G Model

EXIS 329 No. of Pages 9

The Extractive Industries and Society xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

The Extractive Industries and Society

journal homepage: www.elsevier.com/locate/exis

Original article

Assessing releases of mercury from small-scale gold mining sites in

Ghana
Martin J. Clifford
Department of Geography, Pearson Hall, University of Delaware, Newark, United States

A R T I C L E I N F O A B S T R A C T

Article history:
Received 4 December 2015 The small-scale gold mining sector has been and continues to be identified as a major source of mercury
Received in revised form 16 May 2017 contamination. Although many studies have assessed the human and ecological health impacts of
Accepted 16 May 2017 mercury emissions in the vicinity of small-scale mining operations, only a handful have surveyed the
Available online xxx source of the contamination: the sites themselves. This paper reports findings from an environmental
analysis undertaken directly within three small-scale gold mining sites in Ghana. Using targeted soil and
Keywords: water sampling analysed through mass spectrometry, the findings are compared with findings in the
Artisanal mining literature and safety guidelines to gauge the scale of contamination at surveyed sites. The analysis reveals
Mercury
that there is widespread pollution, often several orders higher than background levels. Expanding upon
Ghana
this, the results are extrapolated to provide estimates for mercury emissions accountable to the artisanal
Gold mining
Minamata Convention mining sector for Ghana as a whole.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction: mercury, artisanal and small-scale gold mining
and the Minamata Convention
The artisanal and small-scale mining sector is estimated to
employ in excess of 25 million people (Hruschka and Echavarria,
2011) across 50 countries worldwide. Its workforce is engaged in
the extraction of a diverse range of minerals. Gold is the most
widely sought commodity by artisanal miners owing to its
relatively ubiquity and high value-to-weight ratio. An extremely
buoyant world gold price, in combination with widespread poverty
and wider socio-economic and political instability, has stimulated
the sector’s rapid growth worldwide. The sector has grown to the
extent that it now contributes an estimated 20–30% of total global
gold production (Veiga et al., 2006).
Artisanal gold prospectors almost exclusively use elemental
mercury (locally referred to as ‘med’) to extract gold. Although
efficient (Teschner et al., 2017), mercury can have numerous wide-
ranging deleterious effects on humans, biota and the environment
even at levels marginally exceeding naturally occurring ‘back-
ground’ limits (WHO, 2003; Clifford et al., 2010; Clarkson, 2002). It
is estimated that between at least 1 and 2 grams of mercury is lost
to the environment for every gram of gold that is produced (Veiga
and Baker, 2004). With next to no technology in place to recover
mercury used in refining gold, the vast percentage is emitted into
E-mail address: martinjc@udel.edu (M.J. Clifford).
the environment. The United Nations Environment Programme
(UNEP) estimates that ASM now constitutes the largest ‘intentional
use’ of mercury worldwide. It consumes and emits between 640
and 1350 tons of mercury per year, around one third of all
anthropogenic emissions (UNEP, 2013).
An extensive body of research spanning some 40 years has
drawn attention to how the growth of ASM has been responsible
for increased mercury contamination (e.g. Martinelli et al., 1988;
Malm et al., 1990; Lacerda et al., 1991; Lin et al., 1997; Marrugo-
Negrete et al., 2008; Olivero-Verbal et al., 2016; Gibb and O'Leary,
2014). This body of literature has also helped to raise awareness of
the impacts of mercury and legitimise its seriousness as a global
environmental concern (see Pacyna et al., 2016; Nriagu 1994;
Swain et al., 2007; UNEP, 2008, 2013). Mercury pollution is now on
the radar of a number of development organisations, which have
committed to remediating environments left contaminated by
artisanal and small-scale gold miners. Local interventions made by
bilateral donors1 with the support of small grants from the likes of
the United Nations and European Union, would pave the way for
the Global Mercury Project, a three year, multi-agency initiative
which operated across several countries.2 The work carried out
1
Such as the development agencies of Germany in Ghana (see Hilson and
Clifford, 2010), Denmark in Tanzania and others (see GEUS, 2008) and Canada in
Guyana (see Hilson and Vieira, 2007), for example.
2
See https://unites.uqam.ca/gmf/intranet/gmp/front_page.htm for more infor-
mation.

http://dx.doi.org/10.1016/j.exis.2017.05.007
2214-790X/© 2017 Elsevier Ltd. All rights reserved.

Please cite this article in press as: M.J. Clifford, Assessing releases of mercury from small-scale gold mining sites in Ghana, Extr. Ind. Soc. (2017),
http://dx.doi.org/10.1016/j.exis.2017.05.007
G Model
EXIS 329 No. of Pages 9

