water

Article
Evaluation of Rainfall Temporal Distribution
Models with Annual Maximum Rainfall Events in
Seoul, Korea
Wooyoung Na and Chulsang Yoo *
School of Civil, Environmental and Architectural Engineering, College of Engineering, Korea University,
Seoul 02841, Korea; uoo921227@gmail.com
* Correspondence: envchul@korea.ac.kr; Tel.: +82-10-9326-9168




Received: 17 September 2018; Accepted: 14 October 2018; Published: 17 October 2018


Abstract: This study evaluated five models of rainfall temporal distribution (i.e., the Yen and
Chow model, Mononobe model, alternating block method, Huff model, and Keifer and Chu
model), with the annual maximum rainfall events selected from Seoul, Korea, from 1961 to 2016.
Three different evaluation measures were considered: the absolute difference between the rainfall
peaks of the model and the observed, the root mean square error, and the pattern correlation coefficient.
Also, sensitivity analysis was conducted to determine whether the model, or the randomness of
the rainfall temporal distribution, had the dominant effect on the runoff peak flow. As a result,
the Keifer and Chu model was found to produce the most similar rainfall peak to the observed,
the root mean square error was smaller for the Yen and Chow model and the alternating block
method, and the pattern correlation was larger for the alternating block method. Overall, the best
model to approximate the annual maximum rainfall events observed in Seoul, Korea, was found to be
the alternating block method. Finally, the sensitivity of the runoff peak flow to the model of
rainfall temporal distribution was found to be much higher than that to the randomness of the
rainfall temporal distribution. In particular, in small basins with a high curve number (CN) value,
the sensitivity of the runoff peak flow to the randomness of the rainfall temporal distribution was
found to be insignificant.

Keywords: rainfall temporal distribution; annual maximum rainfall event; sensitivity analysis;
rainfall–runoff analysis

1. Introduction
Once the total rainfall depth (i.e., design rainfall) is determined as a design criterion for a given
return period and rainfall duration, it should be distributed temporally, to be applied in rainfall–runoff
analysis. The temporal distribution of design rainfall is an important factor that affects the result of the
rainfall–runoff analysis, and ultimately, the dimension of hydraulic structures. Thus, a proper model
for temporally distributing the total rainfall depth should be selected, by considering the observed
characteristics of the rainfall temporal distribution. The runoff peak flow and peak time are all affected
by this model of rainfall temporal distribution [1].
Rainfall temporal distribution models may be categorized into three groups: The first group
is composed of those derived by analyzing the observed rainfall data. The Huff model, the SCS
(Soil Conservation Service) model by the NRCS (National Resources Conservation Service), the Yen
and Chow model, the Mononobe model, and the Pilgrim and Cordery model are included in this
group [2–6]. The second group is based on the rainfall intensity-duration-frequency (I-D-F) relation.
The alternating block model, Kiefer and Chu model, and Lee and Ho model are among the models

Water 2018, 10, 1468; doi:10.3390/w10101468 www.mdpi.com/journal/water

Water 2018, 10, 1468 2 of 23

in this group [7–9]. The remaining models are mostly derived based on stochastic or probabilistic
approaches [10–13].
Selecting one of these models of rainfall temporal distribution for application to hydrological
practices is a matter of choice. For example, a dimensionless temporal distribution based on the Huff
model is provided by the NOAA (National Oceanic and Atmospheric Administration) in the US [14];
this is the same in China, Singapore and South Korea [15–17]. Some countries select the model based on
the I-D-F relation, like the Chicago model in New Zealand and Italy [18,19] and the alternating block
model in Philippines [20]. Hong Kong also used a similar method that uses the rainfall intensity
information for given rainfall durations [21]. The observed temporal distribution of an extreme event
is also directly considered in hydrologic analysis in Japan [22].
Verification of the selected rainfall temporal distribution model is generally carried out by
comparing it with the observed rainfall events [23–27]. For example, Aron et al. (1987) evaluated
the SCS model with the observed rainfall data in Pennsylvania to confirm its applicability [24].
A similar result was also found by Guo and Hargardin (2009) [25]. Nguyen et al. (2010) compared
the eight models of rainfall temporal distribution to select the best one for Quebec [27]. In Korea,
MOCT (Ministry of Construction and Transportation) (2000) conducted a comparison study with the
Yen and Chow model, the Mononobe model, the alternating block model, the Huff model and the
Keifer and Chu model and recommended the use of the Huff model in hydrological practices [28].
MLTM (Ministry of Land, Transport and Maritime Affairs) (2011) also confirmed the recommendation
by MOCT (2000) [29]. Additionally, MLTM (2011) suggested that the Huff model should be derived
by analyzing the rainfall events with durations longer than, or equal to, six hours [29]. It was also
suggested to consider only those rainfall events whose rainfall peaks are located in the most frequent
quantiles. The second or third quantile is the most frequently selected in design practice.
However, the Huff method is known to over-smooth the rainfall distribution, even though
the concept of the most frequent quantile is applied. Compared to the other models of rainfall
temporal distribution, the runoff peak from the Huff model is generally lower by 10% or more [30–33].
This problem becomes even more severe when the rainfall duration is short [34]. The effect of
infiltration is also different, depending on the quantiles selected. The effect of infiltration is the
largest for the first quantile of the Huff model, and the smallest for the fourth quantile of the model.
The difference is directly reflected in the runoff peak flow and peak time. In fact, MLTM (2011) also
advised choosing the third quantile, to avoid possible confusion in the selection of the most frequent
rainfall quantiles [29].
This study also focused on the evaluation of the models of rainfall temporal distribution. The five
models considered in MOCT (2000) and MLTM (2011) [28,29] were the Yen and Chow model,
the Mononobe model, the alternating block model, the Huff model and the Keifer and Chu model
were also considered. For the comparison, only the annual maximum rainfall events were used in
this study, as the rainfall distribution model is mostly used for hydrological practices. The annual
maximum rainfall events used in this study are those selected from Seoul, Korea from 1961 to
2016, based on bivariate frequency analysis [35]. The model and observed rainfall distribution
were compared by considering the location of the rainfall peak and the randomness of the rainfall
distribution. The sensitivity of the runoff peak flow to the model and randomness of the rainfall
temporal distribution were also analyzed. The question was addressed of whether the model or the
randomness of the rainfall temporal distribution had the dominant effect on the runoff peak flow.
Finally, as a result, the best model for the rainfall temporal distribution was suggested.

2. Models of Rainfall Temporal Distribution

2.1. The Yen and Chow Model

Yen and Chow (1980) proposed a triangular-shaped temporal distribution model for the design
rainfall [6]. For a basin with the design rainfall depth R (mm) and rainfall duration T (h), the rainfall
Water 2018, 10, 1468 3 of 23

temporal distribution (design hyetograph) is made as a triangle of area R and height h = 2R/T (mm/h).
The peak time of rainfall can vary, which is quantified by the storm advancement coefficient.
The coefficient of the storm advancement (r) is defined as the ratio between the rainfall peak time
and the rainfall duration. If r = 0.5, the rainfall peak is located at the center of the rainfall event,
which is called a centered type. If r < 0.5, the rainfall event is called an advanced type, and if r > 0.5,
a delayed type.

2.2. The Mononobe Model

Mononobe proposed a model to distribute the design rainfall depth by analyzing the rainfall
data observed in Korea [4]. This method is based on the following Equation (1), which shows the
cumulative rainfall distribution up to time t:
 n
R T
Rt = T t (1)
T t

where RT is the design rainfall depth (mm), Rt is the cumulative rainfall depth (mm) up to time t,
and T is the rainfall duration (h). The constant n is assumed to be 2/3, in general. As Equation (1) is
the cumulative rainfall depth, the rainfall intensity data to make the rainfall temporal distribution
is derived as the difference between the current and previous time cumulative rainfall depth.
After locating the rainfall peak, the next largest rainfall intensity is located alternately around the
rainfall peak in turn.

