LEARNING PACKAGE FOR HYDROLOGY
Hydrology means the science of water. It is the science that deals with the occurrence, circulation and distribution of water of the earth and earth's atmosphere. As a branch of earth science, it is concerned with the water in streams and lakes, rainfall and snowfall, >snow and ice on the land and water occurring below the earth's surface in the pores of the soil and rocks. In a general sense, hydrology is a very broad subject of an inter-disciplinary nature drawing support from allied sciences, such as meteorology, geology, statistics, chemistry, physics and fluid mechanics.
Hydrology is basically an applied science. To further emphasis the degree of applicability, the subject is sometimes classified as
- Scientific hydrology - the study which is concerned chiefly with academic aspects.
- Engineering or applied hydrology- a study concerned with engineering applications.
In a general sense engineering hydrology deals with
- Estimation of water resources
- The study of processes such as precipitation,runoff,evapotranspiration and theirinteraction and
- The study of problems such as flood and droughts and strategies to combat them.
Water occurs on the earth in all its three states, viz. liquid, solid and gaseous, and in various degrees of motion.Evaporation of water from water bodies such as oceans and lakes, formation and movement of clouds, rain and snowfall,streamflow and groundwater movement are some examples of the dynamic aspects of water. The various aspects of water related to the earth can be explained in terms of a cycle known as the hydrologic cycle.
A convenient starting point to describe the cycle is in the oceans. Water in the oceans evaporates due to the heat energy provided by solar radiation. The water vapour moves upward and form clouds. While much of the clouds condense and fall back to the oceans as rain, a part of the clouds is driven to the land areas by winds. There they condense and precipitate onto the landmass as rain, snow,hail, sleet, etc. A part of the precipitation may evaporate back to the atmosphere even while falling. Another part may be intercepted by vegetation, structures and other such surface modifications from which it may be either evaporated back to atmosphere or move down to the ground surface.
A portion of the water that reaches the ground enters the earth's surface through infiltration , enhance the moisture content of the soil reach the groundwater body. Vegetation sends a portion of the water from under the ground surface back to the atmosphere through the process of transpiration. The precipitation reaching the ground surface after meeting the needs of infiltration and evaporation moves down the natural slope over the surface and through a network of gullies, streams and rivers to reach the ocean. The groundwater may come to the surface through springs and other outlets after spending a considerably longer time than the surface flow. The portion of the precipitation which by a variety of paths above and below the surface of the earth reaches the stream channel is called runoff. Once it enters a stream channel,runoff becomes stream flow.
The sequence of events as above is a simplistic picture of a very complex cycle that has been taking place since the formation of the earth. It is seen that the hydrologic cycle is a very vast and complicated cycle in which there are a large number of paths of varying time scales. Further, it is a continuous re-circulating cycle in the sense that there is neither a beginning nor an end or a pause. Each path of the hydrologic cycle involves one or more of the following, aspects:
- Transportation of water,
- Temporary storage and
- Change of state.
(a) the process of rainfall has the change of state and transportation and
(b) the groundwater path has storage and transportation aspects.
The quantities of water going through various individual paths of the hydrological cycle can be described by the continuity equation known as water-budget equation or hydrologic equation
For a given problem area, say a catchment, in an interval of time At,
Mass inflow-mass outflow = change in mass storage
if the density of the inflow, outflow and storage volumes are same.
Vi - Vo = D S 9; 9; (1.1)
Where Vi inflow volume of water into the problem area during the time period, Vo outflow volume of water from the problem area during the time period, and D S = change in the storage of the water volume over and under the given area during the given period. In applying this continuity equation to the paths of the hydrologic cycle involving change of state, the volumes considered are the equivalent volumes of water at a reference temperature.
It is important to note that the total water resources of the earth are constant and the sun is the source of energy for the hydrologic cycle. A recognition of the various processes such as evaporation, precipitation and groundwater flow helps one to study the science, of hydrology in a systematic way. Also, one realises that man can interfere with virtually any part of the hydrologic cycle, e.g. through artificial rain, evaporation suppression, change of vegetal cover and land use, extraction of ground. water, etc. Interference at one stage can cause serious repercussions at some other stage of the cycle.
The hydrological cycle has important influences in a variety of fields including agriculture, forestry, geography, economics, sociology and political science. Engineering applications of the knowledge of the hydro-logic cycle, and hence of the subjects of hydrology, are found in the design and operation of projects dealing with water supply, irrigation and drainage,water power,flood control, navigation, coastal works, salinity control and recreational uses of water.
APPLICATIONS IN ENGINEERING
Hydrology finds its greatest application in the design and operation of introduction water-resources engineering projects, such as those for irrigation, water supply, flood control, water power and navigation. In all these projects hydrological investigations for the proper assessment of the following factors are necessary.
The hydrological study of a project should of necessity precede structural and other detailed design studies. It involves the collection of relevant data and analysis of the data by applying the principles and theories of hydrology to seek solutions to practical problems.
Many important projects in the past have failed due to improper assessment of the hydrological factors. Some typical failures of hydraulic structures are:
- Overtopping and consequent failure of an earthen dam due to an inadequate spillway capacity,
- Failure of bridges and culverts due to excess flood flow and
- Inability of a large reservoir to fill up with water due to overestimation of the stream flow. Such failure, often-called hydrologic failure underscore the uncertainty aspect inherent in hydrological studies.
Various phases of the hydrological cycle, such as rainfall, runoff, evaporation and transpiration are all non-uniformly distributed both in time and space. Further, practically all hydrologic phenomena are complex and at the present level of knowledge, they can at best be interpreted with the aid of probability concepts. Hydrological events are treated as random processes and the historical data relating to the event are analysed by statistical methods to obtain information on probabilities of occurrence of various events. The probability analysis of hydrologic data is an important component of present-day hydrological studies and enables the engineer to take suitable design decisions consistent with economic and other criteria to be taken in a given project.
The term "precipitation" denotes all forms of water that reach the earth from the atmosphere. The usual forms are rainfall, snowfall,hail,frost and dew. Of all these, only the first two contribute significant amounts of water. Rainfall being the predominant form of precipitation causing stream flow, especially the flood flow in a majority of rivers in India, unless otherwise stated the term "rainfall" is used in this book synonymously with precipitation. The magnitude of precipitation varies with time and space. Differences in the magnitude of rainfall in various parts of a country at a given time and variations of rainfall at a place in various seasons of the year are obvious and need no elaboration. It is this variation that is responsible for many hydrological problems, such asfloods and droughts.
The study of precipitation forms a major portion of the subject of hydrometeorology. In this chapter, a brief introduction is given to familiarize the engineer with important aspects of rainfall and, in particular, with the collection and analysis of rainfall data. For precipitation to form:
- The atmosphere must have moisture,
- There must be sufficient nuclei present to aid condensation,
- Weather conditions must be good for condensation of water vapour to take place,
- The products of condensation must reach the earth.
Under proper weather conditions, the water vapour condenses over nuclei to form tiny water droplets of sizes less than 0.1 mm in diameter. The nuclei are usually salt particles or products of combustion and are normally available in plenty. Wind speed facilitates the movement of clouds while its turbulence retains the water droplets in suspension. Water droplets in a cloud are somewhat similar to the particles in a colloidal suspension. Precipitation results when water droplets come together and coalesce to form larger drops that can drop down. A considerable part of this precipitation gets evaporated back to the atmosphere. The net Precipitation at a place and its form depend upon a number of meteorological factors, such as the weather elements like wind, temperature, humidity and pressure in the volume region enclosing the clouds and the ground surface at the given place.
FORMS OF PRECIPITATION
Some of the common forms of precipitation are rain,snow, drizzle,glaze, sleet and hail.
It is the principal form of precipitation in India. The term "rainfall" is used to describe precipitation in the form of water drops of sizes larger than 0.5 mm. The maximum size of a raindrop is about 6 mm. Any drop larger in size than this trends to break up into drops of smaller sizes during its fall from the clouds. On the basis of its intensity, rainfall is classified as:
Trace to 2.5 mm/h
2.5 mm/h to 7.5 mm/h
> 7.5 mm/h
Snow is another important form of precipitation. Snow consists of ice crystals which usually combine to form flakes. When new, snow has an initial density varying from 0.06 to 0.15 g/cm3 and it is usual to assume an average density of 0. 1 g/cm3. In India, snow occurs only in the Himalayan regions.
A fine sprinkle of numerous water droplets of size less than 0.5 mm and intensity less than 1 mm/h is known as drizzle. In this the drops are so small that they appear to float in the air.
When rain or drizzle come in contact with cold ground at around OoC, the water drops freeze to form an ice coating called glaze or freezing rain.
It is frozen raindrops of transparent grains which form when rain falls through air at subfreezing temperature. In Britain, sleet denotes precipitation of snow and rain simultaneously.
It is a showery precipitation in the form of irregular pellets or lumps of ice of size more than 8 mm. Hails occur in violent thunderstorms in which vertical currents are very strong.
WEATHER SYSTEMS FOR PRECIPITATION
For the formation of clouds and subsequent precipitation, it is necessary that the moist air masses cool to form condensation. This is normally accomplished by adiabatic cooling of moist air through a process of being lifted to higher altitudes. Some of the terms and processes connected with the weather systems associated with precipitation are given below.
A front is the interface between two distinct air masses. Under certain favourable conditions when a warm air mass and cold air mass meet, the warmer air mass is lifted over the colder one with the formation of a front. The ascending warmer air cools adiabatically with the consequent formation of clouds and precipitation.
A cyclone is a large low-pressure region with circular wind motion. Two types of cyclones are recognized:tropical cyclones and extratropical cyclones.
A tropical cyclone, also called cyclone in India,hurricane in USA and typhoon in South-East Asia, is a wind system with an intensely strong depression with MSL pressures sometimes below 915 mbars. The normal areal extent of a cyclone is about 100-200 km in diameter. The isobars are closely spaced and the winds are anti-clockwise in the northern hemisphere. The centre of the storm,called the eye, which may extend to about 10- 50 km in diameter, will be relatively quiet. However, right outside the eye, very strong winds reaching to as much as 200 kmph exist. The wind speed gradually decreases towards the outer edge. The pressure also increases outwards (Fig. 2.1). The rainfall will normally be heavy in the entire area occupied by the cyclone.
During summer months, tropical cyclones originate in the open ocean at around 5-10° Latitude and move at speeds of about 10-30 kmph to higher latitudes in an irregular path.
They derive their energy from the latent heat of condensation of ocean water vapour and increase in size as they move on oceans. When they move on land the source of energy is cut off and the cyclone dissipates its energy very fast. Hence, the intensity of the storm decreases rapidly. Tropical cyclones cause heavy damage to life and property on their land path and intense rainfall and heavy floods in streams are its usual consequences. Tropical cyclones give moderate to excessive precipitation over very large areas, of the order of 10³ km² for several days.
These are cyclones formed in locations outside the tropical zone. Associated with a frontal system, they possess a strong counter-clockwise wind circulation in the northern hemisphere. The magnitude of precipitation and wind velocities are relatively lower than those of a tropical cyclone. However, the duration of precipitation is usually longer and the areal extent also is longer.
These are regions of high Pressure, usually of large areal extent. The weather is usually calm at the centre.Anticyclones cause clockwise wind circulations in the northern hemisphere. Winds are of moderate speed, and at the outer edges, cloudy and precipitation conditions exist.
In this type of precipitation a packet of air which is warmer than the surrounding air due to localised heating rises because of its lesser density. Air from cooler surroundings flows to take up its place thus setting up a convective cell. The warm air continues to rise, undergoes cooling and results in precipitation. Depending upon the moisture, thermal and other conditions light showers to thunderstorms can be expected in convective precipitation. Usually the areal extent of such rains is small, being limited to a diameter of about 10 km.
The moist air masses may get lifted-up to higher altitudes due to the presence of mountain barriers and consequently undergo cooling, condensation and precipitation. Such a precipitation is known as Orographic precipitation. Thus in mountain ranges, the windward slopes have heavy precipitation and the leeward slopes light rain fall.
CHARACTERISTIC OF PRECIPITATION ON INDIA
From the point of view of climate the Indian subcontinent can be considered to have two major seasons and two transitional periods as:
- South-west monsoon (June-September)
- Transition-1, post-monsoon (October-November)
- Winter season (December-February)
- Transition-11, Summer, (March-May)
South-West Monsoon (June-September)
The south-west monsoon (popularly known as the monsoon) is the principal rainy season of India when over 75% of the annual rainfall is received over a major portion of the country. Excepting the south-eastern part of the peninsula and Jammu and Kashmir, for the rest of the country the south-west monsoon is the principal source of rain with July as the rainiest month. The monsoon originates in the Indian ocean and heralds its appearance in the southern part of Kerala by the end of May. The onset of monsoon is accompanied by high south-westerly winds at speeds of 20-40 knots and low-pressure regions at the advancing edge. The monsoon winds advance across the country in two branches; Arabian sea branch and Bay of Bengal branch.
The former sets in at the extreme southern part of Kerala and the latter at Assam, almost simultaneously in the first week of June. The Bay branch first covers the north-eastern regions of the country and turns westwards to advance into Bihar and UP. The Arabian Sea branch moves northwards over Karnataka, Maharashtra and Gujarat. Both the branches reach Delhi around the same time by about the fourth week of June. A low-pressure region known as monsoon trough is formed between the two branches. The trough extends from the Bay of Bengal to Rajasthan and the precipitation pattern over the country is generally determined by its position. The monsoon winds increase from June to July and begin to weaken in September. The withdrawal of the monsoon, marked by a substantial rainfall activity starts in September in the northern part of the country. The onset and withdrawal of the monsoon at various parts of the country are shown in Fig. 2.2.
The monsoon is not a period of continuous rainfall. The weather is generally cloudy with frequent spells of rainfall. Heavy rainfall activity in various parts of the country owing to the passage of low pressure region is common. Depressions formed in the Bay of Bengal at a frequency of 2-3 per month move along the trough causing excessive precipitation of a 100-200 mm per day. Breaks of about a week in which the rainfall activity is the least is another feature of the monsoon. The south-west monsoon rainfall over the country is indicated in Fig. 2.3.
As seen from this figure the heavy rainfall areas are Assam and the north-eastern region with 200-400 cm; west coast and western ghats with 200-300 cm; West Bengal with 120-160 cm, UP, Haryana and the Punjab with 100-120 cm.
As the south-west monsoon retreats, low-pressure areas form in the Bay of Bengal and a north-easterly flow of air that picks UP moisture in the Bay of Bengal is formed. This air mass strikes the East Coast of the southern peninsula (Tamilnadu) and causes rainfall. Also, in this period, especially November, severe tropical cyclones form in the Bay of Bengal and Arabian Sea. The cyclones formed in the Bay of Bengal are about twice as many as in the Arabian Sea. These cyclones strike the coastal areas cause intense rainfall and heavy damage to life and property.
Winter Season (December-February)
By about mid-December, disturbances of extra tropical origin travel; wards across Afghanistan and Pakistan. Known as western disturbances, they cause moderate to heavy rain and snowfall (about 25 cm) in Himalayas and Jammu and Kashmir. Some light rainfall also occurs in northern plains. Low-pressure areas in the Bay of Bengal formed in the months cause 10-12 cm of rainfall in the southern parts of Tamilnadu.
Summer (Pre-monsoon) (March-May)
There is very little rainfall in India in this season. Convective cells cause some thunderstorms mainly in Kerala, West Bengal and Assam. Some cyclone activity, dominantly on the cast coast, also occurs.
The annual rainfall over the country is shown in Fig. 2.4. Considerable areal variation exists for the annual rainfall in India with high rainfall the magnitude of 200 cm in Assam and north-eastern parts and the western ghats, and scanty rainfall in eastern Rajasthan and parts of Gujarat, Maharashtra and Karnataka. The average annual rainfall for the entire country is estimated as 119 cm. It is well known that there is considerable variation of annual rainfall in time at a place. The coefficient of variation,
of the annual rainfall varies between 15 to 70, from place to place with an average value of about 30. Variability is least in regions of high rainfall and largest in regions of scanty rainfall. Gujarat, Haryana, the Punjab and Rajasthan have large variability of rainfall.
Precipitation is expressed in terms of the depth to which rainfall water would stand on an area if all the rain were collected on it. Thus 1 cm of rainfall over a catchment area of 1 km² represents a volume of water equal to 104 m³. In the case of snowfall, an equivalent depth of water is used as the depth of precipitation. The precipitation is collected and measured in a raingauge. Terms such as pluviometer, ombrometer and hyetometer are also sometimes used to designate a raingauge. A raingauge essentially consists of a cylindrical-vessel assembly kept in the open to collect rain. The rainfall catch of the raingauge is affected by its exposure conditions. To enable the catch of raingauge to accurately represent the rainfall in the area surrounding the raingauge standard settings are adopted. For setting a raingauge the following considerations are important:
- The ground must be level and in the open and the instrument must present a horizontal catch surface.
- The gauge must be set as near the ground as possible to reduce wind effects but it must be sufficiently high to prevent splashing, flooding etc.
- The instrument must be surrounded by an open fenced area of at least 5.5 m X 5.5 m. No object should be nearer to the instrument than 30 m or twice the height of the obstruction.
Raingauges can be broadly classified into two categories as non-recording raingauges and recording gauges.
The nonrecording gauge extensively used in India is the Symons' gauge. It essentially consists of a circular collecting area of 12.7 cm (5.0 inch) diameter connected to a funnel. The rim of the collector is set in a horizontal plane at a height of 30.5 cm above the ground level. The funnel discharges the rainfall catch into a receiving vessel. The funnel and receiving vessel are housed in a metallic container. Figure 2.5 shows the details of the installation.
Water contained in the receiving vessel is measured by a suitably graduated measuring glass, with an accuracy up to 0.1 mm. Recently, the India Meteorological Department (IMD) has changed over to the use of fibreglass reinforced polyster raingauges, which is an improvement over the Symons' gauge. These come in different combinations of collector and bottle. The collector is in two sizes having areas of 200 and 100 cm² respectively. Indian Standard (IS : 5225-1969) gives details of these new raingauges.
For uniformity, the rainfall is measured every day at 8.30 AM (IST) and is recorded as the rainfall of that day. The receiving bottle normally does not hold more than 10 cm of rain and as such in the case of heavy rainfall the measurements must be done more frequently and entered. However, the last reading must be taken at 8.30 Am and the sum of the previous readings in the past 24 h entered as total of that day. Proper care' maintenance and inspection of raingauges, especially during dry weather to keep the instrument free from dust and dirt is very necessary. The details of installation of non-recording raingauges and measurement of rain are specified in Indian Standard (IS : 4986-1968). This raingauge can also be used to measure snowfall. When snow is expected, the funnel and receiving bottle are removed and the snow is allowed to collect in the outer metal container. The snow is then melted and the depth of resulting water measured. Antifreeze agents are some times used to facilitate melting of snow. In areas where considerable snowfall is expected, special snowgauges with shields (for minimizing the wind effect) and storage pipes (to collect snow over longer durations) are used.
Recording gauges produce a continuous Plot of rainfall against time and provide valuable data of intensity and duration of rainfall for hydrological analysis of storms. The following are some of the commonly used recording raingauges.
This is a 30.5 cm size raingauge adopted for use by the US Weather Bureau. The catch from the funnel falls onto one of a pair of small buckets. These buckets are so balanced that when 0.25 mm of rainfall collects in one bucket, it tips and brings the other one in position. The water from the tipped bucket is collected in a storage can. The tipping actuates an electrically driven pen to trace a record on clockwork-driven chart. The water collected in the storage can is measured at regular intervals to provide the total rainfall and also serve as a check. It may be noted that the record from the tipping bucket gives data on the intensity of rainfall. Further, the instrument is ideally suited for digitalizing of the output signal.
In this raingauge the catch from the funnel empties into a bucket mounted on a weighing scale. The weight of the bucket and its contents are recorded on a clockwork-driven chart. The clockwork mechanism has the capacity to run for as long as one week. This instrument gives a plot of the accumulated rainfall against the elapsed time, i.e. the mass curve of rainfall. In some instruments of this type the recording unit is so constructed that the pen reverses its direction at every preset value, say 7.5 cm (3 in.) so that a continuous plot of storm is obtained.
