Water, Energy, Food, and Terrorism: Towards a Nexus Approach

Water, energy, and food are highly interdependent resources fundamental to both the sustainment of life and the modern global economy. While our understanding of their interdependence is hardly new, it was only in 2011 that scholars, practitioners, and policymakers first adopted an explicit water, energy, and food nexus, or “WEF nexus,” approach in an effort to more effectively address the complex issues related to these resources.

Describing the WEF nexus as an approach, however, does not capture the varied ways in which it is employed. Some view it conceptually, while others use it as a framework for policy and resource management. Indeed, this definitional ambivalence plays no small part in the most contentious debates regarding the WEF nexus. It is, as Stephen A. Harwood writes, “clothed in ambiguity.”

Regardless, the interdependent nature of water, energy, and food is well-established. Their significance in our daily lives is self-evident irrespective of variables such as income level or location. They are equally important to the Global North as they are to the South. Therefore, one does not need to commit to any single WEF nexus approach, concept, or framework in order to acknowledge that considering these resources separately is sub-optimal. Yet research in at least one field has remained stubbornly compartmentalized: the relationship between water, energy, food, and terrorism.

A typical study might explore the link between water scarcity and terrorism, for instance, yet 72 percent of the world’s freshwater withdrawals can be attributed to agriculture and 90 percent of power generation is “water-intensive.” It is easy to imagine how a shock to the water supply might impact the ability of local farmers to properly irrigate their land. Therefore, it appears insufficient to explore a link between water scarcity and terrorism absent an analysis of how such scarcity might impact energy and food production, thereby further influencing levels of political violence and terrorism. This is not to say that past and current sectoral focused research lacks value. Rather, the point is that bringing a nexus approach to the study of water, energy, food, and terrorism decompartmentalizes data and clarifies the broader relationship. Such a study is overdue.

Broadly speaking, there are two ways in which we might think of this relationship. The first is expressed in those types of studies previously mentioned, which consider how the resources of the WEF nexus contribute to the conditions necessary for a change in the frequency and/or intensity of political violence broadly and terrorism specifically. Relatedly, much of the research on climate change and terrorism approaches the relationship in this manner, as well.

The second way is essentially the reverse of the first: how does terrorism impact water, energy, and food? This might lead to research that is broadly scoped, such as an exploration of the second- and third-order effects of terrorism on the WEF nexus (e.g. – how the occurrence of terrorism influences the level of displaced persons, thereby impacting local water, energy, and food production). A more specific and narrowly scoped inquiry, which is the focus here, would be to assess how terrorists directly target and attack water, energy, and food resources. This, in other words, is an analysis of the first-order effects of terrorism on the WEF nexus.

We can call this WEF-targeted terrorism and define it as any terrorist incident in which a person, place, or object critical to the extraction, production, transportation, and/or end-user consumption of water, energy, or food is targeted. Following 9/11, interest in this type of relationship grew exponentially and for good reason. It became quite clear that we ought to consider when, where, how, and why terrorists attack water, energy, and food infrastructure. Yet a global analysis of this sort does not yet exist, and the threat is often discussed hypothetically and detached from data.

WEF-Targeted Terrorism: An Initial Survey

This gap in research is worth addressing and with over 210,000 recorded incidents of terrorism between 1970-2020, the University of Maryland’s Global Terrorism Database (GTD) is an excellent place to start. After omitting incidents from 1993 due to insufficient data, as well as filtering out post-1997 incidents for which there is doubt regarding whether they meet the database’s criteria for inclusion, the total number of terrorist incidents between 1970-2020 stands at 174,471.[1]

To identify only incidents that meet the definition of WEF-targeted terrorism, the most relevant data codes in the GTD are those for “target/victim subtype,” which I will refer to as simply target subtype. Of the 113 target subtypes, nine were applicable and can be found in Table 1 below (along with the more broadly defined “target/victim type” that each is associated with).[2]

Table 1

After narrowing the database down to only those incidents of WEF-targeted terrorism, what does the GTD say about when, where, and how it occurs?

To begin, of the 174,471 incidents under consideration between 1970-2020, WEF-targeted terrorism accounted for 5.39 percent of the total, or 9,398 incidents.[3] While it is not a perfect comparison, it is nevertheless instructive to set this figure side-by-side with the percentage of attacks against categories of non-WEF targets. For example, the target type “telecommunications”—under which there is not a single WEF-related subtarget type—accounted for under 1 percent of the total.[4]

Additionally, incidents of WEF-targeted terrorism were damaging to both life and property. These incidents resulted in almost 7,000 fatalities and caused over $1 million in property damage in 153 incidents for a total estimated damage amount of roughly half-a-billion U.S. dollars.[5] Regardless of severity or estimated dollar amount, damage to property was positively identified in roughly 80 percent of cases.

However, it is important to note that WEF-targeted terrorism was less lethal than attacks against other categories of targets. For example, attacks against the target subtypes associated with “transportation” accounted for a smaller share of the total number of incidents yet produced over twice as many fatalities. A likely explanation for this is that many WEF-related targets, such as electrical utilities, are often designed to keep people at a safe distance whereas buses, airplanes, and trains are – if not dedicated to hauling cargo – by nature full of people.

Nevertheless, nearly 7,000 fatalities and hundreds of millions of dollars in property damage can hardly be ignored. This level of damage to life and property is unsurprising given that explosives served as the primary method of attack in nearly three-quarters (71.96 percent) of incidents. Other explicitly violent methods (coded in the GTD as “armed assault,” “assassination,” “facility/infrastructure attack,” and “unarmed assault”) account for an additional nearly 20 percent.[6] According to the data, when terrorists attacked WEF-related targets, they largely did so by applying violence and not simply threatening the use of it.

Again, it is worth noting differences between WEF-targeted terrorism and attacks against all other categories of targets. When non-WEF targets were hit, explosives were the primary method of attack in only 47.77 percent of incidents, roughly 25 percent less than the figure for WEF-targeted terrorism. Additionally, suicide attacks against water, energy, and food are extremely rare: they account for just 0.51 percent of all incidents. In all other cases of terrorism considered in this study, suicide attacks accounted for 3.58 percent.

What begins to emerge from the data is an initial picture of what a “typical” attack against WEF resources looks like: it is an attack utilizing some sort of explosive device that intends to inflict damage primarily to infrastructure, although it is not uncommon for terrorists to target corporate property (i.e. – office buildings) or humans themselves (i.e. – farmers, corporate employees). In addition to the results already surveyed, this archetypal attack is informed by the fact that electricity, oil, and gas utilities accounted for 66.85 percent of all WEF-targeted incidents of terror. When attacks against the water and food supplies are added, this figure is just over 70 percent.

Additionally, explosives served as the primary method of attack in almost exactly 90 percent of incidents against electricity, oil, and gas utilities. Despite accounting for 66.85 percent of all incidents, attacks on utilities produced only 27.71 percent of all fatalities. Again, this is likely due to the fact that these targets are not often densely populated. Table 2 below provides a complete breakdown of WEF-targeted incidents by subtarget type from 1970-2020.

Table 2

The most instructive findings, however, come when the results are analyzed across space and time. Toward that end, we can periodize 1970-2020 as follows: Cold War era (1970-1989, a period of twenty years), post-Cold War era (1990-2001, a period of twelve years), and post-9/11 era (2002-2020, a period of nineteen years). Additionally, the GTD codes each incident for the country and region in which it took place, which facilitates identifying trends across space.[7] When we analyze the dataset through both this periodization and the given regionalization, the results are more useful.[8]

For example, during the Cold War era there were 3,287 incidents of WEF-targeted terrorism that caused 1,575 fatalities, which works out to an average of 164.25 incidents and 78.75 fatalities per year. However, during the post-9/11 era, both numbers were significantly higher at 233.16 incidents of WEF-targeted terrorism per year and an average of 226.74 deaths annually. While the most frequently observed fatality count in every era was zero, WEF-targeted terrorism—mirroring broader trends—grew more lethal over time. As can be seen, during the post-9/11 era there was nearly one fatality for every incident.

