From the Saudi Coast of the Arabian Gulf

Clams as biomonitors of oil-related metal pollution in sea waterChapter Two

CHAPTER TWO

From the Saudi coast of the Arabian Gulf

A field study of Meretrix meretrix as a biomonitor of nickel and vanadium

2.1 Introduction

Comparisons of laboratory studies with established monitors in field trials provide useful information in the assessment of the monitoring capabilities of an organism. Therefore, this chapter assesses the monitoring capabilities of the venus clam Meretrix meretrix through a field monitoring study. This species possesses most of Butleret als (1971) and Haug et als(1974) requirements or an organism to act as a monitor, (see Section 1.5.3). In particular, it is tolerant of considerably elevated metal levels, it is sessile, and is abundant in the study area. Meretrix meretrix can also tolerate salinities up to 60 0/00 (Sadiq et al., 1992) and has also been found in estuaries (Jayabal and Kalyani, 1986). In addition, this species has a similar niche to R. philippinarum, which is used in my laboratory and field studies conducted in the UK (Chapters 2 and 3)

Meretrix meretrix has a triangular shell that rarely exceeds 60 mm in length, with strong hinge teeth and a broad pallial sinus. The lunule and escutcheon are weakly defined. This clam varies in colour from white to deep purple-brown and is often patterned with darker colours, especially near the umbos (Smythe, 1982).

The pilot monitoring survey was performed at selected locations on the Saudi coast of the Arabian Gulf. The main objectives were:

1. to assess the ability of M. meretrix to identify spatial and temporal changes in the bioavailability of the oil–related metals Ni and V at these sites, in relation to known sites of contamination; and
2. to examine the influence of size class on the accumulation of Ni and V.

2.2Study area

The Arabian Gulf forms a shallow extension of the Arabian Sea, which separates Iran from the Arabian Peninsula, and is linked to the Arabian Sea by the Strait of Hormuz and the Gulf of Oman. It is a semi-enclosed sea with an axis length of some 1,000 km, a width varying from 200 to 300 km, and an area comprising about 226,000 km2.

The Gulf coastline of Saudi Arabia extends from Ras Al Khafji in the north to Salwah in the south, and follows a fairly steady south easterly direction. As the coastline is a continuum, all types of shores grade into one another and a variety of mud, muddy sand, sandy mud, sand, sand-rock flats, and rocky shore habitats exist, often in close proximity. The subtidal areas consist of sand and rock, sea grass, and some areas of coral reef.

The near-shore waters are shallow, with depths rarely exceeding 25 m. Salinities are high owing to high evaporation and little fresh-water influx, with values ranging from 38 to 60 0/00. In some restricted areas such as the Gulf of Salwah, the salinity can reach 70 0/00 (Basson et al., 1977), and in enclosed bays and shore pools even higher values can be found. Offshore surface water temperatures range from 17 to 33C with small vertical temperature gradients (Emery, 1956). In some near shore areas, temperatures can rise above 40C (Basson et al., 1977).

Offshore winds predominantly blow from directions between west and north, and most strongly from the north-west.

Currents are mainly tidally generated, with an average maximum tidal range of 2.5 m and an average velocity of less than 0.5 m/s, but the offshore topography and many coastal restrictions cause swift tidal currents locally (Williams, 1979). The tidal currents generally flow westward and north-westward during flood tides and in the opposite directions during ebb tides. Residual currents usually flow from north-west to south-east, approximately parallel to the coast, and are driven by wind and density imbalances (Hunter, 1982).

The annual average of residual wind-driven velocities ranges from 0.1 to 0.3 m s-1 (Henaidi, 1984). Wind-generated waves are usually small (approximately 1 m), but have reached heights of 3.5 m during storms. Strong winds are more frequent in winter, early spring, and late autumn (Williams, 1979).

2.3 Materials and methods

2.3.1 Survey sites

Five sampling sites were selected for this study and are shown in Figure 2.1. Out of these, four sites (sites 2-5) were selected in Tarut Bay which where exposed to several outfalls. In addition, site 1 was chosen as the control site, it is outside the Bay and about 50 km away from the other sites, and in proximity to the public beach facilities of Al Khobar City. Table 2.1 gives further details of each site.

Table 2.1: Description of sampling sites

Location / Salinity (0/00) / Sediments / Description
1 / Alkhobar city / 47 / Muddy sand / Community beach.
2 / Tarut island / 47 / Thin layer of sand on rocky bed / Adjacent to a land filled area.
3 / Safwa city / 51 / Sandy mud / Near to agriculture and sewage outfalls. Close to rusty metallic stands of old pipelines and a land filled area.
4 / Rahimah city / 51 / Thin layer of sand on rocky bed / Adjacent to sewage outfall of Rahimah city. and a land filled area.
5 / Ras Tanura / 47 / Muddy sand / Near to the backside of Ras Tanura refinery.

