Environ Monit Assess (2020) 192: 502

https://doi.org/10.1007/s10661-020-08475-z

Distribution of metal contamination and grain size

in the sediments of Nakdong River, Korea
Shin Kim & Deuk Seok Yang & Yong Seok Kim

Received: 2 April 2020 / Accepted: 5 July 2020 / Published online: 9 July 2020
# The Author(s) 2020

Abstract To assess distribution of metal contamination were found at NS19 (adjacent to the Haman weir). Thus,
and grain size in the sediments of Nakdong River (South the accumulation of fine sediment increased due to the
Korea), surface sediments were collected from 21 sites influence of the weirs, thereby increasing the overall
and analyzed. Within the study area, sand was typically amount of metal contamination.
the dominant grain size. However, because of the re-
duced flow rate and flow velocity, sites adjacent to weirs Keywords Nakdong River . Sediment . Grain size .
were composed of relatively fine sediments. A compar- Metal contamination
ison of sediment metal concentrations with sediment
quality guidelines proposed by the USA, Canada, and
South Korea revealed that sites adjacent to weirs had
Introduction
concentrations that exceeded the standard values. The
enrichment factor, index of geo-accumulation, and pol-
A river is classified into a mainstream and tributaries,
lution load index calculation results that the sites adja-
both of which are directly related to human activities. In
cent to weirs showed high contamination, with Cd ac-
recent years, the natural purification capability of rivers
counting for the highest contamination levels. The
has decreased and the river environment has gradually
metals in the study area varies due to the effect of fine
deteriorated due to increased residential and industrial
sediments; therefore, high concentrations of metals ac-
sewage and wastewater linked to population growth,
cumulated adjacent to weirs where fine sediments were
improved living standards, and industrial progress
distributed in greater proportions. Furthermore, Cd ex-
(Kim et al. 2015a, b). Furthermore, river environments
hibited the greatest contribution to metal contamination
are largely affected by artificial structures, such as weirs,
in the study area and the highest contamination levels
as well as cities and industrial complexes built adjacent
to rivers (Ahn et al. 2014).
S. Kim (*) : D. S. Yang : Y. S. Kim Contaminants flowing into a river are discharged to
Nakdong River Environment Research Center, National Institute the environment through various paths. Contaminants
of Environmental Research, 24-11, Gukgasandan-daero 52-gil, flowing into the water system typically accumulate in
Guji-myeon, Dalseong-gun, Daegu, Republic of Korea
e-mail: sinima@korea.kr sediments transported and deposited by the flow of
water, waves, tidal currents, and wind. Contaminants
deposited in rivers or lakes then accumulate on the
D. S. Yang bottom and influence the river ecosystem. To obtain a
e-mail: yds7055@korea.kr
complete understanding of the river environment, it is
Y. S. Kim important to analyze the geochemical components (e.g.,
e-mail: nierkys@korea.kr metals) that accumulate in the sediments as well as the
502 Page 2 of 15 Environ Monit Assess (2020) 192: 502

water quality environment (Thornton 1983). Water determine the relationships between each variable and
quality analysis is key for understanding the current the main factors influencing the contamination level of
short-term environmental conditions; however, sedi- sediments in the study area. As a result, the objective of
ments contain higher metal concentrations and exhibit this study is to understand distribution of metal contam-
smaller temporal and spatial changes than waster be- ination and grain size in sediments and provide useful
cause of their limited movement. Therefore, river sedi- data for future management and contamination assess-
ments are a useful tool for evaluating continuous envi- ments of river sediments.
ronmental effects (Ra et al. 2013). Specifically, metals
in sediments always exist in the aquatic environment
and have an important influence on benthic organisms. Material and methods
Moreover, they have harmful effects on the hydro-
ecosystem when released into the water and can lead Study area and sediment sampling
to physical/chemical changes (Alloway et al. 1988;
Dekov et al. 1997). Thus, determining the distribution The study area, i.e., the Nakdong River in South Korea,
and behavior of chemical components in sediments, has the basin area of 23,384.21 km2, a mainstream river
including metals, reveals the sediment environment of length of 400.7 km, and a river length of 510.36 km. It is
a river and can be used to provide efficient response located in the southeast of South Korea at a longitude and
measures, such as controls on various environmental latitude of 127° 29′ 19″–129° 18′ 00″ and 34° 59′ 41″–37°
factors (Kim et al. 2001). 12′ 52″. It lies adjacent to the Han River basin to the north
Previous studies on river sediments in South Korea and the Geum River and Seomjin River basins to the west
have involved various analysis methods and evaluations (Jung et al. 2016). The Taebaek Mountains form the east
of organic matter and heavy metals inside the sediments, sea coast basin and watersheds in the east, and the southern
with a particular focus on core and surface sediments (Kim sea area of the Nakdong River lies to the south. The
et al. 2010; Park et al. 2011; Kim et al. 2015a, b). In administrative areas include three metropolitan cities (Bu-
addition, researchers have evaluated contamination levels san, Daegu, and Ulsan Metropolitan Cities) and parts of
using the concentration of metals distributed in the sedi- five provinces (Gyeongsangnam-do, Gyeongsangbuk-do,
ments. These studies typically assess the contamination Jeonlanam-do, Jeonlabuk-do, and Gangwon-do). Further-
level by comparing with the absolute baseline values of more, during the Four River Refurbishment Project, which
sediment quality guidelines (SQGs) derived for each coun- was conducted in order to control floods and secure water
try. Another typical assessment method employs the con- resources that were deemed insufficient for the continu-
centration of metals among crustal materials or in an ously increasing water demands of recent years, river
uncontaminated area as the background concentration for channels were dredged and a total of 16 multi-purpose
comparison (Sekabira et al. 2010; Kim and Um 2013; Md weirs were constructed. Of these, eight were built in the
et al. 2015; Han et al. 2016). Nakdong River (NIER 2017).
Unlike conventional studies that are performed with- This study selected a total of 21 sites in the main-
in a limited spatial area, this study aims to investigate the stream and tributaries of the Nakdong River and collect-
complete sediment environment of the Nakdong River ed surface sediments from September to November
by determining the grain size distribution and metal 2016. For the sediment samples, a Ponar grab was used,
contamination in the mainstream and tributaries of the a type of gravity corer in which the bottom blade closes
Nakdong River system. Moreover, to assess the contam- when the corer touches a basal surface, releasing the
ination level of metals, a comparison is made not only tension. Surface sediments were collected in this way
with the SQGs of South Korea but also with those of from the upper 1–3 cm of sediments. Among the 21
USA and Canada. Based on the measured metal con- sites, 11 are located in the mainstream and 10 in the
centrations, we calculate the EF (enrichment factor), tributaries. Eight of the mainstream sites (NS04, NS07,
Igeo (index of geo-accumulation), and PLI (pollution NS08, NS09, NS10, NS11, NS12, and NS19) are adja-
load index). Then, by assessing metal contamination at cent to weirs (Sangju Weir, Nakdan Weir, Gumi Weir,
each site, we determine the area with the highest con- Chilgok Weir, Gangjeong-Goryeong Weir, Dalseong
tamination for further analysis. Furthermore, the princi- Weir, Hapcheon-Changnyeong Weir, and
pal component analysis (PCA) method is used to Changnyeong-Haman Weir, respectively) (Fig. 1).
Environ Monit Assess (2020) 192: 502 Page 3 of 15 502

