Reflection of Intercontinental Freshwater Resources on Geopolitical Risks: Time Series Analysis

Abstract

Water, an indispensable resource for life and not a complete substitute, is indispensable for energy production, industry, agriculture, and ecosystem sustainability. In particular, the limited and unequal distribution of freshwater reserves makes water a strategic power element on a global scale, making competition inevitable. Increasing water demand and decreasing water resources increase regional and global security risks, causing water to go beyond being a vital natural resource and become a determining factor in diplomacy, conflict, and the balance of power. This study aimed to examine the relationship between freshwater resources and geopolitical risk between 1961 and 2021 using the ARDL model. All models had long-run relationships between water resources and geopolitical risk. In the long-run, a 1% decrease in water resources increased geopolitical risk by 0.37% in Chile, 0.30% in Colombia, 0.46% in the Netherlands, 0.42% in Thailand, 0.44% in Ukraine, and 0.29% in Venezuela. The adjustment rates for the prior period imbalances were estimated to be 0.75% in Switzerland, 0.68% in Chile, 0.28% in Colombia, 0.45% in the Netherlands, 0.86% in Thailand, 0.14% in Ukraine, and 0.59% in Venezuela.

1. Introduction

The world, which is made of more than two-thirds water, consists of only 2.5% fresh water. The ratio of easily accessible earth water is 1.2%. The share of rivers in the waters of the Earth, consisting of rivers, lakes, wetlands, and swamps, is 2% (share in freshwater) [1]. Although the share of rivers in the total fresh water is relatively low, drinking water is extremely valuable in terms of drinking water, agriculture, industry, hydroelectricity, ecosystem, and boundary water flow properties. The fact that rivers are a part of the water supply and a border between countries increases the importance of being a geopolitical risk element.

The establishment of civilizations around rivers also highlights the importance of freshwater resources. For example, Egyptian civilization was around the Nile, and Mesopotamian civilization was around the Euphrates and Tigris rivers. It can be said that rivers, which have been a vital source of transportation, trade, and agriculture since ancient times, have played a similar role to today’s highways. Delli Priscoli (1998) stated that water is “a training ground for civilization” [2,3]. This serves as a critical way to combine people, facilitate transportation, and support the economy. Water paths such as the Gangji River in India, the Yellow River in China, and the Meander River in Ancient Greece are the centers of daily life and economic activities.

A vital source, water, and water resources, as well as geopolitical risks, are essential. Water has caused many conflicts and has been a strategic power throughout history. In ancient Mesopotamia, the Sumerians and Akkadians fought for irrigation channels. The Nile has caused tensions between Egypt and Ethiopia. In Central Asia, disputes exist between the Amu Darya and the Siri Darya. Israel, Palestine, Jordan, and Syria have continued to compete for water resources in the Middle East. These examples show that water is not only a natural source but also a geopolitical tool. Although no war on water resources is so essential in every respect, many regional disputes, conflicts, and tensions are experienced. Water resources often arise as a means of tension, and sometimes as a result. The fact that water resources have become a result of conflict has been clearly observed in the Syrian Civil War. The destruction of infrastructure and lack of water resources during the war accelerated internal migration and deepened the human crisis. This decrease in water access adversely affects agricultural production and daily life and causes people to migrate to large cities and border regions. This deepened regional instability by increasing economic and social pressure in Syria and neighboring countries. Water resources have become an important part of the disputes between Israel and Palestine. The control of water in the West Bank is an example of long-standing tensions between parties.

Another example is that in the 17th and 18th centuries, the Dutch deliberately left strategic areas underwater to defend themselves against Spanish and French invasions [4]. This method is an important example of using water as a means of defense to slow the enemy’s progress. During World War II, the British bombed dams in Germany’s Ruhr region, causing floods and disrupting industrial production [5]. These examples show that water is not only the cause of conflict but also a strategic weapon in the war.

Currently, despite the development of modern transportation technologies, the convenience of rivers remains valid. International waterways are important routes and among the most important arteries in world trade. For example, transportation on the Mekong River, which passes through six countries (China, Burma, Laos, Thailand, Cambodia, and Vietnam), is one factor that increases Southeast Asia’s economic dynamism [6].

Because of water, although there were tensions and conflicts among the countries, there were no wars. However, this does not imply that there will be no future wars. Factors such as global climate change, population growth, and increased water demand increase the likelihood of tension over future water resources.

Many studies have shown that tensions, conflicts, and disputes over water resources may occur and/or increase. Gleick (1993) stated that the Nile River may be the center of conflicts and tensions; Westing (1986) stated that there may be disputes and even water wars for freshwater resources; and Remans (1995) stated that water resources are at risk in the Middle East, South Asia, and South America [7,8,9]. Homer-Dixon (1994) emphasized that rivers can cause natural resource wars between states [10]. Wolf (1998) suggested that a possible water war has turned into a dispute between Lagos and Ummah on the Tigris-Euphrates coast [11,12]. Wolf et al. (2003) stated that the combination of changes in water resources and conflict may make tomorrow’s water disputes look very different from today’s [11].

After winning his second presidential election in 2024, Donald Trump reiterated his desire to include Greenland in the United States. It is noteworthy that the United States has Greenland, and controlling it is absolutely necessary for worldwide national security and freedom.

Greenland, which has large freshwater reserves, can be considered a strategic region for future global water-scarcity scenarios. Trump’s interest in this region was not limited to its geopolitical location. Due to climate change, new paths created by melting glaciers will create economic opportunities. This situation shows that water may become an essential natural resource in international disputes in the future. In particular, the potential for increasing tensions between countries experiencing water scarcity and those with rich water resources increases the risk of possible water wars. It is increasingly likely that regions with strategic freshwater reserves, such as Greenland, will attract more attention from the global powers. Water will be used as a tool for geopolitical pressure in the coming years, particularly in regions with limited water resources. Therefore, Trump’s interest in Greenland should be evaluated not only as an economic or military issue but also from a global water security perspective. Such initiatives regarding access to water resources may trigger future diplomatic crises related to water and potential conflicts between the major powers.

• Why Is This a Problem?

Water is a basic resource for all life, but it is rapidly decreasing due to limited use, misuse, and significant threatening factors such as climate change. A decrease in water resources poses environmental and strategic risks. Access to and control of water can profoundly affect relations between countries, lead to conflicts, and cause geopolitical risk.

It is essential to examine the concepts of water resources and geopolitical risk, which are important in almost every field, together on a global scale, and conduct econometric analyses to understand the relationships between these two factors better and develop effective strategies. However, the fact that no studies have yet been conducted on this scale in this field reveals the need for research on the subject.

• What Are the Differences and Contributions of the Study?

Instead of a single country, analyses covering more than one were conducted. However, instead of addressing countries as a whole using panel analysis, each country was examined in line with its dynamics and characteristics. Analyses were performed separately for each country and evaluated nationally in line with the results obtained. Since the dataset started in 1961 and was monthly, analyses were estimated using high-frequency data.

Although some studies have examined water resources and geopolitical risks separately with different variables, no study has been found to address these two variables within this framework. In this context, this study has not been conducted before in this dimension and will significantly contribute to the literature with the above-mentioned features.

• What Were the Aims and Objectives of This Study?

The primary purpose of this study is to econometrically analyze the relationship between water resources and geopolitical risk across continents using the ARDL model. It aims to analyze how changes in water resources affect the geopolitical risks of countries in the short and long-run and to investigate the cointegration relationship between them. Another goal of this study is to analyze the long-run adjustment process of shocks in the short-run to balance, and how long it will take to restore balance. It also aimed to examine the disagreements and conflicts that the countries included in the analysis experienced in the past.

This paper consists of six sections. After the introduction, the second section describes the disagreements and conflicts experienced by the countries. The third section includes the econometric analyses used in the study. The fourth section consists of the ARDL estimation results. The fifth section contains the results, and the sixth section presents recommendations and policy implications.

2. Freshwater Resources of the Countries and Their Courses over Time

This section presents freshwater resources and their places in world rankings for countries included in the analysis, and significant figures in the world rankings. Furthermore, changes over the years and the relative proportions of the countries included in the analysis compared to other countries were examined.

Table 1. Renewable freshwater resources, total (billion cubic meters).

Rank	Countries	Freshwater Resources	%	Rank	Countries	Freshwater Resources Per Capita
1	Brazil	5661	13.22	1	Iceland	456,351
2	Russia	4312	10.07	2	Guyana	295,531
3	Canada	2850	6.66	3	Suriname	160,221
4	United States	2818	6.58	4	Caribbean small states	114,705
5	China	2813	6.57	5	Bhutan	100,588
6	Colombia	2145	5.01	6	Papua New Guinea	79,997
7	Indonesia	2019	4.72	7	Canada	74,530
8	South Asia	1982	4.63	8	Norway	70,633
9	Africa Western and Central	1898	4.43	9	Gabon	69,003
10	Peru	1641	3.83	10	New Zealand	63,976
11	India	1446	3.38	11	Solomon Islands	58,616
12	Myanmar	1003	2.34	12	Peru	49,493
13	Democratic Republic of the Congo	900	2.10	13	Chile	45,486
14	Chile	885	2.07	14	Colombia	41,904
15	Venezuela	805	1.88	15	Pacific Island small states	41,695
16	Papua New Guinea	801	1.87	16	Small states	39,564
17	Malaysia	580	1.35	17	Belize	38,599
18	Australia	492	1.15	18	Liberia	38,028
19	Philippines	479	1.12	19	Democratic Republic of the Congo	37,677
20	Ecuador	442	1.03	20	Vanuatu	32,694
21	Japan	430	1.00	21	Panama	31,436
22	Mexico	409	0.96	22	Fiji	31,144
39	Thailand	225	0.52	23	Russia	29,790
89	Ukraine	55	0.13	24	Venezuela	28,508
93	Switzerland	40	0.09	25	Uruguay	27,144
130	The Netherlands	11	0.03	26	Brazil	27,014
	World	42,809

Note: Source: World Bank database.

presents the renewable freshwater resources in the world, their percentages, and the renewable freshwater resources per capita. Brazil and Russia rank first in the world in terms of freshwater resources, with 13.22% and 10.07%, respectively. Although Canada ranks third, the percentage share is quite different between these two countries. It is seen that the countries that do not change much in terms of renewable freshwater resources and per capita renewable freshwater resources are Canada (3, 7) and Chile (14, 13). The rankings of other countries in terms of total renewable freshwater resources and per capita water amounts do not always coincide. In particular, countries with high total water resources may be ranked lower in per capita water amount, depending on their population density. These differences show that water resources should be evaluated in terms of quantity and population density, usage patterns, and management strategies. For example, Thailand is in a relatively good position regarding total renewable freshwater resources. However, it is lower in terms of per capita water amount owing to its population size.

