THE SEARCH FOR EL DORADO: AN ASSESSMENT OF THE MERCURY POLLUTION CAUSED BY ILLEGAL OPEN-PIT MINING IN MADRE DE DIOS

(1)

T H E S E A R C H F O R E L D O R A D O :

A N A S S E S S M E N T O F T H E M E R C U RY

P O L LU T I O N C AU S E D BY I L L EG A L O P E N

-P I T M I N I N G I N M A D R E D E D I O S

Date of submission: 22nd_{May 2016} Authors: Supervisors:

Daniel Kooij J. Rothuizen

Sofie van Gessel A. J. Gilbert

Thya van den Berg

Course:

(2)

↑ figure reprinted from: “VIA PANAM: PART III – A bomb in the jungle” by Kadir van Lohuizen, from: http://blog.iamnikon.com/

ABSTRACT

Mercury is the backbone of the illegal gold mining industry in Madre de Dios and both are on the increase. Careless handling of this substance carries high risks for human health in the form of mercury and methylmercury poisoning. This research assesses, in an

interdisciplinary structure, the pathway of mercury from the moment it enters the system to the humans. Based on this chain analysis, this research presents two scenarios for the coming five years. The “Business As Usual” scenario describes a situation when no changes will be implemented and trend of extraction will continue. The “Best Case” scenario describes a situation if the most effective reduction strategies, such as better techniques and policies, are implemented. From these analyses could be concluded that the technological

improvements have the largest impact on the likeliness of reaching the best case scenario. However, in order to implement these, education about the dangers of mercury and the techniques are necessary. International programs, on the other hand, are more suitable as long term solutions.

(4)

INTRODUCTION

Worldwide, the small-scale gold mining sector accounts for one third of the global mercury pollution. Impacts are not always evident, but several studies predict that the careless handling of this substance will carry high risks for both ecosystems and human health (WHO, 2016; Yard et al., 2012; Ashe, 2012). A worrying notion is that the health risks also apply for those who are not in direct contact with mercury. Consequently, the World Health

Organization (WHO) has listed mercury as one of the top ten chemicals that form major public health threats.

In Madre de Dios, an area in the Peruvian Amazon rainforest that is rich in both biodiversity and hidden gold, an estimated 30 to 40 tons of mercury are dumped into the environment annually (Buccella, 2014). This mercury pollution stems mainly from the small-scale mining sector, which, for the largest part, consists of unregulated mining. This industry has

increased in size since the gold prices rose due to the 2008 economic crisis and the mercury contamination in areas where informal mining is prevalent has reached hazardous levels since this “gold-rush” (Ashe, 2012; Yard et al., 2012). The conflicts which occur over access to natural resources, land rights, and subsequent environmental and socio-economic impacts are creating an unstable situation, leaving Madre de Dios as a haven of violence, over-exploitation, health damages and environmental degradation (Cremers, Kolen & Theije, 2013). At first glance, these problems seem to be a specific domestic issue; however, Peru is the fifth largest gold producer in the world, and is mainly driven by global demand. An

estimated 15 to 20 % of the gold is extracted from Madre de Dios (Buccella, 2014). This implies that a substantial fraction of globally available gold comes from Madre de Dios. Considering that gold is used for many products, such as electronics, the Peruvian mercury problem may be sitting in every person’s pocket.

Up until now, the Peruvian authorities have not been able to bring the mercury pollution to a halt. In reality, these problems go beyond the political dimension of illegality. Authorities are insufficiently equipped to grasp the complex societal dynamics underlying small-scale mining communities. The small-scale miners represent a small but economically important group for the region; however, the communities differ strongly from one each other, mainly due to the time of immigration to the region (Damonte, 2013). Moreover, multiple studies show that the quantities and scope of the illegal small-scale mining activities are enlarging, resulting in increased released mercury in the environment (UNEP, 2012; Asner et al., 2013; Swenson et al., 2011). Given the attractive gold price, one may expect that the gold extraction industry will remain active in the coming years and mercury pollution will be a serious danger for the environment and human health.

Mercury is the backbone of the regional economy of Madre de Dios and mainly used to separate gold from sediment. Through the ineffective techniques used, a large amount of mercury is dumped and distributed into rivers near mining operations (UNEP, 2012). In this sense, mercury is entering the food web, forming potential danger for the inhabitants of Madre de Dios. This research will endeavour to assess the impact mercury concentrations on the human health and how this may evolve over time based on two scenarios. Therefore, the research question of this report is; What are the implications - and solutions - of the elevated mercury concentrations caused by unregulated small-scale open-pit gold mining on human health in the Madre de Dios region?

The impact of the gold mining activities can be seen as a complex and profound issue, calling for an interdisciplinary approach. Due to the socio-political, geological and ecological dynamics and interactions, it is impossible to view this problem from a monodisciplinary

(5)

angle. After all, multi- or interdisciplinary feedback loops and influences would be missed and render the conclusions redundant (Rutting et al., 2014). To capture the complexity of this problem, this research will focus on the different geological, ecological, political and social dimensions of this phenomenon. Moreover, this research will help to indicate where different disciplines overlap or influence each other and how important it is, while looking for solutions, to take these interactions into account. This interdisciplinary approach will enrich the debates on what must be done and will help to overcome a biased perspective (Rutting et al, 2014). Hence, this study will shed a new light on the problem from where all recent studies that were limited to one discipline left off.

(6)

METHODS

INTERDISCIPLINARY STRUCTURE

In the 60’s, Rachel Carson’s Silent Spring (1962) introduced a manner of description of a problem that gives a certain succession to it. She first gave a clear explanation of the mechanisms by which a pollutant enters the system, followed by its movement through the system, the pollution of biota and eventually the poisoning of man, and then described how the problem may be solved. While her research was extensive and far more time consuming than we can afford, her book provides a structure that is very suitable for an interdisciplinary description of a pollutant. After all, similarly to DDT as described by Carson, the issue of mercury from gold mining is interdisciplinary in nature. The problems, and their causes, are of various disciplines, including, but not limited to, technical, ecological, geological and political fields. All the processes, both anthropogenic and natural, interact with each other in a complex manner, making it impossible to view from a monodisciplinary angle (Rutting et al., 2014). Hence, this research follows mercury through the different stages of its life cycle in a chain analysis.

Unlike with DDT, it is not possible to simply stop the use of mercury in gold mining. For DDT, ample replacements were available and big corporations were the source of the problem, while with mercury, the replacements are limited, and not to mention equally dangerous, and the artisanal mining industry in Madre de Dios is mainly illicit, making it hard to enforce regulations. In this situation, one can only try to decrease the mercury pollution through improved education, raising awareness and implementation of better technologies. The estimations for how much mercury these improvements differ are assessed in two different scenarios. Therefore, in addition the impact description, a business-as-usual (BAU) scenario and a best-case (BC) scenario are described and finally discussed to determine how realistic they are.

GEOGRAPHICAL INFORMATION SYSTEMS

In this research, information from geological maps and remote sensing data was used in order to calculate the total mining area expansion and its related mercury emission for both scenarios over a time span of five years. The following steps were chronologically taken in order to calculate the total mercury pollution.

Firstly, the geological layers in which the gold-rich sediments are available are analysed. Several reports gave information about which layers in Madre de Dios contained gold and are referred to within the chapter itself. The geological and sedimentary layers of Madre de Dios were mapped in ArcGIS using data from the mining and metallurgical geological institute (El Instituto Geológico Minero y Metalúrgico; INGEMMET). The different geological layers were cut into polygons, in order to visualize the layers and perform calculations with these areas.

