Water-related conflicts: definitions, data, and trends from the water conflict chronology

Violence associated with freshwater resources has occurred throughout recorded history, with water triggering violence and armed conflict, water or water systems being used as weapons, and water or water systems becoming casualties during conflicts. Understanding the causes of water-related violence and regional and temporal trends is critical for identifying priority areas for conflict resolution and strategies to reduce the risk of future conflicts. The purpose of this paper is to introduce the structure, content, and design of The Water Conflict Chronology, an open-source online database, which tracks water-related conflicts from around the globe. Some analysis is provided with caveats about the limitations of the data, but further analysis will be forthcoming. The database is updated approximately annually using information from other conflict-related databases, news reports, eyewitness accounts, and a review of historical documents. As of October 2022, the Chronology had 1298 entries, from the earliest events around 2400 BCE through early 2022. Initial analysis of data shows trends in the nature, location, and social and political characteristics of water-related violence, including a sharp increase in water-related violence in recent years, especially where water has been a trigger or casualty of violence, a concentration of events in the Middle East, southern Asia, and the Sahel, and the increased targeting of civilian water systems during civil and regional armed conflicts.


Introduction
Not every resource issue or environmental problem is a threat to national or international security, but for several decades, it has been increasingly apparent that certain regional and global environmental problems produce conditions that enhance the risk of violence or even trigger violence. In the approach to security studies that dominated in the mid-to latetwentieth century, economic factors were secondary considerations and environmental or resource factors were largely ignored in favor of 'realist' or 'realpolitik' approaches to international relationships, where the conventional meaning of security was 'the defense of territorial and political integrity… as the fundamental, the immutable objective of states in the international system' [1,2].
By the 1980s, however, in parallel with the shift in the relationships between the traditional superpowers of the United States and the weakening Soviet Union, there was growing evidence of regional and global environmental and resource disruptions, including the depletion of Antarctic ozone, transboundary acid rain, depletion and contamination of international rivers, and anthropogenic climate change, and the realization that these problems required international cooperation and in some cases were increasingly affecting international tensions and security [3][4][5]. This led to a dramatic expansion of efforts to understand and study environment and resource issues that transcended traditional political borders and to develop concepts for integrating those issues into geopolitical studies and practice-what became the expanding field of environmental security [6][7][8][9]. The field has also benefited from efforts to develop policies of transboundary cooperation and negotiation, making it even more important to identify remaining triggers for violence [10,11].
Three related, but distinct aspects of security studies in this area are (a) evaluating the role that access to, control of, or a change in natural resources plays, as one of many variables, as a trigger of violence or armed conflict; (b) identifying where resources or the environment are used as military tools or weapons of violence; and (c) assessing the environmental consequences of war or violent conflict where resources or the environment are casualties or targets of violence.
Early efforts in the area of environmental security focused on the first category with particular attention to the role that access to energy resources, especially oil, plays in influencing the security objectives of states because of both the central part it plays in global economies and the massive and vulnerable international trade required to move oil from a small number of producers to a much larger number of consumers [12,13]. These factors have made oilproducing states, especially the Middle East, a locus for violence and conflict internally and involving powerful external players for much of the latter half of the 20th century. In this case, resources are strategic goals and triggers to violence, critical to economic prosperity and political stability.
