Watershed Characteristics January 28, 2000 Watershed: Definition and Delineation

The concept of a watershed is basic to all hydrologic designs. Since large watersheds are made up of many smaller watersheds, it is necessary to define the watershed in terms of a point. This point is usually the location at which the design is being made and is referred to as the watershed outlet. With respect to the outlet, the watershed consists of all land area that sheds water to the outlet during a rainstorm. Using the concept that water runs downhill, a watershed is defined by all points enclosed within an area from which rain falling at these points will contribute water to the outlet. Figure 1. depicts the delineation of a watershed boundary.

Figure 1. Delineation of a watershed boundary.

The Delineation Process:

Information Sources USGS Topographic Maps The fundamental source of data that we use for delineating and studying watersheds is the U.S. Geological Survey Quadrangle map. Each Quad Sheet map covers 7.5 minutes of longitude and latitude. These maps give a wealth of information including topographic contour lines, locations of cities, buildings, roads, road types, railroads, pipelines, water bodies, forested land, stream networks, and USGS stream gauging stations and benchmarks.

Quad sheet maps typically have a scale of 1:24,000 (i.e. 1 inch on the map = 24,000 inches in the real world). Depending on the age of the map, elevation data may be in English or Metric units. Typically, here in the Midwest, the contour intervals of the elevation data are 5 feet or 1.5 meter. For watershed delineation, quad sheet maps offer us the best starting point. More detailed analysis would require a detailed topographic survey of the area of interest. Topographic maps are generally available at all state and federal geological survey offices and can now be ordered over the internet. The cost of a 1:24,000 scale map is $4.00. For more information on USGS mapping products check out http://mapping.usgs.gov/. Digital Elevation Models In this age of computers, geographic data can now be stored electronically. Digital Elevation Models (DEMs) store topographic data in the form of grid cells. Typically, these grid cells have a resolution of 30 meters and elevation intervals of 1 foot or 1 meter. Using a DEM within a Geographical Information System (GIS), we can perform digital terrain analysis (DTA) such as calculating slopes, flow lengths, and delineate watershed boundaries and stream networks. However, there are certain drawbacks to DTA because some algorithms are not very smart, especially in delineating watershed boundaries. To find DEMs for the state of Michigan, check out http://edc.usgs.gov/doc/edchome/ndcdb/7_min_dem/states/MI.html Delineation Steps There are two basic steps to follow in watershed delineation. Step 1: Choose the point of the watershed outlet. This is generally our point of interest for designing a structure or monitoring location. Step 2: Delineate the watershed boundary by drawing perpendicular lines across the elevation contour lines for land that drains to the point of interest. Note - There are a few things to remember when you are working with topographic maps.

a. A watershed boundary always runs perpendicular to the contour lines. b. Arrows that point upstream are valleys. c. Arrows that point downstream are hills.

Important Watershed Characteristics

Drainage Area The drainage area (A) is the probably the single most important watershed characteristic for hydrologic design. It reflects the volume of water that can be generated from rainfall. It is common in hydrologic design to assume a constant depth of rainfall occurring uniformly over the watershed. Under this assumption, the volume of water available for runoff would be the product of rainfall depth and the drainage area. Thus the drainage area is required as input to models ranging from simple linear prediction equations to complex computer models. Watershed Length The length (L) of a watershed is the second watershed characteristic of interest. While the length increases as the drainage increases, the length of a watershed is important in hydrologic computations. Watershed length is usually defined as the distance measured along the main channel from the watershed outlet to the basin divide. Since the channel does not extend to the basin divide, it is necessary to extend a line from the end of the channel to the basin divide following a path where the greatest volume of water would travel. The straight-line distance from the outlet point on the watershed divide is not usually used to compute L because the travel distance of floodwaters is conceptually the length of interest. Thus, the length is measured along the principal flow path. Since it will be used for hydrologic calculations, this length is more appropriately labeled the hydrologic length. While the drainage area and length are both measures of watershed size, they may reflect different aspects of size. The drainage area is used to indicate the potential for rainfall to provide a volume of water. The length is usually used in computing a time parameter, which is a measure of the travel time of water through a watershed. Watershed Slope Flood magnitudes reflect the momentum of the runoff. Slope is an important factor in the momentum. Both watershed and channel slope may be of interest. Watershed slope reflects the rate of change of elevation with respect to distance along the principal flow path. Typically, the principal flow path is delineated, and the watershed slope (S) is computed as the difference in elevation ( E) between the end points of the principal flow path divided by the hydrologic length of the flow path (L):

