Heavy metal concentrations in natural and human-impacted sediments of Segara Anakan Lagoon, Indonesia

  • A. D. Syakti
  • C. Demelas
  • N. V. Hidayati
  • G. Rakasiwi
  • L. Vassalo
  • N. Kumar
  • P. Prudent
  • P. Doumenq
Article

DOI: 10.1007/s10661-014-4079-9

Cite this article as:
Syakti, A.D., Demelas, C., Hidayati, N.V. et al. Environ Monit Assess (2015) 187: 4079. doi:10.1007/s10661-014-4079-9

Abstract

The concentrations of eight elements (Cr, Cu, Fe, Mn, Ni, Ti, V, and Zn) in surface sediments from Segara Anakan Nature Reserve (SARN), Indonesia, were determined using inductively coupled plasma–atomic emission spectroscopy following microwave-assisted acid digestion. In general, the heavy metal concentrations of the sediments were found to decrease in the sequence Fe > Ti > Mn > Zn > V > Cu > Cr > Ni. Sediment pollution assessment was carried out using a pollution status index contamination factor, pollution load index, geoaccumulation index, and enrichment factor as well as by comparing the measured values with two sediment quality guidelines, i.e., threshold effect level and probable effect level. The evaluation showed that in the refinery site stations, Cr, Ni, and Zn concentrations found in the SANR sediments may cause the adverse effect to occur over a wider range of organisms and can contribute to a more serious harmful effect.

Graphical Abstract

Keywords

Heavy metals River basin Environmental chemistry Sediment quality assessment Industrial effluents 

Introduction

The Segara Anakan Nature Reserve (SANR) is located on the south coast of Java, surrounded by an area of sloughs, tributaries, mangrove swamps, and intertidal lands converted to rice fields. Such an ecosystem provides a unique and abundant aquatic ecosystem and a productive marine nursery because the influence of the Indian Ocean may enter lagoon troughs in the western and eastern passages (Fig. 1). Consisting of a mangrove-fringed lagoon, SANR and its surrounding environment in the Cilacap coastal area are considered a unique ecological feature in Java, Indonesia. The actual threat to the ecosystem is sediment input, which is generated in highland areas due to volcanic eruption, deforestation, and poor agricultural practices resulting in increased soil erosion. During the last few decades, such valuable ecological resources are being degraded by human activities and the continuous discharge of environmental pollutants such as persistent organic pollutants (Dsikowitzky et al. 2011) and heavy metals (Noegrohati 2005). The heavy metals enter the SANR through fresh water from the volcanic area of West Java (i.e., the Citanduy River basin and sea water from the Indian Ocean), mostly by tidal action through the western side. White et al. (1989) reported various natural heavy metal inputs associated with the runoff from rivers that are continuously entering the SANR. The heavy metals are being settled, redistributed, and accumulated in the ecosystem. Another potential input arises via point source discharges of municipal sewage, industrial facility effluents, and non-point-source discharges from domestic waste, fisheries, and agriculture (Yuwono et al. 2007). At the present time, many techniques and methods of determining the major and trace metal elements are available, such as graphite furnace atomic absorption spectroscopy (GF-AAS), flame atomic absorption spectroscopy (F-AAS), cold vapor atomic fluorescence spectroscopy (CV-AFS), inductively coupled plasma mass spectrometry (ICP–MS), and inductively coupled plasma–atomic emission spectroscopy (ICP–AES) (Bressy et al. 2013). The latter has been widely used because of many advantages such as rapid analysis, wide linear range, low detection limit, and simultaneous determination of multiple elements (Qing-Hua et al. 2012). The aim of this study was to determine heavy metal concentrations in the SANR sediments using ICP–AES after microwave-assisted acid digestion. The analytical performance of the entire procedure (i.e., limit of detection, precision, and accuracy) was evaluated statistically. From the perspective of an environmental monitoring program, involving the element trace metals, sediment quality assessment using sediment quality guidelines (SQGs) is very important in the SANR area, and several indices, including contaminant factors (CFs), pollution load index (PLI), enrichment factors (EFs), and geoaccumulation index (Igeo) (MacDonald et al. 2000; Farkas et al. 2007), presumably designed to protect the aquatic biota from the deleterious effects associated with sediment-bound contaminants, to rank chemicals of concern for further investigation and/or to prioritize an intermediate intervention to ameliorate the contaminated areas, have been used (Angula 1996; Long et al. 2005). Our investigations have focused on the assessment of (1) spatial and temporal trends of Cr, Cu, Fe, Mg, Ni, Ti, V, and Zn load in the SANR superficial sediments, and (2) environmental risks of the actual heavy metal loads in the study area by comparison with several indices indicating the degree of pollution (EFs, Igeo, CFs, PLI) and the sediment quality guidelines (SQGs).
Fig. 1

Map showing detail for sampling sites in the Segara Anakan Nature Reserve, Indonesia

Materials and methods

Sampling

Sampling was carried out in September and November of 2011. The sampling site (Fig. 1) consisted of 34 stations characterizing the four main rivers (11 stations) entering Segara Anakan (Citanduy, R1; Cibeureum, R2; Sapuregel, R3; Donan, R4) and two stations representing human-impacted activities, i.e., oil refinery sites (RS1 and RS2). Twelve stations covered the center of the lagoon receiving water (SA1–SA12), nine stations (MR1–MR9) were in the sea area with different land cover and land use characteristics (i.e., estuary, marine harbor, and Indian Ocean), and two stations were close to an oil refinery and cement plants, which are also located at the Donan River site. A Global Positioning System (GPS; Garmin Etrex Summit HC, Kansas City, USA) and Geographic Information Systems (GIS; MapInfo Professional 7.5, Boulogne-Billancourt, France) were used to target the different sites. Geographical Information System software (Arch-ViewR GIS 3.2a) was used to delineate the metal accumulation areas in the SANR sediments. Superficial sediments (0–5 cm) were collected using a PVC core sampler, then immediately placed in plastic containers and stored in a coolbox to minimize microbial degradation during transport to the laboratory. Sediment samples were subsequently freeze-dried over the period of a week in order to obtain a constant mass, sieved with stainless steel sieves at 200 μm, then the grain size was normalized granulometrically based on isolation of one, exclusively silty-clay fraction (<63 μm) prior to the trace element concentration analysis.