Fig. 1 Location of sampling sites in the study area

Analysis of surface sediments treatment process was performed, which decomposed

the organic matter with hydrogen peroxide (H2O2).
The samples used for grain size analysis were first Subsequently, the grain size was measured by a
collected in plastic bottles and then transported to the Microtrac S3500 grain size analyzer that uses the laser
laboratory for analysis. For the grain size analysis, a pre- diffraction principle and calculates the grain size
502 Page 4 of 15 Environ Monit Assess (2020) 192: 502

distribution by measuring the difference in diffraction organisms living in the sediments, and SEL indicates an
patterns when the sediment grains pass through a laser adverse effect on benthic organisms (CCME 1995). The
beam. The measured results were classified into sand, National Institute Environmental Research sediment
silt, and clay according to the sediment composition. pollution evaluation of South Korea provides four levels
The sand was sub-divided into five sizes: very coarse of classification for the effects of sediment metal content
sand, coarse sand, medium sand, fine sand, and very fine on benthic organisms. Level I indicates almost no pos-
sand. The textural parameters for the sediments, such as sibility of toxicity appearing in the benthic organisms.
mean grain size, sorting, and skewness, were obtained Level II indicates a possibility of toxicity, level III
using the method of Folk and Ward (1957) after indicates a relatively high possibility of toxicity, and
converting the weight percentage in each grain size class level IV indicates a very high possibility of toxicity
of sediments. To express the grade of grain size, because (NIER 2015).
the logarithmic scale is more useful than the
equimultiple scale, Φ (phi) was used, and the relation Calculation methods of EF, Igeo, and PLI
between Φ and grain diameter, D, was expressed as Φ =
− log2 D (size in mm). To evaluate metal contamination in sediments, compar-
Samples for metal analysis (Al, Li, Zn, Cr, Pb, Ni, isons are typically made with the metal contents of
Cu, and Cd) in sediments were first mixed onsite at the crustal material or natural concentrations in an uncon-
time of collection using a non-metallic sample spoon taminated area near the study area. These methods can
and then sieved with a 150-μm sieve. Subsequently, estimate the concentrated amount of measured metal
samples were placed in glass bottles and stored in a contents by using the difference or proportion of
portable cool box and were then transported to the established standard elements and natural metal contents
laboratory for analysis. They were then dried in a natural (Kim and Jang 2014). Accordingly, to eliminate the
state and crushed to obtain ground powder samples. The effect of grain size in the sediments, this study used
powder samples were pre-treated by adding HNO3, Al, which is a major element in the sediments and
HClO4, and HF (in a 2:1:2 ratio), and the analysis was exhibits relatively small content variations and a large
performed by an inductively coupled plasma–atomic distribution in the sediments. For the background con-
emission spectrometry (ICP-AES). This analysis meth- centrations, we used the values provided by the River
od was performed in accordance with the “test standard Sediments Background Concentrations of the NIER of
for sediments of river and lake” one of the “water quality South Korea (NIER 2011), which are average concen-
pollution test standards” of the National Institute of trations of metals distributed in South Korean river
Environmental Research of South Korea (Ministry of sediments, considering the local characteristics of the
Environment 2012). study sites. The contamination level was then assessed
by calculating the EF, Igeo, and PLI. The EF is calcu-
Sediment quality guidelines lated by Eq. (1):



M ðmetalÞ M ðmetalÞ
In this study, the metal analysis results were compared EF ¼ sediment = reference value ð1Þ
M ðreferenceÞ M ðreferenceÞ
with the sediment quality guidelines of USA, Canada,
and South Korea (Table 1). The US EPA sediment where (M(metal) / M(reference)) sediment is the con-
quality standard is a standard of the Regional Environ- centration ratio of the metal to the reference element
mental Protection Agency for freshwater sediment con- (Al) in the sediment sample, and where (M(metal) /
tamination classification. According to the content of M(reference)) reference value refers to the ratio of the
each metal element (Zn, Pb, Cu, Ni, and Cd), there are metal to the reference element in the background con-
three standard classifications: non-polluted, moderately centration (Bruland et al. 1974). An EF value of 1.5 or
polluted, and heavily polluted (US EPA 1999). less indicates an uncontaminated natural environment,
Canada’s Ontario sediment guidelines express the ad- whereas EF greater than 1.5 implies artificial contami-
verse effects of sediments on benthic organisms in terms nation due to inflow through the air and river (Zhang
of probability, with three contamination levels: NEL and Liu 2002; Hyun et al. 2007). And Yongming et al.
indicates no effect on the organisms living in the sedi- (2006) suggest deficiency to minimal metal enrichment
ments, LEL indicates no effect on a large number of when EF values are lower than 2.
Environ Monit Assess (2020) 192: 502 Page 5 of 15 502

Table 1 US EPA sediment quality standard, Ontario sediment quality guidelines and NIER sediment pollution evaluation standard (unit:
mg/kg)

US EPA sediment quality standard Ontario sediment quality NIER sediment pollution evaluation
guidelines standard

