An assessment of selected hydrochemical

parameter trend of the Nakdong River

water in South Korea, using time series
analyses and PCA
S.Y.Chung, S.Venkatramanan, N.Park,
R.Rajesh, T.Ramkumar & B.W.Kim

Environmental Monitoring and

Assessment
An International Journal Devoted to
Progress in the Use of Monitoring Data
in Assessing Environmental Risks to
Man and the Environment
ISSN 0167-6369
Volume 187
Number 1
Environ Monit Assess (2015) 187:1-13
DOI 10.1007/s10661-014-4192-9

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer
International Publishing Switzerland. This eoffprint is for personal use only and shall not
be self-archived in electronic repositories. If
you wish to self-archive your article, please
use the accepted manuscript version for
posting on your own website. You may
further deposit the accepted manuscript
version in any repository, provided it is only
made publicly available 12 months after
official publication or later and provided
acknowledgement is given to the original
source of publication and a link is inserted
to the published article on Springer's
website. The link must be accompanied by
the following text: "The final publication is
available at link.springer.com.

1 23

Author's personal copy

Environ Monit Assess (2015) 187:4192
DOI 10.1007/s10661-014-4192-9

An assessment of selected hydrochemical parameter trend

of the Nakdong River water in South Korea, using time
series analyses and PCA
S. Y. Chung & S. Venkatramanan & N. Park &
R. Rajesh & T. Ramkumar & B. W. Kim

Received: 10 October 2013 / Accepted: 19 November 2014

# Springer International Publishing Switzerland 2014

Abstract Time series analyses (autocorrelation,

spectral density, and cross-correlation) and principal
component analysis (PCA) were used to understand
the characteristics of the selected hydrochemical parameters pH, turbidity, alkalinity, Cl, hardness, total dissolved solids (TDS), and metals Fe and Mn in the
Nakdong River, South Korea. Autocorrelation and spectral density for pH, alkalinity, hardness, and Cl were
very similar to TDS, whereas Fe, Mn, and turbidity
showed different trends from TDS. Cross-correlograms
of pH, alkalinity, hardness, and Cl versus TDS were
very similar to each other. Those of Fe and turbidity

represented the opposite relations with other components. Cross-correlation coefficients had the highest
values at zero lag, indicating that pH, alkalinity, hardness, and Cl are controlling factors for TDS. On the
other hand, Fe and turbidity showed the highest values
at 6-month lag and Mn at a month lag. PCA indicated
that TDS had very close relation with hardness, pH, and
Cl and very small relation with Mn. Turbidity and Fe
had relatively opposite relations with TDS. It was concluded that the geostatistical methods were very useful
for evaluating the hydrochemical characteristics of the
Nakdong River water in South Korea.

S. Y. Chung : S. Venkatramanan (*) : R. Rajesh

Department of Earth & Environmental Sciences, Institute of
Environmental Geosciences, Pukyong National University,
599-1 Daeyeon-dong Nam-gu, Busan 608-737, South Korea
e-mail: venkatramanansenapathi@gmail.com

Keywords Time series analyses . Autocorrelation .

Cross-correlation . Spectral density . PCA .
Hydrochemical parameters

S. Venkatramanan
e-mail: venkat@pknu.ac.kr

Introduction

S. Y. Chung
e-mail: chungsy@pknu.ac.kr

The hydrochemical parameters of surface water can be

indicative of its origin and history of passage through
underground materials which water has been in contact
with, in shallow and deep-seated conditions. Natural
and anthropogenic sources along with chemical and
biogeochemical constituents have been considerably
altering surface water quality in recent years. The computerized geostatistical methods are powerful tools for
understanding the chemical state of natural river waters
and for predicting the behavior of such waters under a
variety of hypothetical conditions (Helgeson et al. 1970;
Shannon et al. 1977; Spostigo and Mattigod 1979;

N. Park
Department of Civil Engineering, Dong-A University,
840 Hadan-dong Saha-gu, Busan 604-717, South Korea
T. Ramkumar
Department of Earth Sciences, Annamalai University,
Annamalai Nagar 608 502 Tamilnadu, India
B. W. Kim
Radioactive Waste Technology Development Division, Korea
Atomic Energy Research Institute,
Yuseong-gu, Daejeon 305-353, South Korea

Author's personal copy

4192, Page 2 of 13

Wolery 1979; Plummer et al. 1983; Felmy et al. 1984;