2.3. The Alternating Block Method

The alternating block method is a method to make the rainfall temporal distribution
(design hyetograph) using the rainfall intensity-duration-frequency (I-D-F) curve. After dividing
the rainfall duration T (h) by the ∆t time-interval, the rainfall intensities for the rainfall durations ∆t,
2∆t, 3∆t, . . . are estimated from the rainfall I-D-F curve. By calculating the product of the rainfall
intensity and the duration (i.e., the rainfall depth) for each rainfall duration, the cumulative rainfall
distribution can be derived. The rainfall intensity data can then be obtained as the difference between
the successive cumulative rainfall depths. After locating the rainfall peak at the center, the next largest
rainfall intensity is located alternately to the right and left of the rainfall peak in turn, similar to the
centered type Mononobe model.

2.4. The Huff Model

Huff (1967) analyzed the rainfall data collected over small basins in Illinois with a basin area of
less than 400 mi2 , and proposed a model of rainfall temporal distribution [3]. This model was to
choose the proper quantile by considering the frequency of the rainfall peak over the rainfall duration.
Then, using only those rainfall events with peaks located at the same quantile, the rainfall temporal
distribution is derived as an average rainfall distribution. This resulting rainfall temporal distribution
is smooth to reflect the average rainfall distribution with time, which does not exhibit the burst
characteristics of observed storms. Usually, the rainfall temporal distribution for 10, 20, . . . , 90 percent
cumulative probabilities of occurrence are derived with the percentage of total storm rainfall for
10 percent increments of the storm duration. The 50 percent histogram represents a cumulative rainfall
pattern that should be exceeded in about half of all storms. The 90 percent histogram can be interpreted
as a storm distribution that is equaled or exceeded in 90 percent or less of all storms.
Water 2018, 10, 1468 4 of 23

When deriving the rainfall temporal distribution model, the observed rainfall events are
transformed into dimensionless ones by the total rainfall depth and rainfall duration, i.e.,

T (i )
PT (i ) = × 100 (2)
T
R (i )
PR(i ) = × 100 (3)
R
where PT(i) is the percentage of time T(i) over the rainfall duration T, and PR(i) is the percentage of
the cumulative rainfall depth R(i) up to the time T(i) over the total rainfall depth R during the rainfall
duration T. Generally, 10 time-intervals are considered (i.e., i = 1, . . . , 10).

2.5. The Keifer and Chu Model

Keifer and Chu (1957) developed a synthetic hyetograph of the instantaneous intensity method
type for use in sewer design for Chicago [8]. The Keifer and Chu model is also based on the rainfall
I-D-F curve. That is, similarly to the alternating block method, the rainfall depth around the rainfall
peak for the rainfall duration td is assumed to be the same as that from the rainfall I-D-F curve.
However, the rainfall intensity in the Keifer and Chu model is the instantaneous value changing
continuously over the rainfall duration. That is, the difference of the Keifer and Chu model from
the alternating block method is that the rainfall intensity is now considered to vary continuously
throughout the storm. Hence, it is also called the instantaneous intensity model. The rainfall intensities
before and after the rainfall peak time (ia and ib , respectively) are expressed as follows:

c (1 − e)(t a /r )e + f
 
ia = 2 (4)
(t a /r )e + f


c (1 − e)[tb /(1 − r )]e + f


ib = 2
(5)
[tb /(1 − r )]e + f


where ta is the alternating time before the rainfall peak time, tb is the alternating time after the rainfall
peak time, and r is the coefficient of storm advancement; and c, e, f are coefficients that vary with
the return period. Also, these alternating rainfall intensities before and after the rainfall peak are
assumed to make a curve.

3. Data and Model Fitting

3.1. Data
This study considered the annual maximum rainfall event data collected in Seoul, Korea, derived
by Park and Yoo (2012) [36]. These annual maximum rainfall events were selected based on a bivariate
frequency analysis of all the independent rainfall events that occurred in a given year. In Park
and Yoo (2012), this bivariate frequency analysis was repeated every year from 1961 to 2010 [36],
and in this study, additionally from 2011 to 2016. As a result, this study could consider 56 annual
maximum rainfall events that covered the period from 1961 to 2016. Figure 1 compares the rainfall
histogram of those annual maximum rainfall events. All of these 56 rainfall events are different from
each other. Most of them have one rainfall peak, but some have two. Also, the location of the peak
varies a lot from the beginning to the end of the rainfall event. This diversity indicates that the rainfall
distribution model may not easily be evaluated with the observed rainfall events.
Water 2018, 10, 1468 5 of 23

Figure 1. Cont.
Water 2018, 10, 1468 6 of 23

Figure 1. Histograms of annual maximum rainfall events observed from 1961 to 2016 in Seoul, Korea.

Table 1 compares the selected annual maximum rainfall events with their rainfall durations, mean
rainfall intensities, and total rainfall depths. Table 1 shows that the range of rainfall duration was wide,
from 1 to 104 h. The largest total rainfall depth 630 mm was recorded in 2011, and the smallest one was
just 30 mm in 2007. The mean rainfall intensity also showed a wide range from 2.7 mm/h in 1981 to
38.9 mm/h in 2004. The mean rainfall duration was estimated to be 24 h, the mean rainfall intensity
11.4 mm/h, and the mean total rainfall depth 154.7 mm.
Water 2018, 10, 1468 7 of 23

Table 1. Characteristics of the annual maximum rainfall events collected from 1961 to 2016 in
Seoul, Korea.

Rainfall Rainfall Rainfall Rainfall Rainfall Rainfall

Year Duration Depth Intensity Year Duration Depth Intensity
(h) (mm) (mm/h) (h) (mm) (mm/h)
1961 3 55.9 18.6 1989 12 82.7 6.9
1962 18 62.0 3.4 1990 35 352.2 10.1
1963 2 45.8 22.9 1991 35 138.1 3.9
1964 6 127.4 21.2 1992 9 129.5 14.4
1965 27 87.0 3.2 1993 3 75.0 25.0
1966 65 361.7 5.6 1994 4 68.4 17.1
1967 4 43.1 10.8 1995 93 331.8 3.6
1968 17 156.7 9.2 1996 16 116.5 7.3
1969 4 83.7 20.9 1997 13 126.3 9.7
1970 20 191.8 9.6 1998 25 359.7 14.4
1971 8 132.8 16.6 1999 75 500.3 6.7
1972 25 394.5 15.8 2000 2 42.9 21.5
1973 18 59.5 3.3 2001 22 180.3 8.2
1974 18 87.0 4.8 2002 54 227.9 4.2
1975 17 112.1 6.6 2003 41 159.5 3.9
1976 53 226.3 4.3 2004 1 38.9 38.9
1977 52 168.2 3.2 2005 18 52.8 2.9
1978 44 244.3 5.6 2006 80 281.3 3.5
1979 4 88.0 22.0 2007 2 30.0 15.0
1980 18 116.0 6.4 2008 44 192.2 4.4
1981 58 156.2 2.7 2009 10 129.8 13.0
1982 22 80.2 3.6 2010 10 83.5 8.4
1983 1 31.4 31.4 2011 104 630.0 6.1
1984 39 317.5 8.1 2012 4 76.5 19.1
1985 3 66.0 22.0 2013 2 45.0 22.5
1986 8 106.2 13.3 2014 4 42.5 10.6
1987 29 323.8 11.2 2015 29 119.5 4.1
1988 14 67.8 4.8 2016 3 58.0 19.3

