A Hydrogeological Perspective of the Status of Ground Water Under the UN Watercourse Convention
Gabriel Eckstein
I. Introduction
International ground water resources are a primary source of water for millions of people worldwide. Yet, attention to ground water has long been secondary to surface water resources, especially among legislatures and policymakers. The emphasis on surface water was and is due, in large part, to the prevalence and importance that streams, rivers and lakes have had on the course of human development. Nevertheless, ground water today is a significant resource that its importance cannot be overlooked. As a global society, we can no longer ignore the implications of the hydrologic cycle and the interrelationship between surface and ground water resources; we can no longer ignore mismanagement or non-management of ground water resources; we can no longer ignore the transboundary implications of ground water; and, we can no longer ignore the need to develop comprehensive water resource management schemes that encompass all hydraulically linked water resources.
Global reliance on ground water resources emerged more than a half century ago. Improvements in pumping technology and tremendous growths in industry, agriculture, and global population spurred ground water use to escalate from meeting strictly local needs to serving as a primary source of freshwater for whole nations. Today, a majority of the world’s population is dependent on ground water for its basic needs. Ground water now accounts for at least seventy-five percent of drinking water in Europe, and exceeds ninety percent in Austria, Croatia, Denmark, Hungary, Italy, Lithuania, and Slovenia. In the United States, ground water is the source of approximately one half of all drinking water, including ninety-seven percent of that used in rural areas.
Ground water resources hold a promise of tremendous opportunity considering the sheer volume of freshwater available under the ground. Global ground water supplies, in fact, dwarf all of those found in rivers, lakes and other freshwater surface sources by a factor of 100 (see table). Freshwater in lakes, streams, wetlands, and other surface bodies of freshwater comprise 0.0081 percent of global water reserves and less than 0.03 percent of the global volume of freshwater. Fresh ground water, on the other hand, constitutes slightly more than thirty percent of global freshwater resources. While not all groundwater resources are easily accessible – due to the depth at which the groundwater is found, or the geology of the surrounding strata – those that are accessible constitute more than thirty-three times the volume of water found in the world’s lakes and streams.
Global Water Supply
|
As a Percent of All Water |
As a Percent of All Freshwater |
|
|
Salt Water Stocks |
||
|
96.54 % |
|
|
0.006 % |
|
|
0.93 % |
|
|
Fresh Water Stocks |
||
|
1.74 % |
68.70% |
|
0.76 % |
30.10 % |
|
0.02 % |
0.86 % |
|
0.007 % |
0.25 % |
|
0.001 % |
0.05 % |
|
0.001 % |
0.04 % |
|
0.0009% |
0.03 % |
|
0.0002% |
0.006 % |
|
0.0001% |
0.003 % |
Note: Numbers may not add to 100% due to rounding.
Notwithstanding this great promise, ground water resources today are experiencing considerable stress. Overexploitation, pollution, salinization, and mismanagement are all contributing to serious water quality and quantity problems for the world’s ground water resources. Aquifers surrounding many of the world’s largest cities are becoming severely depleted as withdrawals exceed natural recharge. For example, London, Copenhagen, and Mexico City all face chronic problems of falling ground water levels as a result of over pumping. In the Punjab region of India, excessive withdrawals have resulted in salinization of the aquifer, while in West Bengal and western Bangladesh, the consequence is arsenic contamination. Moreover, diversions and large hydro-projects have also affected ground water levels, as was the case in 1992 when Slovakia diverted the Danube River into a side canal and reservoir for the purpose of generating hydroelectric power.
While some of these examples appear to be of a national character, all of them have much broader transnational implications. This is because there is scarcely a country in the world (except for most island-nations) that is not somehow linked hydrologically to another country. Among the members of the Economic Commission for Europe, a recent study identified eighty-nine aquifers shared by two or more states. Another study classified eighteen transboundary aquifers in the Mexico-United States border area, many of which are directly related to international watercourses. Moreover, as limited resources are taxed beyond their sustainable capabilities, nations are now looking beyond their borders to supplement dwindling supplies to adequately meet their societal needs and developmental objectives. Accordingly, there is a transboundary context that must be considered whenever addressing the subject of ground water use, management, regulation, allocation, protection, and conservation.
When the United Nations General Assembly adopted in 1997 the Convention on the Non-Navigational Uses of International Watercourses (Watercourse Convention), ground water resources officially emerged as a proper subject of international law. Prior to this event, ground water received little attention in international relations and international law. While agreements focusing on transboundary rivers and lakes have been relatively common, there was a paucity of treaties and international norms squarely addressing shared ground water resources. Ground water was neglected in bilateral and multilateral agreements, ignored in projects with transboundary implications, and cursorily misunderstood in much of legal discourse. The International Law Association’s Helsinki Rules of 1966 and Seoul Rules of 1986 are among the few international documents to formally address the status of ground water under international law. Due to their non-binding nature, however, the influence of the two Rules on state practice and treaty development has been rather limited. Nevertheless, they provided significant inspiration and motivation within academic and environmental circles.