Where the attacks occurred also changed over time. In the Cold War and post-Cold War eras, it was distinctly a South American and Central American/Caribbean issue. Of the incidents of WEF-targeted terrorism that occurred from 1970-1989, over 78 percent took place in these two regions alone. However, for non-WEF-targeted terrorism, they accounted for only 40 percent of the total incidents.

The data from the post-9/11 years tell a very different story: the Middle East and North Africa, South Asia, Southeast Asia, and Sub-Saharan Africa accounted for over 80 percent of WEF-targeted terrorism, which is much closer to the 90 percent these regions accounted for of non-WEF-targeted terrorism during this same period. Figures 1.1 and 1.2 below fully illustrate the changes in regional occurrences of WEF- and non-WEF-targeted terrorism.

Figure 1.1
Figure 1.2

We can also analyze how target selection shifted over time. While electricity utilities were the most targeted subtype in any given period, Figure 2 below shows how this changed over the fifty years under consideration.

Figure 2

While these results are informative, and the accompanying dataset provides even further insight, we have not yet come to perhaps the most important question, as well as the most vexing one: Why do terrorists target water, energy, and food? In reality, answering this involves working through a series of more narrowly scoped inquiries. Is there a relationship between terrorism writ-large and the targeting of WEF resources? How does WEF nexus (in)security impact target selection, if at all? How does ideology influence whether terrorists target water, energy, and food? How about location?

Providing a satisfactory answer to these questions and more ought to be a long-term research goal for those interested in gaining a data-driven understanding of the threat of terrorism to our critical resources. That is clearly a task beyond the scope of this initial study. However, even a brief investigation reveals insight. By setting the number of non-WEF-targeted incidents of terrorism per country between 1970-2020 as the independent variable, WEF-targeted incidents of terrorism per country during the same period as the dependent variable, and conducting a simple linear regression test, one can begin to assess if a relationship between the two exists. Put another way, one can test to what degree WEF-targeted terrorism is associated with and explained by all other incidents of terrorism.

The results of this test are intriguing. The correlation coefficient, which indicates the strength and direction of the relationship between two variables, is 0.637. This represents a moderate-to-strong and positive association. In simpler terms, countries with higher numbers of non-WEF-targeted incidents of terrorism tended to experience more WEF-targeted terrorism, as well. However, equally important is the coefficient of determination, which speaks to how well one can explain one variable simply by taking another into account. In this instance, the question becomes: to what degree is the number of WEF-targeted incidents of terrorism in a country explained solely by incidents of non-WEF-targeted terrorism. The test yielded a coefficient of determination of 0.406, meaning that only 40 percent of the variation in WEF-targeted terrorism can be explained by the number of non-WEF-targeted incidents in a country.

Therefore, despite the moderate-to-strong and positive association between the two, we cannot simply explain WEF-targeted terrorism by highlighting the presence of terrorism generally in a country. That only takes us so far. While WEF-targeted terrorism appears to rise with terrorism writ-large, it does so to a greater extent under certain currently unknown conditions.

Clarifying those conditions is critical to mitigating WEF-targeted terrorism and there are several avenues available for future research. First, this study offers only an initial description and examination of WEF-targeted terrorism. Second, given the limited scope, correlative relationships were not addressed in-depth. Future research ought to conduct more rigorous testing and explore additional variables such as ideology or WEF nexus security levels. For instance, it is possible that WEF nexus security, not insecurity, is positively associated with higher incidence rates of WEF-targeted terrorism. Infrastructure must exist for it to serve as a target.