States go to war for many reasons, including both material and ideological ones, and it is critical to understand the links between them. Resources are unevenly distributed around the world, including classic mineral resources such as precious or rare metals, energy resources such as fossil fuels, and even renewable energy resources like wind, solar, and freshwater. Lipschutz argued in the late 1980s that post-Cold War foreign policies should be understood as a synthesis of ideology and material interests [14]. As acknowledged by the U.S. military of the time, resource issues were already playing a role in superpower politics, triggering competition and conflict: "US and Soviet interests focus to a large degree on the less-developed countries, many of which are resources rich.. The resources supply patterns.. demonstrate why access to these regions is important to the US and its allies, and how the Soviet Union, by intrusion therein, gains political and economic leverage against the West." [15].
A major shift in recent years has been the recognition that the availability of local substitutes for traditional fossil fuel resources, such as renewable energy, can greatly reduce the risk of energy-related conflicts, while constraints on natural resources for which there are no substitutes, such as water, may pose a greater threat to peace and stability [16].
The second area of concern is the targeting of strategic resources, including energy or water distribution systems, during conflicts that may start for other reasons, an action that violates the Geneva Principles on the Protection of Water Infrastructure and the 1977 supplemental protocols to the 1949 Geneva Convention [17]. Destroying these resources, or denying access, to them (water as a 'casualty' of conflict), is often considered a strategic military objective because of the multiple roles they play in power dynamics, military operations, and in supporting national economic well-being. Saddam Hussein targeted the oil production infrastructure of Kuwait during the Gulf War [18,19]. Access to and control of the nuclear power plants in Ukraine have been a focus of the war there since early 2022. The targeting of water and water systems also has a long history. Large hydroelectric dams were regularly attacked during World War II; civilian water systems in Yemen, Syria, and Iraq were persistently attacked in the late 2010s [20,21]. During periods of colonization, natural resources were often targeted (or used as weapons of war-the next category) with water resources being appropriated or poisoned. Examples from the database include water rights violence against both Native American Tribes and early Hispanic communities in the southwestern US in the 1800s and wells poisoned during slave revolts in the Caribbean [22,23].
The third category is the use of resources or the environment as 'weapons' of conflicts. This includes non-military tools such as trade embargoes, sanctions, or the direct manipulation of environmental factors in conflicts or wars that again, may start for traditional political, economic, or ideological reasons. After the invasion of Afghanistan by the Soviet Union, the US imposed a grain embargo on the Soviet Union in 1980, prompting debate about the practical and ethical implications of using food as a weapon [24]. Saddam Hussein intentionally released oil into the Persian Gulf to try to cripple Kuwaiti desalination plants during the Gulf War, leading to calls for international treaties to prohibit intentional ecological destruction as a weapon of war [25,26]. The control and manipulation of the flow of oil and natural gas from Russia to Europe has played a key role in dynamics of the European response to the 2022 Russia-Ukraine war [27].
While energy resources have been an early focus of much of the environmental security literature, the link between energy and security are weakening as pressures to cut the use of fossil fuels to address the growing risk of climate change has led to a massive expansion in the availability and use of renewable energy that can be produced locally. Fossil fuels are likely to continue to play a major role in international trade and economics and to influence political and strategic actions by major economic players dependent on imports of oil, but other resourcesespecially freshwater-are now increasingly tied to environmental security challenges.
Water is critical for all aspects of national economies, the production of goods and services, growing food, energy production, and the health of both humans and natural ecosystems. Fresh water is also unevenly distributed around the world, with natural variability in the stocks and flows of the hydrologic cycle [28]. These two factors make access to and control of water a top social and political objective, as well as a strategic objective during conflicts. As such, the links between water and conflict are becoming central to concerns about environmental security [11,29,30]. To investigate these links, the Pacific Institute, an independent water research institute in Oakland, California, began collecting data on water-related violence more than 35 years ago and created an open-source database-the Water Conflict Chronology (WCC)-to categorize and analyze water conflicts [31]. Below we introduce the structure, content, and design of the WCC, provide an initial review how water has been a factor in conflicts, explore the way water-related conflicts have changed over time and location, and identify some initial trends in water-related violence that deserve attention by decision makers and policy actors in the environmental security sphere.