Non-polluted Moderately polluted Heavily polluted LEL SEL I II III IV

Zn < 90 90–200 > 200 120 820 ≤ 363 ≤ 1170 ≤ 13,000 > 13,000
Pb < 40 40–60 > 60 31 250 ≤ 59 ≤ 154 ≤ 459 > 459
Cu < 25 25–50 > 50 16 110 ≤ 48 ≤ 228 ≤ 1890 > 1890
Cr - - - - - ≤ 112 ≤ 224 ≤ 991 ≤ 991
Ni < 20 20–50 > 50 16 75 ≤ 40 ≤ 87.5 ≤ 330 > 330
Cd - - > 8.0 0.6 10 ≤ 0.4 ≤ 1.87 ≤ 6.09 > 6.09

The Igeo shows the relative concentration level of number of factors (Grant 1990). This is the most com-
metals. Unlike the load coefficient, it can be used to mon multivariate statistical method used in environmen-
evaluate the level of contamination by directly grading tal studies to reduce data and extract a small number of
the contamination level of sediments. Igeo = 0–1 indi- latent factors for analyzing relations among the ob-
cates unpolluted/moderately polluted, 1–2 indicates served variables (Chen et al. 2007). To investigate the
moderately polluted, 2–3 indicates moderately relationship between all factors analyzed in this study,
polluted/strongly polluted, and > 3 indicates strongly grain size analysis (i.e., sediment composition, mean
polluted. Igeo is calculated by Eq. (2), where M(sedi- grain size, sorting, and skewness), analysis of the con-
ment) is the corresponding metal concentration, centration and contamination (i.e., EF, Igeo, and PLI) of
M(background) is the background concentration of the metals in the study area, and principal component anal-
corresponding metal, and a constant of 1.5 is multiplied ysis were conducted using SPSS 20.0 software.
for compensation (Muller 1979).
M ðsedimentÞ
Igeo ¼ log2 ð2Þ
M ðbackgroundÞ  1:5 Result and discussion
The PLI can assess the overall contamination in an
Distribution of grain size
area, including all analyzed metals, unlike Igeo and EF.
It is calculated by Eq. (3), calculated as a ratio of the
According to the grain size analysis results of surface
metal concentration, M(sediment), and the background
sediments in the study area, sand is the dominant grain
concentration of the natural environment state M(back-
size overall (average 88.02%). At the majority of sites,
ground). When PLI is higher than 1, it implies contam-
the sand content exceeds 90% with medium sand the
ination. As the value increases, the contamination level
most dominant class (average 34.96%). At sites NS04
increases (Tomlinson et al. 1980).
(61.7%), NS09 (69.1%), and NS10 (71.9%), the areas
PLI adjacent to weirs constructed in the study area, slightly
s
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi


 ffi
n M ðsedimentÞ M ðsedimentÞ M ðsedimentÞ smaller sand contents are observed compared with other
¼ 1 2… n
M ðreferenceÞ M ðreferenceÞ M ðreferenceÞ sites. Specifically, site NS07 is predominantly com-
ð3Þ posed of fine-grained sediments, with 27.5% sand,
60.2% silt, and 12.1% clay (Fig. 2). Grain size is signif-
icantly influenced by the flow rate and flow velocity
Principal component analysis (Kim et al. 2017). Indeed, flow velocity changes and
temporary stagnation of flowing streams are linked to
Principal component analysis is a statistical analysis the riverbed slope and river channel direction changes
method that groups multiple variables with similar com- caused by the installation of artificial structures (Oh
mon dimensions and reduces them to a relatively small et al. 2003). This agrees with our results, whereby finer
502 Page 6 of 15 Environ Monit Assess (2020) 192: 502

Fig. 2 Sediment composition of surface sediments in the study area

grain sizes are found at sites adjacent to weirs than at and positive skewness is shown. Conversely, in a
other sites because of the effect of decreased flow rate high-energy environment, coarse sediments with
and flow velocity close to the weirs. good sorting tend to show positive or negative
The mean grain size is 1.80 Φ, which corresponds to skewness (Kim et al. 2018). Furthermore, a large
medium sand, and the sand grains are predominantly positive value of skewness indicates dominant ero-
distributed in the ranges of 0–1 Φ (coarse sand) and 1–2 sion due to physical effects such as flow velocity.
Φ (medium sand). However, at sites NS04 and NS07, On the other hand, when the skewness is small,
where silt is dominant, the mean grain size is 3.43 Φ and the area can be classified as one where sedimen-
5.44 Φ, respectively, which are finest grain sizes in the tation is dominant because the physical effects are
study area. Sorting indicates the homogeneity level of relatively weak (Folk 1980). In other words, the
grains through a standard deviation of the grains. A value very high silt content adjacent to the weirs in the
of less than 0.35 indicates very well sorted grains, 0.35–0.5 study area, as well as the poor sorting and low
Φ indicates well sorted, 0.5–1 Φ indicates moderately well skewness, indicates a reduced flow rate and flow
sorted, 0.7–1.0 Φ indicates moderately sorted, 1–2 Φ indi- velocity, which can have large effects on the trans-
cates poorly sorted, 2–4 Φ indicates very poorly sorted, and port of sediments, thereby indicating a low-energy
above 4 Φ indicates extremely poorly sorted. The average environment where sedimentation is active.
value for the study area is 1.40 Φ, which corresponds to
poorly sorted grains. In general, grains are moderated Contamination and concentration of metal
sorted and poorly sorted, whereas sites NS04, NS09, and
NS10 correspond to very poorly sorted grains. Skewness The concentrations of the eight analyzed metals (Al, Li,
shows the asymmetry of grain distributions and indicates Zn, Pb, Cu, Cr, Ni, and Cd) are shown in Table 2. The
the existence or non-existence of coarse and fine fractions. average concentration of Al is 6.9% and exceeds 8% at
As the skewness approaches 0, coarse and fine sediments four sites. The average concentration of Li is 25.6 mg/kg,
are distributed more symmetrically. The average value is exceeding 40 mg/kg at three sites. The average concen-
0.16 for the study area, and positive values are observed at tration of Zn is 120.2 mg/kg and exceeds 200 mg/kg at
all sites except for six that exhibit negative values (Fig. 3). NS09 and NS19. The average concentration of Cr is
The textural parameters derived from the grain 48.5 mg/kg and is highest at site NS19 (83.6 mg/kg).
size analysis reflect the energy conditions in the The average concentrations of Pb and Cu are 24.6 mg/kg
sedimentation environment. In a low-energy envi- and 16.9 mg/kg, respectively. The highest concentra-
ronment, the sediments are fine, sorting is poor, tions of Pb (over 40 mg/kg) and Cu (43.3. mg/kg) are
Environ Monit Assess (2020) 192: 502 Page 7 of 15 502

