statistics & flood frequency chapter 3
DESCRIPTION
Statistics & Flood Frequency Chapter 3. Dr. Philip B. Bedient Rice University 2006. Predicting FLOODS. Flood Frequency Analysis. Statistical Methods to evaluate probability exceeding a particular outcome - P (X >20,000 cfs) = 10% Used to determine return periods of rainfall or flows - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/1.jpg)
Statistics & Statistics &
Flood FrequencyFlood Frequency Chapter 3Chapter 3
Dr. Philip B. BedientDr. Philip B. Bedient
Rice University 2006Rice University 2006
![Page 2: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/2.jpg)
Predicting FLOODS
![Page 3: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/3.jpg)
Flood Frequency Analysis Statistical Methods to evaluate probability exceeding Statistical Methods to evaluate probability exceeding
a particular outcome - P (X >20,000 cfs) = 10%a particular outcome - P (X >20,000 cfs) = 10%
Used to determine return periods of rainfall or flowsUsed to determine return periods of rainfall or flows
Used to determine specific frequency flows for Used to determine specific frequency flows for
floodplain mapping purposes (10, 25, 50, 100 yr)floodplain mapping purposes (10, 25, 50, 100 yr)
Used for datasets that have no obvious trendsUsed for datasets that have no obvious trends
Used to statistically extend data setsUsed to statistically extend data sets
![Page 4: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/4.jpg)
Random Variables Parameter that cannot be predicted with certaintyParameter that cannot be predicted with certainty
Outcome of a random or uncertain process - flipping Outcome of a random or uncertain process - flipping
a coin or picking out a card from decka coin or picking out a card from deck
Can be discrete or continuousCan be discrete or continuous
Data are usually discrete or quantizedData are usually discrete or quantized
Usually easier to apply continuous distribution to Usually easier to apply continuous distribution to
discrete data that has been organized into binsdiscrete data that has been organized into bins
![Page 5: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/5.jpg)
Typical CDF
Discrete
Continuous
F(x1) = P(x < x1)
F(x1) - F(x2)
![Page 6: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/6.jpg)
Frequency Histogram
Probability that Q is 10,000 to 15, 000 = 17.3%
Prob that Q < 20,000 = 1.3 + 17.3 + 36 = 54.6%
17.3
36
1.3
27
8 1.39
![Page 7: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/7.jpg)
Probability Distributions
€
F(x) = P(X ≤ x) = P(x i)i
∑
€
F(x1) = P(−∞ ≤ x ≤ x1) = f (x)−∞
x 1
∫ dx
CDF is the most useful form for analysis
€
P(x1 ≤ x ≤ x2) = F(x2) − F(x1)
![Page 8: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/8.jpg)
Moments of a Distribution
€
μN' = x i
N
−∞
∞
∑ P x i( )
€
μN' = xN
−∞
∞
∫ f x( )dx
First Moment about the Origin
€
E(x) = μ = x i∑ P(x i)
€
E(x) = μ = xf (x)−∞
∞
∫ dx
€
nthmoment
Discrete
Continuous
![Page 9: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/9.jpg)
Var(x) = Variance Second moment about
mean
€
Var (x) = σ 2 = (x i − μ)2
−∞
∞
∑ P(x i)
€
Var (x) = (x − μ)2 f (x)−∞
∞
∫ dx
€
Var (x) = E(x 2) − (E(x))2
€
cv =σ
μ = Coeff. of Variation
![Page 10: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/10.jpg)
Estimates of Moments from Data
€
x =1
nx i
i
n
∑ ⇒ Mean of Data
€
sx2 =
1
n −1(x i − x )2∑ ⇒ Variance
Std Dev. Sx = (Sx2)1/2
![Page 11: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/11.jpg)
Skewness CoefficientUsed to evaluate high or low data
points - flood or drought data
€
Skewness →μ3
σ 3→ third central moment
€
Cs =n
(n −1)(n − 2)
(x i − x )3∑sx
3 skewness coeff.
