American Society of Civil Engineers.

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

. (page 112 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 112 of 167)
Font size
QR-code for this ebook

it is not so widely known and will not be pointed to so frequently
as a precedent as will the Dayton project. The failure to utilize the
power of the Scioto River and its tributaries, which, with permanent
reservoirs of greater capacity, might have been made available, will
set a precedent which will undoubtedly lead to the utilization of this
plan elsewhere, where the conditions are not the same and where
complete utilization and conservation would have been practicable.

Flood Warning. — The worst feature of floods is the damage they
do to life and property. There are two ways of reducing this damage,
one is to control the floods and lessen their height, and the other is to
get out of the way of them and remove perishable property to places
above flood height. The first proposition has been much discussed,
but the latter, it seems, has not received the attention it deserves.
This involves flood warnings. By timely warnings many lives can
be saved and much valuable, perishable property can be removed beyond
the danger zone. Until the far-distant day when flood control is gen-
eral, more attention should be devoted to early and accurate flood

The Federal Government maintains a flood warning service which
has proved itself of great value. It is a Herculean task, however.
-ti) send flood warnings over this vast land of ours, and the means at
the disposal of this bureau have confined its work to certain points
land a limited number of streams.

Following the flood of March, 1913, which visited streams in West-
ern Pennsylvania, a law was enacted, and $10 000 was appropriated
by the State Legislature, with which to give to the people a more
intensified flood warning service than was possible by the United
States Government. This law was drafted with the approval of a
United States Weather Bureau representative, and in the 3 years
during which it has been enforced by the Water Supply Commission,
a great quantity of property has been saved from damage. A larger
appropriation would make greater savings possible. In Pennsylvania,
however, this service is hampered by the lack of similar service on
the head-waters of the Susquehanna, Allegheny, and Delaware Rivers,


which rise in New York and pass through or along the border of Mr.
Pennsylvania for hvmdreds of miles. sarnie

Either by vastly increasing the resources of the United States
Weather Bureau for flood warnings, so that it could give attention
to other than the few great rivers of the country, or by adopting
uniform flood warning legislation in the several States, much loss in
life and property could be prevented while we are waiting for control
works to be built.

An important duty of such a service is also that of warning pros-
pective constructors of railroads, highways, bridges, factories, mills,
and dwellings of the limits of the flood zones of rivers, so that they
may not locate within such area, or if they do, that the building site
may be elevated above flood level. If there were such an organization
in each State, vmder State or Federal control, and the people knew
of its existence, as they would if proper publicity was given to it
and its work, then it would be possible to reduce greatly the increase
in the damageable property which is constantly being placed in the
way of floods.

In conclusion the writer would suggest that it is the province of
this Society to do three things:

1.- — Place at the disposal of its members full and accurate flood

2. — Change the popular fear of dams into an understanding of
their security;

3. — Formulate and push through the adoption of legislation along
these lines :

A. — To control obstructions and encroachments in or along
streams ;

B. — To permit of co-operation between corporations. States,
municipalities, and the Federal Government in the construc-
tion and operation of storage reservoirs for water power, water
supply, and other commercial uses, so as to adapt them for
flood control as well :

C. — To extend the Federal, or establish a State, flood warning

C. E. Grunsky,* M. Am. Soc. C. E. (by letter). — The discussion Mr._
which the Committee's report has provoked is of a character fully
justifying the report, notwithstanding the fact that the Committee
was not in a position to attempt to say the last word on the subject
of flood control.

Some of the statements made by those who point oiit the incom-
pleteness of the report merit reply, in order that wrong impres-

* San Francisco, CaL


Mr. sions may be corrected. Mr. Eakin says: "It is indirectly stated that


' the use of levees effects an increased channel storage that induces a
lowering of flood levels." No one who knows anything about the propa-
gation of a flood wave could by any possibility intentionally make
the statement that the reduction of the flooded area which results from
the construction of levees will reduce the elevation to which water
will thereafter rise between the levees. The language used by the
Committee (of which the writer was a member) is therefore unfor-
tunate. The levees are put on the banks of rivers for the purpose of
confining the water to a restricted channel and thereby increasing
the rate of flow in this channel, and this cannot be done without
causing the water between the levees and within the channel to rise
to greater height than that which would have been attained without
levees. It takes the crest of a flood wave 10 days, or about that time,
to travel in the Mississippi River from Cairo to New Orleans. Let
it be assumed that at flood the volume of flow is such that levees confine
the water to the river and to bank-land areas between the levees. It
is evident that in such a case the discharge of the river in excess of
ordinary flow for about 10 days before the cresting of the flood at
New Orleans has been consumed in filling the river channel above
that point from a low to a high stage. It is also evident that this
channel storage is greater than it would have been if there had been
no levees. To the extent that this increase of channel storage is made
possible by the levees it is a factor making for a reduction of the
maximum flow at all down-stream points.*