The data: the WCC
The WCC is an online, open-source database that includes historical information on violence over water categorized by region and type, conflict time and date, locale, and other variables. An interactive map is also presented that shows the geographic location of and information about each event. Entries are collected from a wide variety of sources, including historical records, newspaper and magazine reporting, military reports, international governmental and non-governmental organizations, social media, and field work. Internet searches on terms such as 'conflict database' and 'war database' were used to discover online repositories of violent conflict events that were related to or involved fresh water. For every entry in the Chronology, a detailed source or reference is included. Table 1 lists a set of online databases related to armed conflict, terrorism, or regional violence that were reviewed for entries involving water resources, water systems, and people involved in the conflict, such as the workers. These sources were typically created for other (non-water) purposes and their entries were reviewed and evaluated for inclusion in the WCC. Additional sources beyond those listed in the table below will be evaluated for inclusion in the future, including other data sets and media archives, as time and resources allow. For a water-related conflict event that we identified from one of these databases, we have cited the database itself as the source within the WCC.

Event categories and selection
Events are included in the database when there is armed conflict or physical violence related to water systems or water resources involving injuries or deaths, or threats of violence, including verbal threats, military maneuvers, and shows of force. Events are only included if the harm caused is or would have been immediate. Instances are excluded if unintentional or not directly physically harmful, such as impacts to individuals or communities associated with water-management decisions (e.g. populations displaced by dam construction). If, however, a watermanagement decision leads to a violent response, such as violent protests of water restrictions, then the event is included. Events are also excluded when the physical impacts are caused by weather or climate events such as flooding or droughts, and not by humans or direct, human-created activity. Events where water resources or water systems are incidental and not the defining factor, for example, the extensive use of water cannons as a tool to suppress social unrest and riots, are not included. All entries are categorized in groups that parallel the framework described above for broader environmental security discussions, specifically water as a trigger of conflict; water as a military tool or weapon of conflict; and attacks on water or water systems, i.e. casualties of violence. Table 2 summarizes the definitions of these categories and provides selected examples.

Trigger
Water is considered a trigger or root cause of conflict where there is a dispute over the control of water or water systems, or where economic or physical access to water, or scarcity of water, triggers violence. Thousands of people have been killed in recent years over competition for water and land in the Sahel where farmers and nomadic pastoralists clash over access to watering holes and land. Growing use of fences around land have reduced mobility and options for nomadic herders, and changes in climate and water conditions have forced herders further south in search of water, further sowing divisions and igniting religious extremism [32]. While some efforts have been made to reduce tensions over rural-urban water disputes, with negotiations and compensation, this area remains a key one for violence [33].

Weapon
Water or water infrastructure have also been used as intentional 'weapons' during conflicts. Water has been weaponized to poison or cut the supply of water to vulnerable populations or to flood areas for strategic military purposes (see table 2). More than 30 people were reported killed in Somalia in 2017 by water poisoned by al Shabaab, a militant group affiliated with Al-Qaeda [34]. In 2022 in the Russian-Ukrainian war, Ukraine intentionally released water from a dam on the Dnieper River, flooding land

Category Description Examples
Trigger Water as a trigger or root cause of conflict, where there is a dispute over the control of water or water systems or where economic or physical access to water, or scarcity of water, triggers violence.
The water dispute between the Mesopotamian cities of Umma and Lagash begins around 2450 BCE and lasts a century. It includes fights over the control of irrigation water ("trigger") and cut offs of water supply ("weapon") [57]. In 2017, fighting between two clans near Lagawa, Sudan over ownership of a water point leaves six dead and others wounded [39]. Weapon Water as a weapon of conflict, where water resources, or water systems themselves, are used as a tool or weapon in a violent conflict.
From 720 to 705 BCE, King Sargon II of Assyria and his opponents both use water as a "weapon" and a target, damaging wells, diverting and destroying irrigation canals, and flooding lands for offensive and defensive purposes. (Also categorized as "casualty") [57,58].
In 2020, an armed group attacked a control station at Shwerif, Libya, cutting off the water supply to more than 2 million people in the capital Tripoli [59]. Casualty Water resources or water systems as casualty of conflict, where water resources, water systems, or people involved with the resource or the system, such as employees, are intentional or incidental casualties or targets of violence.
During the 1991 Gulf War, attacks on water treatment plants leave whole cities, such as Basra, without water or wastewater treatment [60]. In January 2022, Russian warplanes conducted airstrikes on the water system of the town of Sijer-Bqesemtoh, near Idlib, Syria, rendering the system inoperable [39].
north of Kyiv to slow the advance of Russian armored columns [35].

Casualty
Water resources or water systems that provide water or wastewater services are 'casualties of conflict' where water resources, water systems, or people involved with the resource or the system, such as employees, are intentional or incidental casualties or targets of violence (see table 2). Attacks on civilian infrastructure, including water systems, violate longstanding international codes of conflict, yet these events still occur.

