Rapidly increasing mining in rivers across the global tropics has major, interrelated consequences for ecosystems and human health. River mining involves intensive excavation and sediment processing in river corridors, altering river form and releasing excess sediment to river waters. Contaminants such as mercury and cyanide are used in some operations and are also released to the environment. Although river mining has been investigated at local and regional scales, no global synthesis of its physical footprint and impacts on hydrologic systems exists, so its full environmental consequences are not known. We assemble and analyze a 37 yr satellite database showing that river mining is pervasive globally. We identify 381 mining districts in 49 countries, concentrated in tropical waterways that are almost universally degraded. Of 173 mining-affected rivers, 80% have suspended sediment concentrations double pre-mining levels. In countries where mining affects large (>50 m wide) rivers, 23 ± 19% of large river length is altered by mining-derived sediment, a globe-spanning effect representing 5–7% of all large tropical river reaches. Mining intensity has rapidly increased during 21st-century global financial insecurity and rising demand for precious minerals. The ubiquity and intensity of mining degradation in tropical river systems is a global crisis.
Physical Sciences - Article
The global crisis of mining in tropical rivers
https://doi.org/10.21203/rs.3.rs-1875606/v1
This work is licensed under a CC BY 4.0 License
Journal Publication
published 23 Aug, 2023
Version 1
posted
You are reading this latest preprint version
Mining in rivers has been a common method for the extraction of gold, silver, and other precious minerals for centuries1,2. We define such river mining, often called alluvial mining, as operations that excavate river or floodplain sediments or soils that have a direct surface-water connection to streams or rivers. Particularly appealing to miners lacking significant capital or access to land, artisanal and small-scale river mining continues to be an attractive wealth-generating and poverty-alleviating occupation for artisanal and small-scale operators3,4. Major corporations also engage in river mining in many countries, and minerals derived from river mining of all scales are major contributors to global markets. However, river mining also results in deforestation and increased erosion. The surface water connections between river mines and important water resources often transmit considerable negative environmental and health outcomes through river networks4. Despite localized efforts to reduce environmental impacts5, we find that a global mining boom has occurred in the past two decades with only limited mitigation of environmental consequences. The widespread release of sediment from river mines is a major contributor to documented changes to river suspended sediment transport across the tropics6, part of a global freshwater crisis7.
Several modes of environmental degradation result from river mining directly, including deforestation and forest degradation8-11, threats to human health12-14, and major alterations to water quality, availability, and presence on landscapes15-18. However, environmental monitoring is hindered by the vast global reach of river mining, difficulties in accessing affected regions, and a lack of historical reference data. The global extent, intensity, and chronology of mining-associated environmental impacts on global rivers are thus largely unknown. To address these challenges and standardize monitoring efforts, we developed a suite of satellite remote-sensing tools that document the global environmental expression of river mining. The resulting 37-year global record comprises more than 7 million measurements of 173 mining-affected rivers, revealing the increasing threats to these vital aquatic ecosystems.
Environmental effects of river mining
We find that river mining is widespread globally, with 381 discrete mining areas active in the Landsat 5 and 7 satellite record (1984–present) (Fig. 1). Despite historical presence at all latitudes, such mining now predominantly occurs within 20° of the Equator, in a belt spanning South America, Africa, Asia, and Oceania. More than 80% of active sites globally are in this zone (n = 319 of 381), distributed relatively evenly across the major land masses. At the country-level, Indonesia (n = 51), Myanmar (n = 36), the Democratic Republic of the Congo (n = 29), and Colombia (n = 24) have the most individual mining areas. In total, we document river mining areas in 49 countries (Fig. 1, Extended Data Fig. 2), affecting 173 rivers wide enough (~60 m or greater) to identify and sample using Landsat satellite imagery (30 m resolution). These rivers flow through nearly one-third of countries in the Tropics that have rivers 60 m wide (31%, 27/86 countries).
At most sites, river suspended sediment concentration (SSC) increased rapidly and significantly immediately following the onset of mining (Extended Data Fig. 3). Sediment transport is a natural process in rivers and is critical for the physical maintenance of river form and ecological habitat. However, high SSC, which gives rivers an appearance of cloudiness or muddiness, decreases water clarity, can be a vector for contaminant transport, and may lead to detrimental siltation downstream in rivers. Such siltation is a documented contributor to increased flood risk, rates of river migration19, and may contribute to more frequent avulsions or crevasse splays. Though these rare events are difficult to attribute directly to mining, as flood flows and other land use changes are also important contributors, we observe at least three avulsion events downstream of mining areas since 2010, with no similar events occurring in the same period upstream on the same rivers.