S = E/L

The elevation difference E may not necessarily be the maximum elevation difference within the watershed since the point of highest elevation may occur along a side boundary of the watershed rather than at the end of the principal flow path. Watershed Shape Basin shape is not usually used directly in hydrologic design methods; however, parameters that reflect basin shape are used occasionally and have a conceptual basis. Watersheds have an infinite variety of shapes, and the shape supposedly reflects the way that runoff will bunch up at the outlet. A circular watershed would result in runoff from various parts of the watershed reaching the outlet at the same time. An elliptical watershed having the outlet at one end of the major axis and having the same area as the circular watershed would cause the runoff to be spread out over time, thus producing a smaller flood peak than that of the circular watershed. A number of watershed parameters have been developed to reflect basin shape. The following are a few typical parameters: 1. Length to the center of area (Lca): the distance in miles measured along the main channel from the basin outlet to the point on the main channel opposite the center of area. 2. Shape Factor (Ll) Ll = (LLca)0.3 Where L is the length of the watershed in miles 3. Circularity ratio (Fc): Fc = P/(4 A)0.5 Where P and A are the perimeter (ft) and area (ft2) of the watershed, respectively. 4. Circularity ration (Rc): Rc = A/Ao Where A0 is the area of a circle having a perimeter equal to the perimeter of the basin. 5. Elongation Ration (Re): Re = 2/Lm(A/ )0.5

Where Lm is the maximum length (ft) of the basin parallel to the principal drainage lines. Generally, the shape factor (Ll) is the best descriptor of peak discharge. It is negatively correlated with peak discharge (i.e. as the Ll decreases, peak discharge increases).

Other Important Watershed Factors Land Cover and Use

Surface Roughness Soil Characteristics Texture Soil Structure Soil Moisture Hydrologic Soil Groups

Channel Geomorphology
Channel Length In addition to the drainage area and the watershed length, the channel length is used frequently in hydrologic computation. Two computational schemes are used to computer the channel length:

1. 1. The distance measured along the main channel from the watershed outlet to the end of the channel as indicated on the figure below, which is denoted as Lc. 2. 2. The distance measured along the main channel between two points located 10 and 85% of the distance along the channel from the outlet, which is denoted at L10-85. These definitions along with the watershed length are illustrated below. The watershed length is defined by extending a line on the map from the end of the main channel to the divide. This requires some subjective assessment and is often a source of inaccuracy. The definitions for channel length also involve a measure of subjectivity because the endpoint of the channel is dependent on the way the map was drawn.

Channel Definition 1: L Definition 2: Lc Definition 3: L10-85

Channel Slope The channel slope can be described with any one of a number of computation schemes. The most common is

Sc = Ec/Lc

In which Ec is the difference in elevation between the points defining the upper and lower ends of the channel and Lc is the length of the channel between the same to points. The 10-85 slope can also be used:

S10-85 = E10-85/L10-85
For cases where the channel slope is not uniform, a weighted slope may provide an index that better reflects the effect of slope on the hydrologic response of the watershed. Drainage Density The drainage density (D0 is the ratio of the total length of streams within a watershed to the total area of the watershed; thus D has units of the reciprocal of length (1/L). A high value of the drainage density would indicate a relatively high density of streams and thus a rapid storm response. Values typically ranges from 1.5 to 6 mi/mi2. D = Lt/A Hortons Laws Our old pal Horton (from Hortons infiltration equation fame) developed a set of laws that are indicators of the geomorphological characteristics of watershed. The stream order is a measure of the degree of stream branching within a watershed. Each length of stream is indicated by its order (for example, first-order, second-order, etc.). A first-order stream is an unbranched tributary, a second-order stream is a tributary formed by two or more first-order streams. A third-order stream is a tributary formed by two or more second-order streams and so on. In general, an nth order stream is a tributary formed by two or more streams of order (n-1) and streams of lower order. For a watershed, the principal order is defined as the order of the principal channel. The figure below gives an example of stream ordering.