Channel storage between levees does not reduce the total run-off,
and it does not reduce the water surface elevation of the river at
flood, unless this be in the exceptional case of scour, which may result
from the higher velocities in the channel between levees than existed
under natural conditions. No claim can be made, and none was
intended to be made by the Committee, that the confinement of water
by levees will decrease the original maximum stream flow at the
mouth of the river. Levees are built for the purpose of reducing
flooded areas. They are built to keep water away from land that with-
out them would be less desirable for agricultural, industrial, and other
uses. Consequently, for the same volume of flood flow, the opportunity
for storage, taking the river basin in its entirety, is less after levees
are built than before. The flood wave reaches the mouth of the leveed
river with less elongation than under natural conditions, and con-
sequently the discharge at the peak must be greater. As an illustration,
the Sacramento Flood Control project may be referred to. Under
natural conditions the floods of Sacramento River inundated broad
lateral basins, and the lower reaches of the river had a maximum

• See Transactions, Am. Soc. C. E., Vol. L.XI, p. 332 ; and also Report of Com-
missioner of Public Works of California, 1895, p. 130.


flow of about 200 000 sec-ft. The withdrawal from inundation of Mr.

Grunsk V
the islands of the delta region of this river and of the various flood-
basins which border the river will increase the momentary flood maxi-
mum to more than 500 000 sec-ft. Of this quantity, in the latitude
of Sacramento, it is proposed to hold as much as possible in the river,
20 to 30%, confined between levees of reasonable height, but at
elevations which at extreme flood will be about 7 ft. higher than the
original river flood-plain. It is the additional channel storage repre-
sented by this added depth of water which the Committee had in
mind when it referred to "the resultant reduction of flood height.''
What the Committee desired to do was to call attention to the added
storage as a factor making for elongation of the flood wave and con-
sequent reduction of maximum discharge, without any intention of
claiming that channel flood heights as originally existing would be
reduced by the construction of levees.

In connection with this matter, and in reply to the remarks of
Gen. Chittenden, the general principle should be kept in mind that
the elongation of a flood wave as it passes down stream is due to the
effect of storage. If there were no change possible in the volume of
water in storage between an up-stream and a down-stream point, as
in the case of a closed conduit, there could be no elongation of the
flood wave. The accession of water at the upper point would at once
be manifest by an increased discharge in equal quantity at the lower
point. Storage, whether in the channel or in overflow basins, has
the effect of elongating in some measure every flood wave that travels
down a river, and by such elongation the time of passage is increased
and the flow at the peak is correspondingly reduced. It follows, as
demonstrated by the writer in a paper prepared in 1880 while he was
Assistant State Engineer of California, and published in 1895 in the
Report of the Commissioner of Public Works of California,* that the
maximum discharge of a river which receives no accessions from
tributaries will decrease with distance down stream. Every decrease
of available storage space, such as may result from a contraction of
the waterway, will reduce the effect of storage and will tend to
increase the maximum discharge at down-stream points. Every
increase of storage space, resulting from the raising of levees or the
adding of retention reservoirs, on the other hand, will decrease the
maximum flow at down-stream points.

The Committee shares the view expressed by Mr. Eakin when
he says:

"No individual or organization has as yet commanded such pro-
ficiency in all these sciences as to enable them to outline a programme
of river treatment with dependable authority."

* See also Transactions, Am. Soc. C. E., Vol. LXI, p. 332.

1298 Discussioisr on floods and flood pkeventiox

Mr. Herein lies the justification of the Committee in presenting a

runs y. pppQj.-(- jj-^ which the shortcomings were known to the members, and

which was not put forward as exhaustive on the subject of flood control,

a subject which necessarily presents as many different aspects as there

ai'e river systems.

