See

discussions, stats, and author profiles for this publication at:

http://www.researchgate.net/publication/15589528

Mercury in environmental and biological

samples from a gold mining area in the
Amazon region of Brazil
ARTICLE in SCIENCE OF THE TOTAL ENVIRONMENT JUNE 1995
Impact Factor: 4.1 DOI: 10.1016/0048-9697(95)04533-7 Source: PubMed

CITATIONS

READS

66

63

2 AUTHORS, INCLUDING:
Dulcideia Palheta
Universidade Federal Rural da Amazni
8 PUBLICATIONS 80 CITATIONS
SEE PROFILE

Available from: Dulcideia Palheta

Retrieved on: 09 December 2015

The Science of the Total Environment 168 (1995) 63-69

Mercury in environmental and biological samples from a

gold mining area in the Amazon region of Brazil
Dulcideia Palheta, Andrew Taylor* b
a Trace Elements

Laboratory,
bDepartmertt

Robens Institute of Health and Safeiy, Universiry of Surrey, Guildford,

Surrey GU.2 5XH,
of Clinical Biochemtim,
St. Lukes Hospital, Guildford,
Surrey GUI 3NT UK

UK

Received 2 September 1994; accepted 1 November 1994

Abstract

A massiveincreasein gold mining in the Amazon region of Brazil has led to an enormousdischargeof metallic
mercury into the aquatic ecosystem.To investigatethe dispersion,total and inorganic mercury concentrationswere
measuredin water, fish and animal tissues,and in blood, urine and hair from membersof the local populations.
Mercury concentrationsin river water, sedimentsand fishwere high comparedwith thoseof non-contaminatedareas.
Cattle and pigskept in the area and with accessto the contaminatedrivers had concentrationsof mercury of 0.1-1.28
pg/g and 11.7-15.7 pg/l in hair and blood, respectively.These resultsare approximately twice those measuredin
specimensfrom control animals.Mean mercury concentrationsin blood, urine and hair of residentswere increased
at 11.4pg/l, 22.8 pg/l and 4.3 pg/g, respectively,and the urine mercury of workersfrom the gold mining siteswere
up to 155 pg/l. The resultsdemonstratewidespreadcontamination of the environment by mercury with transfer of
the metal to fish and animalsusedfor food, and into the inhabitantsof the region. Further investigationsfor possible
adversehealth effects need to be undertaken.
Keywords:

Mercury; Gold mining; Amazon region; Blood; Hair; Urine

1. Introduction
Methylmercury
is highly toxic to the central
nervous system and to the developing fetus [l]. It

*Corresponding
author, Trace Elements Laboratory,
Robens Institute of Health and Safety, University of Surrey,
Guildford, Surrey GU2 5XH, UK.
0048-9697/95/$09.50
SSDI

has a long biological half-life and accumulates

through the food chain as was demonstrated in
the Minamata Bay incident where waste mercury
discharged into sea water was methylated and
concentrated into fish 5000-50 OOO-fold [2,3]. Individuals who ate the fish had symptoms of
dysarthria, ataxia and tunnel vision and there
were increased rates of miscarriage and fetal malformations
among pregnant women. Toxicity

0 1995 Elsevier Science BV. All rights reserved.

0048-9697(95)04533-S

64

D. Palheta, A. Taylor

/ The Science of the Total Environment

caused by exposure to elemental mercury with

damage to the peripheral and central nervous
systems, the kidney and the fetus, has long been
recognised [4,5].
In the last 10 years there has been a large
increase in gold mining activity in the Amazon
region of Brazil [6,7]. Extraction of alluvial gold
requires the use of metallic mercury to form an
amalgam of gold and mercury, and amounts in

I68 (1995) 63-69

excess of 100 tonnes/year are employed in Brazil.

The amalgam is subsequently heated with release
of mercury vapour into the air. There is aIso
discharge of enormous amounts of mercury into
the Amazonian aquatic ecosystem with an immediate risk to the environment and to fish. Domestic and farm animals who drink river water are
also exposed to increased amounts of mercury.
Humans in these areas will be exposed to the

ATLANTIC

OCEAN

Fig. 1. Diagramatic plan of the Gurupi gold field, Para, Brazil with water mercury concentrations represented by the size of the
oatterned blocks