Parkhurst et al. 1990; Venkatramanan et al. 2014a, b).
Many researchers have been applying a time series
analysis and principal component analysis (PCA) for the
elucidation of the characteristics of hydrogeological processes. The time series analysis and PCA provide useful
information about various patterns of spatial and temporal variations of hydrologic processes of the aquifer
system (Duffy and Gelhar 1986; Oh 1994; Ryu et al.
2011; Triki et al. 2014). According to Mangin (1984),
Padilla and Pulido-Bosch (1995), and Laroque et al.
(1998), the methods were very useful to study the relation between groundwater flow rates and precipitation
for the better understanding of global flow dynamics in
karst aquifers. Lee and Lee (2000) and Kim et al. (2000,
2005) revealed that time series analysis was helpful for
the delineation of the hydraulic relationship between
groundwater level and rainfall. In this study, the time
series analysis and PCA were used for the identification
of relationship between total dissolved solids (TDS) with
other hydrochemical parameters. Time series of surface
water quality parameters usually exhibit complex fluctuations owing to interactions of many factors. Time series
analysis has been widely used in surface water resource
evaluation, forecast, and management because of the
simple, easy, and practical features. Most of the research
articles decompose the surface water level variations into
trend, periodicity, and random components to understand
their characteristics and combine the three components
together as an additive model to forecast the future trend
(Hu et al. 2001; Zhao et al. 2007; Zhou et al. 2007;
Erdogan and Gulal 2009; Yang et al. 2009; Liang 2011).
The purpose of this research is to identify the characteristics of surface water quality influenced by both
natural and anthropogenic sources. The time series analyses of autocorrelation, spectral analysis, and crosscorrelation clarify the relationship between the
hydrochemical parameters of pH, turbidity, alkalinity,
Cl, hardness, TDS, and the metal components of Fe and
Mn. PCA techniques are also used to establish the same
interrelation of hydrochemical parameters and metals.

Geological and environmental settings

The study area is located at the downstream of the
Nakdong River (Fig. 1), which is the longest river in
South Korea. The length of the Nakdong River is
525 km, and the total watershed area is 24,000 km2.

Environ Monit Assess (2015) 187:4192

The river has eight main tributaries. The river flows

from Andong City, Gyongbuk Province to the Busan
Metropolitan City, Gyongnam Province, and it is finally
connected to East China Sea. During the summer seasons, flood frequently happens in the study area in case
of heavy rain and also many sediments and organic
deposit in the river banks. The delta deposits are about
6090 m deep and composed of backfill, upper clay,
sand, lower clay, and gravel in sequence. The sediments
began to deposit from the late Pleistocene Epoch, i.e.,
the end of fourth glacial period (Oh 1994; Ryu et al.
2011). The basal gravel bed indicates an unconformity
between delta sediments and bedrock. The bedrock
consists of granite, andesite, and rhyolite. The thickness
of upper clay ranged from 5 to 10 m, sand from 10 to
40 m, lower clay from 10 to 30 m, and gravel 5 to 40 m,
respectively. The upper clay is relatively soft and loose,
but lower clay is relatively stiff and dense. An unconfined aquifer of sand layer and a confined aquifer of
sandgravel layer coexist in the study area. The industrial complex is located beside the right side of the
Nakdong River in the study area, and many factories
are managed at the upstream of the Nakdong River.
Many rice paddies and greenhouses of fruits and vegetables are also managed in the delta as well as its
upstream area. Many houses, roads, and airport are also
located in the deltaic region. Thus, some contaminants
such as nitrate, phosphate, and some inorganic and
organic materials come into the Nakdong River, even
though the city controls the contaminants well. The
mouth of the Nakdong River is connected to East
China Sea. By the way, the barrage constructed at the
river mouth prevents seawater intrusion of the study area.

Materials and methods

Measurement of water quality
Surface water was sampled at a depth of 30 cm depth of
the Nakdong River from January to December during
the year 2012. Water samples were measured in the field
using portable instruments: ThermOrion 250A+, USA
for pH, Kasahara TR-5Z, Japan for turbidity, and TOA
CM-14P, Japan for electrical conductivity (EC) and
TDS. Alkalinity was measured by titration method using
0.02 N H2SO4 in the field. Fe and Mn were analyzed in
the laboratory using atomic absorption spectrometer
(AAS, Unicam 989 method). Hardness was expressed