€
Coeff of Var =σ
μ
![Page 12: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/12.jpg)
Mean, Median, Mode• Positive Skew moves mean to right
• Negative Skew moves mean to left
• Normal Dist’n has mean = median = mode
• Median has highest prob. of occurrence
![Page 13: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/13.jpg)
Skewed PDF - Long Right Tail
![Page 14: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/14.jpg)
Brays Bayou at Main (1936-2002)
0
5000
10000
15000
20000
25000
30000
35000
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67
Years
Peak Flow (cfs)
01
'9876 83
![Page 15: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/15.jpg)
Skewed Data
![Page 16: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/16.jpg)
Climate Change Data
![Page 17: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/17.jpg)
Siletz River Data
Stationary Data Showing No Obvious Trends
![Page 18: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/18.jpg)
Data with Trends
![Page 19: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/19.jpg)
Frequency Histogram
Probability that Q is 10,000 to 15, 000 = 17.3%
Prob that Q < 20,000 = 1.3 + 17.3 + 36 = 54.6%
17.3
36
1.3
27
8 1.39
![Page 20: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/20.jpg)
Cumulative Histogram
Probability that Q < 20,000 is 54.6 %Probability that Q > 25,000 is 19 %
![Page 21: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/21.jpg)
PDF - Gamma Dist
![Page 22: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/22.jpg)
Major Distributions Binomial - P (x successes in n trials)
Exponential - decays rapidly to low
probability - event arrival times
Normal - Symmetric based on μ and
Lognormal - Log data are normally dist’d
Gamma - skewed distribution - hydro data
Log Pearson III -skewed logs -recommended
by the IAC on water data - most often used
![Page 23: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/23.jpg)
Binomial Distribution
€
P(x) =n!
x!(n − x)!px (1− p)n−x
x = 0, 1, 2, 3, ..., n
The probability of getting x successes followed by n-x failuresis the product of prob of n independent events: px (1-p)n-x
This represents only one possible outcome. The number of waysof choosing x successes out of n events is the binomial coeff. Theresulting distribution is the Binomial or B(n,p).
Bin. Coeff for single success in 3 years = 3(2)(1) / 2(1) = 3
![Page 24: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/24.jpg)
Binomial Dist’n B(n,p)
![Page 25: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/25.jpg)
Risk and ReliabilityThe probability of at least one success in n years, where the probability of success in any year is 1/T, is called the RISK.
Prob success = p = 1/T and Prob failure = 1-p
RISK = 1 - P(0)= 1 - Prob(no success in n years)
= 1 - (1-p)n
= 1 - (1 - 1/T)n
Reliability = (1 - 1/T)n
![Page 26: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/26.jpg)
Design Periods vs RISK and Design Life
Risk Risk %%
55 1010 2525 5050 100100
7575 4.14.1 7.77.7 18.518.5 36.636.6 72.672.6
5050 7.77.7 14.914.9 36.636.6 72.672.6 144.8144.8
2020 22.922.9 45.345.3 112.5112.5 224.6224.6 448.6448.6
1010 4848 95.495.4 237.8237.8 475.1475.1 949.6949.6
Expected Design Life (Years)
x 2
x 3
![Page 27: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/27.jpg)
Risk ExampleWhat is the probability of at least one 50 yr flood in a 30 year mortgage period, where the probability of success in any year is 1/T = 1.50 = 0.02
RISK = 1 - (1 - 1/T)n = 1 - (1 - 0.02)30
= 1 - (0.98)30 = 0.455 or 46%If this is too large a risk, then increase design level to the 100 year where p = 0.01
RISK = 1 - (0.99)30 = 0.26 or 26%
![Page 28: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/28.jpg)
Exponential Dist’n
Poisson Process where k is average no.of events per time and 1/k is the average time between arrivals
f(t) = k e - kt for t > 0Traffic flowFlood arrivalsTelephone calls
![Page 29: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/29.jpg)
Exponential Dist’nf(t) = k e - kt for t > 0
F(t) = 1 - e - kt
€
E(t) = (tk)e−kt
0
∞
∫ dt
Letting u = kt
Mean or E(t) =1
kue−u
0
∞
∫ du =1
k
Var =1
k 2
Avg Time Between Events
![Page 30: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/30.jpg)
Gamma Dist’n
€
Qn =1
KΓ(n)
t
k
⎛
⎝ ⎜
⎞
⎠ ⎟n−1
e−t / K
Mean or E(t) = nK
Var = nK 2 where Γ(n) = (n −1)!