Secondly, the expansion of the mining areas over the past decade was mapped. This was done using data from Asner et al. (2013). Asner is the principal investigator of the Carnegie Landsat Analysis System-lite (CLASlite) team, a team which monitors forested areas in Madre de Dios using high resolution satellite remote sensing imagery. The mining areas, which are deforested, are clearly distinguishable from forested areas and thus usable to

(7)

map. Using ArcGIS, the CLASlite data was visualized and rasterized using unsupervised clustering tools, provided within ArcGIS.

The surface area of the separated mined patches of each year were added together in order to acquire the total excavated area of that year. Thereafter, the change in expansion of the mined area was calculated using these numbers. It was then possible to extrapolate the mining area expansion for the following five years based on the past increase. The BAU scenario mining area expansion is based on the past five years and the BC scenario mining area expansion is based on the past two years.

Using the expansion of mining area per year, the average amount of gold per square kilometre can be calculated when the area is divided by the gold production of that same year, which was found in literature. If this extent of the futuristic mining area is then combined and multiplied with the amount of gold per square kilometre, the expected amount of gold that will be mined is calculated. Since the amount of mercury that is necessary to recover an amount of gold is 2,8:1, the total mercury emission for the coming years will be 2,8 times as much as the extracted amount of gold.

The mercury which is being evaporated and discharged due to the processing techniques will precipitate into the environment. According to Álvarez (2011), the airborne mercury usually precipitates within a 1,5 kilometre radius around the furnace. Using the mapped mining areas from the previous paragraph, the area in which the mercury will precipitate was assessed by creating a buffer zone around the expected mining areas in ArcGIS. Finally, the area of the possible mercury pollution was divided by the different geological layers, which allocates the pollution to these different layers. The floodplain was assigned and generalized as being a flooded soil, the other layers are expected to only contain normal soils.

MERCURY POLLUTION AND POISONING

Following these two chapters about the causes of the mercury pollution problem in Madre de Dios, the different pathways that mercury can take to humans, and their respective health effects in Madre de Dios, are explained. It describes the manner by which the mercury reaches humans and what health effects it causes. In addition, a description that

distinguishes between a typical “high risk” and “low risk” resident is included, derived from the best predictors for elevated mercury levels in Madre de Dios.

The following things were calculated in this chapter: The current amount of people in danger of mercury- and methylmercury poisoning, and the amount of people in danger of being poisoned according to the BAU and BC scenarios. BC was divided into area induced, technology induced, and total change in order to make clear the different effects.

For the estimations of current danger, the results of the papers by Ashe (2012; high estimate; from hair concentration) and Yard et al. (2012; Low estimate; from urine concentration) were multiplied by the current estimated mining population and the current total population of Madre de Dios. This resulted in a total of four estimates of the amount of people that currently have concentrations that exceed safety guidelines. For the methylmercury, only data from Yard et al. (2012; from blood concentration) was used, again multiplied by the mining population and the total population and resulting in two estimates. The amount of estimates is the same for the subsequent calculations.

For BAU and BC-area, the mining area expansion data from the previous chapter was used to calculate the relative increase in mining area. Then, the increase and relative increase in mining- and total population were calculated for 2021, based on data from Knoema (2013). To correct for population growth, the relative mining expansion was divided by the relative population increase. Relative mining expansion was then multiplied by the same data from Ashe (2012) and Yard et al. (2012) to make an estimation for the percentage of people in

(8)

danger, and finally multiplied by the 2021 population. For BC-technology, the same

calculation was used, with the data on relative potential decrease of mercury emission due to technological improvement from the previous chapter instead of the relative increase in mining area. For the BC-total scenario, BC-technology and BC-area were combined.

SOCIO-POLITICAL LANDSCAPE

The next chapter emphasizes on the socio-political landscape of the Madre de Dios. It describes whether, and how, the power relations and stakeholder network hinder change in the mercury pollution. Additionally, it touches upon the discussion about to which extent formalization could contribute to the reduction of mercury related problems and how the legislation concerning the small-scale gold mining might evolve in the future. The description is based on literature research and an analyses of the regional, national and international laws and regulations concerning both gold mining and mercury pollution.

DISCUSSION OF THE CHALLENGES

Finally, the knowledge of the causes and effects will be combined with ways in which the problems can be solved. Future implementation of technological solutions for a safer mining process, possible influences of new regulations, and better behaviour have been combined into a best-case-scenario, and current trends were extrapolated for a business-as-usual scenario. Both of these scenarios are made for five years into the future. This chapter contains an in depth discussion about the realization and the likeliness of both scenarios.

(9)

1- SETTING THE SCENE

GEOGRAPHICAL DESCRIPTION

Madre de Dios is a region of Peru, lying in the south-eastern part. The name, which

translates to Mother of God, is shared by the main river of which the catchment covers nearly the entire region. In the west, the region is bordered from the rest of Peru by the Andes mountain range, and inlands from there the

Amazon rainforest spreads out to the east into Brazil. Due to its windward location next to the mountain range and around the equator, the area receives an average 1673 mm of precipitation per year and temperatures are on average between the 21 and 34 degrees Celsius. Because the region consists for a large part of pristine rainforest, it is one of the biodiversity hotspots of the world, counting multiple endemic species. Yet, in the southernmost part, the area is rapidly disturbed by humans around the Interoceanic highway. According to recent investigations, the region is inhabited with around 100.000 people, from which almost ⅓ is actively involved in the small-scale mining industry (Yard et al., 2012; Ashe, 2012). The other industries mainly consist of activities concerning the extensive natural resource availability of Madre de Dios, such as fishing, poaching and agriculture.

Figure 1.1The location of Madre de Dios on the map, from: Wikimedia commons

MINING TECHNIQUES

As stated in the introduction, the mining industry in Madre de Dios is mainly small-scale and illicit. However, one should not misinterpret the word ‘small-scale’ for having a small impact on the environment and human health. Due to the large number of miners involved, the numerous small-scale mining operations cause pollution on a vast scale. Yet, small-scale does imply that the individual operations typically consist out of two to ten miners, who normally do not own large machinery.

The mining activity itself is a form of open-pit mining, which means the whole surface area where the mining activity takes place is removed to enable access to the gold-rich

sediments. The topsoil is removed by high-pressure water hoses and the underlying sediments are loosened and mixed with the water. Subsequently, a pump sucks up the muddy water and sprays it over a synthetic mat, in order to trap the gold rich sediments (Webster, 2012). The gold then is retrieved by merging these sediments with mercury, which are usually mixed by blending the sediment and mercury with their feet in a bucket. This

(10)

process forms a solid amalgam of the two metals, which can be taken out of the sediment. The residual sediment, which now also contains mercury, is usually discharged into a nearby river. Next, the amalgam is heated, often in frying pans over open flames in non-ventilated spaces. The mercury evaporates through the heating process and the gold is left behind (Gardner, 2012).