Author's personal copy

Environ Monit Assess (2015) 187:4192

Page 3 of 13, 4192

Mo nitoring
S tation

Fig. 1 Location map of a monitoring station with geological section

as the equivalent of the calcium carbonate (CaCO3) and

calculated by the equation HT =2.5Ca+4.1Mg.
Time series analysis
A short overview of the theory of autocorrelation, spectral density, and cross-correlation analysis is presented.
The mathematical expressions of the functions are
shown in detail in Jenkins and Watts (1968), Mangin
(1984), Box et al. (1994), Padilla et al. (1994), and
Laroque et al. (1998). The simple autocorrelation analysis quantifies the linear dependency of successive
values over a time period. Autocorrelation function,
r(k), is expressed as autocovariance C(k) and variance
C(0) of time series data like
1X


xt X xtk X
n t1

determined based on the interval of the analysis and

the given circumstance. If the time series has strong
interdependency and a long memory effect, the autocorrelation function shows a gently decreasing slope and
nonzero values over a long time lag. However, if the
time series is uncorrelated, such as rainfall, the autocorrelation function decreases very quickly and reaches a
zero value in a short time (Lee and Lee 2000). The
simple spectral density analysis is complementary to
the autocorrelation analysis. The spectral density function corresponds to change from a time mode to a
frequency mode through a Fourier transformation of
the autocorrelation function. The interpretation of the
spectral density function, S(f), through the identification
of the different peaks representing periodical phenomena, leads to the characterization of the system:

nk

rk Ck =C0 C k

where k is the time lag (k=0m), n is the number of

events, xt is a single event, X is the mean of events, and
m is the cutting point. The cutting point is usually

"
S f 2 1 2

Xm
k1

#
Dk rk cos2 f k Dk


1 cos

k
m

where f is the frequency, and D(k) ensures that estimated

values S(f) are not biased. The cross-correlation analysis

Author's personal copy

4192, Page 4 of 13

Environ Monit Assess (2015) 187:4192

is used to establish a link between the input time series

and the output time series. If the input time series is
random, the cross-correlation function, rxy(k), corresponds to the impulse response of the system. In other
cases, the cross-correlation function provides information on the casual and noncasual relationships between
the input and the output time series data as well as the
importance of these relationships. The definition of a
cross-correlation coefficient is as follows:

C xy k
1X

C xy k
xt X ytk 
y
n t1
x y
nk

rxyk

where Cxy(k) is the cross-covariance of x(t) and y(t) time

series, and x, y are the standard deviations of two time
series, respectively.
PCA
The principal components (PCs) are the uncorrelated
(orthogonal) variables, obtained by multiplying the original correlated variables with the eigenvector (loadings
or weightings). The eigenvalues of the PCs are the
measure of their associated variance. The participation
of the original variables in the PCs is given by the
loadings, and the individual transformed observations
are called scores (Vega et al. 1998; Helena et al. 2000;
Wunderlin et al. 2001; Singh et al. 2004). PCA was
performed on normalized (z-scale transformation) variables for eight parameters after sorting out the highly
correlated variables from the data sets. PCs with eigenvalue >1 were retained. The contribution of each factor at
every site (factor scores) was computed, and score plots
of first two PCs (PC1 and PC2) were constructed. PCA
was performed with a view to assess the compositional
differences among different hydrochemical parameters,
spatial variations in water quality, and influence of anthropogenic activities. The results of time series analysis
and PCA were produced by STATISTICA (Ver.7).

Results and discussion

Temporal variation of hydrochemical parameters
The monthly average values of selected hydrochemical
parameters and metals for the Nakdong River quality
during the year 2012 are presented in Table 1. The
trends of these parameters are shown in Fig. 2. During

the study period, pH value of water samples varied from

7 to 9.4, indicating that water is alkaline in nature. pH
was influenced by the variable effluent input, which is
the prime pathway to bring the effluents from the industrial sector (Kura et al. 2014). Turbidity and alkalinity
values ranged from 5.2 to 72.1 mg/L and 30 to 59 mg/L
due to the irrigation return flow. Chloride concentration
of water samples ranged from 10 to 48 mg/L, and it
might be due to the influences of saline water intrusion
to the coastal aquifer. This was similar to observation of
Appelo and Willemsen (1987) who studied geochemical
calculation and observations on saline water intrusions.
Another reason for the increase of chloride concentration is related with the usage of ammonium chloride
fertilizer in the nearby cultivated area. Further, chloride
concentration of surface water is derived from dissolution of evaporate salts as a probable source of salinity
(Bouzourra et al. 2014; Venkatramanan et al. 2014c).
The hardness of water samples varied from 42 to
101 mg/L, indicating the Ca and Mg are known to occur
naturally in water due to its passage through mineral
deposits of rock strata. TDS value ranged from 176 to
272 mg/L, showing good correlation with pH, hardness,
Cl, and alkalinity. Higher concentration of TDS is observed in March, due to the impact of solubility of lime
and gypsum which were added during agriculture operations (Jain et al. 1997; Nas and Berktay 2010). Further,
the study area is a coastal region, and seawater incursion
is a main factor for the increased amount of TDS in
surface water. It is supported by a slightly high value of
Cl. Heavy metals Mn and Fe were dominant during the
months of April and September. Fe and Mn ranged from
0 to 0.55 mg/L and from 0 to 0.027 mg/L, respectively.
In general, these elements mainly originated from terrestrial rock-forming materials. Fe is one of the most
common elements in the earths crust. Mn originated
from factory as well as terrestrial soil materials because
many factories were located around the Nakdong River.
Eight components, pH, turbidity, alkalinity, Cl, hardness, TDS, Fe, and Mn, were measured at every month
in 2012. Figure 2 shows the variations for the average
values of eight components. The monthly variations for
five components pH, alkalinity, Cl, hardness, and TDS
were very similar to each other. High values in February
or March decreased to the lowest values in September,
and then, they increased to next March. These variations
were related with seasonal variations, i.e., increased in
dry season and decreased in wet season. In South Korea,
dry season (fall and winter) ranges from September to

