2.1 UNCLOS

The legal basis for boundaries fluctuating by operation of law is distinct depending on whether one considers, on one hand, boundaries in the territorial sea and, on the other hand, boundaries in the EEZ and continental shelf.

Articles 74 UNCLOS states that:

Article 83 UNCLOS is identical, save for referring to the continental shelf instead of the EEZ. Owing to their vagueness, these two provisions seem to give no indication of a possible legal basis for EEZ and continental shelf boundaries to fluctuate by operation of law. ^{Footnote 27} The position under Articles 74 and 83 seems to be that a boundary ought to be equitable at the time when the relevant states reach the agreement to delimit that boundary.

The situation may be different in relation to the territorial sea. With respect to territorial sea delimitation, Article 15 UNCLOS states that:

According to Dr. Árnadóttir, this provision ‘establishes a bilateral limit to the territorial sea and explains “how States must act failing an agreement on delimitation”’, ^{Footnote 28} from which it would follow that ‘an equidistance line can function as a provisional arrangement until either state initiates the delimitation of a more definitive boundary’. ^{Footnote 29} This equidistance line ‘would assumedly fluctuate in accordance with ambulatory baselines to continuously reflect relevant coastlines, unless and until otherwise agreed’. ^{Footnote 30} These remarks appear to convey that Article 15 UNCLOS establishes median lines as provisional boundary arrangements in the territorial sea, which coastal states can change only by agreement to the contrary. These provisional boundary arrangements could be understood as de facto provisional territorial sea boundaries.

Article 15 UNCLOS does not expressly state that territorial sea boundaries exist irrespective of agreement between the relevant states. All that Article 15 explicitly does is defining the scope ratione loci of the rights and duties of states in the territorial sea: if a coastal state has concluded no agreement with its neighbouring states delimiting the territorial sea, it enjoys none of the rights that coastal states have in the territorial sea, and other states owe it none of the duties that flag states have in the territorial sea, beyond the median line between that state’s coast and the coast of its neighbouring states. Article 15 does not govern the relationship only between the states whose coasts generate maritime entitlements overlapping in the territorial sea, but also between those states and other states whose main interest is ensuring that ships flying their flag can enjoy innocent passage through undelimited territorial sea entitlements. Article 15 does not establish median lines as provisional territorial sea boundaries, which can be changed only by agreement of the coastal states concerned. Stating that Article 15 creates ‘a provisional arrangement’ in the form of a median line seems to force the meaning of its text.

This understanding finds support in the absence, in Article 15, of a provision comparable to Articles 74(3) and 83(3) of UNCLOS. ^{Footnote 31} Articles 74(3) and 83(3) state that, pending agreement on the delimitation of the EEZ or continental shelf, ‘the States concerned … shall make every effort to enter into provisional arrangements of a practical nature and, during this transitional period, not to jeopardize or hamper the reaching of the final agreement’. Discussing Article 83(3), the Special Chamber of the International Tribunal for the Law of the Sea (ITLOS) in Ghana/Côte d’Ivoire stated that it includes an obligation ‘designed to promote interim regimes of a practical nature pending delimitation’. ^{Footnote 32} According to the Special Chamber, the text of this provision ‘clearly indicates that it does not amount to an obligation to reach an agreement on provisional arrangements’. ^{Footnote 33} Because UNCLOS provides expressly for provisional arrangements in relation to the EEZ and continental shelf, it seems odd to suggest that similar arrangements exist in relation to the territorial sea despite the lack of a clear provision to that effect. ^{Footnote 34} Furthermore, the suggested automaticity of provisional arrangements under Article 15 UNCLOS seems inconsistent with the character of the territorial sea as a maritime zone under the coastal states’ sovereignty. ^{Footnote 35} With respect to the EEZ and continental shelf, where coastal states only exercise sovereign rights and jurisdiction, ^{Footnote 36} UNCLOS imposes no obligation to reach agreements on provisional arrangements. The inexistence of such an obligation in relation to maritime zones over which states do not have sovereignty suggests the lack of automatic provisional arrangements in relation to the only maritime zones over which states exercise sovereignty.