Brazil and Russia have 13% and 10% of renewable freshwater resources, respectively. These countries are followed by Canada, the United States, China, and Colombia (Figure 1). As shown in Figure 1, freshwater resources are not equally distributed worldwide, and some countries have more advantageous positions in terms of water resources (Appendix A Figure A22). However, having total water resources does not provide an advantage, and water efficiency, accessibility, and sustainable water management are much more critical. On the other hand, although Brazil and Russia have significant water resources, their geographical distribution and usage policies may limit access to water. However, even in countries with abundant water resources, population growth, climate change, and industrialization may pose long-run risks to water security.

Figure 2 shows the course of renewable freshwater resources of the countries included in the analysis between 1992 and 2021, and their rates compared to other countries. Colombia and Chile have high shares among these countries, but it is noteworthy that their share has decreased over time.

The course of the geopolitical risks of countries in the analysis over time is included in additional Appendix A Figure A1, Figure A2, Figure A3, Figure A4, Figure A5, Figure A6 and Figure A7. The geopolitical risk levels of each country are significantly affected by history, geographical position, regional conflicts, and internal political instability. Access to strategic natural resources such as energy, water, and border disputes, and fragility in international relations has led to a large increase in these risks.

3. Disputes and Conflicts Experienced by Countries

This section includes the causes and consequences of disputes, tensions, and conflicts experienced by the countries included in the analysis. It also examines the geopolitical effects and risks of factors such as sharing water resources, access to water, and the use of water as a strategic power element in relations between countries.

Although Switzerland has not experienced major international disputes over water resources, conflicts and tensions have occasionally occurred in neighboring countries. Issues such as using transboundary waters, water quality in rivers, and the effects of hydroelectric power plants have led to diplomatic negotiations between Switzerland and neighboring countries. Some of the problems are listed below.

‐

The discharge of waste from the Basel region, where Switzerland’s chemical and pharmaceutical industries are concentrated, into the Rhine River caused major environmental disasters. These situations have caused tensions in Germany, France, and the Netherlands over water pollution [13]. The International Commission for the Protection of the Rhine (ICPR) was established in Germany, France, and the Netherlands to protect and manage the Rhine.

‐

There have been occasional disagreements over water levels and flow regulations in the Rhône and Doubs rivers, which Switzerland shares with France. Issues have been particularly focused on the agenda regarding the impact of hydroelectric power plants on water regimes.

‐

Switzerland and Italy share Lakes Maggiore and Lugano. These two countries have occasionally disagreed on agricultural irrigation and drinking water use, especially during dry periods [14]. Switzerland values international cooperation in the management of water resources and has pursued policies aimed at resolving such disagreements through diplomatic means.

Chile shares water resources with Argentina, Bolivia, and Peru. Transboundary water resources, such as rivers and lakes, are essential resources that pose risks to national security and regional stability. Due to rivers flowing across their borders and shared water basins, these countries have experienced tensions, disagreements, and conflicts over the use and management of water resources. Some of the conflicts in this region are as follows.

‐

The Silala River, located in southern Bolivia, flows into Chile. Bolivia claimed that Chile had diverted water and unfairly used Chile’s water resources. In 2018, Chile applied to the International Court of Justice (ICJ) and requested a legal solution for the use of the Silala River’s water resources. In 2022, the ICJ ruled that human intervention in the Silala River’s water resources did not affect Bolivia’s water rights, but stated that both countries had the right to share Silala’s water [15,16,17].

‐

The Pacific War (1879–1884) between Chile, Bolivia, and Peru, although not directly related to water resources, caused problems with natural resources (guano and nitric minerals) and access to the sea, leading to a five-year war. As a result of the war, Bolivia lost access to the sea and became landlocked. Additionally, the rights to water resources in the Atacama Desert were transferred to Chile’s control [18,19].

The Netherlands has experienced disagreements and tensions with neighboring countries, such as Germany and Belgium, regarding water resources. However, these disagreements have generally not escalated into international armed conflicts and have been resolved diplomatically. For example:

‐

The Netherlands disagreed with Switzerland, Germany, and France over the water quality and quantity of the Rhine River. Germany had problems with increasing pollution due to industrial waste, which threatens drinking water and agriculture. These problems were resolved through international agreements, such as the Rhine Protection Convention 1976 and the Rhine Action Plan 1986.

‐

The Netherlands and Belgium disagreed over the water sharing and water quality of the Meuse (Maas) River in the 19th and 20th centuries, and the decrease in water and increased pollution due to Belgium’s agricultural and industrial use negatively affected the Netherlands. The problem was resolved using the Meuse Water Agreement signed in 1994.

‐

The Netherlands has had diplomatic disagreements with England, Germany, and Denmark over maritime jurisdictions in the North Sea, oil and natural gas fields, and fishing rights but has generally resolved these issues through international law and diplomatic negotiations.

Thailand has experienced regional disputes and tensions over water resources in the past, but these disputes have been mainly resolved diplomatically and have not escalated into an armed conflict. The most prominent water resource disputes concern Thailand’s water sharing with neighboring countries, especially Cambodia, Laos, and Myanmar.

‐

Disputes over water resources between Thailand and Cambodia have led to tensions, especially over water sharing on the Mekong River and Tonle Sap Lake. Thailand has expressed concerns about Cambodia’s water use; however, these disputes have generally been resolved diplomatically.

‐

Significant disputes over water resources between Thailand and Laos have intensified, particularly since 2008, over dam projects on the Mekong River and water sharing. In 2010, Thailand and Laos agreed to cooperate on water management and the environmental impacts of the Mekong River. During this period, diplomatic negotiations continued between the two countries, and some disputes were resolved through international cooperation and environmental protection measures.

‐

Some tensions have occurred between Thailand and Myanmar, particularly over water use rights on the Salween River and hydroelectric power plant projects. However, these disputes have been resolved through diplomatic and environmental cooperation agreements.

Ukraine has experienced many disputes and conflicts over water resources throughout history. For example;

‐

Ukraine closed the North Crimean Canal (the central canal carrying water from the Dnieper River to Crimea), which provided approximately 85% of Crimea’s water needs, creating a severe water crisis in Crimea. Russia developed alternative projects to solve the water problem in Crimea. It took control of the North Crimean Canal to restore the water flow to Crimea when it attacked Ukraine in 2022.

‐

Since 2014, tensions in the Donbas region in eastern Ukraine have frequently targeted water infrastructure, causing significant problems in the drinking water supply. Dams and treatment plants in the Donetsk and Luhansk regions have been damaged, and public access to water has been restricted.

‐

The use of the Dniester River, which plays an important role in areas such as the drinking water supply, agricultural irrigation, and hydroelectric power generation between Ukraine and Moldova, has caused disagreements between the two countries. The dams on the Dniester River, which are critical to the region’s energy and water security, were damaged during the Russian–Ukrainian war, causing the region to experience water-related problems.

Regional disputes and problems in Venezuela have occurred, especially regarding water wells and hydroelectric power plants. For example:

‐

Because a large portion of the country’s electricity needs are met by hydroelectric power plants, the decrease in water levels in dams has led to an energy crisis in the country.

‐

The Orinoco and Catatumbo Rivers, which pass between Venezuela and Colombia, are important water resources for both countries. Although they have caused disagreement between the two countries, they have been resolved through diplomatic negotiations.

‐

The Essequibo River, which caused the dispute between Venezuela and Guyana, is one of the region’s water resources.

‐

The environmental impacts and water level changes of the Orinoco River, an important water source used jointly by Venezuela and Brazil, occasionally cause problems managing water resources in both countries. Falling water levels, the impact on hydroelectric projects, and local water use lead to tension [20].

These country examples show the importance of water as a strategic resource and how control over water resources causes tensions in the relations between many countries. Water resources are often not the sole cause of conflict but rather a part of broader geographical, political, and economic tensions. Even if a full-fledged war has not occurred, water resource conflicts have the potential to threaten regional and global peace. The disputes mentioned above, in general terms, as well as the problems experienced by countries regarding water resources, the causes of these problems, solution methods, and geopolitical effects, are summarized in Appendix A,

Table A7. Long-run estimate of the ARDL model.

Models	Coefficient	Std. Error	t-Statistic	Prob.
CHE	−3.62	227.38	−2.06	0.03
CHK	−0.36	0.00	−67.49	0.00
COL	−0.29	0.01	−28.19	0.00
NLD	−0.46	0.01	−36.99	0.00
THA	−0.41	0.00	−48.24	0.00
UKR	−0.44	0.06	−6.63	0.00
VEN	−0.28	0.00	−32.32	0.00

.

In particular, managing transboundary water resources and water security issues are among the main factors shaping these countries’ geopolitical risks. These problems often lead to tensions in international relations, and solution processes have progressed through diplomatic negotiations and legal arrangements between parties. The sharing of water resources is of critical importance, not only in terms of environmental and economic aspects but also in terms of regional security and stability.

