In Re Tri-State Water Rights Litigation

639 F. Supp. 2d 1308, 39 Envtl. L. Rep. (Envtl. Law Inst.) 20160, 2009 U.S. Dist. LEXIS 69895, 2009 WL 2371506

• Court: District Court, M.D. Florida
• Decided: July 17, 2009
• Docket: 2:07-cr-00001
• Status: Published

Opinion

MEMORANDUM AND ORDER

PAUL A. MAGNUSON, District Judge.

In the Rivers and Harbors Acts of 1945 and 1946 (“1945 RHA” and “1946 RHA”), Pub.L. No. 79-14, 59 Stat. 10, 10-11 (1945 RHA); Pub.L. No. 79-595, 60 Stat. 634, 640 (1946 RHA), Congress authorized the United States Army Corps of Engineers (the “Corps”) to begin construction of a dam and reservoir on the Chattahoochee River north of Atlanta, Georgia. Construction on the project finished in approximately 1960. The dam was christened the Buford Dam; the reservoir was named Lake Sidney Lanier.

At issue in this Multi-District Litigation (“MDL”) is the Corps’s operation of Buford Dam and Lake Lanier. The parties to the various member cases are the states of Alabama, Florida, and Georgia; the Southeastern Federal Power Customers (“SeFPC”); the cities of Apalachicola, Florida, and Atlanta, Columbus, and Gainesville, Georgia; the Georgia counties of Gwinnett, DeKalb, and Fulton; the Atlanta Regional Commission (“ARC”); the Cobb County-Marietta Water Authority; the Lake Lanier Association; 1 the Alabama Power Company (“APC”); the Columbus Water Works (“CWW”); the Middle Chattahoochee River Users; and the Corps and several Corps officers. 2

After the cases were consolidated by the Judicial Panel on Multidistrict Litigation, the parties agreed that the Court should consider the claims in two phases. Because some of the claims were similar or identical to claims pending before the United States Court of Appeals for the District of Columbia Circuit, the Court scheduled the proceedings on those claims second, awaiting that court’s resolution of the claims. Thus, the first scheduling orders in the MDL case contemplated that the Court would first entertain environmental claims, such as claims that the Corps’s operations in the Apalachicola-Chattahoochee-Flint (“ACF”) river basin violate the Endangered Species Act (“ESA”), 16 U.S.C. § 1531 et seq., and other environmental laws and regulations. Left for phase two were the overarching claims of the Corps’s authority (or lack thereof) for its operations in the basin in general, such as claims that the Corps is violating the Water Supply Act and the Flood Control Act.

The D.C. Circuit ruled on claims similar to the so-called “overarching” claims in 2008. Thereafter, the “overarching” claims became ripe for this Court’s resolution, and the Court therefore ordered that the phases be “flipped” so that the parties would present the statutory authorization and related issues first. (Aug. 11, 2008, Order.) The issues for resolution in the new Phase One include: (1) whether the Corps’s operations in the ACF basin, including the execution of water-supply contracts and installation of water intake structures in Lake Lanier, the alleged preference of water supply over other purposes, and the denial of Georgia’s water-supply request violate the Administrative Procedures Act (“APA”), 5 U.S.C. § 701 et seq., the National Environmental Policy Act (“NEPA”), 42 U.S.C. § 4321 et seq.; the Flood Control Act (“FCA”), 33 U.S.C. § 708 et seq.; the Water Supply Act (“WSA”), 43 U.S.C. § 390 et seq.; the Coastal Zone Management Act (“CZMA”), 16 U.S.C. § 1451 et seq.; and other congressional enactments; and (2) whether the water control plans and manuals, reservoir regulation manuals, action zones, recreation impact levels, and the Upper Chattahoochee Management Plan/River Management System violate federal law.

The fundamental question in the case is whether, by taking or failing to take the actions complained of in the various lawsuits, the Corps violated § 301 of the WSA, which provides:

Modifications of a reservoir project heretofore authorized, surveyed, planned, or constructed to include storage [for water supply] which would seriously affect the purposes for which the project was authorized, surveyed, planned, or constructed, or which would involve major structural or operational changes shall be made only upon the approval of Congress....

43 U.S.C. § 390b(d). In general, Florida, Alabama, APC, and the SeFPC contend that the Corps was obligated to seek Congressional approval for actions the Corps has taken with respect to water supply in Lake Lanier, because those actions allegedly affect the purposes for which the Buford Dam project was authorized or constitute major structural or operational changes. The Georgia parties and the Corps argue that Congressional approval is not required because the project’s purposes include water supply and because, in any event, the Corps’s operations have not amounted to a major structural or operational change in the project. To resolve these differences, the Court must examine the history of the Buford Dam and Lake Lanier.

BACKGROUND

A. Legislative History

1. Authorization

Although the 1945 and 1946 RHAs officially authorized the construction of Buford Dam, the Corps had been examining the feasibility of such a project for many years prior to 1945. Indeed, as early as 1925, 3 Congress asked the Corps to work with the Federal Power Commission (the predecessor to the Federal Energy Regulatory Commission) to examine the development of hydroelectric facilities on waterways nationwide, including in the ACF basin. River & Harbor Act of 1925, Pub.L. No. 68-585, ch. 467, 43 Stat. 1186, 1186, 1194 (March 3, 1925). In 1938, in response to a House resolution regarding the ACF basin, a Corps district engineer, Colonel R. Park, prepared a report to Congress outlining in great detail the geography and history of the basin and making recommendations for potential improvements in the basin. See H.R. Doc. No. 76-342, at 9-87 (1939) [hereinafter “Park Report”] (ACF000126-65). 4 It was in the Park Report that the project as eventually completed began to take shape.

The Park Report discussed a multitude of options for the development of rivers in the ACF basin and detailed eleven sites that could support a dam project to benefit hydroelectric power plants and navigation on the rivers. One of the eleven sites was the “Roswell” site “located on the Chattahoochee River 16 miles north of Atlanta, Ga., and about 2.5 miles upstream from the highway bridge at Roswell.” Park Report ¶ 196, at 66 (ACF000155). The Roswell site is approximately where Buford Dam was eventually located.

The Park Report detailed both the costs and benefits of each of the eleven sites. Colonel Park considered the following “direct benefits” for all of the proposed sites:

(a) Savings to the public in transportation charges.

(b) Value of hydroelectric power developed.

(c) Value as a facility for national defense.

(d) Increased commercial value of riparian lands.

(e) Recreational value.

(d) Value as a source of industrial and municipal water supply.

Id. ¶ 243, at 77 (ACF000160). The Park Report assigned an approximate dollar value to each “direct benefit.” For example, in Colonel Park’s estimation, the value of hydroelectric power if all eleven projects had been built would have been worth $6.5 million annually. Id.. ¶ 247, at 78 (ACF000161). Similarly, Colonel Park assigned a value of $25,000 to national defense, and $50,000 as a two-reservoir system’s recreational value. Id. ¶¶ 250-51, 259, at 79, 80 (ACF000161-62). For the proposed projects’ value as a water-supply source, however, Colonel Park assigned no monetary value, noting that “[t]here is apparently no immediate necessity for increased water supply in this area though the prospect of a future demand is not improbable.” Id. ¶ 260, at 80 (ACF000162). Water supply was the only potential benefit assigned no monetary value in the Park Report. Id. ¶ 261, at 81 (ACF000162).

After the Park Report was submitted to Congress, the Corps continued to evaluate the ACF basin for potential improvements. A so-called “interim” plan was submitted to the Chief of Engineers in December 1942, but was never submitted to Congress. See Sherk, Buford Dam, at 45 & n. 190 (noting that the interim report itself is not available, likely because it was withdrawn before being submitted to Congress). The interim report recommended two potential dam sites, including the Lanier site, “ ‘principally in the interest of hydropower.’ ” Id. at 45 (quoting Memorandum from P.A. Feringa, Colonel, Corps of Eng’rs, to Chief of Eng’rs (Oct. 28, 1943)). The Chief of Engineers sent the report back to the district engineer, asking him to revise the report to include an analysis of the benefits to navigation and flood control. Id.