The UNCLOS drafting history also suggests that Article 15 does not provide for the automatic establishment of territorial sea boundaries unless and until coastal states reach an agreement to delimit those boundaries. Article 15 UNCLOS is based on Article 12 of the 1958 Convention on the Territorial Sea and Contiguous Zone, ^{Footnote 37} which, in turn, is based on Articles 12 and 14 of the 1956 Draft Articles concerning the law of the sea prepared by the International Law Commission (ILC). ^{Footnote 38} The interpretation of Article 15 UNCLOS should be informed by an understanding of its 1956 and 1958 predecessors.

The 1956 Draft Articles distinguished territorial sea delimitation between opposite coasts (Draft Article 12) and between adjacent coasts (Draft Article 14), but the ILC had envisaged the median line to be the delimitation method in both scenarios unless special circumstances justified a departure from it. ^{Footnote 39} The ILC’s commentary to Draft Article 12 recognized that departures from the median line on account of special circumstances would have been frequent, adding that it had seemed anyway advisable ‘to adopt, as a general rule, the system of the median line as a basis for delimitation’. ^{Footnote 40}

The explanation that the median line was a mere ‘basis for delimitation’ falls short of suggesting that the ILC had conceived Draft Article 12 as automatically creating a provisional boundary. The ILC appears to have written Draft Article 12 either as a guideline for negotiating states, or as a rule to be applied by international tribunals. The ILC also stated that there would have been frequent departures from the median line on account of special circumstances. ^{Footnote 41} The very presence of special circumstances may complicate, and even prevent, reaching an agreement on territorial sea delimitation. If the rule governing territorial sea delimitation provided for the automatic creation of provisional median-line boundaries, the complexity of a case due to special circumstances could result in the effective permanence of a provisional median-line boundary. This mechanism would contradict the rationale justifying the introduction of the notion of special circumstances: although special circumstances warrant departing from median-line boundaries, the complexity determined by special circumstances could prevent states from reaching agreements delimiting territorial sea boundaries departing from the median line, which therefore would result in permanent median-line boundaries.

It is not entirely convincing to understand Article 15 UNCLOS as automatically establishing median-line boundaries unless and until the relevant states agree to the contrary. Consequently, one cannot consider that changes in the baselines determine the automatic fluctuation of territorial sea boundaries. UNCLOS does not provide a convincing legal basis for boundaries fluctuating by operation of law, either in the territorial sea, or in the EEZ and continental shelf.

2.2 Judicial decisions

As maritime boundaries do not fluctuate by operation of law, the question becomes whether they can fluctuate as a result of judicial decisions. At first glance, the answer to this question is in the affirmative. No rule of international law expressly prevents international tribunals from delimiting fluctuating maritime boundaries. Presumably, international tribunals could delimit fluctuating boundaries, provided that states have agreed to request them to do so, for example by special agreement. On closer analysis, this question raises an additional issue, namely whether there is judicial authority for the delimitation of fluctuating boundaries.

Dr. Árnadóttir wrote that two decisions of the International Court of Justice (ICJ or Court) are exceptions to the result of maritime delimitation being a stable boundary. ^{Footnote 42} In its 2007 judgment in Nicaragua v. Honduras, the Court would have ‘established a fluctuating boundary for the first segment of the boundary, from the land boundary terminus and seaward’, ^{Footnote 43} meaning that ‘the first segment was left to fluctuate unless and until otherwise agreed by the parties’. ^{Footnote 44} Nicaragua v. Honduras was a case in which a most prominent feature was high coastal instability at the mouth of the Rio Coco, ^{Footnote 45} a river constituting the eastern half of the land boundary between Nicaragua and Honduras. The Rio Coco empties into the Caribbean Sea forming a markedly convex coastal protrusion resulting from continued sediment accretion, known as Cabo Gracias a Dios.