3.2. Model Fitting

All of the rainfall temporal distribution models considered in this study—the Yen and Chow
model, Mononobe model, alternating block method, Huff model, and Keifer and Chu model—were
fitted to each of the annual maximum rainfall events observed in Seoul, Korea. Six rainfall events were
excluded in this application, as their durations were less than three hours. The third quantile was
considered for the Huff model, which is recommended in Korea [29]. The third quantile is also the
most probable quantile of the observed annual maximum rainfall events.
Somewhat different information was used for the fitting of each rainfall temporal distribution
model. For example, the Yen and Chow model and Mononobe model require the total rainfall depth,
but the alternating block method and Keifer and Chu model require the rainfall intensity from the
rainfall I-D-F curve. The estimated return period of a rainfall event is also needed to handle the rainfall
I-D-F curve. In this study, the rainfall I-D-F curve developed for the Seoul rain gauge station was used
for this purpose [29]. For example, the return period of the annual maximum rainfall in 1984 was
estimated to be about 11 years, and that in 2001 to be about three years.
With the estimated return period along with the total rainfall depth and rainfall duration,
the rainfall temporal distribution models could be fitted for each annual maximum rainfall event.
For example, Figure 2 shows the fitting results of the models to the annual rainfall events in 1984 and
2001. This figure shows that, basically, the peak time of a model is not coincident with that of the
observed. This result is quite natural, considering the randomness of the peak time. Concerning the
rainfall peak, the Huff model derived a smaller rainfall peak than the observed, which was also the
same for the Yen and Chow model. The rainfall peak of the alternating block method and that of the
 With the estimated return period along with the total rainfall depth and rainfall duration, the rainfall
temporal distribution models could be fitted for each annual maximum rainfall event. For example,
Figure 2 shows the fitting results of the models to the annual rainfall events in 1984 and 2001. This figure
shows that, basically, the peak time of a model is not coincident with that of the observed. This result
is quite natural, considering the randomness of the peak time. Concerning the rainfall peak, the Huff
Water 2018, 10, 1468 8 of 23
model derived a smaller rainfall peak than the observed, which was also the same for the Yen and
Chow model. The rainfall peak of the alternating block method and that of the Keifer and Chu model
were found
Keifer and Chu to be quite were
model similar to the
found to observed, while that
be quite similar ofobserved,
to the the Mononobe
whilemodel
that ofwas found to be
the Mononobe
much higher
model was found thanto
the
beobserved.
much higher than the observed.

(a) 1984 (b) 2001

Figure 2.
Figure 2. Comparison
Comparison of of the
the rainfall
rainfall temporal
temporal distribution
distribution models
modelsand
andobserved
observedannual
annualmaximum
maximum
rainfall events
rainfall events in
in 1981
1981 (a)
(a) and
andin in2004
2004(b)(b)(Case
(Case1).
1).

4. Evaluation of the Rainfall Temporal Distribution Models

4.1. Evaluation Methods and Evaluation Measures

In this part of the study, the rainfall temporal distribution models were evaluated by comparing
them to the observed annual maximum rainfall events in Seoul, Korea. This study considered three
different cases for the evaluation. The first case (Case 1) was to compare the model and the observed
rainfall event as it is. The second case (Case 2) was to compare the model and observed rainfall event,
after reconciling the rainfall peak time of the specific model to the observed. That is, by moving the
rainfall temporal distribution of the model in parallel, it is possible to fix the rainfall peak time of
the model to the observed. The third case (Case 3) was to compare the model and observed rainfall
event, after rearranging the observed rainfall temporal distribution to be most similar to the model.
In this case, the randomness of the observed rainfall temporal distribution is removed, to make its
temporal distribution most approximate that of the specific model.
In fact, the rearrangement of the rainfall temporal distribution of the third case may cause a
distortion of the autocorrelation structure of an observed rainfall event [37]. However, after the
rearrangement, mean autocorrelation coefficients of observed rainfall events and rainfall temporal
distribution models were found to be similar to each other. That is, the mean autocorrelation
coefficient of all the rainfall events considered in this study was about 0.35, but became about 0.65
after the rearrangement. The mean autocorrelation coefficient of the five rainfall temporal distribution
models considered in this study was estimated to be about 0.56. The effect of randomness in the rainfall
temporal distribution will also be evaluated in Section 5.3 with an emphasis on the runoff calculation.
Three different evaluation measures were considered in this study. The first measure was the
absolute difference between the rainfall peaks of the model and the observed (DRpeak ), which is defined
as follows:
DR peak = S p − O p (6)

where Sp is the rainfall peak of the model, and Op is the observed. The second measure is the root
mean square error (RMSE). This RMSE can be assumed as the standard deviation of the differences
Water 2018, 10, 1468 9 of 23

between the model and the observed rainfall temporal distribution. The RMSE is calculated using the
following equation: v
u n
u ∑ ( S i − Oi ) 2
u

RMSE = i=1
t
(7)
n
where Si is the rainfall intensity at time i of the model, and Oi is the observed rainfall intensity at time i.
Finally, the third measure is the pattern correlation coefficient (R). This pattern correlation
coefficient is a correlation coefficient to measure the similarity of the pair data. The pattern correlation
coefficient is calculated as follows:
n
∑ S i − S Oi − O

R = s i =1 s (8)
n
2 n
2
∑ Si − S ∑ Oi − O
i =1 i =1

where S and O represent the means of the model and the observed rainfall temporal distribution.

4.2. Evaluation Results

As mentioned in the previous section, three different cases were considered for the evaluation of
the rainfall temporal distribution models. Figure 2 is for Case 1 for the annual maximum rainfall
events
Water observed
2018, 10, x in 1984 and 2001. Case 2 and Case 3 are presented in Figures 3 and 4, respectively.
9 of 22

(a) 1984 (b) 2001

Figure 3.
Figure 3. As for
for Figure
Figure 2,
2, but
butfor
forCase
Case2.2.

Table 2 summarizes the evaluation results with the annual maximum rainfall events in 1984 and
2001. For each model, DRpeak should be estimated to be the same for Cases 1, 2 and 3. Additionally, a
positive (or negative) sign was given to the value of DRpeak to distinguish the higher (or lower) rainfall
peak of a model from the observed. As the rainfall peak of the Yen and Chow model and the Huff
model was mostly smaller than the observed, a negative sign was given to the estimates of DRpeak .
On the other hand, a positive sign was given to the results of the Mononobe model application.
 (a) 1984 (b) 2001
Water 2018, 10, 1468 10 of 23
Figure 3. As for Figure 2, but for Case 2.

(a) 1984 (b) 2001

Figure 4. As
Figure 4. As for
for Figure
Figure2,2,but
butfor
forCase
Case3.3.

The RMSEs
Table for the application
2 summarizes of the
the evaluation alternating
results with the block method
annual maximumand the Keiferevents
rainfall and Chu model
in 1984 and
were For
2001. estimated to be smaller
each model, than for
DRpeak should bethe other model
estimated to be applications. As can1,be
the same for Cases expected,
2 and the RMSE a
3. Additionally,
for Case(or
positive 1 was the largest,
negative) andgiven
sign was that to
forthe
Case 3 was
value of DRthepeaksmallest. This result
to distinguish was (or
the higher alsolower)
the same for peak
rainfall the
pattern correlation coefficient. The highest value of the pattern correlation coefficient
of a model from the observed. As the rainfall peak of the Yen and Chow model and the Huff model was was obtained
in the application
mostly smaller than ofthe
theobserved,
alternating block method
a negative sign wasandgiven
the Keifer
to theand Chu model,
estimates and
of DRpeak thethe
. On lowest
other
value in the application of the Mononobe model. Overall, among
hand, a positive sign was given to the results of the Mononobe model application. the five models considered in this
study,
ThetheRMSEs
alternating
for theblock method and
application thealternating
of the Keifer and blockChu model
method wereandfound to be most
the Keifer and Chusimilar to
model
the observed.
were estimated to be smaller than for the other model applications. As can be expected, the RMSE for
Case Table
1 was2. the largest,results
Evaluation and that for Case
of rainfall 3 was
temporal the smallest.
distribution modelsThis
withresult was also
the maximum the same
rainfall eventsfor the
pattern correlation
observed in 1984coefficient.
and 2001. The highest value of the pattern correlation coefficient was obtained
in the application of the alternating block method and the Keifer and Chu model, and the lowest
value inEvaluation
the application Yen and Alternating Keifer considered in this
Year of the Mononobe
Case model.Mononobe
Overall, among the five models Huff
Measure Chow Block and Chu
study, the alternating block method and the Keifer and Chu model were found to be most similar to
the observed.
DRpeak 1984 - −29.8 67.8 33.0 29.0 −28.5
2001 - −42.7 24.4 −4.0 −10.0 −41.3
1 9.45 16.13 11.73 13.17 11.47
1984 2 9.51 12.86 6.57 8.44 9.48
3 8.30 12.20 6.32 7.53 8.13
RMSE
1 14.28 21.34 16.27 14.34 15.79
2001 2 14.05 11.89 10.55 4.81 14.62
3 13.33 9.93 6.57 3.35 13.05
1 0.58 0.45 0.56 0.58 0.24
1984 2 0.59 0.70 0.83 0.86 0.61
3 0.77 0.73 0.88 0.90 0.79
R
1 0.46 0.14 0.33 0.52 0.19
2001 2 0.45 0.73 0.73 0.92 0.50
3 0.64 0.82 0.93 0.99 0.50
Water 2018, 10, 1468 11 of 23