Water scarcity increases geopolitical risk levels in countries. In creating geopolitical risks from water resources, the type of water scarcity is as important as the amount of water. Falkenmark (1989), who considered absolute water scarcity, determined scarcity entirely according to numerical values [21]. He determined the amount of water per person according to a certain threshold value. Allan (1998), on the other hand, argues that the amount of water should be determined by political, institutional, and economic values beyond numerical values [22]. Allan considered water scarcity to be a management scarcity rather than a numerical scarcity. Irrespective of how water scarcity is viewed, it is highly likely to create geopolitical risks, and this probability value is increasing daily. Water scarcity can trigger problems, such as conflict risk, migration movements, economic fragility, and political instability between countries, societies, and regions. This situation can go beyond tension and disagreement and turn into conflict, military tension, or even water wars.

Hydropolitics, the combination of water, politics, and power, is a political field that examines competition and cooperation between countries over valuable water resources such as transboundary rivers, lakes, and groundwater [23,24]. As the importance of water and its resources increases, so does the importance of this field. 260 international river basins account for approximately 60% of the world’s freshwater and are shared by neighboring countries. These rivers cover almost half of the world’s surface and host 40% of the world’s population [25]. Nile, Euphrates–Tigris, Mekong, Zambezi, and Amu Darya are important transboundary rivers. No formal agreement guarantees equal shares in 60% of international water basins [25]. Climate change requires serious management and cooperation in regions where water scarcity and hydropolitical tensions are increasing. The water in river basins creates asymmetric power relations between upstream and downstream countries [26]. Water can be expressed not only as a natural resource, but also as an element of geopolitical power.

Disputes between Türkiye, Syria, and Iraq exemplify a hydropolitical problem. The Euphrates River originates in Türkiye and flows through Syria and then Iraq. The Tigris River originates in Türkiye and flows through Syria and Iraq. 90% of the water flowing through the Euphrates and 50% of the water flowing through the Tigris originated from Türkiye. As a downstream country, Iraq provides more than 90% of its water surface area to neighboring countries (approximately 80% from Türkiye) [27]. Iraq and Syria are downstream countries. Iraq is largely dependent on the Euphrates and Tigris rivers for its water resources.

While Türkiye, in the upper basin, controls most of the water, Iraq, in the lower basin, has occasionally considered this situation an element of geopolitical pressure. There have been disputes between countries, especially regarding the Southeastern Anatolia Project (GAP) and the Atatürk Dam. According to the protocol signed between Türkiye and Syria in 1987, Türkiye released 500 cubic meters of water per second from the Euphrates River into Syria [28].

Türkiye’s stance on these important water resources was clear. Türkiye defines the Euphrates and Tigris rivers as “transboundary rivers” rather than “international rivers” and rightfully argues that each country has sovereign rights over the resources within its own borders [29,30].

An example is the Mekong River. The longest river in Asia, the Mekong River, originates in China and flows through six Southeast Asian countries. Another example is the Nile River, the longest river in the world. The basin, which borders 11 African countries, covers one-tenth of the continent. The Nile River, which flows from south to north, originates in Burundi’s East African lake region and flows into Lake Victoria. Ethiopia is an upper basin that makes the country very important, but it also makes it vulnerable to geopolitical risk. The Grand Ethiopian Renaissance Dam (GERD), built by Ethiopia on the Nile, is an important power source. The Nile is a transboundary river that concerns 11 countries, and it has caused geopolitical crises between countries, as well as intense political and strategic tensions shaped by water resources management, regional sovereignty, development goals, and historical agreements. These countries are in a strong hydropolitical position, with the power they derive from rivers (Appendix A Figure A23).

Zeitoun and Warner (2006) explained the asymmetric relations that emerge in water sharing with the concept of “hydro-hegemony” [31]. It has been stated that powerful countries with an upstream location can establish hegemonic superiority over water with strategies such as unilaterally managing and controlling resources and structuring them in their own favor. Hydropolitically strong countries are those with an upstream location in transboundary water resources, high technological and institutional capacity, and strong regional influence. These countries can also be defined as “hydro-hegemons” because they have the power to determine the direction and amount of water flow. Nuclear energy and water resources are two important components of geopolitical power.

4. Empirical Framework

In this study, which aimed to investigate the effect of water resources on geopolitical risk, short- and long-run analyses were conducted using a cointegration test. Rather than exhibiting a static structure, as it affects the instantaneous (t) time between geopolitical risk and water resources, it is much more likely that the past values (t-n) of both variables affect it and exhibit a dynamic structure [32]. Changes in water resources do not immediately trigger the risk of disagreement, conflict, or war; however, over time, they may reveal these risks in line with economic, political, and strategic dynamics. Therefore, more accurate and reliable estimates were obtained considering the dynamic effects rather than the static relationship between the variables. In this direction, analyses were performed using the distributed autoregressive model (ARDL) with a delay in the study.

4.1. Autoregressive Distributed Lag Model

The distributed lag autoregressive (ARDL) model, developed to analyze short- and long-run relationships between time series, is a regression model that includes both the lagged values of the dependent variable (autoregressive structure) and the lagged values of the independent variables (distributed lags) [33]. The ARDL (p, q) model:

$$∆{\mathrm{Y}}_{\mathrm{t}}={\mathsf{\eta }}_0+\sum _{\mathrm{i}=0}^{\mathrm{p}}{\mathsf{\alpha }}_{\mathrm{i}}{∆\mathrm{Y}}_{\mathrm{t}-\mathrm{i}}+\sum _{\mathrm{i}=1}^{\mathrm{q}}{\mathsf{\beta }}_{\mathrm{i}}{∆\mathrm{X}}_{\mathrm{t}-\mathrm{i}}+{\mathsf{\epsilon }}_{\mathrm{t}}$$

(1)

where;

${\mathsf{\eta }}_0$ is the constant term

p and q are the optimum lag lengths

Y is the dependent variable; α is the coefficient of the dependent variable’s own lagged values

X is the independent variable, β is the coefficients of the lagged effects of the independent variables

${\mathsf{\epsilon }}_{\mathrm{t}}$ represents the error term.

The ARDL model has several advantages. First, unlike classical cointegration tests that require all variables to be at the same level of stationarity, the ARDL model can be estimated with variables with different stationarity degrees. For this model, it is sufficient for the variables to be stationary at I(0) and/or I(1) at the first difference. In addition, the ARDL model offers the opportunity to determine the long-run equilibrium time of short-run shocks using a vector error-correction model (VECM). In addition, it is pretty powerful, with its ability to simultaneously estimate short- and long-run effects. Another important advantage of this model is that it can obtain reliable and valid estimates even for small sample sizes. The ARDL model offers both long-and short-run estimation opportunities. It contains two important components: the autoregressive distributed lag bound test and the vector error-correction model (VECM).

4.1.1. Autoregressive Distributed Lag Bound Test

The cointegration tests developed by Engle and Granger (1987), Phillips and Ouliaris (1990), Johansen (1991, 1995), Shin (1994), and Stock and Watson (1988) focus on cases in which the variables are integrated with a first-degree I(1). However, this approach requires preliminary tests to determine the degree of stationarity of variables. This requirement adds additional uncertainties to analyzing the relationships between level variables, making the modeling process more complex [33]. This uncertainty can lead to errors in model selection and incorrect results, mainly when applied to series whose stationarity levels are not precisely known. Even slight differences in stationarity test results can cause misleading results in the analysis process. In addition, because traditional cointegration tests require all variables to be at the same level of integration, these tests cannot be applied to series with different degrees of stationarity.

The bounds test approach developed by Pesaran et al. (2001) overcomes this problem [33]. It allows long-run relationships to be tested without knowing whether the variables are I(0) or I(1). It also provides significant flexibility in the time series analysis by providing reliable results. The second difference is that I(2) was not stationary. The linear two-variable ARDL model is expressed as follows:

$$∆{\mathrm{Y}}_{\mathrm{t}}=\mathsf{\eta }+\sum _{\mathrm{i}=0}^{\mathrm{p}}{\mathsf{\delta }}_{\mathrm{i}}{∆\mathrm{Y}}_{\mathrm{t}-\mathrm{i}}+\sum _{\mathrm{j}=1}^{\mathrm{q}}{\mathsf{\vartheta }}_{\mathrm{j}}{∆\mathrm{X}}_{\mathrm{t}-\mathrm{j}}+\mathsf{\alpha }{\mathrm{Y}}_{\mathrm{t}-1}{+\mathsf{\beta }{\mathrm{X}}_{\mathrm{t}-1}+\mathsf{\epsilon }}_{\mathrm{t}}$$

(2)

Y_t is the dependent variable, X_t is the independent variable, δ and ϑ are the short-run coefficients, α and β are the long-run coefficients, Δ is the difference operator, p and q are the optimum lag lengths, and ε_t is the error term. Pesaran et al. (2001) developed an approach based on standard F statistics to test the significance of lagged levels of variables in a univariate balance correction mechanism [33]. This method tests the following hypotheses to determine whether cointegration exists among variables:

• H₀: α=β=0
• H₁: α≠β≠0

The H₀ hypothesis states that there is no long-run relationship between variables, and the H₁ hypothesis states that there is a long-run relationship between the variables. The bounds test developed by Pesaran et al. (2001) compares the calculated F-statistic to the lower and upper bound values [33]. Three possible situations exist. If the F-statistic > the lower bound value, H₀ is rejected, and cointegration occurs. There is a long-run relationship between the variables. If the F-statistic < the lower bound value, H₀ cannot be rejected without cointegration. There is no long-run relationship between variables. If the F-statistic is between the lower and upper bounds, the result is uncertain, and a clear decision cannot be reached.

4.1.2. Vector Error-Correction Model (VECM) in ARDL Model

If cointegration exists, a vector error-correction model can be established to examine the short-run relationship and how long-run balance is achieved. This model consists of two basic components, short-run dynamics and an error-correction term, and is written as follows:

$$\mathrm{T}={\displaystyle \raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$\left|\mathsf{\lambda }\right|$}\right.}$$

(3)

where:

• T is the return to equilibrium time
• λ is the error-correction term coefficient (VECM coefficient).

5. Econometric Analysis Results

This section explains the data used in the analyses, the sources from which these data were obtained, and all variables included in the analyses. In addition, the estimation results of the ARDL model, cointegration test, ARDL bounds test, and vector error-correction model are included.

5.1. Dataset and Variables

The data for Russia and Ukraine cover 1992–2021, and those for all other countries cover 1961–2021. Geopolitical risk, “Geopolitical Risk Index” calculated by Dario and Iacoviello (2024) [34], the variable “Renewable internal freshwater resources, total (billion cubic meters)” was selected to represent water resources and compiled from the World Bank database. The Geopolitical Risk Index (GPR) was calculated every month. Because freshwater resource data are calculated annually, the data were converted to monthly data using the EViews 12 program. To prevent the loss of observations in the analyses and to obtain results compatible with high-frequency series, annual data were converted to quarterly frequencies using the interpolation method [35,36].

All data were logarithmized, and all analyses were estimated using monthly data. For the analyses in the study, 41 countries covered by the “Geopolitical Risk Index” and “Renewable Internal Freshwater Resources” datasets were selected. The country abbreviations in this study were written according to the three-letter code system by international standards, and these abbreviations were used throughout the study. All abbreviations and their explanations for this study are presented in the Abbreviations table in the back.

The models used in the analyses are shown with country abbreviations; the abbreviations for renewable internal freshwater resources of countries are shown as “fw”; the abbreviations for geopolitical risk of countries are shown as “gpr.” For example, CHE is the model for Switzerland, fw is Switzerland’s renewable internal freshwater resources, and gpr is Switzerland’s geopolitical risk.

Before starting the ARDL estimation, Augmented Dickey–Fuller (ADF) and Phillips–Perron (PP) tests were applied to determine whether the series had a unit root. If a clear decision could not be made between the two tests, the Elliot–Rottenberg–Stock (ERS) test was used.

All geopolitical risk series of the countries selected for analysis are stationary at the level; the freshwater series are stationary at different levels in each country. The stationary countries in the first difference of the freshwater series are CHE, CHL, COL, DEU, DNK, IND, ISR, NLD, SAU, THA, UKR, USA, VEN, and WLD. The stationary countries in the second difference of the water series are ARG and AUS. BEL, BRA, CAN, CHN, EGY, ESP, FIN, FRA, GBR, HUN, IDN, ITA, JPN, KOR, MEX, MYS, NOR, PER, PHL, POL, PRT, RUS, SWE, TUN, VNM, ZAF. A summary of the unit root test results is presented in Appendix A,

Table A1. Summary of unit root tests.

Countries	Variables	I(0)	I(1)	I(2)	Countries	Variables	I(0)	I(1)	I(2)
ARG	gpr	+			KOR	gpr	+
	fw			+		fw			+
AUS	gpr	+			MEX	gpr	+
	fw			+		fw			+
BEL	gpr	+			MYS	gpr	+
	fw			+		fw			+
BRA	gpr	+			NLD	gpr	+
	fw			+		fw		+
CAN	gpr	+			NOR	gpr	+
	fw			+		fw			+
CHE	gpr	+			PER	gpr	+
	fw		+			fw			+
CHL	gpr	+			PHL	gpr	+
	fw		+			fw			+
CHN	gpr	+			POL	gpr	+
	fw			+		fw			+
COL	gpr	+			PRT	gpr	+
	fw		+			fw			+
DEU	gpr	+			RUS	gpr	+
	fw		+			fw			+
DNK	gpr	+			SAU	gpr	+
	fw		+			fw		+
EGY	gpr	+			SWE	gpr	+
	fw			+		fw			+
ESP	gpr	+			THA	gpr	+
	fw			+		fw		+
FIN	gpr	+			TUN	gpr	+
	fw			+		fw			+
FRA	gpr	+			TUR	gpr	+
	fw			+		fw			+
GBR	gpr	+			UKR	gpr	+
	fw			+		fw		+
HUN	gpr	+			USA	gpr	+
	fw			+		fw		+
IDN	gpr	+			VEN	gpr	+
	fw			+		fw		+
IND	gpr	+			VNM	gpr	+
	fw		+			fw			+
ISR	gpr	+			ZAF	gpr	+
	fw		+			fw			+
ITA	gpr	+			WLD	gpr	+
	fw			+		fw		+
JPN	gpr	+
	fw			+

Note: gpr geopolitical risk; fw freshwater; I(0) shows the level, I(1) shows the first-degree difference; I(2) shows the second-degree difference.

.

Although the initial study began with 41 countries, analyses using country models at different stages could not be conducted. Although these processes were mentioned in this study, Figure 3 summarizes them.

It is impossible to apply traditional cointegration tests because both geopolitical risks are stationary at the same level, and the two variables are stationary at different levels. The relationship between the variables was estimated using the ARDL model developed by solving these constraints.

5.2. Estimation Results of the ARDL (Autoregressive Distributed Lag) Model

The ARDL model, which provides reliable results in cases where variables have different stationarity levels and allows cointegration testing, aims to analyze the short- and long-run relationships between time series. Pesaran et al. (2001), within the framework of the ARDL model, developed a cointegration test with a series that does not require all variables to be at the same level of integration and is stationary at level (I(0)), unlike traditional cointegration tests, with the bounds test method [33].

Although the ARDL model has a flexible structure, it requires the stationarity level not to be stationary in the second difference. Therefore, although the study started with data from 41 countries, stationary countries in the second difference were excluded from the ARDL model estimation. The ARDL, CHE, CHL, CHN, COL, DEU, DNK, IND, ISR, NLD, SAU, THA, UKR, USA, and VEN models were estimated using data from countries and the world. The equations for these countries were estimated with different lag lengths, using the ARDL model. Lag length was determined according to the Akaike Information Criterion (AIC).

Although the ARDL model is estimated, the estimates must pass the diagnostic tests successfully to ensure accuracy, reliability, and consistency. These tests are important criteria for evaluating heteroscedasticity, autocorrelation, error term distribution, model specification, and general stability of the model (CUSUM and CUSUM of squares).

The next stage of the estimation was not passed for models that did not pass even one of the diagnostic tests. For example, although the DEU, IND, ISR, and USA WLD models did not provide the feature that only the error terms had a normal distribution, the ARDL bound cointegration test did not continue with these models. The country models that passed all the diagnostic tests, such as heteroscedasticity, autocorrelation, normal distribution of error terms, no error in determining the model, and no structural change in the model, are CHE, CHL, COL, NLD, THA, UKR, and VEN. Therefore, the other stages of the ARDL model were continued by using these models (

Table 2. Diagnostic tests belong to the models.