Reagents, solutions, and samples

All laboratory ware was soaked in a 10 % (v/v) HNO3 solution bath for 24 h and was rinsed with high-purity water. All materials were subsequently dried under clean air conditions at ambient temperature. All solvents and reagents were of the highest commercially available purity grade. Deionized water (resistivity 18 MΩ cm−1) that was obtained from a Milli-Q Pluswater purification system (Millipore, Molsheim, France) was employed to prepare all standard and sample solutions. Suprapur grade 65 % (m/m) HNO3 and 37 % (m/m) HCl (Merck, Germany) were used for sample mineralization. Monoelemental high-purity grade 1 g L−1 stock solutions of Cr, Cu, Fe, Mn, Ni, V, and Zn were purchased from Merck (Darmstadt, Germany). The purity of the plasma torch argon was greater than 99.99 %.

Digestion procedure and analyses

The acid digestion of the sediments was performed using a commercial high-pressure laboratory microwave oven (Milestone start D 5 Microwave Labstation, Sorisole, Italy) operating at maximum temperature and pressure (300 °C and 100 bar). This microwave digestion system was equipped with ten 100-mL tetrafluoromethoxy vessels and a ceramic vessel jacket. To protect the unit, the cavity and the door were plasma-coated with Polytetrafluoroethylene (PTFE). Approximately 0.5 g of dry sediment was digested with aqua regia, a mixture of HNO3 and HCl (volume proportion ratio 1:2) according to the AFNOR NF X31-151 (AFNOR 1994) standard. The heating program was performed in four successive steps. After the digestion procedure and subsequent cooling, the digested samples were diluted to a final volume of 25 mL with water.

Inductively coupled plasma–atomic emission spectroscopy (ICP–AES)

The mineralization products were filtered with a 0.45-μm mesh, and extracts were analyzed for metals (Cr, Cu, Fe, Mg, Mn, Ni, Ti, V, and Zn) content using ICP–AES (JY 2000 Jobin Yvon Horiba). The emission lines for the analysis by ICP–AES were (nm): Cr (267.716), Cu (324.754), Fe (259.940), Mn (257.610), Ni (231.604), Ti (334.941), V (311.071), and Zn (213.856). Blanks were prepared for each lot of samples, and triplicate analyses were performed for each sample. Accuracy of the method was tested by analyzing Trace Metals-Sandy Clay 1 certified reference materials (CRM 049-050 from RTC, USA).

Energy dispersive x-ray fluorescence spectrometry

The major mineral phases were identified with an x-ray diffractometer (X’Pert Pro MPD; Panalytical) using Co Kα radiation (λ = 1.79 Å) running at 40 kV and 40 mA, with a linear detector X’Celerator and secondary flat monochromator. Samples were placed on a zero-background silicon plate and spun at 15 rpm. A counting time of 2,500 s per 0.033° step was used for 2θ in the 5–70° range. The International Center of Diffraction Data PDF-2 database with the X’Pert Highscore Plus software (Panalytical) was used to identify the mineral phases from the x-ray diffraction (XRD) pattern obtained.

Assessment of sediment contamination

In this study, four different indices were used to assess the degree of heavy metal contamination in sediments from the SANR, Indonesia. Because there were no data on geochemical background levels for the Segara Anakan sediment and soils, the reference baseline of background values used the abundance of elements in crustal rocks as follows (mg kg−1): 122, 68, 62000, 1060, 99, 6320, 136, and 76 for Cr, Cu, Fe, Mn, Ni, Ti, V, and Zn, respectively (Greenwood et al. 1997).

Contamination factor (CF)

The CF is the ratio obtained by dividing the concentration of each metal in the sediment by the baseline or background value (Hakanson 1980).
$$ \mathrm{C}\mathrm{F}=\frac{\left[\mathrm{heavy}\kern0.5em \mathrm{metal}\right]}{\left[\mathrm{background}\right]} $$

As suggested, the CF values were classified into four groups: CF < 1 indicates low contamination, 1 < CF < 3 is moderate contamination, 3 < CF < 6 is considerable contamination, and CF > 6 is very high contamination.

Pollution load index (PLI)

For the entire sampling site, PLI has been determined as the nth root of the product of the n CFs:
$$ \operatorname{PLI}={\left(\operatorname{CF}1\times \operatorname{CF}2\times \operatorname{CF}3 \times \dots \times {\operatorname{CF}}_n\right)}^{1/n} $$

This empirical index provides a simple comparative means for assessing the metal contamination status. When PLI >1, pollution exists; otherwise, if PLI <1, there is no metal pollution (Tomlinson et al. 1980).

