CHAPTER 3 PRECIPITATION Gok aglamayinca yer gilmez. (The earth cannot smile unless the heavens cry.) Turkish proverb The term precipitation as used in hydrology includes all forms of water coming from the atmosphere to the Earth. The principal forms are rain, snow, hail, drizzle and sleet. Detailed explanation of these forms and some others are given in Table 3.1. Issues related to the formation of precipitation in the atmosphere and to the falling of precipitation on the ground are addressed by meteorologists rather than hydrologists. However, precipitation becomes one of the major elements of hydrology after it falls on the ground, since it is the source of the water on Earth. This chapter deals first with the formation and measurement of precipitation, and then with the analysis of collected precipitation data. 3.1 FORMATION OF PRECIPITATION Condensation of atmospheric vapor results, in general, in the formation of clouds, but not all clouds produce precipitation. Air should be saturated in order for precipitation to occur, but even 100% saturation cannot produce precipitation if the air is pure. For the formation and condensation of precipitation, four processes should take place one after the other. These are cooling, condensation, the growth of droplets, and the afflux of moist air to the precipitation area 3.1.1 The Process of Cooling For air or clouds to reach saturation, they have to cool by some means. The cooling of air may be by dynamic (or adiabatic) means, by the mixing of two air masses of different temperatures, or by the contact of air with the cooler surface of earth or water. Types of precipitation are also determined by the types of cooling, three major ones being convective storms, orographic storms and cyclonic storms. When unsaturated air at or near the Earth's surface is carried to higher levels, either through convection or through other means, expansion will occur due to42 Engineering Hydrology tion (Chorley, 1969) Table 3.1 The major categories of precipi Type Characteristics | Typical amount Dew Deposited on surface, particularly a vegetation | 0.1 ~ 1.0 canopy (frozen form-hoar frost) min/night | Fog-drip Deposited on vegetation and other obstacles | Up to 4 mnv/hr | from fog (frozen form-rime) Dnizvle Droplets < 0.5 mm in diameter (freezing 0.2 -0.5 mm/hr drizzle when surface temperature is below ore) Rain Drops > 0.5 mm diameter, Light<2 mm/hr typically 1-2 mm diameter Heayy>7mnvhr | Steet Partly melted snow or a rain and snow mixture (Great Britain) | Snowflakes Aggregations of ice crystals up to several em, | across: |" Snow grains Very small, flat opaque grains of ice; the solid (granular snow) _| equivalent of drizzle ‘Snow pellets | Opaque pellets of ice 2-5 mm diameter, falling | (graupel. or soft | in showers hail) | Ice pellet (small ] Clea hail) Ice pellets (sleet_| Frozen rain or drizzle drops inthe U.S.) Hail Roughly spherical Jumps of ice, 5-50 mm or more in diameter, showing a layered structure of opaque and clear ice in cross-section ice encasing a snowflake or snow pellet the reduction of pressure with altitude. During this process no heat is added to the air from outside sources and none is deducted from it, However, its temperature is lowered because of the heat energy that is transferred into work in the process of expansion. This is called dynamic or adiabatic cooling and convective storms occur as a result. It is the principal cause of condensation, and is directly responsible for most rainfall If moist air is forced, by a topographic barrier, to rise to higher elevations, expansion, cooling, and precipitation will follow. The heaviest precipitation occurs where conditions are favorable for the production of this type of rainfall, which is referred to as orographie storms When the air moves from a high pressure to a low-pressure area the colder air lifts the warmer air and condensation results. The boundary between air massesPrecipitation 43 with different pressures is called a front. If cold air is moving under warm air it is called cold front, if warm air is moving over cold air it is called a warm front. In cyclonic storms frontal precipitation occurs. Cold fronts move faster and produce high intensity rainfalls. 3.1.2. The Process of Condensation As the air cools, water vapor condenses and changes into liquid form Hygroscopic particles, called condensation (freezing or sublimation) nuclei have to be present in the air for condensation to occur. In the absence of these particles, condensation will not take place even in 100% saturation. The hygroscopic particles that exist in the atmosphere are mostly sodium chloride and sulphur three oxide, which are less than 10 1 in size 3.1.3 The Process of the Growth of Droplets Condensation does not necessarily cause precipitation. It forms fog or clouds, which consist of small droplets of water having an average diameter of 40 Raindrops, on the other hand, have diameters varying from 500 to 4000 jl. It follows that some process, which will increase the size of drops, is necessary before precipitation can occur There are two means for the increase of the size of droplets. For one of them, there should be ice and water droplets as a mixture in a cloud. For subfreezing temperatures the saturation vapor pressure is lower on the ice surface than on the water surface. The air in the cloud will have a vapor pressure somewhere between these two saturation pressures. As a result, water droplets will evaporate and at the same time condensation will occur on the ice particles Large droplets will thus be built up and begin to fall. This is called the ice crystals effect. While falling, larger particles will collide and combine with smatler droplets on their path to further increase their size. This is called the collusion effect. 3.1.4 The afflux of Moist Air to the Pre pitation Area It is found from the observations that the amount of water in clouds is very low A cloud does not contain more than approximately 2-3 gr of liquid water per m’ of volume. For this reason, in order to have long-duration precipitation, in addition to the occurrence of the three processes mentioned above, new water- bearing clouds have to come to the precipitation area. If these four processes take place one after the other, then a considerably long-duration precipitation can occur44 & gineering Hydrology 3.15 Artificial Rainfall When the four processes necessary for the production of precipitation are examined, it can be seen that man can interfere only in the condensation step to obtain artificial rainfall. This process is also called cloud seeding, cloud modification, weather control or weather modification. The process of condensation can be accelerated artificially by injecting condensation or sublimation nuclei into the cloud. This injection can be made from an aircraft or from the ground in the form of smoke. If the temperature of the cloud is about -15 °C, silver iodide is fully effective in producing ice crystals, It is more efficient in mountainous regions. Seeding can generally increase the precipitation from a cloud under favorable conditions. But it is difficult to judge the effectiveness of seeding at a certain location of a large area, since it requires a thorough study to determine whether there is an actual increase in the rainfall or there is only a redistribution of the rainfall in the whole area. Another important point about artificial precipitation is the difficulty of its control. 