SILBERMAN, Senior Circuit Judge,
concurring in the judgment:I agree with the majority’s conclusion that, notwithstanding our limited scope of review of a district court’s approval of a settlement agreement, we are obliged to reject this one. I write separately to discuss issues appellants raise which I think should be disposed of — and should be rejected so as not to complicate any further possible litigation — and to disagree with my colleagues on one important point.
Appellants argued that the Agreement violated the Flood Control Act (“FCA”), as well as the Water Supply Act (“WSA”). I think that alternative claim is quite weak. The relevant provision of the FCA states:
*1326Sale of surplus waters for domestic and industrial uses; disposition of moneys — The Secretary of the Army is authorized to make contracts with States, municipalities, private concerns, or individuals, at such prices and on such terms as he may deem reasonable, for domestic and industrial uses for surplus water that may be available at any reservoir under the control of the Department of the Army: Provided, that no contracts for such water shall adversely affect then existing lawful uses of such water....
33 U.S.C. § 708. By its plain terms, this provision sets the conditions under which the Secretary may sell “surplus water.” However, the Corps does not contend that the Settlement Agreement disposes of “surplus” water. The Agreement does reallocate a certain amount of reservoir capacity to water storage, but reallocations are governed by the Water Supply Act, not the Flood Control Act. Section 301(d) of the WSA requires Congressional approval of “[mjodifications of a reservoir project ... which would involve major structural or operational changes.... ” 43 U.S.C. § 390b(d). It is abundantly clear, then, that the Water Supply Act, not the Flood Control Act, is the statute that governs the Corps’ actions in this case, and I would accordingly explicitly reject the appellants’ FCA claims.
Turning to the WSA, appellants argued — indeed, it was their main argument — that the Agreement was unlawful under that statute, not just because it constituted a “major operational change,” but also because it was inconsistent with the project’s authorized purposes. 43 U.S.C. § 390b(d). The Buford Dam was constructed to improve navigation, generate hydroelectric power, and control flooding. Alabama v. U.S. Army Corps of Engineers, 424 F.3d 1117, 1122 (11th Cir.2005). (For many years, the Corps has maintained that an incidental benefit of the project was to provide metropolitan Atlanta with water supply.) Id. One of the project’s primary purposes, thus, was to provide hydroelectric power to downstream users. The Agreement, it is contended by Alabama and Florida, will reduce the amount of water released from the reservoir which will, in turn, reduce the water available for Alabama’s and Florida’s power requirements. Appellees responded that the Agreement’s compensation mechanisms met the hydroelectric purposes of the project.
Under those mechanisms, the water supply providers will pay substantially higher rates for water storage, and the resulting revenue will be credited to hydropower customers to compensate them for the reduced water flows through the dam. The Corps, the power customers, and the water supply providers all agree that this compensation mechanism will ensure that the Agreement does not have an adverse effect on hydropower generation.
I would not reach the merits of this argument because I do not think Florida and Alabama have standing to raise it. The two states have not identified any cognizable injury attributable to this claim. They do not assert that they or their citizens will pay any more for electricity as a result of the Agreement. Indeed, the hydroelectric companies supplying Florida and Alabama customers — the members of the Southeastern Federal Power Customers — support the Agreement because the compensation mechanism does adequately offset the reduction in water supply. To be sure, Florida and Alabama do have standing — as the panel concludes — to object to the alleged “major operational change” because the decreased water supply will have environmental impacts on Florida and Alabama. However, standing must be established for each claim, The *1327Wilderness Society v. Norton, 434 F.3d 584, 591 (D.C.Cir.2006), and appellants lack standing to assert that the Agreement will “seriously affect” the project purposes of the reservoir.
My fundamental disagreement with my colleagues’ determination that the Agreement works a “major operational change” is with their conclusion that the appropriate baseline for measuring the impact of the Agreement’s reallocation of water storage is zero. That seems to imply that the project was never intended to provide water to the city of Atlanta, which is in tension with the 11th Circuit’s observation mentioned infra, and is an issue which the settling parties agreed was not determined by the Agreement; it is an open question that has not really been briefed.
Beginning in the 1970s, the Corps allocated a steadily increasing volume of storage space to the water supply providers. Alabama v. U.S.A.C.E., 424 F.3d at 1122. It does not appear that Alabama and Florida challenged this policy until 1990, when the Corps was seeking Congressional approval to enter into permanent water supply contracts. Id. at 1122-23. Thus, for over a decade, the appellants acquiesced to a policy of increasingly large withdrawals. Even after Florida and Alabama initiated litigation in 1990, the states entered into two agreements that allowed the Corps to increase water withdrawals “to satisfy reasonable increases in [ ] demand” while settlement negotiations were pending.1
By asserting that the baseline is zero, the majority implicitly suggests that for many years some amount of water stored for (and supplied to) the city of Atlanta was illegal. That is a draconian conclusion I do not think warranted by the record.