This type of recording raingauge is also known as float-type gauge. Here the rainfall collected by a funnel-shaped collector is led into a float chamber causing a float to rise. As the float rises, a pen attached to the float through a lever system record the elevation of the float on a rotating drum driven by a clockwork mechanism. A syphon arrangement empties the float chamber when the float has reached a pre-set maximum level. This type of raingauge is adopted as the standard recording-type raingauge in India and its details are described in Indian Standard (IS : 5235-1969). A typical chart from this type of raingauge is shown in Figure 2.6.
This chart shows a rainfall of 53.8 mm in 30 h. The vertical lines in the pen trace correspond to the sudden emptying of the float chamber by syphon action which resets the pen to zero level. It is obvious that the natural syphon-type recording raingauge gives a plot of the mass curve of rainfall.
These raingauges are of the recording type and contain electronic transmit the data on rainfall to a base station both at regular inter on interrogation. The tipping-bucket type raingauge, being ideally suited is usually adopted for this purpose. Any of the other types of recording raingauges can also be used equally effectively. Telemetering gauges are utmost use in gathering rainfall data from mountainous and genera inaccessible places.
Radar Measurement of Rainfall
The meteorological radar is a powerful instrument for measuring the are extent, location and movement of rainstorms. Further, the amount rainfall over large areas can be determined through the radar with a go degree of accuracy. The radar emits a regular succession of pulses of electromagnetic radiation in a narrow beam. When raindrops intercept a radar beam, it has be shown that
where Pr = average echo power, Z = radar-echo factor, r = distance target volume and C = a constant. Generally the factor Z is related to the intensity of rainfall as
Where, a and b are coefficients and I = intensity or rainfall in mm/h. The values a and b for a given radar station have to be determined by calibration with the help of recording raingauges. A typical equation for Z is
Z = 200 I 1.60
Meteorological radars operate with wavelengths ranging from 3 to 10 cm, the common values being 5 and 10 cm. For observing details of heavy flood-producing rains, 10 cm radar is used while for light rain and snow a 5-em radar is used. The hydrological range of the radar is about 200 km. Thus a radar can be considered to be a remote-sensing super gauge covering an areal extent of as much as 100,000 km². Radar measurement is continuous in time and space. Present-day developments in the field include (i) On-line processing of radar data on a computer and (ii) Doppler-type radars for measuring the velocity and distribution of raindrops.
Since the catching area of a raingauge is very small compared to the areal extent of a storm, it is obvious that to get a representative picture of a storm over a catchment the number of raingauges should be as large as possible, i.e. the catchment area per gauge should be small. On the other hand, economic considerations to a large extent and other considerations, such as topography, accessibility, etc. to some extent restrict the number of gauges to be maintained. Hence one aims at an optimum density of gauges from which reasonably accurate information about the storms can be obtained. Towards this the World Meteorological Organisation (WMO) recommends the following densities.
- In flat regions of temperate, Mediterranean and tropical zones:
ideal-1 station for 600-900 km², acceptable-1 station for 900-3000 km²;
- In mountainous regions of temperate, Mediterranean and tropical zones: ideal-1 station for 100-250 km² acceptable-1 station for 250-1000 km²; and
- In arid and polar zones: 1 station for 1500-10,000 km² depending on the feasibility.
Ten per cent of raingauge stations should be equipped with self-recording gauges to know the intensities of rainfall.
PREPARATION OF DATA
Before using the rainfall records of a station, it is necessary to first check the data for continuity and consistency. The continuity of a record may be broken with missing data due to many reasons such as damage or fault in a raingauge during a period. The missing data can be estimated by using the data of the neighboring stations. In these calculations the nor-mal rainfall is used as a standard of comparison. The normal rainfall is the average value of rainfall at a particular date, month or year over a specified 30-year period. The 30-year normals are recomputed every decade. Thus the term "normal annual precipitation" at station A means the average annual precipitation at A based on a specified 30 years of record.
Estimation of Missing Data
Given the annual precipitation values, P1, P2, P3, ... Pm at neighbouring M stations 1, 2, 3,..., M respectively, it is required to find the missing annual precipitation Px at a station X not included in the above M stations. Further, the normal annual precipitations NI, N2, ..., Ni... at each of the above (M + 1) stations including station X are known.
If the normal annual precipitations at various stations are within about 10% of the normal annual precipitation at station X, then a simple arithmetic average procedure is followed to estimate Px. Thus
If the normal precipitation vary considerably, then Px is estimated by weighing the precipitation at the various stations by the ratios of normal annual precipitation. This method, known as the normal ratio method gives Px as
Mass Curve of Rainfall
The mass curve of rainfall is a plot of the accumulated precipitation against time, plotted in chronological order. Records of float type and weighing-bucket type gauges are of this form. A typical mass curve of rainfall at a station during a storm is shown in Fig. 2.8. Mass curves of rainfall are very useful in extracting the information on the duration and magnitude of a storm. Also, intensifies at various time intervals in a storm can be obtained by the slope of the curve. For non-recording raingauges, mass curves are prepared from a knowledge of the approximate beginning and end of a storm and by using the mass curves of adjacent recording gauge stations as a guide.
A hyetograph is a plot of the intensity of rainfall against the time in The hyetograph is derived from the mass curve and is usually represented as a bar chart (Fig. 2.9). It is a very convenient way of represents characteristics of a storm and is particularly important in the development of a design storms to predict extreme floods. The area under a hyetograph represents the total precipitation received in that period. The time in used depends on the purpose; in urban-drainage problems small durations are used while in flood-flow computations in larger catchments the intervals are of about 6 h.
Point Rainfall MEAN PRECIPITATION OVER AN AREA Arithmetical-Mean Method
As indicated earlier, raingauges represent only point sampling of the areal distribution of a storm. In practice, however, hydrological analysis requires distribution of f the rainfall over an area, such as over a catchment. To convert the point rainfall values at various stations into an average value over a catchment the following three methods are in use:
When the rainfall measured at various stations in a catchment show little variation, the average precipitation over the catchment area is taken as the arithmetic mean of the station values. Thus if P1, P2,........., Pi......Pn are the rainfall values in a given period in N stations within a catchment, then the value of the mean precipitation ` P over the catchment by the arithmetic mean method is
MEAN PRECIPITATION OVER AN AREA
In practice, this method is used very rarely.
In this method the rainfall recorded at each station is given a weightage on the basis of an area closest to the station. The procedure of determining the weighing area is as follows: Consider a catchment area as in Fig. 2.10 containing three raingauge stations. There are three stations outside the catchment but in its neighbourhood. The catchment area is drawn to scale and the positions of the six stations marked on it. Stations 1 to 6 are joined to form a network of triangles. Perpendicular bisectors for, each of the sides of the triangle are drawn. These bisectors form a polygon around each station.
The boundary of the catchment, if it cuts the bisectors taken as the outer limit of the polygon. Thus for station 1, the bounding polygon is abcd. For station 2, kade is taken as the bounding polygon. These bounding polygons are called Thiessen polygons. The areas of these six Thiessen polygons are determined either with a planimeter or by using an overlay grid. If P1, P2, ..., P6, are the rainfall magnitudes recorded by the stations 1, 2, ..., 6 respectively, and A1, A2,.. .... A6, are the respective areas of the Thiessen polygons, then the average rainfall over the catchment ` P is given by
Thus in general for M stations,
The ratio Ai /A is called the weightage factor for each station.
The Thiessen-polygon method of calculating the average precipitation over an area is superior to the arithmetic-average method as some weightage is given to the various stations on a rational basis. Further, the raingauge stations outside the catchment are also used effectively. Once the weightage factors are determined, the calculation of ` P is relatively easy for a fixed network of stations.
An isohyetal is a line joining points of equal rainfall magnitude. In the isohyetal method, the catchment area is drawn to scale and the raingauge stations are marked. The recorded values for which area] average ` P is to be determined are then marked on the plot at appropriate stations. Neighbouring stations outside the catchment are also considered. The isohyets of various values are then drawn by considering point rainfalls as guides and interpolating between them by the eye (Figure 2.11). The procedure is similar to the drawing of elevation contours based on spot levels.
The area between two adjacent isohyets are then determined with planimeter. If the isohyets go out of catchment, the catchment boundary is used as the bounding line. The average value of the rainfall indicated by two isohyets is assumed to be acting over the inter-isohyet area. Thus P1, P2, .... Pn, are the values of isohyets and if a1, a2, ..., a n-1, are the inter-isohyet areas respectively, then the mean precipitation over the catchment of area A is given by
The isohyet method is superior to the other two methods especially when the stations are large in number.
For a rainfall of a given duration, the average depth decreases with the area in an exponential fashion given by
&#` P = Po exp (- KAn) (2.10)
where ` P = average depth in cms over an area A km², Po= highest amount of rainfall in cm at the storm centre and K and n are constant for a given region. On the basis of 42 severe most storms in north India, Dhar and Bhattacharya (1975) have obtained the following values for K and n for storms of different duration.
Since it is very unlikely that the storm centre coincides over a raingauge station, the exact determination of Po is not possible. Hence in the analysis of large area storms the highest station rainfall is taken as the average depth over an area of 25 km².
Maximum Depth-Area-Duration Curves
In many hydrological studies involving estimation of severe floods, it is necessary to have information on the maximum amount of rainfall of various duration occurring over various sizes of areas. The development of relationship, between maximum depth-area-duration for a region is known as DAD analysis and forms an important aspect of hydro-meteorological study.
FREQUENCY OF POINT RAINFALL
In many hydraulic-engineering applications such as those concerned with floods, the probability of occurrence of a particular extreme rainfall, e.g. a 24-h maximum rainfall, will be of importance. Such information is obtained by the frequency analysis of the point-rainfall data. The rainfall at a place is a random hydrologic process and the rainfall data at a place when arranged in chronological order constitute a time series. One of the commonly used data series is the annual series composed of annual values such as annual rainfall. If the extreme values of a specified event occurring in each year is listed, it also constitutes an annual series. Thus for example, one may list the maximum 24-h rainfall occurring in a year at a station to prepare an annual series of 24-h maximum rainfall values. The probability of occurrence of an event in this series is studied by frequency analysis of this annual data series. A brief description of the terminology and a simple method of predicting the frequency of an event is described in this section and for details the reader is referred to standard works on probability and statistics. The analysis of annual series, even though described with rainfall as a reference is equally applicable to any other random hydrological process, e.g. stream flow.
First, it is necessary to correctly understand the terminology used in frequency analysis. The probability of occurrence of an event (e.g. rainfall) whose magnitude is equal to or in excess of a specified magnitude X is denoted by P. The recurrence interval (also known as return period is defined as
T = 1/P
The purpose of the frequency analysis of an annual series is to obtain a relation between the magnitude of the event and its probability of exceedence. The probability analysis may be made either by empirical or by analytical methods.
A simple empirical technique is to arrange the given annual extreme series in descending order of magnitude and to assign an order number m. Thus for the first entry m = 1, for the second entry m = 2 and so on till the last event for which m = N = Number of years of record. The probability P of an event equalled to or exceeded is given by the Weibull formula
The recurrence interval T = 1/P = (N + 1)/m.
LOSSES FROM PRECIPITATION
Evaporation and transpiration form important links in the hydrologic cycle in which water is transferred to the atmosphere as water vapour. In engineering hydrology, runoff is the prime subject of study and evaporation and transpiration phases are treated as "losses". Evaporation from water bodies and soil masses together with the transpiration from vegetation is termed as evapotranspiration and is also known variously as water loss, total loss or total evaporation.
Before the rainfall reaches the outlet of a basin as runoff, certain demands of the catchment such as interception, depression storage and infiltration have to be met. If the precipitation not available for surface runoff is defined as "loss", then these processes are also "losses". In terms of groundwater the infiltration process is a "gain". Aspects of interception, depression storage and infiltration that are important from the point of view of engineering hydrology.
Evaporation is the process in which a liquid changes to the gaseous state at the free surface, below the boiling point through the transfer of heat energy. Consider a body of water in a pond. The molecules of water are in constant motion with a wide range of instantaneous velocities. An addition of heat causes this range and average speed to increase. When they cross over the water surface. Similarly, the atmosphere in the immediate neighborhood or the water surface contains water molecules within the water vapour in motion and some of them penetrate the water surface. The net escape of water molecules from the liquid state to the gaseous state constitute evaporation. Evaporation is a cooling process in that the latent heat of vaporization (at about 585 cal/g of evaporated water) must be provided by the water. The rate of evaporation is dependent on
- Vapour pressure at the water surface
- Air and water temperatures,
- Wind speed,
- Atmospheric pressure,
- Quality of water and
- Size of the water body
- Using evaporimeter data,
- Empirical evaporation equations and
- Analytical methods.
- They differ in the heat-storing capacity and heat transfer from the sides and bottom. The sunken pan and floating pan aim to reduce this deficiency. As a result of this factor the evaporation from a pan depends to a certain extent on its size. While a pan of 3 m diameter is known to give a value which is about the same as from a neighbouring large lake, a pan of size 1.0 m diameter indicates about 20% excess evaporation than that of the 3 m diameter pan.
- The height of the rim in an evaporation pan affects the wind action over the surface. Also, it casts a shadow of variable magnitude over the water surface.
- The heat-transfer characteristics of the pan material is different from that of the reservoir.
- Arid zones-One station for every 30,000 km²,
- Humid temperate climates-one station for every 50,000 km², and
- Cold regions-One station for every 100,000 km².
- It may be retained by the vegetation as surface storage and returned to the atmosphere by evaporation; a process termed interception loss
- It can drip off the plant leaves to join the ground surface or. the surface flow; this is known as throughfall;and
- The rainwater may run along the leaves and branches and down the stem to reach the ground surface. This part is called stemflow.
- Interception loss is solely due to evaporation and does not include transpiration, through fall or stemfiow.
- The type of soil,
- The condition of the surface reflecting the amount and nature of depression,
- The slope of the catchment and
- The antecedent precipitation, as a measure of the soil moisture. Obviously,
general expressions for quantitative estimation of this loss are not available.
Qualitatively, it has been found that antecedent precipitation has a very
pronounced effect on decreasing the loss to runoff in a storm due to depression.
Values of 0.50 cm in sand, 0.4 cm in loam and 0.25 cm in clay can be taken as
representatives for depression-storage loss during intensive storms.
- Flooding-type infiltrometer
- Rainfall simulator.
- Raindrop-impact effect is not simulated;
- Driving of the tube or rings disturbs the soil structure;
- Results of the infiltrometer depend to some extent on their size with the larger
- meters giving less rates than the smaller ones; this is due to the border effect.
- There is no need for costly stifling wells;
- A large change in the stage, as much as 30 m, can be measured;
- The recorder assembly can be quite far away from the sensing point; and
- Due to constant bleeding action there is less likelihood of the inlet getting blocked or choked.
- Vertical-axis meters, and
- Horizontal-axis meters.
- The stream should have a well-defined cross-section which does not change in various seasons.
- It should be easily accessible all through the year.
- The site should be in a straight, stable
- The gauging site should be free from backwater effects in the channel.
- The segment width should not be greater than 1/15 to 1/20 of the width of the river.
- The discharge in each segment should be less than 10% of the total discharge.
- The difference of velocities in adjacent segments should not be more than 20%.
- It should not be absorbed by the sediment, channel boundary and vegetation.should not chemically react with any of the above surfaces and also should not be lost by evaporation.
- It should be non-toxic.
- It should be capable of being detected in a distinctive manner in smalle concentrations.
- It should not be very expensive.
- Chemicals (common salt and sodium dischromate are typical);
- Fluorescent dyes (Rhodamine-WT and Sulpho-Rhodamine B Extra are typical);
- Radioactive materials (such as Bromine-82, Sodium-24 and Iodine-132).
- It is rapid and gives high accuracy;
- It is suitable for automatic recording of data;
- It can handle rapid changes in the magnitude and direction of flow as in tidal rivers;
- The cost of installation is independent of the size of rivers.
- Unstable cross-section í fluctuating weed growth,
- High loads of suspended solids, í air entertainment and Salinity and temperature changes.
- Flow measuring structures, and
- Slope area methods.
- Thin-plate structures are usually made from a vertically set metal plate. The V-notch, rectangular full width and contracted notches are typical examples under this category.
- Long-base weirs, also known as broad-crested weirs are made of concrete or masonry and are used for large discharge values.
- Flumes are made of concrete, masonry or metal sheets depending on their use and location. They depend primarily on the width constriction to produce a control section.
- Plot Q vs G on an arithmetic graph paper and draw a best-fit curve.
- By extrapolating the curve by eye judgement find a as the value of G corresponding to Q = 0. Using this value of a, plot log Q vs log (G-a) and verify whether the data plots as a straight line. If not, select another value in the neighbourhood of previously assumed value and by trial and error find an acceptable value of a which gives a straight line plot of log Q vs log (G-a).
- A graphical method due to Running' is as follows: The Q vs G data are plotted to an arithmetic scale and a smooth curve through the plotted points are drawn. Three points A, B and C on the curve are selected such that their discharges are in geometric progression, (Fig. 4.23) i.e.
- The ordinate at F is the required value of a, the gauge height corresponding to zero discharge. This method assumes the lower part of the stage-discharge curve to be a parabola.
- Plot Q vs G to an arithmetic scale and draw a smooth good-fitting curve by eye-judgement. Select three discharges Q1, Q2 and Q3, such that Q1/Q2 = Q2/Q3 and note from the curve the corresponding values of gauge readings G1, G2 and G3. From Eq. (4.27)
A number of optimization procedures that are based on the use of computers are available to estimate the best value of a. A trial-and-error search for a which gives the best value of the correlation coefficient is one of them.
- Changing characteristics caused by weed growth, dredging or channelencroachment, aggradation or degradation phenomenon in an alluvial channel,
- Variable backwater effects affecting the gauging section and
- Unsteady flow effects of a rapidly changing stage.
- Annual hydrographs showing the variation of daily or weekly or 10 daily mean flows over a year;
- Monthly hydrogaphs showing the variation of daily mean flows over a month;
- Seasonal hydrograps depicting the variation of the discharge in a particular season such, as the monsoon season or dry season; and
- Flood bydrographs or hydrographs due to a storm representing stream flow due to a storm over a catchment.
- Calculating the surface water potential of stream,
- Reservoir studies and
- Drought studies.
- The ranifall characteristics, such as magnitude intensity, distribution time and space and its variability;
- Catchment characteristics such as soil, vegetation, slope, geology, shape and drainage density;
- Climatic factors which influence evapotranspiration.
- The seasonal variation of rainfall is clearly reflected in the High stream discharges occur during monsoon months and 1ow which is essentially due to base flow is maintained during the year.
- The shape of the storm hydrograph and hence the peak flow is essentially controlled by the storm and physical characteristics the catchment. Evapotranspiration plays a minor role in this.
- The annual runoff volume (yield) of a stream is mainly controlled the amount of rainfall and evapotranspiration. The geology of catchment is significant to the extent of deep percolation losses.
- Correlation of stream flow and rainfall,
- Empirical equations, and
- Watershed simulations.
- Iinglis and DeSouza Formula As a result of careful stream gauging in 53 sites in Western India, Inglis and DeSouza (1929) evolved two regional formulae between annual runoff R in cm and annual rainfall p in cm as follows:
- For Ghat regions of western India R = 0.85 P - 30.5 (5.10)
- For Deccan plateau
- The slope of a flow-duration curve depends upon the interval of selected. For example, a daily stream flow data gives a steeper curve than a curve based on Monthly data for the same stream. This is due to the smoothening' of small peaks in monthly data.
- The presence of E.I.reservoir in a stream considerably modifies the virgin-flow-duration curve depending on the nature of flow regulation.
- The chronological sequence of occurrence of the flow is masked in the flow-duration curve. A discharge of say 1000 m³/s in a stream will have the same percentage Pp whether it occurred in January or June. This aspect, a serious handicap, must be kept in mind while interpreting a flow-duration curve.
- The flow-duration curve plotted on a log-log paper (Fig. 5.8) is useful in comparing the flow characteristics of different streams. A steep slope of the curve indicates a stream with a highly variable discharge. On the other hand, a flat slope indicates a slow response of the catchment to the rainfall and also indicates small variability. At the lower end of the curve, a flat portion indicates considerable base flow. A fiat curve on the upper portion is typical of river basins having large flood plains and also of rivers baying large snowfall during a wet season.
- In evaluating various dependable flows in the planning or water-resources engineering projects;
- In evaluating the characteristics of the hydropower potential of a river;
- In the design of drainage systems;
- In flood-control studies;
- In computing the sediment load and dissolved solids load of a stream; and
- In comparing the adjacent catchments with a view to extend the streamflow data.