GOLD OCCURANCE IN SEDIMENTARY LAYERS

The mining areas where the gold mining takes place are not randomly spread throughout the region. The occurrence of gold within the sedimentary layers of the Madre de Dios basin is closely linked to the availability of silt and clay, because gold flakes binds strongly to these particles (Rigsby et al., 2009). The sedimentary layers in which the gold rich silt can be found, originate from the late quaternary period. During this period, primary gold-rich deposits were eroded in the Andean mountains and transported to the lower lying areas of Madre de Dios (Hemric et al., 2005). These deposits now form the youngest river terraces, which were formed during the interglacial periods. As a result, the present-day active floodplains of the Madre de Dios river also contain these sediments (Rigsby et al., 2009). However, since all sedimentary layers contain some form of irregularity (Romans, 2013), the gold-rich sedimentary layers are not uniform in thickness and do not extend homogeneously over Madre de Dios. According to Perupetro (2002) the sedimentary layers are deepest at the side of the Andean mountains and thins regionally to the northeast toward the Brazilian Shield, the side of the Andean mountains is also the place where most of the illicit gold mining occurs.

(11)

2- EXTENDING EXPLOITATION AND PROFOUND POLLUTION

This chapter aims to estimate the total mercury emission in Madre de Dios over a period of five years. In order to accurately estimate the mercury emission, the calculations were performed for a Business As Usual scenario (BAU) and for a Best Case Scenario (BC). Firstly, the mining area expansion will be assessed. Subsequently, the gold occurrence and amount in the sedimentary layers will be estimated. These two together will make it possible to calculate the expected mercury emission. Finally, the possible techniques that are readily available to the miners, which save mercury, will be elaborated on in order to give a final overview of the total expected mercury emission for both scenarios.

INCREASED POLLUTION

Year Area Area increase

Pre 2007 27.097 ha

-2007 - 2008

29.664 ha 2567 ha

2008 - 2009

32.416 ha 2752 ha

2009 - 2010

37.852 ha 5436 ha

2010 - 2011

44.832 ha 6980 ha

2011 - 2012

50.453 ha 5619 ha

2012 - 2014

55.191 ha 4738 ha

Table 2.1. Mining area expansion from 2007 till 2014.

As stated in the introduction, illegal mining activities found an upsurge in mining area expansion after the economic crisis in 2008. In the following three years, average mining area expansion was around 6000 hectares per year, which is more than double the amount of the year before the crisis. After 2012; however, the Peruvian government coordinated multiple raids in the mining areas of Madre de Dios, which lead to a decrease of mining area expansion in these areas (Reuters, 2015). Between 2012 and 2014, the mining expansion decreased to a rate of 2369 hectare per year (Finer & Novoa, 2015; Snelgrove, 2015). Because the increase in mining area is prone to significant fluctuations as caused by economic crises or police raids, it is difficult to determine the expected extent of the mining area in 2021. An estimation of the future five-year extent of mining area was made for both scenarios, which are shown in table 2.2. On average, it is expected that the total mining area extent will be around 80.000 hectares by the end of 2021.

Between 2008 and 2012, the average small-scale gold production of Madre de Dios was estimated at around 16.000 kilogrammes of gold per year (Álvarez et al., 2011). This is, according to Buccella (2014), around 15 to 20% of Peru's total gold production. However, due to the use of outdated gold processing techniques in Madre de Dios, the relative share in Peru's total mercury emission is way higher. Based on the gold production of previous years and that years mining expansion, the average amount of gold after processing is 2,67 kg per

(12)

hectare. If the expected mining area in 2021 will be fully excavated using present-day techniques, the estimated total gold production, according to our BAU scenario, will be

Year BC total mining area BAU total mining area Accumulated BCS mercury emission Accumulated BAU mercury emission 2014-2015 57.600 ha 59.700 ha 18,0 tons 33,7 tons 2015-2016 59.900 ha 64.300 ha 35,2 tons 68,1 tons 2016-2017 62.300 ha 68.900 ha 53,1 tons 102,5 tons 2018-2019 64.700 ha 73.400 ha 71,1 tons 136,1 tons 2019-2020 67.000 ha 78.000 ha 88,3 tons 170,5 tons 2020-2021 69.400 ha 82.500 ha 106,2 tons 204,2 tons 2021-2022 71.800 ha 87.100 ha 124,2 tons 238,6 tons 2016-2021 11.900 ha 22.800 ha 89,0 tons 171,5 tons

Table 2.2. Estimated amount of mining area expansion per year and total accumulated mercury emission for the next five years and the past two years. Estimation are done using the Business As Usual concept (BAU) and the Best Case Scenario (BC).

Figure 2.1. Expected mercury pollution in the southern area of Madre de Dios (shown in the shape of the region at the top left corner). The light-blue area within the light-green zone is the total expected mining area extent in 2021. The light-green area itself is the area in which the mercury is expected to precipitate. The geological layers on which the mercury precipitates are indicated with the blue and red colours underlying the polluted area. Data

(13)

from the geological layers is missing in the south-western part, because it is outside the region of Madre de Dios. However, since these mining areas are located in the Andes, it is certain that there are almost no sedimentary layers available. Information from both Asner (2013) and INGEMMET is used to produce this map.

85,2 tonnes gold over the next five years. Because gold miners in Madre de Dios use 2,8 kg of mercury in order to process one kilogram of gold, the expected total mercury emission over these five years will be 170,5 tons (table 2.2) (Álvarez et al., 2011).

Most of the gold processing in Madre de Dios occurs within the mining areas (Cordy, 2014). Since the vast majority of these miners lack the education or capital in order to own and operate pre-processing techniques or filters, the mercury, which is being evaporated during the heating of the amalgam, is mainly emitted into the air. According to Álvarez (2011), the airborne mercury usually precipitates within a 1,5 kilometre radius around the gold heating process. Figure 2.1 visualizes the total area in which the mercury can be

expected to precipitate. Because there are different reactions that take place with mercury in the different kinds of sinks, it is important to know where the mercury ends up. The impact of mercury in these different sinks will be further elaborated in the following chapter. Due to direct mercury discharges, drainage and overland flow, 45% of the total precipitated mercury is found to end up in nearby rivers (Pfeiffer & de Lacerda, 1988). The remaining amount of mercury remains within the soil it precipitated. According to the map made within this study (figure 2.1), 46% of the residual mercury lands on flooded soils, which are soils that are still being actively flooded by the rivers. The other 54% ends up on the higher river terraces and within the mountainous areas, which are rather unlikely to be flooded. The amount of

mercury which precipitated over the next five years is divided over the different sinks in table 2.3.

Sinks Allocated percentage of total mercury emission

Allocated mercury emission in tons Rivers 45 % 76,7 tons Soils 30 % 51,2 tons Flooded soils 25 % 42,6 tons

Table 2.3. Allocation of the mercury emission to the different sinks with the mercury emissions of the BAU scenario.

MERCURY REDUCTION THROUGH BETTER PROCESSING TECHNIQUES

The results of the GIS study speak for themselves, it is expected that in the following years, mining activities will endure and the continuing mercury pollution accumulates in the area. However, it is not only the scale of the mining practices that determine the extent of the mercury pollution. Small-scale miners use outdated processing techniques, which are rather inefficient and waste mercury (Telmer & Veiga, 2009). Adapting newer technologies saves mercury use and significantly cuts mercury emissions. Using the mercury emissions from the previous paragraph the amount of mercury that is not emitted with implementation of better techniques are calculated and presented below. To assess the mercury savings for different technologies, three different types of techniques are reviewed, namely: Pre-processing techniques, Retorts & fume hoods and Alternative gold-extracting substances.