Because of coastal instability around Cabo Gracias a Dios, both parties requested the Court to identify a fixed point out to sea, instead of their land boundary terminus on Cabo Gracias a Dios, from which a stable boundary would depart. ^{Footnote 46} Honduras requested that fixed point to be 3 nautical miles (nm) out to sea and also provided its geographical co-ordinates. ^{Footnote 47} Nicaragua requested ‘a starting-point located at the current mouth of the River Coco’, ^{Footnote 48} adding that ‘this starting-point, wherever it may be located on any given day, would then be connected by a straight-line single maritime boundary to the start of its proposed bisector line’. ^{Footnote 49} Nicaragua’s was the only solution that would have resulted in a fluctuating boundary. The Court found that ‘Nicaragua’s proposal … is problematic in certain respects and its initial suggestion to start the line some distance out to sea appears a more judicious solution’. ^{Footnote 50} The Court also stated that ‘[t]he Parties are to agree on a line which links the end of the land boundary as fixed by the 1906 Award and the point of departure of the maritime delimitation in accordance with this Judgment’. ^{Footnote 51}

The ICJ expressly opted for starting the maritime boundary at a fixed, not ambulatory, point out to sea. Furthermore, there is no indication that the Court, by directing the parties to agree on a line connecting the fixed starting point to their land boundary terminus, had envisaged that this line would be a fluctuating one. The Court specified that this line had to connect the fixed starting point to ‘the end of the land boundary as fixed by the 1906 Award’ of the King of Spain, ^{Footnote 52} which was also a fixed point, having the geographical co-ordinates identified by a Mixed Commission in 1962. ^{Footnote 53} The ICJ itself had used that point both ‘as the point where the Parties’ coastal fronts meet’, ^{Footnote 54} and as the point from which 3 nm would be measured to identify the starting point of the boundary out to sea. ^{Footnote 55} The only way for Nicaragua v. Honduras to result in a fluctuating boundary is if one accepts that the line connecting the fixed starting point out to sea is established by operation of law under Article 15 UNCLOS. However, maritime boundaries do not fluctuate by operation of law. ^{Footnote 56} Nicaragua v. Honduras is no authority for the establishment of fluctuating boundaries by judicial process.

Costa Rica v. Nicaragua raised issues similar to those in Nicaragua v. Honduras. The land boundary terminus was located at the mouth of the San Juan River. Coastal instability at the land boundary terminus meant that it was difficult to identify a stable starting point for the maritime boundary. ^{Footnote 57} Both parties had requested the Court to identify a starting point on land: for Costa Rica, that point was the one on its coast that was closest to the mouth of the San Juan River, as identified by the Court-appointed expert; ^{Footnote 58} for Nicaragua, that point was the land boundary terminus as it existed on 15 April 1858, when the two parties had concluded the Treaty of Limits demarcating their land boundary, even if that point was submerged by the sea. ^{Footnote 59} It was the Court who asked the parties to indicate, in their oral submissions, their views on a starting point out to sea. ^{Footnote 60} Costa Rica stated that a starting point out to sea could have been a viable alternative, in which case the Court should have connected that point to the mouth of the San Juan River by a mobile segment. ^{Footnote 61} Nicaragua stated that ‘the “hinge” solution is practicable’, ^{Footnote 62} and that ‘the line connecting the fixed point at sea with the actual land territory could be mobile, moving with natural changes in the coastline’. ^{Footnote 63} The Court decided that:

The Court thus fixed a starting point along the median line, between the baselines of both states, located at 2 nm from the land boundary terminus. ^{Footnote 65}

The decision in Costa Rica v. Nicaragua should be understood in the context of the parties’ replies to the Court’s question on the appropriateness of a starting point out to sea. Crucially, both states replied that a starting point out to sea, connected to solid land by a mobile line, was a viable option. In reaching its decision on this point, the Court could build on the parties’ agreement, different from Nicaragua v. Honduras. Conceivably, if the parties had not agreed on the viability of a mobile segment, the Court might have only identified a starting point out to sea, without connecting that point to solid land with a mobile segment, as it had already done in Nicaragua v. Honduras. The ICJ’s judgment in Costa Rica v. Nicaragua does not seem to be authority for the delimitation of fluctuating boundaries, but for the narrower proposition that the Court may establish such boundaries when the parties agree to it.