The 1945 RHA stated specifically that the ACF project was authorized “in accordance with the plans” in the Park Report. 1945 RHA, ch. 19, 59 Stat. at 12, 17. Because the Park Report had not established where in the ACF basin the dam or dams would be built, the Corps continued to study the matter. The first result of this study was the report of Brigadier General James B. Newman, Jr., submitted to Congress in 1947. H.R. Doc. No. 80-300, at 10-40 (1947) [hereinafter “Newman Report”] (ACF000644-74).

General Newman noted that “[t]he principal value of the Chattahoochee River is as a source of power.” Id. ¶ 7, at 13 (ACF000647). He described the Park Report as evaluating the rivers in the ACF basin “in the combined interest of navigation and power.” Id. ¶ 47, at 22 (ACF000656). The majority of the Newman Report consists of detailed evaluations of the hydropower and navigation benefits of the alternatives discussed in the Park Report. General Newman concluded that the locks and dams proposed by the Park Report for the southern portion of the Chattahoochee, below Columbus, Georgia, would not be economically efficient unless a “considerable flow regulation were provided by a large storage-power reservoir upstream.” Id. ¶ 67, at 27 (ACF000661). That reservoir would become Lake Lanier.

General Newman also noted other “incidental” benefits of a reservoir at the Lanier site. 5 He discussed Atlanta’s urging that a reservoir north of Atlanta be constructed before other elements of the ACF basin project, “in order to meet a threatened shortage of water, during low-flow periods, for municipal and industrial purposes.” Id. ¶ 68, at 27 (ACF000661). Specifically, “[i]f the regulating storage reservoir ... could be located above Atlanta, it would greatly increase the minimum flow in the river at Atlanta, thereby producing considerable incidental benefits by reinforcing and safeguarding the water supply of the metropolitan area.” Id. General Newman therefore concluded that the Lanier site should be developed as outlined in the Park Report and in his own report. He determined that the construction of a dam at the Lanier site, along with the proposed developments at Junction and Upper Columbia, would “create an effective and economical system for the production of power, in addition to providing ... for navigation.... The system would also contribute to the reduction of floods and flood damages in the Chattahoochee River valley, and would ensure an adequate water supply for the rapidly growing Atlanta metropolitan area.” Id. ¶ 73, at 28-29 (ACF000662-63). The Newman Report recognized that releases from the proposed dam for downstream water supply might have to be increased as the Atlanta area developed, although the Newman Report emphasized that such an increase “would not materially reduce the power returns from the plant.” Id. ¶ 80, at 34 (ACF000688).

As with the Park Report, the Newman Report estimated the dollar value of the various annual benefits from the construction of a dam and reservoir at the Lanier site. The Newman Report, however, listed only three valuable benefits: power, navigation, and flood control. Id. ¶ 98, at 38 (ACF000672). The Newman Report also allocated the estimated costs of building the Buford project, a total of more than $17 million. Of this, $16 million was allocated to power, none to navigation, and the remainder to flood control. Id. ¶ 97, at 38 tbl. 9 (ACF000672). The Southeastern Power Administration (“SEPA”), from which the SeFPC purchases the power generated by the Buford Dam, ultimately paid approximately $30 million toward the total construction cost of $47 million for the dam. SeFPC Am. Compl. ¶ 32; see also U.S. Army Corps of Eng’rs, Survey Report on Apalachicola, Chattahoochee and Flint Rivers, Alabama, Florida and Georgia ¶49, at 15 (1973) (total cost of Buford Dam was $47,059,711) (ACF003968). The 1946 RHA adopted the Newman Report’s recommendation that the project be limited to three dams, including the Buford dam. U.S. Army Corps of Eng’rs, Definite Project Report on Buford Dam: Chattahoochee River, Georgia ¶ 7, at 4 (1949) [hereinafter “Definite Project Report ”] (ACF001449).

2. Planning

The initial authorization in the 1945 and 1946 RHAs did not end Congress’s involvement in the Buford Dam project. The project required money, and that money had to be appropriated by Congress each year. Thus, once the project entered initial planning stages and during the construction of the project, Congress held yearly hearings on the progress of the project and on the Corps’s use of funds. For fiscal year 1948, Congress considered the Corps’s request for funding for the planning of the project. Georgia Representative James C. Davis asked the Appropriations Committee to recognize the “critical necessity” of the project, which he described as a “multi-purpose dam ... for the purpose of generating power, flood control, and water supply for the city of Atlanta, as well as a regulated flow of water of the Chattahoochee River....” War Dep’t Civil Functions Appropriation Bill 1948: Hearing on H.R. 4002 Before the Subcomm. of the S. Comm, on Appropriations, 80th Cong. 697 (1947) (statement of Rep. James C. Davis, Georgia). In addition, Atlanta’s Mayor William B. Hartsfield testified that the undependable nature of the flow in the Chattahoochee had likely already caused severe economic losses in the Atlanta area. Id. at 700 (statement of William B. Hartsfield, May- or, Atlanta, Georgia). He asked that the Buford project be given priority over other dams proposed for the Chattahoochee. Id.

The following year, the House Subcommittee on Appropriations submitted a report about the funding the Corps had requested for that fiscal year, including funding for the Buford Dam project. See H.R.Rep. No. 80-1420, at 5-8 (1948). The report recommended reducing the Corps’s request for plans for the Buford project by $67,000. According to the report:

While the Buford Dam may be an important part of the comprehensive river system plan for the Apalachicola, Chattahoochee, and Flint Rivers its construction will provide a source of water for the city of Atlanta that witnesses from that part of the country indicate is greatly needed. The city of Atlanta is not, however, providing any contribution toward the construction of this dam and inasmuch as it stands to benefit to a great extent it appears that some substantial contribution should be made toward the ultimate cost of the dam, and in future planning it is suggested that this feature be given careful consideration and an opportunity be afforded the city of Atlanta to make a contribution comparable to the benefits to be received.

Id. at 8.

In January 1948, the House Subcommittee on Appropriations heard testimony from several members of Georgia’s Congressional delegation about the various projects in the ACF basin. Representative Stephen Pace led the delegation and described the ACF projects as having three purposes: navigation, power, and flood control. Civil Functions, Dep’t of the Army Appropriation Bill for 191,9: Hearing on H.R. 5524 Before the Sub-comm. of the H. Comm, on Appropriations, 80th Cong. 723 (1948) (statement of Rep. Stephen Pace, Georgia). Representative Pace also testified that the project had “two additional purposes”: to “serve as a reservoir for the entire system in the event of dry spells and floods,” which would “assure[ ] the navigability of the entire project”; and “to meet the very critical shortage of water in the city of Atlanta.” Id. He also emphasized the area’s “crying need for an abundance of hydroelectric power.” Id. at 724.

Many of the witnesses testified about the navigation and transportation benefits of the projects proposed for the ACF basin. Among them was J.W. Woodruff, for whom the ACF’s southernmost dam, the Jim Woodruff Dam, is named. He envisioned the navigation made possible by the projects in the ACF basin as an economic engine that would drive industrial and commercial development in the region, allowing goods to be shipped from the area around the world. Id. at 750-51 (statement of J.W. Woodruff, Atlanta, Georgia).