Enrichment factor (EF)

Enrichment factor (EF) is a useful tool for determining the degree of anthropogenic heavy metal pollution (Sakan et al. 2009). The EF is computed using the relationship below:
$$ \mathrm{E}\mathrm{F}=\frac{\left[\mathrm{heavy}\kern0.5em \mathrm{metal}/\mathrm{F}\mathrm{e}\right]}{\left[\mathrm{heavy}\kern0.5em \mathrm{metal}\kern0.5em \mathrm{background}/\mathrm{F}\mathrm{e}\kern0.5em \mathrm{background}\right]} $$

In this study, iron (Fe) was used as the reference element for geochemical normalization for the following reasons: (1) Fe is associated with fine solid surfaces, (2) the geochemistry of iron is similar to that of many trace metals, and (3) the natural concentration of iron tends to be uniform (Bhuiyan et al. 2010). Therefore, Fe acts as a normalizer to correct for differences in sediment grain size and mineralogy. EF values were interpreted as suggested by Sakan et al. (2009), where EF <1 indicates no enrichment, 1–3 is minor enrichment, 3–5 is moderate enrichment, 5–10 is moderately severe enrichment, 10–25 is severe enrichment, 25–50 is very severe enrichment, and >50 is extremely severe enrichment.

Geoaccumulation index (Igeo)

The geoaccumulation index (Igeo) is defined by the following equation:
$$ {I}_{\mathrm{geo}}= \log {}_2\frac{\left[\mathrm{heavy}\kern0.5em \mathrm{metal}\right]}{1.5*\left[\mathrm{background}\right]} $$

A factor of 1.5 is the background matrix correction factor due to lithospheric effects. The geoaccumulation index consists of seven classes (Muller 1981): Class 0 (practically unpolluted), Igeo ≤0; Class 1 (unpolluted to moderately polluted), 0 < Igeo < 1; Class 2 (moderately polluted), 1 < Igeo < 2; Class 3 (moderately to heavily polluted), 2 < Igeo < 3; Class 4 (heavily polluted), 3 < Igeo < 4; Class 5 (heavily to extremely polluted), 4 < Igeo < 5; Class 6 (extremely polluted), 5 > Igeo (Bhuiyan et al. 2010).

Sediment quality guidelines

The guidelines comprise two assessment levels. The lower level is referred to as the threshold effect level (TEL) and represents the concentration below which adverse biological effects rarely occur (i.e., biosentinel species). The higher level, known as the probable effect level (PEL), defines a concentration above which adverse effects are expected to occur over a wider range of organisms. The intermediate level (between TEL and PEL) indicates those concentrations that are rarely, occasionally, and frequently associated with adverse biological effects. The available reference data for TEL and PEL, however, are limited for Cr, Cu, Ni, and Zn (MacDonald et al. 2000).

Statistical analysis

Descriptive data analysis was performed using Excel v.2000 for pseudo-total element concentrations, accompanied by correlation analysis to determine relationships among different metals. Possible associations between different variables were assessed by applying Spearman’s rank correlation analysis with the help of SigmaPlot for Windows® version 11.0 (Systat Software, Inc., wpcubed, GmbH, Germany).

Results and discussion

Analytical performance

Initially, ICP–AES attempted to determine a broader range of elements, including Al, As, Cd, and Pb. However, the aluminum (Al) concentrations were phenomenal (>> 1 g L−1), which can present problems for detector saturation and the plasma interference of some trace element emissions such as Pb, Cd, and metalloid As. A dilution factor did not help to quantify these metals and metalloids because the quantitative analysis was limited by the Limit of Detection (LOD) and Limit of Quantification (LOQ) of other trace elements. Elemental analysis wavelength, the detection limit (n = 10), and the accuracy of the method were determined by analyzing Trace Metals-Sandy Clay 1 certified reference materials (CRM 049-050 from RTC, USA). The results showed good agreement with the certified values of the CRM sample. Accuracy results ranged from 91.4 to 96.1 % (Table 1).

Heavy metals in sediments of the SANR

The basic statistics for the Cr, Cu, Fe, Mn, Ni, Ti, Zn, and V concentration measured at 34 stations (Fig. 1) are summarized in Tables 1 and 2. In general, the heavy metal concentrations of the sediments were found to decrease in the sequence Fe > Ti > Mn > Zn > V > Cu > Cr > Ni. All heavy metals showed no significant spatial variations (ANOVA, p > 0.05), suggesting a uniform distribution of those heavy metals with similar precipitation behavior. The ranges of metals in sediments were as follows: 17–73 mg kg−1 for Cr, 16–84 mg kg−1 for Cu, 26,978–103,031 mg kg−1 for Fe, 440–2,490 mg kg−1 for Mn, 13–32 mg kg−1 for Ni, 1,403–4,849 mg kg−1 for Ti, 117–415 mg kg−1 for V, and 39–154 mg kg−1 for Zn (Fig. 1, Table 2). This calculation (above) does not take into account the station RS (1 and 2) because the highest concentrations of most of the heavy metals except for Cu were found at RS, which is located on the oil refinery wastewater discharge outfall in Cilacap.
Table 1

Elemental analysis wavelength and detection limits (n = 10)

Elements

Wavelength (nm)

Reference value (mg kg−1)

Analytical value (mg kg−1)

Recovery (%)

LOD (μg L−1)

LOQ (μg L−1)

Cr

267.716

85 ± 10

78 ± 2

91.4 ± 2.1

10.71

19.39

Cu

324.754

32 ± 3

31 ± 0.7

98.4 ± 2.1

0.54

1.81

Fea

259.940

34,020 ± 770

29,320 ± 920

86.4 ± 2.4

3.75a

12.49a

Mna

257.610

620 ± 30

570 ± 22

92.0 ± 3.5

0.73a

1.61a

Ni

231.604

32 ± 4

30 ± 2

92.7 ± 5.6

4.18

13.92

Tia

334.941

5,500 ± 250

5,374 ± 159

97.8 ± 2.9

19.43

21.99

Zn

213.856

78 ± 5

69 ± 1

88.2 ± 1.6

8.47

12.81

V

311.071

97 ± 8

93 ± 1

95.4 ± 2.4

0.74

2.47

The accuracy of the method was tested by analyzing Trace Metals-Sandy Clay 1 certified reference materials (CRM 049-050 from RTC, USA)

aTheir respective LOD and LOQ were in milligrams per liter

Table 2

Heavy metal concentrations (mg kg−1) in surface sediments from Segara Anakan nature reserve, Indonesia

