American Society of Civil Engineers.

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

. (page 107 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 107 of 167)
Font size
QR-code for this ebook

held back, these ditches must be about 3 ft. deep and 3 ft. wide,
and require from one-sixth to one-fourth of the total land surface
thus treated. At 30 ft. apart, about 1 450 lin. ft. would be required
per acre, or about 928 000 lin. ft. per sq. mile. The cost would be
from $13 000 to $26 000 per sq. mile. Such works favor irrigation by
keeping up a dampness in the soil and increasing the supply of ground-
water. An advantage of this method is that it may easily be carried
out by private individuals. It is practised by the farmers of the
Cevennes Mountains, in France, and in Tunis.



Mr. The last paragraph under (e) Levees, on page 1226, may be mis-

leading. Unless its exact meaning is studied carefully, it may give
the impression that the decrease in valley storage, by confining the
water between the levees, had been lost sight of. The comparison
between possible methods of protection may be made clearer if it is
granted, in the first place, that the lands behind the levees must be
given the same protection by each method considered.

Levees of suitable size will protect the low lands behind them. The
volume of water stored between the levees will be very great. To give
the same protection to these lands without levees would require that
this water be stored elsewhere.

It might be stored in a channel excavated to give the same cross-
sectional area as that provided between the levees. This would gen-
erally be much more costly to build and maintain than the levees.
There might be serious complications at the mouths of the tributaries
and at the mouth of the river. The ground-water conditions might
be very harmfully affected. The added cost at bridges would be
very great.

It might be stored on the drainage area above, if sites could be
found. The total capacity of such storage basins would have to be
much greater than the storage space between the levees, in order to
take care of variations in distribution of flood rainfall, and to make
up for the reduction in the effect of storage that takes place in
passing down stream.

One statement in the paragraph alluded to may be questioned, unless
the writer has not fully gathered its meaning, namely, at the end
of the first sentence: "the resultant reduction of flood height from
this cause". It is unquestionably true that the artificial storage be-
tween the levees is not so great as the natural storage that was avail-
able over the lowlands along the river before the levees were built,
and that this reduction in storage raises flood heights. At the same
time, the contraction of the flood flow between levees raises flood heights
at and up stream from the stretch of river that has been leveed. When
adequate levees are built to above this increased flood height, and
properly protected and maintained, they afford one of the surest and
safest means of flood protection.

In view of the action of levees in reducing natural storage and
thus increasing flood heights, especially in the river below, proper pre-
cautions should be urged that such improvements be not made on
any reach of a river until adequate provisions have been made at
other points to take care of the changed conditions of flood flow. There
are points here which can well be considered in discussion of the
Report of the Special Committee on A National Water Law.*

• Proceedings, Am. Soc. C. E., for December, 1915, pp. 2747-51.


The French water laws specifically name what rivers or stretches of Mr.
rivers may not be leveed to above maximum flood stage. On the Loire,
wide overflow spillways have been provided on the levees to allow tem-
porary storage in the lowlands behind them during extreme floods.
It has for years been a subject of discussion and investigation in
Austria and Germany as to whether the levees could not be altered so
as to keep out only the lower floods of the growing season, and allow
the great winter and spring floods to overtop the levees and spread over
the lowlands. The reduction of flood heights thus obtainable is one
argument in favor of such treatment; another is the fertilizing action
of these silt-laden waters on the lowlands behind the levees, and great
stress is laid on this feature in all reports on the subject. The over-
flow areas in the German river valleys are to-day only about one-fourth
to one-fifth of their natural and original extent.

The natural storage effect on the overflowed lowlands along a river
may be very considerable, and may have an appreciable effect in lower-
ing flood heights. In the Miami Valley above Dayton, the valley storage
in the 1913 flood amounted to about 300 000 acre-ft., or 32% of the total
run-off from the 2 500 sq. miles above Dayton during the maximum 3
days of the flood.

On some rivers a disadvantage of building levees reaching to above
high water is that the lowlands behind the levees are no longer raised
by deposits, and sometimes even sink lower; and the river itself may
gradually raise its bed by deposits in its middle and lower course, where
the slope is flat. These lowlands can even gradually become swampy,
and the pumps formerly used for freeing them of water in the spring
only will have to be kept at work longer. This is especially to be feared
in the lowlands on low deltas, which are gradually extended into the
sea by deposits of silt, so that the raising of the water surface in the
ri^er is necessary, in order to create the fall required for discharge
through the longer channel.

On the other hand, the narrowing of the channel by levees, by in-
creasing the velocity, is oftentimes not only favorable for transportation
of silt, but indeed forms the only available means for preventing a
raising of the river bed. Without the levees, the banks and land im-
mediately bordering them would be gradually raised, and lowlands
farther back would remain low; and sometimes rivers would break
through and seek a new channel in the lowlands, with disastrous results.
This danger is especially great in the upper courses of rivers, where the
slope is steeper. The levees reaching to above high water there form
the only effective means of preventing this.