Fig. 3 Mean grain size, sorting, and skewness of surface sediments in the study area

observed at site NS19, respectively. The average con- these contaminants (Lai et al. 2013). Additionally, metal
centrations of Ni and Cd are 16.08 mg/kg and concentrations of river sediments found in other coun-
0.38 mg/kg, respectively, and their maximum concen- tries, such as China, India, and Australia, generally
tration is observed at site NS19. showed higher concentration distributions in comparison
Table 3 presents the comparison of the sediment metal with this study area. In particular, the Ganga River of
concentration results of the study area (i.e., Nakdong India can be seen to have significantly higher concentra-
River) with the results of Korea’s other three major tions in comparison with the Nakdong River due to the
rivers: Han River, Geum River, and Yeongsan River. effects of domestic, industrial, and agricultural untreated
As shown in the table, the metal concentrations in the wastewater (Song et al. 2011; Duodu et al. 2016; Pandey
sediments of the Nakdong River, i.e., all study sites for et al. 2014). As such, it can be inferred that the sediments
this investigation, showed similar results as the metal found in the Nakdong River showed lower metal con-
concentrations observed in the Geum River (Lee et al. centrations in comparison with other river sediments due
2014) and Yeongsan River (Shin et al. 2015). However, to the lesser amount of anthropogenic impacts due to
sediments of the Han River, which lies more closely domestic, industrial, and agricultural environmental
adjacent to urban areas in comparison with the other factors.
rivers, showed relatively high metal concentrations, in- Table 4 shows the comparison of metal concentra-
cluding Pb and Cu, demonstrating high pollution by tions in the study area with the US EPA sediment quality
502 Page 8 of 15 Environ Monit Assess (2020) 192: 502

Table 2 Metal concentrations of surface sediments in the study area

Al (%) Li (mg/kg) Zn (mg/kg) Cr (mg/kg) Pb (mg/kg) Cu (mg/kg) Ni (mg/kg) Cd (mg/kg)

NS01 7.98 12.0 160.0 11.6 14.8 11.4 8.0 0.72

NS02 7.29 22.2 82.2 12.4 23.0 10.5 10.2 0.12
NS03 4.62 21.9 64.3 36.3 20.8 9.2 12.6 0.17
NS04 7.70 29.2 118.5 44.3 25.4 37.9 17.5 0.36
NS05 4.79 11.6 67.0 29.5 18.2 9.5 10.0 0.19
NS06 4.74 20.7 49.0 52.1 18.0 8.5 11.6 0.32
NS07 6.76 42.3 145.3 55.8 30.4 25.3 27.0 0.39
NS08 7.34 36.0 137.1 50.5 27.2 19.7 20.9 0.39
NS09 6.39 42.1 217.1 51.8 31.9 29.4 28.3 0.42
NS10 8.54 30.9 120.4 50.5 26.6 18.2 19.0 0.46
NS11 4.17 17.2 81.6 62.4 23.8 13.7 11.2 0.53
NS12 5.23 19.3 75.6 49.2 22.5 10.9 12.1 0.52
NS13 8.14 21.2 87.6 23.6 22.5 8.5 4.8 0.26
NS14 7.13 28.6 151.4 29.2 23.8 13.9 8.2 0.22
NS15 5.71 24.7 120.4 62.6 22.3 13.3 15.7 0.36
NS16 8.30 17.8 86.0 72.5 17.6 10.2 17.5 0.34
NS17 8.06 25.6 127.7 79.9 24.4 20.7 26.3 0.34
NS18 7.17 19.6 62.1 61.6 18.7 9.8 13.9 0.29
NS19 8.25 43.9 213.1 83.6 35.5 43.3 32.4 0.51
NS20 6.97 22.4 95.9 56.9 25.6 8.5 13.5 0.45
NS21 7.61 26.7 126.7 28.0 28.0 12.2 15.3 0.40

standards, Canada’s Ontario sediment guidelines, and The EF calculation of surface sediments in the study
South Korea’s NIER sediment pollution evaluations. Ac- area reveals Pb, Zn, Cu, Cr, and Ni values of less than
cording to the US EPA sediment quality standard, ten and 1.5 at all sites, indicating no anthropogenic contamina-
two sites are moderately polluted with Zn and Pb, respec- tion and minimal metal enrichment. However, Zn rela-
tively. Moreover, sites NS09 and NS09 are heavily pollut- tively higher contamination at site NS09 (1.30), adjacent
ed for Zn and Pb, respectively. With respect to Cu and Ni, to the Chilgok Weir. Moreover, Cr values are relatively
no sites are heavily polluted, but six and ten sites are higher at four sites, and Cd, the metal with the highest
moderately polluted, respectively. According to the Ontar- EF values, exhibits values over 1.5 at all except NS01,
io sediment guidelines, one, five, and seven sites corre- NS12, and NS13(adjacent to the weir), indicating that
spond to LEL for Zn, Pb, and Cu, respectively. Further- anthropogenic contamination is pervasive. These results
more, for Cr, all sites excluding the four positions corre- imply that, in general, contamination is relatively high
sponding to NEL correspond to LEL. For Ni, 13 sites and large impacts in weir-constructed areas. Among all
correspond to LEL. For Cd, all sites excluding NS01 metals, contamination by Cd is the highest (Fig. 4).
correspond to NEL. According to the NIER sediment The Igeo value for Cd is 0.26 at site NS01, which
pollution evaluation, the majority of positions correspond corresponds to partially unpolluted/moderately polluted.
to level I, which is an unpolluted level. However, at site Besides this, no other metals or sites exceed Igeo = 0,
NS09, Pb corresponds to level II, and at four sites (NS01, indicating that the contamination level is practically
NS07, and NS08), Cd corresponds to level II, which is the unpolluted. However, in the calculation results, Zn ex-
highest contamination level among all the metals. In the hibits relatively high values at sites NS09 (− 0.57) and
comparative analysis with the SQGs of three countries, the NS22 (− 0.34), and Cu and Cr exhibit relatively high
majority of metal contents indicate high contamination values (− 0.60 and − 0.58, respectively) at site NS19.
levels at sites adjacent to weirs. Because these results show similar trends to the EF
Environ Monit Assess (2020) 192: 502 Page 9 of 15 502