The Watercourse Convention, however, is not intended to serve as a comprehensive elucidation of the status of ground water under international law. In fact, it leaves numerous questions unanswered and generates new concerns about the status of ground water under international law. This is particularly troubling as many of these questions and issues implicate current scientific knowledge, and actually make little sense in the face of hydrologic reality.
Nonetheless, the adoption of the Watercourse Convention marks a unique milestone in the development of international law related to ground water resources. While the scope of the Convention does not include all forms of ground water within its span, it does provide substantial clarification on the current status of certain ground waters within the context of an "international watercourse." Moreover, it both elevates the level of international attention on this "invisible," but valuable natural resource, and provides ground water with status and recognition as a legitimate subject of international law.
The Convention defines ‘watercourse’ as "a system of surface waters and groundwaters constituting by virtue of their physical relationship a unitary whole and normally flowing into a common terminus." This definition is qualified by the meaning given to "International watercourse," which is "a watercourse, parts of which are situated in different States."
Focusing on these definitions, this article examines critically the treatment of ground water under the Convention from a hydrogeological perspective, and identifies the types of ground water that are included within, and those excluded from, the scope of the Convention. Based on hydrogeology and the science of ground water, as well as actual examples, the article identifies six general models in which ground water resources can have transboundary implications. Moreover, the article considers the bases for the inclusion/exclusion of certain ground waters, and assesses the rationale vis a vis basic principles of hydrogeology.
II. Ground water 101
A. The Hydrologic Cycle
The hydrologic cycle is the system in which water, in its various shapes and forms, travels from the atmosphere, to the Earth, and back again in a constant cycle of renewal. Generally, water falls from the atmosphere in the form of precipitation, such as rain, snow, and sleet. Water that falls on land either runs off over the land into streams, rivers, and lakes, or percolate into the earth. Throughout its surface travels and especially when it reaches large bodies of water, like lakes and oceans, water evaporates through the effects of solar energy and returns to the atmosphere where it continues in the cycle.Some water also is absorbed by plants, which then transpire it through their leaves back into the atmosphere.
Water that percolates into the earth typically percolates vertically until it reaches the water table and an aquifer, where it flows in a more lateral direction. Normally, such water eventually emerges in natural discharge sites, such as springs, rivers, lakes, lagoons, swamps, and the sea. While considerably slower than surface water flow, percolation into the subsurface and ground water flow within aquifers are very consistent processes.
Between recharge from the ground surface and surface waters, and discharge into streams and other surface bodies of water, it is evident that ground water is a significant component of the hydrologic cycle. This is especially evident given the exponentially vast quantity of water found under the ground. Thus, from a hydrological point of view, it is improper to consider ground water and surface water as dissimilar, or even "similar," water resources. They are, in fact, the same thing, namely, water moving through the various stages of the hydrologic cycle. Any bifurcation or piecemeal governance of surface and underground water resources is inappropriate and inadequate for optimal productivity and sustainable use, objectives espoused in the preamble of the UN Watercourse Convention.
B. Understanding Ground Water
The term ground water generally refers to any water located beneath the earth's surface that saturates a geologic formation. Ground water comprises only 0.75 percent of the total volume of water found in nature, but it makes up nearly ninety-seven percent of the fresh water readily available to man. With few exceptions, ground water is found in various types of aquifers throughout the world.
1. Aquifers
An aquifer is a relatively porous geologic formation (such as sand or gravel) that has sufficient water storage and transmitting capacity to provide a useful water supply via wells and springs. The upper limit of the saturated area is known as the water table. All aquifers have an impermeable base layer that prevents water from seeping deeper to lower lying strata, creating a natural water reservoir. At any given location, the land surface maybe underlain by one or more distinct aquifers, depending on the composition of the underlying strata.
A confined aquifer (also known as an artesian aquifer) is an aquifer contained within impermeable base and ceiling strata that subject the stored water to pressure exceeding atmospheric pressure. In contrast, an unconfined aquifer (also known as water-table aquifer) is bounded only by an impermeable base layer of rock, and overlain by layers of permeable materials extending from the land surface to the impermeable base of the aquifer. An unconfined aquifer typically is directly related to a surface body of water, such as a river or lake. Rivers, for example, tend to have interrelated unconfined aquifers directly underneath and following the course of the riverbed. Depending on the strata underneath and beside the river, such aquifers also are spread out laterally on both sides of and below the river.
Confined aquifers are not necessarily devoid of any connection to surface water or other water resources. Confined aquifers must have a source for their water and often are recharged through lateral flow of water from recharge zones located at distant higher elevations, such as mountains or high plateaus, where the aquifer crops out on the land surface. Furthermore, confined aquifers can themselves discharge into rivers and lakes. Hence, confined aquifers are very much a part of the hydrologic cycle.
In addition, confined aquifers often are confined in only a portion of the aquifer. The San Pedro Basin Aquifer underlying Mexico and the United States, for example, is a mix confined/unconfined aquifer with transboundary implications. Linked hydraulically to the San Pedro River, both the river and the related groundwater flow northward into the United States. While most of the aquifer is unconfined, in the border region of the basin in the Palominas-Hereford and the St. David-Benson areas, the aquifer is confined.