Additional factors related to data characterization
In some instances, more than one category may apply, such as where water scarcity may trigger violence during riots over access to water, but where water systems are also attacked (i.e. 'casualty'). In such cases, the WCC database categorizes the event under all applicable categories. In some instances, it can be challenging to distinguish between whether a weaponized form of water or water system also creates a casualty (of the resource or infrastructure). For example, in 1978, a water tank was poisoned in protest of airport construction in Japan; assuming that the poisoning was done with the intent to harm one or multiple people, this is listed as a weapon and a casualty as the water supply was no longer usable [36]. In 1938, Chiang Kai-shek ordered the destruction of flood control dikes along the Huang He (Yellow) River to flood areas threatened by the invading Japanese army. The resulting floodwaters covered an area variously estimated at between 3000 and 50 000 km 2 , and while it delayed the Japanese, it also killed Chinese estimated in numbers between 'tens of thousands' and 'one million' [37,38]. This is also listed as a weapon and a casualty. In 2021, an unidentified group attacked a water tanker and killed four people delivering water to pastoralists affected by drought near Buulo Xaaji village, Lower Juba, Somalia [39]. In this instance, the drought was a trigger contributing to a heightened levels of conflict over water, and the water system and four people killed were casualties in the conflict. While most entries in the WCC include violence-the death or injuries of parties involved and/or destruction of infrastructure-a few instances of threats of violence are also included where there were plans of violence that were thwarted by law enforcement or where the threat was sufficiently serious to merit a reaction by authorities, including cyberattacks of computerized operating systems for dams, drinking water systems, or wastewater treatment plants. For example, in 1985, law enforcement authorities disrupted the plan of a religious cult to poison the water supplies of New York, Chicago, and Washington D.C [40]. In the early 2000s, a series of threats to poison water supplies were broadcast by Islamic fundamentalists [41]. In 2020, Israeli authorities thwarted an attempted cyberattack aimed at disrupting water supplies in at least two locations in Israel, linking those attacks to Iranian hackers [42].

Location and timing of events
To map the entries, Google Earth Pro is used to identify event locations by their latitude and longitude. Where a specific location is identified, such as a dam or building, the precise coordinates are used. When information on only the general area or a city or a nation was available, the mapping system's preset place location is recorded, typically in the centroid of that location. Many locally used place names are not captured by geographic information systems such as Google Earth Pro, and/or by the western, Englishspeaking community in which this chronology was created, likely reducing the accuracy of some assigned locations. Sometimes events occur over multiple days or longer or extend over a wide area, complicating both dating and locating an event. For these events, a generalized location representing a single location may be used. Earlier historical incidents that persisted over time are more difficult to place compared to more current events. For example, the 'Fence Wars' in the United States represent more than a decade of violence over water and land in the 1870s-1890s, but we have only one well-documented entry for the conflict. In this instance, there is a single point placed in the general region where the wars were known to have occurred. Conversely, the availability of more detailed information and reporting in recent years produces many entries for a single, multiday conflict like the war in Yemen, where individual attacks on urban water systems can be fairly precisely located.
Events were 'lumped' into a single entry if the timing, event description, parties involved, and geographic location were sufficiently close in time to suggest that they were the same event or sufficiently connected to be part of a related action. This kind of lumping and splitting is also subject to the challenges of identifying older entries. For example, the earliest major water conflict-the water war between Umma and Lagash in ancient Mesopotamia-certainly involved multiple events spread out over almost a century and many generations, but only one entry is included in the Chronology. Comparable violence today over such an extended period would result in numerous separate entries.
Each entry in the Chronology has a field listing the country or countries where the event occurred, with multiple countries listed when the event includes a transboundary action or more than one national party. The location where the event occurred is used to assign the event to a region based on the UN Statistics Division M49 regional classifications (shown in table 3). Each event was assigned to a Level II region.

Sources of uncertainty
There are several kinds of uncertainty associated with conflict databases and the WCC and researchers should be aware of them and use caution in using the data. Here we briefly discuss the diverse sources of uncertainty.

Reporting biases
The most important reporting bias is the dramatic increase in the ability to collect information on waterrelated violence in the past several decades associated with access to real-time reporting, remote sensing, social media accounts, and access to digital resources. Older entries typically depended on historical accounts in books, news reports, and written records. Almost certainly, earlier instances of smallerscale water-related violence were not recorded, or earlier entries are not easily accessible from stored archives. An example may be the relative paucity of entries from the First World War and World War II, when there were certainly many more attacks on water infrastructure, intentional or not, compared to the larger number of entries from the recent war in Yemen and the conflict with the Islamic State.
While we search for and include entries from as many sources as possible, collection methods may undercount or underreport events that are reported in non-English speaking sources. Recent online translation services have reduced the likelihood of missing, or misinterpreting these events, but we note a language bias to the entries. Interpretation of events, especially attacks on water systems, is also vulnerable to the biases of those reporting such events. For example, attacks on civilian water systems during conflicts may be intentional or accidental but determining intentionality can be subjective. Improving the ability to determine intentionality is important from a policy perspective, since intentional attacks on civilian water infrastructure are explicitly prohibited by international laws of war, including the 1977 Protocols to the 1949 Geneva Convention [43].
Another important bias in the data is the lack of documentation of violence against women and girls while collecting water. Globally, women and girls bear the disproportionate burden of collecting, distributing, and using water at the household level, which has been linked to experiences of physical and sexual violence [44]. Traditional sources of media used for conflict research, such as terrorism datasets, fail to capture this type of household-level violence. The UNDP has also found that survey instruments tracking corruption and related impacts fail to track gender-related factors and that much of the violence against women in low socio-economic classes responsible for water access at the household scale goes unreported [45]. This bias is also likely linked to under-reporting by the victims themselves who may lack the rights of protection against violence, depending on their country, or fear more the outcome from reporting than the incident itself [46]. As improvements in the area of gender-based water violence are made, the WCC will be updated.

