RIVER AND CANAL ENGINEERING

By the Same Author

HYDRAULICS WITH WORKING TABLES. Second Edition. 160 illustrations, xii + 311 pp., 8vo (1911).

12/- net.

PUNJAB RIVERS AND WORKS. Second Edition. 47 illustrations, viii +64 pp., folio (1912).

8/- net.

IRRIGATION WORKS (in the press). 27 illustrations, vi + 130 pp., 8vo (1913).

6/- net.

THE SUCTION CAUSED BY SHIPS. Explained in Popular Language. 2 plates, 26 pp., 8vo, sewed (1912).

1/- net.

E. & F. N. SPON, LTD., LONDON

RIVER AND CANAL
ENGINEERING
THE CHARACTERISTICS OF OPEN FLOWING STREAMS, AND THE PRINCIPLES AND METHODS TO BE FOLLOWED IN DEALING WITH THEM
BY
E. S. BELLASIS, M.Inst.C.E.
RECENTLY SUPERINTENDING ENGINEER IN THE IRRIGATION BRANCH OF THE PUBLIC WORKS DEPARTMENT OF INDIA
72 ILLUSTRATIONS

London
E. & F. N. SPON, Ltd., 57 HAYMARKET, S.W.
New York
SPON & CHAMBERLAIN, 123 LIBERTY STREET
1913

PREFACE

The object of this book is to describe the principles and practice adopted in the Engineering of Open Streams. If the book seems to be somewhat small for its object, it will, it is hoped, be found that this is due to care in the arrangement and wording.

Sources of information have been acknowledged in the text, but special mention may be made of lectures given by Professor Unwin at Coopers Hill College, of Harcourt’s large work on Rivers and Canals, of the papers[1] by Binnie on rainfall, by Shaw on the closing of the river Tista, by Harcourt on movable weirs and on estuaries, by Strange on reservoirs, and by Ottley and Brightmore, Gore and Wilson, and Hill on the stresses in masonry dams; of the articles by Bligh[2] on weirs with porous foundations and by Deacon[3] on reservoir capacity, of the Indian Government paper by Spring on “River Control on the Guide Bank System,” and of the Punjab Government paper containing Kennedy’s remarks on silting and scour in the Sirhind Canal. The two papers last mentioned are not easily accessible, and they contain matter of great interest. The important points, often obscured by masses of detail or figures, have been extracted.[4]

Silting and scour ([Chap. IV.]) had already been dealt with in Hydraulics,[5] but some further information has since come to light and the subject has been treated afresh and the matter re-written.

E. S. B.

Cheltenham, 1st May 1913.

RIVER AND CANAL ENGINEERING

CHAPTER I
INTRODUCTION

1. Preliminary Remarks.—River and Canal Engineering is that branch of engineering science which deals with the characteristics of streams flowing in open channels, and with the principles and methods which should be followed in dealing with, altering, and controlling them. It is not necessary to make a general distinction between natural and artificial streams; some irrigation canals or other artificial channels are as large as rivers and have many of the same characteristics. Any special remarks applicable to either class will be given as occasion requires.

2. Résumé of the Subject.—[Chap. II.] of this book deals with the collection of information concerning streams, a procedure which is necessary before any considerable work in connection with a stream can be undertaken, and often before it can even be decided whether or not it is to be undertaken. [Chap. III.] deals with rainfall, and describes how rainfall figures and statistics can be utilised by the engineer in dealing with streams.

[Chap. IV.] explains the laws of silting and scouring action, a subject of great importance and one to which the attention ordinarily given is insufficient. The general characteristics of streams, being due entirely to silting or scouring tendencies, are included in this chapter. [Chap. V.] describes how silting or scouring may be, under some circumstances, artificially induced or retarded.

[Chap. VI.] deals with various methods of protecting banks against erosion or damage. [Chap. VII.] treats of diversions or the opening out of new channels, and with the opposite of this, viz. the closing of channels, a feat which, when the stream is flowing, is sometimes very difficult to achieve. This chapter also deals with dredging and excavation.

[Chap. VIII.] discusses the subject of the training of streams, a class of work which is generally undertaken in order to make them navigable or to improve their existing capacities for navigation, but may be undertaken for other reasons. The main features of this kind of work are the narrowing and deepening of the stream, often the reduction of the velocity and slope, and generally the raising of the water-level. In this kind of work a channel may be completely remodelled and even new reaches constructed. [Chap. IX.] deals with artificial channels of earth or masonry, and includes navigation canals.[6]

In [Chaps. X.] and [XI.] the chief masonry works or isolated structures—as distinguished from general works which extend over considerable lengths of channel—are dealt with, and those principles of design discussed which affect the works in their hydraulic capacities. General principles of design applicable to all kinds of works, such as the thicknesses of arches or retaining walls, are not considered; they can be found in books on general engineering design.

[Chap. XII.] treats of storm waters and river floods, and shows how works can be designed for getting rid of flood water and how floods can be mitigated or prevented, one of the chief measures, the widening of the channel and the lowering of the water-level, being the opposite of that adopted for training works. Embankments for stopping flooding are also dealt with. [Chap. XIII.] deals with reservoirs, including the design of earthen and masonry dams.

[Chaps. XIV.] and [XV.] deal with tidal waters, river mouths and estuaries, and works in connection with them, viz. the training of estuaries and the methods of dealing with bars, the object being in all cases the improvement of the navigable capacities of the channels.

3. Design and Execution of Works.—After obtaining full information concerning the stream to be dealt with, careful calculations are, in the case of any large and important work, made as to the effects which will be produced by it. These effects cannot always be exactly foreseen. Sometimes matters can be arranged so that the work can be stopped short at some stage without destroying the utility of the portion done, or so that the completed work can be altered to some extent.

In works for controlling streams there is, as will appear in due course, a considerable choice of types of work and methods of construction. In practice it will generally be found that there are, in any particular locality, reasons for giving preference to one particular type or kind of work or, at all events, that the choice is limited to a few of them, either because certain kinds of materials and appliances can be obtained more cheaply and readily than others, or because works of a particular type have already been successfully adopted there, or because the people of the district are accustomed to certain classes of work or methods of construction. In out-of-the-way places it is often undesirable to avoid any type of work which cannot be quickly repaired or readily kept in order by such means as exist near the spot.

It is sometimes said that perishable materials, such as trees, stakes, and brushwood, cannot produce permanent results. They can produce results which will last for a long time and which may even be permanent. By the time the materials have decayed, the changes wrought may have been very great, deposits of shingle or silt may have occurred and become covered with vegetation, and there may be little tendency for matters to revert to their former condition. If the expense of using more lasting materials had had to be incurred, the works might never have been carried out at all. On the Mississippi enormous quantities of work have been done with fascines.

4. The Hydraulics of Open Streams.—When any reach of a stream is altered, say by widening, narrowing, or deepening, so that the water-level is changed, there will also be a change in the water-level, a gradually diminishing change, for some distance upstream of the reach. Also in the lowest portion of the reach the change will gradually diminish and it will vanish at the extreme downstream end of the reach. In the next lowest reach there is no change. Thus if it is desired that the change in the water-level shall take full effect throughout the whole of a reach, the change in the channel must be carried further down. If a weir is built there is no change in the water-level downstream of it except such as may be due to loss of water in the reach upstream of it. The above points are mentioned here because, although they are really questions of hydraulics, they are of much importance and of very general application.

Matters connected with the hydraulics of open streams seem to lend themselves in a peculiar way to loosely expressed remarks and fallacious opinions. The set of a stream towards a bank is sometimes supposed to profoundly affect the discharge of a diversion or branch. Its effect is simply that of “velocity of approach,” which, as is well known, is quite small with ordinary velocities, and is merely equivalent to a very small increase of head. Narrow bridges or other works are sometimes said to seriously “obstruct” a stream without any observations being made of the fall in the water surface through the bridge. This fall is the only measure of the real obstruction.[7]

CHAPTER II
RAINFALL

1. Rainfall Statistics.—The mean annual rainfall varies very greatly according to the locality. In England it varies from about 20 inches at Hunstanton in Cambridgeshire, to about 200 inches at Seathwaite in Cumberland; in India, from 2 or 3 inches in parts of Scinde, to 450 inches or more at Cherrapunji in the Eastern Himalayas.

Rain is brought by winds which blow across the sea. Hence the rainfall in any country is generally greatest in those localities where the prevailing winds blow from seaward, provided they have travelled a great distance over the sea. Rainfall is greater among hills than elsewhere, because the temperature at great elevations is lower. Currents of moist air striking the hills are deflected upwards, become cooled, and the water vapour becomes rain. This process, if the hills are not lofty, may not produce its full effect till the air currents have passed over the hills, and thus the rainfall on the leeward slopes may be greater than elsewhere, but on the inner and more lofty ranges the rainfall is generally greatest on the windward side.

Thus the rainfall may vary greatly at places not far apart. An extreme instance of this occurs in the Bombay hills, where the mean annual falls at two stations only ten miles apart are respectively 300 inches and 50 inches.

In temperate climates the rainfall is generally distributed over all the months of the year; in the tropics the great bulk of the rain often falls in a few months.

The fall at any one place varies greatly from year to year. To obtain really reliable figures concerning any place, observations at that place should extend over a period of thirty to thirty-five years. The figures of the mean annual fall will then probably be correct to within 2 per cent. The degree of accuracy to be expected in results deduced from observations extending over shorter periods is as follows:—

These figures were deduced by Binnie (Min. Proc. Inst. C.E., vol. cix.) from an examination of rainfall figures obtained over long periods of time at many places scattered over the world. The errors may, of course, be plus or minus. They are the averages of the errors actually found, and are themselves subject to fluctuations. Thus the 15 per cent. error for a five-year period may be 16 or 13, the 8 per cent. error for a ten-year period may be 8½ or 7½, with similar but minute fluctations for the other periods.

Binnie’s figures also show that the ratio of the fall at any place in the driest year to the mean annual fall, averages ·51 to ·68, with a general average of ·60, and that the ratio in the wettest year to the mean annual fall averages 1·41 to 1·75, with a general average of 1·51. For India the general averages are ·50 and 1·75. These figures are useful as a means of estimating the probable greatest and least annual fall, but they are averages for groups of places. The greatest fall at any particular place may occasionally be twice the mean annual fall. At some places in India, in Mauritius, and at Marseilles it has been two and a half times the mean annual fall. The least annual fall may, in India, be as low as ·27 of the mean. In England the fall in a dry year has, once at least, been found to be only ·30 of the mean annual fall. The mean fall (average for all places) in the three driest years is, from Binnie’s figures, about ·76 of the mean annual fall. The figures given above, except when a particular country is mentioned, apply to all countries and to places where the rainfall is heavy, as well as to those where it is light. But in extremely dry places the fluctuations are likely to be much greater. At Kurrachee, with a mean annual fall of only 7·5 inches, the fall in a very wet year has been found to be 3·73 times, and in a very dry year only ·07 times the mean annual fall.

In the United Kingdom the probable rainfall at any place in the driest year may be taken as ·63 of the mean annual fall. For periods of two, three, four, five and six consecutive dry years, the figures are ·72, ·77, ·80, ·82, and ·835. These figures are of importance in calculations for the capacity of reservoirs ([Chap. XIII., Art. 2]).

When accurate statistics of rainfall are required for any work, the rainfall of the tract concerned must be specially studied and local figures obtained for as many years as possible. Very frequently it is necessary to set up a rain-gauge, or several if the tract is extensive or consists of several areas at different elevations. Sometimes there is only a year or so in which to collect figures. In this case the ratio of the observed fall to that, for the same period, at the nearest station where regular records are kept, is calculated. This ratio is assumed to hold good throughout, and thus the probable rainfall figures for the new station can be obtained for the whole period over which the records have been kept at the regular station. The volumes of the British Rainfall Organisation contain a vast amount of information regarding rainfall. For a large area there should be one rain-gauge for every 500 acres, for a small area more. In the case of a valley there should be at least three gauges along the line of the deepest part—one at the highest point, one at the lowest, and one midway as regards height—and two gauges half way up the sides and opposite the middle gauge (Ency. Brit., Tenth Edition, vol. xxxiii.). Some extra gauges may be set up for short periods in order to see whether the regular gauges give fair indications of the rainfall of the tract. If they do not do so some allowances can be made for this.

2. Available Rainfall.—The area drained by a stream is called its “catchment area” or “basin.” The available rainfall in a catchment area is the total fall less the quantity which is evaporated or absorbed by vegetation. The evaporation does not chiefly take place directly from the surface. Rain sinks a short distance into the ground, and is subsequently evaporated. The available rainfall does not all flow directly into the streams. Some sinks deep into the ground and forms springs, and these many months later augment the flow of the stream and maintain it in dry seasons. The available rainfall of a given catchment area is known as the “yield” of that area.