Other participants addressed their testimony specifically to the proposed Buford project. Representative John S. Wood of Georgia spoke about the need for flood control in an area that could receive more than eight inches of rain in a 24-hour period. Id. at 777 (statement of Rep. John S. Wood, Georgia). Both Mayor Harts-field and Representative Davis again testified about the multi-purpose nature of the project, pointing out its benefits for power, navigation, flow regulation, and pollution control, and as a source of water supply for Atlanta. Id. at 778 (statement of Rep. James C. Davis, Georgia), 782 (statement of William B. Hartsfield, Mayor, Atlanta, Georgia). In his statement to the Senate Subcommittee on Appropriations, however, Mayor Hartsfield de-emphasized Atlanta’s need for water supply from the Chattahoochee. He characterized Atlanta’s need for the water as “necessary” but stressed that Atlanta should not “be put in the category with such cities as are in arid places in the West or flat plain cities where there is one sole source of water .... “ Civil Functions, Dep’t of the Army Appropriation Bill 1919: Hearing Before the Subcomm. of the S. Comm, on Appropriations, 80th Cong. 644 (statement of William B. Harts-field, Mayor, Atlanta, Georgia). Mayor Hartsfield stated:

We need flood control; we need the increased power; we need badly some sort of water recreation in that section. We need, of course, the promotion of navigation....

We need the promotion of regular flow not only for Atlanta’s water supply, but to enable others, as I said, to use the river, industries to use it, which they are not now able to do.

Id. at 646.

As part of the record at the hearing, the Corps filed a report describing the Buford project:

The Buford Reservoir will provide flood protection to the valley below it; provide a large block of power to an area where there is a power shortage; provide an increased flow which is essential to provision of a 9-foot depth for navigation in the Apalachicola River; assure an adequate water supply for municipal and industrial purposes in the Atlanta metropolitan area; and provide recreational facilities for the area surrounding the reservoir.

Id. at 648 (report of P.A. Feringa, Colonel, Corps of Eng’rs).

At the end of February 1948, apparently in response to the questions raised about Atlanta’s willingness to pay for part of the Buford project, Representative Davis wrote to Mayor Hartsfield. In this letter, Representative Davis stated that the Subcommittee’s desire to have Atlanta fund some of the construction cost of the dam was not unprecedented, noting that the city of Dallas had recently contributed more than $2 million to a reservoir project in Texas. Letter from Rep. James C. Davis, Ga., to William B. Hartsfield, May or, Atlanta, Ga. (Feb. 27, 1948) (SUPPAR000420). Mayor Hartsfield responded negatively to the suggestion that Atlanta should bear some of the costs of the Buford Dam:

Frankly, in our zeal I think we have just laid too much emphasis on the Chattahoochee as a water supply----

In our case the benefit so far as water supply is only incidental and in case of a prolonged drought. The City of Atlanta has many sources of potential water supply in north Georgia. Certainly a city which is only one hundred miles below one of the greatest rainfall areas in the nation will never find itself in the position of a city like Los Angeles....

[I]n view of other possible sources of Atlanta’s future water we should not be asked to contribute to a dam which the Army Engineers have said is vitally necessary for navigation and flood control on the balance of the river.

Letter from William B. Hartsfield, Mayor, Atlanta, Ga., to Rep. James C. Davis, Ga. (Mar. 1, 1948) (SUPPAR001063). Atlanta did not contribute to the construction costs of the Buford Dam.

In preparation for the start of construction, the Corps prepared the Definite Project Report. (ACF001436). This report described the project’s “principle purposes” as: “to provide flood control; to generate hydroelectric power; to increase the flow for open-river navigation in the Apalachicola River below Jim Woodruff dam; and to assure a sufficient and increased water supply for Atlanta.” Id. ¶ 115, at 41 (ACF001486). The Definite Project Report addressed only one specific water-supply issue: the city of Gaines-ville’s water-pumping station, located on the Chattahoochee. Id. ¶ 95, at 29 (ACF001474). The report noted that the entire station would require relocation, as it would be inundated on completion of the dam. Id. ¶ 96, at 29 (ACF001474). On June 22, 1953, the Corps and Gainesville executed a contract whereby the Corps paid Gainesville $300,000 for the land to be taken by the reservoir and Gainesville was given the right to withdraw eight million gallons per day from the reservoir. Contract between U.S. Army Corps of Eng’rs & City of Gainesville, Ga. (June 22, 1953) (ACF014457-63). 6

In its discussion of reservoir regulation, the Definite Project Report noted that the power plant would operate as a “peaking plant to provide maximum possible power during the hours of greatest demand.” Definite Project Report ¶ 120, at 42 (ACF001487). At off-peak times the plant would operate only a smaller generator, “to provide flow to meet municipal and industrial requirements at Atlanta.” Id. When water levels in the reservoir fell below a certain level, however, “only prime power [would] be generated.” Id.

The Definite Project Report estimated the total cost of the Buford project at $35.6 million. See id. ¶ 104, at 32-37 tbl. 1 (ACF001477-82). The “primary benefits” of the project were “flood control and production of hydroelectric power.” Id. ¶ 123, at 44 (ACF001489). The report calculated the flood-control benefit as worth $163,000 annually. Id. Power benefits were valued at $1.7 million on site, with the potential for up to $3.2 million in power benefits if all downstream plants were modified as proposed. Id. ¶ 123, at 45 tbl. 3 (ACF001490). The report calculated the potential annual benefit to transportation at almost $1.4 million. Id. ¶ 124, at 45 (ACF001490). The benefit to recreation was calculated at $196,000 annually. Id. ¶ 125, at 46 (ACF001491). The Definite Project Report noted that a “real benefit will also result from assurance of sufficient water for municipal and industrial requirements at Atlanta” but it did not make any estimate of the value of that benefit. Id. ¶ 124, at 46 (ACF001490).

The question of Atlanta’s contribution to the costs of the Buford project surfaced again in the hearings on the 1952 Army Appropriation Bill, H.R. 4386. Corps officer Colonel Potter testified that “[t]he purpose of the project is flood control, water supply for the city of Atlanta, which is growing by leaps and bounds, and the production of power.” Civil Functions, Dep’t of the Army Appropriations for 1952: Hearings Before the Subcomm. of the H. Comm. on Appropriations, 82d Cong. 118 (1951) (statement of Col. Potter, Corps officer) (SUPPAR026654). A member of the Subcommittee asked Colonel Potter if Atlanta was “cooperating in this project in any way.” Id. at 120 (question of Rep. Davis) (SUPPAR026656). Colonel Potter responded:

No, sir; because this is not a problem of furnishing water directly or furnishing storage for that purpose; it is the regulation of the river that gives [Atlanta] a constant supply over the up-and-down supply now existing during the year.... With this dam letting out a constant supply of water every day their water-supply problem is reduced immensely....

Id. (statement of Col. Potter, Corps officer). Other committee members questioned Colonel Potter further on Atlanta’s need for, and contribution to, the project:

Mr. Ford: Where you have a project such as this particular project and water supply is part of the justification for a community, does not the community make any contribution to the project? Col. Potter: Yes, sir, normally, but not in this case....

This dam furnishes Atlanta with water due to the fact that it regulates the discharge of floods. When a flood comes, it comes down in a certain set period — say a week. We store that week’s terrific runoff of water and then let it out gradually.... Hence we discharge that flood, we will say, for 3 months.

Then, in the production of electricity, we can discharge somewhere in the neighborhood of 4,000 to 5,000 second-feet constantly. That [water] will always be flowing by Atlanta; so that now they won’t have the river partially dry or full of mud in the summer, but they will have a more or less constant flow of the river past their door and will always be able to pull water out of it.

It did not cost the Federal Government 1 cent to supply that service, because it was an adjunct to the power supply and flood control. Had we put in some storage purely for water supply, which they would tell us to release at certain intervals, we would then charge them for it, and they would have to pay for the difference of that construction cost.

Id. at 121-122 (exchange between Rep. Gerald Ford, Michigan, and Col. Potter, Corps officer) (SUPPAR026657-58).