We document changes in SSC of recently unprecedented magnitudes in many mining-affected rivers, leading to major alterations to sediment transport and turbidity regimes. SSCs in river reaches downstream of river mining sites are elevated more than 3 standard deviations above un-mined levels at 92% of rivers (159 of 173). SSCs are more than double the historical norms at 80% of rivers (n = 140), five times higher at 43% (n = 73), and an order of magnitude higher at 19% (n = 33) (Extended Data Table 1). At each of the 33 rivers with an order-of-magnitude increase following mining onset, our seasonal analysis shows that minimum post-mining SSC exceeds maximum pre-mining SSC, indicating a complete transformation from pre-mining conditions (based on by-month averages; Extended Data Figs. 7–11).
The suspended sediment concentrations we observe in many mining-affected rivers have increased turbidity to levels associated with disease and mortality in aquatic life not adapted to such conditions20,21. Suspended sediment and associated siltation, along with mining-associated deforestation and heavy metal contamination from the use of mercury and cyanide as processing aids, has been shown to disrupt the life cycles of riverine and riparian flora and fauna22,23. Fish population threats are highlighted by anecdotal and scientific reports of depleted fisheries and/or fish kills downstream of mining sites in Suriname20, Colombia24 and Sierra Leone14. The widespread mining-caused siltation we document suggests that the true scope of these effects is underreported22. Because many regions with active river mining have critical adjacent or overlapping fisheries, heavy metal contamination poses a public health risk, as heavy metals accumulate in the bodies of fish that are then consumed by humans. Researchers have found high concentrations of mercury in miners25, their families26, the local environment27, and people who live in mining regions across the globe12,18, due in part to mining-related mercury contamination in fish28.
Downstream transmission of mining sediment
We conservatively identify >35,000 kilometers of river length where suspended sediment concentrations have been directly altered by river mining activities (Fig. 2). In most cases, changes to SSC are major. More than 21,000 river kilometers have SSCs greater than double their pre-mining average, while 8,390 km were elevated by five times and 3,500 by 10 times or more. We estimate that there are about 500,000 river kilometers >60 m wide in global Tropics; 6–7% of this length is impacted by river mining, a profound effect that has not previously been reported. For individual countries with active mining, the percentage of medium and large rivers altered by river mining is often much higher, on average 23% ± 19% SD (Fig. 3). More than 40% of moderate and large river length is affected in countries with comparatively small river systems, including 57% in French Guiana, 48% in Guyana, and 40% in Côte d'Ivoire and Senegal. The DRC is home to >32,000 medium and large river kilometers, more than any other country in the tropics except for Brazil; more than 26% of these river kilometers have undergone major mining-caused suspended sediment transport increases (8,600 km). The use of medium-resolution satellite imagery may underestimate the true extent of the alteration. Manual mapping at select sites of headwater tributaries too small for Landsat analysis indicates that as much as 2–10 times as many small stream kilometers may be affected by mining, but water quality assessments were not possible using our Landsat satellite approach.
Although environmental degradation is usually most intensive in waters immediately adjacent to mining areas, rivers transport mining-derived sediment in high quantities and across significant distances. We quantify these transport distances for the first time, showing that high SSCs often persist for 10s–100s of kilometers downstream of active mining areas (Fig. 2). On average, river SSC remains >3 SD elevated over pre-mining levels for 162 ± 16 km (Extended Data Fig. 6). This transport distance may be conservative, as we map numerous river reaches that terminate in the ocean. On several reaches, the effects of mining are detectable more than 500 km downstream of active operations (Fig. 2c). For instance, in the Democratic Republic of the Congo, sediment from river mining near the eastern border dramatically alters between 500–1,000 km of the Aruwimi River, Durba River, and Mongbwalu River, despite minimal mining in the intervening river sections. We do not observe systematic downstream attenuation of the mining SSC signal in rivers except where there is dilution from mixing of major un-mined tributaries. In the years 2015–2020, the average mining-affected river had SSCs elevated for more than two-thirds of its mapped length downstream of active mining (68% ± 2%) (Extended Data Fig. 6). This metric has steadily risen during the Landsat 5 and 7 satellite period (1984–present), reflecting increased sediment loading at mining sites consistent with our observations of intensified mining activity in recent decades.
Recent intensification of river mining
Environmental degradation of rivers associated with mining has expanded rapidly in the past two decades as river mining has proliferated across the global tropics. Recent, major mining detectable by our approach began only after 2000 at more than half (60%; n = 227) of currently active river mining sites and after 2009 at nearly one-third of all sites (33%; n = 126). Attributing these global changes to a single cause is not possible but insight can be gained from documented local, regional, and global phenomena. In aggregate, the observed hydrologic impact of mining that we observe fluctuates with the market price of gold (Fig. 4). A Pearson test between the time series of the average global SSC anomaly and the price of gold in global markets has a correlation of R = 0.88 (95% CI: 0.78–0.94; p = 1.4 x 10-12). Notably, the average SSC relative to a pre-mining baseline increased dramatically across the planet during the 2008–09 Global Financial Crisis and subsequent fallout (Fig. 4). Because the price of gold has recently tracked global financial insecurity and coincided with pro-mining Structural Adjustment Programs in numerous countries, disentangling potential contributing pressures is challenging at a global scale, suggesting that myriad socio-economic and political factors mediate changes in the extent of river mining.