To date, there has been no judicial decision on maritime delimitation establishing fluctuating maritime boundaries. The decisions relating to fluctuating boundaries, Nicaragua v. Honduras and Costa Rica v. Nicaragua, were much narrower in their rationale than would be necessary in order to conclude that there is judicial authority for establishing boundaries that change with changes in the coastlines. Anyway, there could be room for international tribunals to delimit wholly or partly fluctuating boundaries in the future. Whether they will do so seems to depend, to a great extent, on the arguments of the parties to the relevant disputes. The ICJ accepted to delimit a fluctuating boundary where the parties had agreed, but was reluctant to do so where the parties held opposing views on the matter. The lack of express support for delimiting fluctuating boundaries in the judges’ individual opinions suggests that the ICJ is unlikely to delimit anything other than stable maritime boundaries, at least for the foreseeable future.

3.1 Delimiting fluctuating boundaries by judicial process

In the exercise of their treaty-making autonomy, states may agree to delimit fluctuating boundaries. Because boundary negotiations are confidential, the methodologies that states employ to delimit boundaries, fluctuating or not, are rarely in the public domain. ^{Footnote 66} Conversely, the international tribunals’ methodologies for delimiting maritime boundaries are well known.

The methodology for territorial sea delimitation is in two stages: first, international tribunals establish a provisional equidistance line by reference to base points located on the relevant coast of the states concerned; second, international tribunals may adjust that provisional line on account of special circumstances. ^{Footnote 67} With respect to the delimitation of the EEZ and continental shelf, the judicial methodology involves three stages: first, international tribunals establish a provisional equidistance line, similar to delimitation in the territorial sea while including the identification of the relevant coast and relevant area; ^{Footnote 68} second, international tribunals may adjust that line in the light of relevant circumstances; ^{Footnote 69} third, international tribunals assess the equitableness of the boundary by reference to the inexistence of gross disproportionality between the length of the states’ relevant coasts and the maritime areas found to appertain to each state. ^{Footnote 70} International tribunals delimit boundaries in the territorial sea, EEZ and continental shelf on the basis of the geographical conditions at the time of delimitation. ^{Footnote 71}

There appear to be two ways for international tribunals to delimit fluctuating boundaries. First, when implementing the delimitation methodology international tribunals may take into account that coastlines will change in the future. This approach would require abandoning the position that maritime boundaries are established on the basis of the geographical conditions at the time of the delimitation. Second, international tribunals may delimit maritime boundaries as they have done until now and leave it to states to update such boundaries in the future, if and when geographical changes justify it. This approach would be consistent with the position that maritime boundaries are established on the basis of the geographical conditions at the time of the delimitation. ^{Footnote 72} The first way would require international tribunals to adjust how they implement the delimitation methodology. Issues arising in the adjustment process would affect all stages of the delimitation methodology, discussed in turn, and would likely require impracticable solutions.

Accounting for shifting coastlines would complicate identifying the relevant coast. International tribunals identify the relevant coast either along the low-water line, ^{Footnote 73} or by approximating the coast by means of straight lines. ^{Footnote 74} Taking into account shifting coastlines in the former approach would require international tribunals to consider what coastlines will look like in the future. There is no indication as to the appropriate future timespan to be considered. The longer the relevant timespan, the more the coast will likely change, the more different will the relevant coast be with respect to the one at the time of delimitation, whether because of erosion or accretion. Another time-related issue would concern the intervals at which international tribunals would have to ‘freeze’ the future picture of the coastline to assess its changes with respect to earlier times. Assessing the coast at weekly intervals, for instance, is likely too minute and time-consuming an exercise, but to assess the coast at intervals of years, or even decades, risks resulting in a distorted understanding of how changes in the coastline have actually taken place.