n =1
n =2
n =3
Unit Hydrographs
![Page 31: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/31.jpg)
Parameters of Dist’nDistributionDistribution NormalNormal
xx
LogNLogN
Y =logxY =logx
GammaGamma
xx
ExpExp
tt
MeanMean μμxx μμyy nknk 1/k1/k
VarianceVariance xx yy
nknk 1/k1/k22
SkewnessSkewness zerozero zerozero 2/n2/n0.50.5 22
![Page 32: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/32.jpg)
Normal, LogN, LPIII
Normal
Data in bins
![Page 33: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/33.jpg)
Normal Prob Paper
Normal Prob Paper converts
the Normal CDF S curve into
a straight line on a prob scale
![Page 34: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/34.jpg)
Normal Prob Paper
Mean = 5200 cfs
Std Dev = +1000 cfs
Std Dev = –1000 cfs
• Place mean at F = 50%
• Place one Sx at 15.9 and 84.1%
• Connect points with st. line• Plot data with plotting position
formula P = m/n+1
![Page 35: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/35.jpg)
Normal Dist’n Fit
Mean
![Page 36: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/36.jpg)
Frequency Analysis of Peak Flow Data
Year Year RankRank Ordered cfsOrdered cfs19401940 11 42,70042,700
19251925 22 31,10031,100
19321932 33 20,70020,700
19661966 44 19,30019,300
19691969 55 14,20014,200
19821982 66 14,20014,200
19881988 77 12,10012,100
19951995 88 10,30010,300
20002000 ………… …………..
![Page 37: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/37.jpg)
Frequency Analysis of Peak Flow Data
Take Mean and Variance (S.D.) of ranked data
Take Skewness Cs of data (3rd moment about mean)
If Cs near zero, assume normal dist’n
If Cs large, convert Y = Log x - (Mean and Var of Y)
Take Skewness of Log data - Cs(Y)
If Cs near zero, then fits Lognormal
If Cs not zero, fit data to Log Pearson III
![Page 38: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/38.jpg)
Siletz River Example 75 data points - Excel
Tools
MeanMean 20,45220,452 4.29214.2921
Std DevStd Dev 60896089 0.1290.129
SkewSkew 0.78890.7889 - 0.1565- 0.1565
Coef of Coef of VariationVariation
0.2980.298 0.030.03
Original Q Y = Log Q
![Page 39: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/39.jpg)
Siletz River Example - Fit Normal and LogN
Normal Distribution Q = Qm + z SQ
Q100 = 20452 + 2.326(6089) = 34,620 cfsMean + z (S.D.)
Where z = std normal variate - tables Log N Distribution Y = Ym + k SY
Y100 = 4.29209 + 2.326(0.129) = 4.5923
k = freq factor and Q = 10Y = 39,100 cfs
![Page 40: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/40.jpg)
Log Pearson Type IIILog Pearson Type III Y = Ym + k SY
K is a function of Cs and Recurrence IntervalTable 3.4 lists values for pos and neg skews
For Cs = -0.15, thus K = 2.15 from Table 3.4
Y100 = 4.29209 + 2.15(0.129) = 4.567
Q = 10Y = 36,927 cfs for LP IIIPlot several points on Log Prob paper
![Page 41: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/41.jpg)
LogN Prob Paper for CDF
• What is the prob that flow exceeds
some given value - 100 yr value• Plot data with plotting position
formula P = m/n+1 , m = rank, n = #• Log N dist’n plots as straight line
![Page 42: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/42.jpg)
LogN Plot of Siletz R.
Straight Line Fits Data Well
Mean
![Page 43: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/43.jpg)
Siletz River Flow Data
Various Fits of CDFsLP3 has curvatureLN is straight line
![Page 44: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/44.jpg)
Flow Duration Curves
![Page 45: Statistics & Flood Frequency Chapter 3](https://reader035.vdocuments.net/reader035/viewer/2022062314/56813ae1550346895da330be/html5/thumbnails/45.jpg)
White Oak at Houston (1936-2002)
0
5000
10000
15000
20000
25000
30000
35000
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67
Years
Peak Flow (cfs)
01
'989892
Trends in data have to be removed before any Frequency Analysis