Author's personal copy

Environ Monit Assess (2015) 187:4192

Page 5 of 13, 4192

Table 1 Average value of chemical components during the year 2012 (unit mg/L except pH)
Month

pH

Turbidity

Alkalinity

Cl

Hardness

TDS

Fe

Mn

January

8.7

9.5

59

35

99

227

0.08

0.011

February

9.4

9.8

59

45

101

242

0.15

0.016

March

9.2

12.7

55

48

98

272

0.16

0.027

April

7.6

16.8

51

34

82

225

0.15

0.031

May

7.4

12.3

45

20

64

201

0.23

0.024

June

7.5

5.6

51

25

70

179

ND

ND

July

7.3

15.6

48

22

68

205

0.09

0.008

August

7.2

16.5

44

15

60

206

0.09

0.015

September

72.1

30

10

42

176

0.55

0.022

October

7.3

16

35

13

50

185

0.12

0.01

November

8.1

6.7

46

21

68

210

0.07

0.013

December

8.2

5.2

55

33

85

223

0.07

0.023

ND not detected

February, and wet season (spring and summer) ranges

from March to August. In wet season, dissolved ions in
the Nakdong River are diluted by much quantity of
surface water, while in dry season, the concentrations
of dissolved ions become increased due to the small
recharge of surface water. The maximum and minimum
concentrations of five components appeared at 1 month
later of dry season and wet season, respectively. The
drinking standard of pH is 5.8 to 8.5, and the Nakdong
River water was not suitable in dry season. However,
alkalinity, Cl, hardness, and TDS were all in the range of
South Korean drinking standard revised in 2011.
Turbidity and Fe exhibited the maximum values in
September. Suspended materials were high due to the
increase of water quantity in wet season. The increase of
suspended materials produced the increase of turbidity
and Fe. Fe originated mainly from the terrestrial rockforming materials. The drinking standard level of Fe
was 0.3 mg/L, and water of the Nakdong River
exceeded only in September. The higher concentration
of Mn was observed during the months of April,
September, and December. Mn increased continuously
from October to April during the cold period. It is
thought that the variation of Mn was not related only
with natural origin, because it was derived from the
adjacent factory as well as terrestrial soil materials
(Venkatramanan et al. 2014d). The drinking standard
level of Mn was 0.3 mg/L, and water of the Nakdong
River did not exceed the standard limit in drinking.

Time series analysis

Autocorrelation and spectral density
The autocorrelation functions of selected hydrochemical
parameters and metals such as pH, turbidity, alkalinity,
Cl, hardness, TDS, Fe, and Mn are shown in Fig. 3. The
autocorrelation functions of turbidity and alkalinity
reach a null value. This is an indicator for the uncorrelated characteristics of the turbidity and alkalinity with
other parameters (Angelini 1997; Kim et al. 2005). On
the other hand, the autocorrelation functions of pH, Cl,
and TDS show sinusoidal patterns. Thus, these parameters have strong interrelationships with a constant period of investigation. In the case of Fe and Mn, the
autocorrelation coefficients decreased slowly towards a
null value at 3.7 and 4.6 months, respectively. They also
showed sinusoidal behaviors during the study period
(Lu et al. 2013). However, the autocorrelation coefficients decrease negatively after increasing positively,
and they are opposite to those of pH, Cl, and TDS.
These results of the autocorrelation represent that pH,
Cl, and TDS have a strong linear interrelationship, and
they are subjected to periodically changing sources. In
fact, TDS and other parameters mainly are affected by
climatic condition and influences of the anthropogenic
sources (Aflatooni and Mardaneh 2011). The conspicuous variations of parameters are mainly due to precipitation and slightly due to agricultural activities and