Furthermore, assessing how coastlines might shift in the future raises evidentiary questions. Coastlines can change not only by natural phenomena, but also owing to human interference; for example, engineering works to protect ports located near a river mouth could affect the supply of sediments brought downstream by that river. The variety of factors capable of determining changes in the shape of coastlines complicates assessing how such changes are likely to take place. In turn, this assessment raises questions concerning both the type of evidence capable of showing the coastline’s evolution and the method for international tribunals to assess this evidence. Expert evidence appears to be most suited to addressing scientific matters such as changes in coastal morphology. However, scholars have shown that presenting expert evidence can have shortcomings, including the international tribunals’ limited expertise in assessing it. ^{Footnote 75} In relation to climate change, a lot may depend on whether mitigation measures will be implemented and prove successful. Moreover, the processes by which ice sheets melt are not linear and not entirely understood within the scientific community. ^{Footnote 76}

Approximating the relevant coast by means of straight lines might avoid the full extent of the problems arising where the relevant coast is identified along the low-water line. Relevant coasts approximated by straight lines can remain a representation of the coast for longer than relevant coasts identified along the low-water line: changes in the coastal geography may entail significant changes in the shape and length of the low-water line, but they may not be sufficiently extensive to justify adjustments to straight lines representing the relevant coast. Approximating the relevant coast by straight lines could be a safer option for international tribunals willing to take stock of the coastline’s potential for future change. Nonetheless, the questions arising in respect of identifying a changing relevant coast along the low-water line would still persist, to some degree, where the relevant coast were approximated using straight lines. Moreover, the approximation approach is problematic in its own right, primarily because it does not result in an accurate depiction of the coast generating the maritime entitlements to be delimited. ^{Footnote 77} A solution could be to use straight baselines to approximate the relevant coast. ^{Footnote 78} Anyway, this solution would carry the same problems connected with identifying the relevant coast along a changing coastline, given the close relationship between straight baselines and the direction of the coast under Article 7(3) UNCLOS.

Identifying the relevant area would be similarly difficult. Similar problems would arise as would with regard to relevant coast identification, because the landward limit of the relevant area is the relevant coast. Additional problems would follow from the need to determine the limits of the relevant area out to sea. ^{Footnote 79} The outer limit of the relevant area may coincide with the outer limit of the EEZ or continental shelf, which depends on the baseline. Otherwise, the limits of the relevant area may be identified by reference to existing boundaries, but the question would be whether such boundaries would be stable over time or, similar to the relevant coast, also fluctuate. If existing boundaries fluctuated, international tribunals would have to assess not only the potential shifts in the relevant coast, but also the potential shifts in boundaries established with, or only between, third states not before them. A consequence of delimiting fluctuating boundaries would be to promote uncertainty in identifying the relevant coast and the relevant area.

Drawing a provisional equidistance line, both in the territorial sea and beyond, depends on the identification of suitable base points. It seems established that base points must be located on solid land, which excludes low-tide elevations. ^{Footnote 80} Nevertheless, what is solid land today may not be solid land anymore in the future. The maritime feature of New Moore/South Talpatty in Bangladesh v. India and the land boundary terminus identified by General Alexander in Costa Rica v. Nicaragua are examples from recent cases. If international tribunals drew provisional equidistance lines using base points that will exist into the future, they may have to look for such base points far inland. This approach appears unsatisfactory, as it would result in delimiting maritime boundaries with complete disregard for the geographical circumstances at the time of delimitation. Alternatively, international tribunals could draw multiple provisional equidistance lines, each representing what an equidistance line would look like at a given point in time in the future. In doing so, international tribunals would use distinct base points for each line, depending on which would be expected to exist at the relevant future time. It is not unheard for international tribunals to draw more than one provisional equidistance line in the same case. In Qatar v. Bahrain, the ICJ drew two such lines, depending on whether base points were placed on certain maritime features. ^{Footnote 81} However, one thing is to draw two provisional equidistance lines, while drawing several more is quite another. The more provisional equidistance lines an international tribunal were to draw, the more would it be possible for it to determine the boundary with greater precision. Questions of proof arising in respect of the relevant coast and relevant area would arise also in respect of drawing a provisional equidistance line.