3.1.6 The Variation of Precipitation Precipitation varies in area and in time according to the general pattern of atmospheric circulation, and also according to local factors such as topography and wind direction. Higher precipitation occurs near the coasts than inland, because the seas supply the bulk of the atmospheric moisture for precipitation. In addition, the precipitation amount is generally more on the windward sides of mountains than on the lee sides. Large seasonal variations in precipitation occur in regions over which the amount of moisture inflow is varied by annual oscillations in the atmospheric circulation, Seasonal precipitation change in Turkey is given in Figure 3.1, where it is seen that summers are dry almost where except Black Sea Coast and north of Eastern Anatolia, Highest pitation occurs generally in winter in Turkey. The detail of the distribution monthly precipitation in some cities evel pre is given in Figure 3.2 as tota Precipitation amount or rate changes during a storm also, depending on the type and duration of the storm and on the time of the year. The average annual precipitation on the land surfaces of the Earth is about 800 mm, but a large variability exists, from very low in Anica (Chile) with an annual average of 0.5 mm, to very high in Mt, Waialeale (Hawaii), which receives 11680 mm per year on the average. The average annual precipitation depth in Turkey, around 643 mm, is lower than the average value for the entireAoyun |, ut (ww ul yidap) jJeyures jeuoseas Jo uoNNgUIsIG 4¢ a4nBI4 ca] vas NVaNvaYaliaaW vas Nva93av es vas yOvId46 Engineering Hydrology Aayin | ul (ww ul Yydap) |JeJUIes AJYJUOW yo UONNqUIsIQ z’¢ auNBI4 WAS NVANVYYSLIGSW was Ovid vas Nv39avPrecipitation 47 Earth. The lowest annual precipitation is observed in Himmetdede, Kayse 63.3 mm in 1933 and the maximum amount in Rize as 4043.3 mm in 1931 (DSI, 1995). Figure 3.3 shows the distribution of total annual precipitation in Turkey 3.2: MEASUREMENT OF PRECIPITATION Different types of instruments are used in measuring the amount of precipitation that reaches the Earth's surface. Some instruments measure the total depth at a point in the area (point observation value) within a certain period of time, while other instruments measure the variation of precipitation depth at a point over the same period. Measurement is possible only for precipitation depths exceeding 0.1 mm. Precipitation below this threshold is referred to as a trace. There are two main difficulties in precipitation measurement. a) The rain gauge should collect the same amount of precipitation that would have reached the ground if the gauge had not been present. The locations and the number of gauges should be selected so that the total Precipitation over the area can be estimated as accurately as possible. b) A precipitation gauge, a tin can or receptacle placed on the ground, may undergo a number of local disturbances such as splash in or out, evaporation losses, losses in wetting of the gauge surfaces, inaccuracies due to improper leveling of the gauge orifice, and the effect of wind producing turbulent eddies about the orifice. For better results, these disturbances should be prevented as much as possible. Measuring instruments may be classified as recording and non-recording ones for daily measurements. There are also stor weather © gauges for longer periods and ars, which are used in order to help the measurements of precipitation. 3.2.1 Non-recording Rain Gauges Although any cup that can collect water may be used as a rain gauge, standard shapes are used to ensure records, which are comparable with each other. One such gauge. used in the U.S.A. and some other countries, is the US Weather Bureau rain gauge (Figure 3.4a), which consists of a can that has a diameter of 20.32 em (8 in) and a depth of 60.96 em (24 in). Fitted over the top there is a copper receiver, whose top rim is a knife edge and whose bottom consists of a funnel that carries the water into the brass measuring tube. A measuring stick, 60.96 cm (24 in) long by 9.53 mm (3/8 in) wide and about48. Engincenng Hydrology (S661 “12 38 HALquiyy) Aeyun | ul (Ydap Ww! Jejuled fenuue ueaw Jo uoINqUIsig ¢’¢ aunBi4 VaS NVANVYYSLIGSW re cupy * o < em A oe e svete teroweveney vas Nv393vPrecipitation 49 3.18 mm (1/8 in) thick, completes the equipment. The cross sectional area of the measuring tube minus the cross sectional area of the measuring stick is exactly one tenth of the area of the opening of the receiver. In other words, the measured depth of water in the tube is ten times the actual depth of rainfall, ear 16cm ao { tunnet aS ee <> Receiver ain Fn iy 3 . Funnel 3 e leasuring ayo es Tube aly & H Outflow 7 3 can 2 7 § Vessel Measuring tube Outflow a) U.S. Weather Bureau type b) Hellmann type Figure 3.4 Non-recording rain gauges When snow is expected, the receiver and measuring tube are removed and snow is collected in the overflow can. Then, the snow is melted and poured into the measuring tube for measurement. In Turkey, precipitation is mostly measured by a Hellman type standard rain gauge (Figure 3.4b), which ha sa diameter of 16 cm (an area of 200 cm’) and height of 44 cm. Fitted over the top there is a copper or brass receiver whose top rim is a knife edge. Below the rim there is a collecting vessel with a capacity of 70 mm, equivalent to 1.4 liters, The lower part supports the other Parts of the gauge and stores extra water. When the gauge is in operation, the height of the rim is 100 em from the ground surface (150 em when snow depth Is high in the region). A glass measuring tube completes the equipment. When snowfall occurs, it is collected in the receiver and then it is melted and poured into the measuring tube for measurement. 3. Recording Rain Gauges It iy often necessary to know, not only the total amount of precipitation, but also. its intensity. Recording gauges make a continuous record of the precipitation and its times of occurrence, and permit calculation of precipitation intensities.50. Engineering Hydrology Although there are a number of different types of recording rain gauges, only three of them have gained widespread use, which are the tipping bucket, weighing, and float types. Among these, the weighing type is used most often in Turkey [Link] Weighing Gaug: In weighing type instruments (Figure 3.5a), a collecting bucket rests on a weighing scale, which actuates a pen that draws a graph in the form of a mass diagram of the rainfall on a clock-driven chart. A receiver directs the rainfall to the collecting bucket. In the recorded diagram (Figure 3.5b), the abscissa represents the time, and the ordinate represents the weight, which can be converted to the depth of rainfall. The slope of this graph at any time with respect to the horizontal axis gives the intensity of the rainfall at that time. It is, therefore, an easy matter to determine the average intensity of rainfall for any given period from the recorded diagram. cover sheet weighing once if am base Figure 3.5a Recording rain gauge (Weighing type) id (mm) 0 2 4 6 B 10 12 14 16 18 20 22 24 1 (hr, min) Figure 3.5b Recorded diagramPrecipitation 51 Generally a daily chart is used for the diagrams, but for locations where access is not easy, weekly charts can also be used by adjusting the clock so that the revolving drum completes one revolution in seven days instead of one day Weekly charts cannot contain the same detail as daily charts, since the same size paper is used for seven days. [Link] Tipping Bucket Gauges This type of instrument has two buckets attached to each other, as shown in Figure 3.6. When the left-hand bucket is filled up to a specified level, it drops down and right-hand bucket comes beneath the funnel. At the same time an electrical circuit is actuated causing a pen to make a notch on a recorder chart wrapped around a clock-driven cylinder. i catch funnel 2 KI Ly S buckets 8 A y iV tank Bucket A fills Bucket B fills (@) Bucket 8 empties (®) Bucket A empties Figure 3.6 Tipping bucket type recording gauge The electrical pulse indications permit also remote recording, which may be important at certain situations. However during snowfall a supplementary device is necessary to melt snow, and in intense rains, water loss occurs while the bucket is tipping. On the other hand, these instruments are durable and reliable, and their mechanism is not complicated. [Link] Float-type Gauges hese instruments are essentially water level gauges. A float in the gauge receptacle tises as rain water enters the chamber, the rise of the float being recorded by a pen on a suitably calibrated, clock-driven chart. Some float gauges can only record to the capacity of the chamber, but others are equipped with siphoning arrangements to let out the water when the chamber fills and to return the pen to the zero position on the chart (Figure 3.7)52. Engincenng Hydrology 7 pen. recording chart from the pail float syphon | | reservoir Figure 3.7 Float type recording gauge The most important drawback of such gauges lies in their operation at temperatures below the freezing point, although some attempts have been made to overcome the problem of freezing. 