In a prescient question, Representative Ford then asked, “Is it not conceivable in the future, though, when this particular project is completed, that the city of Atlanta will make demands on the Corps because of the needs of the community, when at the same time it will be for the best interests of the over-all picture ... to retain water in the reservoir?” Id. at 122 (SUPPAR026658). Colonel Potter’s response is illuminating: “The first thing we do is to decide, after a study, whether or not the water supply is more valuable to use for the production of electricity. If it is, then we would have to come back, I believe, to Congress to alter the authorization of that project, were it a major diversion of the water.” Id. He noted that the Corps “take[s] a very dim view of changing a project to the subsequent needs without Congress having a hand in it.” Id.

3. Construction

In 1952, at the beginning of construction of the dam, Georgia’s Representative Davis and the Corps’s General Chorpening appeared at a hearing of the House Subcommittee on Appropriations to ask Congress for $8.5 million for the Buford project. Representative Davis described the project as providing flood control, power, and navigation benefits. Civil Functions, Dep’t of the Army Appropriations for 1958: Hearings on H.R. 7268 Before the Subcomm. of the H. Comm, on Appropriations, 82d Cong. 1196-97 (1952) (statement of Rep. Davis, Georgia) (SUPPAR02667980). Neither Representative Davis nor General Chorpening mentioned any water-supply benefits from the project.

The next year, two Corps officers testified before the Senate Subcommittee on Appropriations in support of the Corps’s request for another $8.5 million in funding for the Buford project. Colonel Paules described the project as “a combination flood control — power project which will assist navigation downstream by the regulation of the river flows.” Civil Functions, Dep’t of the Army Appropriations, 1951: Hearings on H.R. 5876 Before the Sub-comm. of the S. Comm, on Appropriations, 83d Cong. 480 (1953) (statement of Col. Paules, Corps officer) (SUPPAR026685). Colonel Paules and General Chorpening also testified before the House Subcommittee on Appropriations regarding the requested funding for the Buford project. Colonel Paules discussed the anticipated completion dates for the project, including when the power plant was expected to be operational. He noted, “[t]he project has a total capacity of some 2 million acre-feet for flood control and power, and incidentally would supply additional water downstream for the benefit of the municipalities along the river....” Civil Functions, Dep’t of the Army Appropriations for 1951: Hearings on H.R. 5876 Before the Subcomm. of the H. Comm, on Appropriations, 83d Cong. 503 (1953) (statement of Col. Paules, Corps officer) (SUPPAR026688). Representative Davis asked whether Atlanta was contributing to the cost of the project. The Corps officers responded, “While the city of Atlanta is not contributing to this, they get benefits from it, incidentally, as the result of the controlled release of floodwaters, and as the water is released through the power-plant.” Id. General Chorpening explained:

[Tjhere would be no legal way to collect payment from the city of Atlanta, since, as was just stated, there is no additional cost being included for the construction of this project to provide the more uniform flow of water which will pass the city of Atlanta. In other words, the building of the project, with its power production and flood control and navigation benefits will not make available any more water than is now going past Atlanta. It is only going to make it flow by at a more uniform rate.

Id. (statement of Gen. Chorpening, Corps officer).

The Corps requested an additional $5.8 million for the Buford project in fiscal year 1955. Civil Functions, Dep’t of the Army Appropriations, 1955: Hearings on H.R. 8367 Before the Subcomm. of the S. Comm, on Appropriations, 83d Cong. 324 (1954) (statement of Col. Whipple, Corps officer) (SUPPAR026698). Colonel Whipple told the Subcommittee that “[t]he project provides a considerable amount of flood control, but its main purpose is the output of power to the area.” Id. He also testified that the project’s “additional benefits” would include “increasing] the flow of water downstream which improves the water supply at Atlanta, and the project is unusually well situated for recreational use.” Id. at 325 (SUPPAR026699). Again, the committee members asked about whether Atlanta would contribute toward the cost of the project. Colonel Whipple responded, “We understand not, sir.... There are no additional costs for [Atlanta’s water supply]. It is purely an incidental benefit on account of the power releases which does not require any storage to be devoted to that purpose.” Id.

In the next several years, Georgia’s Representative Davis appeared in similar hearings before the House and Senate Subcommittees on Appropriations. He testified consistently that the purposes of the Buford project were flood control, navigation, and hydroelectric power, mentioning water supply only occasionally. See, e.g., Public Works Appropriations for 1956: Hearings on H.R. 6766 Before the Subcomm. of the S. Comm, on Appropriations, 84th Cong. 307-09 (1955) (statement of Rep. James Davis, Georgia) (SUPPAR026713-15) (stating, that the Buford project is a “multi-purpose” project that will provide flood protection, “will augment the low water flow of the river” to support navigation and to assist in the generation of power downstream, will generate “810 million kilowatt-hours of electrical energy annually and the additional water supply for the growing metropolis of Atlanta”); Public Works Appropriations for 1957: Hearings Before the Subcomm. of the H. Comm, on Appropriations, 84th Cong. 355-57 (1956) (statement of Rep. James Davis, Georgia) (SUPPAR026720-22) (discussing flood control, navigation, and power benefits but not mentioning water supply benefits).

During construction of the dam, Gwinnett County asked the Corps for permission to withdraw water from the reservoir for water supply. F.G. Turner, Ass’t Chief, Eng’g Div., Report on Withdrawal of Domestic Water Supply from Buford Reservoir ¶1, at 1 (1955) (SUPPAR005459). The Corps responded:

that the primary authorized purposes of the Buford project were flood control, power and low-flow regulation for navigation and other purposes, and that diversion of flows from the reservoir would, in some degree, adversely affect one or more of these purposes. [The Gwinnett County representatives] were informed that additional legislation would be necessary....

Id. ¶ 2, at 1 (SUPPAR005459). The Corps noted that the project “will provide storage for flood control, hydro-electric power and increased flow for water supply at Atlanta, Georgia, and for navigation in the Apalachicola River.” Id. ¶ 5, at 2 (SUPPAR005460). The report examined the “Provision for water supply in the Atlanta Area,” discussing Atlanta’s concern that the Corps maintain minimum flows in the river to meet Atlanta’s water requirements. Id. ¶¶ 7-8, at 3 (SUPPAR005461). It also noted that Gwinnett County requested initial withdrawals of four million gallons per day from the reservoir and ultimate withdrawals of ten million gallons per day. Id. ¶ 9, at 3 (SUPPAR005461). The report predicted that “[t]he granting of permission to Gwinnett County to with draw water for domestic water supply as requested will no doubt establish a precedence [sic] for possible like requests from other communities within the area....” Id. The report commented that Gainesville, Georgia, had been granted permission to withdraw a maximum of eight million gallons per day from the reservoir. Id. f 9, at 4 (SUPPAR005462). As noted above, Gainesville had a water intake structure on the Chattahoochee River that was inundated by Lake Lanier, and thus had a preexisting right to withdraw water.

In 1956, Congress granted the Corps permission to contract with Gwinnett County for the use of up to 11,200 acre-feet of storage in Lake Lanier annually, for a period not to exceed fifty years. Act of 1956, Pub.L. No. 84-841, 70 Stat. 725 (1956) (amending 1946 RHA). There is no evidence in the record that the Corps and Gwinnett County ever entered into the contract contemplated by this statute. Although the Corps and Gwinnett County did execute a water-supply contract in the 1970s, neither the original contract nor any supplement or extension thereto invoked the authority of the 1956 statute but rather relied on the more general authority of the WSA. See, e.g., Supplemental Agreement No. 1 to Contract No. DACW01-973-624 Between the U.S. & Gwinnett County, Ga. for Withdrawal of Water from Lake Sidney Lanier, at 1 (Apr. 29, 1974) (ACF004022) (providing that, on expiration of the contract, Gwinnett County “shall have the right to acquire from the Government, under the provisions of the Water Supply Act of 1958, Public Law No. 85-500, the right to utilize storage space in the project ...”). Moreover, the Corps’s first agreements with Gwinnett County were, by their terms, “interim” contracts pending the completion of a study of the Atlanta area’s water-supply needs. Contract Between the U.S. & Gwinnett County, Ga. for Withdrawal of Water from Lake Sidney Lanier, at 2 (July 2, 1973) (ACF004025); see also infra section C.l. As the Georgia parties admit, all of the Corps’s contracts with Gwinnett County have expired. (Ga.’s Mot. for Summ. J. Factual App. ¶¶ 7.23, 7.33.)