2 M.J. Clifford / The Extractive Industries and Society xxx (2017) xxx–xxx

under the auspices of this project, however, has failed to reduce
mercury emissions significantly. The author, like many other
researchers, has reflected critically on why, to date, this has proved
so elusive (see Clifford, 2010, 2014).
The most recent and certainly most far-reaching effort made to
date to address the problem is the Minamata Convention on
Mercury, an international agreement, ratified in 2013 by 128
countries, which seeks to stimulate a reduction in global mercury
emissions from a wide range of sources. Quite understandably,
ASM is highlighted within the Convention as a particular area of
concern. Each signatory country with ‘more than significant’
artisanal gold mining activity is required to formulate and submit a
‘national action plan’ outlining the steps they propose to take to
tackle mercury pollution in the sector. The Convention was
adopted in October 2013 and the aforementioned plans were
scheduled to be drafted within three years. It is likely that many
signatory countries with notable ASM activity will find drafting
and implementing such plans extremely challenging because of a
lack of vital baseline data, coherent strategies, and resources to
tackle mercury pollution in the sector (Sippl, 2015). However, the
Minamata Convention has catalysed a renewed wave of research
aimed at gathering additional information about mercury pollu-
tion in the ASM sector and proposing ways to tackle the problem
(see, for example, Porgo and Gokyay 2017; Fritz et al., 2016; Bose-
O'Reilly et al., 2016), including Ghana (Basu et al., 2015a, 2015b;
Rajaee et al., 2015a). This paper helps to address ‘the substantial
gaps in data availability’ (Rajaee et al., 2015a: 8974) in this area,
drawing on experiences from Ghana.
2. The case of Ghana: context, previous work and an estimate of
mercury emissions from ASM
Ghana, famous for the abundance of its gold, is home to an
estimated 500,000 to 1,000,000 small-scale gold miners. Accord-
ing to the Ghanaian Minerals Commission, the chief government
body in the country responsible for policymaking in the mining
sector, these operators extracted 39,642,782 oz of gold from 1980
to 2008 and contributed almost of quarter of national production
in 2010 (Tetteh, 2010). Following the AuPRODUCED:HgLOST ratio of 1:1
to 1:2 cited above, processing of this gold released an estimated
1123–2246 tons of mercury into the natural environment. The
2,796,834 oz. (79 tons) of gold produced in the year 2008 alone was
responsible for the release of an estimated 79–158 tons of mercury.
These figures are of serious environmental concern, especially
when considering that estimates of 640–1350 tons have been given
as the worldwide total of mercury emissions from ASM (UNEP,
2013).
Ghana has been a focal point of research on ASM for many years.
It has attracted significant research on mercury in particular, work
which has sought to provide an idea of the scale of contamination
and impacts of the sector on surrounding communities and natural
environments (e.g. Mensah et al., 2012; Ahiamadjie et al., 2011;
Babut et al., 2003; Bonzongo et al., 2003; Golow and Adzei 2002;
Golow and Mingle 2003; Paruchiri et al., 2010; Rambaud et al.,
2000; Rajaee et al., 2015a). Surprisingly, however, studies which
have looked specifically at the pollution of soils and waterbodies
found within the direct catchment areas of the country’s ASM sites
are in short supply. Most research carried out on mercury
contamination in the country has focused on communities
adjacent to those supporting mining activity (e.g. Ahiamadjie
et al., 2011; Golow and Adzei 2002; Golow and Mingle 2003;
Amonoo-Neizer et al., 1996). Data collected from within the
boundaries of sites themselves (e.g. Paruchuri et al., 2010) would
certainly yield a more complete picture of the problem. Such
findings have long served as important reference points in the
ASM-Hg literature.
3. Criteria assessed
Veiga and Baker (2004), who offer a set of Protocols for
Environmental and Health Assessment of Mercury Released by
Artisanal and Small-Scale Gold Miners, list four different ways in
which analysis of soil and water may be used for substantiating
mercury contamination at artisanal gold mining sites (Table 1):
identifying ‘hotspots’, estimating total mercury releases and
assessing water quality against guidelines for drinking and
protection of aquatic life. These are discussed further below (with
the two water-related methods being placed together) and their
implementation elaborated upon in Section 4.
3.1. Identification of mining and environmental ‘hotspots’
‘Hotspots’ are areas which have extremely high mercury
concentration. They are defined by Veiga and Baker (2004) as
having levels 100 times higher than what could be considered
naturally occurring. Their existence is owed largely to current and/
or pre-existing artisanal gold mining activity. Identifying terres-
trial hotspots at active mine sites is a relatively simple procedure,
consisting of sampling locations where miners amalgamate gold,
and analysing these for total mercury content. This strategy formed
the first strand of the research: soil and water collected from
mining sites was analysed and compared with relevant literature
to determine the severity of contamination.
3.2. Estimating mercury releases
Estimating emissions of mercury from ASM operations is, as
Veiga and Baker (2004: 37) explain, a ‘crucial’ part of any
assessment of the impact small-scale mining has in a locality.
Obtaining an estimate involves collating direct and indirect data on
issues such as the quantity of mercury being introduced and
recovered at each stage of processing (through weighing at each
point), purchases of mercury over a period of time, gold production
figures, and the mercury content of amalgam and tailings, to

Table 1
Use of soil and water analysis in assessing environmental impacts of mercury use in ASM.

Subject Purpose Material Sampled Technique

Identification of mining &
environmental hotspots
Estimate Hg releases
Drinking water guidelines
Protection of aquatic life
guidelines
Apply remedial procedures depending
on bioavailability
Estimate the relevance of Hg released
by ASM in the region
Determine quality to meet guidelines;
resolve health issues
Determine contamination to meet
guidelines
Superficial soil and bottom sediments Hg analysis (semi or quantitative) or panning for
visual inspection
Hg introduced and recovered; Hg lost in Hg lost: Au produced ratio estimation through Hg
amalgam, tailings, etc. balance in ASM operations
Filtered and unfiltered water Total Hg analysis
Filtered and unfiltered water Total Hg analysis

Taken from Veiga and Baker (2004).

Please cite this article in press as: M.J. Clifford, Assessing releases of mercury from small-scale gold mining sites in Ghana, Extr. Ind. Soc. (2017),
http://dx.doi.org/10.1016/j.exis.2017.05.007
G Model
EXIS 329 No. of Pages 9
M.J. Clifford / The Extractive Industries and Society xxx (2017) xxx–xxx
Fig. 1. Sketch map representative of site layout and sampling in Bolgatanga.
construct figures from which to extrapolate rough emission
amounts for mining regions as a whole. As Veiga and Baker
(2004: 11) note, a much higher proportion of mercury added to
material is lost during the heating of amalgam than is discarded
with tailings. Measured mercury levels in amalgam and tailings
were used along with secondary evidence (gold production
figures) to provide general estimates of mercury releases for
Ghana’s ASM sector.
3.3. Guidelines for drinking water and protection of aquatic life
The final two tested indicators, analysis of water in relation to
guidelines on drinking water and for the protection of aquatic life,
are given ‘low’ priority in the Global Mercury Project’s Protocols.
This, however, is due to the rare detection rate, not because they
inherently represent significantly less of a risk to health (Veiga and
Baker, 2004Veiga and Baker 2004: 80). Both measurements can be
obtained fairly straightforwardly, through analysis of water
samples using spectrometry. Water samples collected at ASM
sites were compared to background and guideline amounts to
determine the extent of the mercury pollution problem.
4. Methods
4.1. Sampling
Soil and water was collected from three regions in Ghana with
significant ASM activity: the localities of Akwatia, Prestea and
Bolgatanga. Piles of waste tailings discarded after panning and
pools used for washing gold were targeted for sampling. Samples
were collected from various sites across the three different field
locations over a three-month period. The typical layout of mining
operations at each of the three sites visited and the typical pattern
at each location are provided in shown in Figs. 1–3 .
In Prestea and Bolgatanga, a small amount of soil (the largest
being approximately 2 kg) was collected, using a small trowel, from
the following places: the top layer of stockpiled tailings heaps at
washing and processing centres, and the bottom of washing pools,
which were specially constructed, permanent pits of between
Please cite this article in press as: M.J. Clifford, Assessing releases of mercury from small-scale gold mining sites in Ghana, Extr. Ind. Soc. (2017),
http://dx.doi.org/10.1016/j.exis.2017.05.007
3
50 cm to 1 metre in depth. Water samples were collected from
constructed washing pools at ‘sumps’ (processing centres) in the
same locations. The opportunity also arose on three occasions at
these sites to collect water which had been used to rinse the
mercury-gold amalgamation free of gangue material directly from
a miners pan.3 Additionally, in Bolgatanga, following the advice of
the local Minerals Commission officer, samples were taken from
the river which flows through the mining area and is used for
washing of gold during the wet season.
At mine sites in Akwatia, which were more rudimentary, there
were no specifically designated areas where miners refined and
washed gold (see Fig. 3). Miners working at the sites were,
therefore – following the suggestion of Veiga and Baker (2004) –
asked to indicate where these activities were often carried out and
soil and water samples were collected there. Surface soil was
collected from the ground at these locations, and water taken from
flooded pits and pools used for washing gold. Material was also
taken from pools, at a depth of between 50 cm and 1 metre.
4.2. Analysis of targeted soil and water samples
Soil samples were placed in sealable polythene bags at the site,
air dried upon returning from the field location on the same day,
and then returned to a polythene bag. They were stored at room
temperature and ‘double bagged’ to avoid any contamination until
their return to the UK for analysis. Water samples were filtered on
the same day of collection following return from the fieldwork
location. Both soil and water samples were stored at room
temperature in the field, and in a cold store upon arrival in the UK,
where soils were thoroughly dried and then homogenised as much
as possible using a pestle and mortar and passed through a metallic
5 mm sieve. Water samples were acidified to approximately 9%
hydrochloric acid (HCl). All samples were then stored in cold
storage at 3  C until preparation for analysis.
3
These fines are then normally discarded completely or added to tailings heaps
for future reprocessing.
G Model
EXIS 329 No. of Pages 9

