Press "Enter" to skip to content

Romania and Ukraine Maritime Boundary Case

I wanted to follow up on Gentian’s post on recent activity at the ICJ with a more detailed consideration of the judgment in the Case Concerning Maritime Delimitation in the Black Sea delivered by the Court at the beginning of February.

Although both Romania and Ukraine are a party to the 1982 United Nations Convention on the Law of the Sea, Ukraine has opted out of compulsory dispute settlement relating to sea boundary delimitation under Article 298(1)(a)(i) of the Convention. Therefore, it was not possible to settle the dispute according to the provisions of Part XV of the Convention. Instead, the jurisdiction of the Court was based upon a 1997 agreement between the two states. This agreement allowed questions concerning the delimitation of the continental shelf and the exclusive economic zone between the two states to be submitted to the ICJ if Romania and Ukraine were unable to reach an agreement within a reasonable period of time. Following six years of unsuccessful negotiations, Romania initiated proceedings in September 2004.

Although the 1982 Convention did not offer a basis for the jurisdiction of the Court, it did provide the relevant principles on which the Court had to decide the case. The relevant provisions of the 1982 Convention (Articles 74 and 83) are notoriously ambiguous, providing that ‘the delimitation of [the continental shelf and the exclusive economic zone] between States with opposite or adjacent coasts shall be effected by agreement on the basis of international law, as referred to in Article 38 of the Statute of the International Court of Justice, in order to achieve an equitable solution.’ The Court has interpreted this aspect of the Convention on many occasions in the past and it was content in this case to follow the methodology on maritime boundary delimitation which it has developed in its previous jurisprudence. Thus, it started by drawing a provisional equidistance line and it then considered whether there were any relevant circumstances which would justify adjusting the equidistance line so as to achieve an equitable solution. On the facts, the Court decided that there were no relevant circumstances which would justify any such adjustment and the final boundary was therefore based upon the equidistance line.

As with most maritime boundary cases, it is some of the details of the decision which prove to be most interesting. In particular, the Court’s discussion of the following points is of particular note:

(1) Selection of base points

(2) Interpretation of ‘permanent harbour works’

(3) Relevance of islands to base points and delimitation

(4) Selection of relevant coastlines

These points are considered in more detail below…

(1) Selection of base points

Before the Court can draw a provisional equidistance line, it must first select the base points from which such a line can be calculated. Given that the base points can fundamentally influence the location of the line, this stage is of utmost importance. The Court describes it as a process of identifying ‘the appropriate points on the Parties’ relevant coast or coasts which mark a significant change in the direction of the coast, in such a way that the geometrical figure formed by the line connecting all these points reflects the general direction of the coast’ (para. 127). The Court further stated that ‘the issue of determining the baseline for the purpose of measuring the breadth of the continental shelf and the exclusive economic zone and the issue of identifying base points for drawing an equidistance/median line for the purpose of delimiting the continental shelf and the exclusive economic zone between adjacent/opposite states are two different issues’ (para. 137). The extent of both the continental shelf and the exclusive economic zone are measured from the same baselines that are used for measuring the territorial sea. The significance of the Court’s statement is that when identifying base points for the purposes of delimitation, the Court is not obliged to follow the base points used by the coastal states in measuring its territorial sea. Rather, the Court may select the base points independently by reference to the physical geography of the relevant coasts. Moreover, geography includes not only natural features, but also other material factors that are present (see para. 131 and below).

(2) Interpretation of ‘permanent harbour works’

The issue of permanent harbour works arose in this case because Romania had relied on the Sulina dyke, a man-made structure intended to protect ships navigating to and from the Danube River, as a relevant base point for the purposes of drawing the equidistance line. Indeed Romania had already used the dyke in drawing its baselines for delimiting the territorial sea in accordance with Article 11 of the 1982 Convention. Article 11 provides, ‘for the purpose of delimiting the territorial sea, the outermost permanent harbour works which form an integral part of the harbour system are regarded as forming part of the coast…’ In considering whether or not the Sulina dyke fell within this provision, the Court looked at the ordinary meaning of the word and also its drafting history, including the travaux preparatoires of the 1958 Convention on the Territorial Sea from which the provision originates. According to the expert appointed by the 1958 Conference, structures such as dykes and jetties could fall within the meaning of Article 11 provided they are not ‘excessive in length.’ The Court concluded that ‘there are grounds for proceeding on a case-by-case basis and that the text of Article 11 of [the 1982 Convention] and the travaux preparatoires do not preclude the possibility of interpreting restrictively the concept of harbour works so as to avoid or mitigate the problem of excessive length identified by the ILC’ (para. 134). Ultimately, the Court did not have to provide a definitive answer to this question because it held that the base points for maritime boundary delimitation between two states could differ from those base points used by states to unilaterally delimit their maritime zones (see above). Thus, the previous use of the Sulina dyke by Romania as a base point when delimiting its territorial sea did not prevent the Court from refusing to use it as a base point in the context of delimitation between two states. Nevertheless, the ruminations of the Court on this issue may have repercussions in future cases which raise issues under Article 11 of the 1982 Convention.