- Infiltration characteristic : landuse and cover, soil type and geological conditions,lakes, swamps and other storage
- Channel characteristics : crosssection, roughness and storage capacity.
- Storm characteristics: precipitation, intensity, duration, magnitude and movementof Storm.
- Initial loss
- Rising limb,
- Crest segment,and
- Recession limb.
- The unit hydrograph represents the lumped response of the catchment to a unit rainfall excess of D-h duration to produce a direct-runoff hydrograph. It relates only the direct runoff to the rainfall excess. Hence the volume of water contained in the unit hydrograph must be equal to the rainfall excess. As 1 cm depth of rainfall excess is considered the area of the unit hydrograph is equal to a volume given by 1 cm over the catchment.
- The rainfall is considered to have an average intensity of excess rainfall (ER) of 1/D cm/h for the duration of the storm.
- The distribution of the storm is considered to be uniform all over the catchment.
Q5 = R1u5 + R2 u4 + R3 u3
- The development of flood hydrographs for extreme rainfall magnitudes for use in the design of hydraulic structures,
- Extention of flood-flow records based on rainfall records and
- Development of flood forecasting and warning systems based on rainfall.
- Precipitation must be from rainfall only. Snow-melt runoff cannot be satisfactorily represented by unit hydrograph.
- The catchment should not have unusually large storages in terms of tanks, ponds, large flood-bank storages, etc. which affect the line relationship between storage and discharge.
- If the precipitation is decidedly nonuniform, unit hydrographs can be expected to give good results.
- Rational method,
- Empirical method,
- Unit-hydrograph technique
- Flood-frequency studies.
The rate of evaporation is proportional to the difference between the saturation vapour pressure at the water temperature, ew and the actural vapour pressure in the air, ea. Thus EL = C (ew -ea) #9; #9; (3.1)
Where, EL = rate of evaporation (mm/day) and C = a constant; ew and ea, are in mm of mercury. This Equation is known as Dalton's law of evaporation after John Dalton (1802) who first recognised this law. Evaporation continue till ew = ea. If ew > ea, condensation takes place.
Other factors remaining same, the rate of evaporation increases with an increase in the water temperature. Regarding air temperature, although there is a general increase in the evaporation rate with increasing temperature, a high correlation between evaporation rate and air temperature does not exist. Thus for the same mean monthly temperature it is possible to have evaporation to different degrees in a lake in different months.
Wind aids in removing the evaporated water vapour from the zone of evaporation and consequently creates greater scope for evaporation. However, if the wind velocity is large enough to remove all the evaporated water vapour, any further increase in wind velocity does not influence the evaporation. Thus the rate of evaporation increases with the wind-speed up to a critical speed beyond which any further increase in the wind speed has no influence on the evaporation rate. This critical wind-speed value is a function of the size of the water surface. For large water bodies, high-speed turbulent winds are needed to cause maximum rate of evaporation.
Other factors remaining same, a decrease in the barometric pressure, as in high altitudes, increases evaporation.
When a solute is dissolved in water, the vapour pressure of the solution is less than that of pure water and hence causes reduction in the rate of evaporation. The per cent reduction in evaporation approximately corresponds to the percentage increase in the specific gravity. Thus, for example, under identical conditions evaporation from sea water is about 2-3% less than that from fresh water.
Heat Storage in Water Bodies
Deep water bodies have more heat storage than shallow ones. A deep lake may store radiation energy received in summer and release it in winter causing less evaporation in summer and more evaporation in winter com-pared to a shallow lake exposed to a similar situation. However, the effect of heat storage is essentially to change the seasonal evaporation rates and the annual evaporation rate is seldom affected.
Estimation of evaporation is of utmost importance in many hydrologic problems associated with planning and operation of reservoirs and irrigation systems. In and zones, this estimation is particularly important to conserve the scarce water resources. However, the exact measurement of evaporation from a large body of water is indeed one of the most difficult tasks. The amount of water evaporated from a water surface is estimated by the following methods:
Types of Evaporimeters
Evaporimeters are water-containing pans which are exposed to the atmosphere and the loss of water by evaporation measured in them at regular intervals. Meteorological data, such as humidity, wind movement, air and water temperatures and precipitation are also noted along with evaporation measurement.
Many types of evaporimeters are in use and a few commonly used pans are :
Class A Evaporation Pan
It is a standard pan of 1210 mm diameter and 255 mm depth used by theUS Weather Bureau and is known as Class A Land Pan. The depth of water is maintained between 18 em and 20 em (Fig. 3.1). The pan is normally made of unpainted galvanised iron sheet. Monel metal is used where corrosion is a problem. The pan is placed on a wooden platform of 15 cm height above the ground to allow free circulation of air below the pan. Evaporation measurements are made by measuring the depth of water with a hook gauge in a stilling well.
ISI Standard Pan
This pan evaporimeter specified by IS:5973-1970, also known as modified Class A Pan, consists of a pan 1220 mm in diameter with 255 mm of depth. The pan is made of copper sheet of 0.9 mm thickness, tinned inside and painted white outside (Fig. 3.2). A fixed point gauge indicates the level of water. A calibrated cylindrical measure is used to add or remove water maintaining the water level in the pan to a fixed mark. The top of the pan is covered fully with a hexagonal wire netting of galvanized iron to protect the water in the pan from birds. Further, the presence of a wire mesh makes the water temperature more uniform during day and night.
The evaporation from this pan is found to be less by about 14% compared to that from unscreened pan. The pan is placed over a square wooden platform of 1225 mm width and 100 mm height to enable circulation of air underneath the pan.
Colorado Sunken Pan
This pan, 920 mm square and 460 mm deep is made up of unpainted galvanised iron sheet and buried into the ground within 100 mm of the top (Fig. 3.3). The chief advantage of the sunken pan is that radiation and aerodynamic characteristics are similar to those of a lake. However, it has the disadvantages like difficult to detect leaks, extra care is needed to keep the surrounding area free from tall grass, dust etc. and expensive to install.
US Geological Survey Floating Pan
With a view to simulate the characteristics of a large body of water, this square pan (900 mm side and 450 mm depth) supported by drum floats in the middle of a raft (4.25 m X 4.87 m) is set afloat in a lake. The water level in the pan is kept at the same level as the lake leaving a rim of 75 mm. Diagonal baffles provided in the pan reduce the surging in the pan due to wave action. Its high cost of installation and maintenance together with the difficulty involved in performing measurements are its main disadvantages.
Pan Coefficient, Cp
Evaporation pans are not exact models of large reservoirs and have the following principal drawbacks:
In view of the above, the evaporation observed from a pan has to be corrected to get the evaporation from a lake under similar climatic and exposure conditions. Thus a coefficient is introduced as
Lake evaporation = Cp X pan evaporation
in which Cp = pan coefficient. The values of Cp in use for different pan are given in Table 3. I.
TABLE 3.1 VALUES OF PAN COEFFICIFNT Cp
|S. No.||Type of pan||Average value||Range|
|2||ISI Pan (modified Class A)||0.80||0.65-1.10|
|3||Colorado Sunken Pan||0.78||0.75-0.86|
|4||USGS Floating Pan||0.80||0.70-0.82|
It is usual to install evaporation pans in such locations where other meteorological data are also simultaneously collected. The WMO recommend the minimum network of evaporimeter stations as below:
Currently India has about 200 pan-evaporimeter stations maintained by the India Meteorological Department.
Transpiration is the process by which water leaves the body of a living plant and reach the atmosphere as water vapour. The water is taken up by the plant-root system and escapes through the leaves. The important factors affecting transpiration are: atmospheric vapour pressure, temperature, wind, light intensity and characteristics of the plant, such as the root and leaf systems. For a given plant, factors that affect the free-water evaporation also affect transpiration. However, a major difference
exists between transpiration and evaporation. Transpiration is essentially confined to daylight hours and the rate of transpiration depends upon the growth periods of the plant. Evaporation, on the other hand, continues all through the day and night although the rates are different.
While transpiration takes place, the land area in which plants stand also lose moisture by the evaporation of water from soil and water bodies. In hydrology and irrigation practice, it is found that evaporation and transpiration processes can be considered advantageously under one head as evapotranspiration. The term consumptive use is also used to denote this loss by evapotranspiration. For a given set of atmospheric conditions, evapotranspiration obviously depends on the availability of water. If sufficient moisture is always available to completely meet the needs of vegetation fully covering the area, the resulting evapotranspiration is called potential evapotranspiration (PET). Potential evapotranspiration no longer critically depends on soil and plant factors but depends essentially on climatic factors. The real evapotranspiration occurring in a specific situation is called actual evapotranspiration (AET). It is necessary to introduce at this stage two terms: field capacity and permanent wilting point. Field capacity is the maximum quantity of water that the soil can retain against the force of gravity. Any higher moisture input to a soil at field capacity simply drains away. Permanent wilting Point is the Moisture Content of a soil at which the moisture is no longer available in sufficient quantity to sustain the plants. At this stage, even though the soil contains some moisture, it will be so held by the soil grains that the roots of the plants are not able to extract it in sufficient quantities to sustain the plants and consequently the plants wilt. The field capacity and permanent wilting point depend upon the soil characteristics. The difference between these two moisture contents is called available water, the moisture available for plant growth. If the water supply to the plant is adequate, soil moisture will be at the field capacity and AET will be equal to PET. If the water supply is less than PET, the soil dries out and the ratio AET/PET would then be less than unity. The decrease of the ratio AET/PET with available moisture depends upon the type of soil and rate of drying. Generally, for clayey soils, AET/PET» 1.0 for nearly 50% drop in the available moisture. As can be expected, when the soil moisture reaches the permanent wilting point, the AET reduces to zero (Fig.3.5). For a catchment in a given period of time, the hydrologic budget can be written as
P – Rs – Go - Eact = D S ; ; (3.12)
Where, P = precipitation, Rs = surface runoff, Go = subsurface outflow, Eact = actual evapotranspiration (AET) and D S = change in the moisture storage. This water budgeting can be used to calculate Eact by knowing or estimating other elements of above equation. The sum of Rs and Go can be taken as the stream flow R at the basin outlet without much error.
Except in a few specialised studies, all applied studies in hydrology use PET for various estimation purposes. It is generally agreed that PET is a good approximation for lake evaporation.
A lysimeter is a special watertight tank containing a block of soil and set in a field of growing plants. The plants grown in the lysimeter are the same as in the surrounding field. Evapotranspiration is estimated in terms of the amount of water required to maintain constant moisture conditions within the tank measured either volumetrically or gravimetrically through an arrangement made in the lysimeter. Lysimeters should be designed to accurately reproduce the soil conditions, moisture content, type and size of the vegetation of the surrounding area. They should be so hurried that the soil is at the same level inside and outside the container. Lysimeter studies are time-consuming and expensive.
In special plots all the elements of the water budget in a known interval of time are measured and the evapotranspiration determined as
Evapotranspiration = [precipitation + irrigation input - runoff - increase in soil storage
- groundwater loss]
Measurements are usually confined to precipitation, irrigation input, surface runoff and soil moisture. Groundwater loss due to deep percolation is difficult to measure and can be minimised by keeping the moisture condition of the plot at the field capacity. This method provides fairly reliable results.
POTENTIAL EVAPOTRANSPIRATION OVER INDIA
Using Penman's equation and the available climatalogical data, PET estimated for the country has been made. The mean annual PET (in cm) over various parts of the country is shown in the form of isopleths - the lines on a map through places having equal depths of evapotranspiration. It is seen that the annual PET ranges from 140 to 180 cm over most parts of the country. The annual PET is highest at Rajkot, Gujarat with a value or 214.5 cm. Extreme south-east of Tamil Nadu also show high average values greater than 180 cm. The highest PET for southern peninsula is at Tiruchirapalli, Tamil Nadu with a value of 209 cm. The variation of monthly PET at stations located in various climatic zones in the country.
In the precipitation reaching the surface of a catchment the major abstraction is from the infiltration process. However, two other processes, though small in magnitude, operate to reduce the water volume available for runoff and thus act as abstractions. These are the interception process and the depression storage,and together they are called initial loss. This abstraction represents the quantity of storage that must be satisfied before overland runoff begins. The following two sections deal with these two processes briefly.
When it rains over a catchment not all the precipitation falls directly onto the ground. Before it reaches the ground, a part of it may be caught by the vegetation and subsequently evaporated. The volume of water so caught is called interception. The intercepted precipitation may follow one of the three possible routes:
The amount of water intercepted in a given area is extremely difficult to measure. It depends on the species composition density and also on the storm characteristics. It is estimated that of the total rainfall in an area during a plant-growing season the interception loss is about 10 to 20%.
Interception is satisfied during the first part of a storm and if an area experiences a large number of small storms, the annual interception loss due to forests in such cases will be high, amounting to greater than 25% of the annual precipitation. Quantitatively, the variation of interception loss with the rainfall magnitude per storm for small storms is as shown in Fig. 3.7. It is seen that the interception loss is large for a small rainfall and levels off to a constant value for larger storms.
For a given storm, the interception loss is estimated as
Ii = Si + Ki Et ; ; (3.18)
Where Ii = interception loss in mm, Si = interception storage whose value varies from 0.25 to 1.25 mm depending on the nature of vegetation, Ki = ratio of vegetal surface area to its projected area, E = evaporation rate in mm/h during the precipitation and t = duration of rainfall in hours.
It is found that coniferous trees have more interception loss deciduous ones. Also, dense grasses have nearly same interception losses as full grown trees and can account for nearly 20% of the total rainfall in a season. Agricultural crops in their growing season also contribute high interception losses. In view of these the interception process has a very significant impact on the ecology of the area related to silvicultural aspects and in the water balance of a region. However, in hydrological studies dealing with floods interception loss is rarely significant and is not separately considered, The common practice is to allow a lump sum value as the initial loss to be deducted from the initial period of the storm.
When the precipitation of a storm reaches the ground, it must first fill up all depressions before it can flow over the surface. The volume of water trapped in these depressions is called depression storage. This amount is eventually lost to runoff through processes of infiltration and evaporation and thus form a part of the initial loss. Depression storage depends on a vast number of factors the chief of which are :
It is well-known that when water is applied to the surface of a soil, a part of it seeps into the soil. This movement of water through the soil surface is known as infiltration and plays a very significant role in the runoff process by affecting the timing, distribution and magnitude of the surface runoff. Further, infiltration is the primary step in the natural groundwater recharge.
Infiltration is the flow of water into the ground through the soil surface and the process can be easily understood through a simple analogy. Consider a small container covered with wire gauze as in Fig. 3.8. If water is poured over the gauze, a part of it will go tainer and a part overflows. Further, the container can hold only a fixed quantity and when it is full no more flow into the container can take place. This analogy, though a highly simplified one, underscores two important aspects, viz., the maximum rate at which the ground can absorb water, the infiltration capacity and the volume of water that it can hold, the field capacity.
Since the infiltered water may contribute to groundwater discharge in addition to increasing the soil moisture, the process can be schematically modelled as in Fig. 3.9(a) and (b). This figure considers two situations, viz. low-intensity rainfall and high intensity rainfall, and is self explanatory.
The maximum rate at which a given soil at a given time can absorb water is defined as the infiltration capacity. It is designated as fc and is expressed in units of cm/h. The actual rate of infiltration f can be expressed as
f = fc when i > fc ; ; (3.19)
f = i when i < fc
where i = intensity of rainfall. The infiltration capacity of a soil is high at the beginning of a storm and has an exponential decay as the time elapses. The infiltration process is affected by a large number of factor and a few important ones affecting fc are described below.
Characteristics of Soil
The type of soil, viz. sand, silt or clay, its texture, structure,permeability and under drainage are the important characteristics under this category. A loose, permeable, sandy soil will have a larger infiltration capacity than a tight, clayey soil. A soil with good under drainage, i.e. the facility to transmit the infiltered water downward to a groundwater storage would obviously have a higher infiltration capacity. When the soils occur in layers, the transmission capacity of the layers determine the overall infiltration rate. Also a dry soil can absorb more water than one whose pores are already full. The land use has a significant influence on fc. For example, a forest soil rich in organic matter will have a much higher value of fc under identical conditions than the same soil in an urban area where it is subjected to compaction.
Surface of Entry
At the soil surface, the impact of raindrops causes the fines in the soils to be displaced and these in turn can clog the pore spaces in the upper layers. This is an important factor affecting the infiltration capacity.Thus a surface covered by grass and other vegetation which can reduce this process has a pronounced influence on the value of fc.
Water infiltrating into the soil will have many impurities, both in solution and in suspension. The turbidity of the water, especially the clay and colloid content is an important factor as such suspended particles block the fine pores in the soil and reduce its infiltration capacity. The temperature of the water is a factor in the sense that it affects the viscosity of the water which in turn affects the infiltration rate. >Contamination of the water by dissolved salts can affect the soil structure and in turn affect the infiltration rate.
MEASUREMENT OF INFILTRATION
Information about the infiltration characteristics of the soil at a given location can be obtained by conducting controlled experiments on small areas. The experimental set-up is called an infiltrometer.There are two kinds of infiltrometers :
This is a simple instrument consisting essentially of a metal cylinder, 30 cm diameter and 60 cm long, open at both ends. This cylinder is driven into the ground to a depth of 50 cm (Fig.3.10). Water is poured into the top part to a depth of 5 cm and a pointer is set to mark the water level. As infiltration proceeds, The volume is made up by adding water from a burette to keep the water level at the tip of the pointer. Knowing the volume of water added at different time intervals, the plot of the infiltration capacity vs lime is obtained. The experiments are continued till a uniform rate of infiltration is obtained and this may take 2-3 h. The surface of the soil is usually protected by a perforated disk to prevent formation capacity vs lime is obtained. The experiments re continued till a uniform rate of infiltration is obtained and this may take 2-3 h.
The surface of the soil is usually protected by a perforated disk to prevent formation of turbidity and its settling on the soil surface.
A major objection to the simple infiltrometer as above is that the infiltered water spreads at the outlet from the tube (as shown by dotted lines in Fig. 3.10) and as such the tube area is not representative of the area in which infiltration takes place. To overcome this a ring infiltrometer consisting of a set of two concentric rings (Fig.3.11) is used. In this two rings are inserted into the ground and water is maintained on the soil surface, in both the rings, to a common fixed level. The outer ring provides a water jacket to the infiltering water of the inner ring and hence prevents the spreading out of the infiltering water of the inner tube. The measurements of water volume is done on the inner ring only.
The main disadvantages of flooding-type infiltrometer are :
In this a small plot of land, of about 2 m X 4 m size, is provided with a size of nozzles on the longer side with arrangements to collect and measure the surface runoff rate. The specially designed nozzles produce raindrops falling from a height of 2 m and are capable of producing various intensities of rainfall. Experiments are conducted under controlled conditions with various combinations of intensities and durations and the surface runoff is measured in each case. Using the water-budget equation involving the volume of rainfall, infiltration and runoff, the infiltration rate and its variation with time is calculated. If the rainfall intensity is higher than the infiltration rate, infiltration-capacity values are obtained.
Rainfall simulator type infiltrometers given lower values than flooding type infiltrometers. This is due to the effect of the rainfall impact and turbidity of the surface water present in the former.
The typical variation of the infiltration capacity for two soils and for two initial conditions is shown in Fig. 3.12. It is clear from the figure that the infiltration capacity for a given soil decreases with time from the start of rainfall; it decreases with the degree of saturation and depends upon the type of soil. Horton (1930) expressed the decay of the infiltration capacity with time as
fct = infiltration capacity at any time t from start of the rainfall
fco = initial infiltration capacity at t = 0
fcf = final steady state value
td = duration of the rainfall and
Kh = constant depending upon the soil characteristics and vegetation cover.
The difficulty of finding the variation of the three parameters fco, fcf and Kh with soil characteristics and antecedent moisture conditions precludes the general use of Eq. (3.20).
It is apparent that infiltration-capacity values of soils are subjected to wide variations depending upon a large number of factors. Typically, a bare, sandy area will have fc » 1.2 cm/h and a bare, clay soil will have fs » 0.15 cm/h. A good grass cover or vegetation cover increases these values by as much as 10 times.
In hydrological calculations involving floods it is found convenient to use a constant value of infiltration rate for the duration of the storm. The average infiltration rate is called infiltration index and two types of indices are in common use.