Estimation of the available rainfall is necessary chiefly in cases where water is to be stored in reservoirs for town supply or irrigation. The ratio of the available to the total rainfall depends chiefly on the nature and steepness of the surface of the catchment area, on the temperature and dryness of the air, and on the amount and distribution of the rainfall. The ratio is far greater when the falls are heavy than when they are light. Again, when the ground is fairly dry and the temperature high—as in summer in England—nearly the whole of the rainfall may evaporate; but when the ground is soaked and the temperature low—as in late autumn and winter in England—the bulk of the rainfall runs off. In the eighteen years from 1893 to 1900 the average discharge of the Thames at Teddington, after allowing for abstractions by water companies, was in July, August, and September 12 per cent. of the rainfall—6·9 inches—in its basin, and in January, February, and March 60 per cent. of the fall which was 5·9 inches. The total fall in the year was 26·4 inches. Some rivers in Spain discharge, in years of heavy rainfall, 39 per cent., and in years of light rainfall 9 per cent. of the rainfall (Min. Proc. Inst. C.E., vol. clxvii.). The discharge of a river is not always greatest in the month, or even the year, of greatest rainfall.

The table opposite gives some figures obtained by comparison of rainfall figures and stream discharges. The case of the area of 2208 acres near Cape Town is described in a paper by Bartlett (Min. Proc. Inst. C.E., vol. clxxxviii.), and it is shown by figures that part of the rainfall in the rainy season went to increase the underground supply which afterwards maintained the flow in the dry season.

The following statement shows how the available rainfall may vary from year to year. The figures are those of a catchment area of 50 square miles on the Cataract River, New South Wales (Min. Proc. Inst. C.E., vol. clxxxi.):—

The manner in which the available rainfall may vary from month to month is shown in the following statement, which gives the figures for 1905 for the Sudbury River in Massachusetts:—

Rankine gives the ratio of the available rainfall to the whole fall as 1·0 on steep rocks, ·8 to ·6 on moorland and hilly pasture, ·5 to ·4 on flat, cultivated country, and nil on chalk. These figures are only rough. The figures for rocks and pastures are too high. The loss from evaporation and absorption is not proportional to the rainfall. It is far more correct to consider the loss as a fairly constant quantity in any given locality but increasing somewhat when the rainfall is great. The available rainfall in Great Britain has generally been overestimated. Sometimes it has been taken as being ·60 of the whole fall. More commonly the loss is taken to be 13 to 15 inches. This is correct for the western mountain districts, where the rainfall is about 80 inches and the soil consists chiefly of rocks partly covered with moorland or pasture. In other parts of the country, especially where flat, the loss is often 17 to 20 inches. All the above figures are, however, general averages. The proper estimation of the available rainfall at any place in any country depends a great deal on experience and judgment, and on the extent to which figures for actual cases of similar character are available. Regarding the “run-off” from saturated land during short periods, see [Chap. XII., Arts. 1] and [2.]

3. Measurement of Rainfall.—A rain-gauge should be in open ground and not sheltered by objects of any kind. The ordinary rain-gauge is a short cylinder. This is often connected by a tapering piece to a longer cylinder of smaller diameter. In this the rain is stored safely and is measured by a graduated rod. The measurement can be made more accurately than if the diameter was throughout the same as at the top. In other cases the water is poured out of the cylinder into a measuring vessel. If the rain-gauge was sunk so that the top was level with the ground, rain falling outside the gauge would splash into it and vitiate the readings unless the gauge was surrounded by a trench. Ordinarily the top of the gauge is from 1 to 3 feet above the ground. When it is 1 metre above the ground the rain registered is said to be on the average about 6 per cent. less than it should be, owing to the fact that wind causes eddies and currents and carries away drops which should have fallen into the gauge. The velocity of the wind increases with the height above the ground, and so does the error of the rain-gauge. Devices for getting rid of the eddies have been invented by Boernstein and Nipher (Ency. Brit., Tenth Edition, vol. xviii.), but they have not yet come into general use. The Boernstein device is being used experimentally at Eskdalemuir. It would appear that much splashing cannot take place when the ground is covered with grass, and that in such a case the top of the gauge could be 1 foot above the ground, thus making the error very small.

If the ground is at first level, then rises and then again becomes level, a rain-gauge at the foot of the slope will, with the prevailing wind blowing up the slope, register too much, and a rain-gauge just beyond the top of the slope will register too little (Ency. Brit., Tenth Edition, vol. xxxiii.).

4. Influence of Forests and Vegetation.—When the ground is covered with vegetation, and especially forests, the humus or mould formed from leaves, etc., absorbs and retains moisture. It acts like a reservoir, so that the run-off takes place slowly and the denudation and erosion of the soil is checked. The roots of the trees or other vegetation also bind the soil together. Vegetation and forests thus mitigate the severity of floods and reduce the quantity of silt brought into the streams. They also shield the ground from the direct rays of the sun and so reduce evaporation, and thus, on the whole, augment the available rainfall. Forests render the climate more equable and tend to reduce the temperature, and they thus, at least on hills, increase the actual rainfall to some extent.

If a forest is felled and replaced by cultivation, the ploughing of the soil acts in the same way as the humus of the forest, and the crops replace the trees; and it has been stated that in the U.S.A. the cultivation is as beneficial as the forests in mitigating floods and checking denudation of the soil (Proc. Am. Soc. C.E., vol. xxxiv.). But when forests are felled they are not, at least in hilly country, always replaced by cultivation. Measures to put a stop to the destruction of forests or to afforest or reforest bare land may enter into questions of the régime of streams or the supply of water. On the Rhine, increase in the severity of floods was distinctly traced to deforestation of the drainage area.

It is usually said that forests act as reservoirs by preventing snows from melting. This is disputed in the paper above quoted, and it is stated that in the absence of forests the snow forms drifts of enormous depth, and these melt very gradually and act as reservoirs after the snow in the forests has disappeared.

5. Heavy Falls in Short Periods.—When rain water, instead of being stored or utilised, has to be got rid of, it is of primary importance to estimate roughly—exact estimates are impossible—the greatest probable fall in a short time. This bears a rough ratio to the mean annual fall. The maximum observed falls in twenty-four hours range, in the United Kingdom, generally from ·05 to ·10 of the mean annual fall—but on one occasion the figure has been ·20,—and in the tropics from ·10 to ·25. Actual figures for particular places can be extracted from the rain registers, but the probability of their being exceeded must be taken into account. The greatest fall observed in twenty-four hours in the United Kingdom is 7 inches, and in India 30 inches in the Eastern Himalayas.

But much shorter periods than twenty-four hours have to be dealt with. The following figures are given by Chamier (Min. Proc. Inst. C.E., vol. cxxxiv.) as applicable to New South Wales, and he considers that they are fair guides, erring on the side of safety, for other countries:—

The above figures are probably safe for England. For India the case is far otherwise. The following falls have been observed there:—

The falls of 1 inch in ten minutes were frequently observed near the head of the Upper Jhelum Canal, a place where the annual rainfall is not more than 30 inches (see also [Chap. XII., Art. 1]). In some parts of the Eastern Himalayas, where 30 inches of rain has fallen in a day, it is possible that 8 inches may have fallen in an hour. In England 4 inches has fallen in an hour. The heaviest falls in short periods do not usually occur in the wettest years, and they may occur in very dry years. Nor do they always occur on a very wet day.

CHAPTER III
COLLECTION OF INFORMATION CONCERNING STREAMS

1. Preliminary Remarks.—The information which is required concerning streams depends on the character of the stream and on the nature of the work which is to be done. For the present let it be supposed that the stream is large and perennial. Other kinds of streams will be dealt with in Arts. 6 and 7. In dealing with a large perennial stream it is nearly always necessary to know the approximate highest and lowest water-levels, and these can generally be ascertained by local inquiry, combined with observations of water marks; but a higher level than the highest known and a lower level than the lowest known are always liable to occur, and must to some extent be allowed for. If navigation exists or is to be arranged for, the highest and lowest levels consistent with navigation must be ascertained. The highest such level depends chiefly on the heights of bridges. A plan to a fairly large scale is also necessary in most cases.

If an embankment to keep out floods is to be made along a river which is so large that its flood-level cannot be appreciably affected by the construction of the work, it may be necessary to obtain information only as to the actual flood-levels, and as to the extent to which the stream is liable to erode its bank. If a length of the bank of a stream has to be protected against scour, it is necessary to know of what materials the bed and bank are composed, and whether the channel is liable to changes and to what extent. It is also desirable to know to what extent the water transports solids, if any. In some kinds of protective work these solids are utilised.

But in cases where the stream is to be much interfered with, it is necessary to have full information concerning it, not only as regards water-levels, changes in the channel, and transport of solids, but as regards the longitudinal profile and cross-sections, and the discharges corresponding to different water-levels. The collection of some of this information, particularly as to the water-levels and discharges at different times of the year and in floods, may occupy a considerable time.

Methods of ascertaining the quantity of silt carried in the water of a stream are described in [Chap. IV., Art. 4]. Remarks regarding the other kinds of information required—the stream being still supposed to be large and perennial—are given in Arts. 2 to 5 of this chapter. The degree of accuracy required in the information depends, however, on the importance of the work, and sometimes the procedure can be simplified. Detailed remarks on gauges and on the instruments used and methods adopted for observing discharges and surface slopes, are given in Hydraulics, Chap. VIII. and Appendix H.

2. Stream Gauges.—Unless the stream being dealt with is an artificial one, it is unlikely that the flow in the reach with which the work is concerned will be uniform. The rise and fall of the water at one place cannot therefore be correctly inferred from those at another. It will be desirable to have two gauges, either read daily or else automatically, recording the water-level, one near each end of the reach concerned, with intermediate gauges if the reach is very long. If, in or near the reach, there is already a gauge which has been regularly read, it may be sufficient to set up only one new gauge, and to read it only for such a period of time as will give a good range of water-level, and to compare the readings with those of the old gauge. The readings of the new gauge for water-levels outside the range of those observed can then be inferred, but if the stream is very irregular this may involve some trouble (Art. 4).

In the case of a large stream which shifts its course, the reading of a gauge does not give a proper indication of the water-level. In other words, the distance of the gauge from the two ends of the reach is subject to alteration. The case is the same as if the stream was stable and the gauge was shifted about. In such a stream there ought, if accuracy is required, to be a group of two or more gauges for each point where there would be only one if the stream was not a shifting one. Also, owing to erosion of the bank or the formation of a sandbank, it may often be necessary to shift the gauge. When possible it should be kept in a fixed line laid down at right angles to the general direction of the stream. When shifted, its zero level should be altered in such a way that the reading at the new site at the time of shifting is the same as it was before shifting. When the gauge is moved back to the original site its zero should be placed at its original level, though this may give rise to a sudden jump in the reading for the reason given in the first sentence of this paragraph.

3. Plan and Sections.—Making a survey and plan, and laying down on it the lines for longitudinal and cross-sections, and taking levels for the sections, are ordinary operations of surveying. If any land is liable to be flooded, its boundaries should be shown on the plan and on some of the cross-sections. Unless the water is shallow, it is necessary to obtain the bed levels from the water-level by soundings, the level of a peg at the water-level having been obtained by levelling. All the sections should show the water-level as it was at some particular time, but the water-level will probably have altered while the survey was in progress, and allowance must be made for this. The pegs at all the cross-sections and on both banks of the stream—for the water-levels at opposite banks may not be exactly the same—may, for instance, be driven down to the water-level when it is steady, and thereafter any changes in it noted and the soundings corrected accordingly.

In order to ascertain what changes are occurring in the channel it may be necessary to repeat the soundings at intervals and, if there is much erosion of the bank, to make fresh plans.

4. Discharge Observations.—For a large stream it is necessary to observe the discharges by taking cross-sections and measuring the velocity. If there is a sufficient range of water-levels, it will be possible to make actual observations of a sufficient number of discharges. If soundings cannot, owing to the depth or velocity, be taken at high water, they must be inferred from those previously taken, but this does not allow for changes in the channel, which are sometimes considerable and rapid. If there is not a sufficient range of water-level, the discharges for some water-levels must be calculated from those at other water-levels. In this case observations of the surface slope will be required, and the discharge site should be so selected that no abrupt changes in the channel will come within the length over which the observations are to extend. This length should be such that the fall in the water surface will be great enough to admit of accurate observation. If the cross-section of the stream is nearly uniform throughout the whole of this length, or if it varies in a regular manner, being greatest at one end of the length and least at the other end. the differences in the areas of the two end sections being not more than 10 or 12 per cent., then the velocity and cross-section of the stream can be observed in the usual manner at the centre of the length; but otherwise they should be observed at intervals over the whole length, or at least in two places, one where the section is small and one where it is great, and the mean taken. Or the velocity can be observed at only one cross-section and calculated for the others by simple proportion and the mean taken. The coefficient C can then be found from the formula C = V/√(RS). To find the discharge for a higher or lower water-level, the change in the value of C corresponding to the change in R can be estimated by looking out the values of C in tables, and the discharge calculated by using the new values of C and R and the new sectional area, S remaining unaltered. But if the channel is such that, with the new water-level, a change in S is likely to have occurred, this change must be allowed for. Any such change will be due to the changed relative effects of irregularities, either in the length over which the observations extended or downstream of that length. The effect of irregularities in the bed is greatest at low water. The effect of lateral narrowings is greatest at high water. Since a change of 10 per cent. in S causes a change of only 5 per cent. in V, it will usually suffice to draw on the longitudinal section the actual water surface observed and to sketch the probable surface for the new water-level. If the whole channel is fairly regular for a long distance downstream of the discharge site, no slope observations need be made nor need several sections be taken in order to find V. The changed value of C should, however, be estimated in the manner above indicated. For this purpose any probable value of S will suffice.