(14)

IMPROVED PRE-PROCESSING

In order to retrieve the mercury-gold amalgams from the sediments, the sediments are pre-processed. A known method is to separate the gold on weight, which is a commonly used practice for the gold miners, but does not extract much gold. The extent to which the

sediments are pre-processed largely determine the amount of mercury needed to extract the gold from the sediments. This implies, among other things, increasing the surface area of the gold-rich sediments by milling the sediments beforehand and adding mercury to it. Efficient pre-processing techniques reduce the amount of mercury needed, which can save mercury up to 45% during this phase of the extracting process (Veiga, 2009). This results in saving 20,0 ton of mercury over the next five years.

RETORTS & FUME HOODS

Since 50% of the total added mercury remains in the amalgam and ends up in the heater, a large part of the mercury reduction can be achieved during the heating process. This reduction can be made possible through the use of retorts and fume hoods; these filters capture the evaporated mercury from the air when the amalgam is burned. Their

implementation is relatively easy and it requires a moderate amount of education to operate. Only an investment is needed in order to acquire the filters, however, the costs are even for the miners not high. Similar small-scale mining industries to Peru, in Brazil and Indonesia, were able to purchase these capturing systems for around 35 USD (Agrawal, 2007). In both cases, a 90% capturing of mercury was achieved, significantly lowering the mercury

emission in these areas (Telmer & Veiga, 2009). If all miners would implement the use of retorts and fume hoods, the mercury emission would be cut down with 40,1 tons and another 11,1 tons of mercury would be saved through the recycling of mercury (table 2.4).

ALTERNATIVE GOLD-EXTRACTING CHEMICAL SUBSTANCES

In addition to reducing the amount of mercury used in the different phases of the gold processing, it is also possible to avoid the use of mercury. This can be done by using

different extracting substances. The most efficient way to enhance gold extracting and avoid the use of mercury is through chemical leaching (UNEP, 2012). With this method, the

sediments are placed into a large container together with a chemical, usually cyanide, in order to extract the gold. The gold dissolves in the chemical solution and is subsequently absorbed from the liquid by adding charcoal to the solution (Veiga, 2009). This technique is cheap, since cyanide itself is not expensive, and only a container and charcoal are required. However, the use of cyanide requires adequate education. Cyanide itself is very lethal to humans and other species of life. If wastewater is not treated with care, the discharged solution containing cyanide kills all life when absorbed. Yet, cyanide does naturally break down, whereas mercury does not and accumulates into the environment.

The use of cyanide would decrease the mercury emissions to zero. However, since it is almost impossible to due to the high uncertainty of implementation of this technique and the abolishment of mercury by the small-scale mining sector, it’s potential reduction in mercury use is not included in the mercury savings calculations.

If all of the techniques mentioned above would be fully implemented, the total reduction in mercury, using the emissions of the best case scenario, would be 71,2 tons of mercury (table 2.4). This is 80 % of the total mercury emission over the next five years. Figure 2.2 visualizes the difference in mercury emission up to and including 2021 for both scenarios. The following chapter will elaborate on next succession of the chain analysis, which is the impact of the mercury emission on the human health in Madre de Dios.

(15)

Technique Saves mercury during pre-processing Saves mercury during amalgamation % of mercury saved Tons of mercury saved Improved pre-processing techniques X 45 % 20,0 tons

Retorts & fume hoods

X 90 % 40,1 tons

Reuse of

captured mercury

X 25 % 11,1 tons

Table 2.4. Percentages and tons of mercury saved over the following five years with the emissions of the Best Case scenario.

Figure 2.2. Total mercury use in the different phases of the gold processing for the Business As Usual (BAU) scenario and for the Best Case (BC) scenario. Differences in mercury emission due to the difference in mining area expansion for the two scenarios is included in the calculation. Mind that the size of the chart is not proportionate to the reduction in mercury emission.

(16)

3- EMISSION, INHALATION AND CONSUMPTION

The mercury pollution in Madre de Dios emerges from the amalgamation process as

elemental mercury vapour. From there on, there are two main pathways it can take to reach humans: The Inorganic Pathway and the Organic Pathway.

INORGANIC PATHWAY

The inorganic pathway concerns the elemental mercury vapour that miners and the mining communities get directly exposed to. Mainly the miners get exposed to the vapour. Since mercury vapour can travel far without reacting to anything; however, people who do not process the amalgamate themselves also inhale it. Ashe (2012) determined, in a case study in Madre de Dios, that gender and location of residence were two of the largest predictors of mercury concentration in residents of a mining community, likely because of these reasons, as most miners are men. As the amalgamation often takes place in the living space of an artisanal miner, their families are often more exposed that other people in the communities (WHO, 2006; Gardner, 2012; Yard et al., 2012). Exposure of food and drink which is contaminated with elemental mercury is far less of a problem, as only 0.01% is taken up by the gastrointestinal tract when eaten, compared to the 80% that can be taken up by the lungs (Park & Zheng, 2012). Moreover, less than 1% of the elemental mercury is taken up by biota worldwide (King et al., 2000) making it barely bioavailable to humans apart from the direct pathway. An additional danger in Madre de Dios, is that local residents are not often aware of the toxic qualities of mercury. In a survey done by Yard et al. (2012), only 27% of the 103 surveyed residents were aware.

Population increase

year Mining pop. MdD Total pop. MdD Increase (%, from Knoema, 2013)

2012 109555 30000 Base - 0 2013 112338 30762 2.54 2014 114862 31453 2.247 2015 117607 32205 2.39 2016 120336 32952 2.32 2017 123055 33697 2.26 2018 125775 34442 2.21 2019 128479 35182 2.15 2020 131177 35921 2.1 2021 133866 36657 2.05

Table 3.1: population increase based on the initial population from the 2007 population census by the Peruvian government (IMEI, 2007) and the relative increase in population per year from Knoema, 2013

(17)

Poisoning by elemental mercury vapour can take place in two manners, acute and long term. The acute form manifests within a day after very high intake, and shows mostly pulmonary symptoms (Lilis, Miller & Lerman, 1985; Levin, Jacobs & Pollos, 1988). In both the case studies done by Ashe (2012) and by Yard et al. (2012), hair and urine samples showed that intake in mining communities is unlikely to be high enough for acute mercury poisoning. The long term poisoning, however, is indicated as the most often occurring form of mercury poisoning (Yard et al., 2012). This form of poisoning is also known as the “Mad Hatter's Disease”, after the hat makers in London, who were exposed to mercury vapour from their felting. Symptoms include tremors, headaches and an infected mouth, but more importantly Erethism mercurialis, which manifests as a behavioural change that includes paranoia, depression and irritability. In the study by Ashe (2012), 11 of the 100 surveyed residents (11%) of a mining community had dangerous levels of mercury in their hair and 1 person had toxicologically symptomatic levels, according to the safety levels designated by the World Health Organisation. The study by Yard et al. (2012), who investigated urine samples, showed that eight of the 103 residents (7.77%) had urine levels above the American

Conference of Governmental Industrial Hygienists’ recommendation for remediation. Based on this data, 2330 to 12051 people could have mercury concentrations above safe levels (table 3.2). As Yard et al. (2012) made a distinction between mercury and methylmercury, their prediction is lower and likely more accurate in indicating elemental mercury as a source. Further calculation, as depicted in table 3.2, shows that, as the mining area expands for the BAU scenario, the amount of people who could be in danger nearly doubles by 2021. On the other hand, the BC scenario shows a decrease to only about a quarter of the current amount of people in danger, even though the population increases by 24300 people for the total population, and by 6600 people for only the mining population. The table shows a clear distinction between the BC area and BC technology changes, with a slight increase and a large decrease respectively.