“Progress is simply a fact,” Derek Thompson wrote last year in a piece for The Atlantic. Indeed, in nearly every measurable category of development and human progress, the world is a better place today than at any time before. For example, there are significantly less poor and hungry people globally in 2023 than there were in 1990. Between 2000 and 2020, the maternal mortality rate globally declined by 34 percent. The amount of people living without electricity is down from over 1.3 billion in 2012 to roughly 745 million in 2023. These and more are no small accomplishments.

However, the Food and Agriculture Organization of the United Nations also estimates that by 2050 “agriculture will need to produce almost 50 percent more food, fibre and biofuel than in 2012 to satisfy global demand.” This will potentially require “global water withdrawals 30 percent higher than today.” According to the U.S. Energy Information Administration, global energy consumption “will likely increase through 2050 and outpace advances in energy efficiency.” As more countries successfully climb the development ladder, global consumption will inevitably increase. Given the interdependent nature of water, energy, and food, an increase in the demand for one will necessarily affect the demand and availability of the others.

 One can remain skeptical of these projections yet still recognize that the resources of the WEF nexus are critically important. Securing them is of equal importance. However, the concept of resource security is complex and means different things to different people. Does security refer to access, availability, quality, a combination thereof, or something different altogether?

Regardless of the frame around security, this study ought to have made something abundantly clear: the threat of terrorism against water, energy, and food is real. We no longer need to speak of it in the abstract. Instead, we know that terrorists targeted the resources of the WEF nexus no less than 9,398 times in the fifty years between 1970-2020. In addition, we know these attacks were violent and caused nearly 7,000 deaths and over half-a-billion dollars’ worth of property damage. When considering second- and third-order effects in the mid- to long-term, these figures are likely far higher. For instance, a 2015 attack attributed to the Revolutionary Armed Forces of Colombia, commonly referred to by their Spanish acronym (FARC), left “hundreds of thousands of people without power” and “affected more than 300,000 people.” The GTD does not capture the impact of these longer-term effects.

Finally, we know that over the past roughly fifty years, WEF-targeted terrorism was a persistent issue across space and time. It was not just a Cold War, post-Cold War, or post-9/11 problem. Similarly, it was not isolated to one region. However, it did become more dangerous over time, in terms of fatality and incident rates (both at their highest during the post-9/11 period). While WEF-targeted terrorism accounts for just about 5 percent of all terrorist attacks over the past five decades, it nevertheless ought to demand our attention due the significant role that water, energy, and food play in the sustainment of life. This is particularly true today given that the 21st century will be shaped by issues of competition over the very resources that WEF-targeted terrorism destroys.

By incorporating a nexus approach in research on the relationship between water, energy, food, and terrorism, our understanding of WEF-targeted terrorism will become less compartmentalized and more holistic, helping us produce better informed policy.


[1] For more information on incidents from 1993, as well as the way in which the GTD codes for doubt and uncertainty, one can reference the GTD codebook here.

[2] These target subtype codes were selected because all incidents to which they were assigned are axiomatically WEF-targeted terrorism. In other words, while a small number of incidents of WEF-targeted terrorism in the GTD may not be captured here, we can operate with near certainty that 100 percent of the incidents captured are WEF-targeted terrorism.

[3] Some incidents are assigned multiple target subtypes. This study is only concerned with the first – or primary – target subtype code assigned to each incident.

[4] The target/victim type “telecommunications” has five associated target/victim subtypes: “radio,” “television,” “telephone/telegraph,” “internet infrastructure,” and “multiple telecommunication targets.”

[5] Of these 153 incidents, the GTD only includes damage value estimates for only fifty-five. The total amount as a result of these fifty-five cases is $490,850,010, which makes it likely that the total amount of damage is far greater than $500 million.

[6] For instance, for an incident to be assigned the code “armed assault,” it must be an “attack whose primary objective is to cause physical harm or death directly to human beings…” This can be contrasted with the GTD’s definition of “hijacking,” which is an “act whose primary objective is to take control of a vehicle…”

[7] The GTD’s codebook provides a full breakdown of the countries that make up each region and can be found starting on page 21.

