorado into its old
channel : Eventually, it would affect both the levee protection work
of the Yuma Project and the diversion into the Alamo Canal of the
low-water flow of the river.

From the reports of Col. Ockerson and Mr. Randolph it must be
inferred that in 1911 the slope from the head of the Bee River to
Volcano Lake was about 2.2 ft. per mile, while along the old channel
of the river it was 1.08 ft. per mile. Since 1911 enormous deposits
of silt have been carried into the depression of Volcano Lake, so
that the grade to the lake has probably decreased. However, there
must still be a preponderance of grade in that direction.

It is a fact that the diversion into the lake did cause a recession of
grade extending at least as far up as Yuma, and this fact has undoubt-
edly contributed to the difficulty of diverting the low-water flow into
the Alamo Canal. The trouble has become very acute, not only in



DISCUSSION ON CONTEOL OF THE COLORADO EIVER 329

the case of extreme low waters, as in 1915 when the river fell to a Mr.
discharge of 3 000 sec-ft., but also with moderate volumes, as in 1916, Sonderegger.
when a water shortage occurred in the valley, with a low-water average
of 25 000 sec-ft.

It is to be expected that the restoration of the old channel will
cause a raising of the river bed, especially at low-water stages. On
the other hand, the confining of the flow between the Arizona Mesa
and the levee on the west side may cause a concentration of flow at
high-water stages, and, simultaneously therewith, a scouring.

Since low waters occur during August, the question of securing a
sufficient supply for the valley is second in importance only to the
question of defense. It is, furthermore, intimately related to the
silt problem, and the two questions will be discussed together.

While in the employ of the Receiver of the California Development
Company, the writer, in conjunction with Mr. Allison, made extensive
studies regarding the phenomena which affect the diversion of the
river at Hanlon Heading.

It should be remembered that the granite ledge which carries
Hanlon Heading, is not on the banks of the river, but approximately
2 000 ft. below it. The sills of the old gates are at an elevation
of 98 ft., and of the new Stoney gate, at 93; and the low-water stage
of the river, of late years, has never fallen below an elevation of 104.3 ft.
Theoretically, there has been sufficient fall to effect a diversion, espe-
cially since the erection of the Stoney gate in 1914. At the intake
of the canal, water is diverted at an angle of 90°, which naturally
causes a decrease in velocity from that along the axis of the river.
The result is the formation of a bar at the intake.

During high waters this first section of 2 000 ft. of canal above
the gate operates as a settling basin. It is about 100 ft. wide, and
varies in depth from 3 to 15 ft. With a rising river, the bottom of
the canal rises practically at the same time as the water level of the
river, causing a silting up of the canal to a depth of 10 ft. or more.
Apparently, the cross-section of the canal and the velocity of the water
are regulated automatically, corresponding to the volume diverted
and the quantity of silt carried by the river. The swift currents carry
heavy silt, shingle, and driftwood, large quantities of which are diverted
into the canal, having a tendency to pile up in front of the head-gate.
Desirable as it might be to prevent the admission of these silt-laden
waters, yet the present position of the head-gate makes it necessary
to keep its sill clear, so as to prevent the formation of mud banks and
log jams immediately above the sill. Once such a bank is formed,
it cannot be flushed out, and, with a sudden drop in the river, the
heading might be left high and dry. The practice adopted by English
engineers with Indian canals — providing gates with two or three leaves,



330 DISCUSSIOX ON CONTEOL OF THE COLOEADO RIVER

Mr. which enable them, during high water, to skim the water oflf the top —
SoDfieregger. ^.^jj^q^ ]^q adopted for the Alamo heading with the head-gate in its
present location.

Immediately below the heading the canal enlarges to a bowl shape
with a width of about 200 ft. Here the waters, rushing through the
gate, are stilled, and drop their heaviest detritus, the effect being that
a bar is formed a few himdred feet below the heading. The width
of this bar varies from 1 000 to 2 000 ft. During the high water of
1914, this bar reached an elevation of from 102 to 103 ft.

High waters generally occur in June and July, after which the
river drops suddenly. Experience has shown that, if the sill is kept
clear, the section of the canal above the gate is scoured practically
simultaneously with the drop of the water level of the river, leaving,
however, a bar at the entrance. To dispose of this the 10-in. suction
dredge El Centro is used, and no difficulty has been experienced in
keeping the channel open.

Below the gate, however, the bar is not scoured out by a falling river ;
in such a case, the grade of the intake canal is governed by the water
surface in the river, and the elevation of this bar is independent of
the elevation of the sills of the gate. In fact, the maximum elevation
of the bar has been from 4 to 5 ft. higher than the sills of the old
' miter gates, and from 9 to 10 ft. higher than the sills of the Stoney

gate. The only means of eliminating this bar has been the operation
of the 15-in. suction-dredge Imperial. In 1914 the bar was reduced
to an elevation of 99 ft., sufficient to divert the necessary low-water
flow; but in 1915 and 1916 it has required the raising of the water
level in the river by temporary dams.

The situation, therefore, is as follows: During high waters large
quantities of heavy silt and shingle enter the canal, forming a bar
below the head-gate, the elevation of which depends mainly on the
high-water conditions of the river. With a drop in the river, its
velocities decrease, and the volume of silt carried is reduced corre-
spondingly. The canal then begins to scour in the section above the
gate, lowering the bottom automatically. If the drop in the river is
not too sudden, this scouring keeps pace with the lowering of the
water surface, however, leaving a bar at the intake; but, below the
gate, apparently no scouring action occurs, and it requires the operation
of a dredge to reduce the lower bar, or the construction of a dam to
raise the water level in the river.

In view of these facts, the immediate effect of the restoration of
the old channel of the river on the elevation of the low-water level
at Hanlon, and the problem of diverting the low-water flow, cannot
be predicted with certainty. Undoubtedly, in time, it would be bene-
ficial; but how soon the effect would be noticeable, and whether it



DISCUSSION ON COXTEOL OF THE COLORADO EIVEE 331

would be sufficient to eliminate the present difficulty completely, are Mr.
still open questions. In solving the problem, the most important ^°^^^^^ss^^-
feature is to prevent the heavy silt, which rolls along the river bottom,
from entering the canal; especially at high- water stages, only the sur-
face waters of the river should be admitted.

This can be accomplished by constructing head-works along the
banks of the river at a point where the current will strike the gate
permanently and keep the sill clear.

It is generally admitted that the construction of a permanent dam
near Hanlon Heading is excluded, for many obvious reasons. The
writer agrees with Mr. Allison that it is not advisable to construct
a new heading on the banks of the river on an alluvial silt foundation,
even if provided with a rock-fill foundation and extensive rock-fill
aprons. This would probably be the method of construction to be
adopted if a gate were placed, as has been suggested, at the site of
the well-known break of 1906 which was closed by a rock-fill embank-
ment. It is difficult, if not impossible, to estimate the extent of the
boring action which will take place in front of a gate during an
extreme high water. Undoubtedly, the action is quite different from
that occurring along a revetted bank. Provision would have to be made
for the maintenance of the rock fill, by a system of tracks commanding
the entire up-stream apron and its wings. In order to avoid excessive
vibration and a slow wreckage from such forces, an extremely massive
concrete floor under the gate proper would have to be provided. At
least, a gate in such a location would require unceasing vigil, and
present a constant menace to the valley.

However, a location which is worthy of investigation exists on
the west bank about 1^ miles above Hanlon Heading, and the writer
is informed that this locality is now under consideration by the Im-
perial Irrigation District.



Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 27 of 167)