Table 3 Comparison of different metal concentrations in surface sediments in this study and other studies

Location Nation Reference Zn (mg/kg) Cr (mg/kg) Pb (mg/kg) Cu (mg/kg) Ni (mg/kg) Cd (mg/kg)

Nakdong River South Korea This study 49.0–217.0 11.6–83.6 14.8–35.5 8.5–43.3 4.8–32.4 0.12–0.72
Han River South Korea Lai et al. (2013) 52.1–690.7 27.3–146.8 17.1–106.2 5.1–158.5 8.8–57.5 0.05–1.32
Geum River South Korea Lee et al. (2014) 57.5–124.9 37.6–78.6 8.3–19.4 11.8–22.4 9.9–20.9 0.05–0.18
Yongsan River South Korea Shin et al. (2015) 22–167 2–72 14–40 2–26 3–35 0.01–0.09
Brisbane River Australia Duodu et al. (2016) 142–257 82–332 25–126 20–110 20–34 0.6–0.9
Chanjiang River China Song et al. (2015) 50.6–221.0 64.5–126.7 19.0–173.2 10.8–87.9 24.2–52.6 0.35–16.45
Ganga River India Pandey et al. (2014) 137.3–201.2 126.8–196.1 148.8–211.4 12.7–84.0 14.6–82.5 9.5–79.0

results, we confirm that relatively high contamination Unlike the contamination assessments performed
appears at sites adjacent to weirs (Fig. 5). using EF and Igeo, the PLI value can be used to
According to the calculation results of EF and Igeo identify the total contamination level by using the
values, Cd was found to be relatively higher than other concentrations of all analyzed metals. All PLI values
metals, indicating a state of higher pollution. Ganugapenta for all sites in the study area are less than 1, indi-
et al. (2018) previously suggested that the high contami- cating significantly low contamination. Furthermore,
nation degree of Cd is generally attributed to agricultural referring to the findings of one study, which sug-
runoff, industrial activities, and other anthropogenic inputs. gested that the area with contaminated river sedi-
Additionally, the official Group of Experts on the Scien- ments (PLI > 1) was influenced by metal pollutants
tific Aspects of Marine Pollution (GESAMP) ( 1985) from nearby mines, cities, and industrial activities
proposed that the contamination degree of Cd may also (Ahn et al. 2019), the PLI value of the present study
increase due to influential factors, such as agricultural soils, area was shown to be considerably low, indicating
mining waste, and municipal sewage effluents and sludges. that for this area, the anthropogenic impacts on
Accordingly, it was assumed that Cd in sediments found in metals induced by nearby areas were relatively
the NS01 (not adjacent to weirs) showed a relatively high small. However, PLI values also indicate relatively
contamination (EF > 1.5, and Igeo > 0) degree due to high contamination at the sites NS04, NS07, NS08,
anthropogenic impacts. Therefore, additional studies NS09, and NS19, which are located in the weir
should be conducted to explore detailed causes affecting sections. Among these sites, metal contamination is
the Cd concentration levels of sediments observed in the highest at site NS19 (PLI = 0.95), located at Haman
Nakdong River. Weir (Fig. 6).

Table 4 Evaluation for sediment quality guidelines (SQGs) using US EPA, Ontario of Canada, and NIER of Korea

SQGs Site name (metal)

US EPA Non-polluted Other sites (metals)

Moderately polluted NS01, 03, 10, 14, 15 (Zn), NS08 (Zn, Ni), NS07, 08, 09, 17,
19 (Zn, Cu, Ni), NS21 (Zn, Pb, Cu)
Heavy polluted N/D
Ontario of Canada NEL Other sites (metals)
LEL NS14, 15 (Zn), NS16 (Ni), NS01 (Zn, Cd), NS03 (Cu, Ni), NS07,
08, 10, 17 (Zn, Cu, Ni), NS09, 19, 21 (Zn, Pb, Cu, Ni)
SEL N/D
NIER of Korea I level Other sites (metals)
II level NS01, 09, 10, 19, 20 (Cd)
III level N/D
IV level N/D
502 Page 10 of 15 Environ Monit Assess (2020) 192: 502

Fig. 4 Enrichment factors of the metals of surface sediments in the study area

Principal component analysis The factors corresponding to factor 1, which have the
largest effect on the sediment environment of the study
The principal component analysis was performed using area, are PLI, Zn, Pb, Cu, and Ni, EF, Igeo, fine sedi-
the EF, Igeo, and PLI calculation results, which were ments such as very fine sand, silt, and clay, and mean
employed in the grain size, metal concentration, and grain size (Table 5). In the component diagram that
contamination assessment, as the factors analyzed in visualizes the principal components, Zn, Pb, Cu, and
this study. As a result, two principal components were Ni, which are the metals that have no effect on metal
extracted. In the principal analysis results, the eigenval- contamination, are concentrated on the right-hand side
ue, which indicates how much each factor explains the of the diagram, indicating the assessment result of fine
information of existing variables, is 46.08% for factor 1 sediments corresponding to factor 1, PLI (overall con-
and 15.08% for factor 2. tamination of the site), and contamination. Therefore, it

Fig. 5 Index of geo-accumulations of metals of surface sediments in the study area

