Statistical Analysis of Extreme Rainfall Events Over Indiana, USA

World Environmental & Water Resources Congress 2007
Statistical Analysis of Extreme

Rainfall Events over Indiana, USA
Shih-Chieh Kao & Rao S. Govindaraju

Purdue University
May 17, 2007
Background and Motivation
• Extreme rainfall behavior
– Basis for hydrologic design
– Conventionally analyzed only by “depth”
– Pre-specified artificial duration (filter), not the real
duration of extreme rainfall event
– Hard to represent other rainfall characteristics, e.g.
peak intensity
• Definition of extreme event in multi-variate sense
is not clear
• Dependence exists between rainfall
characteristics (e.g. volume(depth), duration,
peak intensity)
• Explore the use of copulas
Difficulties in Constructing joint Distribution
• Given marginal distributions
and correlation coefficient,
can joint-distribution be
constructed?
ex: exponential margins ?

=> Unlimited choices!!
• Correlation coefficient can not correctly describe association

between variables
ρ=0.85 ρ=0.85 ρ=0.85

Introduction to Copulas
• A copula C(u,v) is a function comprised of margins
u=FX(x) & v=FY(y) from [0,1]×[0,1] to [0,1].
– Sklar (1959) showed that for continuous marginals u and
v, there exists a unique copula C such that
H XY ( x, y ) = CUV (FX ( x ), FY ( y )) = CUV (u, v )
– Transformation from [-∞,∞]2 to [0,1]2
– Provides a complete description of dependence structure

τ = 0.66 τ = 0.02 τ = -0.65
ρ = 0.85 ρ = 0.03 ρ = -0.84
Applications of Copulas in Hydrology
• Flood frequency analysis
– Favre et al. (2004), De Michele et al. (2005), Grimaldi and
Serinaldi (2006), Zhang and Singh (2006)
• Return period assessment using bivariate model

– Salvadori and De Michele (2004)
• Probablistic structure of storm surface runoff

– Kao and Govindaraju (2007)
• Rainfall frequency analysis

– De Michele and Salvadori (2003, 2006), Grimaldi and Serinaldi
(2006), Zhang and Singh (2006)
• Unanswered questions in rainfall frequency analysis

– Data used for analysis may not be sufficient
– Definition of “extreme events” in multi-variate sense?
– Can results be applied for a large region?
Data Source & Study Area
• Nation Climate Data Center, #
# #
# #
#
Hourly Precipitation Dataset #
#
(NCDC, TD 3240 dataset) #
#
#
#
#
# #
• 53 Co-operative Rainfall Stations #
# #
#
in Indiana with record length # # # #
# #
greater than 50 years #
#
#
#
#
• Minimum rainfall hiatus: 6 hours # #
#
# ##
• About 4800 events per station #
#
#
#
#
• Selected variables for analysis: # #
#
#
– Depth (volume), P (mm) # #
– Duration, D (hour) #
– Peak Intensity, I (mm/hour) # #
#
• Marginals: #
#
– u=FP(p), v=FD(d), w=FI(i)
Definitions of Extreme Events
• Hydrologic designs are usually governed by
depth (volume) or peak intensity
• Annual maximum volume (AMV) events
– Longer duration
• Annual maximum peak intensity (AMI) events
– Shorter duration
Analysis of Marginal Distributions (I)
• Candidate distributions
– Extreme value type I (EV1)
– Generalized extreme value (GEV)
– Pearson type III (P3)
– Log-Pearson type III (LP3)
– Generalized Pareto (GP)
– Log-normal (LN)
• Parameters estimated primarily by maximum

likelihood (ML) or method of moments (MOM)
• Gringorton formula for empirical probabilities
• Chi-square and Kolmogorov-Smirnov (KS) test
with 10% significance level
Analysis of Marginal Distributions (II)
AMV Rejection rate (%) of Chi-square test Rejection rate (%) of KS test
events EV1 GEV P3 LP3 GP LN EV1 GEV P3 LP3 GP LN
Depth, P 13.2 17.0 41.5 17.0 100 13.2 0.0 0.0 7.5 0.0 52.8 0.0
Duration, D 13.2 15.1 24.5 37.7 100 22.6 1.9 0.0 7.5 0.0 22.6 0.0
Intensity, I 15.1 17.0 45.3 20.8 100 11.3 0.0 0.0 1.9 0.0 54.7 0.0
AMI Rejection rate (%) of Chi-square test Rejection rate (%) of KS test
events EV1 GEV P3 LP3 GP LN EV1 GEV P3 LP3 GP LN
Depth, P 5.7 3.8 62.3 3.8 100 1.9 0.0 0.0 11.3 0.0 45.3 0.0
Duration, D 60.4 39.6 88.7 37.7 100 28.3 15.1 0.0 45.3 0.0 45.3 0.0
Intensity, I 15.1 15.1 34.0 18.9 100 15.1 0.0 0.0 5.7 0.0 71.7 0.0
• EV1, GEV, LP3, LN provided better fit. GP provided