Although broader trends in river mining activity are evident across many mining regions in our study, we find that the changing footprint and intensity of mining activity in any given area are context dependent and varies with local dynamics29 (See Supporting Information, e.g., Extended Data Figs 4 and 5). Recent changes to infrastructure10, market structure or availability30, regulatory environments31, and safety29 have allowed and/or incentivized miners to expand or join existing mining operations and/or push into new territory. We observe that mining sometimes follows road building associated with logging, e.g., in Indonesia, Malaysia, and Madagascar. Mining also commonly replaces acreage used for oil palm cultivation or other agriculture, and we observe this dynamic in numerous tropical countries. In sub-Saharan Africa, we observe increased mining activity that has been attributed to a) civil wars in the 1980s and 1990s that decreased stability, caused economic crisis, and involved violent enforcement of mining to fund war efforts32,33; b) increased physical safety but continued economic insecurity following those civil wars33; and c) and economic challenges and economic restructuring in 1990s and early 2000s associated with International Monetary Fund (IMF) and World Bank Structural Adjustment Programs34.
The increase in river mining in the past fifteen years, at a time of global financial insecurity, has occurred despite governmental bans or restrictions on the practice in many regions. In some regions, governments have reduced the environmental effects of river mining by taking actions committed to aligning stakeholders, providing pathways for continued, lower-impact mining or other employment, and enforcing prolonged, consistent regulations35-37. Coordinated efforts by indigenous groups, advocacy organizations, and the Brazilian government eliminated mining in the Kayapo Preserve from 1994–2014 after a mining and logging boom in the 1980s and early 1990s36,38. We find that the Kayapo intervention resulted in nearly two decades without mining-related deforestation. Despite slow or non-existent revegetation, we find that suspended sediment concentrations declined to levels consistent with unaffected rivers in the region in less than a decade. Other regional or national mining environmental regulation include various administrative actions in the United States, most notably the Clean Water Act, New Zealand’s Resource Management Act35, and China’s ban on river mining in 199637. Although such protections have documented shortcomings in both environmental protection39 and community engagement or support35, we observe minimal downstream transmission of contemporary, mining-derived sediment in these countries. More recently, the Peruvian government in 2019 intervened against illegal mining in the Madre de Dios region, reducing illegal mining in one area to nearly zero for two years40. Similar efforts have enforced regulations and/or short-term moratoria or bans on unsanctioned river mining in Sumatra41 and Ghana42, part of attempts to bring informal mining into the formal sector.
Despite these examples, environmental mitigation is uncommon and the exception to a nearly monotonic global increase in river mining in the Tropics since 2000. Interventions against informal mining activity have been unsuccessful or only partially successful in many regions43,44, hampered by insufficient resources for enforcement, failure to address underlying socio-economic, cultural, and political drivers, and unclear or onerous permitting requirements that stifle formalization42,45-48. Even long-standing efforts can be reversed without sustained action or enforcement. Since 2010, and especially from 2016 to the present, miners have returned to Kayapo mining areas and have resumed deforestation and mining at high rates, leading to river suspended sediment concentrations approaching levels not observed since the 1980s and 1990s mining boom (Extended Data Figs 4 and 5). Interviews with miners49 and a prior study suggest that return to mining in the Kayapo region is the combined result of financial pressures, rising prices of gold, and decreased environmental enforcement from Brazil’s central government50, although we cannot yet untangle the relative importance of these factors. However, these broad dynamics—financial and land insecurity51-53, demand and active markets for gold and other precious minerals50, and limited environmental enforcement9,43,46—are common across much of the global Tropics, and likely help to explain individual, corporate, and government efforts to expand operations in the past two decades.
River mining is seen as a mechanism for poverty alleviation, upward economic mobility, and the diversification of livelihood opportunities, mostly in the global South46. Artisanal and small-scale mining directly or indirectly supports as many as 134 million people54, and many are also employed by large-scale river mining operations run by corporations or governments. Because they view mining as an agent of development55, donor organizations like the World Bank and International Monetary Fund often encourage or require mining expansion or codification before releasing aid34. Gold and other minerals mined from rivers are key to complex global electronics supply chains and integral to ongoing global efforts for decarbonization and increasing electrification54. Thus, addressing the environmental degradation associated with these mining sites is, on several levels, critically important in sustainable national and international development.