4 M.J. Clifford / The Extractive Industries and Society xxx (2017) xxx–xxx

Fig. 2. Sketch map representative of site layout and sampling in Prestea.

Fig. 3. Sketch map representative of site layout in Akwatia.

Please cite this article in press as: M.J. Clifford, Assessing releases of mercury from small-scale gold mining sites in Ghana, Extr. Ind. Soc. (2017),
http://dx.doi.org/10.1016/j.exis.2017.05.007
G Model
EXIS 329 No. of Pages 9
M.J. Clifford / The Extractive Industries and Society xxx (2017) xxx–xxx
Table 2
Guideline limits for elemental Hg in soil and water.
Institution
EA Soil Guideline Values (SGV)
Canadian Environmental
Quality Guidelines (CEQG)
World Health Organisation
Atomic absorption (AA) analysis of aqua regia (AR) digests was
used to determine the total mercury content of the collected
samples. The digests were performed following the University of
Reading’s departmental procedures for which there were numer-
ous quality control measures in place.4 In total, 36 of the 49 soil
samples were selected for analysis: 12 from each of Akwatia,
Prestea, and Bolgatanga. All 50 water samples were used, and were
analysed for total mercury using a Perkin Elmer 1100 B Atomic
Absorption Spectrometer.5
4.3. Analysis of gold-mercury amalgam
An estimate of the loss of mercury from burning gold-mercury
amalgam was made, over a two-day period in Prestea, by weighing
the difference between balls of gold-Hg amalgam that miners
brought to gold shops and the dore gold after burning the amalgam
on a fire. In total, 25 balls of amalgam were tested.
5. Findings
5.1. Mercury use at ASM sites in Ghana: field observations
Mercury or ‘med’ was observed being used extensively in
operations. The techniques observed are similar to those described
by other authors. Mercury is used exclusively at the sites visited to
extract gold from ore. But the more polluting practices docu-
mented elsewhere, such as ‘spiking’ pits with mercury (dispersing
mercury openly into pits, as described by Hilson and Vieira, 2007)
or passing unrefined ore through mercury ‘baths’ (Veiga and Baker,
2004), were not observed: miners rather only added mercury to
more refined gold-bearing material. The resulting amalgam ball
was burned on an open fire in a metal container, often a metal food
tin or dish, until the mercury had volatilised and the ball became a
dull gold dore.
At each of the sites visited, few safety precautions were taken by
those who handle mercury. When burning amalgam, miners would
stand very close to the fire and amalgamate in enclosed spaces and
in residential areas.6 Mercury was observed on numerous
occasions being dropped on to the ground and abandoned. Several
miners also commented that after a period of time, the mercury
would ‘stop working’ (i.e. lose its coalescence to gold), at which
point, it would be discarded at the mine or processing site. The
waste material remaining following amalgamation, which is laced
with mercury, was observed being dumped into rivers and pools
(the latter of which were often located next to and overflowed into
rivers during rainy periods) by many miners, especially those
4
These were based upon British Standard BS 7755 Section 3.9 1995 ISO 11466.
5
The full method for AA analysis has been omitted here for the sake of brevity. An
expanded approach can be obtained by contacting the author directly.
6 7
An extremely dangerous practice within itself, something which Black et al.
(2017) have recently studied in Burkina Faso.
Please cite this article in press as: M.J. Clifford, Assessing releases of mercury from small-scale gold mining sites in Ghana, Extr. Ind. Soc. (2017),
http://dx.doi.org/10.1016/j.exis.2017.05.007
5
Guideline Limits for elemental Hg
‘Residential’ land use: 1 mg/kg
‘Commercial’ and ‘Allotment’ land use: 26 mg/kg
‘Agricultural’ land use: 12 mg/kg for soil contact
‘Residential/Parkland’ land use: 12 mg/kg for soil contact
‘Commercial’ land use: 50 mg/kg for soil contact
‘Industrial’ land use: 50 mg/kg for soil contact
Water for Protection of Aquatic Life: 0.1 mg/l (total mercury)
Upper ‘background’ limit to Hg content of water: 5.5 mg/l
Maximum acceptable limit for drinking water: 6 mg/l
engaged in ‘dig and wash’ operations.7 In more complex
operations, this waste material was stockpiled for future rework-
ing. However, mercury contained in these sizeable heaps of tailings
also poses a serious threat to the wider environment.
5.2. Assessing results of soil and water analysis against environmental
guidelines
The World Health Organisation (WHO) and many Western
environmental and health agencies have established guideline
values for different contaminants in soil and water (Table 2). Of
these, the Soil Guideline Values (SGVs) of the Environment Agency
(EA) of England and Wales (Environment Agency, 2009Morgan
et al., 2009), and the Canadian Environmental Quality Guidelines
(CEQG) (Canadian Council of Ministers of the Environment 1999),
provide particularly comprehensive reference points for a range of
soil contaminants, including mercury. These proved valuable in
assessing and contextualising the severity of mercury contamina-
tion in surveyed ASM sites. Overall, SGVs can be used as ‘a starting
point for evaluating long-term risks to human health from
chemicals in soil’ (Environment Agency, 2009: 5): a ‘representative
site soil concentrations above an SGV might represent a significant
possibility of significant harm’ (Environment Agency, 2009: 8, italics
and bold in original).
Guideline values provided by these agencies have been
established for use in Western scenarios and are informed by a
series of conceptual ‘assumptions’ related to representative levels,
routes, and sources of exposure (Environment Agency, 2009) that
may be exceeded in most ASM scenarios. But, rather than making
them inapplicable for gauging levels of mercury in ASM, they can
instead be used to further highlight the potential severity (e.g.
Rajaee et al., 2015a). For example, the Environment Agency
provides SGVs for ‘commercial’ and ‘residential’ land uses where
the guideline values are 26 mg/kg and 1 mg/kg of mercury in soil,
respectively. These two scenarios differ on account of modelled
factors such as time typically spent at each type of land use and the
likelihood of children being present, for example. The CEQGs
provide an additional ‘industrial’ land use scenario with a guideline
value of 50 mg/kg of mercury. Due to the close proximity of