3.2.3 Storage or Totalizer Gauges Storage gauges are used in mountainous regions or remote areas, which are not accessible during winter or where observers are not available, They are, in general, large tanks collecting precipitation during a whole season, and the collected amount is measured at the end of the season. They are usually equipped with windshields, and antifreeze is added to melt the snow. The solution is covered by some low-viscosity oil, which prevents evaporation but allows the passage of precipitation 3.2.4 Weather Radar Techniques have been developed for using the radar as an aid in the measurement of rainfall. Although some quantitative determinations can be made by means of a radar, its principal value is in the determination of the areal extent, orientation, and movement of rainstorms in time, Radars operate by emitting a regular succession of pulses of electromagnetic radiation in a narrow beam from the antenna of the system. Between pulses, the radar antenna becomes a receiver of the energy of the emitted pulses scattered by various targets. These returned signals are transformed into a visual display on the radar scope, which is called an echo, In general the brightness of the echo is an indication for the intensity of the precipitation. ‘This method is usually used to supplement data obtained from a network of rain gauges on the ground su More information about this subject can be found in Wilson and Brandes (1979).Precipitation 53 Remote Sensing techniques in combination with radars are also used for the estimation of precipitation using satellite images. Images are used for the determination of clouds and their densities as an indicator for rain and its intensity after the calibration with ground observation values (Akytirek, 1998). 3.2.5 Snow Measurement Snow measurements are obtained through the use of standard non-recording and recording rain gauges, seasonal storage precipitation gauges, snowboards and graduated sticks or permanent snow stakes. Rain gauges are usually equipped with shields to reduce the effects of wind on snow (Tirksoy, 1981). Snow measuring sites should be chosen carefully so that the effects of wind and meltwater drainage will be minimum. In an area snowpack depth and density change from one point to the next due to changing snowmelt amounts and drifting caused by winds. Therefore measurements are made not at a single point but at a number of points along a line, which is called snow course (U.S. SCS, 1972) Snow is usually composed of ice crystals and liquid water. Water content of the snow describes the amount of liquid water and it changes with snow accumulation. New snow may have as low as 5% water, but old snow as high as 80% water. The quality of snow is defined as the percentage of ice in snow by weight Depth measurements are made with a rain-gauge measuring stick if the accumulation is not large, and with a permanent snow stake otherwise. Snow stakes are simple rulers inserted in the snowpack at a number of locations near the observing station, to get a representative average depth. In areas unattended by regular observers, such as mountain slopes, snow markers are often used. Such markers are variations of simple poles fixed in the ground with heights marked on them, In hydrolog - Water equivalent of the snowpack, which is defined as the depth of equivalent melt water, is more important than the depth of snow. It is a function of not only the depth but also the density of the snow. Snow density, on the other hand, is the ratio of the volume of water obtained by melting the snow to the original volume of snow sample, Snow density may take any value between 0.004 for fresh snow to 0.91 for compacted snow as ice. Therefore it is Very important to measure all the characteristics of snow (depth, water equivalent and density) to have an idea about how much melt will occur. These characteristics are measured by different methods such as by using snew tubes or pressure pillows (Tiirksoy, 1981)54_ Engineering Hydrology In Turkey DSi and FiEI are making snow observations at 151 and 48 locations respectively (DSI, 1995, EIEI, 1995). In addition to on site measurements of snow, remote sensing techniques using photographs taken from aircrafts or satellite images, can also be used to determine the snow covered area, snow depth and also density. These are especially important for remote areas where on site measurements are difficult, such as the case in Eastern Anatolia (Akyiirek et al., 1999, Sorman and Kaya, 1998, Sorman and Uzunoglu, 1998). 3.2.6 Network of Precipitation Stations Designing a network for precipitation gauging stations in an area is a difficult task, which includes the study of spatial variation of precipitation in the area, the purpose of the data collection and most of all depends on the available funds (Lenton and Rodrigez-Iturbe, 1977). If the problem is to determine annual average precipitation over a large area then there is no need to put a very dense network of gauges in the area (Eagleson, 1967). On the other hand, a very dense network is requited to determine the rainfall pattern in individual thunderstorms, or in basins where topographic variability is large. As it would be expected, as the density of the network increases, the sampling errors in the measurements will decrease. It is the same for the duration of the storm and the size of the area, The average error decreases as the number of stations increases, which is shown in Figure 3.8. Average’ Error (mm) ‘Areal Mean Rainfall (mm) Figure 3.8 Change of error with number of stations: ‘The effect of network density is illustrated by an example in Figure 3.9, where the same basin (Goksu Basin) is depicted with different numbers of stations. As would be expected, the mean monthly precipitation differs considerably between the two cases.Precipitation 55 b) areal mean precipitation = 130.86 mm (with 21 stations) ure 3.9 Change in calculated areal mean precipitation with the number of stations January values of Goksu Basin)56 Engineering Hydrology The following minimum densities of precipitation network have been recommended for general hydrometeorological purposes (WMO, 1974) a) For flat regions of temperate, Mediterranean, and tropical zones, 600 to 900 kin’ per station. b) For mountainous regions of temperate, Mediterranean, and tropical zones, 100 to 200 km’ per station. a For small mountainous islands with irregular precipitation, 25 km" per station, d) Forarid and polar zones, 1500 to 10 000 km” per station. In Turkey precipitation measurements are made mainly by State Meteorological Organization (DMI, 1995), then by State Hydraulic Works (DSi, 1995), and by General Directorate of Rural Services in very small upstream basins. 3.3 ANALYSIS OF PRECIPITATION DATA Precipitation data consists of the following three categories: 1. Daily total precipitation depths, obtained from non-recording gauges. 2. Recorded diagrams of cumulative precipitation with respect to time, obtained from recording gauges. Seasonal total depths or recorded diagrams obtained from storage gauges. These data are used to determine the time and the areal variation as well as the frequency of precipitation. However, as an initial step, the data should be checked for missing parts and errors, which should then be used for estimation or analysis. 