To summarize the evaluation results for all annual maximum rainfall events, DRpeak and RMSE
were standardized. That is, the DRpeak was divided by the observed rainfall peak (Opeak ) of the rainfall
event to create the standardized SDRpeak , and the RMSE was divided by the mean rainfall intensity
(I) to make the standardized SRMSE. That is,

DR peak
SDR peak = (9)
O peak

RMSE
SRMSE = (10)
I
Table 3 summarizes the SDRpeak , SRMSE, and R for all 50 annual maximum rainfall events by
their mean and standard deviation. Similarly to the DRpeak , the SDRpeak of each rainfall event was
estimated to be the same for all cases, but the SRMSE and R were estimated differently for each case.
The sign of SDRpeak was determined by following the same rule as for DRpeak .

Table 3. Evaluation results of rainfall temporal distribution models with all the annual maximum
rainfall events considered in this study (mean value along with standard deviation inside the bracket).

Evaluation Yen and Alternating Keifer and

Case Mononobe Huff
Measure Chow Block Chu

SDRpeak -
−0.499 0.706 0.548 0.473 −0.467
(0.240) (0.593) (0.453) (0.419) (0.245)
1.337 1.894 1.810 2.012 1.398
1
(0.592) (0.731) (0.793) (1.000) (0.572)
SRMSE 1.242 1.299 1.296 1.592 1.189
2
(0.628) (0.629) (0.717) (0.986) (0.616)
0.955 0.914 0.754 1.135 1.039
3
(0.557) (0.310) (0.386) (0.794) (0.544)
0.268 0.208 0.238 0.304 0.170
1
(0.337) (0.264) (0.259) (0.266) (0.350)
R 0.508 0.733 0.732 0.610 0.541
2
(0.304) (0.173) (0.196) (0.252) (0.285)
0.751 0.804 0.877 0.743 0.629
3
(0.171) (0.126) (0.100) (0.201) (0.207)

As a result, the SDRpeak shows that the Yen and Chow model and the Huff model tend to
underestimate the rainfall peak, while the Mononobe model tends to overestimate. Among the five
models considered in this study, the Keifer and Chu model produced the most similar rainfall peak to
the observed. On the other hand, the smallest SRMSEs were estimated for the Yen and Chow model and
the alternating block method. The value of R was larger for the alternating block method, but smaller
for the Yen and Chow model and the Huff model. In particular, in Case 3, the alternating block method
showed the smallest SRMSE and largest R, which indicates that it is the most similar model to the
observed rainfall temporal distribution.

5. Sensitivity of the Runoff Peak to the Rainfall Temporal Distribution

5.1. Preparation of the Rainfall–Runoff Model

In this part of the study, the sensitivity of the runoff peak to the rainfall temporal distribution was
evaluated. The sensitivity analysis was repeated two times; once on the models of the rainfall temporal
distribution, and once on the randomness of the rainfall temporal distribution. Here, the randomness of
the rainfall distribution indicates the transformation of the rainfall temporal distribution from Case 3 to
Case 2 or Case 1. In contrast to the very well-organized design for rainfall based on the rainfall
distribution, and once on the randomness of the rainfall temporal distribution. Here, the randomness
of the rainfall distribution indicates the transformation of the rainfall temporal distribution from Case 3 to
Case 2 or Case 1. In contrast to the very well-organized design for rainfall based on the rainfall temporal
distribution Water
model,
2018, 10,the
1468 observed rainfall event shows very random organization 12 of of
23 the rainfall

intensity values over the rainfall duration.

This study used distribution
temporal the 1-h unit hydrograph
model, the observed (UH) forevent
rainfall the rainfall–runoff
shows very random analysis. TheofUH
organization the was derived
rainfall intensity values over the rainfall duration.
using the Clark instantaneous unit hydrograph (IUH), whose two parameters—the concentration time Tc
This study used the 1-h unit hydrograph (UH) for the rainfall–runoff analysis. The UH
and storage coefficient
was derived usingK—were assumed
the Clark to beunit
instantaneous thehydrograph
same. Three different
(IUH), whose twoUHs were derived with the
parameters—the
concentrationconcentration
time 1, 3,time andTc5and
h. storage
The basin areaK—were
coefficient was also assumed
assumed to be theto be Three
same. linearly proportional
different UHs to the
were derived with the concentration time 1, 3, and 5 h. The basin area was also assumed to be linearly
concentration time. Figure 5 compares the three UHs. In particular, it shows that the peak flow is not
proportional to the concentration time. Figure 5 compares the three UHs. In particular, it shows that
proportional theto peak
the basin area.
flow is not proportional to the basin area.

Figure 5. Comparison of 1-h unit hydrographs derived for the three artificial basins.
Figure 5. Comparison of 1-h unit hydrographs derived for the three artificial basins.
The infiltration loss was considered in the rainfall–runoff analysis by the NRCS-CN (National
The infiltration Conservation
Resources Service-Curve Number)
loss was considered method [38]. Three different
in the rainfall–runoff analysis CN by
values
theof NRCS-CN
60, 80 and (National
100 were considered, to consider the different land cover and soil characteristics. The higher the CN
Resources Conservation Service-Curve Number) method [38]. Three different CN values of 60, 80 and
value, the larger the effective rainfall amount. The CN 100 indicates no infiltration, while the CN value
100 were considered, toeffective
0 indicates no consider the different land cover and soil characteristics. The higher the CN
rainfall.
value, the larger the effective rainfall amount. The CN 100 indicates no infiltration, while the CN
5.2. Sensitivity to the Rainfall Temporal Distribution Models
value 0 indicates no effective rainfall.
Each model considered in this study derives a different rainfall temporal distribution, which
is also related to the shape of the runoff hydrograph. Additionally, the infiltration characteristics
5.2. Sensitivity toCN),
(i.e., the Rainfall Temporal
as well as the Distribution
runoff characteristics (i.e.,Models
Tc and K), will also change the shape of the runoff
hydrograph. To evaluate this difference caused by applying a different model of rainfall temporal
Each model considered
distribution, inperformed
this study this study derives a different
the rainfall–runoff rainfall
analysis with temporal
different basin and distribution,
infiltration which is
characteristics. The annual maximum rainfall event that occurred in 1997 was considered
also related to the shape of the runoff hydrograph. Additionally, the infiltration characteristics (i.e., CN), as an
example case. This rainfall event was continued for 13 h, and the mean rainfall intensity was 9.7 mm/h.
as well as theThe
runoffreturncharacteristics (i.e.,
period of this rainfall Tc and
event K), willtoalso
was estimated change
be two years. the shape of the runoff hydrograph.
To evaluate thisWith difference caused
the mean rainfall by applying
intensity a different
and the rainfall model
duration of the rainfallof rainfall
event in 1997, temporal
five different distribution,
rainfall temporal
this study performed distributions were derived.
the rainfall–runoff Also,with
analysis by considering
different three different
basin and CNinfiltration
values and threecharacteristics.
different UHs, a total of nine combinations could be prepared for each model of rainfall temporal
distribution. That is, for each UH, 15 combinations of rainfall distributions and CNs were prepared.
Table 4 summarizes these combinations for the rainfall–runoff analysis.
of this rainfall event was estimated to be two years.
With the mean rainfall intensity and the rainfall duration of the rainfall event in 1997, five
different rainfall temporal distributions were derived. Also, by considering three different CN values
and three different UHs, a total of nine combinations could be prepared for each model of rainfall
Water 2018, 10, 1468 13 of 23
temporal distribution. That is, for each UH, 15 combinations of rainfall distributions and CNs were
prepared. Table 4 summarizes these combinations for the rainfall–runoff analysis.
Table 4. Combinations of five rainfall temporal distribution models, three CN values and three Tc (=K)
Table
values4.considered
Combinations
in thisofstudy.
five rainfall temporal distribution models, three CN values and three Tc
(=K) values considered in this study.
Model CN Tc = K (h)
Model CN Tc = K (h)
Yen and Chow
Yen and Chow 60 1
Mononobe 60 1
Mononobe
Alternating Block 80 3
Alternating Block 80 3
Keifer and Keifer
Chu and Chu
100100 5 5
Huff
Huff