the worst.
• Fitting for duration of AMI events did not yield very
good result due to the limited temporal precision
(hour)
• EV1 and LN could be recommended for use
Analysis of Dependence Structure (I)
• Candidate Archimedean copulas
– Frank family
– Clayton family
– Genest-Ghoudi family
– Ali-Mikhail-Haq family
• Non-parametric procedure for estimating

dependence parameter
• Examination of Goodness-of-fit
– Distribution function KC(t)=P[C(u,v)≤t]
– Diagonal section of copulas δ(t)=C(t,t)
– Multidimensional KS test (Saunders and Laud, 1980)
Assessment of Copula Performance
Analysis of Dependence Structure (II)
• Statistics of Kendall’s tau
τPD τDI τPI
mean stdev mean stdev mean stdev
AMV events 0.183 0.084 -0.370 0.068 0.260 0.097
AMI events 0.407 0.070 -0.011 0.096 0.405 0.070
• The distribution function KC(t) provides the strictest

examination of copulas
• Clayton and Ali-Mikhail-Haq families performed well for

positive dependence cases (CUV and CUW)
• Frank family of Archimedean copulas

– performed well for both positive and negative dependence
– passed the KS test for entire Indiana at the 10% significant level
– recommended for use in practice
Construct Joint Distribution via Copulas
• Bivariate stochastic models
H PD ( p, d ) = CUV (FP ( p ), FD (d )) = CUV (u, v )
H DI (d , i ) = CVW (FD (d ), FI (i )) = CVW (v, w)
H PI ( p, i ) = CUW (FP ( p ), FI (i )) = CUW (u, w)
• Examples using Frank family and EV1 marginals

Estimate of depth for known duration (I)
• For a known (or measured) d-hour event
H PD ( p, d ) − H PD ( p, d − 1)
FP ( p d − 1 < D ≤ d ) =
FD (d ) − FD (d − 1)
C (F ( p ), FD (d )) − CUV (FP ( p ), FD (d − 1))
= UV P
FD (d ) − FD (d − 1)
• Given return period T, the T-year, d-hour rainfall

estimate pT will satisfy
F (p d −1 < D ≤ d ) = 1 −1 T
P T
• Comparison between bivariate and univariate

depth estimates
– Bivariate using EV1 marginals and Frank family
– Univariate counterpart using GEV distribution (Rao
and Kao, 2006)
Estimate of depth for known duration (II)
• Similar trends were observed for durations greater than

10-hour, close to the univariate counterpart
• For durations less than 10-hour
– AMV estimates gave higher values
– AMI estimates should be better, but fitting problem existed
• Overall, less parameters required for bivariate estimates
Conclusions (I)
• Definition of extreme events
– AMV events are generally of longer duration than AMI
events
– AMV estimates should be better for longer duration,
while AMI estimates being better for shorter duration
– AMI results could be improved by adopting data with
finer temporal resolutions
• Marginal distributions
– EV1, GEV, LP3, LN were found to be appropriate
marginal models for extreme rainfall
– GP was found not appropriate for Indiana extreme
rainfall data
– EV1 and LN are recommended
Conclusions (II)
• Dependence structure
– Between P and D, positive correlated
– Between D and I, generally negative correlated
– Between P and I, positive correlated
– Frank family is recommended
• Estimate of depth for known duration

– Similar results for durations larger than 10 hours
– AMI estimates should be better for durations less than
10 hours
– Bivariate approach is more flexible than univariate
approach
Thank you for listening.
Questions?

Statistical Analysis of Extreme Rainfall Events Over Indiana, USA

Uploaded by

Copyright:

Available Formats

Statistical Analysis of Extreme Rainfall Events Over Indiana, USA

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Statistical Analysis of Extreme Rainfall Events Over Indiana, USA

Uploaded by

Copyright:

Available Formats

World Environmental & Water Resources Congress 2007

Statistical Analysis of Extreme

Shih-Chieh Kao & Rao S. Govindaraju

ex: exponential margins ?

• Correlation coefficient can not correctly describe association

ρ=0.85 ρ=0.85 ρ=0.85

– Provides a complete description of dependence structure

• Return period assessment using bivariate model

• Probablistic structure of storm surface runoff

• Rainfall frequency analysis

• Unanswered questions in rainfall frequency analysis

– Peak Intensity, I (mm/hour) # #

• Parameters estimated primarily by maximum

• EV1, GEV, LP3, LN provided better fit. GP provided

• Non-parametric procedure for estimating

• The distribution function KC(t) provides the strictest

• Clayton and Ali-Mikhail-Haq families performed well for

• Frank family of Archimedean copulas

• Examples using Frank family and EV1 marginals

• Given return period T, the T-year, d-hour rainfall

• Comparison between bivariate and univariate

• Similar trends were observed for durations greater than

• Estimate of depth for known duration

You might also like