Another example of a mixed confined-unconfined aquifer with transboundary implications is the Mountain or West Bank Aquifer underlying the foothills bordering the Israeli coastal plain and the Jordan-Dead Sea rift valley. Beginning as an unconfined aquifer in the highlands of the Judean Mountains, which include the Palestinian Territories of the West Bank, the aquifer is recharged solely by precipitation. As it slopes westward toward the Mediterranean Sea and eastward toward the Jordan Rift Valley, following the downward curvature of the strata, it becomes confined in the lowland area underneath impermeable material. Precipitation falling on the surface, at this point, does not reach the aquifer making it absolutely reliant on recharge from the highlands.
Surface water resources that are hydraulically linked to an aquifer are often described as influent or effluent bodies of water. Where the water table is found below the bottom of a surface body of water, such as a stream or a lake, and where the soil is moderately permeable, water percolates from the stream or a lake downward, recharging the underlying aquifer. This is called an influent (or losing) stream or lake. An effluent (or gaining) stream or lake results where the water table, lying at an elevation higher than the intersected stream channel or lake, recharges the surface water resource. This differentiation is important, especially in the context of water quality and contamination. A polluted river that is effluent will not contaminate the related groundwater on either side of the river. Groundwater that is polluted on one side of an effluent river will contaminate the river, but is not likely to affect the quality of the groundwater on the other side of the river.
Aquifers that are non-renewable (i.e., completely detached from the hydrologic cycle) often are described in the legal literature as "fossil" aquifers. Non-renewable aquifers do not have a source of recharge and cannot discharge naturally. As a result, water in these aquifers is static and has little if any flow. Such aquifers typically contain ground water that is trapped in a geologic formation, either because of physical isolation of the aquifer from sources of recharge, impermeability of overlying strata, or paucity of recharge in an arid region. Typically, water in non-renewable aquifers is hundreds if not thousands or millions of years old.
An example of a transboundary non-renewable, unconfined aquifer is the Nubian Sandstone Aquifer in northeastern Africa, underlying the countries of Chad, Egypt, Lybia, and Sudan. Located at a depth ranging from a few meters to a few hundred meters, the water in this aquifer is estimated to be as much as 35,000 years old. While the overlaying strata is still relatively permeable, present-day recharge rates range from miniscule to nil, contingent on the occasional rain and flash flood. Moreover, this aquifer is not related or connected to any other water resource in the region.
2. Ground Water Flow
Water in aquifers is rarely stagnant and tends to flow toward natural discharge sites, such as springs, rivers, lakes, lagoons, swamps, and the sea. While surface water flow is primarily a function of gravity, ground water flow is a function of hydraulic potential, which is a computation involving gravity as the dominant force, but also soil porosity and permeability (the ability of the soil to transmit water), gradient or slope of the strata, ambient air pressure, and temperature. Ground water generally flows from areas of higher hydraulic potential to areas of lower hydraulic potential. As a result, it is possible to have a stream flowing down the side of a mountain in one geographical direction, while water in a connected underlying aquifer flowing in another direction. For example, surface water in the Danube River, as well as related ground water, generally flows toward a terminus in the Black Sea. In the upper region of the Danube, however, where the river emerges from the Black Forest in Germany, water from the river seeps, on a seasonal basis, into the fractured bedrock underlying the river and travels through the fractures into the Rhine River basin, thus flowing toward a terminus in the North Sea. This scenario was the subject of a well-known case – Donauversinkung – brought by the German states of Württemberg and Prussia against Baden.
3. Aquifer recharge
Aquifers may be recharged from rain-soaked ground, from lakes and streams, and to some extent from other aquifers. Significantly, aquifers may also be recharged from certain human activities, such as irrigation operations, dike and canal building, and damming projects. Aquifer recharge is a function of both gravity and of the permeability of the strata lying between the aquifer and the source of the recharge. As a result, aquifers can also transmit to and serve as a source of water for lakes, streams and other aquifers.
An example of aquifer recharge is visible in the Mimbres Basin Aquifer, which underlies the border-states of New Mexico in the US and Chihuahua in Mexico. Due to the high level of evapotranspiration of this arid environment, only a small percentage of basin-wide precipitation and surface runoff actually reaches the aquifer. Most aquifer recharge occurs in the upland area in the northern part of the basin where temperatures and evapotranspiration are relatively lower. Sources of recharge in the area include the only major perennial stream in the Mimbres Basin system, the Mimbres River, and a few intermittent streams, like the San Vicente Arroyo.
This exchange between surface and subsurface water resources is not unique, and is important in that conditions affecting the quality and quantity of the water on one side of the relationship can have consequences on interrelated water resources. Moreover, it is very common to have mutual relationships between surface and underground water sources that vary in time and space. A river, for example, may discharge water into a related aquifer at one point of its course, and receive water from ground water at another; or a given stretch of a river may be discharged into an aquifer during the autumn season and receive water in the spring.
C. Ground water as an International Resource
With some exception, most aquifers regularly receive and transmit water as part of the hydrologic cycle. This is not to say that all aquifers are interconnected with surface water. Nevertheless, in the international context, it is rare that a transboundary river is not somehow linked to domestic or shared ground water resources.