Stations

Cr

Cu

Fe

Mn

Ni

Ti

V

Zn

(mg kg−1)

R1-1

65

45

74,192

983

28

3,964

333

109

R1-2

55

72

71,778

1,628

28

2,343

230

112

R1-3

50

69

68,100

1,386

27

2,248

212

107

R2-1

56

66

64,452

1,118

32

1,462

196

110

R2-2

51

69

65,323

1,454

29

1,926

199

111

R2-3

49

39

68,490

815

25

3,171

259

119

R3-1

49

39

68,490

815

25

3,171

259

119

R3-2

44

46

46,066

440

23

1,403

146

93

R3-3

49

51

51,342

479

25

1,757

162

102

R4-1

50

78

62,888

653

29

1,801

159

132

R4-2

50

84

57,509

657

30

1,638

159

132

RS-1

340

50

467,813

2,901

73

9,401

1,317

480

RS-2

336

49

456,731

2,828

70

6,302

1,260

472

SA-1

49

66

64,851

1,065

26

2,198

201

106

SA-2

53

65

65,995

1,200

27

2,075

211

105

SA-3

47

69

67,672

1,465

26

2,203

204

102

SA-4

51

59

64,835

1,567

27

2,041

205

100

SA-5

58

68

66,129

1,409

29

2,132

215

109

SA-6

50

63

66,311

940

26

2,377

209

101

SA-7

52

72

66,171

1,517

27

2,047

213

111

SA-8

64

55

72,648

1,284

27

2,959

295

109

SA-9

60

58

71,393

1,145

27

3,266

268

154

SA-10

51

32

103,031

1,332

24

4,849

415

145

SA-11

39

54

75,087

746

28

2,301

222

92

SA-12

27

28

58,072

2,490

24

2,177

158

90

MR-1

33

23

65,622

946

15

4,135

315

95

MR-2

28

19

40,608

1,045

13

2,603

190

58

MR-3

73

42

73,272

1,067

25

3,733

306

111

MR-4

17

18

26,978

1,055

15

1,990

117

39

MR-5

29

22

53,291

1,007

13

3,430

270

75

MR-6

22

16

29,760

1,026

15

2,298

143

42

MR-7

27

18

35,276

1,182

17

2,807

186

51

MR-8

25

25

54,731

1,054

21

2,212

163

90

MR-9

34

35

67,488

2,023

29

2,790

205

107

Minimum

17

16

26,978

440

13

1,403

117

39

Maximum

340

84

467,813

2,901

73

9,401

1,317

480

Mean

63

49

85,659

1,257

27

2,859

282

123

Std

71

20

96,635

577

12

1,540

263

93

Maximum (−RS)

73

84

103,031

2,490

32

4,849

415

154

Mean (−RS)

46

49

62,120

1,156

24

2,547

220

101

Std (−RS)

14

21

14,698

421

5

813

64

26

Indicated values correspond to means of three different analyses. The relative standard deviation did not exceed 14 % of the mean

When we grouped the sediment stations according to their respective land cover and land use (i.e., riverine, lagoon, refinery site, and sea), we demonstrated that the concentration factors of the heavy metals at station RS were 6–10 times greater than the riverine station (R), lagoon (SA), and sea stations (MR) (Table 3). In this study, total metal concentrations followed the order of RS > SA ≈ R > MR. Average contamination at the stations RS and SAL did not show significant differences (ANOVA, p > 0.05). From the standard deviations given in Table 2, the distribution of heavy metals was relatively uniform throughout all stations, suggesting a natural geology of the crustal contribution of the elements in the area, except in the case of RS1 and RS2 where the high concentrations may be related directly to the impact of the oil refinery. Moreover, our particular interest in Ti and V augmentation at this site (3–6-fold) suggested the oil refinery impact on the environment because Ti and V were constant in the common petroleum source rocks and dependent on the geological age. Those ratios were used for tracing source effects (Sainbayar et al. 2012). Khala et al. (1982) proposed V as a tracer for oil pollution because Khala et al. found that the elevated V distributions appeared to be related to the extensive oil loading facilities and commercial vessels in Kuwait Bay.
Table 3

Heavy metal concentrations in the sediment samples from SANR and other selected ecosystems from the literature

Location

Heavy metals (mg kg−1)

 

References

Cr

Cu

Fe

Mn

Ni

Ti

V

Zn

Indonesia

 SANRa

17–73

16–84

26,978–103,301

440–2,490

13–32

1,403–4,849

117–415

39–154

This study

 Semarang coast

33–72

1–29

84–259

Takarina et al. (2004)

 Dumai coast

2–14

21,000–39,200

7–20

31–87

Amin et al. (2009)

 Java sea

6–54

1,100–131,300

33–122

Everaarts (1989)

 Jakarta Bay

3–128

18–36

4–595

William et al. (2000)

Asia

 Gomati river, India

2–89

4–254

66–835

5–76

8–343

Singh et al. (2005)

 Luan river, China

29–153

6–179

21–26

Liu et al. (2009)

 Tigris river, Turkey

76–152

673–5,076

822–1,657

152–288

191–2,396

Varol et al. (2011)

 Kuwait coastal

10–77

Khala et al. (1982)

 Egyptian coastal

8–214

El-Moselhy (2006)

Africa

 Nile river

8–274

10–81

75–2,810

2–112

11–221

Rifaat (2005)

 Nador, Morocco

9–139

4–446

 

2–62

4–1,190

Bloundi et al. (2009)

 El Melah, Tunisia

35–156

nd–309

118–347

20–128

34–228

Prudencio et al. (2007)