Figure 2.1: Map of Tarut Bay on the Arabian Gulf, showing the sampling sites.

The coastline surrounding Tarut Bay has become one of the most heavily developed along the entire Saudi Coast of the Arabian Gulf. Major industrial facilities that border the Bay include King Abdul Aziz Port to the south and Rahima/Ras Tanura oil refinery to the north, which is one of the largest refineries in the world. Land-filling and dredging have affected vast areas of sea-grass and mangrove in the Bay (Sadiq et al., 1993). Unchecked disposals of solid and liquid wastes have increased in recent years, particularly with rapid urbanization and industrialization of the towns and cities around the Tarut Bay.

2.3.2 Field sampling

Sampling involved the collection at each site of five individual Meretrix meretrix of two size classes. Clams ranging up to 10 mm in each class were collected, the sizes of which were as follows: small size (30-39 mm shell length); large size (45-54 mm shell length). Collection of animals was carried out during the winter. Immediately after collection the animals were transferred to the laboratory where they were held in sea-water (47 0/00 salinity) for 48 hours to void the gut contents.

2.3.3 Sample preparation and analysis

The length and soft-tissue dry weight of the collected clams were measured. The shell was removed and the soft parts transferred to a pre-weighed 50 ml graduated digestion tube, dried at 85-95°C for 48 hours, and then weighed. 10 ml of 'ARISTAR' grade conc. HNO3 were added to the tubes and the tissues were digested at room temperature for two days. The material was then further digested at 90-100°C in a digestion block for 3 hours or until the solution was clear and then evaporated to dryness. The residue was dissolved in 0.1 M HNO3, made up to 10 ml using 0.1 M HNO3, and stored in a polypropylene bottle. All glassware and sample bottles were acid-washed in 10 % HNO3.

Nickel and V were analysed in individual clams. The content of Ni and V was measured using inductively coupled plasma emission spectroscopy (ICP). The instrument was calibrated using a blank and two standards (1 mg l-1 and 10 mg l-1) for each element. The calibration routine was performed before each set of results was obtained, and the performance of the instrument was checked after every 15 samples using independent quality control samples. Metal concentrations in the clams are expressed as dry tissue weight expressed as mg kg-1 in the results.

2.3.4 Quality control

Oyster tissue was obtained from the US National Bureau of Standards (NBS No. 1566). The standard reference material was digested using the same procedure as described above. The percentage recovery observed is a mean of 10 replicates and is shown in Table 2.2.

Contamination of the sample by the reagent was tested by preparing reagent blanks at frequent intervals during the study. Ni and V concentrations in the reagent blank were < 3 µgl-1 and < 2 µgl-1 respectively.

Table 2:2. Quality control results (analysis of oyster tissue provided by the National Bureau of Standards (NBS), n = 10).

Metal / Certified NBS concentration (mg kg-1) / Measured concentration (mg kg-1) / Recovery %
Ni / 2.25 / 2.50 / 114
V / 4.68 / 5.10 / 109

2.3.5 Statistical analysis

In view of the data variance heterogeneity and the occasional accidental reduction in the data sets to less than five replicates, the use of a frequency distribution test is inappropriate. Thus, the use of parametric statistics was precluded and the following non-parametric statistics were used instead:

1. The Kruscal-Wallis one-way test for several independent samples. This was used to compare more than two groups of cases on one variable.
2. The Mann-Whitney U test for two independent samples. This was used to compare two groups of cases on one variable.
3. The bivariate rank correlation procedure, which computes Kendal’s tau-b () or Spearman’s correlation coefficient (r).

In all cases the significance level was taken as P<0.05.

2.4Results

Intrasample and intersample variability in tissue dry weight in each size class was minor, except for the relatively high intrasample variability at site 4 and the relatively higher weight in the large size class at the control site (site 1, see Figure 2.2). Although the larger size of the clams from the unpolluted site could affect the metal level, it is unlikely that the weight difference affects the trend, especially when there is no significant correlation between the dry weight and Ni and V levels in each size class at all of the sites, except between the dry weight of large size class clams and Ni at site 4 ( = -1; P< 0.05). Furthermore, no significant correlation between the combined dry weight from both size classes and the tested metals was found at any site, except between V and dry weight at site 3 (= -0.629; P<0.05). Interestingly, despite the high weight difference between small and large size classes at site 1, there was no significant correlation between Ni or V and soft tissue dry weight (Figure 2.2). However, increasing the sample size may verify the effect of size on the tested metals.

Samples of small animals collected from site 4 were accidentally lost. This was unfortunate as the availability of this data would have contributed to clarifying the trend of the contamination in Tarut Bay.