Barberis, in his well-known 1986 study for the UN Food and Agriculture Organization, offers four cases to illustrate the transboundary nuances associated with ground water resources:
- "a confined aquifer is intersected by an international boundary, and is not linked hydraulically with other groundwater or surface water, and, as such, it alone constitutes the shared natural resource;"
- "an aquifer lies entirely within the territory of one State but is hydraulically linked with an international river;"
- "the aquifer is situated entirely within the territory of a single State and is linked hydraulically with another aquifer in a neighboring State;"
- "the aquifer is situated entirely within the territory of a given State but has its recharge zone in another State."
While the underlying premise highlighted by Barberis’s cases is correct (that ground water resources can have substantial international implications), and while Barberis’s study has been widely cited (including in the proceedings if the International Law Commission, which drafted the Watercourse Convention), the cases are imprecise and require additional refinement and clarification to conform to the current knowledge of the science of ground water.
Barberis’s first example is imprecise in that he lumps together all unrelated confined aquifers. Such aquifers, however, must be subdivided into two categories based on their relationship to the hydrologic cycle: those that are a dynamic component of the hydrologic cycle (despite being unrelated to any other body of water), and those that are static bodies of water devoid of any connection to a source of recharge. The basis for this categorization is important as the law applicable to these two aquifer types may not necessarily be the same. As will be discussed below, a static body of water unconnected to the hydrologic cycle may be subject to a legal regime more akin to the law of transboundary oil and gas resources rather than to international water law.
Barberis’s third example is inconsistent with the science of ground water as it suggests that a hydraulic link can exist between two adjacent aquifers. Any "link" described between two adjacent aquifers is necessarily a component of both aquifers. Thus, rather than two "linked" aquifers, the example implicates one large transboundary aquifer. To the extent that Barberis’s first case also suggests the same scenario, it also is scientifically inaccurate.
Barberis’s fourth case, which suggests that an aquifer and its recharge zone are identifiable as a separate water resources, is similarly inconsistent with hydrogeology. An aquifer’s recharge zone is necessarily a component of the aquifer. The fact that the recharge zone lies in another state does not create two identifiable water resources. Again, the result is one aquifer traversing an international boundary.
Barberis intended the four cases to be representative of the main cases in which ground water resources have transboundary implications. Clearly, this list is incomplete. The cases fail to account fully for other aquifer types that have possible transboundary implications, including aquifers: that are unconfined and unrelated to another water resource, that are confined and unrelated to another water resource, and that are non-renewable. Barbaris's examples, however, are useful as a starting point from which to develop more refined and precise models based on principles of hydrogeology and actual examples of aquifers.
In the following, six models are proposed to illustrate the main cases in which ground water resources can have transboundary implications. These six models are representative of the vast majority of aquifers existing on Earth.
1) An unconfined aquifer that is linked hydraulically with a river, both of which flow along an international border (i.e., the river forms the border between two States). Examples of this model include: the Red Light Draw, Hueco Bolson, and Rio Grande aquifers underlying the United States and Mexico; the Danube alluvial aquifer between Croatia and Serbia.
2) An unconfined aquifer intersected by an international border and linked hydraulically with a river that is also intersected by the same or another international border. Examples of this model include: the Abbotsford-Sumas Aquifer shared between Canada and the United States; the Mures/Maros Aquifer underlying Hungary and Romania; the San Pedro Basin Aquifer shared between Mexico and the United States.
3) An unconfined aquifer that flows across an international border and that is hydraulically linked to a river that flows completely within the territory of one state. Examples of this model includes: the Mimbres Basin Aquifer traversing northern Mexico and the U.S. state of New Mexico.
4) An unconfined aquifer that is completely within the territory of one state but that is hydraulically linked to a river flowing across an international border (in such cases, the aquifer is almost always located in the downstream State). Examples of this model include: the Gila River Basin Aquifer underneath parts of Arizona, California, Nevada, and New Mexico in the United States.
5) A confined aquifer that traverses an international boundary with a recharge zone (possibly in an unconfined portion of the aquifer) located in one state; or a domestic aquifer with a recharge zone (possibly in an unconfined portion of the aquifer) that traverses an international boundary. Examples of this model include: the series of deep, confined aquifers in the Syr Darya River Basin; the Mountain Aquifer between Israel and the Palestinian Territories; and the Guarani Aquifer underneath Argentina, Brazil, Paraguay and Uruguay.
6) A transboundary aquifer constituting a non-renewable water resource. Such aquifers contain paleo or ancient waters and may be confined or unconfined, and fossil or connate. The key feature is that they are not a part of the hydrologic cycle. Examples of this model include: the Nubian Sandstone Aquifer underneath Chad, Egypt, Libya, & Sudan; the Complex Terminal Aquifer underlying Algeria and Tunisia, and possibly Libya and Morocco; the Continental Interclaire Aquifer underlying Algeria and Tunisia, and possibly Libya and Morocco; the Qa-Disi Aquifer underlying southern Jordan and northern Saudi Arabia
The purpose of these models is to help in the evaluation of the applicability and scientific soundness of proposed and existing rules governing shared ground water resources. Through such analyses, it is hoped that the models assist in the development of clear, logical, and appropriate norms of state conduct.