5. Discharge Curves and Tables.—Ordinarily it will be possible, by plotting the observed discharges as ordinates, the gauge readings being the abscissæ, to draw a discharge curve and from it construct a discharge table. Unless the channel is of firm material and not liable to change, there are likely to be discrepancies among the observed discharges, so that a regular discharge curve will not pass through all the plotted points. If the discrepancies are not serious, they can be disregarded and the curve drawn so as to pass as near as possible to all the points, but otherwise trouble and uncertainty may arise. The soundings should be compared in order to see whether changes have occurred in the channel. If such changes do not account for the discrepancies, the cause must be sought for in some of the recorded velocities. If no sources of error in these can be found, such as wind, it is possible that the velocity has been affected by a change in the surface slope owing to some change in the channel downstream of the length. Failing this explanation, the discrepancies must be set down to unknown causes. With an unstable channel and where accuracy is required, the sectional areas and velocities should be regularly tabulated or plotted so that changes may be watched and investigated. To do this it may be necessary to take surface slope observations, or to set up extra gauges which will show any changes in the slope.

If, downstream of the discharge site, there is any place where affluents come in and bring varying volumes of water, or where gates or sluices are manipulated, and if the influence of this extends up to the discharge site, the water-level there no longer depends only on the discharge, and a discharge table must be one with several columns whose headings indicate various conditions at the place where the disturbances occur.

In order to show how the gauge readings and discharges vary from day to day throughout the year, a diagram should be prepared showing the gauge readings and discharges as ordinates, the abscissæ being the times in days starting from any convenient date as zero. Such a diagram, showing only gauge readings, is given in [fig. 56], [Chap. XII.]

6. Small Streams.—Small streams will now be considered, those, for instance, which are too small to be navigable and which occasionally run dry or nearly dry. If the water of the stream is to be stored for water supply, power or irrigation purposes, full information as to discharges and silt carried will be required. If the stream is small enough the discharges can be ascertained by means of a weir of planks. The discharge is then known from the gauge readings. Cross-sections and large scale plans will not be required unless the stream is to be altered or embanked. If the water, instead of being stored, is to be got rid of, as in drainage work, the only information required as to discharges is the maximum discharge. Large scale plans, sections, or information as to silt or water-levels (except as a means of estimating the discharge) will not be required unless the stream is to be altered or embanked.

In all these cases of small streams the information required is generally, as has been seen, less than in the case of large perennial streams, but it is generally more difficult to obtain. If the stream is ill-defined or its flow intermittent, especially if it is also very small and the place sparsely inhabited, it may be difficult to obtain any discharge figures except those based on figures of rainfall. The method of obtaining such figures has been stated in [Chap. II.] The figures required are those of the annual and monthly fall when the water is to be stored, and those of the greatest fall in a short period when the water is to be got rid of. Of course a plan of the catchment area is required.

7. Intermittent Streams.—In the case of large streams whose flow is intermittent, the information required will, as before, depend upon the circumstances. Such streams occur in many countries. The difficulty in obtaining information is often very great. To obtain figures of daily discharge a gauge must be set up in the stream and a register kept. The chief difficulty in an out-of-the-way place is likely to be the obtaining correct information as to the maximum discharge. Information, derived from reports or from supposed flood marks, as to the highest water-level, may be inaccurate, and information based on rainfall figures may be extremely doubtful owing to the large size of the catchment area, the absence of rain gauges, and the difficulty, especially if the rain is not heavy, in estimating the available fall. All sources of information must be utilised and, whenever possible, observations should be made over a long period of time.

8. Remarks.—Very much remains to be done in collecting and publishing information concerning the ratio of the discharges to the rainfall. By observing a fall of rain and the discharge of a stream before and after the fall, it is possible to ascertain the figures for that occasion, but they will not hold good for all occasions. Continuous observations are required. The chief obstacle is the expense. Not only have measuring weirs and apparatus for automatically recording the water-level to be provided, but the weirs would often cause flooding of land involving payment of compensation. The most suitable places for making observations are those where reservoirs for water-works exist or are about to be made.

CHAPTER IV
THE SILTING AND SCOURING ACTION OF STREAMS

1. Preliminary Remarks.—When flowing water carries solid substances in suspension, they are known as “silt.” Material is also moved by being rolled along the bed of the stream. The difference between silt and rolled material is one of degree and not of kind. Material of one kind may be rolled and carried alternately. The quantity of silt present in each cubic foot of water is called the “charge” of silt. Silt consists chiefly of mud and fine sand; rolled material of sand, gravel, shingle, and boulders. When a stream erodes its channel, it is said to “scour.” When it deposits material in its channel, it is said to “silt.” Both terms are used irrespective of whether the material is silt or rolled material. A stream of given velocity and depth can carry only a certain charge of silt. When it is carrying this it is said to be “fully charged.”

If a stream has power to scour any particular material from its channel, it has power to transport it; but the converse is not true. If the material is hard or coherent, the stream may have far more difficulty in eroding it than in merely keeping it moving. And there is generally a little more difficulty even when the material is soft.

Silting or scour may affect the bed of a channel or the sides or both. The channel may thus decrease or increase in width or—if one bank is affected more than, or in a different manner to, the other—alter its position laterally whether or not it is altering its bed level, and vice versa.

The cross-section of a stream is generally “shallow,” i.e. the width of the bed is greater than the combined submerged lengths of the sides, and the action on the bed is generally greater than on the sides.

Silting and scouring are generally regular or irregular in their action according as the flow is regular or irregular, that is, according as the channel is free or not from abrupt changes and eddies. In a uniform canal fed from a river, the deposit in the head reach of the canal forms a wedge-shaped mass, the depth of the deposit decreasing with a fair approach to uniformity. Salient angles or places alongside of obstructions are most liable to scour, and deep hollows or recesses to silt. Eddies have exceptionally strong scouring power. Immediately downstream of an abrupt change scour is often severe. An abrupt change is one, whether of sectional area or direction of flow, and whether or not accompanied by a junction or bifurcation, which is so sudden as to cause eddies. The hole scoured alongside of an obstruction may extend to its upstream side, though there is generally little initial tendency to scour there. An obstruction is anything causing an abrupt decrease in any part of the cross-section of a stream, whether or not there is a decrease in the whole cross-section, e.g. a bridge pier or spur.

Most streams vary greatly at different times both in volume and velocity and in the quantity of material brought into them. Hence the action is not constant. A stream may silt at one season and scour at another, maintaining a steady average. When this happens to a moderate extent, or when the stream never silts or scours appreciably, it is said to be in “permanent régime,” or “stable.” Most streams in earthen channels are either just stable and no more, or are unstable.

Waves, whether due to wind or other agency, may cause scour, especially of the banks. Their effect on the bed becomes less as the depth of water increases, but does not cease altogether at a depth of 21 feet, as has been supposed. Salt water possesses a power of causing mud, but not sand, to deposit.

Arts. 2, 3, and 6 of this chapter refer to action on the bed of a stream. Action on the sides will be considered in Art. 7.

Weeds usually grow only in water which has so low a velocity that it carries no silt to speak of, but if any silt is introduced the weeds cause a deposit. The weeds also thrive on such a deposit.

2. Rolled Material.—If a number of bodies have similar shapes, and if D is the diameter of one of them and V the velocity of the water relatively to it, the rolling force is theoretically as V² D², and the resisting force or weight as D³. If these are just balanced, D varies as V², or the diameters of similarly shaped bodies which can just be rolled are as V² and their weights as V⁶. From practical observations, it seems that the diameters do not vary quite so rapidly as they would by the above law, the weights being more nearly as V⁵.

Let a stream of pure water having a depth D, and with boulders on its bed, have a velocity V just sufficient to move them very slowly. Any larger boulders would not be moved. Any smaller boulders would move more quickly. Similarly, fine sand would be rolled more quickly than coarse sand. If the velocity of the stream increases, larger boulders would be moved. Streams are thus constantly sorting out the materials which they roll. If the bed is examined it will be found that large boulders exist only down to a certain point, smaller boulders, shingle, gravel, coarse sand and fine sand following in succession.

If the water, instead of being pure, is supposed to contain silt, this may affect its velocity—it is not, however, known to do so—but, given a certain velocity, it is not likely that the rolling power of the stream is much affected by its containing silt.

It is sometimes supposed that increased depth gives increased rolling power, because of the increased pressure, but this is not so. The increased pressure due to depth acts on both the upstream and downstream sides of a body. It is moved only by the pressure due to the velocity.

When sand is rolled along the bed of a stream there is usually a succession of abrupt falls in the bed. After each fall there is a long gentle upward slope till the next fall is reached. The sand is rolled up the long slope and falls over the steep one. It soon becomes buried. The positions of the falls of course keep moving downstream. The height of a fall in a large channel is perhaps 6 inches or 1 foot, and the distance between the falls 20 or 30 feet. A fall does not usually extend straight across the bed but zigzags.

It has sometimes been said that the inclination of the bed of a stream, when high, facilitates scour, the material rolling more easily down a steep inclined plane. The inclination is nearly always too small to have any appreciable direct effect. The inclination of the surface of the stream of course affects its velocity, and this is the chief factor in the case.

A sudden rise in the bed of a stream does not necessarily cause rolled materials to accumulate there, except perhaps to the extent necessary to form a gentle slope. Frequently even this slope is not formed, especially if the rolled material is only sand. The eddies stir it up and it is carried on. The above remarks apply also to weirs or other local rises in the bed.

3. Materials carried in Suspension.—It has long been known that the scouring and transporting power of a stream increases with its velocity. Observations made by Kennedy have shown that its power to carry silt decreases as the depth of water increases (Min. Proc. Inst. C.E., vol. cxix.). The power is probably derived from the eddies which are produced at the bed. Every suspended particle tends to sink, if its specific gravity is greater than unity. It is prevented from sinking by the upward components of the eddies. If V is the velocity of the stream and D its depth, the force exerted by the eddies generated on 1 square foot of the bed is greater as the velocity is greater, and is probably as V² or thereabouts. But, given the charge of silt, the weight of silt in a vertical column of water whose base is 1 square foot is as D. Therefore the power of a stream to support silt is as V² and inversely as D. The silt charge which a stream of depth D can carry is as V^½. V is called the “critical velocity” for that depth, and is designated as V₀.

The full charge must be affected by the nature of the silt. The specific gravity of fine mud is not much greater than that of water, while that of sand is about 1·5 times as great. Moreover, the particles of sand are far larger than the particles of mud. If two streams of equal depths and velocities are fully charged, one with particles of mud and the other with particles of sand, the latter will sink more rapidly and will have to be more frequently thrown up. They will be fewer in number. From some observations referred to by Kennedy (Punjab Irrigation Paper, No. 9, “Silt and Scour in the Sirhind Canal,” 1904), it appears that in a fully charged stream which carried 1/3300 of its volume of a mixture of mud and sand of various grades, sand of a particular degree of coarseness formed only 1/35,000 of the volume of the water, but that when the same stream was clear and was turned on to a bed of the coarse sand it took up 1/15,000 of its volume. It would thus appear that the full charge of silt is less as its coarseness and heaviness are greater. This is in accordance with the laws mentioned above (Art. 2, par. 1). See also [Chap. V., Art. 2], last paragraph.

It is probable that fine mud is carried almost equally into all parts of the stream, whereas sand is nearly always found in greater proportion near the bed and, as before remarked, some materials may be rolled and suspended alternately. The charge of mixed silt which a stream can carry is, no doubt, something between the charge which it can carry of each kind separately, but the laws of this part of the subject are not yet fully known. From the observations above referred to, Kennedy concludes that a canal with velocity V₀ will carry in suspension 1/3300 to 1/5000 of its volume of silt, according as it is charged with sand of all classes or only with the heavier classes.

Let a stream be carrying a full charge of any kind of silt. Then if there is any reduction in velocity, a deposit will occur—unless there is also a reduction of depth—until the charge of silt is reduced again to the full charge for the stream. The deposit generally occurs slowly, and extends over a considerable length of channel. The heavier materials are, of course, deposited first. If a stream is not fully charged, it tends to become so by scouring its channel. It is generally believed that a stream fully charged with silt cannot scour silt from its channel, or bear any introduction of further silt. This seems to be correct in the main, but the remarks made in the latter part of the preceding paragraph must be taken into consideration.

It has been stated (Art. 2) that a weir or a sudden rise in the bed does not necessarily cause an accumulation of rolled material. It never causes a deposit of suspended material unless it causes a heading up and reduction of velocity to below the critical velocity.

