American Society of Civil Engineers.

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

. (page 108 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 108 of 167)
Font size
QR-code for this ebook

of professional opinion that, in times of excessive rainfall, in which


Mr. really great floods always originate, the restraining efi^ect of forest


cover, whatever it may be in ordinary storms, becomes exhausted and
is of no effect? This is the official opinion in France, even in forestry
circles, as is shown by the following extract from a report by the
Director of the French Forestry Service on the 1910 flood of the Seine :

"In the exceptional circumstances which always produce the floods
of the Seine, the action of forests, whether evergreen or deciduous,
becomes for the moment negligible in its efl^ect on the volume of
water carried by the river,"*

Engineers are apparently agreed on this essential point. In for-
mulating plans for flood control, they would not dare to reduce in
any degree the effectiveness of direct measures, such as reservoirs,
levees, etc., by any possible measures of reforestation. The Committee
brings out this point in the first paragraph of page 1224 of its
report, but it weakens its position by holding out the questionable
hope, implied in the words "not at present susceptible", etc., that the
future may demonstrate a value in forest cover not at present apparent.
Mr. Knowles, on page 1231, is outspoken in the expectation that "quan-
titative determination" of the effects of forests on run-off is a possibility
of the future. The writer dissents wholly from that opinion. The
difficulties in the way of any such result are practically insuperable.
A determination for one situation, or season, or storm, would never
apply reliably to any other. The modifying influences are infinite in
detail, and the utmost that we can hope for is a gross integration of
results, as shown in the records of stream flow. This has already
been carried on in Europe for more than a hundred years, with a
wholly negative result so far as the favorable influence of forests in
reducing run-off is concerned. It seems very desirable, therefore,
that a report of this character should not hold out any false encour-

At this point, it may be well to consider a feature of the subject
not mentioned in the report, yet a most important one, and one con-
cerning which there may be differences of opinion, even among engi-
neers. Reference is made to the influence of drainage on flood run-off.
The obvious and plausible inference is that drainage — roads, pave-
ments, sewerage, farm ditches, tile drains, etc. — expedites the run-off
and fills up the streams more rapidly than formerly; but there are
very strong counter forces which operate to retard run-off, particularly
from originally wet and marshy soil and from swamp areas. This
is the creation of ground storage where none existed before. The very

* "Mais, dans les circonstances exceptionelles dans lesquelles se produisent
toujours les crues de la Seine, I'action des forets, fussent-elles resineuses ou feuillees,
devient momentanement negligible sur les volumes des eaux roulees par le fleuve. Les
pluies intenses et persistantes de la seconde quinzaine de Janvier dernier, ayant saturfi
le sol forestier, commes les autres terrains, les eaux tombees k la surface sont toutes
allees aux thalwegs." M. Daubree, Conseilleur d'Etat, Directeur General des Eaux et
Forets. Report of First Commission on Flood of 1910 in the Seine, p. 515.


purpose of drainage is to get water out of the ground; but when so Mr.
gotten out, space is left for other water. The most effective and Chittenden
instantaneous of all reservoirs, so far as flood control is concerned, is
the top 18 in. of soil, depending on the extent to which it is saturated.
A swamp soil undrained is generally thoroughly saturated. It comes
next to an actual body of water in that respect. The onlj^, or the
main, storage available is surface storage due to the conformation of
the basin. Drainage takes the water out of the swamp soil and creates
vast storage space where before none existed. This effect is fully
recognized by the French. It is, of course, greatest with summer or
early fall storms. Like all surface storage, it may become greatly
reduced or completely exhausted by the long rains of winter.