Environ Monit Assess (2020) 192: 502 Page 11 of 15 502

Fig. 6 Pollution load index of surface sediments in the study area

was determined that grain size has a large influence on or environmental characteristics. As sediment grains
these four metals. Furthermore, coarse sediments (very become finer, the surface area becomes larger. Conse-
coarse sand, coarse sand, and medium sand) have a small quently, more matter can be adsorbed, and the increase
effect on contamination. However, as the concentrations of in ion exchange facilitates the adsorption of metals and
Cd have a relatively weak relationship with the main organic matter (Horowitz 1991). Therefore, we suggest
factor, fine sand, it is determined that Cd has a strong that the study area is predominantly affected by fine
influence on metal contamination in the study area (Fig. 7). sediments. Moreover, because of flow rate and flow
The factors that have large correlation coefficients velocity effects, metal concentrations and contamination
generally exhibit close relationships with geochemical are relatively high at sites adjacent to the weirs, which
502 Page 12 of 15 Environ Monit Assess (2020) 192: 502

Table 5 Component matrix (factors 1 and 2) and eigenvalue loading of surface sediments in the study area (principal component analysis)

Units Factor Factor Units Factor Factor Units Factor Factor Units Factor Factor
1 2 1 2 1 2 1 2

Grain Very coarse − 0.465 − 0.621 Metals Al 0.291 − 0.575 EF Zn 0.752 − 0.172 Igeo Zn 0.772 − 0.437
size sand
Coarse sand − 0.631 − 0.002 Li 0.859 − 0.083 Cr 0.266 0.929 Cr 0.541 0.740
Medium sand − 0.145 0.785 Zn 0.777 − 0.392 Pb 0.484 0.444 Pb 0.832 − 0.025
Fine sand 0.345 0.224 Cr 0.526 0.690 Cu 0.855 0.021 Cu 0.923 − 0.186
Very fine sand 0.726 − 0.110 Pb 0.844 − 0.072 Ni 0.757 0.365 Ni 0.828 0.201
Silt 0.736 − 0.275 Cu 0.879 − 0.169 Cd 0.184 0.434 Cd 0.497 0.149
Clay 0.657 − 0.257 Ni 0.864 0.112 PLI 0.963 0.072 Eigenvalue 46.08% 15.08%
Mz 0.805 − 0.023 Cd 0.395 0.022 loading
So 0.785 − 0.334 -
Sk 0.420 − 0.296

are composed of fine sediments. And the highest con- Conclusions

tamination levels were found at NS19 (adjacent to the
Haman weir). Therefore, structures such as artificially In order to understand the contamination environment
constructed weirs are thought to be the major factor that of surface sediments in the entire Nakdong River sys-
affect grain size distribution and control the contamina- tem, surface sediments were collected from 21 sites in
tion of metals in the sediments of this study area. the mainstream and tributaries of Nakdong River, South

Fig. 7 Plot of principal component analysis of surface sediments in the study area
Environ Monit Assess (2020) 192: 502 Page 13 of 15 502

Korea. Through a grain size analysis of the collected Compliance with ethical standards
surface sediments, the sediment composition, mean
Conflict of interest The authors declare that they have no con-
grain size, sorting, and skewness were obtained. More- flict of interest.
over, the contamination was assessed by analyzing the
metal concentration in the sediments, comparing it with
the SQGs of different countries, and calculating the EF,
Open Access This article is licensed under a Creative Commons
Igeo, and PLI values. Attribution 4.0 International License, which permits use, sharing,
In the study area, sand was the dominant grain size; adaptation, distribution and reproduction in any medium or format,
yet, sites adjacent to weirs were composed of relatively as long as you give appropriate credit to the original author(s) and
fine sediments. This indicated that fine sediments are the source, provide a link to the Creative Commons licence, and
indicate if changes were made. The images or other third party
deposited due to the reduced flow rate and flow velocity material in this article are included in the article's Creative Com-
around weirs. A comparison of the metal concentrations mons licence, unless indicated otherwise in a credit line to the
with the SQGs of USA, Canada, and Korea revealed material. If material is not included in the article's Creative Com-
that contamination exceeded the standard concentra- mons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain
tions at sites adjacent to weirs. EF and Igeo values also permission directly from the copyright holder. To view a copy of
showed high levels of contamination at sites adjacent to this licence, visit http://creativecommons.org/licenses/by/4.0/.
weirs, with Cd contamination the highest. Furthermore,
PLI values revealed that the highest contamination of all
weir-adjacent sites occurred at site NS19, located at References
Haman weir.
In conclusion, anthropogenic contamination in sur- Ahn, J. M., Im, T. H., Lee, K. L., Jung, K. Y., Lee, J. W., Cheon, S.
face sediments of the study area is extremely low, and U., & Park, I. H. (2014). A study on efficiency of water
findings revealed that metal contamination levels are supply through conjunctive operation of reservoirs and
multi-function weir in the Nakdong river. Journal of
influenced by the effect of fine sediments. Moreover, Korean Society Water Environmental, 30(2), 138–147.
metal concentrations tend to accumulate substantially at https://doi.org/10.15681/KSWE.2014.30.2.138.
sites adjacent to weirs, where fine sediments are actively Ahn, J. M., Kim, S., & Kim, Y. S. (2019). Selection of priority
deposited because of reduced flow rate and flow veloc- management of rivers by assessing heavy metal pollution and
ecological risk of surface sediments. Environmental
ity. Among the metals, Cd concentrations exhibited a
Geochemistry and Health. https://doi.org/10.1007/s10653-
weak relationship with fine sediment content and dem- 019-00284-9.
onstrated the greatest effect on metal contamination in Alloway, B. J., Thornton, I., Smart, G. A., Sherlock, J. C., &
the study area. Furthermore, the highest contamination Quinn, M. J. (1988). Metal availability. The Science of the
by metals occurred at site NS19, which is located adja- Total Environment, 75, 41–69. https://doi.org/10.1016/0048-
9697(88)90159-3.
cent to the Haman Weir. Thus, the overall accumulation
Bruland, K. W., Bertine, K., Koide, M., & Golberg, E. D. (1974).
of fine sediments increased due to the influence of weirs, History of metal pollution in southern California coastal
thereby increasing pollution by metal contamination in zone. Environmental Science and Technology, 8(5), 425–
sediments of the study area. 432. https://doi.org/10.1021/es60090a010.
CCME. (1995). Protocol for the derivation of canadian sediment
quality guidelines for the protection of life (report CCME
Acknowledgments The Nakdong River Environment Research
EPC-98E). Canada, Ottawa, https://www.ccme.
Center, National Institute of Environmental Research of Korea,
ca/files/Resources/supporting_scientific_documents/pn_1176_
provided all necessary support for the research.
e.pdf.
Chen, K., Jiao, J. J., Huang, J., & Huang, R. (2007). Multivariate
Authors’ contributions All authors contributed to the study statistical evaluation of trace elements in groundwater in a
conception and design. Material preparation, data collection and coastal area in Shenzhen, China. Environmental Pollution,
analysis were performed by Yong Seok Kim, Deuk Seok Yang, 147, 771–780. https://doi.org/10.1016/j.envpol.2006.09.002.
and Shin Kim. The first draft of the manuscript was written by Dekov, V. M., Komy, Z., Araujo, F., Van Put, A., & Van Grieken, R.
Shin Kim and all authors commented on previous versions of the (1997). Chemical composition of sediments, suspended matter,
manuscript. All authors read and approve the final manuscript. river water and ground water of the Nile (Aswan-Sohag tra-
verse). The Science of the Total Environment, 201(3), 195–210.
Funding information This research was supported by a grant https://doi.org/10.1016/S0048-9697(97)84057-0.
from the National Institute of Environmental Research of Korea Duodu, G. O., Goonetilleke, A., & Ayoko, G. A. (2016). Comparison
(NIER-2019-01-01-078). of pollution indices for the assessment of heavy metal in
502 Page 14 of 15 Environ Monit Assess (2020) 192: 502