First,
First,Figure
Figure66compares
compares thethe 15
15 histograms
histograms of effective rainfallwith
effective rainfall withthe
thedifferent
differentrainfall
rainfalltemporal
temporal
distributionsand
distributions andCNs.
CNs.As Ascan
canbebeexpected,
expected, the
the effect
effect ofof the
the CNsCNs is is concentrated
concentrated inin
thethe early
early part
part of of
the
the rainfall,
rainfall, but
but the the effective
effective rainfallrainfall peak
peak was was
also also decreased
decreased a bit
a bit with withCN.
higher higher CN. Second,
Second, Figure 7 Figure
compares7
compares
the the runoff hydrographs
runoff hydrographs derived by
derived by applying theapplying
1-h UH.theThis1-h UH.shows
figure This figure shows
that the thathydrographs
runoff the runoff
hydrographs
were wereaffected
considerably considerably
by theaffected
CN valueby the
andCNthevalue andrunoff
UH. The the UH. The runoff hydrographs
hydrographs wereas
were all derived
all derived as
would be expected.would be expected.

Yen and Chow Mononobe Alternating block Keifer and Chu Huff
(a) CN = 60

Yen and Chow Mononobe Alternating block Keifer and Chu Huff
(b) CN = 80

Yen and Chow Mononobe Alternating block Keifer and Chu Huff
(c) CN = 100

Figure
Figure6.6.Comparison
Comparisonof ofeffective
effective rainfall
rainfall temporal distributions
distributionsof
ofmodel
modeland
andthe
theannual
annualmaximum
maximum
rainfallevent
rainfall eventin
in1997.
1997.

The key result of Figure 7 lies in the effect of the model of rainfall temporal distribution. First, the
Yen and Chow model and the Huff model produced a far smaller peak flow than the other models.
In particular, the sensitivity of the peak flow to the model of rainfall temporal distribution was higher
for the case with small Tc and K. For the case with larger Tc and K, the sensitivity to the model
was found to be much smaller. The peak flow derived by applying the Mononobe model was the
highest, but was also similar to that of the Keifer and Chu model.
Water 2018, 10, 1468 14 of 23
Water 2018, 10, x 13 of 22

Tc, K = 1 h Tc, K = 3 h

Tc, K = 5 h
(a) CN = 60

Tc, K = 1 h Tc, K = 3 h

Tc, K = 5 h
(b) CN = 80

Figure 7. Cont.
Water 2018, 10, 1468 15 of 23
Water 2018, 10, x 14 of 22

Tc, K = 1 h Tc, K = 3 h

Tc, K = 5 h
(c) CN = 100

Figure7.7.Comparison
Figure Comparison of
ofrunoff
runoff hydrographs
hydrographs derived
derived with
with different
different rainfall
rainfall temporal
temporaldistribution
distribution
models for the annual maximum rainfall event in
models for the annual maximum rainfall event in 1997.1997.

Theabove
The key result of can
results Figure
also7 be
liesevaluated
in the effect of the modelfrom
quantitatively of rainfall temporal
the point of thedistribution.
sensitivity ofFirst, the
the peak
flow to the model of rainfall temporal distribution. For this purpose, the following measure, a ratioInof
Yen and Chow model and the Huff model produced a far smaller peak flow than the other models.
particular,
the peak flow theofsensitivity
a model to of the
the mean
peak flow
of alltofive
the models
model of(Ratio
rainfall temporal distribution was higher for
peak ), was introduced:
the case with small Tc and K. For the case with larger Tc and K, the sensitivity to the model was found
to be much smaller. The peak flow derived by applying O peak
the Mononobe model was the highest, but
Ratio peak = (11)
was also similar to that of the Keifer and Chu model. O peak
The above results can also be evaluated quantitatively from the point of the sensitivity of the
peak
where O flow
peak to
is the peak
modelflowof rainfall temporal
of a model, and O distribution. For this purpose, the following measure, a
peak is the mean of the peak flows of the five models of
ratio of the peak flow of a model to the mean of all
rainfall temporal distribution. Figure 8 summarizes the results infive models (Ratio
boxpeak), was introduced:
plots.
This figure shows that the sensitivity of the peak flow to the model of rainfall temporal distribution
Opeak
became much larger in a small basin (where Ratio
the peak 
concentration time is short). The inter-quantile(11) range
estimated for the case of Tc = 1 h was estimated to beOmore peak than three times of that for the case of
Tcwhere
= 3 orO5peak
h.isOnthethe other
peak flowhand, the difference
of a model, and 𝑂 between the two
is the mean of cases
the peak of Tflows
c =3h ofand Tc =models
the five 5 h was
very small. temporal
of rainfall As the rainfall distribution
distribution. Figurewas made at hourly
8 summarizes intervals,
the results theplots.
in box effect of the rainfall peak on
the peak flow seems to be direct for the case
This figure shows that the sensitivity of the peak of T = 1 h. On the other
c flow to the model of rainfallhand, intemporal of longer Tc ,
the casedistribution
the storage
became effect
much significantly
larger in a smalldampened
basin (where thethe
effect of the rainfall
concentration timepeak.
is short). The inter-quantile range
CN also seemed to have a significant effect on the peak
estimated for the case of Tc = 1 h was estimated to be more than three times flow, even though
of thatthat effect
for the casewas
of Tnot
c = 3as
high as that of the concentration time. As the CN value increased, the inter-quantile
or 5 h. On the other hand, the difference between the two cases of Tc = 3 h and Tc = 5 h was very small. range of the box
plot
As increased,
the rainfall which was also
distribution obvious
was made for theintervals,
at hourly cases ofthe Tc effect
= 3 hofand theTrainfall
c = 5 h. peak Foronthe
thecase
peakof
Tcflow
= 1 seems
h, thetoeffect of CN
be direct seemed
for the case oftoTcbe
= 1minimum.
h. On the other Thishand,
resultin indicates
the case ofthat
longeras Tthe basin
c, the storagearea
increases, the sensitivity of the peak flow to the
effect significantly dampened the effect of the rainfall peak. model of rainfall temporal distribution becomes
smaller.CN More
also generally,
seemed to thehaveeffect of rainfall
a significant distribution
effect on the peak onflow,
the rainfall–runoff
even though thatmodel effect for
wasanot
small basin
as high
as that of the concentration time. As the CN value increased, the inter-quantile range of the box plot
increased, which was also obvious for the cases of Tc = 3 h and Tc = 5 h. For the case of Tc = 1 h, the
Water 2018, 10, x 15 of 22

effect of CN
Water 2018, seemed to be minimum. This result indicates that as the basin area increases, the sensitivity
10, 1468 16 of 23
of the peak flow to the model of rainfall temporal distribution becomes smaller. More generally, the
effect of rainfall distribution on the rainfall–runoff model for a small basin with small Tc and K can be
very small TcOn
with limited. and K other
the can behand,
very in
limited.
a largeOn thewith
basin other hand,
larger Tcin a large
and K, thebasin
effectwith larger
of CN Tc be
could and K,
seen
the effect of
more clearly. CN could be seen more clearly.