III. Ground Water Under the UN Watercourse Convention
A. Developing the Scope of the Convention
The UN Watercourse Convention was designed as a framework Convention to provide guidance for more specific bilateral and regional agreements relating to the use, management, and preservation of transboundary watercourses. It was developed to promote sustainable development and protection of global water supplies, as well as to help prevent and resolve conflicts over transboundary water resources. Significantly, the drafters of the Watercourse Convention, members of the UN’s International Law Commission, were instructed to "take up the study of the law of the non-navigational uses of international watercourses with a view to its progressive development and codification …"
Considering this background, the regulation and management of ground water resources is undeniably a proper subject of the law of transboundary watercourses. This is clearly evident when activities in one state’s waters, whether surface or ground waters, can significantly impact interrelated surface and/or ground waters in other states.
The scope of the Watercourse Convention, including the geographic unit (within which the rules relating to uses of a specific international watercourse would be applicable) and the question of whether and to what extent ground water should be included within the regime, were significant issues for the International Law Commission (ILC or Commission). That the drafting work of the ILC took nearly twenty-five years is just one indication of the intricacies, as well as of the importance that states attribute to the subject matter. As pointed out in the 1974 report of the SubCommittee on the Law of the Non-navigational Uses of International Watercourses, various terms had been used in prior treaties and reports of international organizations and conferences, including "successive international rivers," "contiguous international rivers," "river basins," "international drainage basin," and "hydrographic basin." Debate over the term and scope, in fact, resulted in considerable dissent amongst the participants and required postponement of the subject on at least two occasions.
In 1976, following extensive disagreement within the ILC, the Commission decided "that the question of determining the scope of the term ‘international watercourses’ need not be pursued at the outset of the work. Instead, attention should be devoted to beginning the formulation of general principles applicable to legal aspects of the uses of those watercourses." The sticking point among the members was the use of the "drainage basin" framework as the geographical context for international agreements. Employed by the Helsinki Rules, ground water was specifically included to the extent that it was part of the "system of waters" and flowed to a "common terminus". This approach, however, had already been rejected at the outset of the Commission’s work when it considered using the Helsinki Rules as a model.
Focusing on the boundary implications of international watercourses, many of the States that rejected the drainage basin concept argued for the meaning given to an international river in the Final Act of the Congress of Vienna (1815). The Final Act provides that an international river is "a river that separates or traverses the territory of two or more States." Notably, most of the States supporting the Final Act concept were upper riparians, while those supporting the drainage basin model were downstream States.
In 1979, a "watercourse system" approach was proposed as a substitute for the drainage basin. The watercourse system was defined broadly as a network of rivers, lakes, canals, glaciers, and ground waters. While the focus on the basic unit of the "watercourse" was sufficiently analogous to the channel-based model, its emphasis of a "system" of waters was intended to satisfy those who advocated for a unitary approach toward managing interconnected waters, including ground water. In order to encourage progress in the development of the text, in 1980 the ILC accepted the definition as a working hypothesis, albeit as a provisional measure.
The "watercourse system" approach, however, also proved controversial for its drainage basin-like approach and was criticized as mere semantic maneuvering. As a result of strong resistance, the ILC at one point dropped the word "system," but subsequently reintroduced in square brackets. This was done with the understanding that that the scope and the use of the term "system" would be addressed at a later time.
In 1991, the ILC returned to the issue of the scope of the Convention and considered various definitional alternatives. Following considerable discussion, the Commission dropped the working hypothesis and reached agreement on a definition, which, with some revision, ultimately prevailed in the final version adopted by the General Assembly. The definition in the Watercourse Convention states that a "‘watercourse’ means a system of surface waters and ground waters constituting by virtue of their physical relationship a unitary whole and normally flowing into a common terminus." For the purpose of the Convention, this definition is qualified by the definition of "International watercourse," which "means a watercourse, parts of which are situated in different States."
B. Ground Water and the Scope of the Watercourse Convention
The inclusion of "ground waters" in the Watercourse Convention serves as a substantial clarification of the status of ground water under international law. It is, in fact, the first official codification evidencing that ground water resources can be legitimately regulated by international law. Under the definition of "watercourse, however, it is evident that not all ground water falls under the rubric of the Convention and that there are strict criteria that help define whether one or another type of ground water is included. These criteria will now be explored with regard to what is explicitly included within the scope of the Convention, and what is explicitly excluded.
1. "system of surface waters and ground waters" and "constituting by virtue of their physical relationship a unitary whole"
The first criterion, the phrase "a system of surface waters and ground waters," appears to be a follow through by the rappertour to reintroduce the system approach to water resource management, which had been tabled in the ILC’s discussion in the mid 1980s. As noted above, this approach advocates a unitary or comprehensive management scheme of interconnected waters. While self-limiting, this standard nonetheless acknowledges the interrelationship of surface and underground water within the hydrologic cycle. The emphasis on a comprehensive formula to water resource management is further articulated in the definition by the phrase "constituting by virtue of their physical relationship a unitary whole." When read in concert with the definition for "international watercourse," it appears evident that for the Convention to apply, it is not necessary for a particular aquifer or for an interrelated surface body of water to traverse an international boundary so long as the system, or any one of its interrelated components (i.e., an interrelated aquifer or river or lake) traverses or flows along an international border.