3.3.1 Estimation of Missing Precipitation Data Many precipitation stations have short breaks in their records because of lapses in the observation, instrumental failures, or human errors, It is often necessary to estimate the missing records by using the available records of those surrounding stations, which are as close and as evenly spaced as possible Three methods may be utilized to estimate the missing records of a station [Link] The Arithmetic Mean Method In this method, three stations surrounding the station with missing data are taken as index stations, and the simple arithmetic average of their precipitation values is used to fill in the missing precipitation value (Paulhus and Kohler, 1952). This method is used only if the normal annual precipitation at each of the index stations is within 10% of that of the station with the missing record.Precipitation 57 Let X be the station with missing data, and A, B, and C be the index stations, then when condition in Eqn. [Link] is satisfied, the estimated value will be calculated by the formula given in Eqn. 3.1b. 0.9 t (hr) ht & , Figure 3.13 Hyetograph 3.3.4 Mass Curve of Rainfall ded diagram obtained from the resenting the cumulative depth As it is seen in the section above, the rei recording gauge is a mass curve for rainfall, of rain with respect to time, and yielding as its first time derivative at a point, the intensity depicted in the hyetograph. On the other hand, in non-recording gauges, the only piece of information obtained is the daily total rainfall value,Precipitation 61 which, as mentioned earlier, is not suitable for hydrograph analysis. If possible, the time distribution of this total value should be determined. When there is a recording gauge in the neighborhood of the non-recording gauge, with the condition that the areas have similar hydrologic characteristics, then the recorded diagram of this gauge can be used to find the time distribution of the total value in the non-recording gauge. Assuming that the temporal change in rainfall is the same in both gauges, the ordinates of the mass curve of the recorded diagram are multiplied by the ratio of total daily rainfalls (Py/P,) to obtain the time distribution of precipitation in the non-recording gauge, as shown in Figure 3.14. recorded mass curve d(mm) for Station A derived mass curve for Station B t (br) Figure 3.14 Mass curve of rainfall As mentioned above, time distribution of rainfall is important and this distribution changes from one location to another. Therefore, time distribution of rainfall may also be given in dimensionless mass curve form for different locations in percents with respect to time as seen in Figure 11.3 of Chapter 11 as observed in Antalya station for 24 hour duration, or totally dimensionless as seen in Figure 3.15, where P is the total rainfall depth and D is the total duration of rainfall 1 Eap/P 345678 910 1/0 Figure 3.15 Dimensionless rainfall mass curve cop bh DDO Oe 3.3.5 Areal Mean Precipitation The data gathered at precipitation gauges give only point values of precipitation, whereas areal mean values are necessary in most hydrologic62. Engineering Hydrology studies. The three methods that are commonly used for computing the mean rainfall on an area are the Arithmetic mean, Thiessen polygons and Isohyetal map methods [Link] The Arithmetic Mean Method In this method, as the name implies, the result is obtained by dividing the sum of the precipitation depths recorded at all the stations in the basin by the number of stations. y ZF, (3.5) n Where P,,. is the areal mean precipitation, P, is the rainfall observed at the it* station inside the basin, and n is the number of inside stations. If the stations are uniformly distributed over the basin and the rainfall varies in a regular manner, the results obtained by this method will not differ too much from those obtained by either of the other methods. [Link]. The Thiessen Polygons Method In the application of this method, adjacent stations are joined by straight lines, thus dividing the entire area into a series of triangles. While doing this, it is better to try to obtain triangles as close to equilateral triangles as possible. Then, bisectors (perpendicular lines at midpoints) are drawn to the sides of these triangles forming polygons around each station. It is assumed that the polygon of any station shows the influence area of that station. A better approach is to plot the perpendiculars at the midaltitudes between the stations in order to include the orographic effects, but the difference in the results is small, To find the mean precipitation value of a certain region, the polygon areas falling in the region of interest are multiplied by the rainfall depths of the corresponding stations and the sum of these products is divided by the total area. > Py we ZPa 6) Za, Where P, is the rainfall observed at the i" station and a, is the in-region portion of the area of the polygon surrounding this station, In applying this method all the stations in and out of the area are taken into consideration. To obtain the polygons easily, sometimes it may be necessary to make a few trials while joining the stations to form triangles.Precipitation 63 [Link] The Isohyetal Map Method In this method the points getting the same amount of precipitation are joined to obtain isohyets (equal precipitation lines). A straightforward procedure is then to assume that the change in precipitation depths between stations is linear However, to obtain better results from this method, good judgment is required both in drawing the isohyets and in assigning the proper mean rainfall values to the areas between them. Consideration of the altitude change between stations would also improve the result instead of linear change in precipitation. In finding the mean precipitation, the areas between isohyets are multiplied by the average precipitations between them and the sum of these products is then divided by the total area. se = (G7) a, Where P, is the average precipitation between isohyets, and a, is the area between isohyets. In this method, both inside and outside stations are considered. The advantage of the last two methods is the use, in determining mean precipitation, of not only inside but also outside stations located near the boundary, The influence of outside stations diminishes as their distance from the boundary increases, On the other hand, in the arithmetic mean method, every inside station has equal weight regardless of its location. An example for finding the areal mean precipitation in a basin by these three methods is given in Figure 3.16a and b. The thiessen polygons and isohyets are shown in Fig 3.16a and calculations in Figure 3.16b. As it is seen in the figure the mean value for the same basin varies from one method to the other. The effect of changes in the number of stations on the areal mean value is illustrated by Figure 3.9 with the isohyetal map method, It can be seen that the detail of the areal distribution of the storm is lost when the number of stations decreases. In the case with 15 stations, the right-hand-side eye of the storm is totally lost Figure 3.17 illustrates, again with isohyetal map method, the change in the areal mean value in response to a change in the orientation of a constant number of stations, These figures indicate that the number and location of stations are very important in calculating areal mean precipitation64 Engineering Hydrology Figure 3.16a Thiessen polygons and isohyetal map of a basinPrecipitation 65 1) Arithmetic Me ’ INET PP +Py +P 45.4+28.2+35.2 + 48.5 4 4 2) Thiessen Polygons Method Station | Area (kr) | Ran (mm | ap A 486 a5 78225 B 1890 45.4 35606 c 635. 914 58039 D 158, 6 5688 427926.05 e 15 48.2 723 — F 7680 282 47376 — G 2001.5 35.2 | 1001328 T 1815 485 8027.5 See J 5155 425 | 2198.75 E 10096 427826.05 3) Isohyetal Map Method ~~ Isohyets | i a Areas between Isohyets(krh) p,(mm) | 5 (ap) 24112.4 | 55 106733_| “65_| 7667.5. 