I nevertheless agree with the majority’s determination that the Settlement Agreement is unlawful. To be sure, the definition of major operational change is by no means clear. Typically we would defer to an agency’s interpretation of that ambiguous term, but we cannot do so here because we are not reviewing an agency rulemaking or adjudication, but only a settlement agreement (which does not even purport to interpret the crucial language). See United States v. Mead Corp., 533 U.S. 218, 230, 121 S.Ct. 2164, 150 L.Ed.2d 292 (2001). We have given deference to agency interpretation of settlement agreements when Congress has granted the agency “an active role in approving the agreement.” Nat’l Fuel Gas Supply Corp. v. FERC, 811 F.2d 1563, 1571 (D.C.Cir.1987). But we have also emphasized that such deference is inappropriate where — as here — “the agency itself [was] an interested party to the agreement.” Id. In such cases, “deference might lead a court to endorse self-serving views that an agency might offer in a post hoc reinterpretation of its contract.” Id. The government seems to have implicitly interpreted the term “major” in its brief — as not including incremental changes — but we do not defer to mere litigating positions. Bowen v. Georgetown Univ. Hosp., 488 U.S. 204, 213, 109 S.Ct. 468, 102 L.Ed.2d 493 (1988).
The Agreement appears to me to constitute a “major operational change” because *1328it substantially increases the amount of reservoir space allocated to water supply compared to the allocation in 2002, which is all we have to conclude. The total storage capacity of Lake Lanier is 1,049,400 acre-feet. In a 2002 memorandum regarding Georgia’s request for more water storage, the General Counsel of the Department of the Army stated that, “[cjurrently, municipal and industrial interests, through direct withdrawals and releases from the reservoir, utilize the equivalent of 145,460 acre-feet of storage in Lake Lanier for water supply.” Thus, in 2002, approximately 13.9% of the reservoir’s capacity was being used for water supply. Under the Settlement Agreement, up to 240,858 acre-feet of the reservoir would be set aside for water storage (175,000 acre-feet for Gwinnett County, 20,675 acre-feet for the City of Gainesville, and 45,183 acre-feet for the Atlanta Regional Commission). This represents an increase of 95,398 acre-feet, which is a 65.6% increase over the 2002 level. Put another way, under the Agreement, approximately 9% more of Lake Lanier’s total capacity will be set aside for water storage — in 2002, 13.9% of the total capacity was allocated to water supply, but under the Agreement that figure increased to 22.9%. Like the majority, I also find it noteworthy that the storage levels permitted by the Agreement “would be the largest acre-foot reallocation ever undertaken by the Corps without prior Congressional approval.” Maj. Op. at 1324.
At oral argument, counsel for the Corps acknowledged that the Settlement Agreement would increase the amount of reservoir space allocated to storage by approximately 100,000 acre-feet (or 10% of total reservoir capacity), compared to the status quo prior to the Agreement. Tr. of Oral Arg. at 43:20. Counsel then conceded that a permanent reallocation of 10% of the reservoir’s capacity would constitute a “major operational change.” Id. at 49:08. In a letter dated December 13, 2007, the Corps attempted to retract this concession, noting that it was “in error.” But the logic of this concession was ineluctable. The Corps argued, however, that even if a permanent reallocation of 10% of the reservoir would be deemed “major,” the Settlement Agreement does not require Congressional approval because it is only an interim measure. That is not persuasive. The requirements of the Water Supply Act apply to “major structural or operational changes” — the text of that statute draws no distinction between interim and permanent changes.
The Corps argues that the burden was on Florida and Alabama to show that the Settlement Agreement was unlawful, and that the plaintiffs-appellants failed to offer sufficient evidence to meet this burden. But as explained above, the record — including the Corps’ own documents — shows that the Agreement would allocate an additional 95,398 acre-feet of reservoir capacity to water storage, and would increase the share of the reservoir allocated to water storage from 13.9% to 22.9%. I simply do not see how we can conclude that is not a major change.
. These agreements do contain disclaimers that they "shall not be construed as granting any permanent, vested or perpetual rights to the amounts of water used” during settlement negotiations. (It would appear that the word "used” in the agreements only refers to the water withdrawn during the settlement negotiations, and not to reservoir space that had been allocated to water storage prior to those agreements.) Moreover, the 1992 agreement states that it shall not be construed as "changing the status quo as to the Army’s authorization of water withdrawals.” This implies that — at the very least — Florida and Alabama did not contest the amount of storage that had been authorized by the Corps prior to 1992.