Europe

 Berre, France

38–428

11–48

18–56

50–151

Accornero et al. (2008)

 Millos bay, Greece

119

51

34,000

1685

61

3,000

325

Karageorgis et al. (1998)

 Danube river, Europe

26–556

31–8,088

42–1,655

17–173

78–2,010

Woitke et al. (2003)

 Taranto Gulf, Italia

75–103

42–52

26,313–36,098

552–2,829

48–61

2,343–3,876

 

87–129

Buccolieri et al. (2006)

America

 Almendares river, Cuba

84–234

72–421

86–709

Olivares-Rieumont et al. (2005)

 Estuarine lagoon, Puerto Rico

12–211

16,000–72,000

    

25–1,530

Acevedo-Figueroa et al. (2006)

 South Platte river, USA

33–71

18–480

410–6,700

82–3,700

Heiny and Tate (1997)

aStation RS1 and RS2 were excluded

The average values of metals observed in the SANR sediments (Table 2) were comparable to other studies reported in the literature (Table 3). Cr concentrations were still within the range found in other studies, but we noted that the values observed at station RS were higher than other stations (three to four times). This result was similarly found in Berre Lagoon (France), which is also in the surroundings of an oil refinery (Accornero et al. 2008) and an international waterway, and the Danube River in Europe (Woitke et al. 2003). Figure 2 gives idea representation of the human impact on Cu concentrations in a suburban area where the population lives in coastal areas near the three estuaries (R1, R2, and R3). Regarding Cu, the concentration distributions were comparable to the concentrations found in other geographic areas including Indonesia, but lower than the Danube River (Woitke et al. 2003), Almendares River, Havana City, Cuba (Olivares-Rieumont et al. 2005), and a metallic wastewater discharge point from a copper mine in Maden Township (Varol 2011).
Fig. 2

Trace-element distributions in the SANR surface sediments. All classes of concentrations for each of the studied metals are in milligrams per kilogram of dry weight

Compared to other regions in Indonesia and worldwide, the high Fe concentrations of the SANR sediments are not surprising because the composition of fluvial sediments is controlled predominantly by lithogenic influences through fresh waters from the volcanic area of West Java (i.e., Citanduy) entering SANR, including river runoff from terrigenous sedimentary formations rich in Fe-oxyhydroxides (Avila-Perez et al. 1996). A typical XRD scan is shown as Fig. 3. Magnetite (Fe3O4), hematite (Fe2O3), and quartz (SiO2) were the main crystalline phases identified in the SARN sediment. Traces of ilmenite (FeTiO3) and hornblende (Ca2(Mg,Fe,Al)5(Al,Si)8O22(OH)2) were also detected. Iron appeared to be incorporated in different minerals. However, iron (II) seems to be associated with magnetite and occurs in traces in ilmenite, and iron (III) is associated mainly with a hematite mineral. This finding demonstrated that the sediments constituted mainly of silicates and aluminum silicates in the form of clays. A previous study conducted in the SANR area using x-ray diffraction confirms the presence of iron nanoparticles with estimated sizes of 18–70 nm and an amorphous structure for the samples (Widanarto et al. 2013).
Fig. 3

XRD pattern for SANR sediment showing incorporation of iron into different minerals

Generally, the concentrations of Mn, Ni, and Zn in the SANR sediments were of the same order of magnitude as the concentrations of these metals in other areas, but the concentrations were still three times smaller than in the uranium mining area of the South Platte River in the USA (Heiny and Tate 1997). From Fig. 2, we noted a high concentration of Mn at SA12 and MR9, the stations near the refinery site, cement plants, and recreational coast site near Cilacap downtown. As identified by WHO (2004), the major anthropogenic sources of environmental manganese include sewage sludge, mining and mineral processing, combustion of fossil fuels, and, to a much lesser extent, emission from the combustion of fuel additives.

The average values of Ti observed in the SANR sediments (Table 3) are comparable to values observed in Millos Bay (Greece) and Taranto Gulf (Italy). Ti, which is relatively abundant in the earth’s crust, is estimated as 6,320 mg kg−1 (Greenwood et al. 1997), so the concentration of Ti is of a natural origin. Figure 3, shows titanium as a common lithophile metallic element that forms several minerals, including ilmenite (FeTiO3) and magnetite, which most likely formed during magmatic processes. Ti follows Fe in magmatic crystallization in the volcanoes in the surrounding SANR. A comparison of V concentration with other studies showed two to four times greater magnitude in SANR sediments than in Egyptian sediments (El-Moselhy 2006) and in coastal Kuwait areas (Khala et al. 1982). Such incremental concentrations might be contributed by the anthropogenic influence. The following section describes the use of several indices, including the contaminant factor (CF) and pollution load index (PLI), enrichment factor (EF), and geoaccumulation index (Igeo) to better estimate the contamination status of the studied area (MacDonald et al. 2000; Farkas et al. 2007).

Sediment contamination status (CF, PLI, EF, and Igeo)