(a) CN = 60 (b) CN = 80

(c) CN = 100

Figure Comparisonofofthe
Figure 8. Comparison the box-plots
box-plots of the
of the runoff
runoff peakpeak
flowsflows derived
derived with different
with different CN andCN and Tc
Tc values.
values.
5.3. Preparation of the Rainfall–Runoff Model
5.3. Preparation of the
The result in Rainfall–Runoff
Section 4 shows thatModel
the alternating block method produced the most similar rainfall
temporal distribution
The result in Section to the observed.
4 shows that However, this result
the alternating blockwas limited,
method as the comparison
produced the most similarwas
carried out with the redistributed observed rainfall. That is, the comparison
rainfall temporal distribution to the observed. However, this result was limited, as the comparison did not consider the
randomness
was carried out of the
with observed rainfall distribution.
the redistributed Thus, in
observed rainfall. thisis,
That part
theofcomparison
the study, thedidsensitivity
not considerof the
the
peak flow to the randomness of the rainfall temporal distribution was evaluated.
randomness of the observed rainfall distribution. Thus, in this part of the study, the sensitivity of the As an example case,
the same
peak flowrainfall event in the of
to the randomness previous section
the rainfall was considered.
temporal distribution was evaluated. As an example case,
First, a total of 10 rainfall temporal distributions
the same rainfall event in the previous section was considered. were randomly generated using the rainfall
distribution derived by applying the alternating
First, a total of 10 rainfall temporal distributions were block method (Figuregenerated
randomly 9). Exceptusing for the
theoriginal
rainfall
distribution (H1), all the others seem more realistic, and
distribution derived by applying the alternating block method (Figure 9). Except for thesimilar to the observed. Theoriginal
same
rainfall–runoff
distribution analysis
(H1), all the wasothersalso repeated
seem for each and
more realistic, combination
similar tooftheTcobserved.
, K and CN. TheThat
sameis, rainfall–runoff
nine different
combinations of T
analysis was also repeatedc , K, and CN were considered in the rainfall–runoff analysis for one
for each combination of Tc, K and CN. That is, nine different combinations rainfall temporal
distribution.
of Tc, K, and CN Figure
were10considered
compares the in theresulting runoff hydrographs.
rainfall–runoff analysis for one rainfall temporal distribution.
The hydrographs in Figure 10
Figure 10 compares the resulting runoff hydrographs. show several important factors to be considered in the analysis of
the result. First, whenin
The hydrographs evaluating
Figure 10 the showsensitivity to the randomness
several important factors to ofbe the rainfall distribution,
considered in the analysis the
of
effect of T c (or the size of the basin area) is not that important. This result can
the result. First, when evaluating the sensitivity to the randomness of the rainfall distribution, the effect be seen more clearly
in Tthe
of case of CN = 100. Second, particularly for the rainfall distribution with its peak at the rainfall
c (or the size of the basin area) is not that important. This result can be seen more clearly in the case

of CN = 100.the
beginning, effect particularly
Second, of CN is important. This isdistribution
for the rainfall obvious, aswith mostitsofpeak
the infiltration
at the rainfalloccurs at the
beginning,
beginning
the effect ofof CNa rainfall.
is important.However,
This isifobvious,
we consider onlyofthe
as most therainfall events
infiltration withattheir
occurs peaks on the
the beginning of a second,
rainfall.
third or fourth quantile, which is believed to result in some severe floods,
However, if we consider only the rainfall events with their peaks on the second, third or fourth quantile,the effect of CN may be
excluded in the analysis of the result. Finally, it should be mentioned that the runoff peak time is
which is believed to result in some severe floods, the effect of CN may be excluded in the analysis of
directly related to the rainfall peak time. Thus, in this study, only the runoff peak flow was analyzed.
the result. Finally, it should be mentioned that the runoff peak time is directly related to the rainfall
peak time. Thus, in this study, only the runoff peak flow was analyzed.
Water 2018, 10, 1468 17 of 23
Water 2018, 10, x 16 of 22

H1 H2 H3 H4 H5

H6 H7 H8 H9 H10
(a) CN = 60

H1 H2 H3 H4 H5

H6 H7 H8 H9 H10
(b) CN = 80

H1 H2 H3 H4 H5

H6 H7 H8 H9 H10
(c) CN = 100

Figure 9.
Figure 9. Randomly
Randomly distributed rainfall
rainfall histograms
histograms derived from
from the
the alternating
alternating block
block model
model for
for the
the
annual maximum
annual maximum rainfall
rainfall event
event in
in 1997.
1997.
Water 2018, 10, 1468 18 of 23
Water 2018, 10, x 17 of 22

TC, K = 1 h TC, K = 3 h

TC, K = 5 h
(a) CN = 60

TC, K = 1 h TC, K = 3 h

TC, K = 5 h
(b) CN = 80

Figure 10. Cont.

Water 2018, 10, 1468 19 of 23
Water 2018, 10, x 18 of 22

TC, K = 1 h TC, K = 3 h

TC, K = 5 h
(c) CN = 100

Figure 10. As
As for
for Figure
Figure7,7,but
butwith
withthe
therandomly
randomlydistributed
distributedrainfall histograms
rainfall in in
histograms Figure 9. 9.
Figure

Under
Under the above assumptions
assumptions regarding
regardingconsidering
consideringthe therainfall
rainfalldistributions
distributions with
with their
their peaks
peaks
inthe
in the second,
second, third, or fourth
fourth quantile,
quantile,we wederived
derivedthe thefollowing
followingresults.
results.First, thethe
First, highest
highestpeak
peakflow
flow
was produced
was produced for for the
the rainfall distribution
distribution H1,H1, i.e.,
i.e.,the
theoriginal
originaldistribution.
distribution.OtherOther random
random rainfall
rainfall
distributions produced
distributions producedmore moreororless the
less thesame peak
same peakflow. ThisThis
flow. result could
result be seen
could moremore
be seen clearlyclearly
for thefor
casecase
the of CNof =CN
100.=That
100.is,That
underis,the condition
under of saturated
the condition of soil, a very soil,
saturated well-organized rainfall distribution
a very well-organized rainfall
like the model of rainfall temporal distribution can produce higher peak
distribution like the model of rainfall temporal distribution can produce higher peak flow than a flow than a random
distribution.
random distribution.
However, under
However, underthethe condition
conditionof smaller CN values,
of smaller CN this effectthis
values, of well-organized rainfall distribution
effect of well-organized rainfall
distribution became smaller. For the case of CN = 60, no obvious high peak flow could be foundoffor
became smaller. For the case of CN = 60, no obvious high peak flow could be found for any case
concentration
any times. All thetimes.
case of concentration small peak
All theflows were
small generated
peak by thegenerated
flows were rainfall distribution with their
by the rainfall peak
distribution
in the first quantile (i.e., H5 and H7). This result was also the case for CN = 80. Overall,
with their peak in the first quantile (i.e., H5 and H7). This result was also the case for CN = 80. Overall, the sensitivity
of the
the peak flow
sensitivity to the
of the peakrandomness
flow to theofrandomness
rainfall temporal distribution
of rainfall temporalseemed to be small.
distribution seemed This result
to be small.
can also be confirmed by the box plots (Figure 11) of the Ratiopeak defined by Equation (11).
This result can also be confirmed by the box plots (Figure 11) of the Ratiopeak defined by Equation (11).
The box plots in Figure 11 show several important features of the randomness of the rainfall
The box plots in Figure 11 show several important features of the randomness of the rainfall
distribution and its effect on the runoff peak flow. First, when the CN was high (i.e., CN = 100 in this
distribution and its effect on the runoff peak flow. First, when the CN was high (i.e., CN = 100 in
study), for Tc = 1 h, no sensitivity of the peak flow to the randomness of the rainfall temporal distribution
this study), for Tc = 1 h, no sensitivity of the peak flow to the randomness of the rainfall temporal
was found. For the case of Tc = 3 h and Tc = 5 h, the result was a bit higher, but still only a small sensitivity
distribution was found. For the case of Tc = 3 h and Tc = 5 h, the result was a bit higher, but still
was found. On the other hand, when the CN was small, the sensitivity of the peak flow to the randomness
only a small sensitivity was found. On the other hand, when the CN was small, the sensitivity of
of the rainfall temporal distribution was found to be very high. This was basically because the effect
the peak flow to the randomness of the rainfall temporal distribution was found to be very high.
of infiltration on the effective runoff became significant. However, as Tc increased, this sensitivity
This was basically because the effect of infiltration on the effective runoff became significant. However,
became smaller.
as Tc increased, this sensitivity became smaller.
Water 2018, 10, 1468 20 of 23
Water 2018, 10, x 19 of 22

(a) CN = 60 (b) CN = 80

(c) CN = 100

Figure
Figure 11.
11. As
As for
for Figure
Figure 8,
8, but
but with the randomly distributed rainfall histograms in Figure 9.