It happens frequently, as in the case of the Sacramento and San
Joaquin Rivers, in California, that some of the natural flood basins
along these streams, such as Butte Basin to the northward of the
Marysville Buttes, Upper Colusa Basin on the west side of Sacramento
River, and, to a lesser extent, Sutter Basin, Lower Colusa Basin,
and American Basin, and some of the submersible lands along the
San Joaquin River, partake of the nature of the lands which in con-
nection with flood-control projects on other river systems are selected
for detention reservoirs. The demand is strong in the California case
for the protection of these submersible valley lands against overflow.
It becomes an economic question, then, to determine where the line
should be drawn, and it is by no means certain that complete pro-
tection against floods is in all cases justified. Occasionally, it will be
found advisable to retain a natural overflow basin, not only for the
control of the floods which it helps to effect, but also as a recipient
of some of the silt carried by the stream. The permanent utilization
of a natural flood basin seems particularly desirable when the basin
is situated so that, while acting as a detention reservoir, it can be
drained of its water in time to leave the land available for the culti-
vation of summer crops. In 1850 the United States gave to Arkansas
and a number of other States, for the purpose of reclamation, the land
which was segregated as swamp and overflow, and these several States
thereupon proceeded to divest themselves of ownership as rapidly as
possible. They blundered. The land subject to inundation should have
remained in public ownership until comprehensive plans for flood
control were made. It would then have been a simple matter to have
the plans carried out in proper sequence, and much embarrassment
would have been avoided that now results from the demand that
protection be provided for lands which, if they were not in private
ownership, would to no inconsiderable extent be put in a class re-
quiring only partial protection or fair protection against ordinary
and not extraordinary floods.

Mr. Hill points out that the extreme floods of recent years are not
due to changed climatic or physical conditions. The example which
he supplies of an extreme flood stage in the Great Miami in 1S05
before forests had been removed suggests that it would not be out of
place to refer to some additional facts to show that a connection
between deforestation and stream flow will remain difficult to


The writer, in a paper presented to the Society some years ago,* Mr.
has shown that the dominating cause of changes in the surface eleva- ^™°^'^y-
tion of the Great Lakes is the climate, and that a long period of less
than normal rain and snow may be followed by another with more
than the normal. A still better example to illustrate the variation in
the rainfall and rim-off is the Great Salt Lake. This lake was very
low when visited and sounded by Capt. Stansbury in 1849. It was
low when visited by Fremont some years earlier. In the years pre-
ceding 1869 it rose, ultimately reaching an elevation about 11 ft.
above that of 1849, and then, with numerous minor fluctuations,
decreased in extent and in elevation until 1906, when its water sur-
face was about 4 ft. lower than it had been in 1849. It was supposed
by many that, on account of the increasing utilization of water for
irrigation on its water-shed, the lake would never again rise to any-
where near the high stages of 1868 to 1876, but this view was in
error. The lake is now about 8 ft. higher than it was at its lowest
stage in 1904, having maintained this high stage for several years.
For a long time preceding the low period of 1849, and extending to
about 1861, the rainfall in the region tributary to Salt Lake as a
water-shed must have been less than normal. Then followed a num-
ber of years in which, taken collectively, there was more than normal
rain, until about 1876, after which there came another period with
less than normal rain as the average for a number of seasons. The
consequent decrease of run-off and the increasing use of water for
irrigation brought the lake to its lowest stage in 1906. In the case
of this lake, which has no outlet, the rise and fall is a fairly good
index of the run-off, and therefore of the rainfall, when groups of
consecutive years are considered. The problem, moreover, is not here
complicated by deforestation. It appears from this and numerous
similar cases that very large variation in the seasonal water produc-
tion of any region may be anticipated, and that the persistence of any
tendency which the seasonal run-off may have to increase or decrease
for a number of years is not to be accepted as conclusively demon-
strating either a permanent change of climate, nor yet that such
increase or decrease is in any way related to aforestation or to
deforestation of water-shed areas. The same general statement will
apply also to the possible momentary maximum stream-flow. The
factors affecting the rate of run-off are, moreover, so many and so
inter-related that it will remain difficult if not impossible to make a
conclusive determination of the effect of forest growth on flood

Mr. Leighton is evidently of the opinion that no report on the sub-
ject of flood control can be a well-considered report which does not

• Transactions, Am. Soc. C. E., Vol. LXIII, p. 31.