The concept of stream order is used to computer other indicators of drainage character. The bifurcation ratio (Rb) is defined as the ratio of the number of streams of any order to the number of streams of the next highest order. Values of Rb typically range from the theoretical minimum of 2 to around 6. Typically, the values range from 3 to 5. The bifurcation ratio is calculated as Rb = Ni/Ni+! From this, Horton developed the Law of Stream Numbers which relates the number of streams of order I (Ni) to the bifurcation ratio and the principal stream order (k) Ni = Rbk-1 Example: The bifurcation ratio of a watershed is the average of the bifurcation ratios of each stream order For a watershed with a bifurcation ratio of 2.6 and a fourth-order principal stream, Ni = 2.64-I This would predict 18, 7, and 3 streams of order 1, 2, and 3, respectively. In addition to this Horton proposed a Law of Stream Lengths, in which the average lengths of the streams of successive orders are related by a length ratio RL: RL = Li+1/Li Li = L1rLi-1

By similar reasoning, Schumm (1956) proposed a Law of Stream Areas to relate the average areas Ai drained by streams of successive order RA= Ai+1/Ai Example

Estimating Peak Discharge

Thought of the day: Most natural hydrologic phenomena are so complex that they are beyond comprehension, or exact laws governing such phenomena have not been fully discovered. Before such laws can ever be found, complicated hydrologic phenomena (the prototype) can only be approximated by modeling -Ven Te Chow

Introduction When water entering an area is more than what can be transmitted or routed by established watercourses, flooding occurs. Generating hydrographs from big storms becomes the job of flood prediction. We use the word prediction because we still deal with probabilities of events. Some of the reasons we study runoff peaks and volumes are: -- To quantify the volume and rate of water to be handled by water management facilities -- To predict soil erosion and transport of surface pollutants -- To identify critical non-point source pollution areas Peak Discharge Definition The maximum volume flow rate passing a particular location during a storm event. Peak discharge has units of volume/time (e.g. ft3/sec, m3/sec, acre-feet/hour). The peak discharge is a primary design variable for the design of stormwater runoff facilities such as pipe systems, storm inlets and culverts, and small open channels. It is also used for some hydrologic planning such as small detention facilities in urban areas. There have been many different approaches for determining the peak runoff from an area. As a result many different models (equations) for peak discharge estimation have been developed. Ideally, we would like to have a 50-year flood record available at every site where a peak discharge estimate is needed for design work. If such data were always available, then a frequency analysis of the flood record could be used to characterize the flood potential at the site of the design work. More often than not, flood records are rarely available where peak discharge estimates are needed for design work. Therefore, it is necessary to use either a prediction method that was developed from flood frequency analyses of gaged data in the region or an uncalibrated prediction equation that was designed for use at ungaged sites. There are two general classes of peak discharge hydrologic models. They are Calibrated Models and Uncalibrated Models. Calibrated Models Calibrated models are generally multi-parameter regression models that were derived from a frequency analysis from long-term gaged data from watersheds in the region. Some examples of calibrated models include:

USGS Urban Peak Discharge Formulas Index-Flood Estimation Moment Estimation Calibrated models are based on the analysis of stream gage data. For small watersheds, especially those undergoing urban/suburban development, regional equations that are appropriate for assessing the impact of development of peak discharges are not available, with the possible exception of the USGS regression equations. However, these are not widely used because they do not include variables that are typically used to reflect changes in watershed conditions. Thus, there is a demand for methods that provide peak discharge estimates that use readily available input data such as watershed and designstorm rainfall characteristics. Uncalibrated Models Rational Method The most widely used uncalibrated equation is the Rational Method. Mathematically, the rational method relates the peak discharge (q, m3/sec) to the drainage area (A, ha), the rainfall intensity (i, mm/hr), and the runoff coefficient (C). SI Units q = 0.0028CiA Where q = design peak runoff rate in m3/s C = the runoff coefficient i = rainfall intensity in mm/h for the design return period and for a duration equal to the time of concentration of the watershed. English Units q = CiA Where q ft3/sec i = rainfall intensity (in/hr) A = watershed area in acres C = runoff coefficient, ratio of the peak runoff rate to the rainfall intensity, dimensionless To use the rational method there are a few assumptions assumptions. Rainfall intensity and duration is uniform over the area of study Storm duration must be equal to or greater than the time of concentration of the watershed.