In channel improvement by deepening and straightening, due con-
sideration should be given to the effect on ground-water conditions.
On the Weser River, in Germany, agricultural interests have required


Mr. large expenditures in order to restore the ground-water conditions that
■ obtained before the channel was deepened.

It is important, in connection with later plans for channel improve-
ment, that some suitable agency be authorized to exercise control over
the placing of obstructions in and along streams. The cost of adequate
channel improvement in some localities is greatly increased, and some-
times made prohibitive, by structures that should never have been al-
lowed to be built in their present location or to their present dimensions.

On some rivers, particularly where melting first takes place on the
upper river, as on the Oder and Weichsel in Germany, the formation of
ice jams is a frequent cause of iloods. The design of improved channels
on such rivers should take this into consideration. In special cases
ice-breaking steamers may be found necessary to prepare for spring
freshets from the upper river.

Detention basins as a means of flood control are stated by the report
to be successful in certain localities. This method, as yet little used
in America, should perhaps be discussed in more detail. There are
localities where this is the only kind of storage feasible. The agri-
cultural lands may be too valuable to be dedicated to permanent storage.
Floods may not be confined to any one time of year, so that where the
drainage areas controlled are small and floods collect too rapidly to
permit the emptying of a partly full reservoir in time to provide the
required flood storage space, all the capacity intended for flood control
must be kept empty, and ready for immediate use. In such cases, if
the entire capacity available is required for flood storage, a "dry"
reservoir is the only type feasible. Detention basins, as compared with
reservoirs, have the disadvantage of wasting the stored flood water ; and
in some localities this may be a very important consideration; but,
where the foregoing conditions prevail, and where flood control is a
vital enough issue to warrant their construction, they will sometimes
be found the best solution of the problem.

Where assurance can be had of competent and permanent super-
vision and maintenance, detention basins can be made much more ef-
fective by providing artificial control. Where there are no gates in the
outlet conduits, they must be designed so that the maximum flood will
just reach the spillway when the inflow has fallen to the same rate as
the outflow. This, however, gives a much less degree of control over
smaller floods than could be obtained by the manipulation of gates in
the outlet conduits.

The great advantage of detention basins is that they do not remove
the land in the basins from cultivation. In the Miami Valley, where
the conditions for this method of flood control are particularly favor-
able, the total area flooded in the five proposed basins, with a flood like
that of 1913, would be 29 850 acres. More than 60% of this same land
(18 700 acres) was actually flooded in 1913. After such a flood all the

Fig. 1. — PiNAY Dam on Loike River About 7 Miles Above Roanne, France.

Fig. 2. — Typical Example of Work of Reforestation and Mountain-Torrent
Control on Head-Waters of Elbe River.


reservoirs except one would empty again in a week, and that one would Mr.
take about 3 weeks. In smaller floods, they would flood less land and
empty much more rapidly; and, even in extreme floods, a large part of
the area would be flooded for only a few days. The area actually flooded
in 1913 in the Miami Yalley above Hamilton was 63 000 acres, or more
than twice as much as would have been flooded in the detention basins
if they had been in operation. Of this flooded area, 13 000 acres were
in cities and towns.

Fig. 1 shows the Pinay Dam, on the Loire River, about 7 miles
above the City of Eoanne in France. This is one of the most interesting
examples of a detention basin in Europe. It was built by the French
Government in lYll, and has since repeatedly saved the city from

It should be pointed out, in discussing reservoirs, detention basins,
and levees or other channel improvements, that in many cases no one
method may provide an adequate solution of the flood problem, but
the best plan will consist of a combination of two or more methods.
It may be found unfeasible or uneconomical to reduce flood flow to
existing channel capacities by reservoir or detention-basin control.
It may be fotm.d equally unfeasible or uneconomical to provide an
adequate channel to accommodate maximum floods. Protection by one
method is best carried out up to the point where the remaining neces-
sary protection is cheaper by some other method.

In discussing the function of barriers in mountain streams, their
action in increasing soil absorption should be included. Under certain
conditions, as in Los Angeles County, California, this may be very
considerable, and may bear an important relation to other phases of
the water problem.

The important action of barriers in holding back sand, gravel, and
boulders from improved channels and reservoirs below should be
brought out more definitely. In some of the comprehensive river im-
provements of Europe, reforestation and Wildbachverhauung form an
integral part of the reservoir systems and channel improvements, and
are regarded as essential to the maintenance of such works on the
lower streams. The River Regulation Commission of Bohemia has
been carrying on a remarkably well co-ordinated work of this character
on the upper Elbe, including reforestation, barriers on mountain tor-
rents, flood-control reservoirs and channel improvements.* Fig. 2
shows a typical example of the work of reforestation and mountain
torrent control on the head-waters of the Elbe.