In the context of the six models provided above, the first four clearly fall within the reach of the Convention. All involve a system of surface and underground waters of which, at least, one "part" is "situated in a different State[]". In example one and two, both the aquifer and the linked river are transboundary; in example three, the aquifer is the transboundary part; and in example four, the river is the transboundary part.
Despite the expanse of the terminology, the two phrases in the definition of an international watercourse also impose very specific limitation on the scope of the Convention. In particular, the phraseology appears to restrict the scope of the Convention only to "systems," and only to systems that have a "physical relationship" between the inter-linked components. This phraseology begs a number of questions, pertaining to the applicability of the Convention to different types of ground water resources, that are not easily answered. Foremost of the questions is whether a system relationship must exist between an aquifer and a surface body of water for the aquifer to fall under the Convention. In other words, can a solitary aquifer constitute a "system" in and of itself, and can its interaction within the hydrologic cycle (although not with any surface bodies of water) constitute, by virtue of this interaction, "a unitary whole?"
The comments to the final Draft Articles submitted by the last ILC Rapporteur to work on the Draft Articles provide little guidance on these significant issues. The closest language addressing these questions contend that, "[i]t follows from the unity of the system that the term ‘watercourse’ does not include [water that is] unrelated to any surface water." Thus, it appears that a system may necessarily be composed of more than one identifiable and interconnected water resource. The statement quoted here, however, is provided in the Draft Articles in reference to the exclusion of so-called "confined" ground water from the scope of the Convention (an issue addressed separately below) and only focuses on the "unity" of the system; it does not necessarily shed light on a ground water "system" unrelated to surface water.
In the context of the six models, these limitations would likely exclude the aquifers described in examples five and six. While the aquifers in the two examples both traverse an international boundary, thus fulfilling the "international" criteria, neither are related with or linked to any other identifiable surface body of water. Although exclusion of model six may be justifiable, since the resource is static and unreplenishable, such a result is troubling where the aquifer is a dynamic component of the hydrologic cycle. Model five, like models one through four, describes an aquifer that has the potential for sustainable pumping or discharge, and that has a sustained source of recharge. The Mountain Aquifer shared between Israel and the Palestinians is one such example. As a solitary aquifer fed solely by rain in the highlands of the Judean Mountains, the aquifer is not related to any surface water. Its exclusion from the scope of the Convention would leave the status of these types of transboundary aquifers unclear, or, at best, subject to ad hoc agreements by the aquifers’ riparians. Clearly, such an outcome would contravene the purpose of the UN in commissioning the codification of international water law and developing clear, equitable, and predictable rules of state conduct.
2. "flowing into a common terminus"
The second criterion under the Watercourse Convention applicable to ground water is the phrase "flowing into a common terminus." While this terminology was not included in the definition until very late in the Commission’s work, it is not a new standard. In fact, the same phrase appears in Article II of the Helsinki Rules, which defines the scope of the Rules. The phrase was added to the Convention’s definition, in part, because of concerns raised by various states that a geographic limitation was necessary. In particular, some states argued that, "In a single State where most of its rivers were connected by canals, the absence of the requirement of common terminus would turn all those rivers into a single system and would create an artificial unity between watercourses." By embracing the qualifying phrase "flowing into a common terminus," two different watercourses connected by a canal could not be regarded as a single watercourse for the purposes of the Convention.
Notwithstanding the justification, the limitation created by the phrase is particularly vexing since it perpetuates the illogical distinction all too often made between surface and underground waters. From both a legal and scientific point of view, the flow of water in a stream is no different than the flow of water though an aquifer – it is simply water moving through different stages of the hydrologic cycle. The only noteworthy differences concern water chemistry, rate of flow, different technology of exploitation, and absence of navigability, matters for which there is no reasonable justification to create management regimes for ground water disparate from that for surface water.
Furthermore, this limitation of "common terminus" raises serious concerns in the context of ground water since underground water resources comprising part of a watercourse do not always flow to the same terminus as interrelated surface water. As noted earlier, ground water flow is a factor of hydraulic potential, which in turn is dependent on gravity, soil permeability, gradient, ambient air pressure, and temperature. Thus, ground water may flow in a direction different from that of a related surface body of water. As described in the Danube River example above, the surface water flows to a terminus in the Black Sea, while in the upper region of the Danube, ground water often seeps out and flows to a terminus in the North Sea. Based on the Convention’s definition of watercourse, any use or management scheme developed for the upper reaches of the Danube River would not be bound by the Convention’s principles with regard to the related ground water flowing toward the Rhine River. While this scenario may raise unique complications, consequences to the Rhine River are not inconceivable.