Time of Concentration (Tc) Definition The time required for water to flow from the most remote (in time of flow) point of the area to the outlet once the soil has become saturated and minor depressions filled. It is assumed that when the duration of the storm equals the time of concentration, all parts of the watershed are contributing simultaneously to the discharge at the outlet. There are several methods for computing time of concentration. Two of the most popular methods are the Kirpich equation and the SCS lag formula. Kirpich Formula Tc = 0.0195 L0.77S-0.385 where Tc = time of concentration in minutes. L = maximum length of flow in meters. S = the watershed gradient in m per m or the difference in elevation between the outlet and the most remote point divided by the length L. Note to convert to English units, the constant becomes 0.0078. SCS Lag Formula Tc = 0.00526 L0.8(1000/CN 9)0.7 S-0.5 where L = watershed length (ft) S = watershed slope (ft/ft) CN = curve number Rational Method Runoff Coefficients The rational method uses runoff coefficients in the same fashion as the SCS curve number method for estimating runoff volume. They have been determined over the years and primarily focus on urban watershed applications. Below are several tables for different land conditions.

from Schwab (1981)

from McCuen (1998) Example Problem. Determine the design peak runoff rate using the rational method for a 10-year return period storm from a 120-acre suburban/agricultural watershed in East Lansing, MI, with the following characteristics: Area (acres) 40 80 Percent Slope Flat 2% 4% Soil Group B B Description of Area Single Family Res. Cultivated Land

SCS Triangular Hydrograph Method

In this method, peak runoff is determined using the curve number approach. The assumption of uniform rainfall still applies. The hydrograph takes on a triangular shape with equal peak and flow volume as in the rational method. This can be seen below.

Peak runoff rate is calculated by q = 0.0021QA/Tp where Q = runoff volume in mm depth (from the curve number) q = runoff rate in m3/s A = watershed area in ha. Tp = time of peak in hours In this method, the time to peak does not equal the time of concentration as in the rational method, in this method time to peak Tp equals Tp = D/2 + TL = D/2 + 0.6Tc Where Tp = time to peak (hours) D = duration of excess rainfall TL = time of lag Tc = time of concentration It is assumed that the total time of flow is 2.67 Tp and the recession time of the hydrograph is 1.67 Tp.

Time of concentration is calculated using the SCS Lag formula discussed earlier Tc = 0.00526 L0.8(1000/CN 9)0.7 S-0.5 where L = watershed length (ft) S = watershed slope (ft/ft) CN = curve number

Example 2. Determine the peak runoff rate from an 800 ha watershed in Newaygo County, Michigan experiencing a 10-year 6 hour storm event. The weighted curve number for the watershed is 72. The length of flow in the watershed is 2,500 meters. The elevation drop across the watershed is 35 meters.

Measuring Flow with Weirs

Feb 7, 2000 Broad Crested Weir

Sharp Crested Weir

Basic Weir Equation Q = CLH3/2 Where: Q is discharge in ft3/s L is Length of the Weir Crest (ft) H is the head of water above the crest measured at some distance upstream (ft)

C is a discharge coefficient for the weir. C is often times difficult to determine for different type weirs so standard weirs were developed.

Rectangular Contracted Weir

Cipoletti Weir

V-Notch Weir

Compound Weir

Discharge Equation

where: Q = discharge in ft3/s h1 = head above the point of the V-notch in ft L = combined length of the horizontal portions of the weir in ft h2 = head above the horizontal crest in ft

Design Hydrographs SCS Dimensionless Hydrograph

February 25, 2000

Introduction While peak discharge rates are adequate for many engineering design problems, they are inadequate for design problems where watershed or channel storage is significant. This can be storage in manmade structures, storage at a natural constriction in the flow path, or channel storage. Where storage is significant, design work is usually based on hydrographs rather than peak discharge rates. A hydrograph is a graphical or tabular representation of runoff rate against time. Hydrographs are also used for