In 1958, as the Buford Dam neared completion, the Corps promulgated the Apalachicola River Basin Reservoir Regulation Manual. U.S. Army Corps of Eng’rs, Apalachicola River Basin Reservoir Regulation Manual (1958) [hereinafter “1958 Manual”]. (ACF001640.) The 1958 Manual is a detailed description of the geography and hydrography of the ACF basin, including all federal projects undertaken in the basin. It describes Buford Dam as “a multiple-purpose project with major uses of flood control, flow regulation for navigation, and power.” 1958 Manual ¶ 85, at 27 (ACF001677). The 1958 Manual does not specifically describe the operation of the Buford project; rather, the regulation manual for the Buford project, which was completed in October 1959, is appended to the 1958 Manual as Appendix B. 7 Id. app. B (ACF001776); see also id. app. B ¶ 43, at B-21 (ACF001804) (listing October 1959 completion date). The Corps has never updated the 1958 Manual or the Buford Reservoir Regulation Manual (“Buford Manual”), and thus these manuals are the current regulation manuals for the ACF basin and Buford dam.

The Buford Manual lists the elevation for the top of the flood-control pool as 1085 feet above sea level. Id. app. B, at B-l (ACF001784). The elevation of the top of the power pool is 1070 feet, and the bottom of the power pool is 1035. Id. The reser voir’s flood-control storage (elevation 1085 to 1070) is 637,000 acre-feet. Id. Power storage is listed as 1,049,400 acre-feet. Id. The manual also noted that the reservoir reached full power pool on May 25, 1959, and that the President had signed a bill naming the reservoir Lake Sidney Lanier on March 29, 1956. Id. app. B ¶¶ 8-9, at B-5 (ACF001788).

The Buford Manual describes the project:

Buford is a multiple-purpose project with principal purposes of flood-control, navigation and power. It reduces flood stages in the Chattahoochee River as far downstream as West Point, Georgia, 150 miles below the dam; provides an increased flow for navigation in the Apalachicola River below Jim Woodruff Dam during low-flow seasons; and produces hydroelectric energy, operating as a peaking power plant. The increased flow in dry seasons also provides for an increased water supply for municipal and industrial uses in the metropolitan area of Atlanta, and permits increased production of hydroelectric energy at downstream plants.

Id. app. B ¶ 12, at B-6 (ACF001789). The Buford Manual also details the regulation of the project:

Normally, the Buford project will be operated as a peaking plant for the production of hydroelectric power with minimum releases during the daily and weekend off-peak period which will be sufficient, with local inflows added, to supply the Atlanta area with not less than 600 cfs. During low-water periods such regulation will provide increased flow downstream for navigation, water supply, pollution abatement, and other purposes.... [T]he primary purpose of the project is flood control, and a storage of 637,000 acre-feet between elevations 1,070 and 1,085 has been reserved exclusively for the detention storage of flood waters.

Id. app. B ¶29, at B-13 (ACF001796). The Corps contracted with the SEPA to provide 142,000 kilowatts of “dependable” power capacity from the project. Id. app. B ¶ 31, at B-13 (ACF001796). The Corps gave SEPA minimum declarations of energy the dam would produce each month. Id. app. B ¶ 31, sec. 2. 1, at B-15 (ACF001798). The Corps also noted its commitment to keep the flow at Atlanta at a minimum of 600 cfs. Id. app. B ¶ 33, at B-18 to 19 (ACF001801-02).

Enacted in 1958, the Water Supply Act (“WSA”), Pub.L. 85-500, tit. Ill, 72 Stat. 319, changed the way the Corps funded dam-building projects. Specifically, the WSA required the Corps to allocate the costs of each project to the benefits of the project so that, for example, if a project benefitted primarily hydroelectric power, the power interests would pay a proportionate share of the cost of that project. See WSA § 301(b), 72 Stat. at 319. The Buford project was well into construction by the time the WSA’s cost allocation requirements took effect, but the Corps endeavored to comply with those requirements by issuing cost allocation studies for the projects in the ACF basin in 1959. Mobile Dist., U.S. Army Corps of Eng’rs, Cost Allocation Studies, Apalachicola, Chattahoochee and Flint Rivers Projects, Basis of All Allocations of Costs for Buford and Jim Woodruff Projects Adopted by the Chief of Engineers 21 (1959) (ACF002103) (noting that according to an agreement between the Department of Interior, Department of the Army, and the Federal Power Commission, “costs of a multiple-purpose project shall be allocated among the purposes served in such a manner that each purpose will share equitably in the savings resulting from combining the purposes in a multiple-purpose development”).

This study report introduced the projects in the ACF basin: “The primary benefits provided by the ACF project are flood control, navigation and hydroelectric power. The incidental benefits are low-water regulation for water supply and pollution abatement at Atlanta, Georgia and public use with facilities for recreation. ...” Id. at 2 (ACF002086); see also id. at 5 (ACF002089) (stating that the project “will be operated for the primary purposes of flood control, power, and navigation”). Although the report analyzes in detail the cost of the three benefits of flood control, power, and navigation, it does not attribute any costs to the “incidental” benefits of water supply, pollution abatement, or recreation. Id. at 20 (ACF002102).

The report gives the total “first” cost of the Buford project as $43,601,500. Id. at 23 (ACF002105). The portion of this total allocated to navigation was $1,518,200; to flood control, $3,402,600; and to power, $38,680,100. Id. No portion of the project’s costs was allocated to water supply.

Appendix A of the report is the cost allocation study specifically for the Buford project. It states that “[t]he primary purposes of the Buford project are flood control and the generation of hydroelectric power. Incidental uses attributable to the operation of the project for power include flow regulation for navigation in the Apalachicola River and water supply and pollution abatement in the Atlanta area.” Id. app. A, at A-l (ACF002108). The report notes that full-scale power operation began at Buford in July 1958. Id. app. A, at A-2 (ACF002109).

Table Four of the Appendix shows the “average annual benefits” of the Buford project. The annual benefit to navigation is listed as $75,900, to flood control is $193,000, and total power benefits (including benefits at site and downstream) are $2,476,200. Id. app. A tbl. 4 (ACF002127). There are no benefits calculated for any other purpose.

4. Water Supply

The various Corps reports and Congressional testimony discussed above show the original role of the Buford project in supplying water to Atlanta. At the time Buford Dam was authorized, planned, and constructed, the Corps did not anticipate any water-supply withdrawals from the reservoir itself, with the exception of the water withdrawn by the cities of Gaines-ville and Buford. Nor did the Corps or any other entity set aside any portion of Lake Lanier’s storage for water supply. Rather, the water-supply benefit discussed throughout the legislative history was the regulation of the river’s flow. A more regular flow was seen as providing Atlanta both with a reliable flow in the Chattahoochee from which to withdraw water, and more certainty diluting the wastewater Atlanta discharged into the river. Throughout the 1940s and 1950s, when water supply is mentioned in connection with the Buford project, that water supply is in the form of Atlanta’s withdrawals from the river itself, far below the proposed dam.

In the decades after the Buford Dam was built, however, the Corps’s and the Georgia parties’ definition of water supply in the Buford project changed considerably. The origin of this change is difficult to pinpoint. However, at some point after the dam was completed, both the Corps and the municipal entities in the Atlanta area began to envision the water supply benefit as a storage-and-withdrawal benefit. In other words, water supply came to mean not flow regulation in the river but water withdrawals from the lake.