With the utmost that can be done in attempting to evaluate the
influences of man's occupancy of the soil in modifying the effects of
soil cover and absorption, we find it impossible to say on which side

.the balance lies. It is probably not important, either one way or the
other, and, in any event, becomes wholly obliterated in those excessive
rainfalls which alone produce great floods. This conclusion har-
monizes both with theory and stream-flow records, and should be
accepted as definite and final. The writer emphasizes it, not because
the popular view to the contrary influences the engineer to any great
extent, but because it does influence the legislator to whom the engi-
neer often has to look for fluids, and it thus, in the words of the
income tax law, "deducts" from his power of effective action "at the
source". How great is the importance of taking a definite stand on
this question may be judged when a distinguished United States Sen-

' ator, who pretends to some acquaintance with these subjects, and
whose opinions carry weight in legislation concerning them, is capable
of delivering, in apparent soberness and good faith, an utterance like
the following:

"We have been destroying our forests, great natural reservoirs of
moisture into M'hich the waters fall from the heavens, and where they
are stored in the leaves and in the loose soil, and drunk up by the
thirsty roots of the trees and vegetation, and thence the surplus
gradually makes its way to the creeks and tributaries of our rivers, and
the water which used to be absorbed by these forests is now hurried
on into the creeks and rivers."*

* Hon. Francis G. Newlands, in the United States Senate, Feb. 21st, 1916. It is
interesting to apply this to a particular case. In the country around Dayton. Ohio, in
March, 1913, after heavy winter rains had quenched to satiety any thirst from which
tree or other roots might have been suffering, there fell, in a period of 4 consecutive
days, between 9 and 10 in. of rain, or more than one-fourth of the average for an
entire year. Assuming that the virgin forest had still covered the country, what per-
centage of this deluge does Senator Newlands imagine would have been "drunk up by
the thirsty roots?" And what is Senator Newlands' warrant for calling the soil of the
forest "loose"? Has he ever noted a great tree which has been overturned by the wind
and has carried up with its roots perhaps 100 sq. ft. of soil so densely compacted that
it resists for years exposure to the elements? The truth is that virgin forest soil,
undisturbed for ages, is one of the most compact of soils, whereas cultivated soil,
ploughed and harrowed and worked over, and full of perennially decaying roots, has
the maximum absorptive capacity of which a given quality of soil is susceptible.


Mr. If the clearing and drainage of the soil — in general terms, cultiva-

1 en en. ^^^^^^ — ^^ ^^^ materially affect the development and intensity of great
floods, is there any respect in which man's work does enhance their
destructiveness ? This question admits of a definite answer in the
affirmative. By obstructing Nature's overflow channels Man draws
on himself the chief calamities which floods produce. The bottom
lands along streams are just as much a part of Nature's highway for
carrying run-off to the sea as the normal channels between banks.
They do not come into play so continuously, but they are a part of
the natural waterways, and were formerly so used whenever the dis-
charge rose above a certain volume. Man has pre-empted these chan-
nels entirely. They contain the most fertile lands, they are easiest
to build on, and most convenient for roads and railroads. It is here
that population multiplies and wealth accumulates, and whenever
Nature asserts her old prerogative of passing her floods down over them,
trouble ensues. By cutting off these bottoms, greater work is forced
on the ordinary channel, the flood heights are raised, and if once the
waters break their bomids, they not only find wealth and people in
their way, but are actually far more destructive than in their natural
state. Thus Man is, to a large extent, the author of his own mis-
fortunes in the matter of flood destructiveness. The point here empha-
sized, however, is that Man's interference with natural conditions, which
so often leads to disaster, relates to his trespass on overflow channels
rather than, as is popularly assumed, on the normal channels of the
streams themselves. Contrary to general opinion, Man's operations
have, on the whole, improved the natural carrying capacity of streams.
There are exceptions, of course, but the rule seems to be as just stated.
A striking example is furnished by the recent studies of the Great
Miami River, in Ohio. In the four principal towns, where channel
changes have been most important, the discharge capacity was found
to be from two to eight times as great as in the country above and
below, where Man's interference has been slight. Such a condition,
of course, cannot be accidental.