designs on non-homogeneous watersheds, such as where significant variation in land use, soil types, or topography exists within the watershed. This is especially necessary where parts of watersheds undergo land use change such as urbanization or deforestation. There are many types of design hydrographs that have been developed over the years. We are already familiar with the simple SCS triangular hydrograph for determining peak runoff rate. Other hydrographs that are commonly used are the SCS dimensionless hydrograph, the unit hydrograph, and the synthetic unit hydrograph. Today we are going to discuss the SCS dimensionless hydrograph and then focus next lecture or two on developing unit hydrographs. SCS Dimensionless Hydrograph The SCS dimensionless hydrograph is an idealized shape that approximates the flow from an intense storm from a small watershed. Sometimes it is called a synthetic hydrograph. The dimensionless hydrograph arbitrarily has units of 100 units of flow for the peak and 100 units of time for the duration of flow. The area under a dimensionless hydrograph has 2,620 square units of runoff. The SCS hydrograph has 19 constant ordinates that represent percentages of flow and time. They can be seen on the figure below. To develop the design hydrograph for a watershed, the peak flow and the runoff volume must be known for the desired return period storm (determined from Curve Number Method). The design hydrograph is developed from the dimensionless hydrograph by using approximate conversion factors. This allows us to determine the hydrograph from different sized storms by scaling the hydrograph in both space and time.

Conversion Factors This dimensionless hydrograph is used to scale various size storm events that have various peak runoffs and rates to create a design hydrograph. There are three scaling factors used in the dimensionless hydrograph. The first factor is u and it is the ratio of the total runoff volume to the area under the dimensionless hydrograph. Recall that the area under the dimensionless hydrograph has 2,620 units of runoff. So each single unit has a value of u = Q /2620 where Q is the total storm runoff volume. (Generally found from the Curve Number Method) Runoff (Q) is normally expressed in units of either hectare-meters or acre-feet. The second factor is w and is the ratio of the peak runoff for the design storm to the peak flow of 100 on the dimensionless hydrograph. Each unit of flow on the dimensionless hydrograph has a value of w = q/100 where q is the peak runoff (Generally found with the SCS peak flow equation). The third factor is k and is it the value that each unit of time on the dimensionless hydrograph represents in the design hydrograph. On the design hydrograph 1/100 of the peak flow times 1/100 of the duration of the runoff must equal 1/2620 of the flood

volume just as it does on the dimensionless hydrograph. Since w is equal to 1/100 of the design peak flow, k must be equal to 1/100 of the design duration, and u is 1/2620 of the design flood volume therefore w*k = u and k = u/w When runoff rate is measured in m3/s, runoff volume is measured in hectares-meters, and time is measured in minutes. So, k = u(ha-m) * 10,000 (m2/ha) = 167 u W(m3/s) * 60 (sec/min) w Finally an equation that the units work out on! The coordinates of the design hydrograph are obtained by multiplying the ordinates and abscissas of the dimensionless hydrograph by w and k respectively.

Example 1 Suppose a watershed has a peak runoff of 8.2 m3/s and a flood volume 2.9 ha-m. Step 1: Calculate u u = Q/2620 u = 2.9 / 2620 = 0.00110687 ha-m / unit Step 2: Calculate w w = q / 100 w = 8.2 / 100 = 0.082 m3/s / unit Step 3 Calculate k k = 167 u/w k = 167 * 0.00110687 / 0.082 = 2.2542 min/unit Step 4 Multiply ordinates
k = 2.2542 w = 0.082 Design Hydrograph Coordinates Point a b c d e f g h I j k l m n o p q r s Time Ordinate Flow Ordinate 0 2 6 8 12 14 16 18 20 22 24 26 30 34 38 44 52 64 100 0 3 19 31 66 82 93 99 100 99 93 86 68 46 33 21 11 4 0 k*t 0 4.5084 13.5252 18.0336 27.0504 31.5588 36.0672 40.5756 45.084 49.5924 54.1008 58.6092 67.626 76.6428 85.6596 99.1848 117.2184 144.2688 225.42 w*q 0 0.246 1.558 2.542 5.412 6.724 7.626 8.118 8.2 8.118 7.626 7.052 5.576 3.772 2.706 1.722 0.902 0.328 0

Resulting design hydrograph.

8.5 8 7.5 7 6.5 6 5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 0 25 50 75 100 125 150 175 200 225
Tim (m e inutes)

The result is a graph of flow rate versus time for a watershed with a peak flow rate of 8.2 m3/s with total runoff volume equaling 2.9 ha-m. The length of time is 225.4 minutes.

Example 2 Assuming the dimensionless hydrograph is applicable, determine the duration of flow if the peak runoff is 8.5 m3/s and the volume of flow is 102 mm for a 50 yr storm from an 85 ha watershed. What are the coordinates for point n on the design hydrograph?