Overall, it could be concluded

Overall, concluded that when when thethe CN is small, the effect of the randomness of the
rainfall temporal
rainfall temporal distribution
distribution on on the
the runoff peak flow was found to be significant. The The infiltration,
infiltration,
rather than the rainfall temporal distribution, played an important important role in controlling
controlling the runoff
runoff peak
flow. Thus,
flow. Thus, when
whenthetheeffect
effectofofthe
theinfiltration
infiltrationwas wassmall (or(or
small thethe
CNCNvaluevaluewas high),
was thethe
high), sensitivity of the
sensitivity of
runoff
the peakpeak
runoff flow flow
to theto
rainfall temporal
the rainfall distribution
temporal became became
distribution insignificant. This situation
insignificant. was able to
This situation wasbe
obtained
able to bein urban basins
obtained in urbanor saturated
basins or basins.
saturated basins.
Additionally, it is also important to compare the sensitivity to the randomness and the model of
rainfall temporal
rainfall temporal distribution.
distribution. By Bycomparing
comparingFiguresFigures88and and 11,
11, itit can
can be easily concluded
concluded that the
sensitivityof
sensitivity ofthe
therunoff
runoffpeak
peak flow
flow to to
thethe model
model of rainfall
of rainfall temporal
temporal distribution
distribution was much
was much higher. higher.
With
With
regardregard
to thetoflooding
the flooding condition,
condition, i.e., under
i.e., under the condition
the condition of high
of high CN values
CN values in urban
in urban basins
basins and
and saturated
saturated natural
natural basins,
basins, the the sensitivity
sensitivity of the
of the runoff
runoff peakpeak flowtotothe
flow therandomness
randomnessof of the
the rainfall
temporal distribution
distribution was insignificant.
insignificant. Even Even when
when thethe CNCN values
values were small, the sensitivity of the
runoff peak flow to the randomness of the rainfall temporal distribution was minimal, unless the
rainfall peak was located in the first quantile.

6. Summary
6. Summary and
and Conclusions
Conclusions
Although rainfall
Although rainfalltemporal
temporaldistribution
distributionmodels
models areare generally
generally usedused in hydrological
in hydrological practice,
practice, it is
it is not easy to select the best model among the many models available. This
not easy to select the best model among the many models available. This is basically because, as theis basically because,
as the model
model shows, shows, the observed
the observed rainfall
rainfall distribution
distribution is rather
is rather random,
random, notsystematic,
not systematic,which
which prevents
prevents
valid comparison. This study focused on this point, and evaluated five models
valid comparison. This study focused on this point, and evaluated five models of rainfall temporal of rainfall temporal
distribution (i.e.,
distribution (i.e.,the
theYen
YenandandChow
Chowmodel,
model,thethe Mononobe
Mononobe model,
model, thethe alternating
alternating block
block method,
method, the
the Huff
Huff model,model,
and and the Keifer
the Keifer and and
ChuChu model),
model), withwith the annual
the annual maximum
maximum rainfall
rainfall eventsevents selected
selected from
from Seoul,
Seoul, Korea,Korea, fromto
from 1961 1961 to Three
2016. 2016. Three different
different evaluation
evaluation measures
measures were considered
were considered in this in this
study,
study, namely
namely the absolute
the absolute difference
difference between between thepeaks
the rainfall rainfall peaks
of the of the
model and model and the
the observed observed
(DR peak), the
(DR ), the root mean square error (RMSE), and the pattern correlation coefficient
root mean square error (RMSE), and the pattern correlation coefficient (R). Additionally, the DRpeak was
peak (R). Additionally,
standardized by dividing it by the observed rainfall peak (SDRpeak), and RMSE by the mean rainfall
intensity (SRMSE).
Water 2018, 10, 1468 21 of 23

the DRpeak was standardized by dividing it by the observed rainfall peak (SDRpeak ), and RMSE by the
mean rainfall intensity (SRMSE).
Three different cases were considered for the comparison of the model and the observed rainfall
temporal distribution. The first case (Case 1) compared the model and the observed rainfall event as it
was. The second case (Case 2) compared the model and the observed rainfall event, after reconciling
the rainfall peak time of the model to the observed. The third case (Case 3) compared the model
and the observed rainfall event after rearranging the observed rainfall temporal distribution to be
most similar to the model. Finally, the sensitivity of the runoff peak flow to the model and to
the randomness of the rainfall temporal distribution was analyzed. This analysis was carried out to
determine whether the model or the randomness of the rainfall temporal distribution had the dominant
effect on the runoff peak flow.
As a result, the SDRpeak showed that the Yen and Chow model and the Huff model significantly
underestimated the rainfall peak, while the Mononobe model overestimated it. Of the five models
considered in this study, the Keifer and Chu model produced the most similar rainfall peak to the
observed. On the other hand, the smallest SRMSEs were estimated for the Yen and Chow model and
the alternating block method. The value of R was larger for the alternating block method, but smaller
for the Yen and Chow model and the Huff model. In particular, in Case 3, the alternating block method
showed the smallest SRMSE and largest R, which indicated that it was the most similar model to the
observed rainfall temporal distribution. Overall, the model that best mimicked the annual maximum
rainfall events observed in Seoul, Korea, was found to be the alternating block method.
The sensitivity analysis showed that the sensitivity of the runoff peak flow to the model of rainfall
temporal distribution was much higher than to the randomness of the rainfall temporal distribution.
In particular, under the condition of high CN values in urban basins and saturated natural basins,
the sensitivity of the runoff peak flow to the randomness of the rainfall temporal distribution was
found to be insignificant. Even in the case where the CN values were small, the sensitivity of the runoff
peak flow to the randomness of the rainfall distribution was found to be minimal, unless the rainfall
peak was located in the first quantile.
Based on the above findings, it could be concluded that, for the estimation of the runoff peak flow,
the proper selection of the rainfall temporal distribution model is important. This conclusion was also
supported by the result that the sensitivity to the model of rainfall temporal distribution was found to
be much higher than the sensitivity to the randomness of the rainfall temporal distribution. Particularly
in a small basin with high CN values, the importance of using a proper model of rainfall temporal
distribution cannot be underestimated. This conclusion is important as it shows that any rainfall
temporal distribution model may not satisfactorily mimic the observed rainfall event. A hydrological
design based on any selected rainfall temporal distribution model could thus be far from the real world.
To minimize the difference between the modeling and the real world, the selection of the most probable
model may well be the most important factor. However, simply the comparison of the observed and
model rainfall temporal distribution is not enough. A more systematic approach, as suggested in this
study, can help to find the best model.