In documenting the widespread impacts of river mining, we demonstrate the serious threats to the environmental integrity of tropical rivers across the Earth. The dramatic global increase in river mining in the past two decades has occurred during a period of increased global insecurity and high demand for precious minerals. We show that, despite efforts to regulate and formalize river mining, environmental consequences have proliferated in almost every region where river mining occurs. Mining can open critical pathways for subsistence, diversification of livelihood strategies, and labor force integration for women, and could support regional or national development. Efforts to monitor and mitigate the externalities associated with mining without stifling these benefits of mining are thus essential. Our efforts to document and analyze these consequences may raise awareness of this pressing concern and are useful for governments and policy makers seeking to understand and reduce the environmental consequences of river mining.
Description of river mining
The river mining process, by definition, occurs in or directly adjacent to river channels. Miners remove vegetation, then excavate and process large quantities of sediment, either soil or river bed sediment. Many miners operate informally, without formal government authorization in the form of mineral rights or land titles56,57, and/or despite government interventions against mining activity40,44,58,59. This small- and artisanal-scale mining (often abbreviated ASM or, in the case of gold mining, ASGM) often occurs in watersheds or sites with larger, more formalized mining operations with similar environmental footprints. Because formal and informal mining operations are intermingled in many cases, often with joined or indistinguishable environmental impacts, we do not treat them separately. Across scales, river gold mining techniques have similarities across the globe (e.g., Fig. 1b–e). Although miners operate with varying levels of mechanization, in general the process involves transferring excavated sediment onto a sluice, with gravity and flowing water separating out minerals48,60. Some operations are entirely manual, with miners using their hands and shovels to excavate and transfer sediment. Our study focuses mostly on larger operations, where miners use various equipment to dig and transfer sediment. These include excavators, used for digging, clearing, and loading, and diesel-powered engines, which miners use to pump sediment-laden water out of a river or excavated mining pit. We find that most river mining identified in this study is for gold (342/381 sites). Often, the final stage of the gold mining process involves adding mercury to isolate gold in a mercury-gold amalgam61. Sediment, mercury, and other heavy metals and contaminants from the mining process are either released directly to the river12,15,17 or the air27 or stored on-site in unconsolidated tailings62, which are susceptible to erosion during rain storms and floods. Habitat destruction is likely most acute in mining districts themselves, as rivers are comprehensively reconfigured by mining suction pumps, excavators, and intensive deforestation in riparian areas; however extensive modification of the land surface appears to have lasting impacts on vegetation recovery63. Because river mining thus accelerates sediment delivery processes in rivers, examining affected rivers offers insight into both the magnitude and extent of human and environmental consequences.
Although most mining we identify is for gold, some sites, particularly in Angola and the Democratic Republic of the Congo, focus instead on diamonds64,65. The resource-rich eastern Democratic Republic of the Congo also has substantial cassiterite and coltan reserves, as well as a host of rare earth elements65, some of which are mined in river deposits. Reports indicate that mining for multiple precious minerals co-occurs at some sites in the DRC and the Central African Republic65. We do not focus here on sand mining despite its global significance as an agent of local geomorphic change in rivers. Sand mining, in general, may lead to net removal of sediment from rivers and is not likely to produce the same downstream effects we observe from river mining66. In addition, although major pit mines have important environmental implications, we only focus on those with direct surface water connection to rivers and streams, as is the case for select gold and diamond mines around the world, as well as some nickel mines in New Caledonia, Indonesia, and the Philippines.
Mapping mining sites
Official information about the existence, extent, and type of river mining is exceedingly difficult to find. Large-scale, formal mining is generally documented in corporate releases and investor materials, with capital and resource flow recorded and published by companies, nations, and intergovernmental organizations. By contrast, systematic public documentation about river mining is sparse or absent in many countries and areas due to the illegal and/or informal nature of the activity. We undertook a widespread review of academic literature, government and NGO reports, media stories and miner interviews, and social media posts (See Supporting Information) to identify origins, scope, timing, and other details of river mining activities in this study.
We manually identified mining areas through a combination of literature and popular media review; mining company and national reporting on mining activity; social media investigations using YouTube, Facebook, and Instagram, as well as extensive manual satellite and aerial image review using Google Earth, Bing Maps, Apple Maps, Planet Labs imagery, and Google Earth Engine analysis67 of Landsat and Sentinel-2 data. Discrete mining areas were defined as sites either a) in different watersheds; b) associated with different towns, mining companies, and/or mined metal/mineral; and/or c) beginning more than a decade apart. Once identified, we located each site and recorded its likely dominant mined metal/mineral. We then established the likely timing of mining onset using Google Earth, Google Earth Engine, and Planet Labs time-series tools in concert with the Hansen, et al. 68 deforestation dataset.