residential areas to mining operations in many parts of Ghana, it
would be reasonable to assume that ‘residential’ guideline limits
are applicable. However, the discussion below is informed by all of
the scenarios mentioned above in order to emphasise the potential
severity of the levels of mercury found in the analysed samples. For
instance, if levels of mercury are found in soil near to residential
land uses that exceed industrial guidelines, then this is clearly a
cause for concern.
The guideline figures listed in Table 2 are for elemental mercury
(apart from the aquatic protection guideline, which is for total
This refers to individual or small teams of workers, often women, who reprocess
tailings at sites.
G Model
EXIS 329 No. of Pages 9
6 M.J. Clifford / The Extractive Industries and Society xxx (2017) xxx–xxx
mercury). As Veiga and Baker (2004) suggest, the likelihood is that
most of the mercury found at ASM sites will be of the elemental
species, because this is the form used by miners. Guideline values
for inorganic and methylmercury, which may be formed by
emitted elemental mercury interacting with the surrounding
environment, are often much lower. The soil and water sampled
were measured for total mercury so it was not possible to provide a
breakdown of different compounds.
Thirteen (36%) of the soil samples tested had mercury levels
below detection limits (0.21 mg/kg Hg). All but one of these were
taken from Akwatia. Of the 23 remaining samples obtained from
the other two sites, in Prestea and Bolgatanga, 48% yielded
measurements in excess of what can typically be considered
‘background’ levels of mercury in soils (10 mg/kg Hg). Of these
samples, 64% would be considered to contain concentrations
acceptable in cases of ‘industrial’ land use, the highest limit given
in the guidelines above. Almost one fifth (19%) of the analysed
samples contained significant levels of mercury, concentrations at
levels considerably above all of the guideline limits set as being
acceptable for any use of land. Whilst this figure may not appear
high, it is worth considering that the majority of miners at most of
the surveyed sites live and work in close proximity to the places
where samples were collected: the land use is ‘residential’. This
figure is put at a maximum value of 12 mg/kg (in the CSQG).
Assessing the results against this level would bring the number of
tailings samples analysed in excess of guideline limits to 30%.
All of the water samples contained mercury above detection
limits, with elevated levels (>5.5–6 mg/l) found at all the sampled
sites. Excluding a figure of 77.9 mg/l, which is significantly higher
than the other samples, the average reading was 8.4 mg/l mercury.
Each site yielded at least one sample approaching or exceeding
twice the suggested naturally-occurring maximum.
The measured values in the soil samples correspond well with
mercury levels reported in other research, therefore confirming
their validity. The maximum measurement of 901.5 mg/kg mercury
in tailings (taken from a processing centre in Prestea) compares
favourably to the highest reading reported in the literature:
1414 mg/kg mercury found by Van Straaten (2000: 109) at an ASM
site in Tanzania and 1500 mg/kg found by Odumo et al. (2014) in
Kenya. Interestingly, this reading far exceeds the maximum of
330 mg/kg identified in Ghanaian studies (Rajaee et al., 2015a:
8978). This, and the other very elevated samples, also corresponds
to the amounts of mercury suggested to be ‘frequently’ found in
tailings – between 200 and 500 mg/kg Hg – by Veiga and Baker
(2004: 21). Samples that contained elevated but not excessively
high amounts of mercury also generally matched the average
figure of 10 mg/kg mercury found in artisanal mining tailings
measured at a Brazilian site by de Andrade Lima et al. (2008: 751),
where similar amalgamation processes to those observed in Ghana
were used. If readings exceeding 50 mg/kg mercury are excluded in
order to provide a more representative figure for lower range
samples, then the average mercury content of the sampled tailings
is 7 mg/kg mercury. If the one figure falling outside of the range
noted as often found in contaminated tailings by Veiga and Baker
(2004) – 901.5 mg/kg mercury – is excluded, then the average
figure of the remaining soil samples above detection limits is
80.33 mg/kg mercury. This corresponds relatively well with the
work of Odumo et al. (2014) who found that 90% of their samples
from a Kenyan ASM site measured between 40 and 280 mg/kg.
Comparison of the water samples to findings in the relevant
literature is more difficult than for soils as few studies exist that
8
A Minerals Commission report on gold occurrence in the country classifies ‘high
grade zones’ as 15–45 grams per ton.
9
Figure taken from Styles et al. (2010: 293).
Please cite this article in press as: M.J. Clifford, Assessing releases of mercury from small-scale gold mining sites in Ghana, Extr. Ind. Soc. (2017),
http://dx.doi.org/10.1016/j.exis.2017.05.007
specifically analyse total mercury levels found in standing pools at
ASM sites. However, they do appear to be within the same general
range as studies assessing river water in close proximity to ASM
sites. Rajaee et al. (2015), collating studies conducted in Ghana,
suggested that water samples at almost a quarter of mining sites
exceeded 6 mg/l mercury guideline limits. Malm et al. (1990)
recorded values of up to 9.97 mg/l mercuy in river water around
ASM sites along the Madeira River in Brazil, and Ogola et al. (2002)
recorded levels of 8.0 mg/l mercury in stream water flowing from
mining areas in Kenya. Many of the samples taken from washing
pools were exceeded this, ranging between 10 and 20 mg/l
mercury, and the highest recorded value taken from the Tongo
River in Bolgatanga was 10.86 mg/l mercury.
5.3. Mining ‘hotspots’
The fieldwork sought to determine the persistence of mining
‘hotspots’ at visited sites and to provide a general estimate of
mercury releases accountable to ASM. It seems inevitable that
many ‘mining hotspots’ persist at the study sites. Overall, 10 of the
36 (27%) sampled soils and tailings contained in excess of 100
times what could be attributed to a ‘natural occurrence’. As the
research surveyed a small percentage of operations at a limited
number of sites at only three ASM locations in Ghana, where the
activity is prevalent, it can easily be extrapolated from the findings