B. Operation of Buford Project

1. 1970s

Once construction on the Buford project was complete, the record reflects very lit- tie activity until the early 1970s. In 1974, in accordance with NEPA, the Corps prepared a final environmental impact statement (“EIS”) “for continued operation and maintenance of the existing Buford Dam and Lake Sidney Lanier.” Mobile Dist., U.S. Army Corps of Eng’rs, Final Environmental Impact Statement, Buford Dam and Lake Sidney Lanier, Georgia (Flood Control, Navigation and Power), Statement of Findings (1974) [hereinafter “Final EIS ”] (ACF004338). The preliminary statement in the EIS reported that the “Authorized project purposes provide peaking hydroelectric power, flood control, and low flow augmentation.” Id. The preliminary statement added that “Additional benefits derived from operation of the project are recreation and water supply.” Id. The summary states:

The project provides an average annual benefit of $638,400 in flood control. The hydroelectric facilities have a capacity of 86,000 kw and are operated to meet peak demands for electricity in the service area. Low-flow augmentation provides water for navigation, industrial and municipal uses downstream. The reservoir provides a source of water supply for public water users. Over 15 million visitors utilized the recreational facilities of the lake in 1972. The benefit-to-cost ratio is 3.6 to 1.

Id. at i (ACF004339). The EIS’s description of the project notes that the “principal purposes” of the project are flood control, navigation, and power. Id. at 1 (ACF004342). The description explains the project’s effect on the principal purposes, adding that the “increased flow in dry seasons also provides for an increased water supply for municipal and industrial uses in the metropolitan area of Atlanta, and permits increased production of hydroelectric energy at down-stream plants.” Id. The EIS recognizes that “recreation was not a primary purpose for which the project was authorized,” but that recreation had become a significant part of the use of the reservoir, with Lake Lanier the most used Corps lake in the United States. Id. at 12 (ACF004353).

The total storage of the reservoir is 2,554,000 acre-feet, with 637,000 acre-feet of flood-control storage and 1,049,400 acre-feet of power storage. Id. at 4 (ACF004344). The EIS does not list any storage for water supply, but does note that Gwinnett County, Gainesville, and Buford “obtain water directly from the reservoir.” Id. at 14 (ACF04355). In addition, “[t]he Atlanta metropolitan area increased its water use from the river 37% (from 117 mgd to 160 mgd) 8 between 1960 and 1968.” Id. The EIS also discusses the changes in population in the area around the lake, stating that “[t]he number of residences within 2% miles of the lake ... doubled from the time of completion of the project in 1956 through 1969.” Id. at 15 (ACF004356). Such increases in population are not without consequences, of course: “Wastes [sic] treatment plants in the Atlanta metropolitan area have failed to keep pace with the expanding population, and the increased low flows with a 650 cfs minimum flow at Atlanta have provided some relief in improving stream water quality below Atlanta.” Id. at 17 (ACF004358). The EIS also notes that “increased low flows have created a more dependable water supply for the Atlanta metropolitan area, thus helping to insure an adequate source of water for the expanding population. Storage in Lake Lanier has increased the dependability of a source of water for Gainesville, Gwinnett County, and Buford, Georgia.” Id.

Both the EIS and the comments thereto reference a study of Atlanta’s water quality and water supply underway at the time the EIS was prepared. See, e.g., id. at 26-27 (comments of the Environmental Protection Agency) (ACF004867-68). This study, referred to in the EIS as the “Atlanta Urban Study” or the “Atlanta Water Resources Study,” was a joint project of the Corps, the state of Georgia, and the ARC. Id. at 30 (ACF004371). Because the study was not completed in time for the EIS, the EIS stated that a new EIS should be written when the study was finished. Id. This study was not completed until the early 1980s, and is discussed below as the Metro Atlanta Area Water Resources Management Study (“MAAWRMS”). No new EIS has been completed since 1974.

Also in 1974, the Corps prepared a “Report on Consolidation of Existing Program Documents.” Boyce J. Christiansen, Consultant, U.S. Army Corps of Eng’rs, Report on Consolidation of Existing Program Documents, Lake Sidney Lanier (Buford Dam) Georgia (1974) (ACF004096). This report specifically addresses water supply in three different sections. In the first section on “Facilities,” the report states:

Two cities, Gainesville and Buford, obtain water directly from the reservoir. These cities relocated their water works facilities with new or an addition to these facilities. No storage space is allocated to either Gainesville or Buford in these water supply contracts. The Gainesville and Buford contracts provides [sic] for the maximum withdrawal of 8,000,000 and 2,000,000 gallons of water respectively from the reservoir in any 24 hour period. Gwinnett County on [sic] June 1971 initiated a request which would permit withdrawal direct from the reservoir of 40,000,000 gallons per day by 1990. In a contract dated July 2,1973 no storage space is allocated to the county for water supply, but the user will have the privilege of withdrawing water not to exceed that rate until such time as the Government studies of the areas [sic] water supply needs is [sic] completed. Withdrawal is not expected to be initiated for two or three years. Lake Lanier with its large water storage maintains a minimum flow of 650 cfs on the Chattahoochee River at Atlanta. The City of Atlanta and De Kalb County water systems draw their entire water supply from the Chattahoochee.

Id. at 32 (ACF004149).

In the section on “Trends,” the report noted that neither Gainesville nor Buford pays anything for the water each withdraws from the reservoir, but that “in the event the city desires to exceed th[e] [contractual] limitation an agreement will be necessary to provide payment for additional quantities withdrawn.” Id. at 54 tbl. 15 (ACF004180). The section also discussed Gwinnett County’s request for water supply withdrawals, stating that the requested withdrawal “would require a study of municipal and industrial water needs and a possible redistribution of project costs to include water supply as a project cost therefore a temporary contract on an interim basis was entered into.” Id. at 54-55 tbl. 16 (ACF00418081).

Finally, the report noted in the “Benefits” section that no revenues had yet been collected from Gwinnett County for the water-supply withdrawals. Id. at 63 (ACF004197).

In 1979, scientists at Georgia State University issued a report to the Corps on the environmental impacts of four of the alternatives being considered by the MAAWRMS mentioned above. Ga. State Univ. Team Project No. 834, Preliminary Environmental Impact Assessment of Water Supply Alternatives for the Atlanta Metropolitan Area (1979) (ACF006918). The alternatives under consideration were raising the water elevation of Lake Lanier, phasing out power generation at Buford Dam, constructing a second dam below Lake Lanier to further regulate the flows in the Chattahoochee River for the benefit of Atlanta’s water supply and waste treatment, and dredging the Morgan Falls reservoir. Id. at 33 (ACF006955). The report described Buford dam as a “multipurpose project, built 1) to control floods, 2) to improve water quality by means of flow augmentation, 3) to insure sufficient riverflow in the Chattahoochee River before Columbus, Georgia, and 4) to produce hydroelectric power.” Id. at 21 (ACF006943). The report did not mention Lake Lanier as an independent source of water supply, nor did it discuss the environmental impact of large-scale water-supply withdrawals from Lake Lanier. See, e.g., id. at 37 (“In addition, no attention has been given to the effects of additional water intakes, increases in allowable supplies taken through existing intakes, etc. Such factors will affect the flow in the river and should be analyzed.”) (ACF006959).

2. Drought Operations

Although flood control was a primary concern of both the Corps and Congress before and during construction of the Buford project, a drought in 1980 and 1981 caused the Corps to re-evaluate its operation of the project. The Corps formulated a “Drought Contingency Plan” to examine the operation of the ACF projects during the drought and “to explore alternative operational procedures during future periods of extreme drought.” U.S. Army Corps of Eng’rs, Drought Contingency Plan, Apalachicola, Chattahoochee, and Flint Rivers, Florida and Georgia ¶ 2, at 1 (1982) (ACF008205). The Drought Contingency Plan was required because, during the 1981 drought, “not all project functions were met.... Functions that were not fully provided were navigation and contractual hydropower requirements.” Id. ¶ 18, at 6 (ACF008210). The Plan did not comment on the fact that navigation and hydropower were two of the Congressionally mandated project purposes.