Brisbane River sediment. Environmental Pollution, 219, 1077– comparison of grain-size analysis with sieve-pipette method.
1091. https://doi.org/10.1016/j.envpol.2016.09.008. Journal of the Geological Society of Korea, 46, 197–205.
Folk, R. L. (1980). Petrology of sedimentary rock. Texas: Kim, S., Kim, J. E., Lee, K. C., Lee, K. Y., Jeon, H. L., Yu, J. J.,
Hemphill Publishing Company Austin. Lee, I. J., & Ahn, J. M. (2015a). Distribution and pollution of
Folk, R. L., & Ward, W. C. (1957). Brazos river bar [Texas]: A heavy metals in surface sediments from Nakdong river.
study in the significance of grain size parameter. Journal of Journal of Environmental Science International, 24(8),
Sedimentary Petrology, 27(1), 3–26. https://doi.org/10.1306 969–980. https://doi.org/10.5322/JESI.2015.24.8.969.
/74D70646-2B21-11D7-8648000102X1865D. Kim, S., Lee, G. Y., Kim, J. U., Lee, K. C., Ahn, J. M., Lee, I. J.,
Ganugapenta, S., Nadimikeri, J., Chinnapolla, S. R. R. B., Ballari, Jung, K. Y., & Im, T. H. (2015b). Vertical variation of
L., Rajasekhar, M., Madiga, R., Nirmala, K., & Tella, L. P. sediment structure and geochemical characteristics of core
(2018). Assessment of heavy metal pollution from the sedi- sediment in Nakdong river midstream. Journal of Korea
ment of Tupilipalem Coast, southeast coast of India. Society on Water Environment, 31(3), 304–312. https://doi.
International Journal of Sediment Research, 33, 294–302. org/10.15681/KSWE.2015.31.3.304.
https://doi.org/10.1016/j.ijsec.2018.02.004. Kim, S., Lee, K. C., Kim, J. E., Jung, K. Y., Ahn, J. M., Kim, H.
GESAMP (1985). Cadmium, lead and tin in the marine environ- G., Lee, I. J., Shin, D. S., & Yang, D. S. (2017). Change in
ment, UN/FAO/UNESCO/WHO/WMO/IAEA/UNEP joint geochemical characteristics of surface sediments in the
Group of Experts on the Scientific Aspects of Marine Nakdong river main stream. Journal of Environmental
Pollution, GESAMP Report and Studies No.22. Science International, 26(3), 311–324. https://doi.
Grant, A. (1990). Multivariate statistical analysis of sediment org/10.5322/JESI.2017.26.3.311.
geochemistry. Marine Pollution Bulletin, 21(6), 297–299. Kim, S., Ahn, J. M., Kim, H. G., Kwon, H. G., Kim, G. H., Shin,
https://doi.org/10.1016/0025-326X(90)90594-X. D. S., & Yang, D. S. (2018). The distribution characteristics
Han, G. M., Hong, S. H., Shim, W. J., Ra, G. T., Kim, G. T., Ha, S. of grain size and organic matters of surface sediments from
Y., Jang, M., & Kim, G. B. (2016). Assessment of persistent the Nakdong-Goryeong mid-watershed. Journal of
organic and heavy metal contamination in Busan coast: Environmental Science International, 27(6), 411–423.
Application of sediment quality index. Ocean and Polar https://doi.org/10.5322/JESI.2018.27.5.281.
Research, 38(3), 171–184. https://doi.org/10.4217 Lai, T. M., Lee, W., Hur, J., Kim, Y., Huh, I. A., Shin, H. S., &
/OPR.2016.38.3.171. Lee, J. H. (2013). Influence of sediment grain size and land
Horowitz, A. J. (1991). A primer on sediment-trace element use on the distributions of heavy metals in sediments of the
chemistry. Lewis publish, Chelsea, https://pdfs. Han River Basin in Korea and the assessment of anthropo-
semanticscholar.org/bar12/56aa9766c0d3e0b6ea73bb1 genic pollution. Water, Air, and Soil Pollution, 224(7), 1609–
fa9561fbcf120.pdf. 1621. https://doi.org/10.1017/s11270-013-1609-y.
Hyun, S., Lee, C. H., Lee, T., & Choi, J. W. (2007). Lee, J. B., Hong, S. H., Kim, D. H., Huh, I. A., Huh, Y. J., Khan, J.
Anthropogenic contributions to heavy metal distributions in B., Oh, D. Y., Kim, K. Y., Lee, Y. J., Lee, S. H., & Shin, H.
the surface sediments of Masan bay, Korea. Marine Pollution S. (2014). Measurement of metals in sediment of the Geum-
Bulletin, 54(7), 1059–1068. https://doi.org/10.1016/i. River and their correlation. Analytical Science & Technology,
marpolbul.2007.02.013. 27(1), 11–24. https://doi.org/10.5806/AST.2014.27.1.11.
Jung, K. Y., Lee, K. L., Im, T. H., Lee, I. J., Kim, S., Han, K. Y., & Md, S. I., Md, K. A., Mohammad, R., Md, H. A. M., &
Ahn, J. M. (2016). Evaluation of water quality for the Muhammad, K. I. (2015). Heavy metal pollution in surface
Nakdong river watershed using multivariate analysis. sediment: A preliminary assessment of an urban river in a
Environmental Technology and Innovation, 5, 67–82. developing country. Ecological Indicators, 48, 282–291.
https://doi.org/10.1016/j.eti.2015.12.001. https://doi.org/10.1016/j.ecolind.2014.08.016.
Kim, J. G., & Jang, H. S. (2014). Evaluation of characteristics of Ministry of Environment. (2012). Official test methods of water
particle composition and pollution of heavy metals for bot- quality (report no. 2012-99). Sejong: Ministry of
tom sediments in Cheonsu bay, Korea-Comparison of the Environment.
sediments environment of farming area and non-farming Muller, G. (1979). Heavy metals in the sediment of the Rhine–
area. Journal of Korea Society of Marine Environment and changes seity. Umschau in Wissenschaft und Technik, 79,
Safety, 20(4), 358–371. https://doi.org/10.7837 778–783.
/kosomes.2014.20.4.358. NIER (2011). Baseline concentrations of heavy metals in river
Kim, D. H., & Um, H. H. (2013). Estimation of the sediment pollution sediments in Korea (report no. TRKO201300007691).
in coast of Gwangyang, Mokpo and Shinan, Korea. Journal of Inchun: National Institute Environmental Research.
Korea Society of Marine Environment and Safety, 19(4), 303– NIER (2015). Sediment pollution evaluation standard of river and
308. https://doi.org/10.7837/kossmes.2013.19.4.303. lake (published rulings 2015-687). Inchun: National Institute
Kim, J. Y., Eun, G. Y., Koh, Y. K., Youn, S. T., Oh, K. H., & Kim, D. Environmental Research.
J. (2001). Sedimentary environments and geochemical charac- NIER (2017). Result on target water quality monitoring network
ters of the core sediments near Naju-Yeongam area in the lower operation of Nakdong River System in 2017 (report no. 11-
part of the Yeongsan river, Cheonnam, Korea. Journal of the 1480000-0001481-14). Inchun: National Institute
Korean Earth Science Society, 22(4), 301–316. Environmental Research.
Kim, M. J., Chang, T. S., Rhee, C. W., & Jin, J. H. (2010). Oh, K. H., Kim, J. Y., Koh, Y. K., Youn, S. T., Shin, S. E., Park,
Feasibility assessment of a microtrac tri-laser diffraction par- B. Y., Moon, B. C., & Kim, H. G. (2003). Geochemical
ticle size analyzer for use in sedimentary geology: A characteristics and contamination of surface sediment in
Environ Monit Assess (2020) 192: 502 Page 15 of 15 502