that there are thousands of mining ‘hotspots’ across the country.
5.4. Estimating mercury emissions from ASM in Ghana
The average figure of 7 mg/kg given above can be used to
estimate mercury releases from Ghana. This reading is equivalent
to 7 g of mercury per ton of concentrated tailings which are
stockpiled across mining sites at hundreds of locations, such as
Prestea and Bolgatanga, in the country. With each stockpile of
tailings at every sump in Prestea and Bolgatanga easily weighing
several tons, it is likely that many kilograms of mercury are
contained in piles at these sites alone. For example, assuming that
miners are working deposits of 10 grams of gold per ton8 with
recovery rates of 30–40%9 (i.e. 3–4 grams of gold per ton of
material), the 79,289 kg (79,289,000 grams) of gold produced by
artisanal operators in the country for 2008 would have generated
between 237,867,000 and 317,156,000 tons of tailings in total.
However, it is only the concentrated material (referred to by
miners in Ghana as ‘black sand’) containing a mixture of heavy
metals and other geological compounds that is obtained through
washing and sluicing of material to which mercury is added. The
amount of heavy material recovered via sluicing alongside gold
varies widely, which makes identifying specific numbers impossi-
ble. But a fair assumption from repeated observations of panning
and amalgamation in Ghana would suggest approximately 100
parts of ‘black sand’ to gold: that is, 100 grams of ‘black sand’, to
which mercury is added (and was targeted for sampling in the
analysis), for every gram of gold.
Using this figure would mean that 7,928,900 kg (7928 tons) of
mercury-enriched tailings were generated in Ghana in 2008. Using
the conservative average mercury content of 7 mg/kg, an estimate
of 55 kg (0.055 tons) of mercury per year being released along with
these tailings into the environment across the country is reached.
Adopting the higher average mercury content of soil samples given
above 81.33 mg/kg would give a figure of 644 kg (0.644 tons) of
10
This was arrived at by placing the 78 tons of gold produced in Ghana in 2008
into the ratio of ‘Gold Produced to Hg Lost’ of between 1:1 and 1:2.
11
The Precious Minerals Marketing Company of Ghana purchases artisanally
produced gold via a system of licensed gold buyers.
G Model
EXIS 329 No. of Pages 9
M.J. Clifford / The Extractive Industries and Society xxx (2017) xxx–xxx
mercury being released from mining operations. Both of these
figures for the collected soil samples would, in themselves,
represent only a fraction of the figure of between 79 and 158
tons10 of mercury suggested in the introduction that were emitted
from artisanal and small-scale gold mining activities in Ghana in
2008. It would also appear to be well below the AuPRODUCED:HgLOST
ratio given by Veiga and Baker (2004) of between 1:1 and 1:2.
However, with no mechanisms through which these tailings can be
safely dealt with and disposed, it must still be a concern that the
hundreds of kilograms of mercury contained within them is
released into Ghana’s environment year after year, where they may
be dispersed and pose a significant environmental threat.
However, emissions from tailings is just one way in which
mercury is released into the environment by ASM. The burning of
gold also produces significant quantities of mercury emissions. As
Veiga and Baker (2004: 11) note, a much higher proportion of
mercury added to material is lost during the heating of amalgam
than the amount lost from tailings. As indicated above, the amount
of mercury lost in the heating of amalgam was estimated to be
between 33% and 40% of the initial weight. Although the dore may
still contain up to 20% mercury, most of final gold smelting
undertaken by buyers who then sell directly to the Precious
Minerals Marketing Company (PMMC)11 takes place in the same or
similar conditions (i.e. little ventilation or safety precautions) as
those in which the initial burning of amalgam occurs. It can,
therefore, be assumed that the full mercury content in amalgams is
emitted to the environment by the time it reaches its point of sale.
With the understanding that all 79,289 kg of gold in the year 2008
were produced via this route, it could be estimated that between
26 and 31 tons (26,165–31,715 kg) of mercury were emitted
through the burning of amalgam in this one year. Adding this to the
obtained totals from mercury released with tailings (55 kg or
644 kg) gives an overall estimated figure of 26,220–31,770 kg or
26,809–32,359 kg (approximately 26 to 32 tons) of mercury
released by Ghanaian ASM operations in 2008. This figure is still
below what was speculatively offered in the introduction for the
country. But it does nevertheless represent a sizeable and
seemingly proportionate number: the 26 to 32 tons approximation
would account for between 3 and 4% of global emissions if an
overall global figure of 800 tons (a midrange value from estimates
of between 650 and 1000 tons given by the UNEP (2008)) is
assumed. The estimated number of miners in Ghana (using
750,000 from given figures of 500,000 to 1 million) is around 5% of
the estimated global ASM population (using 14 million from
figures of 13 to 15 million).
6. Discussion
The biogeochemical behaviour of mercury emitted from ASM is
extremely challenging to capture due to the interaction of a
multitude of different factors which create unique and changeable
local environmental conditions (see Rajaee et al., 2015a for an
extended discussion of this in reference to Ghana). For example,
naturally-occurring and/or other anthropogenic sources may
elevate or contribute towards levels and forms of mercury in
the environment. As explained by Berzas-Nevado et al. (2010: 595)
in the case of Brazilian ASM, the range of contextual factors that
influence levels, speciation, and transport of mercury in soils ‘make
it difficult to find a definitive conclusion’ on the role played by the
mercury introduced by ASM in the country. In fact, Roulet et al.
(1998, 1999), among others, are of the view that the largest stocks
of mercury in the Brazilian basin a region that houses thousands
of artisanal gold mines and has been intensively studied are
produced under natural conditions or through non-mining related
human activities such as deforestation and agriculture. Yet, as
important as these non-mining inputs may be, it is well recognised