Author's personal copy

4192, Page 6 of 13

Fig. 2 Temporal variations of hydrochemical components in the study area

Environ Monit Assess (2015) 187:4192

Author's personal copy

Environ Monit Assess (2015) 187:4192

Fig. 3 Autocorrelation functions of hydrochemical components in the study area

Page 7 of 13, 4192

Author's personal copy

4192, Page 8 of 13

Fig. 4 Spectral density patterns of hydrochemical components in the study area

Environ Monit Assess (2015) 187:4192

Author's personal copy

Environ Monit Assess (2015) 187:4192

Page 9 of 13, 4192

Fig. 5 Cross-correlation
functions of hydrochemical
components in the study area

seawater intrusion. Apart from this, there are textile/

dyeing industries around the Nakdong River, which
secretly throw fabric scraps and chemicals into the river
(Chun et al. 2001; Alvarez-Mieles et al. 2013).

Spectral density functions of selected hydrochemical

parameters and metals such as pH, turbidity, alkalinity,
Cl, hardness, TDS, Fe, and Mn are shown in Fig. 4. The
behaviors of these functions are similar to each other, and

Author's personal copy

4192, Page 10 of 13

their highest peaks have a single point and some multiple

points. The function of pH, Cl, and hardness has almost
similar trend with TDS. It suggests that TDS was affected
by Cl and hardness and that TDS played a more vital role
on the surface water quality. The turbidity and alkalinity
show a repeated increasing to decreasing trend. Fe and
Mn exhibit multiple peaks, and it means the presence of
several distinct cyclic mechanisms in them (Diggle 1992;
Lu et al. 2013). The spectral densities of Fe and Mn are
very small, compared to the spectral densities of other
components. Multiple peaks of Fe and Mn resulted from
the human activity as well as seasonal influence. It could
be also due to the textile/dyeing, fertilizer, and domestic
sewage inputs from the tributaries of Nakdong Main
River (Sankaran et al. 2012; Venkatramanan et al. 2014a).
Cross-correlation
The cross-correlations of hydrochemical parameters
(Fig. 5) were calculated to inspect a relationship between parameters of the surface water quality and
TDS. The correlograms of pH, alkalinity, hardness,
and Cl have very similar patterns and those of Mn are
a little similar to four components. Their correlation
coefficients are over 0.5 after or before 1-month lag. It
suggests that five components are the controlling factors
of TDS. Turbidity represents a negative correlation with

Environ Monit Assess (2015) 187:4192

TDS. In wet season, TDS is decreased due to the dilution of heavy precipitation, but turbidity is increased due
to increased sediment loads in Nakdong River water. On
the other hand, TDS is increased, and turbidity is decreased in dry season. This phenomenon is because the
study area is located at the lowest downstream of the
Nakdong River. Fe shows the opposite relation with
TDS.
Principal component analysis
PCA allows to reduce the dimensionality of a
highly dimensional data set by explaining the correlation among a large number of variables in
terms of smaller number of underlying factors
(PCs) without losing much information (Jackson
1991; Helena et al. 2000; Kowalkowskia et al.
2006). PCA was applied to the hydrochemical
components of the Nakdong River water to identify the factors that influence each one. PCA 1 of
77.89 % variance shows positively high loadings
by Fe and turbidity and negatively high loadings
with group of TDS, hardness, pH, and Cl. Mn has
a negatively small loading. TDS has very close
relations with hardness, pH, and Cl and very small
relation with Mn. Turbidity and Fe have opposite
relations with TDS. PCA 2 of 17.03 % variance is