The report makes no reference to the method of reducing floods
by spreading over permeable strata. It would be helpful if it would
take up the advantages and disadvantages of this method and its re-

• Engineering News, October 15th, 1914.


Mr. lation to underground storage, and make some reference to localities
' where it has been or might be applied successfully.

Desynchronization of tributary flood peaks is an effective means
of flood prevention in particular cases. It will generally be insufficient
by itself, but will lessen the work to be done in connection with other
methods with which it may be combined. In mountain streams,
tributary flood peaks that reach the main stream at or after the pas-
sage of the flood peak of the latter can be retarded further by barriers
or small detention basins, and those that arrive early can be accelerated
by the improvement of their channels.

The same process on a large scale can sometimes be carried on with
the larger streams, farther down. A flood wave can be made to travel
twice as fast without any increase in the average velocity of the water,
by establishing the proper cross-section. It can be demonstrated that,
with a parabolic cross-section the speed of propagation is approximately
four-thirds of the average velocity. Tributaries that deliver their flood
peaks early can thus be further accelerated without any increase in
erosive action. Other slower tributaries can be further retarded,
where feasible, by reservoirs or detention basins, or by wide overflow
areas along the banks.

Obviously, in solving the flood problem of any river basin, it
is very important to avoid the more perfect synchronization of tribu-
• tary flood peaks that might be brought about by reversing the treat-
ment just outlined. It would be dangerous, for example, to construct
detention basins or provide overflow areas on tributaries delivering
their flood peaks early, unless the basins were large enough and the
outlets small enough to reduce the flood wave sufficiently to preclude
any possibility of aggravating the flood conditions on the main stream.
In the same way, as a general principle, it would be unwise to speed
up the flood wave on a slow tributary.

The report might well include some reference to flood prediction
and its value in preventing or reducing flood damage. This is a
measure of relief that can be applied at once, pending the construc-
tion of flood prevention works. It has been very thoroughly worked
out in some European countries, but has not yet been developed to its
full possibilities in the United States.

The figure quoted in the report for the flood losses in 1913 in the
Ohio Valley is much too small. The direct property loss in the Miami
Valley alone was about $70 000 000, and to include the loss through
depreciation would greatly increase this figure.

The writer joins most heartily with the Committee when it urges
more thorough, intensive, and co-ordinated studies of matters relating
to flood control. Such investigations should include all data neces-
sary for the solution of problems of water conservation and stream
control, in order that flood prevention works may be co-ordinated with


other uses of the streams. Enormous appropriations for river regu- Mr.
lation and water conservation have been urged before Congress for
some years. If these appropriations were to be made to-day we would
not have the data with which to design, intelligently and economi-
cally, the works to carry out the purposes thereof. The proponents
of such measures would do far more good in the development of the
country's water resources if they fathered successfully a bill appro-
priating several million dollars a year for the use of the present Gov-
ernment agencies now struggling to collect data with pitifully inade-
quate funds for operation.

In nearly every problem of river control attacked by the engineer
in America, he finds himself confronted with almost a blank wall of
lack of data, and is obliged to make safe assumptions that greatly
increase the cost of the work. River gauging stations are too widely
scattered, and lack of funds prevents their efficient operation. At
many of them the gauge is read but once daily, so that there is no
complete record of flood peaks, as they frequently occur between
readings. The number of rainfall stations should be very largely
increased, and hourly readings should be taken during rain storms.
It is not possible, with the present methods of collecting and recording
rainfall data in the United States, to determine successfully the rela-
tion between rainfall and run-off.

The writer would be glad to see the report make definite recom-
mendations to remedy this condition.

B. F. Groat,* M. Am. Soc. C. E. (by letter). — A reading of the Mr.
Report of the Special Committee on Floods and Flood Preven- ^^^^ '
tion, the Minority Report by Morris Knowles, M. Am. Soc. C. E., and
the subsequent discussions by Messrs. Eakin and Hill, discloses the
fact that the report does not seem to cover all the ground expected, and
that possibly suggestions in more definite language as to general
policies and methods would be acceptable.

It appears to the writer that the whole question of flood control is
necessarily indefinite at present and probably will so continue for a long
time. The indefiniteness and uncertainty of such a report, therefore,
is a natural and certain consequence of the circumstances and present
state of knowledge of the subject, rather than a result of any neglect
or lack of thoroughness on the part of the Committee. Under the
circumstances, one can scarcely expect a committee to devise revolu-
tionary constructive policies, or express conclusive opinions, concerning
a question which has perplexed all the rest of the world since the idea
of national flood control developed and raised itself from a special to a
general engineering problem, intimately related to that of national
conservation. At least, this is the opinion of the writer, who is more
or less of an exoteric with regard to this particular subject.

* Pittsburgh, Pa.