Please cite this article in press as: M.J. Clifford, Assessing releases of mercury from small-scale gold mining sites in Ghana, Extr. Ind. Soc. (2017),
http://dx.doi.org/10.1016/j.exis.2017.05.007
7
that ASM can pose a significant addition to this existing
environmental burden.
This ambiguity over the amount of mercury pollution attributed
to ASM makes attempting to capture mercury as close to the source as
possible a very appropriate strategy. This is further justified by the
mercury-ASM contamination literature. Environmental surveys that
have collected random samples from areas impacted by artisanal
gold mining have obtained indefinite results in locating the main
sources of emitted mercury. This is despite the knowledge that very
sizeable amounts have been used by ASM operations. Even if there is
a strong suspicion that ASM is releasing significant amounts of
mercury, locating where exactly mercury originating from the
activity ends up can be extremely difficult. For instance, Richard
et al. (2000), who analysed soils and sediments collected randomly
along rivers known to be influenced by ASM activity, obtained
extremely ‘heterogeneous’ readings of mercury in their analysed
materials. Indeed, one of the admitted limitations of their work was
not being able to locate mine sites. Moreover, those who have taken
samples directly from operational ASM sites appear to have had
some success in locating hotspots and sources of significant
amounts of mercury. The few studies that have measured mercury
levels in tailings and washing pools have captured some extremely
high levels of mercury, which must be accounted for by mining
activities (see de Andrade Lima et al., 2008; Van Straaten, 2000). A
review of the relevant evidence, therefore, indicates that targeted
sampling of existing ASM operations is a valid option for not only
locating the source of mercury emissions but more accurately
assessing their potential severity.
The above is readily demonstrated by the relative differences in
mercury levels found at each fieldwork site. The view here is that
the observable differences between the levels of mercury in soil
and water samples collected at each site are definitely linked to the
organisation of mining. Where tailings were stockpiled and
designated washing pools existed, it was possible to collect soil
and water with a much higher likelihood of contamination. The
difference in sampled soil was more or less directly reflected in the
results of the analysis, with the mercury content being consider-
ably higher both on average and in terms of maximum recorded
values where targeted sampling took place (i.e. for both soil and
water in Bolgatanga and Prestea, and for some water samples from
Akwatia, where certain pools were used for washing gold). What
this indicates is that areas where tailings are concentrated, and
washing pools that are repeatedly used are much more likely to
contain high levels of mercury.
The role of ‘non-mining’ inputs of mercury into the soil analysis
figures can in many instances be safely discounted. Sizeable soil
disturbance in terms of deforestation and pit extraction at the sites
may introduce additional levels of naturally-occurring mercury
into the local environment (Veiga and Baker, 2004; Roulet et al.,
1998) some of which will be inadvertently recovered in
concentrating gold and added to tailings piles and wash-pools
as waste. But the results do show some levels of mercury that are
certain to be accountable to anthropogenic inputs. The highest soil
sample was 18 times over that suggested as being safe for
‘industrial’ land use. In contrast to the soil samples, it is possible
that the elevation of mercury content in some of the water samples
could be attributable to external influences, such as deforestation
and soil disturbance (exacerbated by mining activity) from which
run-off reaches sampled washing pools. They are also likely to have
measured higher than in rivers and catchment systems, owing to
the fact that water in sampled pools was concentrated in a
confined area. However, it is equally likely that water in washing
pools displays elevated mercury content because of repeated
additions of elemental mercury during amalgamation and from
releases of the element from the tailings and sediments lying
underneath it.
G Model
EXIS 329 No. of Pages 9
8 M.J. Clifford / The Extractive Industries and Society xxx (2017) xxx–xxx
7. Conclusion
The environmental analysis described above aimed to quantify
typical mercury contamination from small-scale mining activity in
Ghana, and to use this to draw inferences about total levels of
emissions accountable to the sector in the country as a whole.
Observations of prevalent practice and the findings of the analysis
corresponded with the large body of literature that associates the
existence of ASM with significant levels of mercury use and
associated contamination.
Over half (54%) of the targeted water samples from active AGSM
sites that were analysed contained levels above background levels
and in excess of guidelines for drinking water and protection of
aquatic life. Of the 36 targeted soil samples, 33% contained mercury
levels above background levels and 19% contained mercury levels
that would be deemed unacceptable for any land use (i.e. be seen as
contaminated) according to Western environmental guidelines.
Moreover, 27% of the tailings measured in excess of 100 times what
could be attributed to natural occurrence, indicating the presence
of ‘mining hotspots’.
Gold production figures for Ghana’s ASM have been combined
with average mercury levels in surveyed tailings and gold-mercury
amalgam to offer a figure of 26 to 32 tons of mercury being released
in 2008. The figures reached, though rough, are of obvious concern
considering the amount of gold being produced using mercury
each year in the country and the proximity of mine operations to
communities. The results provide comparable and substantial
evidence that the mercury being used at the surveyed small-scale
gold mine sites in Ghana is loading tailings and washing pools, and
consequently probably the adjacent environment, with sizeable
amounts of the hazardous metal. Indeed, one of the most pressing
concerns associated with mercury pollution is not only the
immediate detrimental effects to human and environmental
health but the legacy that the element’s use in artisanal mining
leaves for future generations: mercury’s biogeochemistry means