The foregoing remarks disclose the reason why flood destructiveness
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. In time, this disparity will dis-
appear, or rather be reversed, and flood destructiveness will then show
a decrease. Some railroads have already carried protective measures
so far that they suffer little, even from the greatest floods.

Reservoirs. — The divergence of view between the Majority and
Minority Reports, concerning the conflict of purpose of reservoirs
and the impracticability of utilizing the same reservoir simultaneously
for flood control and other uses, is probably more apparent than real.


It would seem that the essential principles are too clear to be capable Mr.
of dual interpretation. In the first place, the conflict of purpose in
reservoirs can be completely overcome by giving the reservoir excess
of capacity over storage requirements svifficient to provide for flood
control. This, however, is often impossible for lack of site (ignoring,
for the present, considerations of cost), and inasmuch as industrial
use, mvinicipal supply, etc., must be on a dependable basis, it follows
that they must receive first consideration. Subject to this restriction,
however, storage reservoirs, as such, will generally be useful in flood
prevention. It may happen that a flood-producing storm may come
early, when a reservoir is filling, in which case the reservoir might
be as effective as if designed for flood control only. Again, the flood
storms may arrive late, after the reservoir has become full or nearly
so. In that case, the discharge must go over the spillway; but, as
some elevation of surface is necessary to produce the spillway overflow,
and as the surface of the reservoir, and the storage per unit of rise,
are at a maximum at this stage, there will be, even in this extreme
case, a certain degree of control.

Nevertheless, the general principle laid down in the report holds.
The engineer charged with a project of flood control, in which the
use of a storage reservoir is contemplated, cannot take into consid-
eration possible advantages which may or may not accrue. If there
is a possibility that they may not accrue, he is bound to assume that
they will not. He must make the most unfavorable assumption within
reasonable probabilities. Having thus determined what aid can be
positively depended on from the reservoir, he must make other pro-
visions for such additional control as is required. It is right here,
it seems to the writer, that public supervision of reservoir construction
is demanded, and that public aid in the interests of flood control might
well be given. For example, if a reservoir site is about to be developed
on a stream where the flood problem is important, it might well be
that, by increasing the capacity beyond what would be sufficient for
storage purposes, the reservoir could also be made to serve the additional
purpose of flood prevention. In that case, public aid would seem to
be justifiable.

On the whole, the possibilities of combining flood prevention with
other uses of reservoirs are sufficient to justify Mr. Knowles' contention
that the Committee's report would be strengthened by omitting lines
17 and 18 on page 1225 : "Your Committee, however, does not intend to
condemn in toto the utilization of reservoirs for more than one purpose."

Barriers. — The comment on "Barriers", (c) page 1225, seems to
the writer most judicious in recognizing only the function of arresting
debris. Just now there is current another of those extremist propa-
gandas which have interfered so much with scientific treatment of
many of our problems by confusing the public mind with schemes the



Mr^ _ ^^^y virtue of which is superficial plausibility. Los Angeles, Cal., is
the present victim. A campaign is on there to make barriers the sole,
or at least principal, solution of the flood problem in that locality.
There has been created a ''very wide-spread furore that the only way
to control the flood waters [at Los Angeles] is to build myriads of little
check dams in the mountains, and, through the detritus caught above
each of them acting as a sponge, to soak up all the floods before they
leave the mountains."

So far as the writer has observed in reports on this subject, the
primary purpose of these check dams is to prevent erosion. Not much
stress is laid on their effect in storing water. They soon become filled
with detritus, and this, in the general case, will become well soaked
with the rains leading up to great flood storms, so that the chances
in favor of any material reduction in discharge are altogether un-
certain. Immediately following a dry spell, a heavy run-off might
be considerably reduced, whereas, with a storm of the same intensity
following a series of lesser rains, there might be practically no relief.
In this, as in the case of forests, the engineer cannot build on uncer-