Difficulties resulting from the uncertain identification of the relevant coast would also impact the assessment of certain special or relevant circumstances, at the second stage of the delimitation methodologies. Among the circumstances states most frequently invoke to justify adjustments to a provisional equidistance line, one finds marked disparity in the length of the coasts of the states concerned. ^{Footnote 82} If this disparity is sufficiently marked at the time of delimitation to warrant adjusting the provisional equidistance line, it should also be sufficiently marked at a future time. Nonetheless, the accuracy of this statement depends on the degree of coastal length disparity at the time of delimitation. In cases such as Nicaragua v. Colombia, in which the coasts of the two states stood in a 8.2:1 ratio, ^{Footnote 83} one might confidently expect that, all other circumstances being identical, coastal length disparities will remain sufficiently marked to justify adjusting the provisional equidistance lines for an appreciable duration of time in the future. One could not make the same confident assessment in a case such as Bangladesh/Myanmar, in which the coastal length ratio was 1:1.42 and the delta of the Ganges, along which ran most of the coast of Bangladesh, was known to change relatively rapidly. ^{Footnote 84}

The uncertainty stemming from the identification of the relevant area makes it problematic to assess disproportionality, at the third stage of the process for delimiting the EEZ and continental shelf. ^{Footnote 85} Disproportionality can be criticized for being ‘much ado about nothing’, because it has never led to adjusting boundaries which emerged from the first two stages of the delimitation methodology. ^{Footnote 86} Further criticism centres on the discretionary character of the disproportionality assessment. ^{Footnote 87} The reasons of disproportionality’s critics mainly focus on the indeterminacy of the standard and its limited impact on maritime delimitation by judicial process. Seeking to delimit maritime boundaries by taking into account future changes in coastal geography would likely dilute both the disproportionality standard, and its effects in the delimitation methodology. This problem is additional to the ones, already noted in relation to the previous stages of the methodology, of uncertainty and proof.

The current judicial process of maritime delimitation is ill-suited to being applied in the context of evolving coastlines. This view does not necessarily entail that maritime delimitation should never take place in such a context. Yet, it does entail delimiting maritime boundaries while taking into account changing coastlines requires, at least, considering seriously whether the established delimitation methodology can be applied meaningfully at all.

3.2 Updating fluctuating boundaries

The alternative to international tribunals delimiting maritime boundaries while taking into account future changes in the coastlines, states could take on the task of updating such boundaries after international tribunals have delimited them on the basis of the geographical conditions at the time of delimitation. This updating exercise, which would likely take place decades after the relevant judicial decisions, is an operation extending to maritime boundaries delimited by treaty.

Updating maritime boundaries is nothing short of delimiting an entirely new boundary. In the updating process, states would remain bound by the earlier judicial decision or treaty establishing the boundary to be updated, pursuant to the principles of res judicata and pacta sunt servanda. It would be only when the states concerned have reached an agreement on the updated boundary that they would not be obligated to comply with the earlier judicial decision or observe the earlier treaty. ^{Footnote 88} This legal framework may create imbalance in the negotiating power of the relevant states. Once coastlines change, an original maritime boundary will likely become more advantageous for one of the two relevant states, to the detriment of the other state. Cognisant of its acquired advantage, the former state could simply refuse to negotiate an updated boundary. Against this position, there seems to be little in the way of legal remedies.

Dr. Árnadóttir appears to suggest, on the basis of the ICJ’s judgment in Aegean Sea Continental Shelf, ^{Footnote 89} that maritime boundary treaties could be terminated on the basis of a fundamental change of circumstances, under Article 62 VCLT. ^{Footnote 90} While territorial sea and continental shelf boundaries would fall within the exception to the application of this rule, ^{Footnote 91} the same could not be said for EEZ and all-purpose boundaries. ^{Footnote 92} The problem of this view is that it builds upon a formalistic distinction between boundaries in different maritime zones. It is true that, because coastal states exercise sovereignty in their territorial sea, territorial sea boundaries are theoretically identical to land boundaries, in relation to which Article 62 VCLT had been conceived. In practice, territorial sea boundaries are no different from boundaries in any other maritime zone.