4. Methods of Investigation.—The quantity of silt in water is found by taking specimens of the water and evaporating it or, if the silt is present in great quantity, leaving it to settle for twelve hours—an ounce of alum can be added for every 10 cubic feet of water to accelerate settlement—drawing off the water by a syphon, and heating the deposit to dry it. The deposit is then measured or weighed. It is best to weigh it. If clay is filled into a measure, the volume depends greatly on the manner in which it is filled in. When silt deposits in large quantities in a channel, or when heavy scour occurs, the volume deposited or scoured is ascertained by taking careful sections of the channel.

Fig. 1.

Silt is best classified by observing its rate of fall through still water. A sand which falls at ·10 feet per second is, in India, called class (·1), and mixed sand which falls at rates varying from ·1 to ·2 feet per second is called class ·1/·2. Fig. 1 shows a sand separator designed by Kennedy. The scale is ⅛. It has a syphon action, and the rate of flow can be altered by altering the length of the exit pipe. Suppose it is desired to measure the sand of class (·10) and all heavier kinds. The pipe is adjusted so as to give a velocity of ·1 foot per second to the upward flowing water, which then carries off all silt of class (·10) or finer. All heavier silt falls into the glass tube. It can be separated again by being mixed with water and passed through the instrument again, the velocity of flow through the instrument being increased.

The quantity of silt present at various depths can be found by pumping specimens of water through pipes. At each change of depth the pipe, delivery hose, etc., should be cleaned. Allowance must be made for the velocity of ascent of the water up the pipe. Suppose this to be 1·4 feet per second. Then the velocity of sand of class (·2) would be 1·2 feet per second, and the quantity of sand actually found in the water would have to be increased by one-sixth.

5. Quantity and Distribution of Silt.—The quantity of silt present in water varies enormously. Fine mud, even though sufficient to discolour the water, may be so small in volume that it only deposits when the water is still, and even then deposits slowly. In the river Tay, near Perth, the silt was found to be ordinarily 1/10,000 of the volume of water, and at low water only 1/28,000. In the river Sutlej at Rupar, near where it issues from the Himalayas, the silt in the flood season is extremely heavy. Out of 360 observations, made at various depths, during the flood seasons of four successive years, in water whose depth ranged up to 12 feet, the silt was once found to be 2·1 per cent. by weight of that of the water. It was more than 1·2 per cent. on four occasions, and more than 0·3 per cent. (or 3 in 1000) on sixty-four occasions. Generally about one-half of the silt was clay and sand of classes finer than (·10), about one-third was sand of class ·1/·2, and the residue was sand of class ·2/·3. The sand of the river Chenab is generally coarser than that of the Sutlej. There are very great differences in the degree of coarseness of river sand. The sand in any river becomes finer and finer as the gradient flattens in approaching the sea. Sea sand has been found to be of class (·20). In the Sirhind Canal, which takes out from the Sutlej at Rupar, the maximum quantity of suspended silt observed in the four flood seasons was 0·7 per cent., on one occasion out of 270, and it exceeded 0·3 per cent. on twenty-five occasions. About 80 per cent. of the silt was clay.

In another part of the paper quoted, it is stated that the silt suspended in the canal water averaged, during the whole of one flood season, about 1/1700 of the volume of the water. This would be about 1/1200 by weight. The silt deposited in the bed of the canal, in a period of a few days, was sometimes as much as 1/1000 of the water which had passed along, and occasionally as much as 1/500. It was nearly all sand, only about 3 per cent. being clay. Silt of classes finer than (·1) gave no trouble, and were to be eliminated in future investigations. In a canal, as in a river, the sand on the bed becomes finer the further from the head.

Regarding the distribution of the silt at various depths, in water 5 to 17 feet deep, the quantity of silt near the bed may, when the charge is heavy and consists of mixed silt, be 1¼ to 3 times that at the surface. If the charge is fine mud, there is likely to be as much silt at the surface as near the bed, if sand, there may be none at the surface and little in the upper part of the stream.

In all cases single observations are likely to show extraordinarily discordant results; a number of observations must be made at each point and averaged.

6. Practical Formulæ and Figures.—A stream which carries silt generally rolls materials along its bed. The proportion between the quantities of material rolled and carried is never known, and this makes it impossible to frame an exact formula applicable to such cases, but Kennedy, from his observations on canals fully charged with the heavy silt and fine sand usually found in Indian rivers near the hills, arrived at the empirical formula for critical velocities V = ·84 D^·64
The observations were made on the Bari Doab Canal and its branches, the widths of the channels varying from 8 feet to 91 feet, and the depths of water from 2·3 feet to 7·3 feet. The beds of these channels have, in the course of years, adjusted themselves by silting or scouring, so that there is a state of permanent régime, each stream carrying its full charge of silt, and the charges in all being about equal. From further observations referred to above (Art. 3, par. 2) it appears that this kind of silt forms about 1/3300 of the volume of the water, and that on the Sirhind Canal, sand coarser than the (·10) class, formed 1/35,000 of the volume of water.

The formula gives the following critical velocities for various depths:—

In Indian rivers not near the hills the silt carried is not so heavy, and the critical velocities are supposed to be about three-fourths of the above. Thrupp (Min. Proc. Inst. C.E., vol. clxxi.) gives the following ranges of velocities as those which will enable streams to carry different kinds of silt. It does not appear that the streams would be fully charged except at the higher figure given for each case.

It cannot be said that the exact relations between D and V are yet known, but it is of great practical importance to know that V must vary with D. The precise manner in which it must vary does not, for moderate changes, make very much difference. In designing a channel a suitable relation of depth to velocity can be arranged for, and one quantity or the other kept in the ascendant, according as scouring or silting is the evil to be guarded against.

The old idea was that an increase in V, even if accompanied by an increase in D, e.g. simply running a higher supply in a given channel, gave increased silt-transporting power. In a stream of very shallow section this is probably correct, for V increases faster than D^·64 (Hydraulics, Chap. VI., Art. 2). In a stream of deep section a decrease in D gives increased silt-transporting power. If the discharge is fixed, a change in the depth or width must be met by a change of the opposite kind in the other quantity. In this case widening or narrowing the channel may be proper according to circumstances. In a deep section widening will decrease the depth of water, and may also increase the velocity, and it will thus give increased scouring power. In a shallow section, narrowing will increase the velocity more than it increases D^·64. In a medium section it is a matter of exact calculation to find out whether widening or narrowing will improve matters.

If the water entering a channel has a higher silt-charge than can be carried in the channel, some of it must deposit. Suppose an increased discharge to be run, and that this gives a higher silt-carrying power and a smaller rate of deposit per cubic foot of discharge, it does not follow that the deposit will be less. The quantity of silt entering the channel is now greater than before. Owing to want of knowledge regarding the proportions of silt and rolled material, and to want of exactness in the formulæ, reliable calculations regarding proportions deposited cannot be made.

The channels in which the observations above referred to were made have all assumed nearly rectangular cross-sections, the sides having become vertical by the deposit on them of finer silt; but the formula probably applies approximately to any channel if D is the mean depth from side to side, and V the mean velocity in the whole section.

If the ratio of V to D differs in different parts of a cross-section, there is a tendency towards deposit in the parts where the ratio is least, or to scour where it is greatest. There is a tendency for the silt-charge to adjust itself, that is, to become less where the above ratio is less, but the irregular movements of the stream cause a transference of water among all parts, and this tends to equalise the silt-charge.

Dubuat gives the following as the velocities close to the bed which will enable a stream to scour or roll various materials. The bed velocity is probably less than the mean velocity in the ratio of about ·6 to 1 in rough channels, and about ·7 to 1 in smooth channels:—

The figures for brick, earthenware, and rock can apply only to materials of exceedingly poor quality. Masonry of good hard stone will stand 20 feet per second, and instances have occurred in which brickwork has withstood a velocity of 90 feet per second without injury so long as the water did not carry sand and merely flowed along the brickwork. If there are abrupt changes in the stream, causing eddies, or if there is impact and shock, or if sand, gravel, shingle, or boulders are liable to be carried along, velocities must be limited.

7. Action on the Sides of a Channel.—It has been seen that the laws of silting and scour on the bed of a channel depend on the ratio of the depth to the velocity. The same laws probably hold good in the case of a gently shelving bank, so that here again V ought to vary as D^·64. The velocity near the angle where the slope meets the water surface seems to decrease faster than D^·64. At all events, silt tends to deposit in the angle and the slope to become steep.

When the slope is steep the law seems to be different, the tendency for deposit or scour to occur on the bank depending on the actual velocity without much relation to the depth. The velocity very near to a steep bank is always low relatively to that in the rest of the stream. Thus there is often a tendency for silt to deposit on the bank, especially in the upper part, and for the side to become vertical except for a slight rounding at the lower corner. A bank may receive deposits when the bed may be receiving none, and it may have a persistent tendency to grow out towards the stream. The growth of the bank is generally regular, the line of the bank being preserved, but it may be irregular, especially if vegetation, other than small grass, becomes established on the new deposits.

When scour of the sides of a channel occurs it may occur by direct action of the stream on the sides near the water-level, or by action at or near the toe of the slope, which causes the upper part of the bank to fall in. Such falling in is generally more or less irregular, and the bank presents an uneven appearance. The fallen pieces of bank may remain, more or less intact, especially if they are held together by the roots of grasses, etc., where they fell, and prevent further scour occurring along the toe of the slope. Falling in of banks is most liable to occur in large streams and with light soils. It may be caused by the waves which are produced by steamers and boats or, especially in broad streams, by wind. The action on the banks at bends is discussed in Art. 8.

Thus in designing a channel according to the principles laid down in Art. 6, the question of action on the sides of the channel has to be dealt with as follows. Whether or not the velocity is to be low, relatively to the depth, i.e. whether or not deposit on the bed is more likely to occur than scour, care can be taken not to make it actually too low, and not to make it actually too high, particularly if the soil is light and friable. With ordinary soils a mean velocity of 3·3 feet per second in the channel is generally safe as regards scour of the sides. Any velocity of more than 3·5 feet per second may give trouble. A velocity of less than 1 foot per second is likely to give rise to deposit on the sides.

In channels in alluvial soils the falling in of banks is sometimes said to occur more when the stream is falling than at other times. This has been noticed on both the Mississippi and the Indus. The cause has been said to be the draining out of water which had percolated into the bank, the water in flowing out carrying some sand with it. The effect of this cannot however be great.

8. Action at Bends.—At a bend, owing to the action of centrifugal force and to cross-currents caused thereby, there is a deposit near the convex bank and a corresponding deepening—unless the bed is too hard to be scoured—near the concave bank. The water-level at the concave bank is slightly higher than at the convex bank. The greatest velocity instead of being in mid-stream is nearer the concave bank.

As the transverse current and transverse surface slope cannot commence or end abruptly, there is a certain length in which they vary. In this length the radius of curvature of the bend and the form of the cross-section also tend to vary. This can often be seen in plans of river bends, the curvature being less sharp towards the ends.

When once a stream has assumed a curved form, be it ever so slight, the tendency is for the bend to increase. The greater velocity and greater depth near the concave bank react on each other, each inducing the other. The concave bank is worn away, or becoming vertical by erosion near the bed, cracks, falls in, and is washed away, a deposit of silt occurring at the convex bank, so that the width of the stream remains tolerably constant. The bend may go on increasing, and it often tends to move downstream.

Fig. 2.

In [fig. 2] the deep places are shown by dotted lines. Along the straight dotted line there is no deep place. Such a line would be used for a ford. At low water it becomes a shoal. This is the chief reason why a tortuous stream at low water consists of alternate pools and rapids. It is sometimes said that deep water occurs near to a steep hard bank. Such deepening is due to bends or obstructions which give the current a set towards the bank, or it is due to irregularities in the bank which cause eddies. In a straight channel with even and regular banks there is no such deepening.

When a bend has formed in a channel previously straight, the stream at the lower end of the bend, by setting against the opposite bank, tends to cause another bend of the opposite kind to the first. Thus the tendency is for the stream to become tortuous and, while the tortuosity is slight, the length, and therefore the slope and velocity, are little affected; but the action may continue until the increase in the length of the stream materially flattens the slope, and the consequent reduction in velocity causes erosion to cease. Or the stream during a flood may find, along the chord of a bend, a direct route with, of course, a steeper slope. Scouring a channel along this route it straightens itself, and its action then commences afresh. Short cuts of this kind do not, however, occur so frequently as is sometimes believed. In some streams the bends acquire a horse-shoe shape and the neck becomes very narrow and short cuts may then occur. Otherwise they are not common. V increases only as √(S), and if the country is covered with vegetation it is not easy for a stream to scour out a new channel.