that it typically resides in the environment and may pose a
significant threat for decades after mining has ceased (Nriagu,
1994; Veiga and Hinton, 2002). In fact, Lacerda and Salomons
(1998), in a widely cited work, liken the widespread use of mercury
in artisanal gold extraction, particularly the presence of mercury-
rich tailings, to a ‘chemical time bomb’.
While the findings above have obvious limitations in their
wider applicability specifically, they did not capture the effects
on human and ecological health that has made mercury pollution
such a global concern they do provide depth to, firstly, the
relatively limited number of studies that have surveyed mercury
contamination at active ASM sites in Ghana and, secondly, to the
body of work that points to the sector being a significant emitter of
the metal. It is precisely the kind of indicative findings arising from
studies like this that have made mercury use within ASM a
pressing environmental issue, prompting increasingly unilateral
action for solutions such as the Minamata Convention. As a
signatory nation, it is also an ongoing problem that Ghana, aided by
international partners and researchers, is now obliged to
proactively confront once more a process that already appears
to be encouragingly underway (Basu et al., 2015a, 2015b; Rajaee
et al., 2015a, 2015b). Baseline information will, of course, be key in
formulating effective strategies towards tackling mercury emis-
sions. It is hoped that this study will contribute towards this effort.
Acknowledgements
This paper is based upon an interdisciplinary doctoral research
project jointly funded by the Economic and Social Research Council
and the Natural Environment Research Council. The author is
hugely appreciative of their funding and support. Many thanks go
Please cite this article in press as: M.J. Clifford, Assessing releases of mercury from small-scale gold mining sites in Ghana, Extr. Ind. Soc. (2017),
http://dx.doi.org/10.1016/j.exis.2017.05.007
to the Minerals Commission in Ghana for their receptiveness and
assistance with the research. Finally, but by no means least, the
research would not have been possible without trusted fieldwork
assistants and the openness and involvement of the Ghana’s small-
scale mining community.
References
Ahiamadjie, H., Serfor-Armah, Y., Tandoh, J.B., Gyampo, O., Ofosu, F.G., Dampare, S.B.,
Adotey, D.K., Nyarko, B.J.B., 2011. Evaluation of trace elements contents in staple
foodstuffs from the gold mining areas in south-western part of Ghana using
neutron activation analysis. J. Radioanal. Nucl. Chem. 288, 653–661.
Amonoo-Neizer, E.H., Nyamah, D., Bakiamoh, S.B., 1996. Mercury and arsenic
pollution in soil and biological samples around the mining town of Obuasi,
Ghana. Water Air Soil Pollut. 91 (3–4), 363–373.
Babut, M., Sekyi, R., Rambaud, A., Potin-Gautier, M., Tellier, S., Bannerman, W.,
Beinhoff, C., 2003. Improving the Environmental management of small-scale
gold Mining in Ghana: a case study of Dumasi. J. Clean. Prod. 11 (2), 215–221.
Basu, N., Renne, E.P., Long, R., 2015a. An integrated assessment approach to address
artisanal and small-Scale gold mining in Ghana. Int. J. Environ. Res. Public
Health 12, 11683–11698.
Basu, N., Clarke, E., Green, A., Calys-Tagoe, B., Chan, L., Dzodzomenyo, M., Fobil, J.,
Long, R.N., Neitzel, R.L., Obiri, S., Odei, E., Ovadje, L., Quansah, R., Rajaee, M.,
Wilson, M.L., 2015b. Integrated assessment of artisanal and small-Scale gold
mining in Ghana ? part 1: human health review. Int. J. Environ. Res. Public
Health 12, 5143–5176.
Berzas-Nevado, J.J., Rodrıguez Martın-Doimeadois, R.C., Guzman Bernardo, F.J.,
Jimenez Moreno, M., Herculano, A.M., doNascimento, J.L.M., Crespo-Lopez, M.E.,
2010. Mercury in the tapajos river basin brazilian amazon: a review. Environ.
Int. 36 (6), 593–608.
Black, P., Richard, M., Rossin, R., Telmer, K., 2017. Assessing occupational mercury
exposures and behaviours of artisanal and small-scale gold miners in Burkina
Faso using passive mercury vapour badges. Environ. Res. 152, 462–469.
Bonzongo, J.C., Donkor, A.K., Nartey, V.K., 2003. Environmental impacts of mercury
related to artisanal gold mining in Ghana. J. Phys. IV 107, 217–220.
Bose-O'Reilly, S., Schieri, R., Nowak, D., Siebert, U., William, J.F., Owi, F.T., Ir, Y.I., 2016.
A preliminary study on health effects in villagers exposed to mercury in a small-
scale artisanal gold mining area in Indonesia. Environ. Res. 149, 274–281.
Clarkson, T.W., 2002. The three modern faces of mercury. Environ. Health Perspect.
110 (1), 11–23.
Clifford, M.J., Hilson, G.M., Hodson, M.E., 2010. Tin and mercury chapter 21. In:
Hooda, P.S. (Ed.), Trace Elements in Soils. John Wiley & Sons Ltd., Chichester,
United Kingdom.
Clifford, M.J., 2010. Potential repercussions of a mercury ban on the artisanal and
small-scale gold mining sector: a viewpoint. Int. J. Environ. Pollut. 41 (3/4), 229–
241.
Clifford, M.J., 2014. Future strategies for tackling mercury pollution in the artisanal
gold mining sector: making the Minamata Convention work. Futures 62, 106–
112.
Environment Agency, 2009. Using Soil Quality Guidelines. Environment Agency,
Bristol.
Fritz, M.M.C., Maxson, P.A., Baumgartner, R.J., 2016. The mercury supply-chain,
stakeholders and their responsibilities in the quest for mercury-free gold.
Resour. Policy 50, 177–192.
Geological Survey of Denmark and Greenland(GEUS) (2008) Small-scale mining
http://www.geus.dk/geuspage-uk.htm. Last Accessed 05 September 2012.
Gibb, H., O'Leary, K.G., 2014. Mercury exposure and health impacts among
individuals in the artisanal and small-Scale gold mining community: a
comprehensive review. Environ. Health Perspect. 122 (7), 667–672.
Golow, A.A., Adzei, E.A., 2002. Mercury in surface soil and cassava crop near an
alluvial goldmine at dunkwa-on-Offin, Ghana. Bull. Environ. Contam. Toxicol.
69, 228–235.
Golow, A.A., Mingle, L.C., 2003. Mercury in river water and sediments in some rivers
near dunkwa-On-Offin, an alluvial Goldmine, Ghana. Bull. Environ. Contam.
Toxicol. 70, 379–384.
Hilson, G., Clifford, M.J., 2010. Small-scale gold mining, the environment and human
health: an introduction to the Ghana case. Int. J. Environ. Pollut. 21 (3/4), 185–
194.
Hilson, G., Vieira, R., 2007. Challenges with minimising mercury pollution in the
small-scale gold mining sector: experiences from the Guianas. Int. J. Environ.
Health Res. 17 (6), 429–441.
Hruschka, F., Echavarria, C., 2011. Rock-solid chances: for responsible artisanal