No significant correlation between clam dry weight and shell length in each size class at each site was observed, whilst a significant correlation was found when both size classes were combined at all sites (= 0.628-0.767; P < 0.01-0.05), except at site 4. This may be because the data for small clams at this site is not available.

Figure 2.2: Soft tissue dry weight of small and large clams (30-35 mm and 46-54 mm respecticely) collected from polluted sites (2-5) and an unpolluted site (1) along the Saudi coast of the Arabian Gulf. The mean values  standard error are shown.

2.4.1 Nickel

The mean concentrations of Ni in clams from the polluted and non-polluted sites are shown in Table 2.3. Data for shell lengths and mean dry weight are also included. The concentrations of Ni in small-size clams of M. meretrix collected from the sites in the polluted Tarut Bay were not significantly different from each other and from the control site (site 1), except at site 2 where a significant P < 0.01 increase of 44 % was observed compared to the Ni levels in clams at the control site (Figure 2.3).

In contrast, large clams collected from all sites (sites 2-5) in Tarut Bay contained significantly higher (P < 0.05) Ni concentrations than clams from the control site (Figure 2.3). A maximum increase of 234 % at site 3 was observed, compared to the control, whilst at sites 2, 4, and 5 the increase was 147 %, 131 % and 114 % respectively. The metal concentrations in clams at different sites in the Bay were not significantly different from each other except at site 3, where the mean concentration of Ni was higher by 45% than the level at site 5. However, the mean Ni level in animals at site 3 was higher than by 35 % and 57 % at sites 2 and 4 respectively. Even so, these differences were not significant, perhaps because of the high metal variance, which could mask the trend. Increasing the sample size may reduce the high variance and increase the probability of a significant difference manifesting itself.

Table 2.3: Biometric details and Ni concentrations in Meretrix meretrix

Site No. / Size
Class / Shell length range (mm) / Dry weight range (g) / Mean dry weight (g) / Ni range
(mg kg-1) / Ni mean
(mg kg-1)
1 / Small / 31.4-35.5 / 0.23-0.29 / 0.26 / 6.03-9.85 / 8.01
Large / 47.6-49.4 / 1.05-1.50 / 1.24 / 4.35-7.38 / 5.79
2 / Small / 34.3-35.2 / 0.15-0.36 / 0.27 / 10.04-13.39 / 11.52
Large / 48.7-53.8 / 0.45-0.78 / 0.61 / 11.59-19.40 / 14.33
3 / Small / 29.7-33.2 / 0.14-0.33 / 0.19 / 5.60-13.53 / 8.50
Large / 46.5-52.3 / 0.53-0.73 / 0.60 / 11.84-27.07 / 19.35
4 / Small / NA / NA / NA / NA / NA
Large / 46.3-49.0 / 0.31-0.89 / 0.72 / 2.80-24.45 / 13.36
5 / Small / 32.4-34.7 / 0.17-0.34 / 0.26 / 6.04-13.79 / 9.06
Large / 48.3-49.7 / 0.65-0.77 / 0.72 / 7.98-15.91 / 12.39

NA = not available.

Sadiq et al. (1993) have measured Ni concentrations in water and sediment (1.11 µgl-1 and 13.2 mg kg-1 respectively) at a site 2 km from site 4. The authors’ data are about 3 years old compared to this study and the samples were collected only once so that the examined concentration factors (CF) at site 4 will be approximate. The CF observed in large clams, based on these water and sediment levels, are 12 x 103 and 1 x 103 respectively.

Figure 2.3: Nickel body burden in M. meretrix (with shell length 30-35 and 46-54 mm) collected from polluted sites (2-5) and unpolluted site (1) along the Saudi coast of the Arabian Gulf. The mean values  standard error are shown.

2.4.2 Vanadium

The mean concentrations of V, shell length, and mean dry weight are shown in Table 2.4. The concentrations of V in small-sized clams of M. meritrix collected from Tarut Bay are not significantly different from those recorded at the control site (site 1), except at site 3, where a significant P < 0.01 increase of 210 % was observed (Figure 2.4). However, the mean concentrations of V in clams at other sites in Tarut Bay were higher than at the control site by between 110 % and 57 %, and these insignificant differences may have been due to the high metal variance masking the trend. For example, this high variance may be the reason for the insignificant trend in Tarut Bay (sites 2-5), despite the fact that the V concentration in animals at site 3 was significantly (P<0.05) higher than at site 5 by 94 %.