We manually mapped river courses in watersheds affected by river mining. We located the upstream end of each transect at the point where the river achieved an average width of >60 m, at least during the wet season, to allow for analysis by Landsat images with 30 m spatial resolution. We primarily mapped rivers downstream of mining because most mining occurred in the headwaters, precluding mapping of upstream river reaches unaffected by mining. We extended each transect downstream to its outlet in the ocean, to a confluence with a major river with drainage area greater than the drainage area of the mining-affected river at that confluence. “Pre-mining” reference data could be taken from the mining-affected reach during the pre-mining period for most rivers. However, when mining began before 1984 (the beginning of the Landsat 5 mission), we needed to establish reference reaches elsewhere. In these cases, we a) began transects upstream of mining areas, if those reaches were mappable (i.e., >60 m wide) or b) established a reference reach on a nearby river, either one without mining during the Landsat 5 and 7 satellite record or during the pre-mining period if mining activities began sometime after 1984. Some watersheds contain multiple mining areas, whereas 93 mining areas are on rivers that are too narrow to map with Landsat satellites.
Using satellites to sample mining-affected and reference rivers
After delineating transects on mining-affected and reference rivers, we used Google Earth Engine67 to split each transect into 2-km-long reaches, then sampled the cloud-, shadow-, and snow-free river pixels from each USGS Surface Reflectance L2T Product image in the Landsat 5 and 7 record (here simply called the “satellite record”) from 1984–2021 (endpoint years inclusive)6,17,69. River transects had an average of 595 unique days of images (min: 73, max: 2382) (Extended Data Fig. 1). We used the suspended sediment concentration (SSC) algorithms developed by69 to convert Landsat surface reflectance data to SSC. The entire dataset thus includes 7,346,222 measurements of SSC, allowing for robust comparisons between pre- and post-mining states and, in most cases, analysis on annual timescales. The number of unique images at each river varied due to differences in transect lengths, image losses in the Landsat record due to receiver and/or storage lapses, cloudiness, and snow/ice cover for higher-latitude rivers, i.e., those in Mongolia and Russia.
We analyzed SSC in a variety of timeframes/spatial extent contexts. We defined the reference period for each of the 173 mining-affected river transects as the pre-mining period for rivers affected by mining and, for rivers used exclusively as un-mined references, as the entire satellite record (1984–2021). We calculated reference metrics for each transect as the SSC average and standard deviation for the entire transect during the reference period. For monthly analyses, we computed the reference average and standard deviation of SSC for each calendar month over all the years in the reference period. On an annual basis, to ease cross-comparison among rivers with average SSCs ranging over orders of magnitude, we computed the deviation from the reference SSCs as a Z-Score:
where, respectively, SSCmining–3yr,reach is the moving average of SSC values, σ2mining–3yr,reach is the standard deviation, and nmining–3yr,reach is the number of months with SSC data during a 3-year window and on a reach in question. SSCref, σ2ref, and nref are, respectively, the average, standard deviation, and the number of months with SSC samples in the reference dataset. We computed these values for 10 and 50 km reaches, and for the entire mining-affected transect. In addition, we computed monthly Z-Scores:
with the same formulation and parameters as in Eq. 1, but computed for each individual calendar month, indicated by the subscript i. We also computed monthly analysis on an annual basis, but the monthly results presented here are computed as indicated in Eq. 2, using the entire post-mining record relative to monthly reference SSC.
We also computed fractional deviation from reference SSC as a simple ratio:
Where the parameters are the same as in Eq. 1.
We computed aggregate measures of these data on a country and land-mass basis for various periods. We calculated the median annual SSC Z-Score for all transects for each year to construct a time series of global SSC riverine impacts and made the same calculation for each continent and country. For each five-year period, we computed the number of river kilometers with a) average SSC greater than the reference average SSC; b) SSC Z-Score more than three standard deviations above the reference distribution (i.e., SSC Zannual > 3); c) average SSC greater than 2, 5, and 10 times the reference average SSC (i.e., SSC* > 2, 5, and 10, respectively). For the latter two calculations, we took a conservative approach to minimize the effect of outliers, only designating a river reach as over the threshold if it was over that threshold for more than 3 years of a 5 year period. For instance, if a 10 km reach had an SSC* of >5 in each year between 2015–2019, we would designate it as elevated by 5x during the half decade beginning in 2015. However, if in 2013 and 2014 SSC* was less than 5, even if the half-decade-average SSC* was greater than five, that reach would not be designated as elevated by 5x. To understand the changes over time in river lengths affected by river mining, we summed the number of river kilometers above each threshold on a global, land-mass, and country basis. We also calculated the average river length affected by mining in the period 2015–2020. To determine the attenuation of the mining-caused SSC signal downstream of mining areas, we computed, for each river, the percent of river length along and/or downstream affected by mining at each threshold. We compared the lengths of rivers affected by mining to the total number of river kilometers that meet our Landsat-based width threshold. To do so, we merged the WWF hydrology70 vector dataset and the MERIT Hydro river width dataset to identify all rivers that averaged >60 m wide for each WWF-designated segment71. We summed the lengths of rivers that met this width criteria for each country a) in our study and b) in the global Tropics. Because Myanmar straddles the Tropic of Cancer, we included all rivers in that country rather than restricting it to tropical rivers; this adjustment increased the river length by ~4,000 km.