Population in danger- mercury

Mining population exposed Total population exposed

High estimate Low estimate High estimate Low estimate

Current 3300 2330 12051 8509

BAU 5697 4024 20804 14695

BC-area 4696 3317 17150 12114

BC-tech 779 550 2843 2008

BC-total 907 640 3311 2339

Table 3.2: Potential population with unsafe concentrations of elemental mercury in their system, estimated for a situation in which only the mining population is exposed and a situation in which the entire population is exposed, with a high estimate (based on Ashe, 2012) and a low estimate (based on Yard et al., 2012). The current scenario population is based on a 2007 census by the Peruvian government (IMEI, 2007), and the future scenarios were multiplied with population growth trends from Knoema (2013). The future scenarios are business as usual (BAU), the best case scenario (BC) incorporating only the change in area, BC incorporating only the change in

technology and the total BC. (See appendix 2-6 for more information on the calculations)

(18)

Methylmercury is more toxic than elemental mercury, and more difficult to follow as it both bioaccumulates and -magnifies (see appendix). Conversion from elemental mercury to methylmercury is called methylation and can take place in various wet systems (table 3.3). It is most often a by-product of respiration by microorganisms, of which approximately 50% is caused by sulphate-reducing bacteria (Guimaraes et al., 1998). Contrary to elemental mercury, 90 to 99% is accumulated in biota (King et al., 2000). The most important in this uptake is the aquatic food web, as methylmercury is often formed in an aquatic environment, and aquatic organisms are able to take it up via their skin and consequently pass it on to their predators (Bakir et al., 1973). Kwon et al. (2012) investigated bioaccumulation of mercury in a neotropical stream in Venezuela. Over the trophic levels, (piscivore, generalized carnivore, omnivore, invertivore, algivore, terrestrial herbivores and detritivore), total mercury clearly increased, with only the highest trophic levels crossing the limit that is unsafe for consumption according to WHO guidelines. In addition, Mason & Benoit (1986, p.80) note the following:” ...the fraction of body burden Hg as MMHg is a function of trophic, not

taxonomic position”, meaning that as the total mercury increases, the proportion that is

methylmercury also increases. Piscivorous fish take up about 10-15% of the methylmercury in their prey, and about 60% of the methylmercury in their bodies originates from lower trophic levels. For bottom feeders, however, direct uptake is more important, and only about 25% originates from the food chain. (Jernelöv & Lann, 1971). The typical biomagnification factor for methylmercury is 3 to 5 (Mason & Benoit, 1986). Considering these numbers, it is clear that the amount of fish and which fish are eaten are both factors in determining the risk for poisoning.

Neotropical methylation rates

methylation space

% of total Hg converted

source

water column undetectable Guimaraes et al., 1998

soil 1 Roulet & Lucotte, 1995

flooded soil 5-20 Roulet & Lucotte, 1995

sediment 0.4-1.2 Guimaraes et al., 2000 & Guimaraes et al., 1998

macrophyte mats 6.5-13.8 Guimaraes et al., 2000 & Guimaraes et al., 1998

Table 3.3: Methylation factors for different spaces in a Neotropical environment like Madre de Dios

Methylmercury poisoning is also known as the Minamata disease, after an incident in Japan between 1953 and 1960. Contrary to elemental mercury poisoning, the onset of symptoms caused by methylmercury take up to 60 days after ingesting a high enough dose, and the dose needs more time to build up in an organism. The symptoms, similar to chronic elemental mercury poisoning, are mostly neurological. They include loss of feeling in the limbs, loss of coordination and diminished vision. Severe poisoning may result in blindness, coma, or even death and, as with elemental mercury, methylmercury is capable of crossing the placental barrier and cause neurological defects in foetuses (Bakir et al., 1973). Different governmental and non-governmental agencies have implemented limits of around 0.5 ppm for the market (Ibid.; WHO, 2016), however, as fish in small communities like the mining communities in the Madre de Dios are often caught locally, these guidelines are difficult to follow. Ashe (2012) indicated that the intake of fish of the residents of the researched

community was low, but in the research done by Yard et al. (2012) it was found that 10.7% of the participants had a blood methylmercury of above the EPA (US Environmental Protection

(19)

Agency)’s safety limit. Based on these percentages, the population that would be at risk of methylmercury poisoning is between 3200 and 11700 (table 3.4). Similar to the elemental mercury scenarios, the BAU sees the amount of people above health regulation limits nearly double in size and the BC decrease to a bit more than a quarter. Again the implementation of safer technology appears to have the largest influence in making gold mining safer.

However, one thing that is not included in the calculations is the adaptation of the diet of local residents. While their fish consumption is considered low (Ashe, 2012), the effects are still clearly visible from the calculations in table 3.4, and raising awareness on which aquatic animals are safer to eat, namely the ones of a lower trophic level, as well as decreasing the overall consumption of aquatic animals could be of significant influence in decreasing the amount of people above the health regulation limits, especially in the case of unsuccessful implementation of technology.

Population in danger- methylmercury

Mining population exposed Total population exposed

Current 3210 11722

BAU 5542 20237

BC area 4568 16682

BC tech 758 2765

BC total 882 3221

Table 3.4: Potential population with unsafe concentrations of elemental mercury in their system, estimated for a situation in which only the mining population is exposed and a situation in which the entire population is exposed, based on percentages by Yard et al. (2012). The current scenario population is based on a 2007 census by the Peruvian government (IMEI, 2007), and the future scenarios were multiplied with population growth trends from Knoema (2013). The future scenarios are business as usual (BAU), the best case scenario (BC) incorporating only the change in area, BC incorporating only the change in technology and the total BC. (See appendix 2-6 for more information on the calculations)

RISK IN HUMANS

As indicated in the above two paragraphs, gender, (manner of) fish consumption and location of residence are the biggest predictors of mercury related health risks in Madre de Dios. This would indicate that a male, predatory fish consuming miner is at the highest risk, while a female, who has a diet that does not contain aquatic species and lives away from the gold mining site is at the lowest risk. This is not entirely true, as gender as a predictor is largely dependent on the fact that mining is a male dominated sector, and therefore occupation would make for a better predictor. Inhabitants with (methyl-)mercury intake, as described in figure 3.1 A and B will have the highest and lowest risk of poisoning, respectively. Inhalation was divided into inhalation by use and inhalation by residence, as the miner would have both an input caused by himself and by the mining community around him.

(20)

Figure 3.1: The difference between the intake of (methyl-)mercury for a high risk resident(A) and a low risk resident (B) of Madre de Dios.

(21)

4- DIFFICULTY OF SMALL-SCALE MINING POLITICS

Madre de Dios has been a scene for a continuous battle between small-scale miners and the authorities regarding the use of the land. In 2002 the government started a program to formalize the small-scale mining sector. Despite the intensive negotiations between the large number of stakeholders, the process remains in a deadlock (Damonte, 2013). The main reason for this deadlock stems from the complex stakeholder involvement and an internal division within the Peruvian national politics.

STAKEHOLDER INVOLVEMENT

Since the colonial era, the mineral extraction industry has been of great economic

importance. Due to the global demand, the extraction sector now accounts for over 60% of the total export of Peru (EY, 2014). As of now, the Peruvian government is trapped between a large-scale extraction agenda, which prioritizes large-scale over small-scale activities and is one of the major national revenues, and the conservationist environmental agenda that aims to reduce the environmental impact by controlling both mining activities. As a result of this division, the government has a lack of national prioritization and exhibits inconsistent attitude and irregular repressive action towards the small-scale miners. This behaviour has caused an upsurge of protests and social unrest in Madre de Dios (Oxfam, 2009). As a result of the instability, the area has gained political significance and extensive media attention. In addition, the international community has increased their pressure to formalize the small-scale mining activities.