A recent study conducted by Syakti et al. (2013) concomitantly with this study showed average extractable organic matter (OM) content for riverine stations varied from 630 to 1,680 mg kg−1 (R stations). The highest OM was found at the refinery site (RS) station with the extent up to 45,630 mg kg−1 in the eastern SARN. Lagoon (SA) and marine (MR) stations showed OM varied from 480 to 2,580 mg kg−1 and 450 to 1,080 mg kg−1, respectively (Syakti et al. 2013). On the other hand, Holtermann et al. (2009) reported a study using signature of δ13Corg and demonstrated OM derived from total suspended material in the western SANR consists of riverine and mangrove-derived organic matter with higher concentration while the eastern SANR was marine OM with lower concentration. Unfortunately, because of the large heterogeneity of OM data, our study did not depict a general trend of the relation between OM and heavy metals occurrence (data not shown) due to the different type ecosystem (riverine, lagoon, and marine), land-use and land cover (Ardhli and Wolff 2009), and complex hydrodynamic controlling sedimentation dynamics (water mass exchange and tidal action) (Holtermann et al. 2009) including sampling depth. For this reason, prior to the indices calculations, the trace metal concentrations were normalized (physical and chemical factors) by taking iron as the appropriate element for geochemical normalization of sediments (i.e., grain size proxy) (Daskalakis and O’Connor 1995; Bhuiyan et al. 2010) to determine the degree of anthropogenic influences. We acknowledged that pollutants tend to be associated with the fine particles (clay and fine silt fractions) which have chemical affinity for trace elements, but sieving the <20 μm fraction was technically problematic and may be prone to contamination. Therefore, for physical normalization, the silty-clay (<63 μm) fraction was adopted. Concerning chemical factors such as organic matter or representative elements (e.g., Fe, Al, Li), no single element could cope with broad heavy metals in the different sediment compartments. For instance, anthropogenic Cd and Hg have stronger affinity to organic matter than to clays, whereas natural Ni and Cr may be related to heavy minerals in certain sedimentological provinces (Sandler and Herut 2000). This study used Fe with the reasons as stated in the section “Enrichment factor (EF)”.

We employed the expression given by Hakanson (1980), Tomlinson et al. (1980), Sakan et al. (2009), and the reference baseline of background values used for the abundance of elements in crustal rocks (Greenwood et al. 1997) to obtain the indices for the sediment contamination status (i.e., CFs, PLI, EFs, and Igeo). The results of indices calculations were represented by categorical sampling stations, namely, Av-R, Av-RS, Av-SA, and Av-MR for average values for the riverine, oil refinery site, Segara Anakan Lagoon, and sea stations, respectively (Table 4).
Table 4

Contamination factor (CF), pollution load index (PL), enrichment factor (EF), and geoaccumulation index (Igeo) for Segara Anakan surface sediments of riverine, refinery site, lagoon, and marine stations

Stations

Heavy metals

PLI

Cr

Cu

Fe

Mn

Ni

Ti

V

Zn

CF

 Av-R

0.4

0.9

1.1

1.1

0.3

0.4

1.6

1.5

0.8

 Av-RS

2.8

0.7

7.5

2.7

0.7

1.2

9.5

6.3

2.6

 Av-SA

0.4

0.9

1.1

1.3

0.3

0.4

1.7

1.5

0.8

 Av-MR

0.3

0.4

0.9

1.1

0.2

0.5

1.6

1.1

0.6

 Overall average

0.9

0.8

1.3

1.2

0.3

0.4

1.9

1.5

0.7

 Low contamination (%)

97

64

22

25

100

97

2

7

 Moderate contamination (%)

3

36

75

75

0

3

92

90

 Considerable contamination (%)

0

0

0

0

0

0

3

0

 High contamination (%)

0

0

3

0

0

0

3

3

 

PLI

 No metal pollution (%)

97

 Metal pollution exists (%)

3

EF

 Av-R

0.4

0.9

1.0

1.1

0.3

0.3

1.5

1.4

 Av-RS

0.4

0.1

1.0

0.4

0.1

0.2

1.3

0.8

 Av-SA

0.4

0.8

1.0

1.2

0.2

0.3

1.5

1.3

 Av-MR

0.4

0.9

1.0

1.1

0.3

0.3

1.5

1.4

Igeo

 Av-R

−1.8

−0.8

−0.5

−0.6

−2.4

−2.1

0.1

0.0

 Av-RS

0.9

−1.0

2.3

0.8

−1.1

−0.3

2.7

2.1

 Av-SA

−1.9

−0.9

−0.4

−0.3

−2.5

−2.0

0.2

−0.1

 Av-MR

−2.5

−2.1

−0.9

−0.5

−3.0

−1.7

0.0

−0.6

Av-R, Av-RS, Av-SA, Av-MR: CFs and PLIs average of river, refinery site, Segara Anakan lagoon, and marine stations

CF

The highest CF values for all metals studied, except for Cu and Ni, which are classified as low contamination (CFs < 1), were found at RS, which receives large amounts of oil refinery wastewater discharge. The increasing degree of contamination in the order of Ti, Mn, and Cr is classified as moderate contamination (1 < CFs < 3), and a very high contamination factor was determined for Zn, Fe, and V (CFs > 6). The average for CFs at the riverine stations (R1–R3), lagoon stations (SA1–SA12), and sea stations (MR1–MR9) showed a similar pattern, suggesting that the contamination apportionment in the lagoon has the same input from the rivers. Our finding is supported by the fact that the lagoon can be divided into two major water bodies. First, the lagoon receives the largest freshwater input from the Citanduy River (R1) from approximately 80 % of the overall catchment area, discharging into the western part of Segara Anakan Lagoon. The second input is from the eastern water body with the Donan (R4) catchment, from which each part has a direct connection to the ocean (White et al. 1989; Holtermann et al. 2009). Cr, Cu, Ni, and Ti contamination in the river and Lagoon stations was classified as low contamination, while Fe, Mn, V, and Zn were classified as moderate contamination factors (Table 4). The occurrence of the specific class of CFs in the whole range of sediment samples shows that the percentage of degree of contamination, as highlighted by the metal contaminants, followed the order V > Zn > Mn > Fe > Cu > Cr = Ti > Ni (Table 4).

PLI

The pollution load index (PLI) confirmed the interpretation of the CFs. The average PLI for river stations and lagoon stations was 0.8, a value that denoted no metal pollution in this area (Tomlinson et al. 1980). This value is not significantly different, but the average PLI for sea stations was better at 0.6. Such findings do not suggest that further action is necessary to be taken for environmental improvement, including mitigation action. In general, from most of the stations in SANR, we accounted for 97 % of samples that required no pollution treatment. Our focus is on the RS station that has a PLI of 2.6 (3 % of the overall sampled stations), indicating that a more detailed study is needed to monitor the site concomitantly with an immediate intervention to initiate rehabilitation action as necessary.