The Drought Contingency Plan described an agreement between Georgia Power and the Corps to provide minimum releases from Buford Dam of 1750 cfs at the request of Georgia Power each year between June 15 and September 15, to aid both water supply and water quality. Id. ¶ 14(b), at 5 (ACF008209). The previous release requirements for the dam were 600 cfs, 1958 Manual app. B ¶ 33, at B-18 to - 19 (ACF001801-02), although in the mid-1970s the Corps had agreed to increase the minimum releases to 650 cfs. Final EIS at 17 (ACF004358). In addition, the Drought Contingency Plan described water supply as a “principal function” of the ACF basin projects, stating that water supply “must always have a high priority in drought operations.” Id. ¶ 26, at 12 (ACF008216).

Municipal and industrial water supplies which are derived from the Chattahoochee River can probably be adequately supplied during a drought.....Even if Lake Lanier were drawn to elevation 1035 for other purposes there is still sufficient stored water which could be released through the low level sluice to meet the water supply requirements.

No difficulty is contemplated in meeting water supply volume requirements in a drought that is no worse than those which have occurred in the past. There may be, however, difficulty with particular pumping installations. For example, within Lake Lanier there are several withdrawal facilities which could not get water if the pool were drawn to unusually low levels. River pumping stations could face the same problem. For this reason conservation of water should be promoted by local government.

Id. At the time the plan was drafted, communities surrounding Lake Lanier withdrew approximately fifty-five million gallons per day from the lake. Memorandum from Acting Commander, S. Atl. Div. (Apr. 23,1982) (ACF008230).

In what appears to be an earlier version of the plan, called the “Drought Contingency Report,” the Corps stated that the “project purposes specified in the authorizing document included flood control, hydropower, and streamflow regulation for navigation.” U.S. Army Corps of Eng’rs, Drought Contingency Report, Apalachicola, Chattahoochee, and Flint Rivers, Florida and Georgia ¶ 3, at 1 (ACF008241). Under the heading “Project Purposes” the Drought Contingency Report provided: “Project costs for the Buford project have been allocated between the three legislatively authorized purposes. Prior to May 1979, recreation, water supply and water quality have always been considered to be functions of the Buford project and were accommodated as much as possible.” Id. ¶4, at 1-2 (ACF008241^42). The Drought Contingency Report noted that in a recent public notice the Corps had recommended that “recreation, water supply, and water quality control be acknowledged as full project purposes of the Lake Lanier project....” Id. ¶ 5, at 2 (internal quotation and citation omitted) (ACF008242). The public notice also provided that “[a]ny significant change (in operation) would require reconsideration of cost-sharing requirements for the total project.” Id. (internal quotation and citation omitted). According to the draft Drought Contingency Report, “[i]n other words, any ‘significant’ change favoring recreation, water supply or water quality over the three legislatively authorized purposes would require Congressional approval.” Id.

3. MAAWRMS

As discussed briefly above, in the early 1970s the United States Senate directed the Corps and other entities to engage in a study of Atlanta’s water resources. This study lasted from 1972 to 1981 and was published in September 1981 as the Metropolitan Atlanta Area Water Resources Management Study, or MAAWRMS. U.S. Army Corps of Eng’rs, Metropolitan Atlanta Area Water Resources Management Study: Final Report and Final Environmental Impact Statement (rev. ed. Sept. 1981) [hereinafter “MAAWRMS”] (SUPPAR001951). The final version of the MAAWRMS evaluated three “long-range water supply alternatives.” Id. at 1-16 (SUPPAR001978). The alternatives were: (1) construction of a reregulation dam below Buford Dam, (2) reallocation of storage at Lake Lanier, or (3) dredging the reservoir at Morgan Falls and reallocating Lake Lanier storage. Id.

The MAAWRMS noted that the Chattahoochee and Lake Lanier supply more than 90 percent of the total water supply for the metropolitan Atlanta area. Id. at 11-16 (SUPPAR001996). Lake Lanier “provides storage for flood control, power, navigation, recreation, industrial and domestic water supplies, and low-flow augmentation.” Id. at 11-37 (SUPPAR002017). However, the project costs, totaling more than $55 million, were allocated to only four project purposes or uses: hydropower, navigation, flood control, and recreation. Id. at 11-39 tbl. II — 6 (SUPPAR002019). Of these four purposes/uses, hydropower had borne the lion’s share of the costs, paying more than $44 million. 9

The MAAWRMS recognized that changes in the operation of the Buford project would require Congressional approval. For instance, the study noted that one of the proposed alternatives was construction of a new reregulation dam below Buford Dam. Id. at 11-48 (SUPPAR002028). If this construction was undertaken by local governments and not by the Corps, it would be the only alternative that would not require Congressional approval. See id. (“[I]t was considered that a reregulation dam constructed by local governments would be the most probable alternative to the other long-range alternatives which would require Congressional authorization for changes in operation of the Buford project.”).

The MAAWRMS also contained a list of the existing water-supply contracts for withdrawals from Lake Lanier. Id. at II-51 tbl. II — 8 (SUPPA002031). Despite the fact that only Gainesville and Buford had received Congressional authority to withdraw water from Lake Lanier, both Gwinnett County and the city of Cumming had also contracted with the Corps for these withdrawals. Id. The total withdrawals from the lake for water supply were given as 52.5 million gallons per day, with Gwinnett County receiving the majority of these withdrawals at 40 million gallons per day. Id. Although in this litigation the Corps characterizes Gwinnett County’s and Cumming’s water-supply contracts as “interim,” the MAAWRMS states that “[ojnly the method and rate of payment are of an interim nature.” Id. at 11-51.

In 1975, to meet an immediate increased need for water supply, the state of Georgia asked the MAAWRMS group to develop an interim water-supply plan that would allow the state to approve additional withdrawals from the river and provide a flow of 750 cfs at all times. Id. at 11-60. The Corps agreed to a plan that allowed additional water releases from Buford Dam. Id. That plan required the hydropower interests to schedule peak releases from the dam on weekends. Id. at 11-62 (SUPPAR002042). The power company agreed to this schedule only until 1983. Id.

The interim MAAWRMS, released in 1978, also recommended the imposition of a short-term water-supply plan in the event the MAAWRMS was not completed by 1983. Id. at 11-69 (SUPPAR002049). This short-term plan “include[d] raising the normal pool at Lake Lanier by [one] foot and increasing off-peak releases from Buford Dam.” Id. According to the MAAWRMS, the short-term plan “would necessitate reallocation or joint use of storage at the Lake Lanier project....” Id. at 11-71 (SUPPAR002051). This plan provided for an average annual withdrawal of 33 million gallons per day directly from Lake Lanier, and the maintenance of a 750 cfs flow at Atlanta. Id. at 11-72 (SUPPAR002052). The MAAWRMS recognized that the change in operations would have a negative effect on hydropower generation at Buford Dam, but calculated that hydropower would lose only one percent in benefits, and that raising the pool elevation one foot “would have a mitigating effect on this loss.” Id. at 11-74 (SUPPAR002054). The short-term plan was included as a recommendation of the Corps when the interim MAAWRMS was submitted in 1978. Id. at 11-78 (SUPPAR2058).

The final MAAWRMS stated that the “primary purpose of all long-range alternatives was to enhance water supply benefits through increased water supply availability from the Chattahoochee River.” Id. at II-83 (SUPPAR002063). Thus, “[although consideration was given to the impacts of each alternative on other water related uses of the lake and river,” any harm to those other uses was considered “incidental to formulation of each alternative for water supply.” Id. The MAAWRMS acknowledged, however, that “[a]ny proposed change in the operation of Buford Dam which would significantly impact on authorized project purposes would require Congressional approval.” Id. at III-2 (SUPPAR002074).