D. Palheta, A. Taylor

/ The Science of the Total Enuironment

Table 1
Samples collected for measurement of mercury
Source

Specimen

Specimen

Water
Fish

River
7 diferent
species
Blood
Blood
Blood
Blood

8
38

Tap

13
6
10
23

Hair
Hair
Hair
Hair

15
6
10
35

Pig
Sheep
Cattle
Human

Specimen

Urine

25

metal in an occupational setting and/or via consumption of locally produced fish and meat. To
determine the actual dispersion of mercury and
possible health risks associated with gold extraction, one area has been investigated with the
measurement of total and inorganic mercury in
water, fish and animal tissues, and in blood, urine
and hair from members of the local populations.
2. Materials and methods
2.1. Collection of specimens

Specimens were collected at sites within the

Gurupi gold field in the northeast of the Brazilian
state of Para (Fig. 1). The village of Cachoeira
was selected as a base since it is an important
market and gold trading centre within the area.
Water. Samples of river water from sites along

168 (1995)

63-69

65

the Gurupi, Piria and Macaco Rivers, and drinking

water from the village of Cachoeira were collected into containers with 0.2 ml KBrO, (0.1
M)-KBr (1% m/v) and 0.2 ml HCl per 20 ml
water. Specimens were kept at 4C until analysis.
Fish. Species regularly used as food were caught
in the rivers or bought from markets, their local
names were Piaba, Tucunare, Cachorrinho de
Padre, Acara, Traira, Pacu and Mandii. The
specimens were later identified by the Natural
History Museum, London as Characidae, Cichlu
sp., Cemicichla sp., Geophagus sp., Hophias malabtzricus, c.f. Myleus sp. and Pimelodus sp., respec-

tively. Portions of edible tissues were removed for

analysis and stored at -20C.
Domestic animals. Specimens of hair and blood
were collected from pigs, sheep and cattle which
grazed alongside and drank water from the Gurupi and Piria Rivers. A further series of specimens were obtained from animals at the Para
Agricultural College, not within the Gurupi area.
Human specimens.Blood, urine and hair were
obtained from villagers who live alongside the
Gurupi River, from residents of Cachoeira and
from garimpeiros - the men who work at the
gold fields. All blood, urine and hair specimens
were kept at 4C until analysis.
The numbers of specimens collected for analysis are given in Table 1.

Table 2
Measurement of total mercury in blood, tissues and urine
Hg added
( /-G/l)

Hg found
( /G/l)

Blood
Endogenous
Inorganic mercury
Methylmercury

(+ 100)
(+100)

5.2
105.8
116.8

102
110

5.4
3.0
5.6

Urine
Endogenous
Inorganic mercury
Methylmercury

(+50)
(+50)

3.7
51.6
45.5

96
91

13.0
3.1
7.1

Tissue homogenate
Endogenous
Inorganic mercury
Methylmercury

(+ 100)
(+ 100)

94.9
94.6

7.2
11.4
11.0

Accuracy and precision (number of replicates = 10).

11.2
104.9
103.2

Reproducibility
(CV%o)

66

D. Palheta, A. Taylor

/ The Science of the Total Environment

168 (1995)

63-69

Table 3
Speciation of inorganic and organic mercury in urine

Endogenous
Inorganic mercury
Methylmercury

Hg added
( /-G/l)

Hg found
( cLg/l)

Recovery
(%o)

Reproducibility
(CV%o)

(+50)
(+50)

2.8
54.0
4.2

102.5
2.5

9.5
3.3
4.7

Measurement of inorganic mercury. Accuracy and precision (number of replicates = 10)

experiments are shown in Table 2 and demonstrate that the procedure measured mercury accurately and with good precision. The protocol to
differentiate between inorganic and organic mercury species was also demonstrated to be effective
(Table 3).

2.2. Determination of mercury

Tissue samples were homogenised for analysis

[8], hair was dissolved in 40% m/v NaOH [9] and
blood was examined without pre-treatment
[lo].
Concentrations
of mercury were measured by
cold-vapour atomic absorption spectrometry. Total mercury was measured using a stannous chloride-cadmium chloride solution as the reductant.
When a stannous chloride solution was used, only
the inorganic mercury was determined [9].
Urine and water samples were digested at room
temperature with KIMnO,-H,SO,,
for the measurement of total mercury using stannous chloride as the reductant. A second analysis, without
digestion, yielded the inorganic mercury concentrations [ 111.
These procedures were evaluated using specimens spiked with mercuric chloride or methylmercury. Recoveries of added mercury, and precision were determined for all sample types.