Considering the interpretation of the phrase suggests that, as required by the "system" approach and the "physical relationship" standards, here, too, there must be more than one water resource that are interrelated. The term "common" clearly intimates a sharing of characteristics between two or more distinct waters. Moreover, based on the above discussion addressing the "system" approach and the "physical relationship" criteria, one of the linked water resources must be a surface body of water against which flow toward a common terminus is gauged. In the context of the six models, models five and six fall outside the scope of the Convention because the aquifers described have no direct relationship to any surface water; even if such aquifers do flow toward some identifiable terminus, they do not have any interrelated surface water against which to gauge flow toward a common terminus. The application of the Convention to the remaining four models, however, is purely dependent on the flow of the two interrelated water resources.
It should be noted that the limiting phraseology of "flowing to a common terminus" is somewhat attenuated by the addition of the prefix "normally." In placing the term "normally" before the phrase, the ILC sought to achieve a compromise between, on the one hand, those who pushed for the deletion of the phrase "flowing into a common terminus" on the grounds that it was hydrologically incorrect, potentially misleading, and could exclude certain important waters, and, on the other hand, those who urged retention of the phrase in order to maintain some measure of geographic limitation to the application of the Convention. It is unclear whether this attenuation is of consequence since the word "normally" is not defined. Yet, in the case of the Danube River, since the seepage is seasonal, this additional modification merely adds more uncertainty.
3. unrelated ground water and "confined" ground water
One of the specific exclusions of the Convention regards ground water that is unrelated to any surface water. The comments to the final Draft Articles note that, "[i]t follows from the unity of the system that the term ‘watercourse’ does not include ‘confined’ ground water, meaning that which is unrelated to any surface water." Thus, ground water that is not part of a "system of surface waters and ground waters constituting by virtue of their physical relationship a unitary whole" is not encompassed within the scope of the Convention. This intentional exclusion was rationalized on the basis that unrelated ground water cannot have any untoward effects on any other watercourse.
While logical on the surface, the justification is short sighted in that it ignores the transboundary implications of aquifers unrelated to surface waters such as those described in model five. Although the example has no hydraulic relationship to any surface body of water, it clearly can have cross-border implications. Excluding such aquifers would exclude numerous transboundary aquifers of considerable importance, especially in water-poor and arid regions of the world like Northern Africa, the Middle East, and Central Asia. The Mountain Aquifer underlying Israel and the West Bank is but one such example.
A related matter concerns the terms "confined" and "unconfined" ground water as used by the ILC members and as incorporated implicitly into the Convention. Generally, the term "confined" ground water relates to an aquifer overlain and underlain by relatively impermeable strata. The Commission, however, equated the term with ground water that has no hydrological relationship with an international watercourse. While the etymology of the Commission’s use of the term is uncertain, it is clear that their definition does not correspond with the hydrological definition.
As discussed above, confined aquifers often are recharged via portions of the aquifer that are unconfined, or through lateral flow from higher elevations where the aquifer crops out on the land surface. Nevertheless, the term was misapplied through portions of the ILC’s and the Rapporteurs’ comments and discussions, as well as in a number of professional publications, and was later memorialized in the ILC Resolution on Confined Transboundary Ground Water, in which the ILC recognized "confined" ground water as "ground water not related to an international watercourse."
While possibly a simple technical infraction, the result of this and similar misstatements could lead to further misuses of definitions and, possibly, misapplications of international law. That such an error occurred at the level of the United Nations questions the very understanding of basic hydrogeology by the drafters of the Watercourse Convention, as well as by members of the General Assembly. While it is not inconceivable given the technical nature of the subject, the implications of such misunderstanding in the United Nations are significant and far-reaching.
C. Non-Renewable Ground Water
Non-renewable ground water, as discussed above, is water contained in an aquifer that is completely detached from the hydrologic cycle. Such aquifers have little or no possibility for natural recharge and, by definition, cannot be utilized sustainably; any withdrawal from these aquifers eventually will exhaust the resource. The Nubian Sandstone aquifer in northeastern Africa, for example, is not part of the hydrologic cycle and has no source of recharge. Similarly, the Complex Terminal Aquifer underlying Algeria and Tunisia, the Qa-Disi Aquifer underneath Jordan and Saudi Arabia, the Continental Interclaire Aquifer underlying Algeria and Tunisia, and the Northern Sahara Aquifer underlying Algeria, Libya, and Tunisia are all depleatable transboundary aquifers. While these aquifers are pumped by many of the countries sharing their waters, they are slowly being depleted and could eventually dry up.
Due to these particular characteristics, non-renewable ground water is described as akin to other depletable natural resources, like oil and natural gas. Like oil and gas, non-renewable ground water is a static, depleatable, fluidic natural resource. Accordingly, the legal regime applicable to replenishable ground water is not applicable to non-renewable ground water. Rather, rights, allocation, and use regime should follow the model of other non-renewable natural resources law, like oil and natural gas.