46402.5 [19125 | 5452 [4582.2 4315.6 ) | 2184 2014 pe ESR ~ 100960 pata: _ Dla,.p,) 481794 Ya, 10096 = 47.72mm Figure 3.16b Determination of areal mean by precipitation by different methods66. Engineenng Hydrology 10 Stations 10 Stations 80 mm b) Areal mean precipitation = 168.20 mm Figure 3.17 Change in area jon with orientation of stations: mean precipPrecipitation 67 3.3.6 Resultant Hyetograph In hydrograph analysis, the variation of the areal mean precipitation over time is of as much interest as the mean value its ff. The mean hyetograph for a basin, called the resultant hyetograph, is obtained by superposing the hyetographs in the stations and taking a suitable time interval such as minute, hour, or day. The two methods used for the superposition of hyetographs are the Arithmetic mean and the Thiessen polygons methods. In the arithmetic mean method, only the inside stations are considered, and the average intensity for each time interval is found by taking the arithmetic mean of the intensities observed in the stations for that time interval In the Thiessen polygons method, all the stations inside and outside the basin are considered. For any time interval, the observed intensities in that time interval are multiplied by the polygon areas of the corresponding stations and the average intensity is found by dividing the sum of these products by the total area. An example for obtaining the resultant hyetograph by these two methods is given in Figure 3.18, As seen in the figure only four stations (B, F, G, and 1) are used in arithmetic mean method since only those were in the basin. The others are outside the basin, and they are used in Thiessen polygons method with the inside stations, whatever the size of their influence areas are (i.e. station E has a very small influence area in the basin). 3.3.7. Depth-Area-Duration Curves The distribution of storm precipitation in time is as important as its distribution in the area for some hydrologic studies. To determine the maximum depths of precipitation over different size areas and for different durations, the Depth- Area-Duration (D-A-D) curves are obtained (WMO, 1969). For this purpose, a storm of given duration with a single center is chosen and its isohyetal map is obtained over the area, Then the areal mean precipitation within each isohyet is determined for the corresponding area. The plot of these average depths against the areas gives the Depth-Area curve for that particular storm duration. This procedure is explained in Figure 3.19. To obtain the maximum amounts of precipitations for a certain duration, it is necessary to take as many storms as possible with that duration and plot an enveloping curve to all the points obtained by the analysis explained above. An ample for such an analysis may be found in Figure 3.2068 Engineering Hydrolo; a) Arithmetic mean method 1, =(8.2+6547.249.1)/4 75 mrvhr 7.8+4.6+6.1+7.6)/4 =6.525 mmr (5.2+3.44+4.2+48)/4 4 mmhr b) Thiessen polygons method >t (hn) Station | Area Precipitation (mm) Zap, km? | T 2 7 T z 3 a | a6 [61 [49 [37 | 2906 a3a1a} 17989 B | 1890 | 82 | 78 | 52 | isaosof 147420) 982801 C |} 635 | 124 | 92 | 61 | 78740) sear} 3873.5 D | ise | 84 | 63 | 43 | 13272) 995.4) — 679.4 E is] 53 | ar | as m5] 61s] 72.) F | wso | 65 | 46 | 34 | 109200! 77280] 57120 G f2o1s] 72 | 61 | 42 | 208908) 17699.2) 121863 1 Jasis | 91 | 76 | 48 | 165165) 137940] 8712.0] J _[siss [| 87 | 62 | sa | a4saof 31961] 2629.1 = | 10096 BOSSES] ony ef 454903} 1 7.98 451 Figure 3.18 Computation of resultant hyetogPrecipitation 69 area between isohyets ‘mean precipitation in a Pas + Pray a,+a, Pa, + Prat Pads a,+a,+a, eal mean value for the whole basin 4 Asata, Reorascry a A, = total area of the basin A Eig AA, Ay A, (km?) Figure 3.19 Preparation of Depth - Area curve for a storm storm 1 All the storms have the same duration d (mm) pep o+o A (km?) Figure 3.2 Depth - Area curve fora certain duration70. Engineering Hydrology When the same analysis is repeated for storms of different durations, which occurred in the same area, a family of curves is obtained as shown in Figure 3.21. Depthi (mm) S hours 2 hours duration Area (km? or %) Figure 3.21 Depth - Area - Duration curves As seen in Figure 3.21, depth-area-duration curves give two distinct types of information for design purposes. One is the change of depth with different durations for the same area (as the duration increases, the depth also increases), and the second is the change of depth with area for the same duration (as the area increases, the depth decreases) 3.3.8 Intensity-Duration-Frequency Curves In general, the higher the intensity of the rainfall the shorter the duration of it will be. The intensity-duration-frequency relationship, of importance to engineers in designing hydraulic structures, is also depicted by a family of curves as shown in Figure 3.22. Each curve is drawn for a certain frequency, and indicates the change of intensity with respect to the time interval called the reference time interval, which shows the duration of the storm. Intensity (mmr) frequency Reference time (min or hr) Figure 3.22 Intensity - Duration - Frequency curvesPrecipation 7 The relationship between the rainfall intensity and its duration and then the required frequency can be found using different statistical distributions, such as Gumbel distribution. It can be seen in the figure that, for any frequency, intensity decreases as the reference time interval increases. When a certain time interval is considered, intensity increases with frequency, as would be expected Since these curves are obtained by the statistical analysis of the maximum storms, observed in the study area, found in the long time series of rainfall data, they should be generated for all the areas or basins where small hydraulic structures are to be planned and constructed. As an example, intensity-duration- frequency curves obtained for Ankara with 47 years of data are shown in Figure 3.23 on linear scale and in Figure 3.24 on semi-log scale. 200. 3 9. 8 8 1 lei frequencies 2 & intensity (mm/hr) 25 yr | 10 40 ao Cer Bee 0 M1Beckenee S$ oh. 4 45 6 OBS duration (hr) Figure 3.23 Intensity-Duration-Frequency curves for Ankara (on linear scale) Observed greatest point rainfalls are given in Table 3.3 with standard durations for Turkey, and in Table 3.4 with actual durations of the storms for the world 3.4 RATIONAL METHOD Rational method is one of the oldest methods to relate rainfall on a basin to the corresponding runoff, It is introduced to American literature by Kuichling72 Engineering Hydrology 200 + } 160 3S 8 2 s i Intensity (mm/hr) 5 8 0 0.01 Frequencies 100 yr 0.10 50 yr 25 yr 1.00 Time (hr) Figure 3.24 Intensity-Duration-Frequency curves for Ankara Table 3.3 Greatest Observed Point Rainfalls in Turkey* (on Semi-log scale) Duration Depth (am) Location Date (yea) mi 274 Mardin 1982 10 min 36.7 Mardin 1982 15min 456 Zonguldak 1955 30 min 116 Yalova 1977 Thr 110.0 Rize 1957 2hr 159.0 Rize 1957 3 hr 1740 Rize 1957 thr 180.1 mgulda 1955 Shr 181.6 Zonguldak 1955 he 193.4 Zonguldak. 1955 Shr 283.5 Zonguldak 1955 12hr 3174 Zonguldak 1955 18 hr 4131 Zonguldak 1955 24 hr 438.8, Zonguldak 1955 © Data are gathered from DSI (1990) (1889). It is mainly used to estimate the peak discharge from a basin or from a contributing area, which is smaller than 100 km’. The idea behind the rational method is that if a rainfall of certain intensity begins instantaneously and continues indefinitely, the rate of runoff wil == increase until the time of concentration, when the entire basin is contribu"! ———«~ flow at the outlet, then stays constant. The peak flow is calculated by b= tormula given in Eqn. 3.8 10.00Table 3.4 World's Greatest Observed Point Rainfalls (Linsley et al., 1988) Duration Depth mm Location Date 1 min 38 Barot, Guadeloupe Nov. 26, 1970 8 min 126 Fussen, Bavaria May 25, 1920 15 min 198 Plumb Point, Jamaica May 12, 1916 20min 206 ‘Curtea-de-Arges, Roumania July 7, 1889 42 min 305 Holt, Mo. June 22, 1947 2hr 10 min 483 Rockport, W. Va July 18, 1889 2 hr 45 min 559 Danis, Tex. (Imi NNW) May 31, 1935 4hr 30 min 782+ ‘Smethpon, Pa July 18, 1942 ohr 1087 Belouve, Reunion Feb. 28, 1964 12 hr 1340 Belouve, Reunion Feb. 28-29, 1964 18 hr 30min 1.689 Belouve. Reunion Feb. 28.29, 1964 24 he 1870 Cilaos, Reunion Mar. 15-16, 1952 2 days 2500 Cilaos, Reunion Mar. 15-17, 1952 3 days 3240 Cilaos, Reunion Mar. 15-18, 1952 4 days 3721 Cherrapunji India Sep. 12-15, 1974 5 days 3854 Cilaos, Reunion Mar. 13-18, 1952 6 days 4055 Cilaos, Reunion Mar. 13-19, 1952 7 days 4110 Cilaos, Reunion Mar. 12-19. 