Accepting that states cannot terminate treaties establishing territorial sea and continental shelf boundaries pursuant to Article 62 VCLT, while could do so in respect of treaties establishing EEZ and all-purpose boundaries, would complicate the relevant states’ exercise or rights and duties in the maritime zones concerned. Treaty-made territorial sea and continental shelf boundaries would become immutable, allowing states only to update EEZ or all-purpose boundaries. There are two problems with this situation. First, the EEZ conceivably incorporates the continental shelf upon being declared by a coastal state. ^{Footnote 93} States willing to update continental shelf boundaries thus could find a legal justification for invoking the termination of earlier treaties delimiting such boundaries simply by declaring EEZs. Second, there are numerous treaties still in force specifying that they delimit boundaries in the continental shelf, with or without also delimiting boundaries in the water column. ^{Footnote 94} On Dr. Árnadóttir’s line of argument, one should understand such treaties to establish continental shelf boundaries distinct from water column boundaries. The states parties to these treaties could only delimit, or update, water column boundaries on the basis of the future coastal geography. This exercise in delimiting or updating could result in water column boundaries which follow a course different from continental shelf boundaries. The fact that such boundaries have been delimited only on three occasions to date ^{Footnote 95} indicates the states’ aversion to such arrangements, presumably owing to their difficult workability. ^{Footnote 96}

The additional issue is that fundamental change of circumstances is not a ground for updating boundaries established by judicial decision. The only legal basis to achieve that result is to file an application for revision, in accordance with Article 61 of the ICJ’s Statute or Articles 127–128 of ITLOS’ Rules of Procedure. An application for revision can only be filed before the lapse of ten years from the date of the relevant judgment. Ten years appear unlikely to be a sufficiently long time period for coastlines to change in such a manner that would justify updating maritime boundaries. Future changes in the coastline also could not justify applications for revision anyway, because a ‘fact’ on the basis of which revision proceedings can start is one that already existed before the date of the judgment to be revised. ^{Footnote 97}

Assuming that states are willing to update a previously-established maritime boundary because of changed coastlines, the same issues may arise that already arise in relation to the delimitation of present-day boundaries. Such issues may include agreeing on the delimitation method, selecting base points and assessing resource-related factors. The suggestion that ‘[s]tates are usually in good positions to monitor fluctuations of their coastlines and they can adjust boundaries accordingly’ ^{Footnote 98} assumes that these issues would present no obstacles to future delimitations. There are numerous examples of maritime boundary negotiations which are protracted for decades because of the states’ inability to agree on such issues. For example, the negotiations between Bangladesh and Myanmar continued from 1974 to 2010, without resulting in a treaty. ^{Footnote 99} There is no telling that similar cases would not take place also in future boundary negotiations, even if such negotiations were to build upon maritime boundaries already delimited by judicial process or treaty decades earlier.

Base point selection could be the most contentious problem in updating existing boundaries. Whether coastlines receded or advanced, base points used by international tribunals or states to delimit maritime boundaries would disappear as a result of erosion or accretion. In the accretion scenario, base points that used to be situated along the coast would be found inland; for this reason, they could not be seen as suitable base points anymore. Erosion is the more likely phenomenon, given the scientific predictions concerning sea level rise. Receding coastlines would mean that base points that used to be located on solid land would become submerged and thus rendered entirely unsuitable as base points for maritime delimitation. This likely future situation would cause limited problems for the unilateral establishment by states of the outer limits of their maritime zones, but states may not delimit maritime boundaries unilaterally. Selecting base points along the changed coastline is a matter for the relevant states’ subjective appreciation and on which they must reach an agreement, ^{Footnote 100} but judicial delimitation cases show that states can hardly ever agree on base point selection. International tribunals may have selected base points at an earlier point in time, but the very need for updating boundaries shows that such base points are now unsuitable to construct an equidistance line. Therefore, the international tribunals’ selection of base points is no guarantee that states will have an easier task in selecting their own updated base points years on.