In contrast to Ni, the patterns of V body burden by small and large clams are similar. The mean concentration of V (Table 2.4) observed in the large-size class of M. meretrix collected from all sites of Tarut Bay was significantly elevated (P< 0.05) compared to the control. However, the significant increase at site 4 compared to the control was close to the level of significance (P < 0.06). A maximum increase of 556 % in clams at site 3 was observed, whilst at sites 2, 5, and 4 the increment was 459 %, 238 %, and 147 % of the control respectively. The levels of the metal in M. meretrix at sites 2 and 3 were significantly (P < 0.05) higher (by approximately 100 %) than at sites 4 and 5, but each of the sub-groups (2 and 3, 4 and 5) were not significantly (P > 0.05) different from each other.

Table 2.4. Biometric details and V concentrations in Meretrix meretrix

Site no. / Size
Class / Shell length range (mm) / weight range (g) / Mean weight (g) / V range
(mg kg-1) / V mean
(mg kg-1)
1 / Small / 31.4-35.5 / 0.23-0.29 / 0.26 / 0.00-2.05 / 1.26
Large / 47.6-49.4 / 1.05-1.50 / 1.24 / 0.00-0.77 / 0.34
2 / Small / 34.3-35.2 / 0.15-0.36 / 0.27 / 0.84-7.01 / 2.64
Large / 48.7-53.8 / 0.45- 0.78 / 0.61 / 1.16-2.86 / 1.90
3 / Small / 29.7-33.2 / 0.14-0.33 / 0.19 / 2.72-5.32 / 3.90
Large / 46.5-52.3 / 0.53-0.73 / 0.60 / 1.70-3.52 / 2.23
4 / Small / NA / NA / NA / NA / NA
Large / 46.3-49.0 / 0.31-0.89 / 0.72 / 0.29-1.33 / 0.84
5 / Small / 32.4-34.7 / 0.17-0.34 / 0.26 / 0.69-3.50 / 1.98
Large / 48.3-49.7 / 0.65-0.77 / 0.72 / 0.78-1.75 / 1.15

NA = not available.

Figure 2.4: Vanadium body burden in M. meretrix (with shell lengths of 30-35 and 46-54 mm) collected from polluted sites (2-5) and an unpolluted site (1) along the Saudi coast of the Arabian Gulf. The mean values  standard error are shown.

Vanadium concentration was measured by Sadiq et al. (1993) in water and sediment (2.99 µgl-1 and 9.45 mg kg-1 respectively) at a site 2 km away from site 4. On account of the limitations of the above workers’ data (see Section 2.4.1), the examined concentration factors (CF) at site 4 will be approximate. The CF observed in large clams based on the above water level is 283 and much smaller (84) when related to sediment concentration.

2.5Discussion

The field survey showed that collecting the clam M. meretrix from the intertidal zone was a relatively simple task. The clams were abundant at most of the sites.

Analysis of the clams revealed spatial changes in the bioavailability of Ni and V among the survey sites, especially between the polluted sites in Tarut Bay and the control site. However, the high variance of Ni and V in the clams (see Figures 2.3 and 2.4) may have masked any trends in concentrations and there were few significant P0.05 differences between sites. Other workers have found a large variation in metal concentrations in bivalves. Boyden (1977) found that oysters from contaminated locations exhibit equivalent variability for Zn (1816 to 11185 mg kg-1), while apparently homogeneous samples of M. edulis display a fourteen-fold variation in Zn content, with some individuals seemingly demonstrating regulation of the metal and others apparently capable of assimilating unusually high levels (Lobel et al., 1982), (see section 1.4.2.1).

2.5.1Nickel

Nickel body burden in M. meretrix is shown in Figure 2.3. As mentioned in the results section, the animals collected from the Bay showed higher levels of Ni compared to animals from the control site. The mean Ni concentrations are 8.01 and 11.52 mg kg-1 in small clams and 5.79 and 19.35 mg kg-1 in large clams at the control and contaminated sites respectively. However, small clams for some reason showed no significant differences between the various sites, except at site 2 where tissue concentrations were slightly higher than at the rest of the sites. No satisfactory explanation has been found for this apparent discrepancy.

The animals at the west side of the Bay (sites 2 and 3) seem to contain relatively higher Ni concentration than at other sites (4 and 5). The trend is more apparent at site 3. The hydrography in particular, current velocity, direction and level of particulates within the Bay, and the increasing number of outfalls near to these locations may contribute to increasing the concentration of contaminants around these sites (Sadiq et al., 1993).

As mentioned in the results section, the sea-water levels of Ni and V recorded by Sadiq et al. (1992) and Sadiq and McCain (1993) may not be representative of the average metal contamination in Tarut Bay. This is mainly owing to the short period of sampling and the limited number of sites (1 and 2) monitored by these workers. Therefore, the reported sea-water metal levels in the Bay are equivocal. The sediment metal levels reported by Sadiq et al. (1993), which were collected from 31 sites are more representative of the contamination than those from the sea-water levels and they may show anthropogenic impact on the Bay. However, Ni and V body burden in clams sampled during this study did not correlate with the metal levels in the sediment.