Unit Hydrograph Derivation

The discrete convolution equation allows the computation of direct runoff Qn given excess rainfall Pm and the unit hydrograph Un-m+1.
n M

m =1

PmU

n m +1

The reverse process, called deconvolution, is needed to derive a unit hydrograph given data on Pm and Qn. Suppose that there are M pulses of excess rainfall and N pulse of direct runoff in the storm considered; then N equation can be written for Qn, n = 1,2, .,N, in terms of N M + 1 unknown values of the unit hydrograph. If Qn and Pm are given and Un-m+1 is required, the set equations is overdetermined, because there are more equations (N) than unknowns (N M + 1).
n M

This table shows the set of equations for discrete time convolution n = 1, 2, , N
Q1 = Q2 = Q3 = .. QM = QM+1 = . QN 1 = QN = P1U1 P2U1 + P3U1 + PMU1 + 0+ P1U2 P1U2 + PM-1U2 + PMU2 P1U3 + + + + + + + + P1UM P2UM + P1UM+1

m =1

PmU

n m +1

0+ 0+

0+ 0+

0+ 0+

0+ 0+

+ + + +

PMUN-M +1 0

PM-1UN-M+1 PMUN-M+1

Lets do an example to illustrate this set of equations. Try to work it out on your own to cement this concept. Example 1. Suppose we have the observed hydrograph shown below. The time interval is 6 hours between readings. The excess rainfall is also shown. Hour 1 2 3 4 5 6 7 8 9 10 11 12 13 Excess Rainfall (inches) 0.5 1.2 0.9 Observed Direct Discharge (ft3/s) 5 35 90 203 816 1602 1138 302 275 158 65 47 12

This is the observed hydrograph for this event.

1800 1600
Observed Discharge (cfs)

1400 1200 1000 800 600 400 200 0 30 80 130 180 230 280 330 380
Time (30 minute interval)

Lets now derive the unit hydrograph for this watershed.

Remember, the unit hydrograph is defined as a direct runoff hydrograph resulting from 1 inch of excess rainfall generated uniformly over the drainage area at a constant rate for an effective duration. 1st define our number of equations. There are 3 pulses of rainfall so M = 3. There are 13 pulses of observed direct runoff so N = 13. The total number of unit hydrograph ordinates are N M +1 == 13 3 + 1 = 11 ordinates. So we have to solve 11 linear equations. Lets set them up. Ill show the entire table to illustrate the matrix type approach this takes.
Q1 = Q2 = Q3 = Q4 = Q5 = Q6 = Q7 = Q8 = Q9 = Q10 = Q11 = P1U1 P2U1 + P3U1 + 0+ 0+ 0+ 0+ 0+ 0+ 0+ 0+ +0 P1U2 P2U2 + P3U2 + 0+ 0+ 0+ 0+ 0+ 0+ 0+ +0 +0 P1U3 P2U3 + P3U3 + 0+ 0+ 0+ 0+ 0+ 0+ +0 +0 +0 P1U4 P2U4 + P3U4 + 0+ 0+ 0+ 0+ 0+ +0 +0 +0 +0 P1U5 P2U5 + P3U5 + 0+ 0+ 0+ 0+ +0 +0 +0 +0 +0 P1U6 P2U6 + P3U6 + 0+ 0+ 0+ +0 +0 +0 +0 +0 +0 P1U7 P2U7 + P3U7 + 0+ 0+ +0 +0 +0 +0 +0 +0 +0 P1U8 P3U8 + P3U8 + 0+ +0 +0 +0 +0 +0 +0 +0 +0 P1U9 + P2U9 + P3U9 + +0 +0 +0 +0 +0 +0 +0 +0 +0 P1U10 P2U10 + +0 +0 +0 +0 +0 +0 +0 +0 +0 +0 P1U11

Explanation of the Table

What the unit hydrograph says is that each pulse of runoff (1-13 in this example) is generated by some linear combination of the excess rainfall. For example, the very first pulse of runoff (5 ft3/s) is caused by the very first pulse of rainfall (0.5 inches). Because it is the only rainfall that occurred during that time interval, it alone is responsible for the runoff that is occurring. Similarly, The second pulse of rainfall is caused by rainfall pulse 1 and 2 (0.5 and 1.2 inches) because they alone occurred during that time interval. So the unit hydrograph is merely a solution to a set of linear equations that determine the contributions of rainfall over time to the direct runoff hydrograph. The unit hydrograph becomes normalized during the deconvolution process to represent the flow that would occur from one unit of rainfall occurring during the 1st time interval. Now plug in the numbers and see what we get.