Comparisons to the price of gold
We plotted the monthly price of gold72 against the global median and distributions of SSC Z-Scores. To quantify the relationship between these time series, we used the parametric Pearson Correlation Test. Because the price of gold was computed monthly, we used only the price of gold from January of each year.
Code and Data Availability
All code necessary for accessing the raw data used for this study and re-producing our results is available at https://github.com/evandethier/global-alluvial-mining.
Acknowledgments
This research was funded by NASA award 20-LCLUC2020-0025.
Contributions
E.D, M.S., D.L., F.M., and C.R. devised the project. E.D., D.L., C.R., and F.M. obtained project funding. E.D. analyzed the data. E.D. drafted the manuscript and all authors contributed to the final version.
1 Bowie, A. J. A practical treatise on hydraulic mining in California. 11th edn, (D. Van Nostrand company, 1910).
2 Cleary, D. & St. Antony's College (University of Oxford). Anatomy of the Amazon gold rush. (Macmillan in association with St. Antony's College, Oxford, 1990).
3 Hilson, G. M. The future of small-scale mining: environmental and socioeconomic perspectives. Futures 34, 863-872 (2002).
4 Schwartz, F. W., Lee, S. & Darrah, T. H. A Review of the Scope of Artisanal and Small-Scale Mining Worldwide, Poverty, and the Associated Health Impacts. Geohealth 5 (2021).
5 Barney, K. Reassembling informal gold-mining for development and sustainability? Opportunities and limits to formalisation in India, Indonesia and Laos. THE PICK, 335 (2018).
6 Dethier, E. N., Renshaw, C. E. & Magilligan, F. J. Rapid changes to global river suspended sediment flux by humans. Science 376, 1447-1452 (2022).
7 Best, J. Anthropogenic stresses on the world's big rivers. Nat. Geosci. 12, 7-21 (2019).
8 Asner, G. P. & Tupayachi, R. Accelerated losses of protected forests from gold mining in the Peruvian Amazon. Environmental Research Letters 12 (2017).
9 Papworth, S. et al. The impact of gold mining and agricultural concessions on the tree cover and local communities in northern Myanmar. Sci Rep-Uk 7 (2017).
10 Espejo, J. C. et al. Deforestation and Forest Degradation Due to Gold Mining in the Peruvian Amazon: A 34-Year Perspective. Remote Sens-Basel 10 (2018).
11 Giffin, L. et al. Clinical Impact of Mercury Toxicity among Indigenous Peoples of the Peruvian Amazon. Am J Trop Med Hyg 99, 161-162 (2018).
12 Diringer, S. E. et al. River transport of mercury from artisanal and small-scale gold mining and risks for dietary mercury exposure in Madre de Dios, Peru. Environ Sci-Proc Imp 17, 478-487 (2015).
13 de Santi, V. P. et al. Malaria in French Guiana Linked to Illegal Gold Mining. Emerg Infect Dis 22, 344-346 (2016).
14 Marcantonio, R. A., Field, S. P., Sesay, P. B. & Lamberti, G. A. Identifying human health risks from precious metal mining in Sierra Leone. Reg Environ Change 21 (2021).
15 Telmer, K., Costa, M., Angélica, R. S., Araujo, E. S. & Maurice, Y. J. J. o. e. m. The source and fate of sediment and mercury in the Tapajós River, Pará, Brazilian Amazon: Ground-and space-based evidence. Journal of Environmental Management 81, 101-113 (2006).
16 Lobo, F. D., Costa, M., Novo, E. M. L. D. & Telmer, K. Distribution of Artisanal and Small-Scale Gold Mining in the Tapajos River Basin (Brazilian Amazon) over the Past 40 Years and Relationship with Water Siltation. Remote Sens-Basel 8 (2016).
17 Dethier, E. N., Sartain, S. L. & Lutz, D. A. Heightened levels and seasonal inversion of riverine suspended sediment in a tropical biodiversity hot spot due to artisanal gold mining. P Natl Acad Sci USA 116, 23936-23941 (2019).
18 Gerson, J. R. et al. Artificial lake expansion amplifies mercury pollution from gold mining. Science Advances 6 (2020).
19 Constantine, J. A., Dunne, T., Ahmed, J., Legleiter, C. & Lazarus, E. D. Sediment supply as a driver of river meandering and floodplain evolution in the Amazon Basin. Nat. Geosci. 7, 899-903 (2014).
20 Mol, J. H. & Ouboter, P. E. Downstream effects of erosion from small-scale gold mining on the instream habitat and fish community of a small neotropical rainforest stream. Conservation Biology 18, 201-214 (2004).