(3) Relevance of islands to base points and delimitation

Serpent’s Island is subject to the sovereignty of Ukraine and it is located about twenty miles from the coast of the mainland, in the vicinity of the mouth of the River Danube. The status of the island arose in two contexts in this case.

First, there was a question whether the island could be taken into account when choosing base points from which to draw the provisional equidistance line. The Court noted that previous arbitrations had placed base points lying on the low water line of certain fringe islands considered to constitute part of the coastline. However, in this case, the Court held that Serpents’ Island, ‘lying alone and some 20 nautical miles from the mainland, is not one of a cluster of fringe islands constituted “the coast” of Ukraine’ (para. 149).

Second, there was a question whether the island was a relevant circumstance which required adjustment of the provisional equidistance line drawn by the Court. In this regard, Romania had argued that Serpents’ Island was a rock incapable of sustaining human habitation or economic life of its own within the meaning of Article 121(3) of the 1982 Convention and therefore it was irrelevant for the purposes of delimiting the continental shelf or exclusive economic zone. Ukraine, on the other hand, argued that Serpents’ Island was an island within the meaning of Article 121(2) of the 1982 Convention. Again, the Court avoided providing any definitive answer to this question as it held that any maritime entitlements potentially generated by Serpents’ Island would in any case be fully subsumed by other maritime entitlements of Ukraine. Nevertheless, the arguments advanced by the parties may provide some interesting ideas for states faced with similar issues in future litigation.

(4) Selection of relevant coastlines

One of the factors that the Court will take into account in delimiting continental shelf and/or exclusive economic zone between two states is whether or not any disproportion exists in the ratio of the coastal length of each state and the maritime areas falling on either side of the delimitation line. Before it can do so, it must first identify the relevant coasts. Whilst there was no dispute in this case that the whole Romanian coast was relevant to the delimitation, the two states did disagree over the extent of the Ukrainian coast which could be taken into account. This was due to the concave nature of the coast and the existence of several deep indentations. The Court made clear that ‘the coast, in order to be considered as relevant for the purpose of the delimitation, must generate projections which overlap with the projections from the coast of the other party’ (para. 99). The Court was unwilling to accept that the coasts of the Karkinits’ka Gulf, Yahorlyts’ka Gulf and Dnieper Firth were relevant in this context. The Court therefore calculated the relevant length of the Ukrainian coast as 705 kilometres, rather than the 1,058 kilometres which had been suggested by Ukraine. It followed that the ratio of the relevant coasts of Romania and Ukraine was 1:2.8.

It is understandable that coastlines which are inward facing, such as the coast of the Karkinits’ka Gulf can be discounted for the purposes of delimitation. However, it is more controversial that the Court also chose to ignore the straight line which it had drawn over the opening of the gulf when calculating the length of the Ukrainian coastline (see para. 100). The line over the Karkinits’ka Gulf stretches over a not insubstantial distance and it is completely discounted by the Court. The Court reasoned that as the line replaces the coast of the Karkinits’ka Gulf which does not generate a maritime entitlement, the line also cannot do so. However, the line over the opening of the gulf, unlike the coast of the gulf, does project onto the area to be delimited and therefore it could be suggested that it should count for the purposes of calculating the relevant coastline.

It is not clear whether taking into account this line would have actually made a difference to the final decision, however. The Court made very clear that it was not seeking to determine whether the areas apportioned to each state were proportionate but rather the purpose of the test was to ensure that there was no significant disproportion. It refused to make any general comments on what counts as significant disproportion, saying that ‘this remains in each case a matter for the Court’s appreciation, which it will exercise by reference to the overall geography of the area’ (para. 213). In this case, the Court did not find that there were any particularly marked disparities which required it to adjust the provisional equidistance line.

Be First to Comment

Leave a Reply

%d bloggers like this: