Introduction
Terrorist attacks targeting the mining sector represent an under-explored and unique threat. Mining of industrial metals, critical minerals, and other mineral resources is projected to increase over the next decades, both by emerging and developed economies. Mines are valuable economic assets, but are also inherently hazardous environments, often in remote locations with confined spaces, volatile materials, and heavy machinery.1 The resulting patterns of injury from attacks on these locations can include ballistic, blast, and crush injuries.Reference Harjai, Agarwal and Dave2 Mining-related terrorist attacks carry direct public health consequences. These incidents can occur far from definitive medical care and strain limited trauma capacity in remote regions. Workers injured in remote mine sites may face prolonged evacuation times, delayed access to definitive care, and overtaxed local health infrastructure, while communities near mining operations may experience displacement, disrupted health services, and lasting psychological trauma following attacks. However, there exists a gap in the literature regarding common characteristics of attacks on mining sites.
Malicious groups may target the mining industry in attacks—whether on mining sites, quarries, equipment, infrastructure, or workers—for numerous reasons. Benefits to the perpetrating group may accrue through revenue generation or abducting for ransom, while damage against the target may occur through asset destruction and supply chain disruption.Reference Pate3 Ideological functions of mining-related attacks include expression of social grievances, assault on groups in control of resources, or elimination of soft targets (i.e., facilities that are difficult to secure, used by the public in large numbers, or have potential to result in widespread adverse impact4) to cultivate a climate of fear.Reference Iqbal5 However, analyses focusing specifically on mining-related attacks are scarce. This study is an epidemiological examination of terrorism-related events targeting the mining industry from 1970 to 2020.
Methods
Data collection was performed using a retrospective database search of the Global Terrorism Database (GTD). The GTD contains over 200,000 global terrorism incidents and is maintained by the National Consortium for the Study of Terrorism and Responses to Terrorism.6 Acts of state terrorism are omitted from the database. The GTD contains no personal identifiers.
The GTD defines a terrorist attack as the “threatened or actual use of illegal force and violence by a non-state actor to attain a political, economic, religious, or social goal through fear, coercion, or intimidation.” Attacks met inclusion criteria into the GTD if they fulfilled all three of the following attributes, determined within the database and not by the authors6:
-
1. The incident must be deliberate;
-
2. The incident must involve some degree of violence or an immediate threat of violence;
-
3. The perpetrators must be non-state or sub-national actors.
Ambiguous events were excluded when there was uncertainty as to whether the incident fulfilled criteria for GTD inclusion as a terrorist incident. Additionally, 2 of the 3 following criteria must be present6:
-
1. The act must be carried out in pursuit of a political, economic, religious, or social objective;
-
2. There must be evidence that the act was intended to intimidate, coerce, or send a message to a wider audience beyond the direct victims;
-
3. The act must occur outside the bounds of lawful armed conflict.
Starting in 1970, data in the GTD was sourced from a variety of daily media articles across multiple languages using both manual and automated strategies, as described in the GTD codebook.6 Articles reporting on terrorist attacks were scraped, filtered for duplicates, and human-verified before being added to the database. Events were included in the GTD only if documented by at least one high-quality source, defined as independent, reporting externally verifiable content, and relying on primary sources.6
The GTD was downloaded and searched within the date range of January 1, 1970, to December 31, 2020. Pre-coded variables for primary target type “business” and target subtype “mining” were used to identify attacks involving the mining industry. Data on standard GTD codebook variables collected per incident included: date of incident, location (longitude/latitude; country; world region), successfulness of the attack (i.e., whether the intended act was completed), number of victims killed or wounded, number of hostages, method of attack, weapon type, weapon subtype, and property damage value (US dollar amount at time of incident). Additional variables were manually derived from event summary descriptions, including perpetrating terrorist group category (Table S1) and site mineral resource. Although many perpetrating groups espouse more than one ideology, distinguishing between dominant and subordinate ideologies is difficult and therefore the most plausible ideology was selected.Reference Pate3
Data was analyzed using R v4.5.0. Categorical variables were reported using frequencies and percentages, and continuous variables using median and interquartile range (IQR).
Results
There were 541 mining-related attacks documented during the 1970–2020 study period, accounting for 0.26% of all terrorist attacks (209,706 total events). These attacks led to 617 fatalities and 574 non-fatal injuries. 96.9% of these attacks were deemed successful. Attacks were more frequent in years after 2010, with the largest number clustered in the 2010–2020 decade and the highest number of attacks occurring in 2018 (Figure 1). Prior to that, a stand-alone spike in attacks occurred in 1996.
Terrorist attacks on the mining industry by year.
Figure 1. Long description
The X-axis represents the Year from 1970 to 2020 in ten-year increments. The Y-axis represents Frequency from 0 to 60 in increments of 20. The data is presented as blue vertical bars.
From 1970 to the early 1990s, the frequency remains low, generally under 15 attacks per year, with minor fluctuations. A notable isolated spike occurs in the mid-1990s, reaching nearly 30 attacks, followed by a sharp decline to near-zero levels by the early 2000s.
Starting around 2008, there is a significant and rapid upward trend. The frequency climbs steadily, peaking at approximately 58 attacks around 2018. The chart ends with a sharp drop in 2019 followed by a moderate rebound to roughly 28 attacks in 2020.
Geographically, South Asia (23.7%), South America (22.9%), and Southeast Asia (22.7%) accounted nearly equally for the largest share of mining-related attacks (Figure S1, Table S2). Attacks in sub-Saharan Africa, despite comprising 16.6% of total attacks, comprised the largest share of deaths (60.0%) and injuries (61.5%). The top 3 deadliest mining-related terror incidents were all in sub-Saharan Africa, including Burkina Faso, Nigeria, and Angola (Table 1). The countries with the highest number of attacks were the Philippines (17.6%) and India (16.1%). No attacks were reported in North America.
Top 10 deadliest terrorist attacks on the mining industry ranked by total deaths
Table 1. Long description
The table contains eight columns: No., Year, Country, City and province, Description, Perpetrator, Deaths, and Injuries.
* Rank 1: 2019, Burkina Faso, Fada N’Gourma, Est. Explosion of convoy carrying Semafo gold mine workers by Islamic State in the Greater Sahara. 37 deaths, 60 injuries.
* Rank 2: 2016, Nigeria, Bindim, Zamfara. Armed assault on mine by Unknown. 30 deaths, 41 injuries.
* Rank 3: 1998, Angola, Lucapa, Lunda Norte. Armed assault on diamond mine, vehicles destroyed, warehouses looted by National Union for the Total Independence of Angola. 24 deaths, 7 injuries.
* Rank 4: 2018, Venezuela, Bolivar. Armed assault on gold mine by National Liberation Army of Colombia. 20 deaths, no data for injuries.
* Rank 5: 2019, Burkina Faso, Madoudji, Sahel. Armed assault on gold mine by Ansar al-Islam and Jamaat Nusrat al-Islam wal Muslimin. 20 deaths, no data for injuries.
* Rank 6: 2009, Democratic Republic of the Congo, Walikale, North Kivu. Armed assault on Bisie cassiterite mine by Mai-Mai and Democratic Forces for the Liberation of Rwanda. 19 deaths, 45 injuries.
* Rank 7: 2015, Sudan, Talodi, South Kordofan. Armed assault on gold mine by Sudan People’s Liberation Movement North. 18 deaths, 15 injuries.
* Rank 8: 2015, Sudan, Talodi, South Kordofan. Armed assault on gold mine by Sudan People’s Liberation Movement North. 17 deaths, 15 injuries.
* Rank 9: 2015, Kenya, Mandera, Mandera. Explosion of residential compound by Al-Shabaab. 14 deaths, 11 injuries.
* Rank 10: 2018, Nigeria, Kuru-Kuru, Zamfara. Armed assault on mine by Unknown. 13 deaths, 0 injuries.
Anarchist/revolutionary groups (46.0%) took responsibility for the largest percentage of mining-related attacks, while fascist groups comprised the smallest share (0.4%) (Table S3). The New People’s Army, an armed wing of the Philippine Communist Party, was the group responsible for the largest percentage of mining-related attacks (14.0%). While ethno-nationalist/separatist groups were responsible for a smaller share (20.0%) of attacks, their incidents possessed the highest percentages of deaths (35.0%) and injuries (42.5%). Similarly, despite religious extremist groups contributing to a small percentage (6.1%) of total mining-related terror attacks, their incidents carried an outsized impact in terms of deaths (16.4%) and injuries (17.6%).
Armed assault was the most common modality of attack (28.1%) and comprised the highest fraction of deaths (56.9%) and injuries (51.0%). Facility/infrastructure attack (25.9%) and bombing/explosion (19.2%) were other common modalities, whereas hijacking (0.4%) was the least common (Table S4). Hostage-taking was involved in 19.0% of incidents, across which 775 hostages were taken. Suicide attacks consisted 0.4% of attacks. Firearms (38.6%) were the leading weapon type used and comprised the highest percentage of deaths (63.7%) and injuries (52.8%) (Table S5). Incendiaries (24.8%) and explosives (20.9%) were also prominently used. Arson/fire was implicated in a sixth of attacks (16.6%) (Table S6).
Mine sites of attack varied across the mineral resource or commodity they extracted. Coal (20.0%) and gold (10.4%) were the most common minerals, whereas the vast majority (47.7%) of mines’ commodity focus was unrecorded (Table S7). Attacks targeting diamond mines were uncommon (0.9%). Median value of property damage across 48 incidents with reported cost data was US$100,000 (IQR: $40,000 to $1,000,000).
Discussion
The global mining industry employs millions worldwide and is a major contributor to gross domestic product, particularly in commodity-based economies. Mining sites can be attractive targets for terrorist attacks due to (1) low mobility and high concentration of the commodity, (2) high economic/strategic value, and (3) geographical remoteness that makes protection difficult.Reference Lee7 Attacks on the mining industry, like other soft targets, can be highly lethal, inflict significant financial damage, and disrupt local communities.
The increase in mining attacks after 2010 may reflect expanding mining activity in politically unstable regions and intensifying competition over natural resources, although temporal changes in data reporting practices within the GTD may confound the observed trend. The top regions by volume of mining-related terrorist attacks (South Asia, South America, and Southeast Asia) broadly correspond with areas of substantial global mineral production. For instance, in South Asia, India contains some of the world’s largest coal-producing mines in Jharkhand and Chhattisgarh. In South America, Brazil’s Carajas mine is one of the largest iron ore operations globally, while Chile’s Escondida mine is the world’s largest copper producer. In Southeast Asia, Indonesia is a major extractor of nickel in Sulawesi, while China remains one of the leading producers of gold worldwide. The highest-casualty terror incidents occurred in sub-Saharan Africa, potentially reflecting background social instability, weak governance, or contested control of resource-rich areas. In contrast, the high frequency of incidents in South Asia and Southeast Asia with lower casualties may reflect a greater proportion of localized insurgent activity or smaller-scale attacks rather than mass-casualty events.
A combination of armed assault and firearms emerged as the most common modality and weapon type, respectively, in mining-related attacks. Facilities/infrastructure were often targeted, and hostage taking was common. This differed from a previous study of terrorist attacks on energy- and mining-related targets, which likely included power plants and oil refineries and therefore may have involved different motives and attack modalities; that study found 74% of attacks were bombings, while only 11% targeted facilities/infrastructure, 7% involved armed assault, and 5% had hostage taking.Reference Pate3 However, the present results supported the study’s finding that anarchist groups were common perpetrators while fascist groups were uncommon. The current results also provide a new angle to previous research on terrorist attacks on energy security, which concluded that (1) attacks against energy infrastructures are comparatively few, (2) no strong connections exist between terrorist group ideology and proclivity to attack, and (3) attacks are clustered in a handful of countries.Reference Toft, Duero and Bieliauskas8 The present evidence supports points (1) and (3), but remains insufficient to substantiate point (2).