5 = 35 = 90 = 203 = 816 = 1602 = 1138 = 302 = 275 = 158 = 54 =

0.5U1 1.2U1 + 0.9U1 + 0+ 0+ 0+ 0+ 0+ 0+ 0+ 0+

+0 0.5U2 1.2U2 + 0.9U2 + 0+ 0+ 0+ 0+ 0+ 0+ 0+

+0 +0 0.5U3 1.2U3 + 0.9U3 + 0+ 0+ 0+ 0+ 0+ 0+

+0 +0 +0 0.5U4 1.2U4 + 0.9U4 + 0+ 0+ 0+ 0+ 0+

+0 +0 +0 +0 0.5U5 1.2U5 + 0.9U5 + 0+ 0+ 0+ 0+

+0 +0 +0 +0 +0 0.5U6 1.2U6 + 0.9U6 + 0+ 0+ 0+

+0 +0 +0 +0 +0 +0 0.5U7 1.2U7 + 0.9U7 + 0+ 0+

+0 +0 +0 +0 +0 +0 +0 0.5U8 1.2U8 + 0.9U8 + 0+

+0 +0 +0 +0 +0 +0 +0 +0 0.5U9 + 1.2U9 + 0.9U9 +

+0 +0 +0 +0 +0 +0 +0 +0 +0 0.5U10 1.2U10 +

+0 +0 +0 +0 +0 +0 +0 +0 +0 +0 0.5U11

We now have 11 linear equations with 11 unknowns. Each unknown is an ordinate of the unit hydrograph. We have to solve these in a step-wise fashion starting with U1. Lets work them out. EQ 1. rainfall 5 = 0.5 U1 U1 = 5/0.5 = 10 cfs direct runoff / inch excess rearrange and solve for U2 U2 =

EQ 2. 35 = 1.2 U1 + 0.5 U2 (35 1.2 U1)/0.5 = (35 1.2(10))/0.5 = 46

EQ 3. 90 = 0.9 U1 + 1.2 U2 + 0.5 U3 = (90 0.9(10) 1.2(46))/0.5 = 51.6 EQ 4. 203 = 0.9U2 + 1.2U3 + 0.5U4 = (203 0.9(46) 1.2(51.6))/0.5 = 199.4 EQ 5. 816 = 0.9U3 + 1.2U4 + 0.5U5 = (816 0.9(51.6) 1.2(199.4))/0.5 = 1060.7

U3 = (90 0.9U1 1.2U2)/0.5 U4 = (203 - 0.9U2 - 1.2U3)/0.5 U5 = (816 -0.9U3 - 1.2U4)/0.5

EQ 6. 1602 = 0.9U4 + 1.2U5 + 0.5U6U6 = (1602 - 0.9U4 - 1.2U5)/0.5 = (1602 - 0.9(199.4) 1.2(1060.7)/0.5 = 299.4 EQ 7. 1138 = 0.9U5 + 1.2U6 + 0.5U7U7 = (1138 - 0.9U5 - 1.2U6)/0.5 = (1138 0.9(1060.7) 1.2(99.4)/0.5 = 128.2

EQ 8. 302 = 0.9U6 + 1.2U7 + 0.5U8 = (302 0.9(99.4) 1.2(128.2)/0.5 = 117.4 EQ 9. 275 = 0.9U7 + 1.2U8 + 0.5U9 = (275 0.9(128.2) 1.2(117.4)/0.5 = 37.5 EQ 10 158 = 0.9U8 + 1.2U9 + 0.5U10 1.2U9)/0.5 = (158 0.9(117.4) 1.2(37.5)/0.5 = 14.7 EQ 11 54 = 0.9U9 + 1.2U10 + 0.5U11 = (54 0.9(37.5) 1.2(14.7) / 0.5 = 5.2