The effect of bends on the velocity of a stream is not well understood. In case of a bend of 90° the increased resistance to flow when the bend is absolutely sudden (a sudden bend is known as an “elbow”) amounts perhaps to V²/(2g). Whether it is greater or less in the case of a gentle bend of 90° is not known. In the case of a pipe there is a certain radius which gives a minimum resistance (Hydraulics, Chap. V.). The increased resistance at a bend is due partly to the fact that the maximum velocity is no longer in the centre of the stream, and partly to the fact that the velocities at the different parts of the cross-section have to be rearranged at the commencement of the bend and again at its termination. Thus the effect of a bend of 45° is a good deal more than half of that of a bend of 90°. Two bends of 45°, both in the same direction, with a straight reach between them, will cause more resistance than a single bend of 90° with the straight reach above or below the bend. If the two bends of 45° are in different directions the resistance will be still greater. A succession of sharp bends may produce a serious effect, amounting to an increase in roughness of the channel. A succession of gentle bends, of any considerable angle, cannot of course occur within a moderate length of channel.

When there is head to spare there is clearly no objection to bends, except that the bank may need protection. At a place where the bank has in any case to be protected, e.g. at a weir, there is no objection to an elbow.

9. General Tendencies of Streams.—Since the velocity is greater as the area of the cross-section is less, a stream always tends to scour where narrow or shallow, and to silt where wide or deep. The cross-section thus tends to become uniform in size. Suppose two cross-sections to be equal in size but different in shape. The velocities of the two sections will be equal. The tendency of the bed to silt will (Art. 6) be greater at the deeper section and, when silting has occurred on the bed, the section will be reduced and there will be a tendency to scour at the sides. Thus the cross-sections tend to become also uniform in shape. If a bank of silt has formed in a stream, the tendency is for scour to occur. There is also a tendency for silt to deposit just below the point where the bank ends. Hence a silt bank often moves downstream.

Owing to the tendency to scour alongside of, or downstream of, obstructions (Art. 1), it is clear that a stream constantly tends to destroy obstructions.

There is an obvious tendency for silt to deposit where the bed slope of a stream flattens, and for scour to take place where it steepens (Hydraulics, figs. 16 and 17, pp. 24 and 25), and thus the tendency is for the slope to become uniform.

In a natural stream flowing from hilly country to a lake or sea, the slope is steepest at the commencement and gradually flattens. There is thus a tendency for the bed to rise except at the mouth of the stream. This rising tends to increase the slope and velocity in the lower reaches, and this again enhances the tendency, described in the preceding article, of the stream to increase in tortuosity.

When a silt-bearing stream overflows its banks the depth of water on the flooded bank is probably small and its velocity very low, and a deposit of silt takes place on the bank. When the deposit has reached a certain height it acts like a weir on the water of the next flood, which flows quickly over it and, instead of raising it higher, deposits its silt further away from the stream. In this way a strip of country along the stream gradually becomes raised, the raising being greatest close to the stream. The country slopes downwards in going away from the stream. In other words, the stream runs on a ridge. If the bank becomes raised so high that flooding no longer occurs, the raising action ceases, but if, as is likely in alluvial country, the bed of the stream also rises, the action may continue and the ridge become very pronounced.

Some rivers have very wide and soft channels which are only filled from bank to bank in floods, if then. The deep stream winds about in the channel, and the rest of it is occupied by sandbanks and minor arms. The winding is the result of the velocity being too great for the channel. The streams, especially the main stream, constantly shift their courses by scouring one bank or the other. Now and then the main stream takes a short cut, either down a minor arm or across an easily eroded sandbank. This is a very different matter from a short cut across high ground. The sandbanks receive deposits of silt in floods, but are constantly being cut away at the sides. Such rivers frequently erode their banks to an extraordinary extent. The Indus sometimes cuts into its bank 100 feet or more in a day, and it may cut for half a mile or more without cessation. The tortuosity of such a stream increases as it gets nearer the sea. The actual length of the Indus in the 400 miles nearest the sea is 39 per cent. greater than its course measured along the bank. In the reach from the 600th to the 700th mile from the sea, the difference is only 3 per cent. For a detailed description of some such rivers, see Punjab Rivers and Works.

Sometimes general statements are made regarding silting or scour in connection, for instance, with a stream which is confined between embankments or training walls, or has overflowed its banks or is held up by a weir. It is impossible to say that any such condition, or any condition, will cause silting or scour, unless the velocity depth and silt charge are known.

CHAPTER V
METHODS OF INCREASING OR REDUCING SILTING OR SCOUR

1. Preliminary Remarks.—Most important works which affect the régime of a stream have some effect on its silting or scouring action, but this is not generally their chief object. Such works will be dealt with in due course, and the effects which they are likely to produce on silting or scouring will be mentioned. In the present chapter only those works and measures will be considered whose chief object is to cause a stream to alter its silting or scouring action. It does not matter, so far as this discussion is concerned, whether the object is direct, i.e. concerned only with the particular place where the effect is to be produced, or indirect, as, for instance, where a stream is made to scour in order that it may deposit material further down the stream. The protection of banks from scour is considered in [Chap. VI.] Dredging is dealt with in [Chap. VIII.]

2. Production of Scour or Reduction of Silting.—Sometimes the silt on the bed of a stream is artificially stirred up by simple measures, as, for instance, by scrapers or harrows attached to boats which are allowed to drift with the stream, or by means of a cylinder which has claw-like teeth projecting from its circumference and is rolled along the bed, or by fitting up boats with shutters which are let down close to the bed and so cause a rush of water under them, or by anchoring a steamer and working its screw propeller. It is thus possible to cause a great deal of local scour, but the silt tends to deposit again quickly, and it is not easy to keep any considerable length of channel permanently scoured. The system is suitable in a case in which a local shallow or sandbank is to be got rid of and deposit of silt a little further down is not objectionable. It may be suitable in a case in which the bed is to be scoured while a deposit of silt at the sides of the channel is required, especially if some arrangement to encourage silt deposit at the sides is used (Art. 3, par. 4; also [Chap. VI., Art. 3]).

Holding back the water by means, for instance, of a regulator or movable weir, and letting it in again with somewhat of a rush, will, if frequently repeated, have some effect in moving silt on in the downstream reach. Regarding the upstream reach, it has been remarked ([Chap. IV., Art. 3]) that a weir does not necessarily cause silt deposit. If, in a stream which does not ordinarily silt, a regulator or movable weir causes, when the water is headed up, some silt deposit, the cessation of the heading up not only removes the tendency to silt, but the section of the stream, at the place where the deposit occurred, is less than elsewhere, and there is thus a tendency to scour there. If a regulator is alternately closed and opened, no permanent deposit of much consequence is likely to occur.

A stream may be made to scour its channel by opening an escape or branch. This causes a draw in the stream, and an increase in velocity for a long distance upstream of the bifurcation (Hydraulics, Chap. VII., Art. 6). This procedure is sometimes adopted on irrigation canals. The escape is generally opened in order to reduce the quantity of water passing down, but it may be opened solely to induce scour or prevent silting. The floor of the escape head is usually higher than the bed of the canal, but this does not interfere with operations except at low supplies. It may ([Chap. IV., Art. 2]) have some effect on the quantity of rolled material passed out of the escape.

If there is a weir in the river below the off-take of the canal, and if the escape runs back to the river and thus has a good fall, the scouring action in the canal may be very powerful.

If the main channel has a uniform slope throughout, the slope of its water surface is greater upstream of the escape than downstream of the escape, and there is thus an abrupt reduction of velocity and possibly a deposit of silt in the main channel below the escape. This may or may not be objectionable. In the case of an irrigation canal, it is far less objectionable than deposit in the head of the canal. The best point for the off-take of any escape or scouring channel depends on the position of the deposits in the main channel. The off-take should be downstream of the chief deposits, but as near to all of them as possible. A breach in a bank acts of course in the same way as an escape.

A stream of clear water when sent down a channel will scour it if the material is sufficiently soft. In the case of the Sirhind Canal, it has already been mentioned ([Chap. IV., Art. 3]), that when the river water became clear after the floods the proportion of coarse sand, i.e. sand above the (·10) class, carried by the canal water was about 1/15,000 by volume. This was in the period from 22nd September to 7th October. From 8th to 23rd October the proportion averaged 1/32,000 from 24th October to 8th November 1/44,000, and from 9th to 24th November 1/85,000. The reason of this reduction was that the comparatively clear water kept picking up the sand from the bed and moving it on, the finer kinds being moved most quickly. As the coarse sand left on the bed became less in quantity, the water took up less. It appears, however, that the water also picked up some clay which was left, and that the total suspended silt in November was 1/9000 of the water. All the observations mentioned in this paragraph appear to have been made at Garhi, 26 miles from the head of the canal.

3. Production of Silt Deposit.—Works or measures for causing silt deposit may be undertaken in order to cause silt deposit in specific places where it will be useful, or in order to free the water from silt. Sometimes both objects are combined.

If a stream can be turned into a large pond or low ground—a bank being built round it if necessary—it can be made to part with some or all of its silt whether rolled or suspended. Even if the pond is so large that the velocity becomes imperceptible, the whole of the suspended matter will not deposit unless it has sufficient time, but the matter which remains in the water is likely to be extremely small in amount. The silting up of marshes, pools, borrow-pits, etc., is now being effected, or should be effected, in places where mosquitoes and malaria are prevalent.

In the upper or torrential part of a stream, a high dam, provided with a sluice and a high-level waste weir, may be built across it. The space above the dam becomes more or less filled with gravel, etc. This has been done in Switzerland (Min. Proc. Inst. C.E., vol. clxxi.). In the U.S.A. long weirs have been built in order to stop the progress of detritus from gold mines. Such detritus was liable to choke up rivers and damage the adjoining lands. The detritus from hill torrents can also be reduced by afforestation of the hill sides.

Fig. 3.

Fig. 4.

When a stream is in embankment—irrigation channels are frequently so—the bank can be set back ([fig. 3]), and suspended silt will then deposit on the berms. The object of this arrangement is generally to create very strong banks in low ground. A similar plan can be adopted when the berm is only slightly below the water-level and even when it is only occasionally submerged. In this case the deposit of a small bank of silt along the edge of the berm next the stream will prevent the access of fresh supplies of silt-bearing water to the parts further away. Gaps should be cut in the bank of silt at intervals, and cross banks made to form “silting tanks,” as shown in [fig. 4]. The inlets to the tank should be large, and the outlets small, so that the water in the tank may have little velocity. It is not, however, correct to have the outlet so small—unless the water contain very little silt—that there is very little flow through the tank. The tanks will generally be silted up most quickly by allowing a good flow through them, even though only a small proportion of the silt in the water is deposited. Regular banks arranged to form tanks on the above principle can be made behind the original banks of a canal in cases where the original banks were not, for any reason, set back.

When a channel is made in low ground and the excavation is not sufficient to make the banks, borrow-pits can be dug in the bed of the channel. Such pits should not be long and continuous, but wide bars should be left so that a number of short pits will result. These pits will trap rolled material as well as suspended silt. The object in this case is to free the water from silt and to reduce the size of the channel and thus reduce the loss of water from percolation.

On the Indus, where it has a strong tendency to shift westwards, long earthen dams or groynes are run out from the west bank across the sandbanks. One object is to cause silt deposit, and so increase the quantity of material which the river will have to cut away, but whether this result is achieved is doubtful. The sandbanks receive deposits in any case. A groyne may increase the deposit on its upstream side, but it cuts off the flood water from its downstream side and so reduces the deposit there.

4. Arrangements at Bifurcations.—At a bifurcation, as where a branch takes off from a canal, it is possible to reduce the quantity of rolled material entering the canal by raising its bed or constructing a weir or “sill” in its head. This arrangement may have great effect in excluding boulders, shingle, or gravel. As regards rolled sand, it has much less effect than might be expected ([Chap. IV., Art. 2]). If the canal is reduced in width ([fig. 5]) there will be eddies below the bed level of the branch. They will stir up the sand and some of it will enter the branch. If the canal is not reduced in width, eddies will be produced in the surface water, and they will affect the bed.

The above remarks apply also to the case of a canal taken off from a river when there are no works in the river.

Fig. 5.

Fig. 6.

5. A Canal with Headworks in a River.—In the case of a canal taking off from a river and provided with complete headworks, it is possible to do a great deal more. The case of the Sirhind Canal, already referred to ([Chap. IV., Arts. 5] and [6]), is a notable example. The canal ([fig. 6]) is more than 200 feet wide, the full depth of water 10 feet, and the full discharge about 7000 cubic feet per second. In 1893 when the irrigation had developed, and it became necessary to run high supplies in the summer—July, August, and part of September—the increase in the silt deposit threatened to stop the working of the canal. In the autumn and winter, say from 25th September to 15th March, the water entering the canal is clear and much of the deposit was picked up by it, but not all. In the five years 1893 to 1897 inclusive, the following remedial measures were adopted. Increased use was made of the escape at the twelfth mile. This did some good, but there was seldom water to spare. In 1893 to 1894 the sill of the regulator was raised to 7 feet above the canal bed, and it was possible to raise it 3 feet more by means of shutters. This had little effect. The coarsest class of sand was ·4, and the velocity of the water, even of that part of it which came up from the river bed and passed over the sill, was over 2 feet per second, so that all sand was carried over. In 1894 to 1895 the divide wall, which had been only 59 feet long, was lengthened to 710 feet, so as to make a pond between the divide wall and the regulator,[8] but probably the leakage through the under-sluices was often as much as the canal supply, and the water in the pond was thus kept in rapid movement and full of silt. The canal was closed in heavy floods. This did some good, but probably the canal was often closed needlessly when the water looked muddy but contained no excessive quantity of sand. The above comments on the measures taken were made by Mr Kennedy when chief engineer. The above measures did not reduce the silt deposits, but the scour in the clear water season improved, probably because higher supplies were run owing to increased irrigation. The deposit in the upper reaches of the canal, when at its maximum about the end of August of each year, was generally more than twenty million cubic feet. From the year 1900 a better system of regulation was enforced, the under-sluices being kept closed as much as possible, so that there was much less movement in the pond and much less silt in its water. By 1904 the deposit in the canal had been reduced to three million cubic feet, and no further trouble occurred.