1952 8 days 4130 Cilaos, Reunion Mar. 11-19, 1952 15 days 4798 Cherrapunji, India June 24-July 8, 1931 31 days 9300 ‘Cherrapunji, India July 1861 2mo 12.767 Cherrapunji, India June-July 1861 3 mo 16 369 Cherrapunji, India May-July 1861 4 mo 18 738 Cherrapunji, India Apr-July 1861 5 mo 20412 Cherrapunji, India Apr-Aug. 1861 6mo 22.454 Cherrapunji, India Apr-Sep. 1861 Limo 22.990 Cherrapunji India Jan -Nov. 1861 26 461 Cherrapunji, India Aug.1860-July 1867 40 768 Cherrapuny, India 1860-1861 Q,=0.278CiA GB.8) Where Q, = the peak flow rate (m'/s), C = the runoff coefficient (dimensionless), the average rainfall intensity (mnvhr), lasting for a critical period of time, t, the time of concentration (hr), the drainage area (km’). The time of concentration, {,, is defined as the time necessary for raindrops falling at the farthest point of the basin (o flow to the outlet point. The intensity of rainfall i, is assumed to be constant during concentration time t., and the peak flow Q,, occurs after the period t.74 Engineering Hydrology The effect of rainfall duration ts, compared to the time of concentration is explained in Figure 3.25, In the first case ty is less than t, and the maximum, discharge gp is less than the basin maximum discharge Q, which occurs in cases 2 and 3 where duration of rainfall is equal to or larger than the time of concentration Case1 (t,t) Case 2 (t,=t,) Case 3 (t,>t,) iA iA iA te 1 Figure 3.25 Peak discharges for different rainfall durations The runoff coefficient C is the least precise variable of the rational method. Its use in the formula implies a fixed ratio of peak runoff rate to rainfall rate for the drainage basin, which in reality is not the case. The proportion of the total rainfall that will reach the storm drains depends on the percent imperviousness, slope, and ponding character of the surface. Values of C vary from 0.05 for flat sandy areas to 0.95 for impervious urban surfaces, and considerable knowledge of the catchments is needed in order to estimate an acceptable value. Suggested coefficients for various surface types are given in Table 3.5 In urban areas, the drainage area usually consists of subareas or subbasins of Therefore, a composite analysis is used to take reas of subbasins different surface characteristics, care of the various surface characteristics in the basin, If the are denoted by A, and the runoff coefficients of each subbasin are denoted by 2, .... n, where nis the number of subbasins; the peak runoff is then C), forj = computed by using the following form of the rational formula: Q, =0.2781 ECA, G9) The following steps are utilized in the application of the Rational formula for the determination of peak flowTable 3.5 Typical C Coefficients for Design with 5 to 10-yr Frequencies (ASCE, 1970)* Description of area Runoff | Description of area Runoff coefficient coefficient Business ‘Streets Downtown areas 0.70-0.95 | Asphalt 0.70 - 0.95 Neighborhood areas 050-070 | Concrete 0.80 - 0.95 Brick 0.70 - 0.85 Residential Drives and walks 0.75 - 0.85 Single-family areas 030-050 Multiunits, detached 0.40 - 0.60 } Roofs 0.75 - 0.95 Multiunits, attached 0.60 - 0.75 Residential (suburban) 0.25 - 0.40 } Lawns; Sandy soil: Apartment dwelling areas 0.50-0.70 | Flat, 2% 0.05 - 0.10 Average, 2-7% 0.10 - 0.15 Industrial Steep, 7% 0.15 - 0.20 Light areas 0.50 -0.80 Heavy areas 0.60 -0.90 | Lawns; Heavy soil: Railroad yard areas 0.20-0.40 | Flat, 2% 0.13-0.17 Parks and cemeteries 0.10-0.25 | Average,2-7% 0.18 -0.22 Playgrounds 0.20-0.35 | Steep, 7% 0.25 -0.35 Unimproved areas 0.10 -0.30 * For 25 and 100 year frequencies, multiply by 1-1 and 1.25 respectively (product cannot exceed 1.0) 1) Estimation of the time of concentration of the drainage area, 2) Estimation of the runoff coefficient, C. 3) Selection of a return period T,, and finding the intensity of rain that will be equaled or exceeded, on the average, once every T, years. To produce equilibrium flows, this design storm must have a duration equal to t. oF longer. The desired intensity can be found from the Intensity-Duration- Frequency curves of that region using a rainfall duration equal to the time of concentration. 4) Determination of the desired peak flow Q. One of the principal assumptions of the Rational Method is that the predicted peak discharge has the same retum period as the rainfall. Another assumption is the constancy of the runoff coefficient during the progress of individual storms and also from storm to storm. Attempts to verify the Rational Method have not produced encouraging results. It is impossible to observe or measure a uniform average rainfall; times of concentration do not seem to accord with the time to eak; and the ratios of runoff to rainfall do not produce a constant coefficient The shortcomings are due to the adoption of a simple model to express a76 Engineering Hydrology complex hydrologic system. Yet the method continues to be used in practice with results accepted by the designers 3.5 PROBABLE MAXIMUM PRECIPITATION (PMP) As more and more rainfall measurements become available throughout the world with the increasing number of rain gauges and with the continuation of existing series of observations, records of maximum rainfall for various durations continue to be broken. The question arises as to whether records will always continue to be superseded or whether there is a physical upper limit to rainfall. This concept of a finite limit has been named the probable maximum precipitation (PMP). The PMP will of course vary over the Earth's surface according to the climatic or precipitation regime. It would be expected to be much higher in the hot humid equatorial regions than in the colder regions of the mid-latitudes where the atmosphere is not able to hold so much moisture. The concept of PMP, to which no return period or recurrence interval can be attached, is attractive to the engineer responsible for the design of an impounding dam that must neither fail nor be overtopped by flood flows. A definition of PMP has been given by Wiesner as the depth of precipitation which, for a given area and duration, can be reached but not exceeded under known meteorological conditions. When a value of PMP is required for a specific project, it is common practice to estimate it by several methods and then make an engineering judgment on the value to be used ‘There are two main groups of methods for the evaluation of PMP. One group uses statistical techniques applied to the measurements of extreme rainfalls, The second group, generally undertaken by meteorologists, studies the storm mechanisms causing heavy rainfalls. 3.5.1 Statistical Methods The annual maxima for the necessary duration from long orographic rainfall records are fitted to an extreme value distribution. Then for such a distribution, it can be shown that Ry (the rainfall depth in the given duration of return period T) is: (3.10) Where R is the mean of the annual maximum rainfalls, sp is their standard deviation and K(T,), known as the frequency factor, is a function of the return period, T,Precipitation 77 For a PMP, a maximum value of K(T,) has to be specified to give a value of Rr that would not be exceeded. Although in principle K(T,) gets larger indefinitely with increasing T,, it may be considered that K(T,) has some maximum value, Ky. Thus PMP =R+K,,sq B11) It has been suggested that K,, = 15 gives a good estimate of PMP. This is a very useful and simple method for short-duration storms over small areas. but is dependent both on the length of records available for Rand sp and on the assumption of the maximum value of the frequency factor. 3.5.2 Meteorological Methods These methods require considerable meteorological information; they extend the principle of maximizing rainfall values by taking the maximum precipitable water possible and substituting it into individual storm analysis. Where there are areas that have not experienced high-intensity storms, knowledge of extreme storms is transposed from neighboring areas. Storm transposition takes into account storm movement, orographic effects and precipitable water content of the atmosphere available in dew point measurements. 