Author Contributions: C.Y. conceived and designed the idea of this research and wrote the manuscript.
W.N. collected the data and conducted estimation of rainfall temporal distribution models using that data.
Funding: This work was supported by the Korea Agency for Infrastructure Technology Advancement (KAIA)
grant funded by the Ministry of Land, Infrastructure and Transport (Grant 18AWMP-B083066-05).
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Nouh, M.A. Effects of Rainfall Runoff Model Assumptions on Optimal Storm-Sewer System-Design. Arab. J.
Sci. Eng. 1987, 12, 19–35.
Water 2018, 10, 1468 22 of 23

2. U.S. Soil Conservation Service (SCS). A Method for Estimating Volume and Rate of Runoff in Small Watersheds;
U.S. Soil Conservation Service: Washington, DC, USA, 1973.
3. Huff, F.A. Time Distribution of Rainfall in Heavy Storms. Water Resour. Res. 1967, 3, 1007–1019. [CrossRef]
4. Jeong, J.H.; Yoon, Y.N. Design Practices for Water Resources; Goomi Press: Seoul, Korea, 2007.
5. Pilgrim, D.H.; Cordery, I. Rainfall Temporal Patterns for Design Floods. J. Hydraul. Div. ASCE 1975, 101,
81–95.
6. Yen, B.C.; Chow, V.T. Design Hyetographs for Small Drainage Structures. J. Hydraul. Div. ASCE 1980, 106,
1055–1076.
7. Chow, V.T.; Maidment, D.R.; Mays, L.W. Applied Hydrology; McGraw-Hill: New York, NY, USA, 1988.
8. Keifer, C.J.; Chu, H.H. Synthetic storm pattern for drainage design. J. Hydraul. Div. 1957, 83, 1–25.
9. Lee, K.T.; Ho, J.Y. Design hyetograph for typhoon rainstorms in Taiwan. J. Hydrol. Eng. 2008, 13, 647–651.
[CrossRef]
10. Garcia-Guzman, A.; Aranda-Oliver, E. A Stochastic-Model of Dimensionless Hyetograph. Water Resour. Res.
1993, 29, 2363–2370. [CrossRef]
11. Lin, G.F.; Chen, L.H.; Kao, S.C. Development of regional design hyetographs. Hydrol. Process. 2005, 19,
937–946. [CrossRef]
12. Cheng, K.S.; Hueter, I.; Hsu, E.; Yen, H.C. A scale-invariant Gauss—Markov model for design storm
hyetographs. J. Am. Water Resour. Assoc. 2001, 37, 723–735. [CrossRef]
13. Wu, S.J.; Yang, J.C.; Tung, Y.K. Identification and stochastic generation of representative rainfall temporal
patterns in Hong Kong territory. Stoch. Environ. Res. Risk Assess. 2006, 20, 171–183. [CrossRef]
14. National Ocanic and Atmospheric Administration (NOAA). Precipitation-Frequency Atlas of the United States;
NOAA: Silver Spring, MD, USA, 2013.
15. Chan, E.S.; Balmforth, D.; Kan, Y.; Koike, T.; Lampe, L.; Lim, K.K.; Yong, K.K. Report on Key Conclusions and
Recommendations of the Expert Panel on Drainage Design and Flood Protection Measures; Singapore’s National
Water Agency: Singapore, 2012.
16. Ministry of Construction and Transportation (MOCT). River Maintenance Basic Plan and Instructions of River
Maintenance Report; MOCT: Seoul, Korea, 2004.
17. Yin, S.Q.; Xie, Y.; Nearing, M.A.; Guo, W.L.; Zhu, Z.Y. Intra-Storm Temporal Patterns of Rainfall in China
Using Huff Curves. Trans. ASABE 2016, 59, 1619–1632.
18. Nelson City Council. Maitai River Flood Hazard Mapping Modelling Report; Nelson City Council: Nelson,
New Zealand, 2013.
19. Palla, A.; Gnecco, I. Assessing the urban catchment hydrologic response under different environmental
scenarios. In Proceedings of the NOVATECH 2016—9th International Conference on Planning and
Technologies for Sustainable Urban Water Management, Lyon, France, 28 June–1 July 2016.
20. Japan International Cooperation Agency (JICA). Technical Standards and Guidelines for Planning of Flood Control
Structures; JICA: Tokyo, Japan, 2010; pp. 39–42.
21. Drainage Services Department. Stormwater Drainage Manual: Planning, Design and Management;
Drainage Services Department: Hong Kong, China, 2018; Volume 26.
22. Japanese Ministry of Land, Infrastructure, Transport and Tourism (MLITT); Kanazawa, H.; Yoshitani, J.;
Shintaku, S. Technical Criteria for River Works: Practical Guide for Planning; MLITT: Tokyo, Japan, 2009.
23. Al-Rawas, G.A.; Valeo, C. Characteristics of rainstorm temporal distributions in arid mountainous and
coastal regions. J. Hydrol. 2009, 376, 318–326. [CrossRef]
24. Aron, G.; Wall, D.J.; White, E.L.; Dunn, C.N. Regional Rainfall Intensity-Duration-Frequency Curves for
Pennsylvania. Water Resour. Bull. 1987, 23, 479–485. [CrossRef]
25. Guo, J.C.Y.; Hargadin, K. Conservative Design Rainfall Distribution. J. Hydrol. Eng. 2009, 14, 528–530.
[CrossRef]
26. Kimoto, A.; Canfield, H.E.; Stewart, D. Comparison of Synthetic Design Storms with Observed Storms in
Southern Arizona. J. Hydrol. Eng. 2011, 16, 935–941. [CrossRef]
27. Nguyen, V.T.V.; Desramaut, N.; Nguyen, T.D. Optimal rainfall temporal patterns for urban drainage design
in the context of climate change. Water Sci. Technol. 2010, 62, 1170–1176. [CrossRef] [PubMed]
28. Ministry of Construction and Transportation (MOCT). 1999 Report on the Development of Water Resources
Management Techniques: Design Rainfall Temporal Distribution; MOCT: Seoul, Korea, 2000.
Water 2018, 10, 1468 23 of 23

29. Ministry of Land, Transport and Maritime Affairs (MLTM). Research on the Improvement of Probability Rainfall;
MLTM: Seoul, Korea, 2011.
30. Choi, S.; Joo, K.; Shin, H.; Heo, J. Improvement of Huff’s Method Considering Severe Rainstorm Events.
J. Korea Water Resour. Assoc. 2014, 47, 985–996. [CrossRef]
31. Joo, J.; Lee, J.; Jo, D.; Jun, H.; Kim, J. Development of a Rainfall Time Distribution Model for Urban Watersheds.
J. Korea Water Resour. Assoc. 2007, 40, 655–663. [CrossRef]
32. Knapp, H.V.; Terstriep, M.L. Effects of Basin Rainfall Estimates on Dam Safety Design in Illinois; Illinois State
Water Survey: Champaign, IL, USA, 1981; pp. 14–22.
33. Ward, A.B.; Bridges, T.; Barfield, B. An evaluation of hydrologic modeling techniques for determining
a design storm hydrograph. In Proceedings of the International Symposium on Urban Storm Runoff,
University of Kentucky, Lexington, Kentucky, 28–31 July 1980; pp. 28–31.
34. Lee, J.K.; Chu, H.J. Time distribution characteristics of an annual maximum rainfall according to rainfall
durations using Huff’s method. J. Korean Soc. Civ. Eng. 2006, 26, 519–528.
35. Park, M.; Yoo, C.; Kim, H.; Jun, C. Bivariate frequency analysis of annual maximum rainfall event series in
Seoul, Korea. J. Hydrol. Eng. 2014, 19, 1080–1088. [CrossRef]
36. Park, C.; Yoo, C. Review of parameter estimation procedure of Freud bivariate exponential distribution.
J. Korea Water Resour. Assoc. 2012, 45, 191–201. [CrossRef]
37. Loveridge, M.; Babister, M.; Retallick, M. Australian Rainfall and Runoff Revision Project 3: Temporal Patterns of
Rainfall; Department of Industry, Innovation and Science: Canberra, Australia, 2015.
38. Mockus, V. National Engineering Handbook: Section 4, Hydrology; U.S. Department of Agriculture: Washington,
DC, USA, 1965.

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).