Models	Diagnostic Tests
CHE	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	0.92	Prob. F (2, 315)	0.39
		Obs*R-squared	1.85	Prob. Chi-Square (2)	0.39
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	0.37	Prob. F (2, 312)	0.68
		Obs*R-squared	0.77	Prob. Chi-Square (2)	0.67
	Normality	Jarque-Bera	0.56
		Probability	0.75
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	1.37	313	0.16
		F-statistic	1.89	(1, 313)	0.16
		Likelihood ratio	1.92	1	0.16
CHL	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	1.53	Prob. F (2, 315)	0.21
		Obs*R-squared	3.06	Prob. Chi-Square (2)	0.21
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	0.39	Prob. F (2, 313)	0.67
		Obs*R-squared	0.80	Prob. Chi-Square (2)	0.66
	Normality	Jarque-Bera	0.56
		Probability	0.76
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	2.09	344	0.03
		F-statistic	4.38	(1, 344)	0.03
		Likelihood ratio	4.56	1	0.03
COL	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	1.30	Prob. F (7, 234)	0.24
		Obs*R-squared	9.11	Prob. Chi-Square (7)	0.24
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	1.45	Prob. F (2, 232)	0.23
		Obs*R-squared	3.00	Prob. Chi-Square (2)	0.22
	Normality	Jarque-Bera	1.67
		Probability	0.43
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	1.52	233	0.12
		F-statistic	2.32	(1, 233)	0.12
		Likelihood ratio	2.40	1	0.12
DEU	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	1.06	Prob. F (7, 234)	0.37
		Obs*R-squared	28.78	Prob. Chi-Square (7)	0.37
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	1.12	Prob. F (2, 232)	0.32
		Obs*R-squared	2.29	Prob. Chi-Square (2)	0.31
	Normality	Jarque-Bera	419.35
		Probability	0.00
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	0.31	358	0.75
		F-statistic	0.09	(1, 358)	0.75
		Likelihood ratio	0.09	1	0.75
DNK	Heteroskedasticity Test: Breusch–Pagan-Godfrey	F-statistic	0.31	Prob. F (7, 234)	0.98
		Obs*R-squared	3.87	Prob. Chi-Square (7)	0.98
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	1.64	Prob. F (2, 232)	0.17
		Obs*R-squared	5.09	Prob. Chi-Square (2)	0.16
	Normality	Jarque-Bera	18067.41
		Probability	0.00
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	2.33	348	0.02
		F-statistic	5.43	(1, 348)	0.02
		Likelihood ratio	5.60	1	0.01
IND	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	1.27	Prob. F (7, 234)	0.27
		Obs*R-squared	5.09	Prob. Chi-Square (7)	0.27
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	0.99	Prob. F (2, 232)
		Obs*R-squared	2.01	Prob. Chi-Square (2)
	Normality	Jarque-Bera	85.23
		Probability	0.00
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	1.41	364	0.15
		F-statistic	2.00	(1, 364)	0.15
		Likelihood ratio	2.03	1	0.15
ISR	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	0.14	Prob. F (7, 234)	0.70
		Obs*R-squared	0.14	Prob. Chi-Square (7)	0.70
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	0.03	Prob. F (2, 232)	0.96
		Obs*R-squared	0.07	Prob. Chi-Square (2)	0.96
	Normality	Jarque-Bera	3430.74
		Probability	0.00
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	0.24	358	0.80
		F-statistic	0.06	(1, 358)	0.80
		Likelihood ratio	0.06	1	0.80
NLD	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	0.31	Prob. F (6, 324)	0.93
		Obs*R-squared	1.89	Prob. Chi-Square (6)	0.92
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	0.11	Prob. F (2, 322)	0.88
		Obs*R-squared	0.24	Prob. Chi-Square (2)	0.88
	Normality	Jarque-Bera	1.11
		Probability	0.57
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	1.09	323	0.27
		F-statistic	1.20	(1, 323)	0.27
		Likelihood ratio	1.22	1	0.26
SAU	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	3.98	Prob. F (6, 361)	0.00
		Obs*R-squared	22.84	Prob. Chi-Square (6)	0.00
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	0.54	Prob. F (1, 359)	0.57
		Obs*R-squared	1.12	Prob. Chi-Square (2)	0.57
	Normality	Jarque-Bera	282.40
		Probability	0.00
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	3.05	360	0.00
		F-statistic	9.35	(1, 360)	0.00
		Likelihood ratio	9.43	1	0.00
THA	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	0.50	Prob. F (2,b299)	0.60
		Obs*R-squared	1.01	Prob. Chi-Square (2)	0.60
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	1.05	Prob. F (1, 297)	0.30
		Obs*R-squared	1.07	Prob. Chi-Square (1)	0.30
	Normality	Jarque-Bera	3.46
		Probability	0.16
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	0.91	297	0.36
		F-statistic	0.83	(1, 297)	0.36
		Likelihood ratio	0.84	1	0.35
UKR	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	0.55	Prob. F (5, 272)	0.73
		Obs*R-squared	2.80	Prob. Chi-Square (5)	0.73
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	0.90	Prob. F (2, 270)	0.40
		Obs*R-squared	1.85	Prob. Chi-Square (2)	0.39
	Normality	Jarque-Bera	4.76
		Probability	0.09
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	1.61	271	0,10
		F-statistic	2.59	(1, 271)	0.10
		Likelihood ratio	2.65	1	0.10
USA	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	1.05	Prob. F (4, 365)	0.37
		Obs*R-squared	4.24	Prob. Chi-Square (4)	0.37
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	1.08	Prob. F (2, 270)	0.33
		Obs*R-squared	2.21	Prob. Chi-Square (2)	0.32
	Normality	Jarque-Bera	35,506.40
		Probability	0.00
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	0.37	362	0.70
		F-statistic	0.14	(1, 362)	0.70
		Likelihood ratio	0.14	1	0.70
VEN	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	1.72	Prob. F (4, 268)	0.14
		Obs*R-squared	6.86	Prob. Chi-Square (4)	0.14
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	1.65	Prob. F (3, 264)	0.17
		Obs*R-squared	5.03	Prob. Chi-Square (3)	0.16
	Normality	Jarque-Bera	0.98
		Probability	0.61
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	1.04	266	0.29
		F-statistic	1.08	(1, 266)	0.29
		Likelihood ratio	1.11	1	0.29
WLD	Heteroskedasticity Test: Breusch–Pagan–Godfrey	F-statistic	0.65	Prob. F (4, 365)	0.62
		Obs*R-squared	2.64	Prob. Chi-Square (4)	0.61
	Breusch–Godfrey Serial Correlation LM Test	F-statistic	0.89	Prob. F (2, 363)	0.41
		Obs*R-squared	1.80	Prob. Chi-Square (2)	0.40
	Normality	Jarque-Bera	5650.23
		Probability	0.00
	Model Specification	Ramsey Reset Test	Value	df	Probability
		t-statistic	1.28	364	0.19
		F-statistic	1.65	(1, 364)	0.19
		Likelihood ratio	1.67	1	0.19

Note: Null hypothesis: Homoskedasticity; Null hypothesis: No serial correlation at up to 2 lags. The models include estimation results for countries; ‘df’ represents the degrees of freedom. The country models that successfully passed all diagnostic tests are shown in bold in the table.

). Because there are too many countries and too many diagnostic tests in the study, the results are summarized in Appendix A,

Table A4. Summary of ARDL model.

Models	Diagnostic Tests	Long-Run	$\left(\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$\mathit{\lambda }$}\right.\right)$	$\left(\mathsf{\lambda }\right)$
CHE (1, 0)	+	−3.62	1.33	0.75%
CHL (2, 11)	+	−0.36	1.45	0.69%
COL (5, 1)	+	−0.29	3.51	0.28%
DEU	−	−	−	−
DNK	−	−	−	−
IND	−	−	−	−
ISR	−	−	−	−
NLD (5, 0)	+	−0.46	2.19	0.45%
SAU	−	−	−	−
THA (1, 1)	+	−0.41	1.16	0.86%
UKR (4, 0)	+	−0.44	6.99	0.14%
USA	−	−	−	−
VEN (2, 2)	+	−0.28	1.67	0.60%
WLD	−	−	−	−

, to make the results clear. The “+” sign indicates that it passed the diagnostic test, and the “−“ sign indicates that it did not pass the diagnostic test.

5.3. Boundary Test Within the Framework of the ARDL Model

The bounds test was applied to determine cointegration for all diagnostic tests in the country models CHE, CHL, COL, NLD, THA, UKR, and VEN. Pesaran et al. (2001) developed an approach based on standard F statistics to test the significance of the lagged levels of the variables [33]. The advantage of the autoregressive distributed lag ARDL bound test is that it can be applied regardless of whether the series in the model is I(2), all I(0), and I(1), or all mutually cointegrated I(1). This is ineffective for models consisting of stationary series only at the second level. This is because if they are stationary when the second difference is considered, the critical values that can be compared are not derived. Another important feature is that it is a model that helps capture long-and short-run causality relationships. However, the bounds test provides effective and unbiased estimates for small and limited sample sets. Of course, it should not be forgotten that increasing the number of observations increases the possibility of reaching more effective and consistent results.

When the test statistic value is greater than the upper limit of the table [I(1)], the H₀ hypothesis, which states that there is no cointegration in the series, is rejected. For example, because the test statistic value is 64.14 in the CHE model and the upper value of the table is more significant than 5.58 at the 1% significance level, H₀ is rejected. All models have a cointegration relationship between the variables (

Table 3. F-bounds test.

Models		Value	Signif.	I (0)	I (1)
CHE	Test Statistic	64.14	10%	3.02	3.51
	F-statistic	1	5%	3.62	4.16
	k		2.5%	4.18	4.79
			1%	4.94	5.58
CHL	Test Statistic	38.42	10%	3.02	3.51
	F-statistic	1	5%	3.62	4.16
	k		2.5%	4.18	4.79
			1%	4.94	5.58
COL	Test Statistic	4.76	10%	3.02	3.51
	F-statistic	1	5%	3.62	4.16
	k		2.5%	4.18	4.79
			1%	4.94	5.58
NLD	Test Statistic	10.57	10%	3.02	3.51
	F-statistic	1	5%	3.62	4.16
	k		2.5%	4.18	4.79
			1%	4.94	5.58
THA	Test Statistic	79.28	10%	3.02	3.51
	F-statistic	1	5%	3.62	4.16
	k		2.5%	4.18	4.79
			1%	4.94	5.58
UKR	Test Statistic	3.75	10%	3.02	3.51
	F-statistic	1	5%	3.62	4.16
	k		2.5%	4.18	4.79
			1%	4.94	5.58
VEN	Test Statistic	24.72	10%	3.02	3.51
	F-statistic	1	5%	3.62	4.16
	k		2.5%	4.18	4.79
			1%	4.94	5.58

Note: Null Hypothesis: No levels of relationship.

).

5.4. Vector Error-Correction Model (VECM)

The vector error-correction model (VECM) in the ARDL model analyzes how short-run shocks adapt to long-run balance and how long it will take to restore the balance. Equation (4) expresses the CHE model’s ARDL vector error-correction model (VECM).

$$\begin{gathered} \mathrm{L}\mathrm{g}\mathrm{p}\mathrm{r}=2.75-0.76{\mathrm{L}\mathrm{g}\mathrm{p}\mathrm{r}}_{\mathrm{t}-1}-355.10\mathrm{L}\mathrm{f}\mathrm{w}-0.76{\epsilon }_{t-1} \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\left(0.00\right)\hspace{1em}\hspace{1em}\left(0.04\right)\hspace{1em}\hspace{1em}\hspace{1em}\left(0.00\right) \end{gathered}$$

(4)

The CHE model estimated the VECM coefficient (CointEq (−1)) as −0.76. The coefficient is negative and statistically significant at the 1% level, and two assumptions are met. The VECM coefficient should be tested to determine if it is negative, if the t-statistic value is significant, and if it is compatible with the standard t distribution. A bound test should also be performed for the t-statistics. The t-statistics should be greater than the upper critical value of the absolute value. The t-statistic value is more significant than all models’ upper critical values in absolute terms. Because all the assumptions are met, comments can now be made. The CHE model corrected 76% of the previous period’s imbalances. The deviation in the short-run is corrected by 1.33 months (1/0.75). According to the VECM models of other countries, the recovery rates of the previous period imbalances in each period were estimated as CHL 0.69%, COL 0.28%, NLD 0.45%, THA 0.86%, UKR 0.14%, and VEN 0.60%, respectively. When these results are examined, it is seen that the effect of the shock occurring in the short-run is the longest in the UKR model (approximately 7 months) and the shortest in the CHE model (approximately 1.5 months shorter).