3.6 SNOW COVER AND SNOWMELT In many areas, snow is the dominant source of streamflow and in the extreme cases the cause of floods. A very good example in Turkey would be the mountainous regions in Eastern Anatolia, where almost 70% of river flows are due to snowmelt. Hence, it is very important for hydrologists to understand the nature and distribution of snowfall and the mechanisms involved in the snowmelt process, so that they may make adequate estimates of the streamflow which would be produced by this source. Snowmelt usually begins in the spring and ends in the early summer. The tunoft produced by snowmelt is normally out of phase with the periods of greatest water need, therefore various control structures such as storage reservoirs have been developed to minimize this problem. An additional point of significance is that some of the greatest floods result from combined large scale rainstorms and snowmelt, as was the case of Seyhan flood in 1980, which threatened the entire Cukurova region in Souther Turkey Streamflow forecasting is highly dependent upon adequate knowle of the extent and characteristics of snow fields within the watershed, Geographic.78 Engineering Hydrology geologic, topographic, and vegetative factors are as important in the sno accumulation-snowmelt-runoff process as they are in the rinlt_H#_—_—— process, For rainfall-runoff relationships, point rainfall measurements are used estimating areal and time distribution over the basin. A similar approach adopted in snow hydrology also, although the point-areal relation}? —_ usually more complex As the melting of snow is a thermodynamic event like evaporation, ener balance or heat transfer in different media have to be considered. Val’ !_ = are needed for each cm’ of water to melt from snow at 0°C. The ene—— necessary to melt the snow may come from: a) radiation, b) heat by conduction or convection from the air over the snow, c) heat from the condensed vapor over the snow, d) heat from the soil below, e) heat from the rain falling on the snow, or £) compaction due to its own weight. An energy budget equation can be written for snowpack as given in Eqn. 3 where the energy terms have the unit of cal/em*/day. SE=E, +E, +E, +E, +E, G12) Where change in the energy of the snowpack net radiation = heat coming from ground heat from the air latent heat released from condensed water vapor on snow E, = heat from the rain The positive energy change will cause the melting of snow. Howes measurements of these factors are not easy and therefore 1 ——7 computations are difficult. U.S. Army Corps of Engineers developed similar equations after gathering data in Sierra Nevada, California (US. —_ 1956). A simpler computation (degree-day factor) method uses the difference betv air and snow temperatures as the determining factor for snowmelt ay g=—=——— belowPrecipitation 79 M, = Ky (T,-T,)v G13) Where My is the daily snowmelt (in mm), Ky, is the variation coefficient (determined empirically), T, and T, are the temperatures of air and snow respectively (T.= 0 °C or 32°F), and v is the wind speed (mls) As mentioned above rainfall over snow also causes snowmelt. As the drops enter the snowpack, their temperature is reduced to 0°C and an equivalent amount of heat is given to the snow. The melt from the rain is given by the following equation: M,=PT,/ 80 (3.14) Where M, is the snowmelt (mm), P is the depth of precipitation (mm), Ty, is the wet-bulb temperature (°C), and 80 is the latent heat of fusion. U.S. National Weather Service (Anderson, 1973, 1976) developed a similar model for runoff forecasting. Recently remote sensing techniques are also being used for determination of snowmelt hydrographs.80 Engineering Hydrology Yagmur yagsin da varsin kerpigci aglasin, {So long if it rains, who cares if the brickmaker is upset.) Turkish proverb QUESTIONS TO ANSWER 3.1. Explain the principle of artificial precipitation, and under which conditions would it be successful. 3.2. Explain the measurement of snow and its importance in hydrology. 3.3. What are the types of errors in precipitation data and what types of corrective measures can be applied to them? 3.4. What is a point precipitation measurement? How are the point precipitation measurements used to find areal mean precipitation? What are the differences in the methods of finding areal mean precipitation? Discuss which one is best under which conditions. 3.6. Discuss how the number of gaging stations and their orientations influence the areal mean precipitation. 3.7. What is the purpose of depth-area-duration analysi procedure to obtain D-A-D curves for a region. 2 Explain the 3.8. Sketch a set of intensity-duration-frequency curves and explain what type of information you obtain from these curves. 3.9. What is the use of Rational Formula? Explain how each of the elements are determined. SOLVED PROBLEMS Problem 3.1, The cumulative depth diagram of a storm obtained from a recording raingage is given in Figure 3.26. Obtain the corresponding hyetograph for this storm Solution: ‘The cumulative depth diagram is approximated by straight line segments as shown in Figure 3.26 with dotted lines. The slope of each segment gives the | | | |Precipitation 81 rainfall intensity for that period. The solution is given below in tabular form and its graphical representation in Figure 3.27 d(mm) A 70 60 50 40 20 10 t (he) o1 304 6 9 Ww Figure 3.26 Mass curve of precipitation (Problem 3.1) Table 3.6 Solution of Problem 3.1 EDepth | At ad Tn (mm) _|_ (hry (rom) _|_ (mnie) 0 1 10 10 1 10 2 10 3 3 20 ical ie 1 20 20 4 40 2 10 3 6 50 3 10 33 9 60 2 10 S u 70 i(mnvhr), A 20 10 5 5 33 5 L100 Onn S54: 6 9 Ww Figure 3.27 Hyetograph of Problem 3.182. Engineering Hydrology Problem 3.2. For a certain station (X), March 1971 precipitation record is missing due to a failure at the raingage. The stations A, B, and C are as close to and as evenly distributed around X as possible. Estimate the missing monthly record at Station X. Precipitation Normal annual precipitation (mm) 408 [380 | 335 [350 | March 1971 precipitation records (mm) | 35 [25 | 30 Solution: The normal annual precipitation values of the stations should be checked to see whether they differ from each other by more than ten percent or not. Ny _ 350 N 0.86 , 408 Since not all the ratios are in the range 0.9 - 1.1, the simple averaging method | cannot be used, instead the normal ratio method will be used to find the missing value as — +——+——) = 28.1 mm (March 1971 value) 408 380 335 Problem 3.3. The precipitation measuring stations of a basin, the Thiessen polygons for these stations and isohyets of torm which took place on this basin are shown in Figure 3.28. The precipitation amounts, the areas of polygons, and the areas between isohyets are also given (Table 3.7). Determine’ a) b Areal mean precipitation in mm for two days by i) Arithmetic mean method, ii) Thiessen polygons method, iii) Isohyetal map method Resultant hyetograph (i in mm/day, t in day) by i) Arithmetic mean method, ii) Thiessen polygons method.Precipitation 83 70mm -- 60 mm 40mm °F Figure 3.28 Thiessen polygons and isohyets of the storm (Problem 3.3) Table 3.7 Data for Problem 3.3 St[ Precip. (mm) | Area of Thiessen Isohyets | Area between Day 1 | Day 2 | polygon (km?) (mm) _ | Isohyets (km A 15 20 5 >70 20 1B | e575 | E592) 40. 60-70 60 c[ 26 | 36 25 30-60 70 Di 32 [40 30 40-50 30 Bi jrme2a |i 32 50 E 18_| 24 10 G 20 25 40 Solution: | a) Areal Mean Precipitation i) By aritmetic mean method: The stations outside the basin are excluded in this method. For the first day: (dq. = (25+32+24+20)/4 = 5.25 mmviday For the second day: (ds)yy = (32+40+32+25)/4 = 32.25 mmv/day Total areal mean precipitation depth for two days is 25.25 + 32.25 = 50 mm84 Engineenng Hydrology ii) By Thiessen Polygons method: Ics easier to carry out the calculations in a table (Table 3.8). It may be seen in the following table that the total areal mean precipitation depth for two days is 24.3 + 31.6 = 55.9 mm. Table 3.8. Thiessen polygons solution for Problem 3.3 Precip. (mm) | Polygon | Thiessen Sta area Coeff. Day! Day2] (km) S/S A 15 20 5 0.05 0.375_| 0.500 B 25 32 40 0.200 5.00 _| 6.400 calm 36 25 0.125 3.250_| 4.500 D | 2 40. 