This discussion of the status of non-renewable ground water resources under international law is a neglected topic in international legal discourse and regulatory development. This author has found no scholarship on the subject other than secondary remarks of the inapplicability of mainstream international water law to non-renewable ground water resources. Likewise, the Watercourse Convention indirectly omits consideration of such ground water from its scope by excluding water unrelated to surface water, and water that flows to a common terminus. Non-renewable ground water, by definition, does not fall within either criterion. Moreover, it is likely that in considering "confined" ground water, the Commission, in fact, was referring to non-renewable ground water.
Although the global reserves of fresh water stored in such transboundray non-renewable aquifers is unclear, suffice it to say that it constitutes a highly important water source for many nations, and often the only viable source of fresh water. Scientific and legal logic suggest that a legal regime akin to that applicable to oil and natural gas resources may be more appropriate than that applicable to ground water connected to the hydrologic cycle. Given the ambiguity and lack of attention to the subject, more comprehensive and critical consideration is in order.
IV. Conclusion
While the Watercourse Convention was not intended as a treaty specifically for shared ground water resources, their inclusion within the rubric of the Convention is a significant step in expanding our focus to and our knowledge of this unique resource. The importance of the Convention is especially evident in its substantial elucidation of the current status of ground water under international law.
As a means of evaluating the Watercourse Convention vis vis shared ground water resources, six models, representing the vast majority of aquifers existing on Earth, were proposed above. The purpose in devising the models was to illustrate the main cases in which ground water resources can have transboundry implications. Moreover, the models were used as standards against which to consider the applicability and scientific soundness of the Convention’s provisions as they related to ground water.
Taking these models into account, as well as the science of ground water, the above discussion suggests that three general criterion can be identified for an aquifer to fall within the scope of the Watercourse Convention:
a. the aquifer must be hydraulically linked to a system of waters that includes a surface body of water;
b. the aquifer need not traverse an international boundary so long as any part of the system to which it is hydraulically linked crosses an international boundary;
c. the aquifer must normally flow to the same terminus as the hydraulically linked surface water.
Based on the analysis, however, it is also clear that as much as the Watercourse Convention clarifies the status of ground water under international law, it also generated new questions and concerns. The most troubling are those that appear inconsistent with the science of ground water.
- The Convention excludes from its scope all ground water that is unrelated to surface water. Such a blanket exclusion is particularly troubling in that it ignores the transboundary implications of aquifers that, although are not related to surface water, are still dynamic components of the hydrologic cycle.
Aquifers, such as those of the Syr Darya River Basin and the Mountain Aquifer between Israel and the Palestinian Territories, for example, are recharged across political borders in higher elevations where surface runoff percolates into exposed areas of the aquifers. These aquifers, however, are not linked to any surface body of water and therefore are not covered by the provisions of the Convention. Contrary to the purpose of the Convention, this exclusion allows states to embark on strategies designed to use, regulate, or manage a particular water resource without regard to the consequences the action might effect on neighboring states.
- The Convention requires that related ground and surface waters must normally flow to a common terminus to be subject to the Convention’s provisions. This criteria is troubling since it is not uncommon for such water resources to flow to different termini based on hydraulic potential, soil permeability, gradient, and other factors. The fact that such related water resources flow toward disparate termini does not negate the international implications, especially where transboundary pollution is a concern.
A chemical spill in the upper region of the Danube River, for example could have serious consequences for the Rhine River. Nevertheless, because the Danube River in the Black Forest region flows to a termini different than that of the related groundwater (which toward the Rhine River), this scenario is not covered by the provisions of the Convention. This dilemma is further obscured by the modifier "normally," which is undefined in the Convention.
- The ILC inaccurately used the term "confined ground water" to define ground water that is unrelated to surface water. "Confined" ground water, however, is a hydrogeologic term used to describe an aquifer overlain and underlain by impermeable strata.
The use of different terminology in a subject requiring thorough scientific understanding is a significant concern. At the least, it suggests a misunderstanding of the science underlying the subject of water law. The need for thorough understanding of the relevant hydrological and hydrogeological bases, however, is readily apparent given the intricacies of the hydrologic cycle. This is especially important when developing and formulating regulatory schemes and guidelines designed to the sustainable use and protection of scarce water resources.
- The long standing preoccupation with surface waters by statesmen and scholars has prevented a more comprehensive and hydrogeologically sound approach to international water law. This fixation is readily apparent in the Convention’s precise focus on the "watercourse" and surface bodies of water, and a more vague and self-limiting consideration of ground water resources. This focus is also suggested in the positions held by many ILC members who, for example, sought to exclude ground water from the scope of the Convention.
Ultimately, the above analysis suggests that while development of the Watercourse Convention greatly increased international attention on global ground water resources, there is still a great deal of misunderstanding of the subject. An inadequate level of scientific knowledge among legislators, policy-makers, jurists, and scholars may generate rules and legal principles with little scientific underpinning, and as a consequence, can result in inadequate use, management, and protection schemes for shared natural resources.
As science develops and operates on the frontiers of knowledge, the law must keep apace and must continually adapt to new scientific discoveries and developments. Only through the full understanding of the various legal, policy and scientific issues involved will states be able to use, manage and protect their shared resources appropriately, effectively, and in such a way that the resources suffice for present needs and are preserved for future generations.