Mr. Mr. Knowles' objections to the report appear to be based largely on

phraseology which he considers to be single-viewed, and to partake, so
to speak, of the prejudice of the Committee. The writer believes that
a. committee report is no place for the individual opinions of the mem-
bers, except when expressly stated to be such, and that where con-
troversial matter is involved, the evidence on both sides should have
equal publicity. Engineering questions are not to be settled as are
questions of law — by arbitrary rule — but they should lead to the
determination of a rule, or procedure, which shall meet with either
general approval or the specific approval of the Engineering Profession.

As regards breadth of view, there does not appear to be much room
for exception, and it must be observed that, with the amendments
suggested, the report would be even more indefinite than vrithout, as
should be expected in the case of this subject, except that Mr, Knowles
does ask for additional kinds of data, as well as an increase in the
quantity of such data as are now regularly observed by various agencies.
The Government agencies might increase the variety by adding such
information as data on erosion and transportation, precipitation for
various altitudes at the same place, and for different places and circum-
stances, especially in higher altitudes and in sparsely settled districts.
Some systematic method might be recommended for the study of forest

The most important policy suggested by Mr. Knowles is that briefly
outlined in the last paragraph on page 1234. It appears to be admitted
by all that the whole question of floods and flood prevention is of vital
importance, and demands special efforts and expenditures commen-
surate with the difficulties encountered. Why not, therefore, have a
"special agency, supported by adequate appropriations, for the purpose
of studying stream regulation in its largest sense?"
Mr. H. M. Chittenden,* M. Am. Soo. C. E. (by letter).— The writer

does not fully understand the purpose of a report of this character,
whether it is simply to remain as the findings of the Committee, or,
after discussion, to be adopted, with possible modification, as an
official expression by the Society. There are certain reasons why the
latter course, if practical unanimity could be arrived at, would seem,
at the present time, to be particularly desirable. From this point of
view, the writer will contribute his mite toward the further elucidation
of certain features of the Committee's able treatment of the subject.

Terminology. — First, as to terminology and verbiage: Ought not
** control" to replace "prevention" in the title? Prevention is not broad
enough. Levees, for example, do not prevent floods (they actually in-
crease gauge height and discharge), but they control them. Reser-
voirs prevent, or tend to prevent, floods. The comprehensive word is
"control," with a logical subdivision into measures for flood forecast or

* Gen. Chittenden, who lived at Seattle, Wash., died since writing this discussion.


svarniBg, measures of prevention, and measures of protection. The Mr.
distinction between prevention and protection is well brought out in ' ®° ^°'
the reiK)rt of the Pittsburgh Flood Commission.

On page 1219, under "Suggested Methods", Subdivision (a), on
the first line, should not "rainfall" be "run-off"? Man cannot reduce
"extreme variations in rainfall", as the Committee correctly states in
the last paragraph on this page, but he may reduce extreme variations
in the results of rainfall.

Would it not be well for the Committee to recommend the definite
adoption of one or the other of the equivalent terms, "drainage area",
and "water-shed". The term water-shed was formerly used, and occa-
sionally is still used, to signifiy "divide"; but its use is now almost
universally confined to a designation of the territory drained by the
particular stream under consideration. Probably no confusion results
from the use of both terms, but simplicity of terminology is always
desirable. The writer favors the term water-shed.

Several designations have been applied recently to the compara-
tively novel system of flood prevention which is the distinguishing
feature of the great flood-control project being worked up by the Miami
Conservancy District. These include "dry reservoirs", "detention reser-
voirs", "detention basins", and "retarding basins". When the writer
was preparing his paper on Flood Control for the recent International
Engineering Congress in San Francisco, he gave this matter some
little consideration. Inasmuch as reservoirs may properly be classed
in two distinct types, dependent on their purpose and functioning —
those designed to store water for future use, and those designed to
detain it temporarily for flood prevention only — it seemed wholly
logical to use "reservoir" (Latin, re servare, to hold back) as the
generic term, so to speak, and "storage" and "detention" as specific
designations. It was for this reason that he felt particular about
adopting "reservoir" as against "basin". The Miami Conservancy Dis-
trict has definitely adopted "retarding basin", and the prominence of
this example ought, perhaps, to settle the question.

Effect of Human Occupancy of Soil. — The writer feels that the
Committee might advantageously have been more definite and positive
in its discussion of this subject, particularly as it relates to the value
of reforestation. For reasons stated later, a definite pronouncement
on this subject by a professional body like this Society would have a
most important influence. Doubtless, there is some reluctance to
assume a positive attitude on a question which, as Mr. Knowles says,
is "one of the most controversial subjects upon which engineering and
scientific minds have been recently engaged." It may be questioned,
however, whether to-day there is much controversy among well-informed
engineers as to the essential principles involved. Is not the consensus

Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 107 of 167)