Fig. 6 Cross-plot of the PCA for

hydrochemical components in the
study area

1.0

PCA 2 : 17.03%

0.5

Alkalnity
0.0

Hardness
Cl*TDS
pH

*Mn
-0.5
Turbidity

*Fe

-1.0
-1.0

-0.5

0.0
PCA 1 : 77.89%

0.5

1.0

Author's personal copy

Environ Monit Assess (2015) 187:4192

Page 11 of 13, 4192

2.0

Fig. 7 Score plot of the PCA for

hydrochemical components in the
study area

1.5
1.0

June
MayAug Oct
Nov July

PCA 2: 17.03%

0.5

Dec
April

0.0

Jan

-0.5
Feb
Mar
-1.0
-1.5
Sep
-2.0
-2.5
-3.0
-5

-4

-3

-2

-1

PCA 1: 77.89%

characterized by a moderate negative loading of

Mn, whereas alkalinity exhibits the moderate positive loading. It indicates the Mn was derived from
the anthropogenic origin. Contribution of alkalinity
shows natural origin and atmospheric CO2 influence (Fig. 6). PCA 2 components are derived from
mixed sources of agricultural runoff, industries,
and infiltration of domestic sewage from the surrounding area (Kumari et al. 2013). By the score plot
(Fig. 7), monthly variations of the hydrochemical parameters and metals mostly fall in positive factor loadings except January, February, March, and September. It
is thought that the phenomenon is due to the similar
geochemical conditions during the periods. The temporal trends of pH, alkalinity, Cl, and hardness fluctuations
were very similar to those of TDS with a little time gap.
Time series and PCA analyses provided the same
results.

Conclusion
The sources of hydrochemical and metal components in the Nakdong River might be most likely
derived from industrial wastewater, irrigation activity, sediment loadings, and a slight seawater intrusion. The temporal trends of pH, alkalinity, Cl, and

hardness fluctuations were very similar to those of

the TDS with a little time gap. Autocorrelation and
spectral density for pH, alkalinity, hardness, and
Cl were very similar to TDS, but Fe, Mn, and
turbidity showed different trends. PCA was performed on six significant PCs explaining 78 %
of data variance. Two main principal components
contained the variables with origin in geochemical
condition and anthropogenic sources. The information obtained from the hydrochemical study would
be contributed to the decision-making of protection, management, and conservation for the water
quality of the Nakdong River in South Korea. It is
expected that the improvement of water management as well as contaminant control will be accomplished in the Nakdong River and adjacent
deltaic region with the aid of time series analyses
and PCA.
Acknowledgments The authors are grateful to two anonymous
referees for their constructive comments and suggestions which
led to significant improvements to the manuscript. We wish to
gratefully acknowledge the valuable suggestions given by Prof.
Yu-Pin Lin, Associate Editor which greatly helped in the final
presentation of this article. This research was supported by a grant
(Code: 13AWMP-B066761-02) from AWMP Program funded by
the Ministry of Land, Infrastructure and Transport of Korean
government. The water quality data of the Nakdong River were
supplied from the Busan Metropolitan City, South Korea, and
many thanks are given to it.

Author's personal copy

4192, Page 12 of 13

References
Aflatooni, M., & Mardaneh, M. (2011). Time series analysis of
ground water table fluctuations due to temperature and rainfall change in Shiraz plain. International Journal of Water
Resources and Environmental Engineering, 3, 176188.
Alvarez-Mieles, G., Irvine, K., Griensven, A. V., Arias-Hidalgo,
M., Torres, A., & Mynett, A. E. (2013). Relationships between aquatic biotic communities and water quality in a
tropical riverwetland system (Ecuador). Environmental
Science and Policy, 34, 115127.
Angelini, P. (1997). Correlation and spectral analysis of two
hydrogeological systems in Central Italy. Hydrological
Science Journal, 42, 425439.
Appelo, C. A. J., & Willemsen, A. (1987). Geochemical calculations and observations on salt water intrusions. I. A combined
geochemical/mixing cell model. Journal of Hydrology, 94,
313330.
Bouzourra, H., Bouhlila, R., Elango, L., Slama, F., & Ouslati, N.
(2014). Characterization of mechanisms and processes of
groundwater salinization in irrigated coastal area using statistics, GIS, and hydrogeochemical investigation.
Environmental Science and Pollution Research. doi:10.
1007/s11356-014-3428-0. 2014.
Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (1994). Time series
analysis: forecasting and control (3rd ed.). Englewood
Cliffs: Prentice-Hall. 598 p.
Chun, K. C., Chang, R. W., Williams, G. P., Chang, Y. S.,
Tomasko, D., LaGory, K., Ditmars, J., Chung, H. D., &
Lee, B. K. (2001). Water quality issues in the Nakdong
River Basin in the Republic of Korea. Environmental
Engineering Policy, 2, 131143.
Diggle, P.J. (1992). Time series: a biostatistical introduction.
Oxford, London, 257p.
Duffy, C. J., & Gelhar, L. W. (1986). A frequency domain analysis
of groundwater quality fluctuations: interpretation of field
data. Water Resource Research, 22, 11151128.
Erdogan, H., & Gulal, E. (2009). The application of time series
analysis to describe the dynamic movements of suspension
bridges. Nonlinear Analysis: Real World Applications, 10,
910927.
Felmy, A. R., Bervin, D. C., Jenne, E. A. (1984). MINTEQ. A
computer program for calculating aqueous geochemical equilibria. US EPA 84032, Athens GA.
Helena, B., Pardo, R., Vega, M., Barrado, E., Fernandez, J. M., &
Fernandez, L. (2000). Temporal evolution of groundwater
composition in an alluvial (Pisuerga river, Spain) by principal
component analysis. Water Research, 34, 807816.
Helgeson, H. C., Brown, T. H., Nigrini, A., & Jones, T. A. (1970).
Calculation of mass transfer in geochemical processes involving aqueous solutions. Geochimica et Cosmochimica
Acta, 34, 569592.
Hu, K. Z., Zhang, J. Z., & Xing, L. T. (2001). Study on dynamic
characteristics of groundwater based on the time series
analysis method. Water Science Engineering Technology, 5,
3234.
Jackson, J. E. (1991). A users guide to principal components. New
York: Wiley.
Jain, C. K., Bhatia, K. K. S., & Seth, M. (1997). Characterization
of waste disposals and their impact on the water quality of