streams of Gwangju city. Journal of the Korean Earth apportionment of heavy metal contaminants in the bottom
Science Society, 24(4), 346–360. sediments of lower reach of Changjiang River. Catena,
Pandey, M., Tripathi, S., Pandey, A. K., & Tripathi, B. D. (2014). 85(1), 73–81. https://doi.org/10.1016/j.catena.2010.12.009.
Risk assessment of metal species in sediments of the river Thornton, I. (1983). Applied environmental geochemistry.
ganga. Catena, 122, 140–149. https://doi.org/10.1016/j. London: Academic Press.
catena.2014.06.012. Tomlinson, D. L., Wilson, J. G., Harris, C. R., & Jeffrey, D. W.
Park, J. E., Kim, J. G., & Ahn, S. Y. (2011). Distribution and (1980). Problems in the assessment of heavy metals in estu-
source identification of PCDD/Fs and Co-PCBs in sediments aries and the formation pollution index. Helgoland Scientific
from the Geum river. Korea Environmental Engineering Meeresunteruchung, 3(1–4), 566–575. https://doi.
Research, 33, 900–906. https://doi.org/10.4491 org/10.1007/BF02414780.
/KSEE.2011.33.12.900. US EPA (1999). Protocol for developing sediment TMDL (report
Ra, K. T., Kim, E. S., Kim, J. K., Lee, J. M., & Kim, E. Y. (2013). EPA 841-B-99-004). Washington: US EPA. https://purl.
Distribution and pollution assessment of trace metals in core access.gpo.gov/GPO/LPS50983.
sediments from the artificial lake Shihwa. Journal of Ocean Yongming, H., Peixuana, D., Junji, C., & Posmentier, E. S.
Polar Research, 35(2), 69–83. https://doi.org/10.4217 (2006). Multivariate analysis of heavy metal contamination
/OPR.2013.35.2.069. in urban dust of Yi’an, central China. Science of the Total
Sekabira, K., Oryem Origa, H., Basamba, T. A., Mutumba, G., & Environment, 355, 176–186.
Kakudidi, E. (2010). Assessment of heavy metal pollution in
Zhang, J. H., & Liu, C. L. (2002). Riverine composition and
the urban stream sediments and its tributaries. International
estuarine geochemistry of particle metals in China-
Journal of Environmental Science and Technology, 7(3),
Weathering features, anthropogenic impact and chemical
435–446. https://doi.org/10.1007/BF03326153.
fluxes. Estuarine, Coastal and Shelf Science, 54(6), 1051–
Shin, S. E., Youn, S. T., Koh, Y. K., & Oh, K. H. (2015). The
1070. https://doi.org/10.1006/ecss.2001.0879.
geochemical characteristics and environmental changes of
surface sediment in Yeonsan River area. Journal of the
Korean Geomorphological Association, 22(3), 31–42. Publisher’s note Springer Nature remains neutral with regard to
https://doi.org/10.16968/JKGA.22.3.31. jurisdictional claims in published maps and institutional
Song, Y., Ji, J., Yang, Z., Yuan, X., Mao, C., Frost, R. L., & affiliations.
Ayoko, G. A. (2011). Geochemical behavior assessment and