AN OVERALL VIEW OF CADMIUM
AND ZINC CONTAMINATION IN SEPETIBA BAY, RIO DE JANEIRO, BRAZIL
J.E.L. Maddock*, L.D.
Lacerda, Dept. Geoquímica, U.F.F., Niteroi,RJ, Brazil, geojelm, geodrud @
vm.uff.br, J.G. Rees, N. Breward, B.G.S.,Keyworth, Nottingham, U.K., jgre, nbr
@ wpo.nerc.ac.uk
Abstract
A survey of contamination of water and sediments of
Sepetiba Bay, Rio de Janeiro State, Brazil, with zinc and cadmium indicated
that most of the estimated 49,000 tons Zn and 910 tons Cd emitted during the
last 35 years by a hydrometalurgy plant on its north shore has been retain in
the sediments of the bay. However, resuspension and sediment transport have
dispersed the metals over a wide area in the bay.
History of Sepetiba Bay and
its contamination with metals
Sepetiba Bay is a semi enclosed saline coastal lagoon,
about 450km2 in area, separated from the Atlantic Ocean to the south
by a sand spit, and has some characteristics of an estuary. During the last 35
years, it has received metal contamination from an industrial park on its north
shore, especially from a zinc/cadmium hydrometalurgy plant, closed since 1999,
whose effluent and tailings pile runoff drained, and in the case of the
tailings pile, still drains directly into the bay. The bay also receives
sediment load from the Canal de São Francisco which carries about 170m3/s of water diverted, since
1953, from the Paraiba do Sul River, through a hydroelectric scheme, into a
once small (~10 m3/s) water course (Figure 1). Effluent from potable
water treatment for Rio de Janeiro flows into Sepetiba Bay, together with
domestic sewage from several neighbouring towns. Until recently, the natural
catchment area had a relatively low population density, so that only the
sediments near the shore, east of the industrial area, contain sewage material.
The bay has bulk carrier port facilities and is important for shrimp and other
fishing, and bathing beaches.
Metal Contamination Survey
We undertook a comprehensive survey of the contamination
of the sediments and water of Sepetiba Bay in July/August of 1997 (Rees et al.,
1998). Filtered water samples and suspended material (s.s), taken from 43
points spread over the area of the bay (Figure 1), were analysed for contents
of 8 trace metals. Sediment cores, of up to 1.4m. in length, were taken at 29
of these points and grab samples at the remaining points. Cores were sectioned
to study vertical metal distributions. Metal concentrations were determined in
the resulting sub-samples, as well as in the grab samples. Acidified water samples
were analysed by ASV and material in suspension by ICP-MS after total digestion
of the 0.45mm pore membrane
filters. Sediment samples were dried at 60°C and the sieved -150mm fraction (practically the
whole sample in most cases) digested in hot conc. HNO3 for
subsequent analysis by ICP or ICP-MS.
Results
Water and Suspended
Particles
Dissolved cadmium and zinc were both found in greatest
concentrations near the source and to the east of this, along the north part of
the bay, where they were carried by the prevailing clockwise current. However,
the median [Zn] = 4,5mg/l, was not much
higher than “average coastal water” value quoted by Delft Hydraulics Laboratory
(1985) of 2mg/l and median
[Cd] was <0,02mg/l. Despite the
introduction of the metals into the bay in dissolved form, their removal by
absorbtion and deposition must be rapid enough for water from most of the bay
not to contain abnormally high dissolved concentrations. Maximum dissolved
concentrations were 0.341mgCd/l and 64.5mgZn/l, which did not
exceeded national water quality limits. At
the mouth of the bay, dissolved [Cd] was below detection limit (<0.02mg/l) at 3 of 5 points,
[Cd]=0.02mg/l at the other
2, and the mean [Zn] was 3,2mg/l , with little
variation between points.
[Cd] in s.s. was below the effective detection limit of
10ng/l and maximum [Zn] in s.s.
was 31.9mg/l. At the mouth
of the bay, mean [Zn] in s.s. was 1.7mg/l (at 5
points).
Sediments
Provisional reference values for assessing
sediment contamination were promoted via a draft criteria document at an
international sediment quality forum in 1992 (Van Veen and Strotelder 1988).