Despite the growing attention for this problem, the national benefits of formalizing the sector are relatively small because the illegal small-scale mining operations do not significantly contribute to the Peru’s GDP (Damonte, 2013). Moreover, since 2002, the Peruvian political system has been decentralized, which has strongly influenced the mining sector. Javier Arellano-Yanguas (2008) studied the changes in the mining sector stemming from the decentralization process and stated that it caused a shift in political and legal power from the national level to the sub-national and local level. This shift resulted in a strong emergence of the local level as the crucial political area where stakeholders interact; and the deep

involvement of “new” actors in the conflict such as NGO’s and miner federations. The changes had several, mostly unintentional, negative effects for the miner sector.

First of all, the growing importance of the sub-national and local level is causing complex problems for the area. In theory, decentralization could contribute to finding more space specific solution; however, the national government has not been able to provide enough resources, leaving the local institution poorly equipped to perform their new responsibilities. Secondly, the involvement of new actors in the conflict is resulting in a complex actor network. The variety of actors involved, including national and regional authorities, associations of small-scale miners, large scale miners and native communities, all have competing interests. The conflict, which occurs over the access to the natural resources, land rights, and the environmental and social-economic impact, are creating an unstable

economic and living environment, leaving Madre de Dios a violent, over-exploited, health damaging and environmentally degraded region (Cremers, Kolen & Theije, 2013).

(22)

In addition to the negative effects of decentralization, the regulatory framework has not been able to put the excessive mercury pollution to a halt. The mercury pollution is one of the most threatening problems associated with the production of gold and causes both short and long-term health risks. Though the health effects of mercury are well known, it remains difficult to propose a set of rules which will contribute to reduce the pollution. The complexity of the problem makes finding a comprehensive solution for this problem very difficult. However, due to the increased pressure of both the international community and the Peruvian

Environmental Department, there are several attempts to find a synthesis between all the competing interests.

BALANCING TWO POLITICAL NATIONAL AGENDAS

In 2002, the Peruvian state set the formalization process in motion by implementing Law 27651. This law was the first in a series of interventions to get the illegal mining sector under control. It obligates all miners to have a tax number, mining certificate and a mining permit. All small-scale activities also need to carry out an Environmental Impact Assessment (EIA) and based on this assessment submit an Environmental Management and Closure Plan (EMCP) (Damonte, 2013). The Peruvian government has also promoted the formalization process by implementing a series of environmental regulations. In 2009, the Economic Zoning Plan (EZZ) was implemented. This regulation prescribed mining and non-mining zones in Madre de Dios (figure 4.1).

Figure 4.1: The geographic description of the Economic Zoning Plan implemented in 2009. Reprinted from “Peru Battles the golden curse of the Madre de Dios”, by E. Garder, 2012.

Two years later the Peruvian government implemented a more far-reaching legislation. Building upon the EZZ, several “main mining areas” were indicated. The “Plan Nacional de Formalización de la Minería Artesanal” made it possible to ban all the mining activities outside these identified areas (figure 4.2). However, the mining activities outside the areas could not yet be labelled as “illegal”. Only in 2012 was it made possible to declare all mining activities outside of these prescribed areas as illegal (Damonte, 2013).

(23)

Figure 4.2: The geographic description of the mining areas prescribed in Plan Nacional de Formalización de la

Minería Artesanal. Reprinted from “Peru Battles the golden curse of the Madre de Dios”, by E. Gardner, 2012. Despite all the legal attempts to scale down and formalize the mining activities in Madre de Dios, the process has been anything but successful. Several reasons could be identified as possible obstacles to the formalization process. Firstly, the clash of two political agendas resulting in conflicting interests within the Peruvian national government. For years, the (international) environmental community exerted pressure concerning the environmental and health consequences of small-scale gold mining activities. This has led to the development of a Peruvian political conservation agenda to protect the Amazonian environment and its indigenous people. Beside the above-mentioned creation of special mining and non-mining zones, the government also needed to start recognizing the land rights of indigenous

communities by granting communal titles (Damonte, 2013). Secondly, beside the continuous pressure from the (international) environmental community, the conservation agenda is overshadowed by a historically-grounded, large-scale extraction agenda (Damonte, 2008). This agenda highlights the economic opportunities of the resource-rich area. The economic crisis of 2008 has led to a rapid increase of the world’s gold prices. This increase has caused a so-called “Gold Rush” within Madre de Dios and many people migrated to the area.

Moreover, it has provided the national with an additional incentive to prioritize large-scale gold mining extraction programs because it became more economically feasible to exploit the gold. This increased the fear of indigenous people of losing their land to the bigger mining companies. Because many miners and their families operate outside the legal framework, they still do not have any property rights of the land they may have lived on for several generations. Thus, on the one hand the Peruvian government wants to please the international community by showing their commitment to protect the Amazon and its

indigenous communities. On the other hand, the rising global gold prices form an economic incentive for the government to put more pressure on the formalization program so they can expand the large-scale gold mining activities in Madre de Dios.

The clash of the two national political agendas and the extra pressure from the increased number of small-scale gold mining activities have resulted in a poor implementation of current obligations by the regional authorities. This has made those authorities a subject of corruption (ELLA, 2012). They also struggle to negotiate with the local miners, as they are a heterogeneous group with a wide range of cultural backgrounds, socio-economic positions, land property rights and working relations. This lack of communication between the regional authorities and the miners can be seen as another important obstacle that hinders any change.

(24)

REGULATION OF MERCURY USE

One of the questions that remains is to what extent the formalizing of the sector will contribute to the reduction of mercury pollution. Despite the absence of legalization of mercury use, formalizing the small-scale mining sector will to some extent contribute to the reduction of mercury pollution. Because the majority of miners operates outside the legal framework, they are not able to receive government assistance or credit from legitimate sources. This means that miners are in a position where there are unwilling and unable to invest in their business because the lack of resources or the absence of property rights. The lack of formal right causes the miners to have no incentive to comply with environmental legislation, as they do not receive any of the benefits or “encourage investments” for mercury-free mining. Formalizing the small-scale mining sector will not only provide the miners capital and resources, it will also help create an incentive for the small-scale miners to invest in their business. Thus, through promoting new techniques and encouragement investments, the Peruvian authorities could reduce the utilization of mercury in the recovery and refinement of gold.

(25)

5- CHALLENGES AND SOLUTIONS

This chapter is a discussion of the different solutions and their impacts. From this chapter will be concluded which of the solutions should have the most of the focus in order to reach the best case scenario.

CHALLENGE 1: IMPROVING THE COMMUNICATION AND RAISING AWARENESS

To increase the likelihood to reach the BC scenario, an important point of action is to improve the communication and raising awareness among the miners. As previously-mentioned, the upsurge of conflicts and increase of social unrest in the Madre de Dios has drawn the attention of the international community. Despite their lack of legal power, NGOs play a critical part in the international response. NGOs can offer many unique skills and approaches beyond the legal framework, like humanitarian relief, preventive action, conflict resolution, development assistance and institution building (Aall, 1996).