21 Mcleay, D. J., Birtwell, I. K., Hartman, G. F. & Ennis, G. L. Responses of Arctic Grayling (Thymallus-Arcticus) to Acute and Prolonged Exposure to Yukon Placer Mining Sediment. Canadian Journal of Fisheries and Aquatic Sciences 44, 658-673 (1987).
22 Azevedo-Santos, V. M. et al. Negative impacts of mining on Neotropical freshwater fishes. Neotrop Ichthyol 19 (2021).
23 Keovilignavong, O. Mining governance dilemma and impacts: A case of gold mining in Phu-Hae, Lao PDR. Resour Policy 61, 141-150 (2019).
24 Velez-Torres, I., Vanegas, D. C., McLamore, E. S. & Hurtado, D. Mercury Pollution and Artisanal Gold Mining in Alto Cauca, Colombia: Woman's Perception of Health and Environmental Impacts. J Environ Dev 27, 415-444 (2018).
25 Taux, K., Kraus, T. & Kaifie, A. Mercury Exposure and Its Health Effects in Workers in the Artisanal and Small-Scale Gold Mining (ASGM) Sector—A Systematic Review. Int J Env Res Pub He 19, 2081 (2022).
26 Gibb, H. & O’Leary, K. G. Mercury exposure and health impacts among individuals in the artisanal and small-scale gold mining community: a comprehensive review. Environ Health Persp 122, 667-672 (2014).
27 Gerson, J. R. et al. Amazon forests capture high levels of atmospheric mercury pollution from artisanal gold mining. Nat Commun 13 (2022).
28 Kambey, J. L., Farrell, A. P. & Bendell-Young, L. I. Influence of illegal gold mining on mercury levels in fish of North Sulawesi's Minahasa Peninsula, (Indonesia). Environmental Pollution 114, 299-302 (2001).
29 Geenen, S. 'Qui Cherche, Trouve'The political economy of access to gold mining and trade in South Kivu, DRC, Universiteit Antwerpen (Belgium), (2014).
30 Botchwey, G., Crawford, G., Loubere, N. & Lu, J. X. South-South Irregular Migration: The Impacts of China's Informal Gold Rush in Ghana. Int Migr 57, 310-328 (2019).
31 Weng, L. et al. Asian investment at artisanal and small-scale mines in rural Cameroon. The Extractive Industries and Society 2, 64-72 (2015).
32 Le Billon, P. Diamond wars? Conflict diamonds and geographies of resource wars. Ann. Assoc. Am. Geogr. 98, 345-372 (2008).
33 De Boeck, F. Garimpeiro Worlds: Digging, Dying & 'Hunting' for Diamonds in Angola. Review of African Political Economy 28, 549-562 (2001).
34 Hilson, G. 'Formalization bubbles': A blueprint for sustainable artisanal and small-scale mining (ASM) in sub-Saharan Africa. Extr Ind Soc 7, 1624-1638 (2020).
35 Ruckstuhl, K., Thompson-Fawcett, M. & Rae, H. Māori and mining: Indigenous perspectives on reconceptualising and contextualising the social licence to operate. Impact Assessment and Project Appraisal 32, 304-314 (2014).
36 Turner, T. An Indigenous People's Struggle for Socially Equitable and Ecologically Sustainable Production: The Kayapo Revolt Against Extractivism. Journal of Latin American Anthropology 1, 98-121 (1995).
37 Wu, Y. G., Xu, Y. N., Zhang, J. H., Hu, S. H. & Liu, K. Heavy metals pollution and the identification of their sources in soil over Xiaoqinling gold-mining region, Shaanxi, China. Environ Earth Sci 64, 1585-1592 (2011).
38 Zimmerman, B., Peres, C. A., Malcolm, J. R. & Turner, T. Conservation and development alliances with the Kayapo of south-eastern Amazonia, a tropical forest indigenous people. Environ Conserv 28, 10-22 (2001).
39 Palmer, M. A. et al. Mountaintop Mining Consequences. Science 327, 148-149 (2010).
40 Dethier, E. et al. Mining pond water quality recovery and reclamation following abandonment indicated by Copernicus Sentinel multi-spectral and radar sensor fusion analysis. Conserv Lett (2022).
41 Putra, E. I. in Mongabay (2020).
42 Hilson, G. & Maconachie, R. For the Environment: An Assessment of Recent Military Intervention in Informal Gold Mining Communities in Ghana. Land Use Policy 96 (2020).
43 Dondeyne, S., Ndunguru, E., Rafael, P. & Bannerman, J. Artisanal mining in central Mozambique: Policy and environmental issues of concern. Resour Policy 34, 45-50 (2009).
44 Spiegel, S. J. Legacies of a nationwide crackdown in Zimbabwe: Operation Chikorokoza Chapera in gold mining communities. J Mod Afr Stud 52, 541-570 (2014).