Environ Monit Assess (2015) 187:4192

river Kali. Indian Journal of Environmental Protection, 17,
287295.
Jenkins, G. M., & Watts, D. G. (1968). Spectral analysis and its
applications. San Francisco: Holden-Day Inc. 525 p.
Kim, T., Lee, K. K., Ko, K. S., & Chang, H. W. (2000).
Groundwater flow system inferred from hydraulic stresses
and heads at an underground LPG storage cavern site.
Journal of Hydrology, 236, 165184.
Kim, J., Lee, J., Cheong, T., Kim, R. H., Koh, D. C., Ryu, J. S., &
Chang, S. W. (2005). Use of time series analysis for the
identification of tidal effect on groundwater in the coastal
area of Kimje, Korea. Journal of Hydrology, 300, 188198.
Kowalkowskia, T., Zbytniewskia, R., Szpejnab, J., & Buszewski,
B. (2006). Application of chemometrics in river water classification. Environmental Chemistry, 40, 744752.
Kumari, M., Tripathi, S., Pathak, V., & Tripathi, B. D.
(2013). Chemometric characterization of river water
quality. Environmental Monitoring and Assessment, 185,
30813092.
Kura, N. U., Ramli, M. F., Ibrahim, S., Sulaiman, W. N. A., &
Aris, A. Z. (2014). An integrated assessment of seawater
intrusion in a small tropical island using geophysical, geochemical, and geostatistical techniques. Environmental
Science and Pollution Research. doi:10.1007/s11356-0142598-0.
Laroque, M., Mangin, A., Razack, M., & Banton, O. (1998).
Contribution of correlation and spectral analysis to the regional study of a large karst aquifer (Charente, France).
Journal of Hydrology, 205, 217231.
Lee, J. Y., & Lee, K. K. (2000). Use of hydrologic time series data
for identification of recharge mechanism in a fractured bedrock aquifer system. Journal of Hydrology, 229, 190201.
Liang, Y. H. (2011). Analyzing and forecasting the reliability for
repairable systems using the time series decomposition method. International Journal Quality Reliability Management,
28, 317327.
Lu, W. X., Zhao, Y., Chu, H. B., & Yang, L. L. (2013). The
analysis of groundwater levels influenced by dual factors in
western Jilin Province by using time series analysis method.
Applied Water Science. doi:10.1007/s13201-013-0111-4.
Mangin, A. (1984). Pour une meilleure connaissance des
systemes hydrologiques a partir des analyses correlatoire
et sepctrale. Journal of Hydrology, 67, 2543.
Nas, B., & Berktay, A. (2010). Groundwater quality mapping in
urban groundwater using GIS. Environmental Monitoring
and Assessment, 8, 215227.
Oh, G. H. (1994). The paleoenvironment of the northern part of the
Nakdong River delta. Korean Journal of Quaternary
Research, 8, 3342.
Padilla, A., & Pulido-Bosch, A. (1995). Study of hydrographs of
karstic aquifers by means of correlation and cross-spectral
analysis. Journal of Hydrology, 168, 7389.
Padilla, A., Pulido-Bosch, A., & Mangin, A. (1994). Relative
importance of baseflow and quickflow from hydrographs of
karst springs. Groundwater, 32, 267277.
Parkhurst, D.L., Thorstenson, D.C., Plummer, LN. (1990).
PHREEQEa computer program for geochemical calculation. USGS water resources investigations report 8096, 195p.
Plummer, L. N., Parkhurst, D. L., & Thorstenson, D. C. (1983).
Development of reaction models for groundwater systems.
Geochimica et Cosmochimica Acta, 47, 665685.