EF

Contamination of the sediment by heavy metals in the present study was also assessed using the enrichment factor (EF) and geoaccumulation index values. Table 4 shows the EFs and Igeo of the respective heavy metals. Cr, Cu, Ni, and Ti indicated no enrichment within the SANR sediments because their EF values were below 1. Therefore, for those metals, the anthropogenic inputs in the SANR sediments were not significant. Concerning Mn, V, and Zn, the EF values were between 0.4 and 1.5, indicating minor enrichment or lacking enrichment from anthropogenic contributions of metals. Surprisingly, the EF values for RS stations in general were below the R, SA, and MR stations, indicating that the RS stations were more specifically less polluted in copper and nearly unpolluted. Obviously, such observations may be due to the very high iron concentration at the RS stations (1 and 2).

Igeo

Geoaccumulation index (Igeo) showed that the SANR area was not polluted by Cr, Cu, Fe, Mn, Ni, Ti, and Zn but was moderately polluted by V at some river, lagoon, and sea stations (Table 4). Their Igeo range was between 3 and 0.2, which, according to Bhuiyan et al. (2010), are classified as Class 0 (Igeo ≤0) and Class 1 (1 < Igeo < 0). This situation was supported by the fact that fishery activities including lagoon, silvofisheries, and offshore fisheries are still economically important to the people residing in the SANR area. The most important fishery species are finfish, shrimps, crab, and mollusks (personal observation). At the station RS, the values of Igeo for Cu, Ni, and Ti were below 1, suggesting that all stations were designated unpolluted–moderately polluted while these stations were called moderately polluted by Cr (Igeo = 0.9) and Mn (Igeo = 0.8) (Table 4). The two RS stations have a more severe geoaccumulation index for Fe (2.3), V (2.7), and Zn (2.1), belonging to Class 2 (moderately to heavily polluted): 2 < Igeo < 3 (Bhuiyan et al. 2010). Overall, an order of relative risk significance of selected heavy metals in surface sediments from the SANR was V, Fe, and Zn, suggesting point source pollution for anthropogenic activity (Landre et al. 2011).

Correlation matrix

Spearman’s rank correlation analysis was applied to test the relationships among the heavy metals analyzed. Table 5 shows that the correlation matrix among pairs Cr, Cu, Fe, Mn, Ni, Ti, V, and Zn was good, as shown by a p value below 0.05 (significantly correlated). The correlations were negative between Ti versus Cu and Ni (Table 5). In many cases, positive correlations between two pair variables tended to increase the concentration of both compared metals. For the pair with negative correlation coefficients and p values below 0.05 (in case of Ti versus Cu and Ni), one variable tends to increase while the other decreases. For pairs with p values greater than 0.05, there is no significant relationship between the two variables. The high correlation obtained among each pair of metals in SANR sediments can be interpreted as a common pollution source apportionment for these metals; in this case, the source could be linked to the natural geology of the elements of crustal contribution. V versus Cu, Mn, and Ni reflects an unexpected apportionment concentration pattern due to an anthropogenic contribution. In the case of RS1 and RS2, the sites are greatly influenced by the oil refinery wastewater discharge.
Table 5

Spearman rank order correlation between two variables of heavy metals (Cr, Cu, Fe, Mn, Ni, Ti, V, and Zn) in the surface sediment of Segara Anakan nature reserve

 

Cr

Cu

Fe

Mn

Ni

Ti

V

Zn

Cr

<0.0001 (0.510)

<0.0001 (0.630)

0.0317 (0.269)

<0.0001 (0.681)

0.0306 (0.271)

<0.0001 (0.545)

<0.0001 (0.711)

Cu

 

0.0066 (0.337)

0.0022 (0.337)

<0.0001 (0.699)

0.0257 (−0.337)

0.435 (0.099)

<0.0001 (0.579)

Fe

  

0.0005 (0.425)

0.0002 (0.448)

<0.0001 (0.615)

<0.0001 (0.717)

<0.0001 (0.663)

Mn

   

0.0065 (0.338)

0.0892 (0.214)

0.174 (0.171)

<0.0001 (0.658)

Ni

    

0.489 (−0.087)

0.121 (0.197)

<0.0001 (0.658)

Ti

     

<0.0001 (0.649)

0.0637 (0.233)

V

      

0.0011 (0.404)

Zn

       

Correlation coefficients are shown in brackets

Sediment quality guidelines (SQG)

The guidelines comprise two assessment levels. The lower level is referred to as threshold effect level (TEL) and represents the concentration below which adverse biological effects rarely occur (i.e., biosentinel species). The higher level, known as the probable effect level (PEL), defines a concentration above which adverse effects are expected to occur in a wider range of organisms. The intermediate levels (between TEL and PEL) indicate those concentrations that are rarely, occasionally, and frequently associated with adverse biological effects. The indices of SQG used in this study were limited to TEL and PEL, and the available references were for Cr, Cu, Ni, and Zn. Comparing the heavy metal concentrations to the consensus-based TEL and PEL values developed by MacDonald et al. (2000), up to 70 % of the Cr and 82 % of the samples of Zn were below the TEL. For the metals considered, 24 % (Cr) and 12 % (Zn) of the sample concentrations fell between TEL and PEL, while 6 % (both Cr and Zn) concentrations found in the SANR sediment may cause the adverse effects to occur in a wider range of organisms (Table 6). RS1 and RS2 sediment samples clearly contribute to this more serious possible hazard. Our results suggested that Zn is apparently more chronic than Ni because the observed concentrations have already exceeded the background value (ca. 78 mg kg−1), while the background concentrations of Cr (122 mg kg−1) have largely been far exceeded in the case of the SANR sediments.
Table 6