mining. Series on Responsible ASM No. 3 2011. Alliance for Responsible Mining
[ARM].
de Lacerda, L.D., Salomons, W., 1998. Mercury from Gold and Silver Mining: a
Chemical Time Bomb? Springer, Verlag, New York.
de Lacerda, L.D., Pfeiffer, W.C., Marins, R.V., Rodrigues, S., Souza, C.M.M., Bastos, W.
R., 1991. Mercury dispersal in water, sediments and aquatic biota of a gold
mining tailing deposit drainage in pocone, Brazil. Water Air Soil Pollut. 55, 283–
294.
Lin, Y.H., Guo, M.X., Gan, W.M., 1997. Mercury pollution from small gold mines in
China. Water Air Soil Pollut. 97 (3–4), 233–239.
Malm, O., Pfeiffer, W.C., Souza, C.M.M., Reuther, R., 1990. Mercury pollution due to
gold mining in the madeira river basin, Brazil. Ambio 19 (1), 11–15.
G Model
EXIS 329 No. of Pages 9
M.J. Clifford / The Extractive Industries and Society xxx (2017) xxx–xxx
Marrugo-Negrete, J., Benitez, L.N., Olivero-Verbel, J., 2008. Distribution of mercury
in several environment compartments in an aquatic ecosystem impacted by
gold mining in northern Colombia. Arch. Environ. Contam. Toxicol. 55, 305–316.
Martinelli, L.A., Ferreira, J.R., Forsberg, B.R., 1988. Mercury contamination in the
amazon a gold rush consequence. Ambio 17 (4), 252–254.
Mensah, E.K., Afari, E., Wurapa, F., Sackey, S., Quainoo, A., Kenu, E., Nyarko, K.M.,
2012. Exposure of small-scale gold miners in prestea to mercury, Ghana. Pan
Africa Med. J. 1, 6.
Nriagu, J.O., 1994. Mercury pollution from the past mining of gold and silver in the
Americas. Sci. Total Environ. 149, 167–181.
Odumo, B.O., Carbonell, G., Angeyo, H.K., Patel, J.P., Torrijos, M., Martin, J.A.P., 2014.
Impact of gold mining associated with mercury contamination ins soil biota
sediments and tailings in Kenya. Environ. Sci. Pollut. Res. 21, 12426–12435.
Ogola, J.S., Mitullah, W.V., Omulo, M.A., 2002. Impact of gold mining on the
environment and human health: a case study in the Migori gold belt, Kenya.
Environ. Geochem. Health 24, 141–158.
Olivero-Verbal, J., Carranza-Lopez, L., Caballero-Gallardo, K., Ripoll-Arboleda, A.,
Munoz-Sosa, D., 2016. Human exposure and risk assessment associated with
mercury pollution in the Caqueta Rive, Colombian Amazon. Environ. Sci. Pollut.
Res. 23, 20761–20771.
Pacyna, J.M., Travnikov, O., De Simone, F., Hedgecock, I.M., Sundseth, K., Pacyna, E.G.,
Steenhuisen, F., Pirrone, N., Munthe, J., Kindbom, K., 2016. Current and future
levels of mercury atmospheric pollution on a global scale. Atmos. Chem. Phys.
16, 12495–12511.
Paruchuri, Y., Suiniak, A., Johnson, N., Levin, E., Mitchell, K., Goodrich, J.M., Renne, E.
P., Basu, N., 2010. Occupational and environmental mercury exposure among
small-scale gold miners in the Talensi-Nabdam District of Ghana’s Upper East
Region. Sci. Total Environ. 408, 6079–6085.
Porgo, M., Gokyay, O., 2017. Environmental impacts of gold mining in Essakane site
of Burkina Faso. Human Ecol. Risk Assess.: Int. J. 23 (3), 641–654.
Rajaee, M., Obiri, S., Green, A., Long, R., Cobbina, S.J., Nartey, V., Buck, D., Antwi, E.,
Basu, N., 2015a. Integrated assessment of artisanal and small-Scale mining in
Ghana – part 2: natural sciences review. Int. J. Environ. Res. Public Health 12,
8971–9011.
Rajaee, M., Long, R.N., Renne, E.P., Basu, N., 2015b. Mercury exposure assessment
and spatial distribution in ghanaian small-Scale gold mining community. Int. J.
Environ. Res. Public Health 12, 10755–10782.
Richard, S., Arnoux, A., Cerdan, P., Reynouard, C., Horeau, V., 2000. Mercury levels of
soils, sediments, and fish in French Guiana, South America. Water Air Soil Pollut.
124, 221–224.
Please cite this article in press as: M.J. Clifford, Assessing releases of mercury from small-scale gold mining sites in Ghana, Extr. Ind. Soc. (2017),
http://dx.doi.org/10.1016/j.exis.2017.05.007
9
Roulet, M., Lucotte, M., Saint-Aubin, A., Tran, S., Rhéault, I., Farella, N., De Jesus Da
Silva, E., Dezencourt, J., Sousa Passos, C.-J., Santos Soares, G., Guimarães, J.-R.D.,
Mergler, D., Amorim, M., 1998. The geochemistry of mercury in central
Amazonian soils developed on the Alter-do-Chao formation of the lower
Tapajos River Valley, Para state, Brazil. Sci. Total Environ. 223 (1), 1–24.
Roulet, M., Lucotte, M., Farella, N., Serique, G., Coelho, H., Sousa Passos, C.J., Da Silva,
E.J., De Andrade, P.S., Mergler, D., Guimarães, J.-R.D., Amorim, M., 1999. Effects of
recent human colonization on the presence of mercury in Amazonian
ecosystems. Water Air Soil Pollut. 112 (3-4), 297–313.
Sippl, K., 2015. Private and civil society governors of mercury pollution from
artisanal and small-scale gold mining: a network analytic approach. Extr. Ind.
Soc. 2, 198–208.
Swain, E.B., Jakus, P.M., Rice, G., Lupi, F., Maxson, P.A., Pacyna, J.M., Penn, A., Spiegel,
S.J., Veiga, M.M., 2007. Socioeconomic consequences of mercury use and
pollution. Ambio 36 (1), 45–61.
Teschner, B., Smith, N.M., Borrillo-Hutter, T., John, Z.Q., Wong, T.E., 2017. How
efficient are they really? A simple testing method of small-scale gold miners'
gravity separation systems. Miner. Eng. 105, 44–51.
Tetteh, K., 2010. An Overview of Ghana's Artisanal and Small-Scale Mining (ASM)
Sector. Minerals Commission of Ghana, Accra.
United Nations Environment Programme, 2008. The Global Atmospheric Mercury
Assessment: Sources, Emissions and Transport. UNEP, Geneva.
UNEP, 2013. Global Atmospheric Mercury Assessment 2013: Sources, Emissions and
Transport. . http://hdl.handle.net/20.500.11822/7984.
Van Straaten, P., 2000. Mercury contamination associated with small-scale gold
mining in Tanzania and Zimbabwe. Sci. Total Environ. 259, 105–113.
Veiga, M., Baker, R., 2004. Protocols for Environmental and Health Assessment of
Mercury Released by Artisanal and Small-Scale Gold Miners. GEF/UNDP/UNIDO,
Vienna, Austria (Produced for the Global Mercury Project).
Veiga, M.M., Hinton, J.J., 2002. Abandoned artisanal gold mines in the brazilian
amazon: a legacy of mercury pollution. Nat. Resour. Forum 26, 15–26.
Veiga, M.M., Maxson, P.A., Hylander, L.D., 2006. Origin and consumption of mercury
in small-scale gold mining. J. Clean. Prod. 14, 436–447.
World Health Organisation, 2003. Concise International Chemical Assessment
Document 50: Elemental Mercury and Inorganic Mercury Compounds: Human
Health Aspects. (Geneva, Switzerland).
de Andrade Lima, L.R.P., Bernardez, L.A., Barbosa, L.A.D., 2008. Characterization and
treatment of artisanal gold mine tailings. J. Hazard. Mater. 150, 747–753.