Author's personal copy

Environ Monit Assess (2015) 187:4192
Ryu, C. K., Kang, S., Chung, S. G., & Jeon, Y. M. (2011). Late
Quaternary depositional environmental change in the northern marginal area of the Nakdong River delta, Korea. Journal
of the Geological Society of Korea, 47, 213233.
Sankaran, S., Sonkamble, S., Krishnakumar, K., & Mondal, N. C.
(2012). Integrated approach for demarcating subsurface pollution and saline water intrusion zones in SIPCOT area: a
case study from Cuddalore in Southern India. Environmental
Monitoring and Assessment. doi:10.1007/s10661-011-23279. 2012.
Shannon, D. W., Morray, J. R., Smith, P. R. (1977). Use of a
chemical equilibrium model computer code to analyze scale
formation and corrosion in geothermal brines. Proceedings in
Symposium on Oil Field and Geothermal Chemistry
Conference 770609, pp. 2136.
Singh, K. P., Malik, A., Mohan, D., & Sinha, S. (2004).
Multivariate statistical techniques for the evaluation of spatial
and temporal variations in water quality of Gomti River
(India)a case study. Water Research, 38, 39803992.
Spostigo, G., & Mattigod, S. V. (1979). GEOCHEM: a computer
program for the calculation of chemical equilibria in soil
solution and other natural water systems. Riverside:
Department of Soil and Environmental Sciences, University
of California.
Triki, I., Trabelsi, N., Hentati, I., & Zairi, M. (2014). Groundwater
levels time series sensitivity to pluviometry and air temperature: a geostatistical approach to Sfax region, Tunisia.
Environmental Monitoring and Assessment, 186, 15931608.
Vega, M., Pardo, R., Barrado, E., & Deban, L. (1998). Assessment
of seasonal and polluting effects on the quality of river water
by exploratory data analysis. Water Research, 32, 35813592.
Venkatramanan, S., Chung, S. Y., Lee, S. Y., & Park, N. (2014a).
Assessment of river water quality via environmentric multivariate statistical tools and water quality index: a case study
of Nakdong River basin, Korea. Carpathian Journal of Earth
and Environmental Sciences, 9, 125132.

Page 13 of 13, 4192

Venkatramanan, S., Chung, S. Y., Park, N., Ramkumar, T., &
Gnanachandrasamy, G. (2014b). A preliminary investigation
of hydrogeochemistry, metals and saturation indices of minerals in Nakdong surface water and adjacent deltaic groundwater using WATEQ4F geochemical model. International
Journal of Earth Sciences and Engineering, 7, 456466.
Venkatramanan, S., Chung, S. Y., Ramkumar, T., Gnanachandrasamy,
G., Vasudevan, S., & Lee, S. Y. (2014c). Application of GIS and
hydrogeochemisty of groundwater pollution status of
Nagapattinam district of Tamil Nadu. India: Environmental
Earth Sciences. doi:10.1007/s12665-014-3728-1.
Venkatramanan, S., Chung, S. Y., Kim, T. H., Prasanna, M. V., &
Hamm, S. Y. (2014d). Assessment and distribution of metals
contamination in groundwater: a case study of Busan City.
Korea: Water Quality Exposure and Health. doi:10.1007/
s12403-014-0142-6.
Wolery, T.J. (1979). Calculation of chemical equilibrium between
aqueous solutions and minerals: the EQ3/EQ6 software package (41 pp.). Livermore, University of California, Lawrence
Livermore Laboratory (UCRL 52658).
Wunderlin, D. A., Diaz, M. P., Ame, M. V., Pesce, S. F., Hued, A.
C., & Bistoni, M. A. (2001). Pattern recognition techniques
for the evaluation of spatial and temporal variations in water
quality. A case study: Suquia river basin (CordobaArgentina). Water Research, 35, 28812894.
Yang, Z. P., Lu, W. X., & Long, Y. Q. (2009). Application and
comparison of two prediction models for groundwater levels:
a case study in Western Jilin Province, China. Journal of Arid
Environment, 73, 487492.
Zhao, J., Bian, Y. M., & Zhou, X. J. (2007). Application of time
series analysis method in groundwater level dynamic forecast
of Shenyang City. Water Resources Hydropower Northeast,
8, 3134.
Zhou, X. J., Lan, S. S., & Wang, B. (2007). Application of time
series analysis in groundwater level forecast of Siping region.
Water Science Engineering Technology, 8, 3537.