[8] This periodization is hardly unconventional and takes into account commonly used historical inflection points. However, I found the idea for analyzing the data in this way from Dr. Jennifer Vellieux’s and Dr. Shlomi Dinar’s study on water-related terrorism.

Works Cited

Simpson, Gareth B., and Graham P. W. Jewitt. “The Development of the Water-Energy-Food Nexus as a Framework for Achieving Resource Security: A Review.” Frontiers in Environmental Science 7 (February 2019), https://doi.org/10.3389/fenvs.2019.00008.

Hoff, Holger. “Understanding the Nexus.” Stockholm Environment Institute. Accessed December 20, 2023. November 11, 2011. https://www.sei.org/publications/understanding-the-nexus/.

Harwood, Stephen A. “In Search of a (WEF) Nexus Approach.” Environmental Science & Policy 83 (May 2018): 79–85. https://doi.org/10.1016/j.envsci.2018.01.020.

Naylor, Kelly Ann. Blueprint for Acceleration: Sustainable Development Goal 6 Synthesis Report on Water and Sanitation 2023. New York, NY: United Nations, August 2023.

Connor, Richard and Engin Koncagül. The World Water Development Report 2014. Paris, FR: United Nations Educational, Scientific and Cultural Organization, 2014.

National Consortium for the Study of Terrorism and Responses to Terrorism. 2022. Global Terrorism Database 1970 – 2020. https://www.start.umd.edu/gtd.

Thompson, Derek. “The World is Really Getting Better.” The Atlantic, September 13, 2022. https://www.theatlantic.com/newsletters/archive/2022/09/bill-melinda-gates-foundation-goalkeepers-report-poverty/671415/

UNICEF. “Maternal Mortality – UNICEF DATA.” Accessed December 23, 2023. February 2023. https://data.unicef.org/topic/maternal-health/maternal-mortality/.

Cozzi, Laura, Daniel Wetzel, Gianluca Tonolo, Nouhoun Diarra, and Arthur Rogue. “Access to Electricity Improves Slightly in 2023, but Still far from the Pace Needed to Meet SDG7 – Analysis.” International Energy Agency. September 15, 2023. https://www.iea.org/commentaries/access-to-electricity-improves-slightly-in-2023-but-still-far-from-the-pace-needed-to-meet-sdg7.

United Nations, Food and Agriculture Organization. The State of the World’s Land and Water Resources for Food and Agriculture – Systems at breaking point. Main report. Rome, IT: FAO, 2021.

Michel, David. “Water and Food: How, When, and Why Water Imperils Global Food Security.” Center for Strategic and International Studies, October 16, 2023. https://www.csis.org/analysis/water-and-food-how-when-and-why-water-imperils-global-food-security.

Sanicola, Laura, Stephanie Kelly, Laura Sanicola, and Stephanie Kelly. “Global Energy Consumption to Increase through 2050, Outpace Efficiency Gains, EIA Says.” Reuters, October 11, 2023. https://www.reuters.com/business/energy/global-energy-consumption-increase-through-2050-outpace-efficiency-gains-eia-2023-10-11/.Garcia, Cesar. “Rebel Attack in Colombia Leaves 300,000 in the Dark.” Associated Press, June 11, 2015. https://apnews.com/article/10ff1a0425214e9d81cdf6ebad3f5a43.

Benjamin Levine
Benjamin Levine

Benjamin Levine is a first-year M.A., International Relations Candidate concentrating in Security, Strategy, and Statecraft at Johns Hopkins University's School of Advanced International Studies. Prior to SAIS, he served for over eight years as an Officer in the United States Army. Most recently he was selected to serve as a Google Public Policy Fellow at TechNet: The Voice of the Innovation Economy where he focused on federal policy regarding advanced and emerging technologies. Benjamin holds a B.A. in History and Politics from Drake University.