30 0.150 | 4.800_| 6.000 E_ | 24 32 50. 0.250 6.250_| 8.000 F 18 24 10 0.05 0.900_| 1.200 Gc [| 20 25 40 0.200 4.00_|5.000 Total 200 1,000_| 24.325 [31.600 iii) By Isohyetal map method Calculations are given in tabular form (Table 3.9), with the assumption that diy = 80 mm and d,,,=40 mm (found by studying the map). ‘Table 3.9 Isohyetal map solution for Problem 3.3 “ohyets ‘Area between (mm) 70-80 75 20 1500 60-70 65 60. 50-60 55. 70 40-50 45 50 Total 200 Areal mean precipitation depth = 11500 / 200 = $7.5 mm. b) Resultant hyetographs: The daily mean values from arithmetic mean and Thiessen polygons metho are plotted as resultant hyetographs in Figure 3.29.Precipitation 85 i (mmiday) i (mmvday) 3225 3160 25.25 i - —> (day) (cay) a) arithmetic mean b)Thiessen polygons Figure 3.29 Hyetographs of Problem 3.3 Problem 3.4. The isohyetal map of a 10 hour storm, which took place over a basin, is given in Figure 3.30. For this basin, determine: a) areal mean precipitation depth for the storm, b) depth-area curve for 10 hrs obtained from this storm, 10mm Figure 3.30 Isohyets of the storm (Problem 3.4) Solution: a) Maximum and minimum precipitation depths are assumed as d,.., = 75 mm and djj, = 18 mm from the isohyetal map, and solution is given in Table 3.10. As seen in the table, areal mean precipitation for thel0-hr storm is 48.3 mm. b) The depth-area-curve for this 10 hr duration storm (col.7 vs. col. 4) is given in Figure 3.3186 Engincenng Hydrology Table 3.10 Solution of Problem 3.4 OTo®T ®T@T 8 )_{| Tsohyet | Mean | Area | Ace, | Precip. vol. | Acc. Prec. [| Prean limits | depth | betisoh. | area vol. (mm) (mm)_|_ (mm) (mm,km’} 75-70 _| 72.5 7 507. 72. 70-60, 65, 25. 1625 2132.5 | 66.6 60-50 | 55 30 1650_| 3782.5 | 61.0 50-40 45 a5; 1125 4907.5 | 56.4 40-30 35, 20 107, 700 5607.5 | 52.4 30-20, 25. 15, 122 375. 5982.5 | 49.0 19 2. 125 57 6039.5 | 48.3 125 6039.5 | 48.3 d (mmm) 20 60 nae 40 20 0 ‘Area (km?) 0 20 40 60 80 100 120 140 Figure 3.31 Depth-Area curve of Problem 3.4 Problem 3.5. Isohyetal maps of two storms of equal duration are given in Figure 3.32 for a particular basin. Determine the depth-area curve for this basin for the specific storm duration of 90 minutes using the data given in Table 3.11 Table 3.11 Data for Problem 3.5 Tsohyet limits Area (km) for (mm) Storm 1 Storm 2 >70 3 6 60-70 12 4 50 - 60 15 16 40-50 18 4 30-40 2 Is 20-30 20 16 <20 4 B Z| Solution: The two storms should be analyzed separately and mean depths in each isohy-———— should be found for each storm. Calculations are given below in tabular form.Precipitation 87 Figure 3.32 Isohyets of two storms (Problem 3.5) Storm 1: From the isohyetal map, assume Prag = 78 mm and P,,, = 16 mm. Table 3.12 Solution for Storm 1 of Problem 3.5 Mean] Areabet. | Ace. | Precip. vol depth | isoh. betw. iso. (mmm?) ‘Ace. Pree. limits 60-50 4 30-40 45 1s | 48 810 40-30 35 22m E70. 770 3407 30-20 25 20 | 90 500 3907 18 Amn94 72 3979 Storm 2: From the isohyetal map assume Pyaax = 80 mm, Pg = 14 mm.88 Engineering Hydrology Table 3.13 Solution for Storm 2 of Problem 3.5 Isohyet_ [Mean] Areabet. ] Ace. | Precip. vol. | Acc. Prec. | Prem limits] depth | isoh, area_| betw. isoh. | vol (mm) (mm) _| (km? (km?) | (mmm. ([Link]’)_| (mm) >70 [75 6 6 450. 450 | 75 70-60 | 65 14 20 310 1360 | 68 60-50 | 55 16 36 880 2240 | 62.2 50-40 [as |g 30 630 2870_|_57.4 40-30 35 15. 65. 325 3395 52.2 30-20 | 25 16 81 400. 3795 | 46.9 <20 17 13 94 221 4016 42.7 1 Afier plotting the areal mean precipitations with respect to the corresponding areas an enveloping curve is plotted above the points as shown in Figure 3.33 d (mm) 80 © Storm 1 x Storm 2 60. 40 20: 0 Area (km?) 0 20 40 60 80 100 Figure 3.33 Depth-Area curve of Problem 3.5 PROBLEMS TO SOLVE Problem 3.6. The cumulative rainfall values of a storm, which took place over a basin of 120 km? area are given below. Determine the hyetograph of this storm. Plot the mass curve and the hyetograph separately, and determine the total volume of precipitation over the basin. t@) 0 | 1] 2m] 4 [es | 8e]_o Jatt] a2 Zdimm) | 0 | 5 [ 12 [ 36 | 40 [ 49 [ 49 | 03 | 67 Problem 3.7. Cumulative rainfall depth versus time values of a storm over 1 basin of 120 km’ area are given below. a) Plot the mass curve of this storm b) Construct the hyetograph of this stormPrecipitation 89 c) If 35% of total rainfall becomes surface runoff find the volume of runoff in m’ from this basin. “Time ([Link]) 0.0 [3.00 [5.00 J 5.30 | 6.00 | 7.00 [ 8.00 Rainfall depth (mm) o | 24 | 30 [35 [37 [38 [38 Problem 3.8. There are four stations in hydrologically similar regions of a basin. Total annual precipitation values for these four stations are given in Table 3.14 for eight years. Check the data of Station D for errors and if there is any error, correct it using the data gathered for the surrounding stations. Table 3.14 Data for Problem 3.8 Annual Precipitation (mm) Year A B 82 85 1980 68 B 69 75 1979 79 65. 63 65. 1978 84 81 a2% 74 1977 wt 83 4 85 1976 67 69 7 46 1975, 94 92 84 60 1974 80 4 i 50 Problem 3.9. The rainfall data of a storm, which occurred over a basin of 125 km area, is given on the map shown in Figure 3.34, Determine the areal mean Precipitation for this storm by Thiessen polygons and isohyetal map methods. 32mme 35 mm ° 14mm Figure 3.34 Rainfall data for a basin (Problem 3.9) Problem 3.10. Observed hyetographs of an 8 hour storm which took place over a basin of 85 km” area are given in Figure 3.35 for a number of stations90 Engineering Hydrology inside and outside the basin. Determine the resultant hyetograph for this storm by arithmetic and Thiessen polygons methods. mmibr thr Figure 3.35 Thiessen polygons and observed hyetographs (Problem 3.10) Problem 3.11. The isohyetal map of a basin for an 18 hr storm is given in Figure 3.36. Determine the areal mean precipitation and the depth-area curve for this storm. Figure 3.36 Isohyetal map of the basin (Problem 3.11)Precipitation 91 Problem 3.12. The isohyetal maps of two storms of equal duration are given in Figure 3.37 for a particular basin of 120 km? area. Determine the depth-area curve for this basin for the specific storm duration of 3 hours using the data provided in Table 3.15. Figure 3.37 Isohyets of two 3 hr storms of a basin (Problem 3.12) Table 3.15 Data for Problem 3.12 Isohyet limits ‘Area (km) (mm) Storm 1 ‘Storm? >90) 4 7 80-90 2 16 70 - 80 16 18 60-70 2 4 50-60 4 18 40-50 20 20 30-40 16 20 <0 6 Problem 3.13. Isohyetal maps of two 6 hr storms are given for a basin of 100 ki’ area in Figure 3.38. Determine 6 hr depth-area curve for this basin. Problem 3.14. Rainfall depth - duration - frequency (D-D-F) data of an area (1.2 km’) are given in Table 3.16. Determine the maximum surface runoff from this area for 25 year frequency. Assume that runoff coefficients of the area are92 Engincenng Hydrology 0.75 and 0.85 for 60% and 40% regions respectively. Time of concentration is 20 min. for the area ~~ Storm 1 isohyet (mm) _atea (km?) > 60 5 60 - 50 25 50-40 30 0mm 40-30 20 30-20 15 <20 5 2... Storm 2 40 mm. > 50 10 50-40 40 30mm 40-30 20 ~ 20mm 30-20 20 <20 10 Figure 3.38 Isohyets of the storms (Problem 3.13) Table 3.16 D-D-F data for Problem 3.14 10 min 20min 30 min 45 min 95 M45 190 235 270 335 Problem 3.15. A parking lot has a size of 240 m X 500 m and a concentration time of 20 minutes. Runoff coefficients for 30% and 70% of this area are 0.7 and 0.9 respectively. Rainfall intensities obtained from intensity-duration- frequency curves of nearest precipitation station to this lot are given in Table 3.17. Determine the 10-yr peak discharge value at the outlet of this lot in m/s. Table 3.17 1-D-F data for Problem 3.15 Intensities of rainfall (mmv/hr) Frequency ow At__ (min) (return period) 24 B a 48 26 14 65, 34 18, 85, 4S 23