3.2. Mercury concentrations in water and biological

samples

Water mercury concentrations are shown in

Fig. 1. River water collected around the gold
mining areas contained up to 800 rig/l mercury
compared with < 200 rig/l in the drinking water
and in river water upstream from the mining
operations. The highest concentrations were in a
lake which drained the mining area and in the
River Gurupi close to the areas where the mining
activity was at its most intense. Concentrations
fell as tributaries and other sources caused the
river to increase greatly in size, with an effective
dilution of the mercury.
Total mercury concentrations in the fish specimens are given in Table 4. Higher levels occured
in carnivorous compared with plant eating species.
The proportion of organic mercury in the samples
varied from 87 to 100%.

3. Results
3.1. Accuracy and precision of the analytical
procedures

The results for the recovery and reproducibility

Table 4
Total and organic mercury concentrations in fish
--No. of
Species
samples
Characidae
Cichla sp.
Crenicichla sp.
Geophagus sp.
Hoplias malabaricus
cf Myleus sp
Pimelodus biochii

4
3
8
2

13
2
6

Total mercury
( /G/g)

Organic
mercury (%)

0.19
0.11
0.21
0.03
0.61
0.04
0.18

87
91
94
90
95
100
98

D. Palheta, A. Taylor

Exposed animals

/ The Science

of the

Control animals

Total Environment

168 (1995)

67

63-69

Control animals

2.0

1.o
NS

Fig. 3. Concentrations of mercury (pg/g) in hair samples

from pigs, cattle and sheep raised in the Gurupi gold field
(exposed) and the Para Agricultural College (control). NS, no
significant difference between results from the two groups.

Pigs

Cattle

Fig. 2. Concentrations of mercury

from pigs, cattle and sheep raised
(exposed) and the Para Agricultural
cance between results from the two
not significant.

Sheep

(@g/l) in blood samples

in the Gurupi gold field
College (control). Signifigroups: *, P < 0.001; NS,

there was no exposure to mercury. These differences were statistically significant for the cattle.
Blood, urine and hair mercury concentrations
in garimpeiros, residents of Cachoeiro and those
living along the Gurupi River are given in Table
5.
4. Discussion

Total mercury levels in blood and hair from

domestic farm animals are shown in Figs. 2 and
3. Higher concentrations are evident in specimens
from pigs compared with cattle and sheep, but for
all animals the levels were higher than in similar
species living in the agricultural college where

Uncontaminated
freshwater mercury concentrations are < 10 rig/l. Thus, the concentrations
determined in the Gurupi and Piria Rivers confirm significant contamination
from the gold recovery procedures. All the mercury in the water

Table 5
Mercury concentrations in human subjects
Sample

Group

Range
of results

Number
of results

Blood
(/G/O

Garimpeiros
Villagers
River dwellers
(Reference range)

2.0-29.3
2.7-9.6
1.0-64.7
1.7-9.9

10
3
10

Urine
(M/l)

Garimpeiros
Villagers
(Reference range)

1.0-155
1.0-2.5
0.1-6.9

20
5

Hair
@pm)

Garimpeiros
Villagers
River dwellers
(Reference range)

0.4-32.0
0.8-4.6
0.2-15.0
<2

20
5
10

Reference range from Minoia et al. [14] and Taylor (151.

68

D. Palheta, A. Taylor

/The

Science

of the Total Environment

specimens was in the organic form. These increased levels were obtained during the rainy
season when the rivers were exceptionally full and
expansive (several hundred metres wide). Even
higher concentrations might be anticipated at
other times of the year. Levels of up to 600 rig/l
were determined in Minamata Bay while elsewhere in the Amazonian region, river waters have
been found with mercury concentrations as high
as 8.6 pg/l [6]. The Brazilian National Council
for the Environment has set a safety limit of 200
g/l, a level which was clearly breached during
the period of this study. Thus, water used for
drinking, agricultural and recreational purposes
represents a hazard to the inhabitants of this
area.
Previous measurements of mercury in fish muscle of specimens from uncontaminated rivers were
N 0.06 pg/g. The higher concentrations in some
of the fish reported here demonstrate the accumulation of mercury from the water, possibly via
smaller aquatic organisms. The major proportion
of the mercury in fish was in the organic form
indicating probable methylation of the discharged
mercury.
Domestic
animals
foraging,
grazing and
drinking around the contaminated rivers also accumulated mercury, some to much higher concentrations than in unexposed controls. High levels
in hair are more usually derived from external
contamination
where airborne mercury is increased. Mercury in blood reflects recent and
current exposure to mercury and as a consequence of the foraging and scavenging lifestyle of
the pigs these animals accumulated more mercury, both the test and control groups, than did
the sheep and cattle. The cattle grazing around
the river had significantly increased blood mercury concentrations compared with the controls.
Some of the pigs also had very high (> 50 pg/l)
blood mercury levels, but the group data were not
significantly different from the controls. However,
the exposed animals were much younger (< 1
year of age) than the controls and would have
accumulated
less mercury from usual dietary
sources. The results indicate that mercury accumulated in at least some of the domestic animals
and would provide a further source to humans if