U8 = (685 - 0.9U6 - 1.2U7)/0.5 U9 = (275 - 0.9U7 - 1.2U8)/0.5 U10= (158 - 0.9U8 U11= (54 - 0.9U9 - 1.2U10)/0.5

So the resulting unit hydrograph is shown below in tabular and graphical form. So remember that this hydrograph represents the flow that would result from 1 inch of rainfall occurring during the 1st time interval.
Time Interval (1/2 hr) Unit Hydrograph (cfs) 1 10 2 46 3 51.6 4 199.4 5 1060.7 6 299.4 7 128.2 8 117.4 9 37.5 10 14.7 11 5.2
Unit Hydrograph 1200 1000
Discharge (cfs)

800 600 400 200 0 1 2 3 4 5 6 7 8 9 10 11

Time Interval (30 minutes)

Using the Unit Hydrograph to derive a direct runoff hydrograph

Now that we have developed the unit hydrograph, we can use it to determine the direct runoff hydrograph for any rainfall amount with any time distribution. The process is called convolution. Lets use the following rainfall distribution to calculate the direct runoff hydrograph.
Time Interval (1/2 Unit Hydrograph hr) Rainfall (in) (cfs) 1 1.3 10 2 0.5 46 3 51.6 4 199.4 5 1060.7 6 299.4 7 128.2 8 117.4 9 37.5 10 14.7 11 5.2

We can derive a new design hydrograph from the unit hydrograph by setting up a table to help solve the 11 linear equations. The table looks like this.

Unit Hydrograph Ordinates Direct Runo Time Interval Rainfall U1 U2 U3 U4 U5 U6 U7 U8 U9 U10 U11 1 P1 P1U1 0 0 0 0 0 0 0 0 0 0 P1U1 2 P2 P2U1 P1U2 0 0 0 0 0 0 0 0 0 P2U1 + P1U 3 0 P2U2 P1U3 0 0 0 0 0 0 0 0 P2U2 + P1U 4 0 0 P2U3 P1U4 0 0 0 0 0 0 0 P2U3 + P1U 5 0 0 0 P2U4 P1U5 0 0 0 0 0 0 P2U4 + P1U 6 0 0 0 0 P2U5 P1U6 0 0 0 0 0 P2U5 + P1U 7 0 0 0 0 0 P2U6 P1U7 0 0 0 0 P2U6 + P1U 8 0 0 0 0 0 0 P2U7 P1U8 0 0 0 P2U7 + P1U 9 0 0 0 0 0 0 0 P2U8 P1U9 0 0 P2U8 + P1U 10 0 0 0 0 0 0 0 0 P2U9 P1U10 0 P2U9 + P1U 11 0 0 0 0 0 0 0 0 0 P2U10 P1U11 P2U10 + P1U 0 0 0 0 0 0 0 0 0 0 P2U11 P2U11

Here is the table with numbers plugged in. The sum of the columns equals the direct runoff hydrograph.

Time Interval (1/2 hr) Rainfall (in)

10

46

51.6

Unit Hydrograph 199 1061 299

128

117

37.5

14.7

5.2

Direct Ru (cfs)

1 2 3 4 5 6 7 8 9 10 11

1.3 0.5

13 5 0 0 0 0 0 0 0 0 0 0

0 59.8 23 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 67.1 0 0 0 0 25.8 259 0 0 0 0 99.7 1379 0 0 0 0 530 389 0 0 0 0 150 167 0 0 0 0 64.1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 153 58.7 0 0 0

0 0 0 0 0 0 0 0 48.8 18.8 0 0

0 0 0 0 0 0 0 0 0 19.1 7.35 0

0 0 0 0 0 0 0 0 0 0 6.76 2.6

13 64.8 90.08 285.02 1478.6 919.57 316.36 216.72 107.45 37.86 14.11 2.6

Here is the resulting graph shown the unit hydrograph and the storm event totaling 1.8 inches. Remember, the area under the unit hydrograph equals 1 inch of runoff and the area under the direct runoff hydrograph equals 1.8 inches of runoff.
1600 1400 1200
Discharge (cfs) Storm Unit Hydrograph

1000 800 600 400 200 0 1 2 3 4 5 6 7 8 9 10 11 12

Time Interval (30 minutes)

Working with the unit hydrograph takes some practice and a spreadsheet like Excel is almost a must if you dont want to go crazy. These examples should be able to get you through your homework assignment.