During the period from 20th September 1908 to 10th October 1908 the quantity of silt in the canal above Chamkour (twelfth mile) decreased from 19,325,800 cubic feet to 12,477,600 cubic feet. The quantity scoured away was 6,848,200 cubic feet. During this period no silt entered the canal. The quantity which passed out of the reach in question in suspension was 4,183,660 cubic feet, so that 2,664,540 cubic feet of material must have been rolled along the bed. The rolled material was 64 per cent. of the suspended material. During this period the Daher escape, in the twelfth mile, was open, and the mean velocity in the canal just above the escape was about 4 feet per second, the depth of water being about 10 feet. The velocity near the escape was thus greater than the critical velocity for mixed silt ([Chap. IV., Art. 6]), and even a long way up the canal it would be in excess of the critical velocity. The water seems to have carried about 1/1800 of its volume of silt. Whether the above proportions of rolled to suspended matter would hold good in a fully charged stream flowing with the critical velocity it is not easy to say.

As silt deposits in the pond, the velocity of the water in it, along the course of the main current towards the canal, increases and eventually the water begins to carry coarse sand dangerous for the canal. In order to ascertain when this state of affairs has been reached, two methods of procedure are possible. One is to frequently test specimens of the water in the pond along the course of the main current and see when it contains more than 1/15,000 of its volume of coarse sand. This plan would be troublesome and liable to error, and is rejected by Kennedy, who suggests that the depth and velocity of the water in the pond be frequently observed along the course of the main current. As soon as the velocity exceeds the critical velocity for mixed silt, it is time to close the canal and open the under-sluices and scour out the deposit from the pond. The period in which most silt is believed to have been deposited in the canal is the spring and early summer, say from 15th March to 1st July. This is the time when the snows are melting and the river water is clear. It can then carry more sand than in the rains—1st July to 15th September,—when it is muddy.

Kennedy also suggests that some under-sluices should be provided at the far side of the river, i.e. at the right-hand side of the weir. It would then be possible, by opening them, to let floods pass without interfering with the pond.

The two spurs or groynes, shown in the plan, were constructed in 1897 so as to cause the stream to flow along the face of the canal regulator and not allow deposits to accumulate there. The depth of silt deposited in a great part of the pond amounted at times to 8 or 10 feet.

6. Protection of the Bed.—It is possible to afford direct protection from scour to the bed of a stream by constructing walls across it, but unless the walls are near together the protection will not be effective. An arrangement used in some streams in Switzerland consists of tree trunks secured by short piles and resting on brushwood. But as long as the walls are not raised above the bed they cannot entirely stop scour, unless extremely close together. If raised above the bed they form a series of weirs.

The weirs must be so designed that the depth of water in a reach between two weirs is great enough to reduce the velocity down to the critical velocity, or less. The fall in the water surface at each weir being very small, the discharge over the weir can be found by considering it as an orifice extending up to the downstream water surface, and the head being the fall in the surface at the weir.

To stop scour of the bed by direct protection without raising the water-level, the bed can be paved, a plan adopted in artificial channels with very high velocities. The paving can be of stones, bricks, or concrete blocks. The Villa system of protection, which has been used in Italy, France, and Spain, consists of a flexible covering laid on the bed. Prisms of burnt clay or cement are strung on several parallel galvanized iron wires, which are attached to cross-bars so as to form a grid a few feet square. The grids are loosely connected to one another at the corners, and the whole covering adjusts itself to the irregularities of the bed (Min. Proc. Inst. C.E., vol. cxlvii.).

The special protection or paving required in connection with weirs and such-like works is considered in [Chap. X., Arts. 2] and [3].

CHAPTER VI
WORKS FOR THE PROTECTION OF BANKS

1. Preliminary Remarks.—The protection of a length of bank from scour may be effected by spurs, which are works projecting into the stream at intervals, or by a continuous lining of the bank. A spur forms an obstruction to the stream ([Chap. IV.], Art. 1), and when constructed, or even partly constructed, the scour near its end may be very severe, even though there may be little contraction of the stream as a whole. If the bed is soft a hole is scoured out. Into this hole the spur keeps subsiding, and its construction, or even its maintenance, may be a matter of the greatest difficulty. A high flood may destroy it. If it does not do so, it may be because the stream has, for some reason, ceased to attack the bank at that place. A continuous lining of the bank is not open to any objection, and is generally the best method of protection. Spurs made of large numbers of rather small trees, weighted with nets filled with stones, have been used on the great shifting rivers of the Punjab which swallowed up enormous quantities of materials. The use of spurs on such rivers has now, in most cases, been given up. If L is the length of a spur measured at right angles to the bank, the length of bank which it protects is about 7 L—3 L upstream and 4 L downstream,—but the spur has to be strongly built, and its cost is, in many cases, not much less than the expense of protecting the whole bank with a continuous lining.

Whatever method is adopted, a plan, large enough to show all irregularities, should always be prepared, and the line to which it is intended that the bank shall be brought marked on it.

Fig. 7.

Sometimes natural spurs exist as, for instance, where a tree projects into a stream or has fallen into it, and the holes between the spurs may be deep, so that a continuous protection would be expensive. Or there may be trees standing in such positions that, if felled, they will be in good places for spurs. In cases such as the above, spurs may be suitable even in a stream with a soft channel.

Regarding the use of spurs or groynes for diversion works or for reducing the width of a stream, see [Chap. VII., Art. 1], and [Chap. VIII., Art. 3].

2. Spurs.—A spur may be made of—

(a) Loose stone, which may be faced with rubble above low-water level ([fig. 7]).

(b) Layers of fascines weighted with gravel or stones.

(c) Earth or sand closely covered with fascines.

(d) A double line of stakes with fascines or brushwood laid between them ([fig. 8]).

(e) A single line of stakes with planking or basket work on its upstream side, or with twigs or wattle laid horizontally and passed in and out of the stakes, as in [fig. 20].

(f) A single tree with the thick end of the trunk on the bank and with stakes, if necessary, to prevent the current from moving it.

(g) A number of small trees heaped together and weighted with nets full of stones.

(h) A layer of poles and over them a layer of fascines on which are built walls of fascine work so arranged as to form cells or hollow rectangular spaces which become filled with silt.

(i) Large fascines running out into the stream and having their inner ends staked to the bank while the outer ends float, other fascines being added over them and projecting further into the stream, and the whole eventually sinking.

Fig. 8.

Combinations of the above are also used, for instance, (d) or (e) may be used for the upper portion, the foundation being (a) or (c).

Fig. 9.

Fig. 10.

Fig. 11.

Instead of running out at right angles to the bank a spur may be inclined somewhat downstream. This somewhat reduces the eddying and scour round the end. The ends of a system of spurs should be in the line which it is intended that the edge of the stream shall have ([fig. 9]). The tops of short spurs are usually above high flood level. Sometimes spurs are made to slope downwards ([fig. 10]), and they then cause less disturbance of the water and less scour than if built to the form shown by the dotted line. Such spurs are sometimes combined with a low wall running across the bed of the stream, the whole forming a “profile” of the cross-section to which it is intended to bring the channel. Regarding such walls, see [Chap. V., Art. 6]. When a spur is long it may have small subsidiary spurs ([fig. 11]) to reduce the rush of water along it; or its end may have to be protected in the same manner as the advancing end of a closure dam ([Chap. VII., Art. 2]).

The following is a curious case of misconception of the action of spurs. In 1909 the river Indus was eroding its right bank and threatening to destroy the town of Dera Ghazi Khan. A clump of date palms formed a promontory and resisted erosion to some extent. A suggestion was made—by an engineer of eminence who had formerly been consulted in the case—to the effect that the date palms be removed, the reason given being that they caused disturbance and scour. On this principle spurs would have to be made not to protect a bank but to cause it to be eroded.

3. Continuous Lining of the Bank.—The lining or protection of a bank may be of stone or brick pitching (figs. 12 and 13), loose stone ([fig. 14]), fascines ([fig. 15]), turfing, plantations, brushwood, or of other materials laid on the slopes. Before protecting a bank it is best to remove irregularities and bring it to a regular line. This can generally be done most easily by filling in hollows, but sometimes it is done by cutting off projections. It is also necessary to make the side slope uniform. Where the slope is as shown by the dotted lines in figs. 12 to 14, filling in can be effected, but cutting away the upper part of the slope is also feasible. Such cutting away has been proposed as a remedy in itself in cases where the steep upper part of the slope was falling in, but it is not much of a remedy.

Stone pitching may rest, if boats are required to come close to the bank, on a toe wall of concrete, as in [fig. 13],[9] or otherwise on a foundation of loose stone, as in [fig. 12]. When concrete is used the bed is dredged to such a depth as will provide against undermining by scour. Sloping boards attached to piles are placed along the front face and the concrete is thrown in under water. The slope of stone or brick pitching is usually from 2 to 1 to 1 to 1, but it may be as steep as ½ to 1. The earth behind the pitching must be well rammed in layers. In order to prevent the earth from being eaten away by the water which penetrates through the interstices of the stone or brick, a layer, 3 to 6 inches thick, of gravel or ballast is placed over the earth and rammed. When loose stone is used, dredging is not necessary, but the stone is allowed to gradually sink down and more is added at the top. A certain proportion of the stones should be of large size.

Fig. 12.

Fig. 13.

When fascining is used, long twigs are made into bundles and tied up at every 2 feet so as to form fascines about 4 to 6 inches thick, and these are laid on the slopes and secured by pegs driven in at short intervals, between the fascines.

Fig. 14.

Fig. 15.

Sometimes the pitching or loose stone is not carried up to the top of the bank, or even up to high flood-level, and the bank above the pitching is protected by turfing—the pieces of turf being placed on edge normally to the slope if very steep ([fig. 14]) or laid parallel to the slope if it is not very steep—or, above ordinary water-level, by plantations of osiers or willows which obstruct the water and tend to cause silting, and whose roots bind the banks together.

Fig. 16.

Another method of using fascines is to lay them on the slopes with their lengths normal to the direction of the stream. The upper end of a fascine is above low water, and the lower end extends down to the bed of the stream. Sometimes large ropes made of straw, or rough mats made of grass, are laid on the slopes and pegged down, or mattresses of fascines are laid on the slopes and weighted with stones.

Fig. 17.

A deep recess in the bank ([fig. 16]) can be filled in, before the protection is added, with earth well rammed. On the Adige the filling material consisted ([fig. 17]) of faggots filled with stones, small cross dams being made at intervals, as shown by the dotted lines, to arrest flood water and cause it to deposit silt. At the back of the berm, poplar or willow slips were planted, and these grew up and their roots held the bank together. This system succeeded well.

A method of protection which is suitable when the water contains much silt is what is known in India as bushing. Large leafy branches of trees are cut and hung, as shown in [fig. 18], by ropes to pegs. They must be closely packed so as not to shake. At first they require looking after, but silt rapidly deposits and the branches become fixed and no longer dependent on the ropes. If the work is carefully done, the result is a smooth, regular, and tenacious berm, as per dotted line in the figure.

Fig. 18.

Fig. 19.

Another method, used on canals, is to make up the bank with earth and to revet it with twigs or reeds, as shown in [fig. 19]. The foundation must be taken down well below bed-level, otherwise the work may slip. This kind of work cannot be done except when the canal is dry.

If the bank consists of sand or of very sandy soil, it must in any case have a flat side slope such as 3 to 1. If the sand is in layers alternating with firm soil, it is a good plan to dig out some of the sand and to replace it with clods of hard earth.

Staking ([fig. 20]) may be used, the stakes being one or two feet apart from centre to centre, and long twigs laid horizontally being passed in and out of the stakes, or bushing filled in behind the stakes. But bushing alone is cheaper and nearly as good.

Fig. 20.

For protecting the banks of the Indus it has been proposed (Punjab Rivers and Works, Chap. IV.) to use trees in exactly the same manner as bushing, the trees being grown in several rows parallel to the river so that whenever the river, by eroding its bank, comes up to the lines of trees the first row will fall in. The first row would be chained to the second, which would take the place of the pegs used in bushing. The other rows would remain as a reserve.