In the case of the Madre de Dios, only 27% of the miners have actual knowledge on the poisoning dangers of mercury (Yard et al., 2012). Consequently, this is an area where much can be gained. NGO’s can be seen as important links between miners and national and local government. Due to the inconsistent and sometimes violent behaviour of the government, many miners and miner federations are very hesitant to cooperate with government

institution. Improving the communication between legal institution, educational institution and NGO’s might help to create awareness among the miners. Education is also the manner of achieving the diet changes as described in chapter 4. Reaching a significant amount of the population and actually decreasing the amount of fish in the diet as well as educating them about the dangers of mercury is, although it is the least tedious of the challenges, unlikely to show improvement over 5 years and decreasing the actual emission is far more likely to be a significant improvement of the human health.

CHALLENGE 2: CREATING A RELIABLE INTERNATIONAL PLATFORM

In October 2013, 140 countries, including Peru, have adopted the Minamata Convention on Mercury. The convention addresses the harmful impacts of mercury by placing limitations on its use, production, and trade (UNEP, 2013). The Peruvian ratification took place on the 21th of January in 2016. Because the ratification is such a recent event it is not clear yet to which extent the treaty could curtail the massive amounts of mercury pollution in Peru. The majority of the convention does not apply for the use of mercury in the Peruvian gold mining industry it might be a great starting point to enforce change.

The Minamata Treaty encourages nations to reduce or phase out its use of mercury, but no targets or dates are included. Each country that has signed the Minamata Convention, and has a more than “insignificant” artisanal and small-scale gold mining activity is obligated to submit a “national action plan” that presents and evaluates the baseline of the mercury use and practices and the plans and strategies to reduce the use of mercury. The first national action plan must be submitted no later than three years after the ratification and the progress must be reviewed every three years (UNEP, 2013). Although the treaty does not legally ban the utilization of mercury by the production of gold, it will definitely contribute to identifying the scope and significance of the problem in the Madre de Dios.

(26)

In this case, the challenge of the international community is to held Peru accountable if no progress has been made by emphasizing Peru’s international responsibility and increasing the international pressure. However, the Treaty does not give explicit financial support for job creation or workforce education programs. It does not offer guidance on, or stipulate

requirements for, offering incentives to those who rely on illegal gold mining to leave the industry (Buccella, 2014). This limitation decreases the likelihood that the treaty will make a significant difference for the mercury use in the Peruvian gold sector.

From an international perspective, one might make a significant difference by changing the global demand for gold. The global economy could have a positive effect on mercury pollution in Peru Buying fair trade gold, spending resources on researching and developing non-toxic mercury alternatives, or creating technological innovations to the gold mining process could create a solution that does not have such drastic negative consequences (Buccella, 2014). While this solution might not be fully encompassed by the Minamata Convention, the convention is at least a starting point that has given Peru the international platform it needed to enforce national change.

CHALLENGE 3: TECHNOLOGICAL IMPROVEMENTS

In order to reduce the mercury pollution, the adaptation of processing technologies and the use of mercury have to be assessed. In Madre de Dios, gold processing technologies are outdated and miners lack education to understand and acknowledge health risks caused by careless mercury handling (Cordy, 2013). If mercury pollution has to be cut down, other actors, such as the state and NGO’s, have to educate the miners on new processing technologies and techniques.

One of these techniques is the use of pre-processing. This can be relatively easy, but might be time-consuming. Yet, they save money on the mercury, and with the mercury prices going up, there is a strong economic incentive for this practice.

Another technology is the use of filters to capture and recycle the mercury and to cut down on mercury vapour emissions. Some retorts are able to capture 90% of the used mercury, reducing mercury emissions significantly. In addition, similar small-scale miners in mining areas for instance Indonesia were able to buy these filters for only around 35 dollars. However, the drawbacks are that small-scale miners are anxious to make investments because of inconsistent behaviour of regional authorities (Damonte, 2013). Subsequently, the filters are not easy to come by and are usually provided by NGO’s (UNEP, 2012). Next to pre-processing and capturing of mercury, it is also possible to completely avoid the use of mercury. In other parts of Peru, this technique is already used amongst gold miners (Cordy, 2013). The benefits of using cyanide is that it recovers more gold and the additional effort needed is not much more than using mercury. Thus, it provides a good economic incentive to adapt the use of cyanidation (Ibid.). However, the use of cyanide is also lethal to living organisms, thus requiring education in order to use it safely. If all of the techniques would be implemented widespread, it is expected that the total mercury emission would reduce with 80%.

(27)

CONCLUSION

The gold mining industry has expanded significantly over the last few years in Madre de Dios. Based on the amount of gold still available in the sediments and the increasing gold price, this “gold rush” is unlikely to stop. National political advancement is in a deadlock, stuck between the pressure of the international community to preserve the rainforest and the economic benefits of formalising the gold mining sector. Thus, the community currently mining in Madre de Dios employs dangerous artisanal mining techniques which exposes both the population and the environment to large amounts of mercury. Between 2300 and 12100 people may have unsafe concentrations of elemental mercury in their bodies from inhaling mercury vapour, and between 3200 and 11700 may have unsafe concentrations of methylmercury, which is ingested via contaminated fish. If no measurements are taken, this number could increase tremendously up to 20800 and 20000 potential people in danger by 2021, for mercury and methylmercury, respectively. In terms of effect on the population, this means that these people- and their unborn children- are in danger of neurological disorders that could severely harm their social and working condition.

The 2016 Minamata treaty provides an international pressure that could improve the situation, but the fact that the treaty gives only limited guidance and the first national action plan has to be done only by 2019, mean that this effect will likely be long term and irrelevant for our five-year prognosis. The solution that shows the biggest measurable impact, is improvement of the technology used by the artisanal miners. The Best Case scenario

predicts that implementing techniques could bring down the mercury contamination from 238 tons to only 29 tons each year, decreasing the amount of people in danger to at most 3300 and 3200 people for mercury and methylmercury, respectively. While this change is large, the miners themselves are anxious to invest, as they are largely unaware of the dangers of mercury and could be apprehended by regional authorities at any time. Therefore, it is important that not only an international platform such as the Minamata treaty is created, but also a local platform. Education about the dangers of mercury, the avoiding of mercury contamination and the technical improvements to be implemented could help bring about the actual technical change.

Although many of the solutions are easy to implement or even already in motion, the Best Case scenario is very optimistic in the sense that it assumes all calculable changes are implemented within five years. However, since action is being undertaken, it is also unlikely that the Business As Usual scenario will be the outcome. It is therefore most likely that the situation in Madre de Dios, in five years, will be somewhere between these two scenarios.

(28)

REFERENCES

Aall, P. (1996). Nongovernmental organizations and peacemaking. Managing global chaos: Sources of conflict of and responses to international conflict, 433-443.

Arellano-Yangas. J. (2008). A thoroughly modern resource curse? The new resource agenda and mining revival in Peru. IDS Working Paper 300, Brighton: IDS

Ashe, K. (2012). Elevated mercury concentrations in humans of Madre de Dios, Peru. PloS one, 7(3), e33305.

Asner, G. P., Llactayo, W., Tupayachi, R. & Ráez-Luna, E. (2013). Elevated rates of gold mining in the Amazon revealed through high-resolution monitoring. Proceedings. 110, 18454-18459.

Bakir, F., Damluji, S. F., Amin-Zaki, L., Murtadha, M., Khalidi, A., Al-Rawi, N. Y., ... & Doherty, R. A. (1973). Methylmercury poisoning in Iraq. Science,181(4096), 230-241.