Number of events/entries
The number of events included in the Chronology, or any conflict database, depends on two key factors: (a) the quality and accuracy of reports; and (b) the definitions used to determine if an event should be included. Methods of reporting water-related violence have changed over centuries, and especially dramatically in recent decades. The well-documented multi-generation water conflict in ancient Mesopotamia between the city-states of Lagash and Umma was described in several engraved stele that survived for 4500 years and were uncovered in archeological digs. While these historical records offer some insights into early examples of water-related violence, the paucity of these records and the almost certain loss of records over thousands of years make it highly likely that other violence over water occurred but that records of those events have not survived to present day.
Over the past century, reporting strategies and technologies have permitted more global assessments, analysis, and preservation of news and events in the form of newspaper, radio, and television, though even today, what events are reported remains subjective and inconsistent. For example, while several examples of military attacks on water systems during World War II and other conflicts have been reported and are included in the Chronology, underreporting of such events seems certain. Most recently, new tools of social media, global news coverage, and the ability of individuals with cell phone cameras, video capability, and real-time streaming means that even events that may have previously been considered 'not newsworthy' can be reported, studied, and included. These differences over time should be factored into any attempt to quantitatively assess the significance of the recent dramatic trend of increasing water-related violence (see figure 1). As a result, we report here only trends from the most recent periods when reporting has been more consistent.

Results: data analysis and trends
The entries in the WCC can be analyzed for a range of trends, and future research will expand on this area. Below we provide an initial synthesis of the data as of October 2022, assessing some trends over time and looking at the types and regional distribution of the entries. We limit this analysis to the period from 2000 to 2021 because of concern that data availability and reporting methods prior to this period were less comprehensive and consistent, as noted earlier. Figure 1 shows the trends in both the total number of water-related conflicts from 2000 through 2021 as well as trends in the categories of events. For events that fall in two or more categories, they are included in all applicable categories, so the total number of events in the figure is larger than the actual number of events during this period. These data suggest a large increase in the number of events over the past decade, due to the extensive violence in the Middle East in Syria, Iraq, and Yemen, but also with a growing number of events associated with drought and water scarcity in places like Iran, India, and sub-Saharan Africa. Over the second half of the past 22 years, the number of reported events is approximately four times larger than during the 2000-2011 period. Figure 1 also shows a bifurcation in growth rates of event categories, with the number of events where water was a trigger or casualty generally growing beginning in 2012, while the number of events where water was used as a weapon has essentially remained flat or even decreased in this same timeframe. The largest number of entries in the complete database are those where water or water systems are casualties or targets of conflicts; the second largest number involves violence triggered by water scarcity or disputes over access to or control of water. Events where water is used as a weapon were the least common. Despite reporting biases that may favor more recent events, the fact that water-related conflict where water was a trigger and water or water systems were casualties have greatly increased in number while the instances where water has been used as a weapon has stayed constant or even decreased suggests that this observed trend is real. Table 4 summarizes the number in each category for the full database.
For the period 2000-2021, water-related violence has appeared to shift to an increasingly large proportion of conflicts when water was a trigger. From the early 2000s through the mid-2010s water-related violence was proportionally focused on events where water or water systems were a casualty of conflict with nearly all years from 2006 to 2018 having more than 50% of the events in this category. Starting in 2019, more than 60% of the water-related conflict events involved water as a trigger. Since 2011, during all but 2014, water as a weapon remained under 10% of the total number of water-related conflicts. Additional work is needed to understand the drivers of these shifts, such as increasing pressures on water resources due to population growth and climate change, economics and equity, or other factors.
For the full database, figure 2 shows the breakdown of the location of all events through early 2022. By far the largest number of events are recorded in Western Asia, including the countries of the Middle East, Gulf States, and in Sub-Saharan Africa, particularly the Sahel Region. A note of caution: the spatial areas and the populations of the UN regions are vastly different: the Australia and New Zealand region encompasses just those two countries. 'Sub-Saharan Africa' includes 53 different countries.
Western Asia, where the WCC has tracked 380 total events, also has the highest number of events where water is a casualty of conflict. Figure 3 shows water conflicts as a function of type of conflict and key regions. In the four other regions with the highest number of water-related conflict events, water as a trigger is the most common category. This outcome is largely the result of the large number of events from the wars in Yemen, Syria, and Iraq over the past decade (see, for example, the comprehensive reporting from the Yemen Data Project).