' Outlets. — With some hesitation the writer offers a certain criticism
of the conclusion stated in the last paragraph on page 1227. So far as
he has followed the official reports on the subject, it is evident that
there is an important distinction between the outlets proposed, say,
near New Orleans, and natural outlets or accidental crevasses cited
as militating against the outlet theory. It is manifest that the action
on the regimen of a river of a great crevasse must be different from
that of a long overflow weir which comes into play only after the river
has reached a certain stage. The writer cannot understand why, if
the river safely transports its sediment at that stage, it should fail
to do so by preventing any marked rise thereafter through abstraction
of the surface flow by a weir. It is very possible that the outlet
channel itself might silt up and become ineffective, but that would
have nothing to do with the argument just advanced.

Cut-Offs. — Having now reviewed the general features of the report,
the writer desires to devote more detailed consideration to the subject of
cut-offs. In general, his views coincide with those of Mr. Grunsky. He
believes that the ultimate effect of cut-offs is beneficial, providing always
that those reasonable and obvious precautions are taken which the cir-
cumstances of the individual case clearly indicate. He dissents from
the orthodox view, so long held on the Mississippi, that that river has
a sort of intelligent purpose in maintaining a particular length of
channel, and that it always "seeks" to regain that length if suddenly
shortened by the excision of a bend. The truth is that a river in an
alluvial soil may perforin almost any antics which the physical con-
ditions will permit; but, that such a river has a definite length which


it is always seeking to maintain, is an assumption not substantiated Mr.
by fact, at least on other streams.

In illustration of this point, the writer presents a diagram (Fig. 3),
showing the changes of a century in about a 50-mile course of the
Missouri River. The early cut-ofF of some of the enormous bends
that existed in 1804 has never resulted in any "attempt" on the part
of the river, either on this section or above or below, to regain its
original length. On the other hand, successive surveys show if any-
thing a contrary tendency; and there is nothing in the long record
of navigation on that stream which indicates any ill results to
navigation resulting from such cut-ofFs.'^

The writer also questions the correctness of the statement that
cut-offs "increase flood heights below them", while "reducing flood
heights immediately above". This may be the first effect, but it is
not first effects, but permanent results, which must be considered.
In the long run, the result, according to the writer's observation, is
to draw down the channel above the cut-off until it comes practically
to the original grade of the channel below. (It is understood that we
are here considering only the case of alluvial streams in comparatively
unstable channels.) In the course of time, these effects may extend
many miles up stream, and the sharp change of slope at first produced
may be distributed so far as to be very slight per unit of distance.

Fig. 4 is reproduced from a photograph of quite an extensive
cut-off on the Missouri River which took place under the writer's

The channel shortening, as he now recalls (without verification
from the records), was about 7 miles. Much local alarm was felt
lest the cut-off should produce a flood below. The writer tried to
assure the public that no such thing could possibly happen; that the
cut-off in itself would not increase the discharge of the river at all,
except as to the small increment from draining the bend, and that
there was absolutely no danger. Nothing would allay the fear, how-
ever, and there was much talk of taking steps to prevent the cut-off,
when certain interests which wanted it to take place rendered judicious
assistance (with dynamite, mainly, it was said), and the controversy
was promptly settled. At the first gauge below, a rise of a few inches
was detected, but no untoward result whatever happened. In this
case, the writer feels certain that the final result was a lowering of
the old bed above the cut and not a raising of the bed below.

The reports of the Mississippi River Commission abound in refer-
ences to work designed to prevent cut-offs which might (to cite a

* One naturally questions the correctness of the early maps on which these
extreme bends are shown ; but, apart from the general accuracy of Lewis and Clark's
work, as shown by later surveys, there are authentic references in the journals of other
early voyages up the river. It never took less than a day and sometimes two days for
a keelboat to get well past Blackbird Hill, opposite the neck of one of these bends, after
having arrived in its vicinity.