According to VECM models, the country with the highest rate of elimination of imbalances in previous periods is Thailand, followed by Switzerland. When two countries are compared, geographical differences are important factors in managing water resources and adapting to natural resources. Switzerland’s mountainous and cold climate and the tropical and more diversified geography of Thailand differentiate the use of water resources and perception of risk. While Thailand is mountainous only in the north and west, Switzerland’s mountainous location also affects the use and management of water resources and, therefore, the degree and duration of reaching a balance in risks.

In addition to comparing two countries with similar speeds of reaching balance, it is important to compare the countries with the highest and lowest speeds of reaching balance. Thailand’s high ECT coefficient (86%) shows that imbalances between water resources and geopolitical risks are corrected very quickly; that is, the system responds quickly to such risks. However, the very low ECT value in Ukraine (0.14%) shows that it is slow to correct long-run imbalances. Economic and political instabilities in a country also explain this rate. Russia’s military occupation, which began in 2014 and intensified in 2022, seriously weakened Ukraine’s economic infrastructure, institutional structure, and general management capacity. Such a heavy external intervention and conflict complicates water resource management and control of geopolitical risks and prevents the system from adapting to environmental and economic shocks.

The periods in which other countries reach equilibrium are CHL 1.45, COL 3.51, NLD 2.19, THA 1.16, and VEN 1.67 months. While limited water resources increase geopolitical risks, decreasing these resources leads to further pressure intensification. According to the estimation results, the countries with the most extended periods of reaching equilibrium after the shocks are Ukraine (6.69) and the Netherlands (2.19). When the world water resources ranking is examined, it is striking that these countries are not rich in water; Ukraine ranks 89th, whereas the Netherlands ranks 130th (Table 1). The econometric analysis results support these findings. This shows that countries experiencing water scarcity are more vulnerable to external factors and that water management policies are of vital importance for geopolitical stability. Water crises can trigger economic instability and increase international tensions, especially in regions with limited access to water.

While interpreting the ARDL model estimates, evaluations were conducted in the context of the interaction of these factors beyond the numerical data by considering the countries’ water resources, geopolitical dynamics, and economic conditions. The analyses and interpretations comprehensively reveal how these dynamics differ by considering the relationship between water resources and geopolitical risk on a national basis.

The estimation results of the ARDL models belonging to the countries and information on the short-run equations are presented in

Table 4. Short-run forecast results of the ARDL model.

Models	Variable	Coefficient	Std. Error	t-Statistic	Prob.
CHE	C	−2.74	0.23	−11.57	0.00
	Lgpr(−1)	−0.75	0.05	−13.86	0.00
	DLfw	−355.10	174.02	−2.04	0.04
CHL	C	0.008	0.00	1.68	0.09
	Lgpr(−1)	−0.68	0.06	−10.73	0.00
	DLfw(−1)	−2.57	4.84	−0.53	0.59
	D(Lgpr(−1))	−0.07	0.05	−1.54	0.12
	D(DLfw)	12.96	23.66	0.54	0.58
	D(DLfw(−1))	67.24	23.85	2.81	0.00
	D(DLfw(−2))	−0.06	24.09	−0.00	0.99
	D(DLfw(−3))	−12.27	24.05	−0.51	0.61
	D(DLfw(−4))	0.65	23.79	0.02	0.97
	D(DLfw(−5))	0.80	23.77	0.03	0.97
	D(DLfw(−6))	4.18	23.77	0.17	0.86
	D(DLfw(−7))	−19.81	23.77	−0.83	0.40
	D(DLfw(−8))	−10.11	23.79	−0.42	0.67
	D(DLfw(−9))	111.46	23.79	4.68	0.00
	D(DLfw(−10))	−35.45	24.49	−1.44	0.14
COL	D(Lgpr(−1))	−0.19	0.04	−4.04	0.00
	D(Lfw)	223.74	52.52	4.25	0.00
	D(Lfw(−1))	−236.09	52.53	−4.49	0.00
NLD	Lgpr(−1)	−0.51	0.06	−7.50	0.00
	Lfw	−0.23	0.03	−7.43	0.00
	D(Lgpr(−1))	−0.20	0.06	−3.13	0.00
	D(Lgpr(−2))	−0.11	0.05	−2.16	0.03
THA	Lgpr(−1)	−0.40	0.13	−3.02	0.00
	Lfw	−0.16	0.05	−3.03	0.00
	D(Lgpr(−1))	−0.50	0.13	−3.80	0.00
	D(Lgpr(−2))	−0.43	0.12	−3.36	0.00
	D(Lgpr(−3))	−0.30	0.12	−2.51	0.01
	D(Lgpr(−4))	−0.32	0.11	−2.97	0.00
	D(Lgpr(−5))	−0.29	0.09	−3.09	0.00
UKR	Lgpr(−1)	−0.10	0.04	−2.38	0.01
	Lfw	−0.04	0.01	−2.61	0.00
	D(Lgpr(−1))	−0.42	0.06	−6.06	0.00
	D(Lgpr(−2))	−0.38	0.07	−5.32	0.00
	D(Lgpr(−3))	−0.35	0.07	−4.77	0.00
	D(Lgpr(−4))	−0.11	0.07	−1.56	0.11
	D(Lgpr(−5))	−0.09	0.06	−1.41	0.15
	D(Lgpr(−6))	−0.12	0.06	−2.00	0.04
VEN	D(Lgpr(−1))	−0.14	0.05	−2.50	0.01
	D(Lfw)	−718.47	357.32	−2.01	0.04
	D(Lfw(−1))	890.94	357.58	2.49	0.01

Note: Fw: freshwater resources, gpr: geopolitical risk index, L: logarithm, D: difference, C: constant parameter. The p-value is incompatible with the t-bound distribution.

, and the long-run estimation results are presented in Appendix A,

Table A6. VECM regression.

Models	Variable	Coefficient	Std. Error	t-Statistic	Prob.	$\left(\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$\mathit{\lambda }$}\right.\right)$
CHE	CointEq(−1) *	−0.75	0.05	−13.91	0.00	1.33
CHK	CointEq(−1) *	−0.68	0.06	−10.76	0.00	1.45
COL	CointEq(−1) *	−0.28	0.07	−3.79	0.00	3.51
NLD	CointEq(−1) *	−0.45	0.08	−5.64	0.00	2.19
THA	CointEq(−1) *	−0.86	0.05	−15.47	0.00	1.16
UKR	CointEq(−1) *	−0.14	0.04	−3.36	0.00	6.99
VEN	CointEq(−1) *	−0.59	0.06	−8.64	0.00	1.67

Note: CointEq(−1) error-correction term.

. In the CHE model, the Lfw variable reduces the Lgpr variable by 355.10% in the short-run and the Lgpr variable by 0.76% of the lagged value one period ago [Lgpr(−1)]. The findings show that disputes and conflicts related to water resources significantly increase the coefficients of some variables in the short-run. In the long-run, a 1% decrease in Lfw increased Lgpr by 3.62%. In the long-run, a 1% decrease in Lfw increased Lgpr by 3.62%.

In the CHL model, the Lfw variable does not affect the Lgpr variable in the short-run. In the long-run, a 1% decrease in Lfw increased Lgpr by 0.37%. In the COL model, Lfw inversely affects Lgpr in the short-run. A 1% decrease in Lfw increased Lgpr by 236.09%. A 1% increase in the previous value of the Lgpr variable decreases it by 0.20%. In the long-run, a 1% decrease in the Lfw variable increases the Lgpr variable by 0.30%. In the short-run, a 1% decrease in the Lfw variable in the NLD model increased the Lgpr variable by 0.23%; in the long-run, a 1% decrease in the Lfw variable increased the Lgpr variable by 0.46%. A 1% decrease in the previous lag of the Lgpr variable [Lgpr(−1)] increased the Lgpr variable by 0.52%. In the THA model, a 1% decrease in Lfw increased Lgpr by 0.41% in the short-run and 0.42% in the long-run. The first lag of the Lgpr variable [Lgpr(−1)] affected the Lgpr variable by approximately 0.51%, which continued for five periods. In the UKR model, a 1% decrease in the Lfw variable increased the Lgpr variable by 0.04% in the short-run and 0.44% in the long-run. The sixth lag of the Lgpr variable [Lgpr(−6)] explains 0.12% of the Lgpr variable.

In the VEN model, a 1% decrease in the Lfw variable increased the Lgpr variable by 718% in the short-run. In the Venezuela model, water resources have a high impact on geopolitical risks in the short-run (718%). Although this impact level is high, no model violations are indicated in the model in diagnostic tests, such as autocorrelation, heteroscedasticity, and normal distribution. In addition, the successful passing of the CUSUM and CUSUM of squares tests shows that the model is structurally stable. It can be evaluated as a temporary response to Venezuela’s fragile economic structure and periodic shocks (e.g., hyperinflation, currency crises, political uncertainty, sudden oil price shocks, and external dependency). In addition, the fact that a large part of the population (69.2%) does not have regular access to safe drinking water increases this risk factor. In addition to these negative factors, more than 34 oil spills in 2024 polluted Lake Maracaibo by 70% [37]. High levels of mercury pollution were detected in more than five water basins in the Bolívar and Amazon regions. These negative developments that the country is exposed to reflect the strong short-run impact of water resources on geopolitical risks.

If the estimation results of the ARDL model are summarized:

- In all models, cointegration is a long-run relationship between water resources and geopolitical risk.

- According to the CHE model, Switzerland’s water resources have a very high impact on geopolitical risk in the short-run. In the long-run, a 1% decrease in Lfw increased Lgpr by 3.62%.

- According to the CHL model, water resources do not cause geopolitical risk in Chile in the short-run, while a 1% decrease in water resources increases geopolitical risk by 0.37% in the long-run.