The F index is the average rainfall above which the rainfall volume is equal to the runoff volume. The F index is derived from the rainfall hyetograph with the knowledge of the resulting runoff volume. The initial loss is also considered as infiltration. The F value is found by treating it
as a constant infiltration capacity. If the rainfall intensity is less than 0, then the infiltration rate is equal to the rainfall intensity; however, if the rainfall intensity is larger than F the difference between rainfall and infiltration in an interval of time represents the runoff volume (Fig. 3.13). The amount of rainfall in excess of the F index is called rainfall excess. The F index thus accounts for the total abstraction and enables runoff magnitudes to be estimated for a given rainfall hyetograph.
Streamflow representing the runoff phase of the hydrologic cycle is the most important basic data for hydrologic studies. It was seen in the previous chapters that precipitation, evaporation and evapotranspiration are all difficult to measure exactly and the presently, adopted methods have severe limitations. In contrast the measurement of streamflow is amenable to fairly accurate assessment. Interestingly, streamflow is the only part of the hydrologic cycle that can be measured accurately.
A stream can be defined as a flow >channel into which the surface runoff from a specified basin drains. Generally, there is considerable exchange of water between a stream and the underground water. Streamflow is measured in units of discharge (m³/s) occurring at a specified time and constitutes historical data. The measurement of discharge in a stream forms an important branch of Hydrometry, the science and practice of water measurement. This chapter deals with only the salient streamflow measurement techniques to provide an appreciation of this important aspect of engineering hydrology.
Streamflow measurement techniques can be broadly classified into two categories as (i) direct determination and (ii) indirect determination.
1. Direct determination of stream discharge:
(b) Dilution techniques,
(c) Electromagnetic method, and
(d) Ultrasonic method.
2. Indirect determination of stream flow:
(a) Hydraulic structures, such as weirs, flumes and gated structures
(b) Slope-area method.
Barring a few exceptional cases, continuous measurement of stream discharge is very difficult to obtain. As a rule, direct measurement of discharge is a very time-consuming and costly procedure. Hence, a two step procedure is followed. First, the discharge in a given stream is related to the elevation of the water surface (stage) through a series of careful measurements. In the next step the stage of the steam is observed routinely in a relatively inexpensive manner and the discharge is estimated by using the previously determined stage -discharge relationship. The observation of the stage is easy, inexpensive, and if desired, continuous readings can also be obtained. This method of discharge determination of streams is adopted universally.
MEASUREMENT OF STAGE
The stage of a river is defined as its water-surface elevation measured above a datum. This datum can be the mean-sea level (MSL) or an arbitrary datum connected independently to the MSL.
The simplest of stage measurements are made by noting the water surface in contact with a fixed graduated staff. The staff is made of a durable material with a low coefficient of expansion with respect both temperature and moisture. It is fixed rigidly to a structure, such an abutment, pier, wall, etc. The staff may be vertical or inclined with clearly and accurately graduated permanent markings. The markings a distinctive, easy to read from a distance and are similar to those or surveying staff. Sometimes, it may not be possible to read the entire range of water-surface elevations of a stream by a single gauge and in such cases the gauge is built in sections at different locations. Such gauges called sectional gauges (Fig. 4.1). When installing sectional gauges, must be taken to provide an overlap between various gauges and to all the sections to the same common datum.
If is a gauge used to measure the water-surface elevation from above the surface such as from a bridge or similar structure. In this a weight is lowered by a reel to touch the water surface. A mechanical counter measures the rotation of the wheel which is proportional to the length of
the wire paid out. The operating range of this kind of gauge is about 25 m.
Automatic Stage Recorders
The staff gauge and wire gauge described earlier are manual gauges. While they are simple and inexpensive, they have to be read at frequent intervals to define the variation of stage with time accurately. Automatic stage recorders overcome this basic objection of manual staff gauges and find considerable use in stream-flow measurement practice. Two typical automatic stage recorders are described below.
The Float-operated stage recorder is the most common type of automatic stage recorder in use. In this a float operating in a stifling well is balanced by means of a counterweight over the pulley of a recorder. Displacement of the float due to the rising or lowering of the water-surface elevation causes an angular displacement of the pulley and hence of the input shaft of the recorder. Mechanical linkages convert this angular displacement to the linear displacement of a pen to record over a drum driven by clockwork. The pen traverse is continuous with automatic reversing when it reaches the full width of the chart. A clockwork mechanism runs the recorder for a day, week or fortnight and provides a continuous plot of stage vs time. A good instrument will have a large-size float and least friction. Improvements over this basic analogue model consists of models that give digital signals recorded on a punched tape, magnetic tape or transmit directly onto a central data-processing centre.
To protect the float from debris and to reduce the water surface wave effects on the recording, stifling wells are provided in all float-type stage-recorder installations. Figure 4.2 shows a typical stifling well installation. Note the intake pipes that communicate with the river and flushing arrangement to flush these intake pipes off the sediment and debris occasionally. The water-stage recorder has to be located above the highest water level expected in the stream to prevent it from getting inundated
during floods. Further, the instrument must be properly housed in a suitable enclosure to protect it from weather elements and vandalism. On account of these, the water-stage-recorder installations prove to costly in most instances. A water-depth recorder is shown in Fig.4.3 (Plate 1).
In this gauge compressed air or gas is made to bleed out at a very small rate through an outlet Placed at the bottom of the river [fig.4.4, 4.5(plate 1) and 4.6 (Plate 11)]. A pressure gauge measures the gas pressure which in turn is equal to the water column above the outlet. A small change in the water-surface elevation is felt as a change in pressure from the present value at the pressure gauge and this in turn is adjusted by a servo-mechanism to bring the gas to bleed at the original rate under the new head. The pressure gauge reads the new water depth which transmitted to a recorder.
The bubble gauge has certain specific advantages over a float operated water stage recorder and these can be listed as under :
The stage data is often presented in the form of a plot of stage against chronological time (Figure 4.7) known as stage hydrograph. In addition to its use in the determination of stream discharge, stage data itself is of importance in flood warning and flood-protection works. Reliable long-term stage data corresponding to peak floods can be analysed statistically to estimate the design peak river stages for use in the design of hydraulic structures, such as bridges, weirs, etc. Historic flood stages are invaluable in the indirect estimation of corresponding flood discharges. In view of these multifarious uses, the river stage forms an important hydrologic parameter chosen for regular observation and recording.
MEASUREMENT OF VELOCITY
The measurement of velocity is an important aspect of many direct stream-flow measurement techniques. A mechanical device, called current meter, consisting essentially of a rotating element is probably the most commonly used instrument for accurate determination of the stream-velocity field. Approximate stream velocities can be determined by floats.
The most commonly used instrument in hydrometry to measure the velocity at a point in the flow cross-section is the current meter. It consists essentially of a rotating element which rotates due to the reaction of the stream current with an angular velocity proportional to the stream velocity. Historically, Robert Hooke (1663) invented a propeller-type current meter to measure the distance traversed by a ship. The present-day cup-type instrument and the electrical make-and-break mechanism were invented by Henry in 1868. There are two main types of current meters.
These instruments consist of a series of conical cups mounted around a vertical axis [Figs. 4.8 and 4.9 (Plate 111)]. The cups rotate in a horizontal plane and a cam attached to the vertical axial spindle records generated signals proportional to the revolutions of the cup assembly. The Price current meter and Gurley current meter are typical instruments under this category. The normal range of velocities is from 0.15 to 0.40 m/s. The accuracy of these instruments is about 1.50% at the threshold value and improves to about 0.30% at speeds in excess of 1.0 m/s.
Fig. 4.8 Vertical-axis current meter
Vertical-axis instruments have the disadvantage that they cannot be used where there are appreciable vertical components of velocities. For example, the instrument shows a positive velocity when it is lifted vertically in still water.
These meters consist of a propeller mounted at the end of a horizontal shaft [Fig. 4.10 (Plate 111) and 4.11]. These come in a wide variety of sizes with propeller diameters in the range 6 to 12 cm and can register velocities in the range of 0.15 to 4.0 m/s. Ott-, Neyrtec= [Fig. 4.12 (Plate IV)] and Watt-type meters are typical instruments under this kind.
Fig. 4.11 Propeller,type current meter
These meters are fairly rugged and are not affected by oblique flows of as much as 15°. The accuracy of the instrument is about 1% at the threshold value and is about 0.25% at a velocity of 0.3 m/s and above.
A current meter is so designed that its rotation speed varies linearly with the stream velocity v at the location of the instrument. A typical relationship is
v = a Ns+ b (4.1)
where v = stream velocity at the instrument location in m/s,
Ns = revolutions per second of the meter and
a, b = constants of the meter. Typical values of a and b for a standard size 12.5 cm dia Price meter (cup-type) is a = 0.65 and b = 0.03. Smaller meters of 5 cm. diameter cup assembly called pigmy meters run faster and are useful in measuring small velocities. The values of the meter constants for them are of the order of a = 0.30 and b = 0.003. Further, each instrument has a threshold velocity below which Eq. (4.1) is not applicable. The instruments have a provision to count the number of revolutions in a known interval of time. This is usually accomplished by the making and breaking of an electric
circuit either mechanically or electro-magnetically at each revolution of the shaft. In older model instruments the breaking of the circuit would be counted through an audible sharp signal ("tick") heard on a headphone. The revolutions per second is calculated by counting the number of such signals in a known interval of time, usually about 100 s. Present-day models employ electromagnetic counters with digital or analogue displays.
The relation between the stream velocity and revolutions per second of the meter as in Eq. (4.1) is called the calibration equation. The calibration equation is unique to each instrument and is determined by towing the instrument in a special tank. A towing tank is a long channel containing still water with arrangements for moving a carriage longitudinally over its surface at constant speed. The instrument to be calibrated is mounted on the carriage with the rotating element immersed to a specified depth in the water body in the tank. The carriage is then towed at a predetermined constant speed (v) and the corresponding average value of revolutions per second (Ns) of the instruments determined. This experiment is repeated over the complete range of velocities and a best-fit linear relation in the form of Eq. (4.1) obtained. The instruments are designed for rugged use and hence the calibration once done lasts for quite some time. However, from the point of view of accuracy it is advisable to check the instrument calibration once in a while and whenever there is a suspicion that the instrument is damaged due to bad handling or accident. In India excellent towing-tank facilities for calibration of current meters exist at the Central Water and Power Research Station, Pune and the Indian Institute of Technology, Madras.
This method of discharge measurement consists essentially of measuring the area of cross-section of the river at a selected section called the gauging site and measuring the velocity of flow through the cross-sectional area. The gauging site must be selected with care to assure that the stage-discharge curve is reasonably constant over a long period of about a few years. Towards this the following criteria are adopted:
At the selected site the section line is marked off by permanent survey markings and the cross-section determined. Towards this the depth at various locations are measured by sounding rods or sounding weights. When the stream depth is large or when quick and accurate depth measurements are needed, an electro-acoustic instrument called echo-depth recorder is used. In this a high frequency sound wave is sent down by a transducer kept immersed at the water surface and the echo reflected by the bed is also picked up by the same transducer. By comparing the interval between the transmission of the signal and the receipt of its the distance to the bed is obtained and is indicated or recorded in the instrument. Echo-depth recorders are particularly advantageous in high-velocity streams, deep streams and in streams with Oft or mobile beds. For purposes of discharge estimation, the cross-section is considered to be divided into a large number of subsections by verticals (Fig. 4.14). The average velocity in these subsections are measured by current meters or floats. It is quite obvious that the accuracy of discharge estimation increases with the number of subsections used.
However, the larger the number of segments, the larger is the effort, time and expenditure involved. The following are some of the guidelines to select the number of segments:
It should be noted that in natural rivers the verticals for velocity measurement are not necessarily equally spaced. The area-velocity method as above using the current meter is often called as the standard current meter method.
Discharge measurement of large alluvial rivers, such as the Ganga, by the standard current meter method is very time-consuming even when the flow is low or moderate. When the river is in spate, it is almost impossible to use the standard current meter technique due to the difficulty of keeping the boat stationary on the fast-moving surface of the stream for observation purposes. It is in such circumstance that the newly developed moving-boat techniques prove very helpful.
In this method a special propeller-type current meter which is free to move about a vertical axis is towed in a boat at a velocity vb at right angles to the stream flow. If the flow velocity is vf the meter will align itself in the direction of the resultant velocity vR making an angle q with the direction of the boat (Fig. 4.15). Further, the meter will register the velocity vR. If Vb is normal to vf,
&#vb = vR cos q and vf = vR sin q
If the time of transit between two verticals is D t, then the width between the two verticals (Figure 4.8) is
W = vb D t
The flow in the sub-area between two verticals i and i+1 where the depths are yi and yi+1. respectively, by assuming the current meter to measure the average velocity in the vertical, is
Thus by measuring the depths yi, velocity vR and q in a reach and the time taken to cross the reach D t, the discharge in the sub-area can be determined. The summation of the partial discharges D Qi over the whole width of the stream gives the stream discharge
Q = å D Qi (4.12)
In field applications a good stretch of the river with no shoals, islands, bars, etc. is selected. The cross-sectional line is defined by permanent land marks so that the boat can be aligned along this line. A motor boat with different sizes of outboard motors for use in different river stages is selected. A special current meter of the propeller-type, in which the velocity and inclination of the meter to the boat director 0 in the horizontal plane can be measured, is selected. The current meter is usually immersed at a depth of 0.5 m from the water surface to record surface velocities. To mark the various vertical sections and know the depths at these points, an echo-depth recorder is used.
In a typical run, the boat is started from the water edge and aligned to go across the cross-sectional line. When the boat is in sufficient depth of water, the instruments are lowered. The echo-depth recorder and current meter are commissioned. A button on the signal processor when pressed marks a distinctive mark line on the depth vs time chart of the echo-depth recorder. Further, it gives simultaneously a sharp audio signal to enable the measuring party to take simultaneous readings of the velocity vR and the inclination q . A large number of such measurements are taken during the traverse of the boat to the other bank of the river. The operation is repeated in the return journey of the boat. It is important that the boat is kept aligned along the cross-sectional line and this requires considerable skill on the part of the pilot. Typically, a river of about 2km stretch takes about 15 min for one crossing. A number of crossings are made to get the average value of the discharge. The surface velocities are converted to average velocities across the vertical by applying a coefficient [Eq. (4.5)]. The depths yi and time intervals D t are read from the echo-depth recorder chart. The discharge is calculated by Eqs. (4.11) and (4.12). In practical use additional coefficients may be needed to account for deviations from the ideal case and these depend upon the actual field conditions.
DILUTION TECHNIQUE OF STREAMFLOW MEASUREMENT
The dilution method of flow measurement, also known as the chemical method depends upon the continuity principle applied to a tracer which is allowed to mix completely with the flow. Consider a tracer which does not react with the fluid or boundary.
Let Co be the small initial concentration of the tracer in the streamflow. At section 1 a small quantity (volume V1) of high concentration C1 of this tracer is added (Fig. 4.16). Let section 2 be sufficiently far away on the downstream of section 1 so that the tracer mixes thoroughly with the fluid due to the turbulent mixing process while passing through the reach. The concentration profile taken at section 2 is schematically shown in Fig. 4.16. The concentration will have a base value of CO, increases from time t1 to a peak value and gradually reaches the base value of Co at time t2. The streamflow is assumed to be steady.
Fig. 4.16 Sudden-injection method
By continuity of the tracer material M1 = mass of tracer added at section 1 = V1 Cl
Thus the discharge Q in the stream can be estimated if for a known M1 the variation of C2 with time at section 2 and Co are determined. This method is known as sudden injection or gulp or integration method.
Another way of using the dilution principle is to inject the tracer of concentration C1 at a constant rate Qt at section 1. At section 2, the concentration gradually rises from the background value of Co at time t1 to a constant value C2 (Fig. 4.17). At the stready state, the continuity equation for the tracer is
This technique in which Q is estimated by knowing C1, C2, Co and Q is known as constant rate injection method or plateau gauging.
It is necessary to emphasise here that the dilution method of gauging is based on the assumption of steady flow. If the flow is unsteady and the flow rate changes appreciably during gauging, there will be a change in the storage volume in the reach and the steady-state continuity equation used to develop Eqs. (4.13) and (4.14) is not valid. Systematic errors can be expected in such cases.
The tracer used should have ideally the following properties:
The tracers used are of three main types:
Common salt can be detected with an error of ± 1% up to a concentration of 10 ppm. Sodium dichromate can be detected up to 0.2 ppm concentrations. Fluorescent dyes have the advantage that they can be detected at levels of tens of nanograms per litre (~1 in 1011) and hence require very small amounts of solution for injections. Radioactive traces are detectable up to accuracies of tens of picocuries per litre (~1 in 1014) and therefore permit large-scale dilutions. However, they involve the use of very sophisticated instruments and handling by trained personnel only. The availability of detection instrumentation, environmental effects- of the tracer and overall cost of the operation are chief factors that decide the tracer to be used.
Length of Reach
The length of the reach between the dosing section and sampling section should be adequate to have complete mixing of the tracer with the flow. This length depends upon the geometric dimensions of the channel cross-section, discharge and turbulence levels. An empirical formula suggested by Rimmar (1960) for estimation of mixing length for point injection of a tracer in a straight reach is
where L = mixing length (m), B = average width of the stream (m), d = average depth of the stream (m), C = Chezy coefficient of roughness which varies from 15 to 50 for smooth to rough bed conditions and g = acceleration due to gravity. The value of L varies from about 2 km for a mountain stream carrying a discharge of about 1.0 m³/s to about 100 km for river in a plain with a discharge of about 300 m³/s. The mixing length becomes very large for large rivers and is one of the major constraints of the dilution method. Artificial mixing of the tracer at the dosing station may prove beneficial for small streams in reducing the mixing length of the reach.
The dilution method has the major advantage that the discharge is estimated directly in an absolute way. It is a particularly attractive method for small turbulent streams, such as those in mountainous areas. Where suitable, it can be used as an occasional method for checking the calibration, stage-discharge curves, etc. obtained by other methods.
The electromagnetic method is based on the Faraday's principle that an emf is induced in the conductor (water in the present case) when it cuts a normal magnetic field. Large coils buried at the bottom of the channel carry a current I to produce a controlled vertical magnetic field, (Fig.4.18). Electrodes provided at the sides of the channel section measure the small voltage produced due to flow of water in the channel.
It has been found that the signal output E will be of the order of millivolts and is related to the discharge Q as
where d = depth of flow, I = current in the coil, and n, K1 and K2 are system constants.
The method involves sophisticated and expensive instrumentation and has been successfully tried in a number of installations. The fact that this kind of set-up gives the total discharge when once it has been calibrated, makes it specially suited for field situations where the cross-sectional properties can change with time due to weed growth, sedimentation, etc. Another specific application is in tidal channels where the flow undergoes rapid changes both in magnitude as well as in direction. Present-day commercially available electromagnetic flowmeters can measure the discharge to an accuracy of ±3%, the maximum channel width that can be accommodated being 100 m. The minimum detectable velocity is 0.005 m/s.
This is essentially an area-velocity method with the average velocity being measured by using ultrasonic signals. The method was first reported by Swengel (1955); since then it has been perfected and complete systems are available commercially.
Consider a channel carrying a flow with two transducers A and B fixed at the same level h above the bed and on either sides of the channel (Fig. 4.19). These transducers can receive as well as send ultrasonic signals. Let A send an ultrasonic signal to be received at B after an elapse time t1.
Similarly, let B send a signal to be received at A after an elapse time t2.
If C = velocity of sound in water,
where L = length of path from A to B and vp = component of the flow velocity in the sound path = v cos q . Similarly, from Fig.4.19 it is easy to see that
Thus for a given L and q , by knowing t1 and t2, the average velocity along the path AB, i.e. v can be determined. It may be noted that v is the average velocity at a height h above the bed and is not the average velocity V for the whole cross-section. However, for a given channel cross-section v can be related to V and by calibration a relation between v/V and h can be obtained. For a given set-up, as the area of cross-section is fixed, the discharge is obtained as a product of area and mean velocity V. Estimation of discharge by using one signal path as above is called single-path gauging. Alternatively, for a given depth of flow, multiple single paths can be used to obtain v for different h values. Mean velocity of flow through the cross-section is obtained by averaging these v values. This technique is known as multi-path gauging.
Ultrasonic flowmeters using the above principal have frequencies of the order of 500 kHz. Sophisticated electronics are needed to transmit, detect and evaluate the mean velocity of flow along the path. In a given installation a calibration (usually performed by the current-meter method) is needed to determine the system constants. Currently available commercial systems have been installed successfully at many places and accuracies of about 2% for the single-path method and 1% for the multipath method are reported. The systems are currently available for rivers up to 500 m width.