The most targeted mineral was coal, the most extracted mineral worldwide and a key facet of many nations’ energy mix. Surprisingly, gold was the only metal of the 4 so-called conflict minerals (tantalum, tin, tungsten, gold)9 that produced a signal of terrorist attacks. Gold commands the highest value and price per ounce of this group of minerals, potentially explaining these trends. However, unreported mineral types, which comprise the largest percentage of data, may be absorbing the signal for minerals like tantalum, tin, or tungsten.
From a medical and public health perspective, these findings underscore the need for enhanced emergency preparedness at mining sites. Priorities include establishing site-specific emergency response protocols, expanding mass casualty triage training, implementing hemorrhage control and “Stop the Bleed” initiatives, strengthening coordinated evacuation planning, and investing in prehospital trauma infrastructure. Occupational health frameworks for the mining sector should incorporate terrorism-related hazards. Mental health services should be integrated into post-incident response to address the psychological consequences of terrorist attacks.
From a corporate and operational perspective, mining companies should strengthen site security through risk-based investments in measures such as armed facility security, drone surveillance, and resilient supply chain protection in high-risk regions. These investments may not only mitigate operational disruptions and financial losses but also enhance the resilience of the surrounding communities in which mining operations are embedded. Given the transnational nature of contemporary terrorist threats, effective risk mitigation will also require intelligence sharing, technological innovation, and coordinated policy implementation across industry and government stakeholders.Reference Shortland, Forest, JA and AJ10
Limitations
The GTD is regarded as one of the most comprehensive and high-quality datasets on global terrorism events.6 However, there exists a risk of registration bias, where several terrorist events may have been systematically omitted from inclusion into the dataset. Given that media publications are the sole primary sources for inclusion into the GTD, there is concern that lower-quality media outlets in specific geographies may lead to under-reporting or misreporting of terrorist incidents in those areas.
The descriptive study design precludes causal interpretation of findings. There exists potential misclassification in perpetrating group categories, and underlying intent cannot be assessed. All available years were included to capture rare but high-impact attacks on mining. However, trends over time should be interpreted with caution, as there was retrospective data collection from 1998 to 2007, a gap in 2003 data, and incomplete data collection in the earliest decades (1970–1989).6 The GTD does not capture denominator data such as number of mines, workforce size, or regional mining output; therefore, attack rates relative to mining activity could not be calculated or compared across countries and time periods.
Conclusion
The mining sector is a strategic target for terrorist attacks. Mining-related terrorist attacks have increased in recent years and are characterized by high operational success, politically motivated perpetrating groups, and disproportionate lethality in sub-Saharan Africa. South Asia, South America, and Southeast Asia are regions with particularly high numbers of mining-related attacks. Armed assaults and facility/infrastructure attacks were common, and hostage-taking was frequent. Coal and gold were the most targeted minerals. Tailored response strategies, including site security and emergency preparedness, are needed to mitigate the impacts of mining-sector terrorism.
Supplementary material
To view supplementary material for this article, please visit http://doi.org/10.1017/dmp.2026.10409.
Acknowledgments
TSA was supported by grant number T32GM144273 from the National Institute of General Medical Sciences. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of General Medical Sciences or the National Institutes of Health.
Author contribution
TSA contributed to the conception, design, analysis, and drafting of the article. DT and GRC participated in critically revising the manuscript for important intellectual content. GRC supervised and gave final approval of the version to be submitted.
Funding statement
TSA was supported by grant number T32GM144273 from the National Institute of General Medical Sciences. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of General Medical Sciences or the National Institutes of Health.
Competing interests
The authors have no conflicts of interest to disclose.
Ethical approval
IRB approval was not required since our study utilizes publicly available and de-identified data from the Global Terrorism Database.
AI technology
The authors did not use generative AI in this manuscript.