Bhattacharyya, S., Chaudhuri, P., Basu, S., & Santra, S. C. (2015). Bioaccumulation of Mercury and Assessment of Suitable Biomarker in Vegetables Collected from East Calcutta Garbage Farming Area (Dhapa), India. Int. J. Curr. Res. Biosci. Plant Biol, 2(5), 148-154.

Brooks, William E, Esteban Sandoval, Miguel A Yepez, and Howell Howard. (2007). Peru mercury inventory 2006: US Geological Survey.

Cremers, L., J. Kolen and M. de Theije. (2013). Small-scale Gold Mining in the Amazon, CEDLA, 26, 9-99. Retrieved from http://www.cedla.uva.nl/50_publications/pdf/cuadernos/cuad26.pdf

Damonte, G. (2013). Formalizing the Unknown [online article]. Accessed on: 20 April 2016, Retrieved from http://www.thebrokeronline.eu

Damonte, G. H. (2008). The constitution of political actors. Vdm Verlag Dr Mueller EK. Damonte. G. D. (2013) Formalizing the unknown. The broker online. Retrieved from http://www.thebrokeronline.eu

Evidence and Lessons from Latin America (ELLA). (2012) Small-scale and Informal mining: A Big Problem for Latin American States [policy brief].Accessed on 20 April 2016, Retrieved from http://r4d.dfid.gov.uk/

Gardner, E. (2012). Peru Battles the golden curse of the Madre de Dios. Nature, (486), p 306-307.

Guimaraes, J. R. D., Meili, M., Hylander, L. D., e Silva, E. D. C., Roulet, M., Mauro, J. B. N., & de Lemos, R. A. (2000). Mercury net methylation in five tropical flood plain regions of Brazil: high in the root zone of floating macrophyte mats but low in surface sediments and flooded soils. Science of the Total Environment, 261(1), 99-107.

Guimaraes, J. R., Meili, M., Malm, O., & de Souza Brito, E. M. (1998). Hg methylation in sediments and floating meadows of a tropical lake in the Pantanal floodplain, Brazil. Science of the total environment, 213(1), 165-175.

Hemric, E. M., Rigsby, C. & Baker, P. A. (2005). Sedimentology and geomorphology of the Rio Madre de Dios from Diamante to Puerto Maldonado, Amazon basin, Peru. Department of Geology, East Carolina University, Greenville.

Instituto Geológico, minero y metalúrgico, INGEMMET. Mapa geólogico. Accessed on april 8, 2016 from: http://www.ingemmet.gob.pe/mapa-geologico-100-000-boletines-y-franjas'

Jernelöv, A., & Lann, H. (1971). Mercury accumulation in food chains. Oikos, 22(3), 403-406. King, J. K., Kostka, J. E., Frischer, M. E., & Saunders, F. M. (2000). Sulfate-reducing bacteria methylate mercury at variable rates in pure culture and in marine sediments. Applied and

Environmental Microbiology, 66(6), 2430-2437.

Knoema (2013), Madre de Dios - Annual Average Growth Rate of Population, accessed on: 21-05-2016, from: https://knoema.com

Kwon, S. Y., McIntyre, P. B., Flecker, A. S., & Campbell, L. M. (2012). Mercury biomagnification in the food web of a neotropical stream. Science of the Total Environment, 417, 92-97.

(29)

Levin, M., Jacobs, J., & Polos, P. G. (1988). Acute mercury poisoning and mercurial pneumonitis from gold ore purification. CHEST Journal, 94(3), 554-556.

Lilis, R., Miller, A., & Lerman, Y. (1985). Acute mercury poisoning with severe chronic pulmonary manifestations. CHEST Journal, 88(2), 306-309.

Mason R.P. & Benoit J.M., (1986). Organomercury compounds in the environment. In: Multiple authors, (2003), Organometallic compounds in the environment, publisher: Wiley, edited by: Craig, P.J., second edition, first edition 1986.

Moreno-Jiménez, E., Esteban, E., Carpena-Ruiz, R. O., & Peñalosa, J. M. (2009). Arsenic-and mercury-induced phytotoxicity in the Mediterranean shrubs Pistacia lentiscus and Tamarix gallica grown in hydroponic culture. Ecotoxicology and Environmental Safety, 72(6), 1781-1789.

O'Carroll, R. E., Masterton, G., Dougall, N., Ebmeier, K. P., & Goodwin, G. M. (1995). The neuropsychiatric sequelae of mercury poisoning. The Mad Hatter's disease revisited. The British Journal of Psychiatry, 167(1), 95-98.

Oxfam America. (2009). Mining conflicts in Peru condition critical [policy brief].Accessed on: 19 April 2016, Retrieved from http://www.oxfamamerica.org

Park, J. D., & Zheng, W. (2012). Human exposure and health effects of inorganic and elemental mercury. Journal of Preventive Medicine and Public Health, 45(6), 344-352.

Rigsby, C., Hemric, E. M. & Baker, P. A. (2009). Late Quaternary Paleohydrology of the Madre de Dios River, southwestern Amazon Basin, Peru. Geomorphology. 113, 158–172.

Romans, B. (2013). Strata are not flat. In: 52 things you should know about geology. Edited by Matt Hall.

Roulet, M., & Lucotte, M. (1995). Geochemistry of mercury in pristine and flooded ferralitic soils of a tropical rain forest in French Guiana, South America. Mercury as a Global Pollutant, 80(1), 1079-1088.

Rutting, L., Post, G., Roo de, M., Blad, S., Greed de, L. (2014) An introduction to interdisciplinary research. Institute of Interdisciplinary Studies. University of Amsterdam.

Swenson, J. J., Carter, C. E., Domec, J. C., & Delgado, C. I. (2011). Gold mining in the Peruvian Amazon: global prices, deforestation, and mercury imports. PloS one, 6(4), e18875.

United Nation Environmental Program (UNEP). (2013) Text of the Minamata Convention on mercury for adoption by the conference of plenipotentiaries [Report]. Accessed on: 18 April 2016, Retrieved from: http://www.unep.org/

United Nations Environment Programme [UNEP]. (2012). Analysing of formalization approaches in the artisanal and small-scale gold mining sector based on experience in Ecuador, Mongolia, Peru, Tanzania and Uganda: Peru Case Study. Division of Technology, Industry and Economics (DTIE) Chemicals Branch. Geneva, Switzerland.

Veiga, M. M., Nunes, D., Klein, B., Shandro, J. A., Velasquez, P., & Sousa, R. N. (2009). Mill leaching: a viable substitute for mercury amalgamation in the artisanal gold mining sector? Journal of Cleaner Production, vol. 17, 1373-1381.

World Gold Council (2016). Historical prices. Accessed on 18 April 2016 from: https://www.gold.org/statistics#group1

World Health Organization [WHO], (2016). Mercury and Health, accessed on: 02-03-16, from: http://www.who.int/

Yard, E. E., Horton, J., Schier, J. G., Caldwell, K., Sanchez, C., Lewis, L., & Gastaňaga, C. (2012). Mercury exposure among artisanal gold miners in Madre de Dios, Peru: a cross-sectional study. Journal of Medical Toxicology, 8(4), 441-448.

THE SEARCH FOR EL DORADO: AN ASSESSMENT OF THE MERCURY POLLUTION CAUSED BY ILLEGAL OPEN-PIT MINING IN MADRE DE DIOS