The specific advantages of the ultrasonic system of river gauging are:
The accuracy of this method is limited by the factors that affect the signal velocity and averaging of flow velocity, such as
Under this category are included those methods which make use of relationship between the flow discharge and the depths at specified locations. The field measurement is restricted to the measurement of t depths only. Two broad classifications of these indirect methods are:
Use of structures like notches, weirs flumes and sluice gates for flow measurement in hydraulic laboratories is well known. These conventional structures are used in field conditions also but their use is limited by the ranges of head, debris or sediment load of the stream and the back-water effects produced by the installations. To overcome many of these limitations a wide variety of flow measuring structures with specific advantages are in use.
The basic principle governing the use of a weir, flume or similar flow measuring structure is that these structures produce a unique control section in the flow. At these structures, the discharge Q is a function of the water-surface elevation measured at a specified upstream location,
Q =f (H) (4.20)
where H = water surface elevation measured from a specified datum.
Thus, for example, for weirs, Eq. (4.20) takes the form
&#Q = K Hn (4.21)
where H = head over the weir and K, n = system constants. Equation (4.20) is applicable so long as the downstream water level is below a curtain limiting water level known as the modular limit. Such flows which are independent of the downstream water level are known as free flows. If the tailwater conditions do affect the flow, then the flow is known as drowned or submerged flow. Discharges under drowned condition are obtained by applying a reduction factor to the free flow discharges. For example, the sumberged flow over a weir (Fig. 4.20) is estimated by the Villemonte formula,
Where, Qs = submerged discharge, Q1 = free flow discharge under head H1, H1 = upstream water surface elevation measured above the weir crest, H2 = downstream water surface elevation measured above the weir crest, n exponent of head in the free flow head discharge relationship [ Eq. (4.21)]. For a rectangular weir n = 1.5.
The various flow measuring structures can be broadly considered under three categories:
The resistance equation for uniform flow in an open channel, e.g. Manning’s formula can be used to relate the depths at either ends of a reach to the discharge. Figure 4.21 shows the longitudinal section of the flow in a between two sections, 1 and 2. Knowing the water-surface elevations at the two sections, it is required to estimate the discharge.
As indicated earlier the measurement of discharge by the direct method involves a two step procedure; the development of the stage-discharge relationship which forms the first step is of utmost importance. Once the stage-discharge (G-Q) relationship is established, the subsequent procedure consists of measuring the stage (G) and read the discharge (Q) from the (G-Q) relationship. This second part is a routine operation. Thus the aim of all current-meter and other direct-discharge measurements is to prepare a stage discharge relationship for the given channel gauging section. The stage-discharge relationship is also known as the rating curve.
The measured value of discharges when plotted against the corresponding stage give a relationship that represents the integrated effect of a wide range of channel and flow parameters. The combined effect of these parameters is termed control. If the (G-Q) relationship for a gauging section is constant and does not change with time, the control is said to be Permanent. If it changes with time, it is called shifting control.
A majority of streams and rivers, especially nonalluvial rivers exhibit permanent control. For such a case, the relationship between the stage and the discharge is a single-valued relation, which is expressed as
Q = Cr (G - a)b (4.26)
in which Q = stream discharge, G = gauge height (stage), a = a constant which represent the gauge reading corresponding to zero discharge, Cr and b are rating curve constants. This relationship can be expressed graphically by plotting the observed stage against the corresponding discharge values in an arithmetic or logarithmic plot [Fig. 4.22 (a) and (b)] . Logarithmic plotting is advantageous as Eq. (4.26) plots as a straight line in logarithmic coordinates. In Fig. 4.22(b) the straight line is drawn to best represent the data plotted as Q vs (G-a). Coefficients Cr and b need not be the same for the full range of stages.
Fig. 4.22 (a) Stage-discharge curve: arithmetic plot
Fig. 4.22 (b) Stage-discharge curve: logarithmic plot
The best values of Cr and b in Eq. (4.26) for a given range of stage are obtained by the least-square-error method. Thus by taking logarithms,
log Q = b log (G - a) + log Cr (4.27)
or Y = b X + b (4.27a)
in which Y = log Q, X = log (G - a) and b = log Cr.
For the best-fit straight line of N observations of X and Y,
In the above it should be noted that a is an unknown and its determination poses some difficulties. The following alternative methods are available for its determination:
At A and B vertical lines are drawn and then horizontal lines are drawn at B and C to get D and E as intersection points with the verticals. Two straight lines ED and BA are drawn to intersect at F.
The control that exists at a gauging section giving rise to a unique stage discharge relationship can change due to:
If the shifting control is due to variable backwater curves, the same stage indicates different discharges depending upon the backwater effect. To remedy this situation another gauge, called the secondary gauge or auxiliary gauge is installed some distance downstream of the gauging section and readings of both gauges are taken. The difference between the main gauge and the secondary gauge gives the fall (F) of the water surface in the reach. Now, for a given main-stage reading, the discharge under variable backwater condition is a function of the fall F, i.e.
&#Q = f (G, F) (4.30)
Schematically, this functional relationship is shown in Fig. 4.24. Instead of having a three-parameter plot, the observed data is normalized with respect to a constant fall value. Let Fo be a normalizing value of the fall
taken to be constant at all stages, F the actual fall at a given stage when the actual discharge is Q. These two fall values are related as
in which Qo = normalized discharge at the given stage when the fall is equal to Fo and m = an exponent with a value close to 0.5. From the observed data, a convenient value of Fo is selected. An approximate Qo vs G curve for a constant Fo called constant fall curve is drawn. For each observed data, Q/Qo and F/Fo values are calculated and plotted as Q/Qo vs F/Fo (Fig.4.25). This is called the adjustment curve. Both the constant fall curve and the adjustment curve are refined, by trial and error to get
the best-fit curves. When finalized, these two curves provide the stage-discharge information for gauging purposes. For example, if the observed stage is G1 and fall F1, first by using the adjustment curve the value of Q1/Qo is read for a known value of F1/Fo. Using the constant fall-rating curve, Qo is read for the given stage G1 and the actual discharge calculated as (Q1/Qo) x Qo.
When a flood wave passes a gauging station in the advancing portion of the wave the approach velocities are larger than in the steady flow at corresponding stages. Thus for the same stage, more discharge than in a steady uniform flow occurs. In the retreating phase of the flood wave the converse situation occurs with reduced approach velocities giving lower discharges than in an equivalent steady flow case. Thus the stage-discharge relationship for an unsteady flow will not be a single-valued relationship as in steady flow but it will be a looped curve as in Fig. 4.26. It may be noted that at the same stage, more discharge passes through the river during rising stages than in falling ones. Since the conditions for each flood may be different, different floods may give different loops.
If Qn is the normal discharge at a given stage under steady uniform flow and Qm is the measured (actual) unsteady flow the two are related as
where So = channel slope = water surface slope at uniform flow, dh/dt = rate of change of stage and Vw = velocity of the flood wave. For natural channels, Vw is usually assumed equal to 1.4 V, where V = average velocity for a given stage estimated by applying Manning's formula and the energy slope Sf. Also, the energy slope is used in place of So in the denominator of Eq. (4.32). If enough field data about the flood magnitude and dh/dt are available, the term (1/Vw So) can be calculated and plotted against the stage for use in Eq. (4.32). For estimating the actual discharge at an observed stage, QM/Qn is calculated by using the observed data of dh/dt. Here Qn is the discharge corresponding to the observed stage relationship for steady flow in the channel reach.
EXTRAPOLATION OF RATING CURVE
Most hydrological designs consider extreme flood flows. As an example, in the design of hydraulic structures, such as barrages, dams and bridges one need maximum flood discharges as well as maximum flood levels. While the design flood discharge magnitude can be estimated from other considerations, the stage-discharge relationship at the project site will have to be used to predict the stage corresponding to design-flood discharges. Rarely will the available stage-discharge data include the design-flood range and hence the need for extrapolation of the rating curve.
Before attempting extrapolation, it is necessary to examine the site and collect relevant data on changes in the river cross-section due to flood plains, roughness and backwater effects. The reliability of the extrapolated value depends on the stability of the gauging section control. A stable control at all stages leads, to reliable results. Extrapolation of the rating curve in an alluvial river subjected to aggradations and degradation is un-reliable and the results should always be confirmed by alternate methods, There are many techniques of extending the rating curve and two well-known methods are described here.
The conveyance of a channel in nonuniform flow is defined by the relation
Q = K Ö St #9; #9; #9; (4.33)
where Q = discharge in the channel, St = slope of the energy line and K = conveyance. If Manning's formula is used,
where n = Manning's roughness, A = area of cross-section and, R hydraulic radius. Since A and R are functions of the stage, the values of K for various values of stage are calculated by using Eq. (4.34) and plotted against the stage. The range of the stage should include values beyond the level up to which extrapolation is, desired. Then a smooth curve is fitted to the plotted points [Fig. 4.27(a)). Using the available discharge and stage data, values of St are calculated by using Eq. (4.33) as St = Q²/K² and are plotted against the stage. A smooth curve is fitted through the plotted points [Fig. 4.27(b)]. This curve is then extrapolated keeping in mind that Sf approaches a constant value at high stages.
In this technique the stage-discharge relationship by Eq.(4.26) is made use of. The stage is plotted against the discharge on a log-log paper. A best-fit linear relationship is obtained for data points lying in the high-stage range and the line is extended to cover the range of extrapolation. Alternatively, coefficients of Eq. (4.26) are obtained by the least-square-error method by regressing X on Y in Eq. (4.27a). For this Eq. (4.27a) is written as
X = a Y + C (4.35)
where X = log (G-a) and Y = log Q. The coefficients a and C are obtained as,
The relationship governing the stage and discharge is now
(G - a) = Cl Qa (4.36)
where Cl= antilog C.
By the use of Eq. (4.36) the value of the stage corresponding to a design flood discharge is estimated.
Runoff means the draining or flowing off of precipitation from a catchment area through a surface channel. It thus represents the output from the catchment in a given unit of time.
Consider a catchment area receiving precipitation. For a given precipitation, the evapotranspiration, initial loss, infiltration and detention-storage requirements will have to be first satisfied before the commencement of runoff. When these are satisfied, the excess precipitation moves over the lend surfaces to reach smaller channels. This portion of the runoff is called overland flow and involves building up of a storage over the surface and daining off of the same. Usually the lengths and depths of overland flow are small and the flow is in the larninar regime. Flows from several small channels join bigger channels and flows from these in turn combine to form a larger stream, and so on, till the flow reaches the catchment outlet. The flow in this mode where it travels all the time over the surface as over landflow and through the channels as open-channel flow and reaches the catchment outlet is called surface runoff.
A part of the precipitation that infilters moves laterally through upper crusts of the soil and returns to the surface at some location away from the point of entry into the soil. This component of runoff is known variously as interflow, through flow, storm seepage, subsurface, storm flow or quick return flow(Fig. 5.1). The amount of interflow depends on the biological conditions of the catchment. A fairly pervious soil overlying a hard impermeable surface is conducive to large interflows. Depending upon time delay between the infiltration and the outflow, the interflow is sometimes classified into prompt interflow, i.e. the interflow with the least time lag and delayed interflow.
Another route for the infiltered water is to undergo deep percolation and
reach the groundwater storage in the soil. The groundwater follows a complicated and long path of travel, and ultimately reaches the surface. The time lag, i.e. the difference in time between the entry into the soil and outflows from it is very large, being of the order of months and years. This part of runoff is called groundwater runoff or groundwater. Ground water flow provides the dry-weather flow in perennial streams.
Based on the time delay between the precipitation and the runoff, the runoff is classified into two categories; as direct runoff and Base flow
It is that part of runoff which enters the stream immediately after the precipitation. It includes surface runoff, prompt interflow and precipitation the channel surface. In the case of snow-melt, the resulting flow entering the stream is also a direct runoff. Sometimes terms such as direct runoff and storm runoff are used to designate direct runoff.
The delayed flow that reaches a stream essentially as groundwater flow is called base flow. Many times delayed interflow is also included under this category.
Runoff, representing the response of a catchment to precipitation reflects the integrated effects of a wide range of catchment, climate and precipitation characteristics. True runoff is therefore stream flow in the natural condition, i.e. without human intervention. Such a stream flow unaffected by works of man, such as structures for storage and diversion on a stream is called virgin flow. When there exist storage or diversion works on a stream, the flow in the downstream channel is affected by structures and hence does not represent the true runoff unless corrected for storage effects and the diversion of flow and return flow.
Each of these types have particular applications. Annual and seasonal hydrographs are of use in
Flood hydrographs are essential in analysing stream characteristics associated with floods.
In annual runoff studies it is advantageous to consider a water year beginning from the time when the precipitation exceeds the average evapotranspiration losses. In India, June Ist is the beginning of a water year which ends on May 31st of the following calendar year. In a water year a complete cycle of climatic changes is expected and hence the water budget will have the least amount of carry over.
RUNOFF CHARACTERISTICS OF STREAMS
A study of the annual hydrographs of streams enables one to classify stream into three classes as perennial, intermittent and ephemeral. A perennial stream is one which always carries some flow (Fig. 5.2). There is considerable amount of groundwater flow throughout the year.
Fig. 5.2 Perennial stream
Even during dry seasons the water table will be above the bed of the stream.
An intermittent stream has limited contribution from the groundwater. During the wet season the water table is above the stream bed and there is a contribution of the base flow to the stream flow. However, during dry seasons the water table drops to a level lower than that of the stream bed and the stream dries up. Excepting for an occasional storm which can produce a short-duration flow, the stream remains dry for the most part of the dry months (Fig. 5.3).
An ephemeral stream is one, which does not have any base-flow contribution. The annual hydrograph of such a river show series of short-duration spikes marking flash flows in response to storms (Fig. 5.4). The stream becomes dry soon after the end of the storm flow. Typically an ephemeral stream does not have any well-defined channel. Most rivers in and zones are of the ephemeral kind,
The flow characteristics of a stream depend upon:
The interrelationship of these factors is extremely complex. However, the risk of oversimplification, the following salient points can be noted:
YIELD (ANNUAL RUNOFF VOLUME)
The total quantity of water that can be expected from a stream in a given period such as a year is called the yield of the river. It is usual for yield to be referred to the period of a year and then it represents the annual runnoff volume. In this book the term yield is used to mean annual runoff volume unless otherwise specified. The calculation of yield is of fundamental importance in all water-resources development studies. The various methods used for the estimation of yield can be listed below:
The relationship between rainfall and the resulting runoff is quite complex and is influenced by a host of factors relating the catchment and climate. Further, there is the problem of paucity of data which forces one to adopt simple correlations for the adequate estimation of runoff. One of the most common methods is to correlate runoff, R with rainfall, P values. Plotting of R values against P and drawing a best-fit line can be adopted for very rough estimates. A better method is to fit a linear regression line between R and P and to accept the result if the correlation coefficient is nearer unity. The equation for straight-line regression between runoff R and rainfall P is
R = a P + b (5.1)
and the values of the coefficients a and b are given by
in which N = number of observation sets R and P. The coefficient of correlation r can be calculated as
The value of r lies between 0 to + 1 as R can have only positive correlalion with P. A value of 0.6 < r < 1.0 indicates good correlation. Further it should be noted that R ³ 0.
For large catchments, it is found advantageous to have an exponential relationship as
R= b Pm (5.5)
where b and m are constants, instead of the linear relationship given by Eq. (5. 1). In that case Eq. (5.5) is reduced to a linear form by logarithmic transformation as
1n R = m 1n p + 1n b (5.6)
and the coefficients m and ln b determined by using the method indicated earlier,
Since rainfall records of longer duration than the runoff data are normally available for a catchment, the regression equation [(Eq. (5.1) or (5.5)] can be used to generate synthetic runoff data by using rainfall data. While this may be adequate for preliminary studies, for accurate results sophisticated methods are adopted for synthetic generation the runoff data. Many improvements of the above basic rainfall-runoff correlation by consider additional parameters such as soil moisture or antecedent rainfall have been attempted. Antecedent rainfall influences the initial soil moisture and hence theinfiltration rate at the start of a storm. For calculation of the annual runoff from the annual rainfall a commonly used antecedent precipitation index, Pa is given by
Pa = a Pi + b Pi-1 + c Pi (5.7)
where Pi, Pi-1 and Pi-2 are the annual precipitation in the ith, (i-1)th (i-2)th year, i=current year, a, b and c are coefficients with their sum equal to unity. The coefficients are found by trial and error to produce best results.
There are many other types of antecedent precipitation indices in us achieve good correlations of rainfall and runoff. The use of coaxial chart with a defined antecedent precipitation index is given by Linsley et al.
The importance of estimating the water availability from, the available hydrologic data for purposes of planning water-resource projects was recognised by engineers even in the last century. With a keen sense of observation in their region of their activity many engineers of the past have developed empirical runoff estimation formulae. However, these formulae are applicable only to the region in which they were derived. These formulae are essentially rainfall-runoff relations with additional third or fourth parameters to account for climatic or catchement characteristics. Some of the important formulae used in various parts of India are given below.
Sir Alexander Bitinic measured the runoff from a small catchment near Nagpur (Area of 16 km2 ) during 1869 and 1872 and developed curves of cumulative runoff against cumulative rainfall. The two curves were found to be similar. From these he established percentages of runoff from rainfall. These percentages have been used in Madhya Pradesh and Vidarbha region of Maharashtra for the estimation of yield.
Barlow, the first Chief Engineer of the Hydro-Electric Survey of India (1915) on the basis of his study in small catchments (area~ 130Km²)in Uttar Pradesh expressed runoff R as
R = Kb P (5.8)
Where Kb = runoff coefficient which depends upon the type of catchment and nature of monsoon rainfall.
Strange (1928) studied the then available data on rainfall and runoff in the border areas, of present-day Maharashtra and Karnataka and obtained the values of the runoff coefficient
&#Ks = R/P (5.9)
as a function of the catchment character. For purposes of calculating the yield from the total monsoon rainfall, the catchments were characterised as "good", "average" and "bad'. Values of Ks for these catchments are shown in Table 5.9. Strange also gave a table for calculating the daily runoff from daily rainfall. In this the runoff coefficient depends not only on the amount of rainfall but also on the state of the ground. Three categories of the original ground state as 'dry', 'damp' and 'wet' are used by him.
Khosla (1960) analysed the rainfall, runoff and temperature data for various catchments in India and USA to arrive at an empirical relationship between runoff and rainfall. The time period is taken as a month. His relationship for Monthly runoff is
Rm = Pm - Lm (5.12)
and Lm = 0.48 Tm for Tm > 4.5°C
where Rm = Monthly runoff in cm and Rm 1 ³ 0
Pm = monthly rainfall in cm
Lm = monthly losses in cm
Tm = mean monthly temperature of the catchment in °C
For Tm £ 4.5°C, the loss Lm may provisionally be assumed as T°c 4.5 - 1 - 6.5, Lm (cm) 2.17 1.78 1.52 and Annual runoff = S Rm
Khosla's formula is indirectly based on the water-balance concept and the mean monthly catchment temperature is used to reflect the losses due to evapotranspiration. The formula has been tested on a number of catchments in India and is found to give fairly good results for the annual yield for use in preliminary studies. The formula can also be used to generate synthetic runoff data from historical rainfall and temperature data.
The hydrologic water-budget equation for the determination of runoff a given period is written as
&#R = Rs + Go = P - Eet - D S (5.13)
in which Rs = surface runoff, P = precipitation, Eet = actual evapotranspiration, Go = net groundwater outflow and D S = change in the soilmoisture storage. The sum of Rs and Go is considered to be given by the total runoff R, i.e. streamflow.
Starting from an initial set of values, one can use Eq. (5.13) to calculate R by knowing values of P and functional dependence of Eet, D S and infiltration rates with catchment and climatic conditions. For accurate results, the functional dependence of various parameters governing the runoff in the catchment and values of P at short time intervals are needed. Calculations can then be done sequentially to obtain the runoff at any time. However, the calculation effort involved is enormous if attempted manually. With the availability of digital computers the use of water budgeting as above to determine the runoff has become feasible. This technique of predicting the runoff, which is the catchment response to a given rainfall input is called deterministic watershed simulation. In this the mathematical relationships describing the interdependence of various parameters in the system are first prepared and this is called the model. The model is then calibrated i.e. the numerical values of various coefficients determined, by simulating the known rainfall-runoff records. The accuracy of the model is further checked by reproducing the results of another string of rainfall data for which runoff values are known. This phase is known as validation or verification of the model. After this, the model is ready for use.