The Villa system of bed protection ([Chap. V., Art. 6]) has also been successfully used for bank protection on the Scheldt, and on the Brussels-Ghent Canal, the prisms being about 10 × 10 × 4 inches, and having overlapping joints. The bands of prisms are placed in position by a boat, the bands unrolling over a drum. The boat is provided with an oscillating platform carrying rollers at its end. A thin layer of gravel is laid over the bank and is pressed down by the rollers before the prisms are laid on it (Min. Proc. Inst. C.E., vol. cxxxiv., and vol. clxxv.).

In the case of the river mentioned in [Chap. XI., Art. 3], where extremely high velocities were met with, cylindrical rolls of wire-netting were made, each 50 feet long and 5 feet in diameter, and filled with boulders. These rolls can be used for bank protection. The netting was made by wires 6 inches apart, crossing each other at right angles and tied together at the crossings by short pieces of wire.

Fig. 21.

On ship canals a berm ([fig. 21]) is frequently made a few feet below the water-level. It serves as a foundation for the pitching, which need not usually extend down to more than 5 feet below the water-level. Below that the wash has little or no effect on the banks. On ordinary navigation canals a similar berm is sometimes made—one or two feet in width and a foot or less below the water-level—and rushes are planted on it.

Sometimes a bank has been protected by a kind of artificial weed, consisting of bushes or branches of trees attached to ropes. The end of the rope is fastened to the bank and the weeds float in the stream alongside the bank.

To protect a bank from ice, which exercises an uplifting force on pitching, use has been made of a covering of a kind of reinforced concrete consisting of slabs of concrete with wires embedded in it, and fastened to the bank by wires, 20 inches long, running into the bank, these wires being embedded in mortar so as to act like stakes.

Fig. 22.

4. Heavy Stone Pitching with Apron.—On the great shifting rivers of India a system of bank protection is adopted, consisting of a pitched slope with an apron ([fig. 22]). The system is used chiefly in connection with railway bridges or weirs, but it has been used in one instance, that of Dera Ghazi Khan, for the protection of the bank near a town. When, as is usual, the flood-level is higher than the river bank, an artificial bank is made. In any case the bank is properly aligned. The pitching has a slope of 2 to 1, and consists of quarried blocks of stone loosely laid, the largest blocks weighing perhaps 120 lbs. The apron is laid at the time of low water on the sandbank or bed of the stream. If necessary, the ground is specially levelled for it. It is intended to slip when scour occurs. The following dimensions of the apron are given by Spring (Government of India Technical Paper, No. 153, “River Training and Control on the Guide Bank System,” 1904). The probable maximum depth of scour can be calculated as explained in [Chap. XI., Art. 3]. If this depth, measured from the toe of the slope pitching is D, and if T is the thickness considered necessary for the slope pitching, then the width of the apron should be 1·5 D, and its thickness 1·25 T next the slope and 2·8 T next the river. It will then be able to cover the scoured slope to a thickness of 1·25 T. This thickness is made greater than T because the stone is not likely to slip quite regularly. The thickness T should, according to Spring, be 16 inches to 52 inches, being least with a slow current and a channel of coarse sand, and greatest with a more rapid current and fine sand; but since the sand is generally finer as the current is slower, it would appear that a thickness of about 3 feet would generally be suitable. Under the rough stone there should be smaller pieces or bricks. Along the top of the bank there is generally a line of rails so that stone from reserve stacks, which are placed at intervals along the bank, can be quickly brought to the spot in case the river anywhere damages the pitched slope.

For the special protection to banks required near weirs and similar works, see [Chap. X., Arts. 2] and [3].

CHAPTER VII
DIVERSIONS AND CLOSURES OF STREAMS

1. Diversions.—When a stream is permanently diverted the new course is generally shorter than the old one, and the diversion is then often called a cut-off. The first result of a cut-off is a lowering of the water-level upstream and a tendency to scour there, and to silt downstream of the cut-off. [Fig. 23] shows the longitudinal section of a stream after a cut-off A B has been made. The bed tends to assume the position shown by the dotted line. If both the diversion and the old channel are to remain open, the water-level at the bifurcation will be lowered still more, and the tendency to scour in the diversion will be reduced.

Fig. 23.

If the material is soft enough to be scoured by the stream, it is often practicable to excavate a diversion to a small section and to let it enlarge itself by scour. This operation is immensely facilitated if the old channel can be closed at the bifurcation. The question whether the scoured material will deposit in the channel downstream of the diversion must be taken into consideration; also the question whether the diversion will continue to enlarge itself more than is desirable. The velocity in the diversion will be a maximum if its section is of the “best form,” i.e. if its bed and sides are tangents to a semicircle whose diameter coincides with the water surface, but this may not ([Chap. IV., Art. 6]) be the section which will give most scour. In order to prevent the enlargement of the diversion taking place irregularly, the excavation can be made as shown in [fig. 24], water being admitted only to the central gullet. The side gullets should not be quite continuous, but unexcavated portions should be left at intervals, so that if the water in scouring out the channel breaks into the gullet, it will not be able to flow along it until it has broken in all along.

Fig. 24.

If a diversion is made, not with the object of lowering the water-level but merely in order to shorten the channel, the increased velocity caused by the steepened slope may be inconvenient. In this case a weir or weirs can be added ([Chap. VIII., Art. 4)].

If the water contains sufficient silt to enable the abandoned loop to be silted up within a reasonable time, it may be desirable to do this. The silting up may, for instance, increase the value of the land. The loop should be closed at its upper end. Water entering the lower end will cause a deposit there. When the lower end is well obstructed by silt, the upper end should be opened.

The set of the stream, due, for instance, to a bend at the point where a diversion takes off has very little to do with the quantity of water which goes down the diversion. The only effect of the set of the stream is a slight rise of the water-level as compared with the opposite bank. Similarly, the angle at which the diversion takes off is only of importance in giving, in some cases, a velocity of approach whose effect is generally small. The distribution of the water between the diversion and the old channel really depends on their relative discharging capacities. If the required quantity of water does not flow down a diversion it can be dredged.

Sometimes a long spur is run out to send the water towards the off-take of a diversion. The effect of this is very small—it merely causes a set of the stream,—unless its length is so great that it amounts to something like a closure dam. It is sometimes said that it is easier to lead a river than to drive it. This remark is probably based on the fact that spurs, such as those under consideration, generally produce little effect, whereas the excavation of a diversion or the deepening of a branch by dredging it, is more likely to produce some result. There is, however, no certainty about this. Sometimes too much is expected of such channels. Calculations are not always made as to the scouring power of the stream, nor is account always taken of the fact that as the cut scours its gradient flattens.

2. Closure of a Flowing Stream.—The closure of a flowing stream by means of a dam is usually attended with some difficulty and sometimes with enormous difficulty. There may be little trouble in running out dams from both banks for a certain distance, but as soon as the gap between the dams becomes much less than the original width of the stream, the water on the upstream side is headed up and there is a rush of water through the gap, which tends to deeply scour the bed and to undermine the dams. The smaller the gap becomes the greater is the rush and scour.

The closure is most easily effected at or near to the place where the stream bifurcates from another. Then, as the gap decreases in width, some of the water is driven down the other stream and it does not rise so much. Eventually all the water goes down the other stream, and the total rise is only so much as will enable this other stream to carry the increased discharge. If the closure is not effected near a bifurcation, the rise of the water will go on even after the closure is completed, and it will not cease, unless the water escapes or breaks out somewhere, until it has risen to the same level as that to which it would have risen if the closure had been at the bifurcation, or perhaps not quite to the same level, since there may still be a slight slope in the water surface and a small discharge which percolates through the dam. Sometimes in such a case it is possible to arrange for temporary escapes or bifurcations, which will be shallow and therefore easily closed, after the main closure has been completed.

A closure is, of course, far more easily effected where the bed is hard than where it is soft. Very often it is best to close temporarily at such a place or near a bifurcation, even if the permanent dam has to be elsewhere, and then to construct the permanent dam in the dry channel, or in the still water, and remove the temporary one or cut a gap in it.

Generally the best method to adopt in a closure is to cover the bed of the channel beforehand—unless it is already hard enough—with a mattress or floor, such that it cannot be scoured as the gap closes. A floor may consist of a number of stones or sandbags dropped in from boats or by any suitable means, and placed with care so that there shall not be gaps or mounds. Sandbags should be carefully sewn up. A mattress may be made of fascines laid side by side and tied together, floated into position, weighted and sunk. Even a carpet made of matting or cloth and suitably weighted has sufficed in some cases. If the scour is likely to be such that stones or sandbags will be carried away, the stones may be placed in nets, baskets, or crates. Sandbags may also be placed in nets. Probably the long rolls of wire-netting filled with stones, described in [Chap. VI., Art. 3], would be better than anything, and the diameter could be reduced somewhat. The floor or mattress need not usually extend right across the stream. It must cover a width much greater than—perhaps twice as great as—the width of the gap is likely to be when scour begins. Its length, measured parallel to the direction of the stream, must be such that severe eddies in the contracted stream will have ceased before the stream reaches its downstream edge. It need not extend to any considerable distance upstream of the line of the dam.

The dams when started from the banks can generally be of simple earth or gravel, or loose stones, but before they have advanced far they will probably require protection at the ends by stones, or by staking and brushwood, or by fascines. As soon as the dams have advanced well onto the mattress and their ends have been well protected, it is best to cease contracting the stream from the sides and to contract it from the bottom by laying a number of sandbags across the gap so as to form a submerged weir. In this way the rush of water is spread over a considerable width of the stream. The weir is then raised until it comes up above water. Leakage can be stopped by throwing in earth, or gravel, or bundles of grass on the upstream side. Sometimes it is best to construct the mattress over the whole width of the stream, and to effect the closure entirely by a weir, carrying each layer right across before adding another. The banks of the stream, if not hard, can be protected by sandbags, stones, staking or fascining.

The chief cause of failures of attempts to close flowing streams is neglect to provide a proper floor or mattress. The stones or other materials may be of insufficient weight or not closely laid, or the extent of the floor may be insufficient. In a soft channel and deep water loose stones in almost any quantity may fail unless a mattress of fascines is laid under them. Another cause of failure is running short of materials, such as sandbags. Allowance should be made for every contingency, including making good any failure of parts of the work. Enormous sums of money have been wasted, and vast inconvenience, loss and trouble incurred, in futile attempts to close breaches in banks, or gaps in dams.

Sometimes the gap is closed by sinking a barge loaded with stones, or by sinking a “cradle” or large mattress made of fascines, taken out to the site by four boats, one supporting each corner, and then loaded with stones and sunk. Another method is to run out a floating mattress of fascines from one side of the gap to the middle and sink it, then to proceed similarly on the other side, and so on.

An excellent plan, when it can be adopted, is to have more than one line of operations, so that the heading up of the water is divided between them.

In India closures of streams having depths of 6 or 8 feet are effected by means of rough trestles made from trunks of small trees and placed at intervals in the stream like bridge piers, one leg of the trestle inclined upstream and one downstream. Each pair of adjacent trestles is connected by a number of rough, horizontal poles. Against these are placed bundles of brushwood. Earth is at the same time collected and is rapidly added at the last. The chief danger is the undermining of the bed by scour. This is prevented by driving in stakes and placing brushwood against them. Closures of small channels or of breaches in the banks of canals are effected by means of staking and brushwood. Where dangerous breaches are liable to occur, it is a good plan to have a barge, fitted up with a small pile-driver and carrying a supply of sheet piles, ready at a convenient spot.

Hurdle dykes, first used on the Mississippi, were employed on the Indus in 1902 to close partially the main channel of the river. There were to be three dykes, each dyke consisting of three lines of very long piles—some were 60 feet long,—driven into the bed of the stream, which was to be protected with mattresses made of fascines and extending right across it, with their heads above flood-level. The idea was not to wholly stop the flow of the water, but to obstruct it so much that silt would deposit, the channel become choked up, and the water find a course down another channel. The work was begun in March 1902, and was in progress in May of the same year when an unusually early flood put a stop to it. The dykes had at this time advanced considerable distances from the right bank of the stream, but none had been completed. Two dykes out of the three were for the most part carried away. The river, however, took a new course, starting from a point far upstream, the western channel became a creek, and the remains of the dykes were soon embedded in silt.

In any case in which the provision of a proper mattress has been omitted, or when the mattress has been destroyed, or when a breach has occurred in an embankment, whenever, in short, it is evident that the gap cannot be closed until some other escape for the water is provided, it may be possible to provide such an escape by cutting partly through the dam or embankment on the downstream side at another place, and thoroughly protecting the place and extending the protection downstream and away from the dam or embankment. The water can then be let in, and the closure of the old gap attempted. If a closure is effected, the protected gap can then be closed. Sometimes it may be desirable to make such a protected gap beforehand and with deliberation.

Dams for closing streams which are dry can be made similarly to flood embankments ([Chap. XII., Art. 7]). Sand does very well, provided it is protected by a covering of clay or by fascining.