Mr. recognize the forest as one of the material factors which influence the
y- magnitude and frequency of floods. Much as the writer has studied
this subject, he has yet to find any convincing demonstration that
there is such an effect. Generally, the individual cases cited are so
complicated by related circumstances that the evidence is not con-
clusive. The writer is not prepared to accept, and does not believe
that the Profession will accept, as final, any mathematical demon-
stration that forest growth will markedly reduce or meliorate flood
conditions. If the statement made by the Committee gives the impres-
sion that the quantitative influence of the forest on the frequency and
magnitude of floods is not proved, this is exactly what one member of
the Committee, at least, intended. This is without prejudice to the
forest, which the writer would like to see extended and maintained to
meet fully at all times the requirements of the country. The writer,
moreover, does not wish to be classed among those who will not con-
cede that it may yet be possible to demonstrate the true relation of the
forest to the flood; and, for that reason, he subscribed to the state-
ment in the report that the exclusion of the forest as a method of
flood prevention had not yet been demonstrated, though he believes
that, except in the rarest cases, conditions will never be such that the
forest will become a material factor.

The maximum stream flow is not dependent solely on the rainfall
and the rate at which snow is melting; it depends also on the condition
of the surface of the ground on which the snow lies or the rain falls.
Usually, the period of intense rain which produces flood stages is pre-
ceded by rain which saturates more or less thoroughly the top layers
of the soil. This is least likely to occur in sandy and gravelly regions
and where loamy soils are well tilled and are deep. It is most likely
to occur in regions where the ground is swampy and naturally full of
water, or where the flood-producing rain falls on frozen ground which
requires but little water for complete saturation. It follows that,
whenever a region which in its natural condition was swamp has been
made arable and brought under cultivation, the probability of com-
plete saturation preceding a flood-producing rain will have been
reduced, and some effect on the frequency and intensity of the flood
condition in such a region will have to be conceded to the reclamation
work. The main effect of any such modification of surface conditions
should be sought in terms of water storage and retardation of the
flood wave, because, if thereby the resultant elongation of the flood
wave can be determined, some basis will exist for estimating the
quantitative effect on the maximum flow. The writer agrees with the
comments of Gen. Chittenden in this regard, and shares his view
that the modification of the effect of soil cover and absorption by Man's
occupancy and use of the soil will be a negligible factor in the exces-
sive rainfalls which produce great floods.


In the light of present information, the writer agrees too with Mr.
Gen. Chittenden when, after referring to the aggravation of flood ^"°'' ^'
damage which results from the obstruction and use by Man of
Nature's overflow channels, he says:

"The foregoing remarks disclose the reason why flood destructive-
ness is increasing. It is not because floods themselves are increasing,
either in frequency or intensity, but because property subject to
destruction is very greatly increasing, and as yet protective work does
not keep pace with this increase."

The writer, though having qualifiedly subscribed to the Com-
mittee report, is not in full accord with its reference to outlets. Con-
centration of the flood flow in one channel is desirable. Separation
into a number of channels is undesirable. The conditions, however,
are not always such that a single channel can be given adequate
capacity to carry all the water presented at flood. The study of the
question what to do when this is the case led the writer to formulate
certain principles for the treatment of Sacramento River floods,* sub-
stantially as follows:

First. — Make the river carry as much water as possible between
levees of reasonable height.

Second. — Let the flood-water in excess of capacity between levees
go out of the channel at selected points under control.

Third. — Confine the outgoing surplus waters to limited areas, and
keep them under control until they may re-enter the river or can be
delivered into the bay.

The writer has not modified his views in this respect, having, sub-
ject to these principles, always been in favor of relief weirs or con-
trolled outlets. These are not to be confounded with ordinary outlets
which begin to function before the river has reached the danger line
and which become secondary branches of the stream. It is outlets
of the latter kind, in the sense of a division of the stream, which with
good reason are disapproved by the Committee.

The writer is pleased to find Gen. Chittenden in accord with the
writer's view that the cut-off is generally a desirable improvement,
and that the fear that shortening a river channel will interfere with
its usefulness as a commercial highway is not ordinarily well grounded.
The sketch, Fig. 5, is from an old dociiment, and shows the extent
to which, by intelligent direction and restraint, the course of the
Rhine in the vicinity of Germersheim was modified during the period
from 1817 to 1861. This sketch shows channel shortening and align-
ment modification quite comparable to that which might be carried
out on the Mississippi, on the Missouri, or on the Colorado, if the
necessary outlay, including permanent bank revetment, were justified,
and the occupancy and use of the bank-lands do not already set a

* Report of Commissioner of Public Works of California, 1894-95, p. 59.







Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 112 of 167)