Crawford and Linsley (1959) pioneered this technique by proposing a watershed simulation model known as the Stanford Watershed Model (SWM). This underwent successive refinements and the Stanford Watershed Model-IV (SWM-IV) suitable for use on a wide variety of conditions was proposed in 1966. The flow chart of SWM-IV is shown in Figure 5.5. The main inputs are hourly precipitation and daily evapotranspiration in addition to physical description of the catchment. The model considers the soil in three zones with distinct properties to simulate evapotranspiration, infiltration, overland flow, channel flow, interflow and baseflow phases of the runoff phenomenon. For calibration about 5 years of data are needed. In the calibration phase, the initial guess value of parameters are adjusted on a trial-and-error basis until the simulated response matches the recorded values. Using an additional of rainfall-runoff of about 5 years duration, the model is verified for its ability to give proper response. A detailed description of the application of SWM to an Indian catchment is given in Ref. 3.
SWM-IV has been tested in a number of applications and it has been found to give satisfactory results for the yield and not so satisfactory results in predicting peak values. However, it requires considerable familiarity with the model to arrive at optimal values in the calibrating stage. An improved version called Hydrocomp Simulation Program (HSP) (1966) gives a package of three simulation modules to solve a variety of water-shed-simulation problems. Another model called the SSARR model (Streamflow Synthesis and Reservoir Regulation Model) developed by Rockwood (1968) for the Columbia river basin, USA has been successfully tested on large watersheds. The Kentucky Watershed model (KWM) (1 970) is a revised and updated version of SWM-IV. KWM is used with an optimization programme called OPSET which generates best-fit parameter estimates. The successful use of KWM to catchments up to 1200 km² size have been reported.
It is well-known that the streamflow varies over a water year.One of the popular methods of studying this streamflow variability is through flow a duration curves. A flow-duration curve of a stream is a plot of discharge against the per cent of time the flow was equalled or exceeded. This curve is also known as discharge-frequency curve.
The streamflow data is arranged in a descending order of discharges, using class intervals if the number of individual values is very large. The data used can be daily, weekly, ten daily or monthly values. If N number of data points are used in this listing, the plotting position of any discharge (or class value) Q is
where m is the order number of the discharge (or class value), Pp = percentage probability of the flow magnitude being equalled or exceeded. The plot of the discharge Q against Pp is the flow-duration curve (Fig. 5.6). Arithmetic scale paper, or semi-log or log-log paper is used depending upon the range of data and use of the plot. The flow-duration curve represents the cumulative frequency distribution and can be considered to represent the streamflow variation of an average year. The ordinate Qp at any percentage probability Pp, represents the flow magnitude in an average year that can be expected to be equalled or exceeded Pp percent of time and is termed as Pp% dependable flow. In a perennial river Q100 = 100% dependable flow is a finite value. On the other hand in an intermittent or ephemeral river the streamflow is zero for a finite part of an year and as such Q100 is equal to zero.
The following characteristics of the flow-duration curve are of interest:
Figure 5.7 shows the typical reservoir regulation effect.
The virgin-flow->duration curve when plotted on a log probability paper plots as a straight line at least over the central region. From this property, various coefficients expressing the variability of the flow in a stream can be developed for the description and comparison of different streams.
Flow-duration curves find considerable use in water-resources planning and development activities. Some of the important uses are :
The flow-mass curve is a plot of the cumulative discharge volume against time Plotted in chronological order. The ordinate of the mass curve,V at any time t is thus
where `to’ is the time at the beginning of the curve and Q is the discharge rate. Since the hydrograph is a plot of Q vs t, it is easy to see that the flow-mass-curve is an integral curve (summation curse) of the hydrograph. The flow-mass curve is also known as Rippl’s mass curve after Rippl (1882) who suggested its use first. Figure 5.9 shows a typical flow-mass curve. Note that the abscissa is chronological time in months in this figure. It can also be in days, weeks or months depending on the data being analysed. The ordinate is in units of volume in million m³. Other units employed for ordinate include m³/s. day (cumec day), ha.m and cm over a catchment area.
The slope of the mass curve at any point represents dV/dt = Q = rate of flow at that instant. If two points M and N are connected by a straight
line, the slope of the line represents the average rate of flow that can be maintained between the times tm and tn if a reservoir of adequate storage is available. Thus the slope of the line AB joining the first and the last points of a mass curve represents the average discharge over the whole period of plotted record.
Consider a concentrated storm producing a fairly uniform rainfall of duration, Tr over a catchment. After the initial losses and infiltration losses are met, the rainfall excess reaches the stream through overland and channel flows. In the process of translation a certain amount of storage is built up in the overland and channel-flow phases. This storage gradually depletes after the cessation of the rainfall. Thus there is a time Jag between the occurrence of rainfall in the basin and the time when that water passes the gauging station at the basin outlet. The runoff measured at the stream gauging station will give a typical hydrograph as shown in Fig. 6.1. The duration of the rainfall is also marked in this figure to indicate the time lag in the rainfall and runoff. The hydrograph of this kind which results due to an isolated storm is typically single-peaked skew distribution of discharge and is known variously as storm hydrograph, flood hydrograph or simply hydrograph. It has three characteristic regions : (i) the rising limb AB, joining point A, the starting point of the rising curve and point B, the point of inflection, (ii) the crest segment BC between the two inflection with a peak P in between, (iii) the falling limb or depletion curve CD starting from the second point of inflection C. Other points of interest are tpk, the time to peak from the starting point A, the time from the centre of mass of rainfall to the centre of mass of hydrograph called lag time TL, the peak discharge Qp and the time base of the hydrograph TB.
The hydrograph is the response of a given catchment to a rainfall input.
It consists of flow in all the three phases of runoff, viz. surface runoff, interflow and base flow, and embodies in itself the integrated effects of a wide variety of atchnient and rainfall parameters having complex inter-actions. Thus two different storms in a given catchment Produce hydrhographs differing from each other. Similarly, identical storms in two catecments produce hydrographs that are different. The interactions of various storms and catchments are in general extremely complex. If one examines the record of a large number of flood hydrographs of a stream, it will be found that many of than will have kinks, multiple peaks, etc. resulting in shapes much different from the simple single-peaked hydrograph of Fig. 6.1. These complex hydrographs are the result of storm and catchment Peculiarities and their complex interactions. While it is theoretically possible to resolve a complex hydrograph into a set of simple hydro-graphs for purposes of hydrograph analysis, the requisite data of accept-able quality are seldom available. Hence, simple hydrographs resulting from isolated storms are preferred for hydrograph studies.
FACTORS AFFECTING FLOOD HYDROGRAPH
The factors that affect the shape of the hydrograph can be broadly grouped into climatic factors and physiographic factors. Each of these two groups contain a host of factors and the important on Table 6.1. Generally, the climatic factors control the rising limb and the recession limb is independent of storm characteristics, being determined by catchment characteristics only. Many of the factors listed in Table 6.1 are interdependent. Further, their effects are very varied and complicated. As such only important effects are listed below in qualitative terms only.
FACTORS AFFECTING FLOOD HYDROGRAPH
Basin characteristics : shape, size, slope, nature of the valley, elevation,
Shape of the Basin
The shape of the basin influences the time taken for water from the remote parts of the catchment to arrive at the outlet. Thus the occurrence of the peak and hence the shape of the hydrograph are affected by the basin shape. Fan-shaped, i.e. nearly semi-circular shaped catchments give high peak and narrow hydrographs while elongated catchments give broad-and low-peaked hydrographs. Figure 6.2 shows schematically the hydrographs from
three catchments having identical infiltration characteristics due to identical rainfall over the catchment. In catchment A the hydrograph is skewed to the left, i.e. the peak occurs relatively quickly. In catchment B, the hydrograph is skewed to the right, the peak occurring with a relatively longer lag. Catchment C indicates the complex hydrography produced by a composite shape.
Small basins behave different from the large ones in terms of the relative importance of various phases of the runoff phenomenon. In small catchments the overland flow phase is predominant over the channel flow. Hence the land use and intensity of rainfall have important role on the peak good. On large basins these effects are suppressed as the channel flow phase is more predominant. The peak discharge is found to vary as An where A is the catchment area and n is an exponent whose value is less than unity, being about 0.5. The time base of the hydrographs from larger basins will be larger than those of corresponding hydrographs from smaller basins. The duration of the surface runoff from the time of occurrence of the peak is proportional to Am, where m is an exponent less than unity and is of the order of magnitude of 0.2.
The slope of the main stream controls the velocity of now in the channel. As the recession limb of the hydrograph represents the depletion of storage, the stream channel slope will have a pronounced effect on this part of the hydrograph. Large stream slopes give rise to quicker depletion of storage and hence result in steeper recession limbs of hydrographs. This would obviously result in a smaller time base.
The basin slope is important in small catchments where the overland low is relatively more important. In such cases the steeper slope of the catchment results in larger peak discharges.
The drainage density is defined as the ratio of the total channel length to the total drainage area. A large drainage density creates situation conducive for quick disposal of runoff down the channels. This fast response is reflected in a pronounced peaked discharge. In basins with smaller drainage densities, the overland flow is predominant and the resulting hydrograph is squat with a slowly rising limb(Fig. 6.3).
Vegetation and forests increase the infiltration and storage capacities of the Soils. Further, they cause considerable retardance to the overland flow. Thus the vegetal cover reduces the peak flow. This effect is usually very Pronounced in small catchments of area less than 150 km². Further, the effect of the vegetal cover is prominent in small storms. In general, for two catchments of equal area, other factors being identical, the peak discharge is higher for a catchment that has a lower density of forest cover. Of the various factors that control the peak discharge, probably the only factor that can be manipulated is land use and thus it represents the only practical means of exercising long-term natural control over the flood hydrograph of a catchment.
Among climatic factors the intensity, duration and direction of storm movement are the three important ones affecting the shape of a flood hydrograph. For a given duration, the peak and volume of the surface runoff are essentially proportional to the intensity of rainfall. This aspect is made use of in the unit hydrograph theory of estimating peak-flow hydrographs, as discussed in subsequent sections of this chapter. In very small catchments, the shape of the hydrograph can also be affected by the intensity.
The duration of a storm of given intensity also has a direct proportional effect on the volume of runoff. The effect of duration is reflected in the rising limb and peak flow. Ideally, if a rainfall of given intensity i lasts sufficiently long enough, a state of equilibrium discharge proportional to iA is reached.
If the storm moves from upstream of the catchment to the downstream end, there will be a quicker concentration of flow at the basin outlet. This results in a peaked hydrograph. Conversely, if the storm movement is up the catchment, the resulting hydrograph will have a lower peak and longer time base. This effect is further accentuated by the shape of the catchment, with long and narrow catchments having hydrographs most sensitive to the storm-movement direction.
The rising limb of a hydrograph, also known as concentration curve represents the increase in discharge due to the gradual building up of storage in channels and over the catchment surface. The initial losses and high infiltration losses during the early period of a storm cause the discharge to rise rather slowly in the initial periods. As the storm continues, more and more flow from distant parts reach the basin outlet. Simultaneously the infiltration losses also decrease with time. Thus under a uniform storm over the catchment, the runoff increases rapidly with time. As indicated earlier, the basin and storm characteristics control the shape of the rising limb of a hydrograph.
The crest segment is one of the most important parts of a hydrograph as it contains the peak flow. The peak flow occurs when the runoff from various parts of the catchment simultaneously contribute the maximum amount of flow at the basin outlet. Generally for large catchments, the peak flow occurs after the cessation of rainfall, the time interval from the centre of mass of rainfall to the peak being essentially controlled by basin and storm characteristics. Multiple-peaked complex hydrographs in a basin can occur when two or more storms occur in close succession. Estimation of the peak flow and its occurrence, being very important in flood-flow studies are dealt in detail elsewhere in this book.
The recession limb which extends from the point of inflection at the end of the crest segment to the commencement of the natural groundwater flow represents the withdrawal of water from the storage built up in the basin during the earlier phases of the hydrograph. The starting point of the recession limb, i.e. the point of inflection represents the condition of maximum storage. Since the depletion of storage takes place after the cessation of rainfall, the shape of this part of the hydrograph is independent of storm characteristics and depends entirely on the basin characteristics.
The storage of water in the basin exists as surface storage, which includes both surface detention and channel storage, interflow storage, and groundwater storage, i.e. base-flow storage. Barnes (1940) showed that the recession of a storage can be expressed as
in which Qo and Qt are discharges at a time interval of t days with Qo being the initial discharge; Kr is a recession constant of value less than unity. Equation (6.1) can also be expressed in an alternative form of the exponential decay as
The recession constant Kr can be considered to be made up of three components to take care of the three types of storage as
Kr = Krs . Kri . Krb (6.3)
where Krs = recession constant for surface storage, Kri = recession constant for interflow and Krb = recession constant for base flow. Typically the values of these recession constants, when t is in days, are
Krs = 0.05 to 0.20
Kri = 0.50 to 0.85
Krb = 0.85 to 0.99
If the interflow is not significant Kri can be assumed to be unity. When Eq. (6.1) or (6.2) is plotted on a semilog paper with the discharge on the log-scale, it plots as a straight line and from this the value of Kr can be found.
For purposes of correlating DRH with the rainfall which produced the flow, the hydrograph of the rainfall is also pruned by deducting the losses. Figure 6.6 shows the hyetograph of a storm. The initial loss and infiltration losses are subtracted from it. The resulting hyetograph is known as effective rainfall hyetograph (ERH). It is also known as hyetograph of rainfall excess or supra rainfall. Both DRH and ERH represent the same total quantity but in different units. Since ERH is usually in cm/h plotted against time, the area of ERH multiplied by the catchment area gives Fig. 6.6 the total volume of direct runoff, which is the same as the area of DRH. The initial loss and infiltration losses are estimated based on the available data of the catchment.
The problem of predicting the flood hydrograph resulting from a known storm in a catchment has received considerable attention. A large number of methods are proposed to solve this problem and of them probably the most popular and widely used method is the unit-hydrograph method. This method was first suggested by Sherman in 1932 and has undergone many, refinements since then.
A unit hydrograph is defined as the hydrograph of direct runoff resulting from one unit depth (1 cm) of rainfall excess occurring uniformly over the basin and at a uniform rate for a specified duration (Dh). The term unit here refers to a unit depth of rainfall excess which is usually taken as 1 cm. The duration, being a very important characteristic, is used as a prefix to a specific unit hydrograph. Thus one has a 6-h unit hydrograph, 12-h unit hydrograph, etc. and in general a D-h unit-hydrograph applicable to a given catchment. The definition of a unit hydrograph implies the following:
Area under the unit hydrograph = 12.92 X 106m³
Hence Catchment area of the basin = 12.92 km²
Two basic assumptions constitute the foundations for the unit hydrograph theory. These are the time invariance and the linear response.
This first basic assumption is that the direct-runoff response to a given effective rainfall in a catchment is time-invariant. This implies that the DRH for a given ER in a catchment is always the same irrespective of when it occurs.
The direct-runoff response to the rainfall excess is assumed to be linear. This is the most important assumption of the unit-hydrograph theory. Linear response means that if an input x1(t) causes an output y1(t) and an input x2(t) causes an output y2(t), then an input x1(t)+x2(t) gives an output Yl(t)+Y2(t). Consequently, if X2(t) = rxl(t), then y2(t) = ry1(t). Thus if the rainfall excess in a duration D is r times the unit depth, the resulting DRH will have ordinates bearing ratio r to those of the corresponding D-h unit hydrograph. Since the area of the resulting DRH should increase by the ratio r, the base of the DRH will be the same as that of unit hydrograph.
The assumption of linear response in a unit hydrograph enables method of superposition to be used to derive DRHs. Accordingly, rainfall excesses of D-h duration each occur consecutively, their combined effect is obtained by superposing the respective DRHs with due cart taken to account for the Proper sequence of events. These aspects resulting from the assumption of linear response are made clearer following two illustrative examples.
DERIVATION OF UNIT HYDROGRAPHS
A number of isolated storm hydrographs caused by short spells of rainfall excess, each of approximately same duration [0.90 to 1.1 Dh] are selected from a study of the continuously gauged runoff of the stream. For each in these surface hydrographs, the is separated by adopting one the methods indicated in Sec.6.4. The area under each DRH is evaluated and the volume of the direct runoff obtained is divided by the catchment area to obtain the depth of ER. The ordinates of the various DRHs are divided by the respective ER values to obtain the ordinates of the unit hydroraph.
Flood hydrographs used in the analysis should be selected to meet the following desirable features with respect to the storms responsible for them:
í The storms should be isolated storms occurring individually.
í The rainfall should be fairly uniform during the duration and should cover the entire catchment area.
í The duration of the rainfall should be 1/5 to 1/3 of the basin lag.
í The rainfall excess of the selected storm should be high. A range ER values of 1.0 to 4.0 cm is sometimes preferred.
A number of unit hydrographs of a given duration are derived by the above method and then plotted on a common pair of axes as shown in Fig.6.1 1. Because of rainfall variations both in space and time due to some departures from the assumptions of the unit-hydrograph theory, the various unit hydrographs thus developed will not be identical is common practice to adopt a mean of such curves as the unit graph of a given duration for the catchment. While deriving the mean curve, the average of peak flows and time to peaks are first calculated. Then a mean curve of best fit, judged by eye, is drawn through the averaged peak to close on an averaged base length. The volume of DRH is calculated and any departure from unity is corrected by adjusting the value of the peak. The averaged ERH of unit depth is customarily drawn in the plot of the unit hydrograph to indicate the type and duration of rainfall causing the unit hydrograph.
By definition the rainfall excess is assumed to occur uniformly over the catchment during duration D of a unit hydrograph. An ideal duration for a unit hydrograph is one wherein small fluctuations in the intensity of rainfall within this duration do not have any significant effects on the run-off. The catchment has a damping effect on the fluctuations of the rainfall intensity in the runoff-producing process and this damping is a function of the catchment area. This indicates that larger durations are admissible for larger catchments. By experience it is found that the duration of the unit hydrograph should not exceed 1/5 to 1/3 basin lag. For catchments of sizes larger than 250 km² the duration of 6h is generally satisfactory.
Unit Hydrograph from a Complex Storm
When suitable simple isolated storms are not available, data from complex storms of long duration will have to be used in unit-hydrograph derivation. The problem is to decompose a measured composite flood hydrograph into its component DRHs and base flow. A common unit hydrograph of appropriate duration is assumed to exist. This problem is thus the inverse of the derivation of flood hydrograph through use of Eq. (6.5). Consider a rainfall excess made up of three consecutive durations of D-h and ER values of R1, R2 and R3. Figure 6.13 shows the ERH. By base flow separation of the resulting composite flood hydrograph a composite DRH is obtained (Fig.6.13). Let the ordinates of the composite DRH be drawn at a time interval of D h. At various time intervals 1D, 2D, 3D, ...from the start of the ERH, let the ordinates of the unit hydrograph be u1, u2, u3, ... and the ordinates of the composite DRH be Ql, Q2, Q3 ...
Q1 = R1u1
Q2 = R1u2 + R2 ul
Q3 = R1u 3 + R2 u2 + R3 ul (6.6)
Q4 = R1u4 + R2 u3 + R3 u2
From the Eq.(6.6) the values of ul, u2, u3, ... can be determined. However, this method suffers from the disadvantage that the errors propagate and increase as the calculations proceed. In the presence of errors the recession limb of the derived D-h unit hydrograph can contain oscillations and even negative values. Matrix methods with optimisation schemes are available for solving Eq. (6.6) in a digital computer.
Ideally, unit hydrographs are derived from simple isolated storms and if the duration of the various storms do not differ very much, say within a band width of ± 20% D, they would all be grouped under one average duration of D h. If in practical applications unit hydrographs of different duration are needed they are best derived from field data. Lack of adequate data normally precludes development of unit hydrographs covering a wide range of durations for a given catchment. Under such conditions a D-h unit hydrograph is used to develop unit hydrographs of differing durations, nD. Two methods are available for this purpose i.e. Method of superposition, and S-curve.
Method of Superposition
If a D-h unit hydrograph is available, and it is desired to develop a unit hydrograph of nD h, where n is an integer, it is easily accomplished by superposing n unit hydrographs with each graph separated from the previous one by D h. Figure 6.14 shows three 4-h
Unit hydrographs A, B and C, Curve B begins 4-h after B. Thus the combination of these three curves is a DRH of 3cm due to an ER of 12h duration. If the ordinates of this DRH art now divided by 3, one obtains a 12-h unit hydrograph. The calculations easy if performed in a tabular form (Table 6.6).