- In the COL model, water resources cause an increase in geopolitical risk in the short-run. In the long-run, a 1% decrease in water resources in Colombia increases geopolitical risk by 0.30%.

- In the NLD model, in the Netherlands, a 1% decrease in water resources increases geopolitical risk by 0.23% in the short-run and by 0.46% in the long-run.

- In the THA model, a 1% decrease in Thailand’s water resources increases geopolitical risk by 0.41% in the short-run and 0.42% in the long-run.

- In the UKR model, a 1% decrease in Ukraine’s water resources increases geopolitical risk by 0.04% in the short-run and by 0.44% in the long-run.

In the VEN model, a 1% decrease in Venezuela’s water resources increases geopolitical risk by 718% in the short-run and 0.29% in the long-run.

- In the long-run, the decrease in water resources increases geopolitical risks, and this effect is powerful in the Netherlands and Ukraine.

- The recovery rate of the previous period imbalances is estimated to be 0.75% in Switzerland, 0.68% in Chile, 0.28% in Colombia, 0.45% in the Netherlands, 0.86% in Thailand, 0.14% in Ukraine, and 0.59% in Venezuela. However, the time it takes for countries to reach equilibrium is 1.33 months in Switzerland, 1.45 months in Chile, 3.51 months in Colombia, 2.19 months in the Netherlands, 1.16 months in Thailand, and 1.67 months in Venezuela.

- While limited water resources increase geopolitical risks, decreasing these resources leads to further pressure intensification. According to the estimation results, the countries with the longest time to reach equilibrium aftershocks are Ukraine (6.69) and the Netherlands (2.19), and it is seen that these countries are not rich in water (Table 1).

The decrease in global water resources and the increase in geopolitical tensions have caused water to become a strategic power element, and this requires an increase in academic studies and new research that examines this dynamic relationship in depth. In this context, several suggestions can be made for future studies. A composite index that measures the connection between water scarcity and geopolitical risks can be developed, and the change in risk levels over time can be analyzed. The impact of water-sharing agreements on regional stability in transboundary water basins such as the Nile, Mekong, and Euphrates–Tigris can be examined. The effect of water scarcity and water-based geopolitical crises on oil prices, agriculture, commodities, and food inflation can be investigated using econometric models. Sustainable solutions can be developed by evaluating the impact of a country’s water management policies on internal conflicts, political instability, and the economy. The effect of global warming on water resources can be analyzed to investigate how changing water balances shape geopolitical risks.

6. Conclusions

Water, a substance constantly cycling in nature, is continuously redistributed through evaporation, precipitation, and other processes and continues to exist in different forms (ice, liquid, and gas). Water, a natural resource that other substances cannot entirely replace, is a fundamental source of life. Water is not only a need but also an essential condition of life and one of the most critical components of sustainability. Water resources are indispensable not only for drinking water and agriculture but also for industry, energy, military strategies, and the sustainability of ecosystems. Proper water management can strengthen economic sustainability and minimize geopolitical risks by increasing the resilience of countries against natural disasters such as droughts and floods.

At the same time, water resources are an important factor shaping political and economic relations between countries. Water management in these rivers can trigger geopolitical risks, leading to cross-border conflicts and diplomatic tensions. In the long-run, competition over water resources can pose more risks, especially in relation to neighboring countries. Water scarcity, cross-border disputes, and social pressures may become more pronounced over time. Water-sharing disputes between countries may increase as water resources decrease over time, increasing geopolitical risk. Freshwater resources, especially surface waters such as lakes and rivers, have become geopolitical elements because they are located within the borders of more than one country. Water resources are not only a source of life but also a factor affecting the regional balance of power. For this reason, managing transboundary water resources is an issue that must be carefully addressed within the framework of regional cooperation and international law. This situation can lead to cooperation, tensions, conflicts, and even wars between countries regarding issues such as water sharing, border security, and economic development. Water management is, therefore, a strategic issue. Some of the rivers with geopolitical importance are the Nile River (Egypt, Sudan, and Ethiopia), Amazon River (Brazil, Peru, Colombia, Ecuador, Bolivia, Venezuela, and Guyana), Euphrates River (Türkiye, Syria, and Iraq), Tigris River (Türkiye, Syria, and Iraq), Mekong River (China, Myanmar, Laos, Thailand, Cambodia, and Vietnam), Aral Sea (Uzbekistan, Kazakhstan), Caspian Sea (Russia, Kazakhstan, Turkmenistan, Iran, and Azerbaijan), and Ganges River (India and Bangladesh).

This study estimated the relationship between water resources and geopolitical risk in Switzerland, Chile, Colombia, the Netherlands, Thailand, Ukraine, and Venezuela in the short-run and the long-run using the ARDL model. There is a cointegration relationship between these variables in all countries, and they move together in the long-run. However, each country’s long-run coefficients (i.e., the degree of impact) are different. Therefore, each country was considered separately when evaluating and interpreting results.

Switzerland, which is at the world average in terms of water resources, is not among the top in terms of total reserves when compared to water-rich countries; however, it has a great advantage in terms of the amount of freshwater per capita. Switzerland’s water is among the best in the world, not only in quantity but also in the quality of the spring waters coming from mountainous regions. It is also one of the few countries to export freshwater.

In Switzerland, water resources can cause geopolitical risks at a very high rate in the short and long-run. Therefore, when determining policies regarding water resources and geopolitical risk, both short- and long-run effects should be considered, and strategies should be shaped accordingly. Sustainable management of Switzerland’s water resources can help the country maintain stability by minimizing possible geopolitical risks due to water scarcity. Switzerland also plays an important role in hydroelectric energy production. Providing hydroelectric energy safely and sustainably can reduce geopolitical risks by increasing energy independence; however, it should not be forgotten that this also depends on water.

In Chile, water resources do not cause geopolitical risk in the short-run, but in the long-run, a 1% decrease in water resources increases geopolitical risk by 0.37%. Although Chile is a developing country, it plays a very important role in water resources, especially in countries with water basins, such as the Andes. Water is a valuable and strategic resource for rivers, underground water, and irrigation. Chile is well below the world average regarding water resources and has serious water problems due to water scarcity, climate change, and regional imbalances. However, Chile needs to take steps in water resource management to alleviate the effects of the water crisis. Although the Netherlands is at a medium-low level in terms of water resources, it has a more advanced level of water management. At the same time, Chile is on the way to developing technological and economic solutions to make water resources management sustainable.

In the ARDL model estimated for Colombia, in the short-run, water resources inversely affect geopolitical risk, and it is seen that when water resources decrease, geopolitical risk increases. In the long-run, a 1% decrease in water resources in Colombia increases geopolitical risk by 0.30%. There have been occasional discussions on the use of water resources and the effects of hydroelectric projects in Colombia and its neighboring countries, especially Brazil, Peru, and the Amazon Basin. In the ARDL model for the Netherlands, a 1% decrease in water resources increases geopolitical risk by 0.23% in the short-run and 0.46% in the long-run. The geopolitical risk in the previous period increased by 0.52 risk in the next period. These results provide important findings for understanding the effect of water resources on geopolitical risk in the Netherlands and the internal dynamics of geopolitical risk. This shows that the decrease in water resources can create geopolitical instability in the short-run and that the policies of the Netherlands regarding water management have instantaneous effects. In the long-run, water scarcity creates increasing political and economic pressure, which makes geopolitical risks more apparent. These results show that even in a country such as the Netherlands, which has advanced water management, a decrease in water resources can increase geopolitical uncertainties. In addition, the fact that the long-run effect is higher than the short-run effect indicates that decreasing water resources creates a greater geopolitical risk over time, and that this situation can have long-run effects on the country’s security strategies, foreign relations, and local government policies.

According to the ARDL model, a 1% decrease in Thailand’s water resources increases geopolitical risk by 0.41% in the short-run and 0.42% in the long-run. This result shows that the decrease in water resources can trigger the country’s short-and long-run geopolitical risks. Water scarcity can directly affect agricultural production, the drinking water supply, and industrial activities, leading to tensions in international relations.

In Ukraine, a 1% decrease in water resources increases geopolitical risk by 0.04% in the short-run and 0.44% in the long-run. Although the decrease in water resources in Ukraine does not have much impact on geopolitical risk in the short-run, it demonstrates the strategic importance of water in the region in the long-run. The lagged values of this variable explain geopolitical risk. The fact that lagged values determine geopolitical risk, that is, geopolitical risk has a self-reinforcing dynamic, shows that geopolitical uncertainties and conflicts are continuous over time and that past risks can shape future risks. A 1% decrease in water resources in Venezuela increases geopolitical risk by 718% in the short-run. This is a very high rate, which shows that reductions in water resources cause geopolitical risks to increase rapidly, especially in countries such as Venezuela, which is sensitive to water resource management.

While limited water resources increase geopolitical risks, the fact that the decrease in these resources causes the pressure to intensify is supported by the results of the econometric analysis (Ukraine and the Netherlands). This situation shows that countries experiencing water scarcity are more vulnerable to external factors and that water management policies are of vital importance for geopolitical stability. Water crises can increase international tensions, particularly in regions with limited access to water. The capacity to jointly manage water resources can strengthen regional peace. Such diplomatic relations reduce geopolitical risks. Similarly, it is possible to use water as a peaceful tool. Since water is a vital natural resource, most countries have experienced serious disagreements or conflicts with their neighbors over water sharing due to transboundary rivers, lakes, and groundwater. Although it is rare to find a country that has not experienced conflicts over water resources, some countries have not experienced serious conflicts over water resources. Countries such as Iceland, Norway, New Zealand, Canada, Japan, and Bhutan, rich in water resources, have low population densities or are geographically located where water sharing is not required, and have not experienced such conflicts. In another example, long-run water agreements with Singapore and Malaysia prevented problems between the two countries despite their insufficient water resources.