168 (1995)

63-69

used as foodstuffs. Additional measurements in

meat and milk should be undertaken to determine the realistic hazard from these products.
The human subjects included in this study were
exposed to several different sources of mercury;
amalgam
ation
and
heating
processes
(garimpeiros),
further gold refining activities
within the villages, eating of contaminated
fish
and meat and intake of contaminated river water.
Our results confirm that occupationally and nonoccupationally exposed subjects accumulate mercury, and in discussions with the sampling team
some individuals described symptoms which were
consistent with mercury toxicity.
Despite the widespread publicity which has attended this topic [12] there are few published
reports of the concentrations of mercury in human subjects living or working in these regions
[13]. The results from this study clearly indicate
that widespread environmental
contamination
with mercury exists and that further monitoring
of environmental, animal and human specimens is
necessary.
References
RI F. Bakir, Methylmercury poisoning in Iraq, Science, 191
(1973) 230-241.

t21 J.O. Nriagu, The Biogeochemical Cycle of Mercury in

the Environment, N.-Holland/Elsevier,
Amsterdam,
1979.
131 M. Harada, Minamata disease: organic mercury poisoning caused by ingestion of contaminated food, in E.P.F.
Jelliffe and D.B. Jelliffe (Eds), Adverse Effects of Foods,
Plenum Publishing, 1982, pp. 135-148.
[41 J.J. Putman, Quicksilver and slow death, Natl. Geographic, October (1972) 507-527.
[51 T.W. Clarkson, Mercury poisoning, in S.S. Brown (Ed),
Clinical Chemistry and Chemical Toxicity of Metals,
Elsevier/N.-Holland, Amsterdam, 1977, pp. 189-200.
[61 WC. Pfieffer, Mercury concentrations in inland waters
of gold mining areas in Rondonia, Brazil. Sci. Total
Environ., 87/88 (1989) 233-240.
[71 L.D. Lacerda, W.C. Pfieffer, R.V. Marins, S. Rodrigues,
C.M.M. Souza and W.R. Bastos, Mercury dispersal in
water, sediments and aquatic biota of a gold mining
tailing deposit drainage in Pocone, Brazil, Water Air
Soil Pollut., 55 (1991) 283-294.
181 L. Magos, Selective atomic absorption determination of
inorganic and methylmercury in undigested biological
samples. Analyst, 96 (1971) 847-853.

D. Palhetu,

A. Tayfor

/ The Science of the Total Enuironment

[9] L. Magos and T.W. Clarkson, Atomic absorption determination of total, inorganic and organic mercury in
blood, J. Off. Assoc. Anal. Chem., 55 (1972) 966-971.
[lo] D.C. Shanna and P.S. Davis, Direct determination of
mercury in blood by use of sodium borohydride, Chn.
Chem., 25 (1979) 769-772.
[ill
G. Lindstedt, A rapid method for the determination of
mercury in urine, Analyst, 95 (1970) 264-267.
[12] MS. Serrill, The poisoned Amazon, Time, January 10
(1994) 30-31.
[13j J. Finkelman, G. Corey and R. Calderon (Eds), Environ-

I68 (1995)

63-69

69

mental Epidemiology: A Project for Latin America and

the Caribbean, Pan American Center for Human
Ecology and Health (WHO), Mexico DF, Mexico, 1993.
[14] C. Minoia, E. Sabbioni, P. Apostoli, R. Pietra, L. Pozzoli, M. Gallorini, G. Nicolaou, L. Alessio and E.
Capodaglio, Trace element reference values in tissues
from inhabitants of the European Community. I. A
study of 46 elements in urine, blood and serum of
Italian subjects, Sci. Total Environ., 95 (1990) 89-105.
[15] A. Taylor, Usefulness of measurements of trace elements in hair, Ann. Clin. Biochem., 23 (1986) 364-378.