If it is desired to develop a unit hydrograph of duration mD, where m is a fraction, the method of superposition cannot be used. A different technique known as the S-curve method is adopted in such cases, and this method is applicable for rational values of m.
The S-curve, also known as S-hydrograph is a hydrograph produced by a continuous effective rainfall at a constant rate for an infinite period. It is a curve obtained by summation of an infinite series of D-h unit hydrographs spaced D-h apart. Figure 6.15 shows such a series of D-h hydrograph arranged with their starting points D-h apart. At any given time the ordinates of the various curves occurring at that time coordinate are summed up to obtain ordinates of the S-curve. A smooth curve through these ordinates result in an S-shaped curve called S-curve.
This S-Curve is due to a D-h unit hydrograph. It has an initial steep portion and reaches a maximum equilibrium discharge at a time equal to the tine base of the first unit hydrograph. The average intensity of ER producing the S-curve is 1/D cm/h and the equilibrium discharge,
where A = area of the catchment in km² and D = duration in hours of ER of the unit hydrograph used in deriving the S-curve. Alternatively
where A is in km² and D is in h. The quantity Qs represents the maximum rate at which an ER intensity of 1/D cm/h can drain out of a catchment of area A. In actual construction of an S-curve, it is found that curve oscillates in the top portion at around the equilibrium value due t magnification and accumulation of small errors in the hydrograph. When it occurs, an average smooth curve is drawn such that it reaches a value Qs at the time base of the unit hydrograph.
Consider two D-h S-curves A and B displaced by Th (Fig.6.16) If the ordinates of B are subtracted from that of A, the resulting curve is a DRH produced by a rainfall excess of duration T h and magnitude (1/D x T) cm. Hence if the ordinate differences of A and B, i.e. (SA – SB) are divided by T/D, the resulting ordinates denote a hydrograph due to an ER of 1 cm and of duration T h, i.e. a T-h unit hydrograph. The derivation of a T-h unit hydrograph as above can be achieved either by graphical mean arithmetic computations in a tabular form.
USE AND LIMITATIONS OF UNIT HYDROGRAPH
As the unit hydrographs establish a relationship between the ERH and DRH for a catchment, they are of immense value in the study of the hydrology of a catchment. They are of great use in
Unit hydrographs assume uniform distribution of rainfall over the catchment. Also, the intensity is assumed constant for the duration of the unfair excess. In practice, these two conditions are never strictly satisfied. Nonuniform areal distribution and variation in intensity within a storm are very common. Under such conditions unit hydrographs can still be used if the areal distribution is consistent between different storms. However, the size of the catchment imposes an upper limit on the applicability of the unit hydrograph. This is because in very large basins the centre of the storm can vary from storm to storm and each can give different DRHs under otherwise identical situations. It is generally felt that about 500 km² is the upper limit for unit-hydrograph use. Flood hydrographs in very large basins can be studied by dividing then into a number of smaller subbasins and developing DRHs by the unit-hydrograph method. These DRHs can then be routed through their respective channels to obtain the composite DRH at the basin outlet.
There is a lower limit also for the application of unit hydrographs. This limit is usually taken as about 200 ha. At this level of area, a number of factors affect the rainfall-runoff relationship and the unit hydrograph not accurate enough for the prediction of DRH.
Other limitations to the use of unit hydrographs are:
In the use of unit hydrographs very accurate reproduction of results should not be expected. Variations in the hydrograph base of as much as ± 200% and in the peak discharge by ± 10% are normally considered acceptable.
DURATION OF THE UNIT HYDROGRAPH
The choice of the duration of the unit hydrograph depends on the records. If recording raingauge data, are available any convenient depending on the size of the basin can be used. The choice is not only daily rainfall records are available. A rough guide for the choice of duration D is that it should not exceed the least of
(i)Time (ii) and (iii) Time of concentration.
A value of D equal to about 1/4 of the basin lag is about the best choice. Generally, for with areas more than 1200 km² a duration D = 12 hrs is preferred.
The distribution graph introduced by Bernard (1935) is a variation of unit hydrograph. It is basically a D-h unit hydrograph with ordinates showing the percentage of the surface runoff occurring in successive periods of equal time intervals of Dh. The duration of the rainfall excess (Dh) is taken as the unit interval and distribution-graph ordinates indicated at successive such unit intervals. Figure 6.17 shows a typical distribution graph. Note the ordinates plotted at 4-h intervals and the area under the distribution graph adds up to 100%. The use of the distribution graph to generate a DRH for a known ERH is exactly the same that of a unit hydrograph (Example 6.11). Distribution graphs are useful in comparing the runoff characteristics of different catchments.
SYNTHETIC UNIT HYDROGRAPH
To develop unit hydrographs to a catchment, detailed information about the rainfall and the resulting flood hydrograph are needed. However, such information would be available only at a few locations and in a majority of catchments, especially those which are at remote locations, the data would normally be very scanty. In order to construct unit hydrographs for such areas, empirical equations of regional validity which relate the salient hydrograph characteristics to the basin characteristics are available. Unit hydrographs derived from such relationships are known as synthetic-unit hydrographs. A number of methods for developing synthetic-unit hydrographs are reported in literature. It should, however, be re-numbered that these methods being based on empirical correlation’s are applicable only to the specific regions in which they were developed and could not be considered as general relationships for use in all regions.
INSTANTANEOUS UNIT HYDROGRAPH(IUH)
The unit-hydrograph concept discussed in the preceding sections considered a D-h unit hydrograph. For a given catchment a number of unit hydrographs of different durations are possible. The shape of these different unit hydrographs depend upon the value of D. Figure 6.20 shows a typical variation of the shape of unit hydrographs for different values of D. As D is reduced, the intensity of rainfall excess being equal to 1/D increases and the unit hydrograph becomes more skewed. A finite unit hydrograph is indicated as the duration DÕ 0. The limiting case of a unit hydrograph of zero duration is known asinstantaneous unit hydrograph (IUH). Thus IUH is a fictitious, conceptual unit hydrograph which
Represents the surface runoff from the catchment due to an instantaneous precipitation of the rainfall excess volume of 1 cm. IUH is designed u (t) or sometimes as u (0, t). It is a single-peaked hydrograph with a finite base width and its important properties can be listed as below:
1. 0 £ u (t) £ a positive value, for t > 0;
2. u (t) = 0 for t £ 0;
3. u (t) ® 0 as t ® a ;
4. = unit depth over the catchment; and
_. time to the peak < time to the centroid of the curve.
Consider an effective rainfall I(t ) of duration to applied to a catchment as in Fig.6.21. Each infinitesimal element of this ERH will operate on the IUH to produce a DRH whose discharge at time t is given by
where t' = t when t < to
and t' = to when t ³ to
Equation (6.22) is called the convolution integral or Duhamel integral. The integral of Eq.(6.22) is essentially the same as the arithmetical computation of Eq. (6.5).
The main advantage of IUH is that it is independent of the duration of ERH and thus has one parameter less than a D-h unit hydrograph. This fact and the definition of IUH make it eminently suitable for theoretical analysis of rainfall excess-runoff relationship of a catchment. For a given catchment IUH, being independent of rainfall characteristics, is indicative of the catchment storage characteristics.
A flood is an unusually high stage in a river-normally the level at which the river overflows its banks and inundates the adjoining area. The damages caused by floods in terms of loss of life, property and economic loss due to disruption of economic activity are all too well-known. Crores of rupees are spent every year in flood control and flood forecasting. The hydrograph of extreme floods and stages corresponding to flood peaks provide valuable data for purposes of hydrologic design. Further, of the various characteristics of the flood hydrograph, probably the most important and widely used parameter is the flood peak. At a given location in a stream, flood peaks vary from year to year and their magnitude constitutes a hydrologic series which enable one to assign a frequency to a given flood-peak value. In the design of practically all hydraulic structures of the peak flow that can be expected with an assigned frequency (say 1 in 1 00 years) is of primary importance to adequately proportion the structure to accommodate its effect. The design of bridges, culvert waterways and spillways for dams and estimation of scour at a hydraulic structure are some examples wherein flood-peak values are required. To estimate the magnitude of a flood peak the following alternative methods are available:
The use of a particular method depends upon (i) the desired objective, (ii) the available data and (iii) the importance of the project. Further the rational formula is only applicable to small-size (< 50km²) catchments and the unit-hydrograph method is normally restricted to moderate-size catchments with areas less than 5000 km².
Consider a rainfall of uniform intensity and very long duration occurring over a basin. The runoff rate gradually increases from zero to a constant value as indicated in Fig.7.1. The runoff increases as more and more flow from remote areas of the catchment reach the outlet. Designating the time taken for a drop of water from the farthest part of the catchment to
reach the outlet as the time of concentration, it is obvious that if the rainfall continues beyond tc, the runoit will be constant and at the peak value. The peak value of the runoff is given by Qp = C A i; for t ³tc (7.1)
where C = coefficient of runoff = (runoff/rainfall), A = area of the catchment and i = intensity of rainfall. This is the basic equation of the rational method. Using the commonly used units, Eq. (7.1) is written for field application as
Where, Qp = peak discharge (m³/s)
C = coefficient of runoff
itcp = the mean intensity of precipitation (mm/h) for a duration equal to tc and
an exceedence probability P
A = drainage area in km²
The use of this method to compute Qp requires three parameters: tc, (ict p) and C.
The empirical formulae u3ed for the estimation of the flood peak are essentially regional formulae based on statistical correlation of the observed peak and important catchment properties. To simplify the form of the equation, only a few of the many parameters affecting the flood peak are used. For example, almost all formulae use the catchment area as a parameter affecting the flood peak and most of them neglect the flood frequency as a parameter. In view of these, the empirical formula are applicable only in the region from which they were developed and when applied to other areas they can at best give approximate values.
By far the simplest of the empirical relationships are those which relate the flood Peak to the drainage area. The maximum good discharge Qp from a catchment area A is given by these formulae as Qp = f(A)
The unit-hydrograph technique described in the previous chapter used to predict the peak-flood hydrograph if the rainfall producing flood, infiltration characteristics of the catchment and the appropriate hydrograph are available. For design purposes, extreme rainfall situations are used to obtain the design storm, viz., the hyetograph of the rainfall excess causing extreme floods. The known or derived unit hydrograph of the catchment is then operated upon by the design storm to generate the desired flood hydrograph.
Hydrologic processes such as floods are exceedingly complex natural events. They are resultants of a number of component parameters and are therefore very difficult to model analytically. For example, the floods in a catchment depend upon the characteristics of the catchment; rainfall and antecedent conditions, each one of these factors in turn depend upon a host of constituent parameters. This makes the estimation of the flood peak a very complex problem leading to many different approaches. The empirical formulae and unit-hydrograph methods presented in the previous sections are some of them. Another approach to the prediction of flood, lows, and also applicable to other hydrologic processes such as rainfall etc. is the statistical method of frequency analysis. The values of the annual maximum flood from a given catchment area for large number of successive years constitute a hydrologic data series called the annual series. The data are then arranged in decreasing order of magnitude and the probability P of each event being equalled to or exceeded (plotting position) is calculated by the plotting-position formula
where m = order number of the event and N = total number of events in the data. The recurrence interval, T (also called the return period or frequency) is calculated as
The relationship between T and the probability of occurrence of various events is the same as described in Sec. 2.10. Thus, for example, the probability of occurrence of the event r times in n successive years is given by
where q = 1-p
Consider, for example, a list of flood magnitudes of a river arranged in blending order is shown in Table 7.2. The length of the record is 50 years.
The last column shows the return Period T of various flood magnitude, Q. A plot of Q Vs T yields the probability distribution, for small return periods (i.e. for interpolation) or where limited extrapolation is required, a, simple best-fitting curve through plotted points can be used as the probability distribution. A logarithmic scale for T is often advantageous. However, when larger extrapolations of Tare involved, theoretical probability distributions have to be used. In frequency analysis of floods the usual problem is to predict extreme flood events. Towards this, specific extreme-value distributions are assumed and the required statistical parameters calculated from the available data. Using these the flood magnitude for a specific return period is estimated. Chow (1951) has shown that most frequency-distribution functions applicable in hydrologic studies can be expressed by the following equation known as the general equation of hydrologic frequency analysis:
XT, = ` x + K s #9; #9; #9; (7.13)
Where, XT = value of the variegate X of a random hydrologic series with a return period T, ` x = mean of the variegate, s = standard deviation of the variegate, K = frequency factor which depends upon the return period, T and the assumed frequency distribution. Some of the commonly used frequency distribution functions for the predication of extreme flood values are Gumbel's extreme-value distribution, Log-Pearson Type III distribution, and Log normal distribution.
In the annual hydrologic data series of floods, only one maximum value flood per year is selected as the data point. It is likely that in s catchments there are more than one independent floods in a year many of these may be of appreciably high magnitude. To enable all the large flood peaks to be considered for analysis, a flood magnitude than an arbitrary selected base value are included in the analysis. Such a data series is called partial-duration series.
In using the partial-duration series, it is necessary to establish that all events considered are independent. Hence the partial-duration adopted mostly for rainfall analysis where the conditions of independency of events are easy to establish. Its use in flood studies is rather rare. The recurrence interval of an event obtained by annual series (TA) an partial duration series (Tp) are related by
From this it can be seen that the difference between TA and Tp is significant for TA < 10 years and that for TA > 20, the difference is negligibly small.
REGIONAL FLOOD FREQUENCY ANALYSIS
When the available data at a catchment is too short to conduct frequency analysis, a regional analysis is adopted. In this a hydrologically homogeneous region from the statistically point of view is considered. Available long-time data from neighbouring catchments are tested for homogeneity and a group of stations satisfying the test are identified. This group of stations constitutes a region and all the station data of this region are pooled and analysed as a group to find the frequency characteristics of the region. The mean annual flood Qma, which corresponds to a recurrence interval of 2.33 years is used for nondimensionalising the results. The variation of Qma with drainage area and the variation of QT/Qma with T, where QT is the discharge for any Tare the basic plots prepared in this analysis.
LIMITATIONS OF FREQUENCY STUDIES
The flood-frequency analysis described in the previous sections is a direct means of estimating the desired flood based upon the available flood-flow data of the catchment. The results of the frequency analysis depend upon the length of data. The minimum number of years of record required to obtain satisfactory estimates depends upon the variability of data and hence on the physical and climatological characteristics of the basin. Generally a minimun of 30 years of data is considered as essential. Smaller lengths of records are also used when it is unavoidable. However, frequency analysis should not be adopted if the length of records is less than 10 years.
Flood-frequency studies are most reliable in climates that are uniform from year to year. In such cases a relatively short record gives a reliable picture of the frequency distribution. With increasing lengths of flood records, it affords a viable alternative method of flood-flow estimation in most cases.
In the design of hydraulic structures it is not practical from economic considerations to provide for the safety of the structure and the system It the maximum-possible flood in the catchment. Small structures such as culverts and storm drainages can be designed for less severe floods as the consequences of a higher-than-design flood may not be very serious. It can cause temporary inconvenience like the disruption of traffic and very rarely severe property damage and loss of life. On the other hand, storage structures such as dams demand greater attention to the magnitude of floods used in the design. The failure of these structures causes large loss of life and great property damage on the down-stream of the structure.
From this it is apparent that the type, importance of the structure and economic development of the surrounding area dictate the design criteria for choosing the flood magnitude. This section highlights the procedures adopted in selecting the flood magnitude for the design of some hydraulic structures.
Flood adopted for the design of a structure.
Spillway Design Flood
Design flood used for the specific purpose of designing. the spillway of a storage structure. This term is frequently used to denote the maximum discharge that can be passed in a hydraulic structure without any damage or serious threat to the stability of the structure.
Standard Project Flood (SPF)
The flood that would result from a severe combination of meteorological and hydrological factors that are reasonably applicable to the region. Extremely rare combinations of factors are excluded.
Probable Maximum Flood (PMF)
The extreme flood that is Physically possible in a region as a result of severest combinations, including rare combinations of meteorological and hydrological factors.
The PMF is used in situations where the failure of the structure would result in loss of life and catastrophic damage and as such complete security from potential floods is sought. On the other hand, SPF is often used where the failure of a structure would cause less severe damages. Typically, the SPF is about 40 to 60% of the PMF for the same drainage basin. The criteria used for selecting the design flood for various hydraulic structures vary from one country to another.
To estimate the design flood for a project by the use of a unit hydrograph, one needs the design storm. This can be the storm-producing probable maximum precipitation (PMP) for deriving PMF or a standard project (SPS) for SPF, calculations. The computations are performed by experienced hydrometeorologists by using meteorological data. Various methods ranging from highly sophisticated hydrometeorological methods to simple analysis of past rainfall data are in use depending on the availability of reliable relevant data and expertise.
The following is a brief outline of a procedure followed in India:
The duration of the critical rainfall is first selected. This will be the basin lag if the flood peak is of interest. If the flood volume is of prime interest, the duration of the longest storm experienced in the basin is selected. Past major storms in the region which conceivably could have occurred in the basin under study are selected. DAD analysis is performed and the enveloping curve representing maximum depth-duration relation for the study basin obtained.
Rainfall depths for convenient time intervals (e.g.6h) are scaled from the enveloping curve. These increments are to be arranged to get a critical sequence which produces the maximum flood peak when applied to the relevant unit hydrograph of the basin.
In critical sequence of rainfall increments can be obtained by trial and error. Alternatively, increments of precipitation are first arranged in a of relevant unit hydrograph ordinates such that
- the maximum increment is against the maximum unit hydrograph ordinate, he second highest rainfall increment is against the second largest unit hydrograph ordinate, and so on, and
- the sequence of rainfall increments arranged above is now reversed, with the last item first and first item last. The new sequence gives the design storm.
The design storm is then combined with hydrologic abstractions most conducive to high runoff, viz. low initial loss and lowest infiltration rate to get the hyetograph of rainfall excess to operate upon the unit hydrograph.
The flood hydrograph discussed in Chap.6 is infact a wave. The stage and discharge hydrographs represent the passage of waves of river depth and discharge respectively. As this wave moves down the river, the shape of the wave gets modified due to various factors, such as channel storage, resistance, lateral addition or withdrawal of flows, etc. When a flood wave passes through a reservoir, its peak is attenuated and the time base is enlarged due to the effect of storage. Flood waves passing down a river have their peaks attenuated due to friction if there is no lateral inflow. The addition of lateral inflows can cause a reduction of attenuation or even amplification of a flood wave.
Flood routing is the technique of determining the flood hydrograph at a section of a river by utilizing the data of flood flow at one or more upstream sections. The hydrologic analysis of problems such asflood forecasting, flood protection, reservoir design and spillway design invariably include flood routing. In these applications two broad categories of routing can be recognised. These are Reservoir routing, and channel routing.
In reservoir routing the effect of a flood wave entering a reservoir is studied. Knowing the volume-elevation characteristic of the reservoir and the outflow-elevation relationship for the spillways and other outlet structures in the reservoir, the effect of a flood wave entering the reservoir is studied to predict the variations of reservoir elevation and outflow discharge with time. This form of reservoir routing is essential in the design of the capacity of spillways and other reservoir outlet structures and in the location and sizing of the capacity of reservoirs to meet specific requirements.
In channel routing the changes in the shape of a hydrograph as it travels down a channel is studied. By considering a channel reach and an input hydrograph at the upstream end, this form of routing aims to predict the flood hydrograph at various sections of the reach. Information on the flood-peak attenuation and the duration of high-water levels obtained by channel routing is of utmost importance in flood-forecasting operations and flood-protection works.
A variety of routing methods are available and they can be broadly classified into two categories as: (i) hydrologic routing and (ii) hydraulic routing. Hydrologic-routing methods employ essentially the equation of continuity. Hydraulic methods, on the other hand, employ the continuity equation together with the equation of motion of unsteady flow. The basic differential equations used in the hydraulic routing, known as St. Venant equations altord a better description of unsteady flow than hydrologic methods.
HYDROLOGIC CHANNEL ROUTING
In reservoir routing presented in the Previous sections, the storage was a unique function of the Outflow discharge S=f(Q). However, in channel routing the storage is a function of both Outflow and inflow discharges and hence a different routing method is needed. The flow in a river during a flood belongs to the category of gradually varied unsteady flow. surface in a channel reach is not only not parallel to the channel bottom but also varies with time (Fig.8.7). Considering a channel reach having a flood flow, the total volume in storage can be considered under two categories as Prism storage and wedge storage.