Heavy metal average concentrations (mg kg−1 dw), TEL, PEL, and SQG values for heavy metals in the SANR sediments

Heavy metals

Average concentrations at the stations (mg kg−1)

SQG

Av-R

Av-RS

Av-SA

Av-MR

<TEL (%)

TEL

TEL-PEL (%)

PEL

>PEL (%)

Cr

52

338

50

34

70

52.3

24

160.4

6

Cu

62

49

58

26

9

18.7

91

108.2

0

Ni

27

71

26

19

15

15.9

79

42.8

6

Zn

114

476

110

234

82

124

12

271

6

Av-R, Av-RS, Av-SA, Av-MR: average of river, refinery site, Segara Anakan lagoon, and marine stations

Concerning Cu and Ni, 9–15 % of the overall sediments were still below TEL quotients, but regarding their respective background values (which are 68 mg kg−1 for Cu and 99 mg kg−1 for Ni) compared to the average value for the metal concerned (52 and 27 mg kg−1 for Cu and Ni, respectively), particularly for Cu, the trace metal concentrations might have adverse biological effects for some sensitive species such as mangrove crab (Neosermatium sp.) (Sastranegara et al. 2003). The broad range of the Cu (91 %) and Ni (79 %) concentrations in the SANR sediments ranged between the TEL and PEL quotients. Based on this classification approach, those concentrations indicated that the adverse biological effects may occur rarely, occasionally, and frequently for the broad range of biota. This situation would rank more than 80 % of the SANR sediments as low to moderate toxicological risks due to the presence of trace metals (Cu and Ni). With regard to the refinery site (RS), the contamination ranking of the Ni load in the sediments against the regional background values seems to be more reliable (≈72 versus 42.8 mg kg−1). These consensus-based sediment quality guidelines evaluate the degree to which the sediment-associated heavy metal contamination status might adversely affect aquatic organisms in the study area. To the best of our knowledge, this study reported for the first time the pseudo-total of heavy metal concentrations in the sediments combined with using the SQGs to correctly classify field-collected sediments as nontoxic or toxic to one or more aquatic organisms.

Conclusions

The concentrations of the heavy metals (Cr, Cu, Fe, Mn, Ni, Ti, V, and Zn) have been determined in surface sediment samples collected in the Segara Anakan Nature Reserve, Indonesia. Analytical results have been elaborated by using Geographical Information System (GIS) software to show accumulation areas for metals (Fig. 2). The heavy metal contamination of sediments was assessed with respect to metal pollution load, enrichment of heavy metal concentrations, and geoaccumulated risk and ecological risk. In general, the heavy metal concentrations of the sediments were found to decrease in the sequence Fe > Ti > Mn > Zn > V > Cu > Cr > Ni. For various metals (Fe, Mn, V, and Zn), the calculated CF and Igeo indices showed the observed concentrations crossing the threshold value of background levels, while Cr, Cu, Ni, and Ti concentrations in sediments could be considered near the background levels found in the SANR. The presence of those metals was due mostly to natural origins, but we noted that some metals (i.e., Fe, Mn, V, and Zn) had been enriched anthropogenically. Significantly higher concentrations of metals (except for Cu) were found at RS1 and RS2 sites that may be due to wastewater discharge from an oil refinery. Sediment pollution assessment was carried out by comparing with two sediment quality guidelines: threshold effect level (TEL) and probable effect level (PEL). The evaluation based on the TEL and PEL quotients showed the concentrations of Cr and Zn were the lowest, while the concentrations of Cu and Ni are likely to result in adverse effects on sediment-dwelling organisms. At a specific site (the RS stations), Cr, Ni, and Zn concentrations found in the SANR sediments may cause the adverse effects to occur in a wider range of organisms and could contribute to more serious harmful effects. In perspective, our results should be a baseline study because it will be important for the future risk assessment of the heavy metal load in sediment samples. More elaborate studies in ecotoxicology (e.g., toxicity of various metals vis-à-vis the indicator species) may reinforce the available data for designing better monitoring programs and ecological risk assessments as well as mitigation action for environmental remediation.

Acknowledgments

The authors specially thank Mr. Gabriel Tirtawidjaya and Cahyadi, who participated in the sampling trips. We also thank the anonymous reviewers for their constructive comments. This research was financially supported by Grants-in-Aid from the National Ministry of Education of Indonesia (DIKTI and BPKLN Kemdiknas Grants) and the Laboratory of Environmental Chemistry, Aix-Marseille University. We acknowledge the support of the Institut Français d’Indonésie in the form of an International Travel Grant.

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • A. D. Syakti
    • 1
  • C. Demelas
    • 2
  • N. V. Hidayati
    • 1
  • G. Rakasiwi
    • 3
  • L. Vassalo
    • 2
  • N. Kumar
    • 4
  • P. Prudent
    • 2
  • P. Doumenq
    • 5
  1. 1.Fisheries and Marine Sciences FacultyJenderal Soedirman UniversityPurwokertoIndonesia
  2. 2.Aix-Marseille Université, CNRS, Laboratoire Chimie EnvironnementMarseille Cedex 3France
  3. 3.Center for Coastal and Marine Resources StudiesBogor Agricultural UniversityBogorIndonesia
  4. 4.Aix Marseille Université, CEREGE/UMR 6635 Europôle de l’ArboisAix-En-Provence Cedex 4France
  5. 5.Aix Marseille Université, CNRS, Laboratoire Chimie Environnement, FRE 3416, équipe MPO. Europôle de l’ArboisAix-en-Provence Cedex 4France

Personalised recommendations