“Target”, “standard” and “limit” metal concentrations were proposed. Mean [Cd]
in surface sediments, of 3.04mg/kg, was near the “standard” value, 4mg/kg, and
locally “limit” values (30 mg/kg) were approached. [Zn] at most sites, and the
mean concentration, exceeded the standard value, 750mg/kg, and at site 1,
[Zn], = 5100mg/kg, greatly exceeded the “limit” value of 2500mg/kg.
Many of the cores showed distinct step
decreases in Cd and Zn concentrations, at depths where sediment ages
corresponded to the startup of the hydrometalurgy plant, in 1965, p.ex.: Figure
2. Thus background concentrations in the sediments could be estimated from the
concentrations in the lower parts of cores. This “step” may have corresponded to the lower limit of
bioturbation but background concentrations would still be indicated in the same
way.
Metal concentrations in suspended material sampled at the
mouth of the bay, [Cd] < 0.1mg/kg (not detected) and range [Zn] = 5.0 –
15.5mg/kg, were lower than the background concentrations found in cores [Cd] ~
0 (n.d.) and [Zn] = 156mg/kg. Concentrations in bottom sediments from this area
were also extemely low [Cd] < 1mg/kg (n.d.) and range [Zn] = 26 –30mg/kg,
except near the north shore, where higher values were found: [Cd] = 2mg/kg and
[Zn] =529mg/kg. These figures, together with the values for dissolved metals,
indicate that little export occurs of these metals from the bay.
Quantities of
metals in the upper parts of the cores, in excess of those calculated from
background concentrations, i.e. due to contamination from the hydrometalurgy
plant, were estimated. The area mass density distribution of Zn is shown in
figure 3 and that for Cd was similar but slightly less dispersed. The total tonnages of excess metals deposited in the
bay’s sediments, 110,000 tons Zn and 670 tons Cd, were calculated by krigging
these data.
Discussion and Conclusions
Maximum total
emissions of Zn and Cd were calculated by Barcellos et al (1991), from the
differences between total contents of the ore processed and tonnages of metals
produced, as 49,000 tons Zn and 910 tons Cd. Extrapolation of these figures,
assuming 1990 loss rates up to 1994, when pollution control measures were
implanted, resulted in estimated total losses of 64,000 tons Zn and 1330 tons
Cd.
The calculated
quantities of Zn and Cd found in the sediments in this survey correspond to
170% and 50%, respectively, of these estimates of total losses from the
refinery. It is probable that part of the cadmium losses is still be stockpiled
in the tailings. If this is the case, part of the zinc must also be, and the
excess zinc in the sediment must be attributed to other recent sources.
Concentrations of Zn in the water and suspended sediments of the Canal de São
Francisco indicate this as a source. On the basis of data published in other
countries for typical source strengths and an audit of human activities present
in the area, Barcellos & Lacerda
(1994) attributed 30% of the Zn
entering the Bay to sewage and urban drainage, solid waste (rubbish tip)
leachate, port activities and paper manufacture effluent.
We conclude that
not all the zinc and cadmium lost from the ore processing has reached the bay
but virtually all that has was rapidly taken up by suspended matter and has
been spread over a large part of the bay where it is retained in the sediments.
Barcellos C., Rezende C.E.
& Pfeiffer W.C. (1991) “Zn and Cd production and pollution in a Brazilian
Coastal Region.” Marine Poll. Bull. 22:
558-561.
Barcellos C. & Lacerda
L.D. (1994) “Cadmium and zinc source assessment in the sepetiba Bay and region
basin.” Env. Monit. & Assess. 29:
183-199.
Delft Hydraulics Laboratory
(1985) “Coastal zone information system and water quality modelling for
management of coastal waters”. BTA-60, Rep. R1182, Delft, Holand
Rees J.G., Breward N.,
Maddock J.E.L., Ferguson A.J. & Williams T.M. (1998) “Assessment of
contamination by metals and selected organic compounds in coastal sediments and
waters of Sepetiba Bay, Rio de Janeiro State, Brazil.” Tech. Rep. WC/98/30 BGS,
Keyworth, Notts., UK. 93p.
Van Veen R.J. &
Strotelder P.B.M. (1988) “Research on contaminated sediments in the
Netherlands.” In: Wolf K., Van de Brink W.J. & Colon F.J. (Eds.).
Contaminated Soil. p 1263-1275. Acad. Publ.



Contours from 0
to 2kg Zn.m-2 at intervals of
400g Zn.m-2