It is the volume that would exist if uniform flow occurred at the downstream depth, i.e. the volume formed by an imaginary plane parallel to the channel bottom drawn at the outflow section water surface.
It is the wedge-like volume formed between the actual water surface profile and the top surface of the prism storage. At a fixed depth at a downstream section of a river reach the prism storage is constant while the wedge storage changes from a positive value at an advancing flood to a negative value during a receding flood.
The prism storage Sp is similar to a reservoir and can be expressed as a function of the outflow discharge, Sp =f(Q). The wedge storage Sw can be accounted for by expressing it as Sw = f(I). The total storage in the channel reach can then be expressed as
where K and x are coefficients and m = a constant exponent. It has been found that the value of m varies from 0.6 for rectangular channels to a value of about 1.0 for natural channels.
Using m = 1.0, Eq. (8.8) reduces to a linear relationship for S in terms of I and Q as
S= K [x I + (1-x) Q] (8.9)
and this relationship is known as the Muskingum equation. In this the parameter x is known as weighting factor. When x = 0, obviously the storage is a function discharge only and the Eq. (8.9) reduces to
S = K Q (8.10)
Such a storage is known as linear storage or linear reservoir. When x = 0.5 both the inflow and outflow are equally important in determining the storage. The coefficient K is known as storage-time constant and has the dimensions of time. It is approximately equal to the time of travel of a flood wave through the channel reach.
HYDRAULIC METHOD OF FLOOD ROUTING
The hydraulic method of flood routing is essentially a solution of the basic St Venant equations [Eq's(8.4) and (8.5)] . These equations are simultaneous, quasi-linear, first-order partial differential equations of the hyperbolic type and are not amenable to general analytical solutions. Only for highly simplified cases can one obtain the analytical solution of these equations. The development of modern, high-speed digital computers during the past two decades has given rise to the evolution of many sophisticated numerical techniques. The various numerical methods for solving St Venant equations can be broadly classified into two categories, Approximate methods and Complete numerical methods.
These are based on the equation of continuity only or on a drastically curtailed equation of motion. The hydrological method of storage routing and Muskingum channel routing discussed earlier belong to this category. Other methods in this category are diffusion analogy and kinematic Wave models.
Complete Numerical Methods
These are the essence of the hydraulic method of routing. In the direct method, the partial derivatives are replaced by finite differences and the resulting algebraic equations are then solved. In the method of characteristics (MOC) St Venant equations are converted into a pair of ordinary differential equations (i.e. characteristic forms) and then solved by finite difference techniques. In the finite element method (FEM) the system is divided into a number of elements and partial differential equations are integrated at the nodal points of the elements.
The numerical schemes are further classified into explicit and implicit methods. In the explicit method the algebraic equations are linear and the dependent variables are extracted explicitly at the end of each time step. In the implicit method the dependent variables occur implicitly and the equations are nonlinear. Each of these two methods has a host of finite-differencing schemes to choose from.
The term "flood control" is commonly used to denote all the measures adopted to reduce damages to life and property by floods. As there is always a possibility, however remote it may be, of an extremely large flood occurring in a river the complete control of the flood to a level of zero loss is neither physically possible nor economically feasible. The flood control measures that are in use can be classified as:
1. Structural methods
- Storage and detention reservoirs Levees (flood embankments),
- Channel Improvement,Flood ways (new channels), and
2. Non-structural methods: Flood plain zoning, #9; Flood warning and evacuation.
Storage reservoirs offer one of the most reliable and effective methods of control. Ideally, in this method, a part of the storage in the reservoir kept apart to absorb the incoming flood. Further, the stored water is deceased in a controlled way over an extended time so that downstream channels do not get flooded. Figure 8.13 shows an ideal operating plane of flood control reservoir. As most of the present-day storage reservoirs aye multipurpose commitments, the manipulation of reservoir levels to satisfy many conflicting demands is a very difficult and complicated task. It so happens that many storage reservoirs while reducing the floods and food damages do not always aim at achieving optimum benefits in the flood-control aspect. To achieve complete flood control in the entire length of a river, a large number of reservoirs at strategic locations in the catchment will be necessary.
The Hirakud and Damodar Valley Corporation (DVC) reservoirs are examples of major reservoirs in the country which have specific volumes earmarked for flood absorption.
A detention reservoir consists of an obstruction to a river with an un-controlled outlet. These are essentially small structures and operate to reduce the flood peak by providing temporary storage and by restriction of the outflow rate. These structures are not common in India.
Levees, also known, as dikes or flood embankments are earthen banks constructed parallel to the course of the river to confine it to a fixed course and limited cross-sectional width. The heights of levees will be higher than the design flood level with sufficient free board. The confinement of the river to a fixed path frees large tracts land from inundation and consequent damage (Fig. 8.14).
Levees are one of the oldest and most common methods of flood protection works adopted in the world. Also, they are probably the cheapest of structural flood-control measures. While the protection offered by a levee against flood damage is obvious, what is not often appreciated is the potential damage in the event of a levee failure. The levees, being earth embankments require considerable care and maintenance. In the event of being overlapped, they fail and the damage caused can be enormous. In fact, the sense of protection offered by a levee encourages economic activity along the embankment and if the levee is overlapped the loss would be more than what would have been if there were no levees. Confinement of flood banks of a river by levees to a narrower space leads to higher flood levels for a given discharge. Further, if the bed level of the river also rises, as they do in agrading rivers, the top of the levees has to be raised at frequent time intervals to keep up its safety margin.
The design of a levee is a major task in which costs and economic benefits have to be considered. The cross-section of a levee will have to be designed like an earth dam for complete safety against all kinds of saturation and drawdown possibilities. In many instances, especially in locations I where important structures and industries are to be protected, the water aside face of levees are protected by stone or concrete revetment. Regular maintenance and contingency arrangements to fight floods are absolutely necessary to keep the levees functional.
Masonry structures used to confine the river in a manner similar to levees are known as flood nails. These are used to protect important Structures against floods, especially where the land is at a premium.
Floodways are natural channels into which a part of the flood will be diverted during high stages. A flood way can be a natural or man-made channel and its location is controlled essentially by the topography. Generally, wherever they are feasible, floodways offer an economical alternative to other structural flood-control measures. To reduce the level of the river Jhelum at Srinagari a supplementary channel has been constructed to act as a floodways with a capacity of 300 m3/s. This channel is located 5 km upstream of Srinagar city and has its outfall in lake Wullar.
The works under this category involve:
- Widening or deepening of the channel to increase the cross-sectional area;
- Reduction of the channel roughness, by clearing of vegetation from the channel perimeter;
- Short-circuiting of meander loops by cutoff channels, leading to increased slopes.
All these three methods are essentially short-term measures and require continued maintenance.
Soil-conservation measures in the catchment when properly planned and effected lead to an all-round improvement in the catchment characteristics affecting abstractions. Increased infiltration, greater evapotranspiration and reduced soil erosion are some of its easily identifiable results. It is believed that while small and medium floods are reduced by soil-conservation measures, the magnitude of extreme floods is unlikely to be affected by these measures.
Forecasting of floods in advance enables a warning to be given to the people likely to be affected and further enables civil-defense measures to be organized. It thus forms a very important and relatively inexpensive nonstructural flood-control measure. However, it must be realised that a flood warning is meaningful if it is given sufficiently in advance. Also, erroneous warnings will cause the populace to loose faith in the system. Thus the dual requirements of reliability and advance notice are the essential ingredients of a flood-forecasting system. The flood forecasting techniques can be broadly divided into three categories:
- Short-range forecasts,
- Medium-range forecasts
- Long-range forecasts.
In this the river stages at successive stations on a river are correlated with hydrological parameters, such as precipitation area,antecedent precipitation index, and variation of the stage at the upstream base point during the travel time of a flood. This gives an advance warning of 12-40 h for floods. The flood forecasting currently being used for the metropolitan city of Delhi is based on this technique.
In this rainfall-runoff relationships are used to predict flood levels with a warning of 2-5 days. Coaxial graphical correlations of runoff with rainfall and other parameters like the time of the year, storm duration and antecedent wetness have been developed to a high stage of refinement by the US Weather Bureau.
Using radars and meteorological satellite data, advance information about critical storm-producing weather systems, their rain potential and time of occurrence of the event are predicted well in advance.
FLOOD CONTROL IN INDIA
In India the Himalayan rivers account for nearly 60% of the flood damage in the country. Floods in these rivers occur during monsoon months and usually in the months of August or September. The damages caused by foods are very difficult to estimate and a figure of Rs. 10 billion as the annual flood damage in the country gives the right order of magnitude. It is estimated that annually on an average 40 Mha of land flooding and of this about 12 Mha have some kind of flood control measure. There are about 12,500 km of levees and about 25600 km drainage chennels affording protection from floods. About 60 to 80% of flood damages occur in the states of UP, Bihar, West Bengal, Assam and Orissa.
Flood forecasting is handled by CWC in close collaboration with the IMD, which lends meteorological data support. Nine flood Met offices both the aid of recording raingauges provide daily synoptic situations, actual rainfall amounts and rainfall forecasts to CWC. The CWC has 141 flood-forecasting stations situated on various basins to provide a forecasting service to a population of nearly 40 million.
A national programme for flood control was launched in 1954 and an amount of about 976 crores has been spent since then till the beginning of the Sixth Five-Year Plan. The Planning Commission has provided an outlay of 1045 crores in the sixth Five-Year Plan for flood control. These figures highlight the seriousness of the flood problem and the efforts made towards mitigating flood damages. The experience gained in the flood control measures in the country are embodied in the report of the Rashtriya Barh Ayog (RBA) (National Flood Commission) submitted in March 1980. This report, containing a large number of recommendations on all aspects of flood, control forms the basis for the evolution of the present national policy on floods.
In the previous chapters various aspects of surface hydrology that deal with surface runoff have been discussed. Study of subsurface flow is equally important since about 22% of the world's fresh water resources exist in the form of groundwater. Further, the subsurface water forms a critical input for the sustenance of life and vegetation in and zones. Because of its importance as a significant source of water supply, various aspects of groundwater dealing with the exploration, development and utilization have been extensively studied by workers from different disciplines, such as geology, geophysics, geochemistry, agricultural engineering, fluid mechanics and civil engineering and excellent treatises are available.
This zone, also known as groundwater zone is the space in which all the pores of the soil are filled with water. The water table forms its upper limit and marks a free surface, i.e. a surface having atmospheric pressure.
Zone of Aeration
In this zone the soil pores are only partially saturated with water. The space between the land surface and the water table marks the extent of this zone. Further, the zone of aeration has three subzones:
Soil Water Zone
This lies close to the ground surface in the major root band of the vegetation from which the water is lost to the atmosphere by evapotranspiration.
In this the water is held by capillary action. This zone extends from the water table upwards to the limit of the capillary rise.
This lies between the soil water zone and the capillary fringe. The thickness of the zone of aeration and its constituent subzones depend upon the soil texture and moisture content and vary from region to region. The soil moisture in the zone of aeration is of importance in agricultural practice and irrigation engineering. The present chapter is concerned only with the saturated zone.
All earth materials, from soils to rocks have pore spaces. Although these pores are completely saturated with water below the water table, from the groundwater utilization aspect only such material through which water moves easily and hence can be extracted with ease are significant. On this basis the saturated formations are classified into four categories: Aquifer, Aquitard, Aquiclude >Aquifuge
An aquifer is a saturated formation of earth material which not only stores water but yields it in sufficient quantity. Thus an aquifer transmits water relatively easily due to its high permeability. Unconsolidated deposits of sand and gravel form good aquifers.
It is a formation through which only seepage is possible and thus the yield is insignificant compared to an aquifer. It is partly permeable.
It is a geological formation which is essentially impermeable to the flow of water. It may be considered as closed to water movement even though it may contain large amounts of water due to its high porosity. Clay is an example of an aquiclude.
It is a geological formation which is neither porous nor permeable. There are no interconnected openings and hence it cannot transmit water. Massive compact rock without any fractures is an aquifuge.
The definitions of aquifer, aquitard and aquiclude as above are relative. A formation which may be considered as an aquifer at a place where water is at a premium (e.g. and zones) may be classified as an aquitard or even aquiclude in an area where plenty of water is available.
The availability of groundwater from an aquifer at a place depends upon the rates of withdrawal and replenishment (recharge). Aquifers play the roles of both a transmission conduct and a storage. Aquifers are classified as unconfined aquifers and confined aquifers on the basis of their occurrence and field situation. An unconfined aquifer (also known as water table aquifer) is one in which a free surface, i.e. a water table exists (Fig. 9.2). Only the saturated zone of this aquifer is of importance in groundwater studies.Recharge of this aquifer takes place through infiltration of precipitation from the ground surface. A well driven into an unconfined aquifer will indicate a static water level corresponding to the water table level at that location.
A confined aquifer, also known as artesian aquifer, is an aquifer which is confined between two impervious beds such as aquicludes or aquifuges (Fig. 9.2). Recharge of this aquifer takes place only in the area where it is exposed at the ground surface. The water in the confined aquifer will be under pressure and hence the piezometric level will be much higher than the top level of the aquifer.
At some locations: the piezometric level can attain a level higher than the land surface and a well driven into the aquifer at such a location will flow freely without the aid of any pump. In fact, the term "artesian" is derived from the fact that a large number of such free-flow wells were found in Artois, a former province in north France. Instances of free-flowing wells having as much as a 50-m head at the ground surface are reported.
A water table is the free water surface in an unconfined aquifer. The static level of a well penetrating an unconfined aquifer indicates the level of the water table at that point. The water table is constantly in motion adjusting its surface to achieve a balance between the recharge and outflow from the subsurface storage. Fluctuations in the water level in a dug well during various seasons of the year, lowering of the groundwater table in a region due to heavy pumping of the wells and the rise in the water table of an irrigated area with poor drainage, are some common examples of the fluctuation of the water table. In a general sense, the water table follows the topographic features of the surface. If the water table intersects the land surface the groundwater comes out to the surface in the form of springs or seepage. Sometimes a lens or localised patch of impervious stratum can occur inside an unconfined aquifer in such a way that it retains a water table above the general water table (Fig. 9.3). Such a water table retained around the impervious material is known as perched water table. Usually the perched water table is of limited extent and the yield from such a situation is very small. In groundwater exploration a perched water table is quite often confused with a general water table.
The important properties of an aquifer are its capacity to release the water held in its pores and its ability to transmit the flow easily. These properties essentially depend upon the composition of the aquifer.
The amount of pore space per unit volume of the aquifer material is called porosity. It is expressed as
where n = porosity, Vv = volume of voids and Vo = volume of the porous medium, In an unconsolidated material the size distribution, packing and shape of particles determine the porosity. In hard rocks the porosity is dependent on the extent, spacing and the pattern of fracturing or on the nature of solution channels. In qualitative terms porosity greater than 20% is considered as large, between 5 to 20% as medium and less than 5% as small.
While porosity gives a measure of the water-storage capability of a formation, not all the water held in the pores is available for extraction by pumping or draining by gravity. The pores hold back some water by molecular attraction and surface tension. The actual volume of water that can be extracted by the force of gravity from a unit volume of aquifer material is known as the specific yield, Sy. The fraction of water held back in the aquifer is known as specific retention, Sr. Thus porosity
n = Sy + Sr (9.2)
The representative values of porosity and specific yield of some common earth materials are given in Table 9.1.
TABLE 9.1 POROSITY AND SPECIFIC YIELD OF SELECTED FORMATIONS
It is seen from Table 9.1 that although both clay and sand have high porosity the specific yield of clay is very small compared to that of sand.
GEOLOGICAL FORMATIONS AS AQUIFERS
The identification of a geologic formation as a Potential aquifer for ground Water-development is a specialized job requiring the services of a trained hydrogeologist. The geologic formations of importance for possible use as an aquifer can be broadly classified as unconsolidated deposits and consolidated rocks. Unconsolidated deposits of sand and gravel form the most important aquifers. They occur as fluvial alluvial deposits, abandoned channel sediments, coastal alluvium and as lake and glacial deposits. The yield is generally good and may be of the order of 50-100 m³/hr. In India, the Gangetic alluvium and the coastal alluvium in the states of Tamil Nadu and Andhra Pradesh are examples of good aquifers of this kind. Among consolidated rocks, rocks with primary porosity such as sandstones are generally good aquifers. The state of weathering of rocks and occurrence-of secondary openings such as joints and fractures enhance the yield. Normally, the yield from these aquifers is 1ess than that of alluvial deposits and typically may have a value of 20-50 m³/h. Sandstones of Kathiawar and Kutch areas of Gujarat and of Lathi region of Rajasthan are some examples.
Limestones contain numerous secondary openings in the form of cavities formed by the solution action of flowing subsurface water. Often these form highly productive aquifers. In Jodhpur district of Rajasthan, cavernous limestones of the Vindhyan system are providing very valuable ground-water for use in this arid zone.
The volcanic rock basalt has permeable zones in the form of vesicles, joints and fractures. Basaltic aquifers are reported to occur in confined as well as underunconfined conditions. In the Satpura range some aquifers of this kind give yields of about 20 m³/h.
Igneous and metamorphic rocks with considerable weathered and fractured horizons offer good potentialities as aquifers. Since weathered and fractured horizons are restricted in their thickness these aquifers have limited thickness. Also, the average permeability of these rocks decreases with depth. The yield is fairly low, being of the order of 5-10 m³/h. Aquifers of this kind are found in the hard rock areas of Karnataka, Tamil Nada, Andhra Pradesh and Bihar.
Wells form the most important mode of groundwater extraction from an aquifer. While wells are used in a number of different applications, they find extensive use in water supply and irrigation engineering practice. Consider the water in an unconfined aquifer being pumped at a constant rate from a well. Prior to the pumping, the water level in the well indicates the static water table. A lowering of this water level takes place on pumping. If the aquifer is homogeneous and isotropic and the water table horizontal initially, due to the radial flow into the well through the aquifer the water table assumes a conical shape called cone of depression. The drop in the water table elevation at any point from its previous static level is called drawdown. The areal extent of the cone of depression is called area of influence and its radial extent radius of influence (Fig. 9.5). At constant rate of pumping, the drawdown curve develops gradually with time due to the withdrawal of water from storage. This phase is called unsteady flow as the water-table elevation at a given location near the well changes with time. On prolonged pumping, an equilibrium state is reached between the rate of pumping and the rate of inflow of groundwater from the outer edges of the zone of influence. The drawdown surface attains a constant position with respect to time when the well is known to operate under steady-flow conditions. As soon as the pumping is stopped, the depleted storage in the cone of depression is made good by groundwater inflow into the zone of influence. There is a gradual accumulation of storage till the original (static) level is reached. This stage is called recuperation or recovery and is an unsteady phenomenon. Recuperation time depends upon the aquifer characteristics.
Changes similar to the above take place due to a pumping well in a confined aquifer also, but with the difference that, it is the piezometric surface instead of the water table that undergoes drawdown with the development of the cone of depression. In confined aquifers with considerable piezometric head, the recovery into the well takes place at a very rapid rate.
The quantum of groundwater available in a basin is dependent on the inflows and discharges at various points. The interrelationship between inflows, outflows and accumulation is expressed by the water budget equation
å I D t - å Q D t = D S (9.39)
- å I D t = all forms ofrecharge and includes contribution by lakes, streams, canals, precipitation and artificial recharge, if any, in the basin
- å Q D t = net discharge of groundwater from the basin and includes pumping, surface outflows, seepage into lakes and rivers and evapotranspiration
D S = change in the groundwater storage in the basin over a time D t
Considering a sufficiently long time interval, At of the order of a year, the capability of the groundwater storage to yield the desired demand and its consequences on the basis can be estimated. It is obvious that too large a withdrawal than what can be replenished naturally leads ultimately to the permanent lowering of the ground water table. This in turn leads to problems such as drying up of open wells and surface storages like swamps and ponds, change in the characteristics of vegetation that can be supported by the basin. Similarly, too much of recharge and scanty withdrawal or drainage leads to waterloggii3g and consequent decrease in the productivity of lands.
The maximum rate at which the withdrawal of groundwater in a basin can be carried without producing undesirable results is termed safe yield. This is a general term whose implication depends on the desired objective. The "undesirable" results include permanent lowering of the groundwater table or piezometric head, maximum drawdown exceeding a preset limit leading to inefficient operation of wells and salt-water encroachment in a coastal aquifer. Depending upon what undesirable effect is to be avoided, a safe yield for a basin can be identified.