Z _Z_L— 180i Lewis and Clark 68 Aliles

r.":.-;:'.-V:- 1852 C.S.LandSm-vey 49 "

: ; ; '■'.'. I879 Mo. RIv. com. survey 46 "

: — — ~.1894 County Sui-vey 44 "

Fig. 3.


particular case) "disturb the regimen of the river a long distance above Mr.
and below, and bring disaster to towns, levees, and other works along ' ^° ^°'
the river banks." The results of a cut-off that did take place are
traced in considerable detail, showing how one ill effect led to another,
until a heavy toll of misfortune had accumulated. All this appeals
to the writer, however, as a clear case for the Franklinian philosophy
contained in the parable of the horse-shoe nail. If the occurrence of
this cut-off had been anticipated by timely protective work above
and below, probably nearly all the ill effects which actually developed
might have been avoided. The writer would not for a moment criti-
cize official policy in preventing cut-offs on the Mississippi River; he
simply criticizes the theory on which that policy is ostensibly based.
He believes that the real objection to cut-offs on that stream is not
the effect on the regimen of the river (which can be controlled effec-
tively), but on riparian interests. Ports and landings might be ruined
and other interests jeopardized in a way that would result in serious
hardship ; but, aside from these considerations, viewed only as a
problem in river hydraulics, the writer believes that a material short-
ening (say 125 to 150 miles) of the Mississippi River between Cairo
and Red River would not, in the long run, injure the regimen, even
for purposes of navigation, and it would certainly make the river a
better flood carrier and would reduce materially the length of channel
to be maintained and of bank to be protected.

The writer would dissent, further, from the distinction which the
Committee draws as against alluvial streams in this connection. It
is, in many instances, on alluvial streams that this measure can be best
applied. The difference between an alluvial stream and one with
stable banks is that, in one case, disturbance of regimen may result,
and, in the other, it may not, and that, therefore, in using the cut-off
on alluvial streams, auxiliary measures are necessary. If these are
provided, however, the shortening of the channel should certainly
facilitate the natural movement of sediment down stream. Cut-offs
have been used extensively on great alluvial rivers, notably on the
Danube and other streams of the Hungarian plain.

The cut-off method finds its most effective application in low flat
tracts in which stream sinuosity dates from the remote period of
emergence of the soil from a status of continuous overflow, and is
not a result of present-day processes. Generally, in such locations,
the sinuosity is excessive, the slope very slight, and the channel capacity
wholly deficient. Extensive cut-offs do not produce excessive slopes
and troublesome erosion, but facilitate drainage and carry off flood
waters more rapidly. Wliether in any particular case a cut-off or
series of cvit-offs is desirable depends on local circumstances. The
most important obstacles are cost and the likelihood of riparian com-
plications. Another consideration relates to the excised portions of


Mr. the channel. If these can be filled up, there is a net gain of land area,
■ for the excised portion is always larger than the new channel. If they
cannot be filled, there is a loss of land equal to the new channel. Even
in that case, there may be some utility in the excised sections in
connection with drainage. By placing flood-gates between them and
the new channel, they can be drawn down to the level of low water
in the dry season and then be utilized as immense sumps to maintain
effective drainage during high-water periods.

Minority Report, — Concerning the several substitutes proposed in
the Mino:i*ity Report, some of them seem to be in the interest of com-
pleteness and precision, and some appear to the writer as undesirable.
Among the latter are the proposed changes in the forestry discussion.
As to (10), on page 1233, it would seem that Mr. Knowles confuses
two distinct things. The fact that levees cut off valley storage pre-
viously available does not in any way alter the other fact that levees,
by raising the flood surface between them, do develop a certain amount
of channel storage.

It is important, however, to observe a distinction, which neither
the Committee nor Mr. Knowles has brought out, but which justifies
to some extent Mr. Knowles' contention that "such storage is ineffec-
tive", although for a reason quite different from that which he has
assigned. It is entirely true that the volume of water at flood stage
stored in stream channels, like the Mississippi and Ohio, is prodigious,

Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 108 of 167)