3. Instances of Closures of Streams.—In 1904 the Colorado River broke into the Salton Sink—a valley covering 4000 square miles. Unsuccessful attempts were made to close the stream by two rows of piles with willows and sandbags between them, by a gate 200 feet long, supported on 500 piles, and by twelve gates each 12 feet wide. A “rock-fill” dam was then constructed on a mattress 100 feet wide and 1·5 feet thick. The river, which was 600 feet wide, broke through, but was stopped by the construction of three parallel rock-fill dams in the gap (Min. Proc. Inst. C.E., vol. clxxi.).

Fig. 25.

At the site of the railway bridge over the river Tista in Bengal, it was necessary to close the main stream ([fig. 25]), which flowed at the left side of the channel, while the bridge had been built at the right. The bed was of sand, width 500 feet, depth 6 feet, and discharge 3700 cubic feet per second. The first attempt to close the stream was made at M N, a floor of stone 200 feet long, 20 feet wide, and 2 feet thick, being laid in the middle of the stream, and dams of earth, sandbags, and stones being run out from each bank. As the gap decreased in width the bed was torn up and the work failed. The heading up was 3 feet 9 inches. It was recognised later that the site should have been at the bifurcation higher up, and that the stone floor should have been laid on a mattress.

In the next working season the dams C D and E F G were made. The dam C D was of earth. Two walls, each consisting of a double line of bamboos with the spaces between the lines filled with bundles of grass weighted with earth, were run out 50 feet in advance of the earthwork near the lines of the toes of the slopes. Along the line of the upper wall a mattress of broken bricks 10 feet in width, and 1 foot thick, was laid, and was kept 50 feet in advance of the wall. A total length of 1000 feet of embankment was made in five months and pitched on its upstream side. The end was strongly protected by a mass of stone. The embankment F G was of earth. The dam E F consisted of three lines of piles driven 10 feet into the bed. A mattress weighted with stones extended for 20 feet upstream of the dam and 40 feet downstream. A gap of 150 feet was left at D E, and was not protected by a floor of any kind. A channel, parallel to F G and extending to K, had been dug to a width of 200 feet. During the floods the heading up at D E was about 2·5 feet, and the water was 30 feet deep. The line E F was greatly damaged and was repaired. The cut F G K gradually enlarged, and by the end of the floods more water was going down it than down the main stream. The gap D E was finally closed by means of a line of bamboos and grass, the bed being protected by a carpet, 100 × 50 feet, made of common cloth weighted with sandbags. The success of the operations turned on the scouring out of the cut F G K. It is remarkable that the gap D E did not become wholly unmanageable in the floods (Min. Proc. Inst. C.E., vol. cl.).

CHAPTER VIII
THE TRAINING AND CANALISATION OF RIVERS

1. Preliminary Remarks.—When a stream is trained or regularised it is generally made narrower, but sometimes narrow places have to be widened. Deepening has also very frequently to be effected. The object of training is generally the improvement of navigation, but it may be the prevention of silt deposit. Some natural arms of rivers which form the head reaches of canals in the Punjab are wide and tortuous, and they are sometimes trained. Training often includes straightening or the cutting-off of bends, as to which reference may be made to [Chap. VII.]

2. Dredging and Excavating.—When a flowing stream is to be deepened, the work is usually done by dredgers. Dredgers can remove mud, sand, clay, boulders, or broken pieces of rock. The “bucket ladder” dredger is the commonest type. The “dipper” dredger is another. Both these can work in depths of water ranging up to 35 feet. The “grab bucket” dredger can work up to any depth and in a confined space. The “suction dredger” drawls up mud or sand mixed with water. A dredger may be fitted with a hopper or movable bottom, by means of which it can discharge the dredged material—this, however, involves cessation of work while the dredger makes a journey to the place where the material is to be deposited—or it can discharge into hopper barges or directly on to the shore by means of long shoots. For small works in comparatively shallow water the “bag and spoon” dredger, worked by two men, can be used.

When rock has to be removed under water it is blasted or broken up by the blows of heavy rams provided with steel-pointed cutters.

In widening a channel the excavation can be carried down in the ordinary way to below the water-level, a narrow piece of earth, like a wall, being left to keep the water out. If the channel cannot be laid dry, the work can be finished by dredging.

Regarding methods by which the stream is itself made to deepen or widen its channel, reference may be made to [Chap. V.]

3. Reduction of Width.—If a channel which is to be narrowed is not a wide one, the reduction in width can be effected by any of the processes described under bank protection ([Chap. VI.]). But in a wide channel, reduction of the width by any direct process is generally impracticable. The expense would generally be prohibitive. Earth, if filled in, is liable to be washed away unless protected all along. Reduction in the width of a large channel is nearly always effected either by groynes ([fig. 26]) or by training walls ([fig. 27]). Spurs or short groynes for bank protection have been already described ([Chap. VI., Art. 2]). Groynes for narrowing streams are made in the same way and of the same materials, but are longer. They are at right angles to the stream or nearly so. Groynes in the river Sutlej have been mentioned in [Chap. V., Art. 5], and are shown in [fig. 6], p. 55. Whether groynes or training walls are used, the object is to confine the stream to a definite zone and to silt up the spaces at the sides. These spaces when partly silted can be planted with osiers or with anything which will grow when partly submerged, and this will assist in completing the silting.

Fig. 26.

Fig. 27.

A training wall can be made of any of the materials used for groynes. In order to silt up the spaces between each wall and the adjacent bank of the stream, other walls are run at intervals across them. Usually the training walls and cross walls are carried up only to ordinary water-level, sometimes only to low-water level. Floods can thus spread out and submerge the walls and deposit silt. If the walls are carried up too high it may be necessary, in order to give room for floods, to space them too far apart, and this, as will be seen below, is objectionable.

The difference between training walls and groynes is one of degree rather than one of kind. The material most commonly used is, in either case, loose stone—with pitching, if desired, above low-water level,—but it may be wattled stakes. If the water of the stream contains silt at all stages of the supply, gaps can be left in training walls so that silt deposit may occur at all times and not only in floods. If the walls are of wattled stakes, water will pass through them, and it may not be necessary to leave any gaps. Groynes are frequently made with T-heads ([fig. 26]), and they are thus equivalent to training walls with long gaps in them. The edge of the narrowed channel usually forms somewhat as shown in the figure. If the groynes are placed so near together as to give a regular channel, the cost is not likely to be much less than that of training walls.

The alignment of training walls or groynes should be such as will give the best channel consistent with economy in cost. The best channel is generally that which is most free from sharp bends. It is assumed for the present that no cuts or diversions of such lengths as to materially alter the gradient are to be made, but that a certain amount of choice of alignment is afforded by the reduced width of the trained channel and by small diversions or easings of bends. It is sometimes said that straight reaches are objectionable because the stream will tend to wander from side to side and cause shoals, whereas in a bend there will be no such tendency. The difficulty as to shoaling will be greatest at low water, but it is likely to be serious only when the width between the training walls is too great. If the width cannot be reduced to such an extent as to do away with the trouble, it may be better to adopt a curved course. The width between the training walls should generally be the same throughout, whether the reaches are straight or curved, but in view of the preceding remarks it may be desirable, where a reach cannot be otherwise than straight and where shoaling is feared, to give the straight portion a reduced width with of course a greater depth, and similarly to reduce the width at reverse changes of curvature. In curves which are at all sharp the curvature should be rather sharper in the middle of the curve than at the ends ([Chap. IV., Art. 8]).

4. Alteration of Depth or Water-Level.—When the width of a stream is altered, the depth of water—the gradient being supposed to be unchanged—must alter in the opposite manner. A narrowing of the channel by training necessitates an increase in the depth of water, and the same remark applies if an arm of the stream is closed. The increase in depth may be effected either by raising the water-level or by lowering the bed—as may be convenient—or both. If the bed is to be lowered and is of hard clay, it may be necessary to dredge it and, when this has been done, training may be unnecessary. If the bed is of soft mud, a dredged channel is likely to fill up again, and training alone will be the method to adopt. If the bed is moderately hard, say compact sand, it may be suitable to train the channel first and then to dredge if necessary. In any case, shoals of hard material may have to be dredged or rocks, whether these form shoals or lateral obstructions, to be blasted or otherwise broken up (Art. 2). In cases where it is desired to raise the water-level without any lowering of the bed, training is of course necessary. In any case in which the bed is likely to scour to a lower level than is desired, or if the bed is to be raised, the measures described in [Chap. V., Art. 6], may be adopted, but they are hardly likely to be suitable and satisfactory in all cases.

5. Training and Canalising.—The steps so far described, together with any of those described in [Chaps. V.] and [VI.], exhaust the list of what can be done so long as only the cross-section of a stream is dealt with. This is often called the “regulation” of a stream, though “training” is a more satisfactory term.[10]

Fig. 28.

A mere alteration of the cross-section of a stream will not always afford a solution of the problem to be solved. Frequently a change of gradient is required. The gradient can be steepened by means of straightenings, or flattened by introducing weirs, or perhaps by adopting a course somewhat more circuitous than was intended. This extended scope of operations is known as canalising in the case of a river, and remodelling in the case of a canal.

Suppose that it is desired to alter the cross-section of a stream, at ordinary water-level, so as to reduce the width and increase the depth ([fig. 28]). If the mean depth is doubled, the new width will be about equal to 1/(3·2) of the old width (Hydraulics, Chap. VI., Art. 2). If this gives too narrow a channel, it may be desirable to flatten the gradient. If it gives too wide a channel, the gradient can be steepened or a greater depth adopted. While the width and depth of the stream will be fixed so as to be suitable for the navigation, the ratio of depth to velocity should be so arranged, if this is possible, as to minimise trouble connected with silting or scour ([Chap. IV., Art. 6]). A remodelled channel is, in short, designed in exactly the same way as a new channel. The depth of water exercises the greatest effect on the discharge, and the gradient the least. The weak point in a scheme which includes weirs is the difficulty of dealing with floods. A scheme perfect in all other respects may be vitiated because of the obstruction, caused by weirs, to the passage of floods. The difficulty is got over by means of movable weirs. The whole subject of weirs is dealt with in [Chap. X.]

Training or canalising should not be effected in any reach of a stream without regard to other reaches. A mere local lowering of the water-level by dredging may accentuate the effect of a shoal at the upper end of the reach.

When the water-level is raised by a weir or by narrowing the channel—though in the latter case the raising may not be permanent—it is generally best to commence the work from the upstream end. The raising of the water-level will then not interfere with the execution of the rest of the work. But in a case of widening, where the water-level upstream of the work is lowered, the work can conveniently be begun at the downstream end, and the remark applies also to a case of straightening, provided that the new channel is not so small that it at first causes no lowering. In any case in which there is a doubt whether the whole of the scheme will be carried out, the reach to be dealt with first can be decided on according to circumstances. There is no general reason for selecting an upstream or downstream reach, except that any raising or lowering of the water-level will extend upstream of the reach and not downstream of it ([Chap. I., Art. 4]).

Training walls and groynes, if made with stakes or fascines or any materials except stone, require careful watching and maintenance.

CHAPTER IX
CANALS AND CONDUITS

1. Banks.—All banks which have to hold up water should be carefully made. The earth should be deposited in layers and all clods broken up. In high banks the layers should be moistened and rammed. The dotted lines in [fig. 29] show two possible courses of percolation water. The vertical height—from the water-level to the ground outside the bank,—divided by the length of the line of percolation is the hydraulic gradient, as in the case of a pipe, and this gradient is more or less a measure of the tendency to leakage. A bank which has water constantly against it nearly always becomes almost water-tight in time. The time is less or greater according as the soil is better, and according to the amount of care with which the bank is made.

Fig. 29.

The side slopes of banks vary with the soil. Generally they are 1½ to 1, but they are sometimes 2 to 1 or even 3 to 1 if the soil is bad or sandy, or if great precautions against breaches are to be taken.

Leakage can sometimes be stopped by throwing chaff or other finely divided substances into the water at the site of the leak. In other cases it is necessary to dig up part of the bank, find the channel by which the water is escaping, and fill it up by adding earth and ramming. On some navigable canals in France it was at one time the custom to lay the reach dry, when a bad leak occurred, and to dig away the bank and lay slabs of concrete or puddle over the place. This plan was abandoned, and instead of it sheet piles are driven in. They are then withdrawn one at a time and, if any leakage occurs, the space is filled with concrete.

The dimensions of a bank should depend to some extent on the head of water against it and on the volume of the stream whose water it holds up. A breach is obviously more serious the greater the volume of the escaping water. This volume depends on the size of the stream and on its velocity. In navigation canals in England the bank on the side opposite the towing-path is usually 4 to 6 feet wide and 1½ feet above the water. In irrigation canals in India the bank of a very large canal is 2 feet above the water and 20 feet wide, while that of a small canal with 6 feet of water is 8 or 10 feet wide and 1½ feet above the water, and that of a small distributary channel with 3 feet of water is 4 feet wide and 1 foot above the water. The soil is often poor.