45 Diaz, F. A., Katz, L. E. & Lawler, D. F. Mercury pollution in Colombia: challenges to reduce the use of mercury in artisanal and small-scale gold mining in the light of the Minamata Convention. Water Int 45, 730-745 (2020).
46 Verbrugge, B. Voices from below: artisanal-and small-scale mining as a product and catalyst of rural transformation. Journal of rural studies 47, 108-116 (2016).
47 Spiegel, S. J. Governance institutions, resource rights regimes, and the informal mining sector: Regulatory complexities in Indonesia. World Dev 40, 189-205 (2012).
48 Diaz, J. Ecologies of Gold: Understanding the social, political, and ecological impacts of mercury use in informal, small-scale gold mining in Madre de Dios, Peru, University of California, Berkeley, (2021).
49 Anderson, J. L. Blood gold in the Brazilian rain forest. New Yorker, 639-649 (2019).
50 Alvarez-Berríos, N. L. & Mitchell Aide, T. Global demand for gold is another threat for tropical forests. Environmental Research Letters 10, 014006 (2015).
51 Kelly, J. T. D. "This mine has become our farmland": Critical perspectives on the coevolution of artisanal mining and conflict in the Democratic Republic of the Congo. Resour Policy 40, 100-108 (2014).
52 Geenen, S. 'Who Seeks, Finds': How Artisanal Miners and Traders Benefit from Gold in the Eastern Democratic Republic of Congo. Eur J Dev Res 25, 197-212 (2013).
53 Maclin, B. J., Kelly, J. T. D., Perks, R., Vinck, P. & Pham, P. Moving to the mines: Motivations of men and women for migration to artisanal and small-scale mining sites in Eastern Democratic Republic of the Congo. Resour Policy 51, 115-122 (2017).
54 World Bank. (eds J McQuilken & R Perks) (World Bank, Washington, DC, 2020).
55 De Haan, J., Dales, K. & McQuilken, J. Mapping Artisanal and Small-Scale Mining to the Sustainable Development Goals. Newark DE: University of Delaware (Minerals, Materials and Society program in partnership with PACT). Available online via http://www. pactworld. org and https://sites. udel. edu/ceoe-mms/.(Accessed 10th January 2021) (2020).
56 Dondeyne, S. & Ndunguru, E. Artisanal gold mining and rural development policies in Mozambique: Perspectives for the future. Futures 62, 120-127 (2014).
57 Salomon, M. A nova corrida do ouro na Amazônia. (Instituto Escolhas, 2020).
58 Gallwey, J., Robiati, C., Coggan, J., Vogt, D. & Eyre, M. A Sentinel-2 based multispectral convolutional neural network for detecting artisanal small-scale mining in Ghana: Applying deep learning to shallow mining. Remote Sens Environ 248 (2020).
59 Ayelazuno, J. A. & Mawuko-Yevugah, L. Between the Africa Mining Vision and the neo-patrimonial state: The agency gap in Ghana's regulation of artisanal and small-scale gold mining. S Afr J Int Aff 28, 555-582 (2021).
60 Vieira, R. Mercury-free gold mining technologies: possibilities for adoption in the Guianas. J Clean Prod 14, 448-454 (2006).
61 Davies, G. R. A toxic free future: Is there a role for alternatives to mercury in small-scale gold mining? Futures 62, 113-119 (2014).
62 Veiga, M. M. & Gunson, A. J. Gravity Concentration in Artisanal Gold Mining. Minerals 10, 1026 (2020).
63 Kalamandeen, M. et al. Limited biomass recovery from gold mining in Amazonian forests. Journal of Applied Ecology 57, 1730-1740 (2020).
64 le Billon, P. Angola's political economy of war: The role of oil and diamonds, 1975-2000. Afr Affairs 100, 55-80 (2001).
65 Muller, T. (ed International Peace Information Service) (2020).
66 Hackney, C. R. et al. River bank instability from unsustainable sand mining in the lower Mekong River. Nat Sustain 3, 217-+ (2020).
67 Gorelick, N. et al. Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sens Environ 202, 18-27 (2017).
68 Hansen, M. C. et al. High-Resolution Global Maps of 21st-Century Forest Cover Change. Science 342, 850-853 (2013).
69 Dethier, E. N., Renshaw, C. E. & Magilligan, F. J. Toward Improved Accuracy of Remote Sensing Approaches for Quantifying Suspended Sediment: Implications for Suspended-Sediment Monitoring. Journal of Geophysical Research: Earth Surface 125, e2019JF005033 (2020).
70 Grill, G. et al. Mapping the world’s free-flowing rivers. Nature 569, 215-221 (2019).
71 Yamazaki, D. et al. MERIT Hydro: A High-Resolution Global Hydrography Map Based on Latest Topography Dataset. Water Resour. Res. 55, 5053-5073 (2019).
72 World Gold Council. (ed World Gold Council) (2022).
There is NO Competing Interest.