Discussion and future work
The WCC database is designed to serve as the basis for ongoing research on the risks of water-related violence, and improvements and extensions will continue. Some efforts have already been made to use WCC data to help evaluate these risks and to identify strategies for reducing those risks, but as noted, the complexity of definitions and limitations in conflict data reporting demand caution in assessing trends. Key gaps and uncertainties include reporting biases by region and language, lack of documentation around gender-related water violence, and better reporting of newer events.
As noted in the section 2, efforts are underway to collect additional data, including older historical data, from a range of records, peer-reviewed research, other conflict databases, and media reports. The database will be updated on a regular basis as new information is processed. Additional analysis is also underway, including comparing trends in water-related violence with overall trends in conflict to identify if water resource conflicts are disproportionately represented, assessing detailed causal information to help in further identifying possible trends, and evaluating the nature of water problems that lead to conflict rather than cooperation. The top five regions based on total number of water-related conflict events for the full database, with the total number of identified events broken out by category: weapon, casualty, and trigger. Note: events in two or more categories are counted in each relevant category such that the total number of events in the WCC is smaller than the total of the rows above.
Recent work by the Water Peace and Security Partnership (an international collaboration founded in 2018 among researchers in several countries, including the Netherlands Ministry of Foreign Affairs, IHE Delft, the World Resources Institute, Deltares, the Hague Centre for Strategic Studies, the Pacific Institute, and others) included an assessment of four categories of strategies to reduce water conflict risks: natural resources, science and engineering approaches, political and legal tools, and financial and economic strategies. The report also included six in-depth case studies from the African Sahel, Yemen, Iraq, India, Iran, and Central America with recommendations for audiences from the globaldevelopment, diplomatic, disaster-response and water-management communities [32]. The WCC, including future analysis that it enables, will be used to advance solutions to help prevent and mitigate water-related conflict.

Conclusions
Social, economic, and political challenges associated with freshwater resources pose a variety of severe risks to communities around the world, from water-related diseases, to crop failures, to ecological destruction, to actual violence. The risks and incidences of waterrelated conflicts in recent years, when data are more available and consistent, as tracked by the WCC-an open-source database on water violence-are on the rise, and the factors driving such violence appear to be worsening. Additional research is needed to assess the role of key drivers of water-related violence, such as growing populations and economies putting more pressure on fixed water resources; accelerating climate changes that are worsening extreme hydrologic events such as droughts and floods; poor, weak, or corrupt water management and institutions; and increasing targeting of civilian water infrastructure in conflicts that start for other, non-resource-related reasons.
Data from the Chronology indicates that the frequency of water-related conflicts has grown in the past two decades, especially as a result of violence in the Middle East, growing disputes during severe droughts over access to water in regions like India and Iran, and worsening confrontations between nomadic pastoralists and farmers in sub-Saharan Africa. The database entries are classified in three categories: water as a trigger of conflict; water or water systems used as weapons of conflicts; and water or water systems that are casualties or targets of conflicts. The bulk of the entries are in the first and third of these groups. Additional work is underway to improve the historical coverage of the data, to identify additional forms of water-related violent conflict, to analyze the trends identified relative to conflict more generally, and to evaluate the drivers for each event. By understanding the root causes of water conflicts, more effective strategies for reducing their probability and consequences can be developed and implemented.

Data availability statement
The data used for this analysis come from the Water Conflict Chronology, created and maintained by the Pacific Institute at www.worldwater.org/waterconflict/. The specific data used in the current analysis can be found at: Gleick, Peter; Shimabuku, Morgan (2022), 'Water Conflict Chronology Data' , Mendeley Data, V1, doi: 10.17632/wjhw4xxgbr.1.
The data that support the findings of this study are openly available at the following URL/DOI: 10.17632/ wjhw4xxgbr.1.