The impact of water resources on geopolitical risk varies according to each country’s geographical, social, and economic dynamics and neighboring countries. However, in general, a decrease in water resources causes an increase in geopolitical risk, leading to more careful strategies in countries and foreign policies. When examining long-run relations, countries’ water resource management policies, infrastructure investments, and regional cooperation should be considered. For example, the tendency of water-rich countries to use this resource as a strategic tool can change their economic and political balances in the long-run. Short-run dynamics should be evaluated within the framework of factors such as sudden geopolitical crises, natural disasters such as droughts, or disputes regarding water sharing.

These findings reveal that the unbalanced distribution of freshwater resources is one of the main factors that increases geopolitical risks between countries. Water management strategies play a critical role in reducing these risks. If water security cannot be ensured, geopolitical tensions will inevitably increase, especially in water-dependent regions. Competition over water resources, disputes, population growth, drought, migration, and energy production can lead to significant conflicts in international relations when added to factors such as energy production. Therefore, developing international cooperation and sustainable water management policies plays a critical role in minimizing future geopolitical risks.

The decrease in water resources can negatively affect agriculture, industry, and energy production, increasing economic fragility, which, in turn, can feed geopolitical tensions. In addition, the protection of water resources is not only a part of environmental awareness but also a critical necessity for the economic, political, and long-run well-being of societies. Every drop of water is a factor that directly affects not only the continuity of a resource but also the balance of ecosystems, food security, health conditions, and economic growth. With the increasing importance of water crises, fair and effective management of resources has become a global responsibility. A joint effort should be made not only to use water correctly but also to prevent large-scale human interventions that threaten the sustainability of the environment and water resources (such as ineffective dams and hydroelectric power plants, excessive irrigation and ineffective agricultural practices, and industrial waste) that will lead to water depletion.

It is insufficient to use water effectively and efficiently; protecting water is a responsibility to ensure the lives of both today and future generations. Every drop of water is valuable for the sustainability of life. Therefore, global cooperation and management are necessary to protect the water. Using water correctly and carefully protects natural resources and ensures the continuity of all ecosystems. This responsibility should become a social and global movement, not just an individual one. Water loss poses an irreversible threat to humanity. Establishing this awareness is a critical step not only for today but also for the world of tomorrow.

7. Recommendations and Policy Implications

From the past to the present, water resources have maintained their value as a strategic element. The reduction in water amounts, including water resources, and the rapid increase in the effects of climate change make water resources more valuable, while on the other hand, they are an important risk element.

In particular, pressure, disagreements, and conflicts regarding water resources are triggered in regions where water is strategically important. Therefore, governments must effectively protect and manage water resources to ensure national security.

Reviewing and strengthening regional water-sharing agreements in countries with limited water resources, such as the Netherlands, Thailand, and Chile, is important. In these countries, if necessary, monitoring and updating international water agreements will ensure that water is shared fairly and sustainably. For example, agreements with other countries in the river basin should be made to prevent situations similar to the Meuse River dispute between the Netherlands and Belgium. To prevent international disputes and ensure peaceful water use, it is necessary to strengthen the diplomatic negotiations between countries. More regional and international agreements should be established to manage water resources jointly.

Comprehensive policies should be developed to ensure the sustainable use of water resources. These policies should include not only the protection of water but also the application of more sustainable methods in agriculture, industry, and all other areas to promote the efficient use of water.

In countries such as Thailand and Venezuela, where water resources are exhausted and affected by climate change, the state plays an active role in the fight against climate change. Governments should take steps to protect water resources effectively by developing strategies to adapt to climate change. This will ensure the efficient use of existing water resources and prevent future water crises and conflicts.

Preparing for possible crises related to water resources should be a long-run strategy. Early warning systems should be installed, especially in areas where water resources are reduced due to famine, water pollution, and other environmental threats. In countries such as Venezuela, infrastructure monitoring should be strengthened to detect decreases in water resources.

Sustainability of water use is directly related to awareness. Countries should initiate training programs and public information campaigns to educate people to save water and emphasize the importance of protecting water resources. This will create an important consciousness for them to understand the importance of water and its resources. The exhaustion of water resources, such as climate change, is not only a problem for countries, but also a global problem. Therefore, a country that provides water efficiency should also fulfill its responsibility on a worldwide scale. Countries with high water efficiency, such as Switzerland and the Netherlands, should lead globally to increased awareness of protection and efficient use.

Public education programs for water conservation should be expanded in these countries, and efficient water consumption should be encouraged using developing technologies. Water consumption efficiency should be increased, especially in industry and agriculture, as these sectors consume many water resources. Modern irrigation techniques can minimize water loss in agriculture by ensuring more efficient water use. Similarly, recycling and reusing water in the industry are important steps for reducing water consumption. These innovative approaches contribute to water conservation and increase the sustainability of production processes.

Switzerland, which can be considered the water tower of Europe, feeds many of Europe’s major rivers (Rhine, Po, Danube) with water originating from the Alps. The process of collecting and analyzing hydrological data, such as water levels, flow rates, water quality, glacial melting, and precipitation rates, is important. These data play an important role in understanding environmental risks and planning water sharing. However, planning and implementing these systems jointly with neighboring countries (such as Italy, Germany, and Austria) is much more important.

According to UN data, Chile experienced the highest increase in drought severity among South American countries between 2010 and 2019. Unfortunately, increasing temperatures and decreasing precipitation are not the only problems in the country. Chile clearly states in its constitution that water rights are private. In Chile, where excessive privatization has resulted in inequalities in access to water, new legal regulations that limit the commercialization of water rights and prioritize rural/local communities need to be implemented.

Although the Netherlands, a downstream country, has a strong governance structure supported by flood control systems and advanced infrastructure technologies against the possibility of sea level rise and flood risk due to climate change, it is also extremely important for the development of multilateral hydrological monitoring systems and protocols with neighboring countries. In the Rhine River Basin, flood prevention and water quality monitoring protocols within the framework of the International Rhine Commission, which the Netherlands jointly conducts with Germany and Belgium, can be updated in line with climate change scenarios and strengthened with digital early warning systems that provide more frequent data sharing.

Considering the uncertainties experienced in Thailand’s water regime on the Mekong River and the impact of China’s upstream dam construction in Thailand, a multilateral joint flow information agreement under the roof of Thailand’s Mekong River Commission, especially with Laos and Cambodia, which requires transparent data sharing on water flow patterns, dam management protocols, and seasonal flow regimes, will ensure that important problems are resolved before they occur.

Due to the serious damage to water infrastructure in Ukraine, especially in the Donbas and Kherson regions, cooperation needs to be implemented to reconstruct and protect strategic water infrastructure. In addition, to prevent water resources from being targeted, “environmental protection protocols” within the scope of the Geneva Convention should be implemented with a special monitoring mechanism for water infrastructure. The management deficiencies and hydrological data shortages experienced in Venezuela’s transboundary water resources, which it shares with Guyana and Colombia via the Orinoco Basin, have become more critical, especially with the changes in flow regimes caused by climate change. Therefore, Venezuela needs to implement effective regional policies for the management of transboundary water basins under the umbrella of the Amazon Cooperation Treaty Organization.

Water resources that are not equally distributed on the surface of the world are not equally distributed across the borders of some countries. Water resources that are not equally dispersed may create geopolitical risks in the international arena and tensions within the country’s borders. Therefore, measures should be taken before this possibility occurs. Water access should be provided between income groups, particularly for those living in low-income and rural areas, and water accessibility should be guaranteed.

It is clear that access to water is a right for people to survive. Water resources should not be considered commercial goods. The commercialization of water, which may lead to its commodification, can cause significant problems. In 2010, the UN General Assembly recognized water access as the main human right. Commercialization can increase water inequality by making this right, depending on market conditions.

In a world where the amount of water is decreasing, the increase in water prices can deprive low-income groups of water services when they think of water as a commodity and are left to market conditions. This could lead to health crises and social unrest.

On the other hand, companies with profit aims can waive long-run environmental sustainability. Water is extremely valuable and indispensable for its commercialization and modification. Water and water resources should never be left to the mercy of market power.

These specific policies will provide critical steps for effectively managing water resources and reducing geopolitical risk. Developing solutions tailored to each country’s unique geographical, economic, and social conditions can increase water security and prevent potential conflicts. This approach will not only promote efficient water use but also foster cooperation among countries, enabling greater resilience to the global water crisis. International collaboration and diplomatic solutions for water management and sharing among countries are critical to global security and peace.

Further research is needed to guide national policies. International academic collaborations and policy analyses should be encouraged better to understand the relationship between water resources and geopolitical risk. This will provide valuable strategies for states to predict and prevent future water crises. These policies can help reduce geopolitical risks between countries and strengthen international peace by improving water resource management. The fact that most freshwater comes from transboundary rivers complicates water sharing between countries and increases external dependency. In addition, the risks of saltwater intrusion and flooding are increasing in low-lying areas. This situation leads to a further scarcity of water resources and endangers water security. In particular, with climate change, fluctuations in water levels, drought, or overuse can turn water into a strategic weapon between countries, leading to tension and conflict in international relations. Joint studies should be conducted through international agreements and commissions on managing transboundary rivers and lakes. These collaborations are of great importance for the sustainable use of water resources and the prevention of possible disputes.

The proper management of water resources is not only an environmental responsibility but also critical for national security and international stability. Transboundary water resources can create tension in relations between countries, and water as a strategic weapon can create a greater risk of conflict in the future. Therefore, global cooperation is essential for protecting and efficiently using water resources. At the same time, minimizing the effects of factors such as climate change and population growth on water resources will help ensure a sustainable future for today’s world and the future. Water is not just a natural resource; it guarantees the sustainability of all life, and therefore, protecting it is humanity’s most important responsibility.