The MAAWRMS considered in detail three alternatives. The first alternative was the construction of a 4100 acre-foot reregulation reservoir and dam on the Chattahoochee six miles below Buford Dam. The second was a reallocation of storage at Lake Lanier. The third was a combination of dredging the Morgan Falls reservom (to increase storage at that reservoir) combined with a reallocation of storage at Lake Lanier. Id. at IV-3 (SUPPAR002105). In evaluating these alternatives, the MAAWRMS considered as a “base” condition average annual water-supply withdrawals of 14.6 million gallons per day from Lake Lanier, with projected average annual withdrawals from the lake of 53 million gallons per day by 2010. Id. at IV-7 tbl. IV-2 (SUPPAR002109). 10 The study also pointed out “how the increasing withdrawals from Lake Lanier result in a decrease in the dependable peak energy from Buford Dam.” Id. at IV-7 to -8 tbl. IV-3 (SUPPAR002109-10). According to the MAAWRMS, such decreases were acceptable because the Corps report on which Congress based its initial authorization of the project contemplated some increase in releases from the dam to support water supply. Id. at TV-9 (SUPPAR002111). The Corps report from 1947 stated that these increased releases “would not materially reduce the power returns” from the dam, and opined that the benefits of an assured water supply “ ‘would outweigh any slight decrease in system power value.’ ” Id. (quoting Newman Report ¶ 80, at 34 (ACF000668).). Although withdrawals of 53 million gallons per day would result in average annual power loss of $583,700 (in 1980 dollars), the MAAWRMS nevertheless considered 53 million gallons per day as the “base” withdrawal and “an integral part of each alternative.” Id. at IV-17 (SUPPAR002119).

The first alternative the MAAWRMS suggested was the construction of a reregulation dam 6.3 miles below Buford Dam. Id. at IV-24 (SUPPAR002126). This reregulation dam would store outflows from Buford Dam until those outflows were needed for water supply. Id. at IV-31 (SUPPAR002133). The total cost of this alternative was estimated at approximately $17.5 million. Id. at TV-44 tbl. TV-10 (SUPPAR002146). However, this alternative mitigated somewhat the lost power benefits assumed by the “base” scenario of increasing water-supply withdrawals from Lake Lanier. The power benefits gained in this alternative were estimated at $1.2 million annually. Id. at TV-50 tbl. IV-14 (SUPPAR002152). The MAAWRMS estimated that the increase in net benefits (for water supply, recreation, and power) under the first alternative would be $1.2 million annually, for a benefit-cost ratio of 1.75. Id. This alternative had the greatest annual net benefits of any of the final three alternative plans. Id. at IV-64 to 65 (SUPPAR002166-67). It also received the most support, from both federal and state agencies. Id. at VI-1 (SUPPAR002238). Should the federal government build the reregulation dam, the Corps would be re quired to seek Congressional authorization for the project. Id. at IV-81 (SUPPAR002183).

The second alternative called for reallocating storage at Lake Lanier from power to water supply. Id. at IV-86 (SUPPA002188); see also id. at IV-97 (SUPPAR002199) (describing alternative as involving “reallocating storage in Lake Lanier from power to water supply”). According to the MAAWRMS, water-supply storage in Lake Lanier amounted to 10,512 acre-feet in 1980, with 14.6 million gallons per day withdrawn from the lake. Id. at IV-87 tbl. IV-19 (SUPPAR002189). Under the second alternative, such storage would increase to 141,-685 acre-feet by 2010, with 53 million gallons withdrawn. Id. Power generation would decrease from more than 123 million kilowatt-hours in 1980 to 97.7 million kilowatt-hours in 2010. Id. “Losses in power benefits ... would occur primarily due to the need for scheduling additional weekend releases from Buford Dam for water supply.” Id. at IV-92 (SUPPAR002194). The MAAWRMS estimated the total net annual benefit of the second alternative as $475,100, with a benefit-cost ratio of 1.42. Id. Finally, the MAAWRMS acknowledged that “Congressional authorization would be required for reallocation of storage to water supply.” Id. at IV-97 (SUPPAR002199).

The final alternative was to dredge the downstream Morgan Falls Reservoir and also reallocate storage at Lake Lanier for water supply. Id. at IV-99 (SUPPAR002201). The loss of power benefits under this alternative was not as great as under the second alternative, from 124.5 million kilowatt-hours in 1980 to 112 million kilowatt hours in 2010. Id. at IV-100 tbl. TV-23 (SUPPAR002202). The annual net loss in power benefits would be $284,600. Id. at IV-112 (SUPPAR002214). The Lake Lanier storage reallocated to water supply would rise from zero acre-feet in 1980 to 48,550 acre-feet in 2010. 11 Id. at IV-100 tbl. IV-23 (SUPPAR002202). The dredging of the Morgan Falls Reservoir would result in increasing that reservoir’s storage capacity to 3200 acre-feet, with maintenance dredging required to maintain that capacity. Id. at IV-100 (SUPPAR002202). The MAAWRMS calculated the net annual benefit of the third alternative as either $875,000 if the dredged material could be sold, or $312,000 if that material could not be sold. Id. at IV-114. The benefit-cost ratio varied from 2.13 to 1.23. Id. As with the second alternative, “Congressional authorization would be required for reallocation of storage to water supply.” Id. at IV-119 (SUPPAR002221).

As discussed above, the MAAWRMS assumed a baseline for all alternatives of phased-in reallocations of storage at Lake Lanier from power to water supply. This phased-in reallocation, however, can skew the benefit-cost analysis. See id. at V-4 (SUPPAR002228) (“Non-phasing also reflects a more equal comparison of the costs of the three plans____”). When the MAAWRMS analyzed the three alternatives, assuming that all reallocations would occur at present to meet the future water-supply needs, the costs of the second and third alternatives rose and the net benefits decreased significantly. Id. at V-4 to -5 tbl. V-3 (SUPPAR002228-29). Thus, if the Corps did not have the authority to reallocate storage even in the limited way envisioned by the “baseline” of the MAAWRMS, the alternatives the study rec ommended were in general no longer cost-effective.

Ultimately, the MAAWRMS recommended the adoption of the first alternative. Id. at VI-4 (SUPPAR002241). According to the Corps, the first alternative was best suited “to provide a long-range water supply, improvement in water quality and the net positive contribution to the goal of National Economic Development.” Id. at EIS-1 (SUPPAR002256). The final recommendation of the MAAWRMS included a recommendation to Congress that “[r]ecreation, water supply, and water quality control should be acknowledged as full project purposes of the Lake Lanier project along with power, flood control, and navigation, and ... all of these purposes [should] be fully considered in future decisions affecting the use or operation of the project.” Id. ¶ 15, at IX-4 (SUPPAR002327).

4. Reregulation Dam to Reallocation of Storage

Congress considered the Corps’s recommendation with respect to the reregulation dam in 1982. Proposed Water Resources Dev. Projects of the U.S. Army Corps of Eng’rs: Hearings Before the Subcomm. on Water Resources of the H. Comm, on Public Works and Transp., 97th Cong. 713 (1982). At least one member of the Subcommittee expressed an unwillingness for the federal government to fund a project primarily for local water supply. See id. at 718 (statement of Rep. Edgar, Pennsylvania) (SUPPAR001443) (“Can you tell me when the corps got involved in providing local water supplies?”); id. at 719 (SUPPAR001443) (asking why the people of Atlanta did not construct the proposed dam “rather than the Federal Government coming in and providing the construction costs”); id. at 720 (SUPPAR001444) (“I am just wondering whether or not we are providing a subsidy to Atlanta at the Federal expense.... ”). Unlike during the construction of Buford Dam, however, Atlanta expressed its intention to share in the costs of the reregulation dam. Id. at 2459 (SUPPAR001451) (Letter from Andrew Young, Mayor, Atlanta, Ga., to Harry West, Exec. Dir., ARC (July 14, 1982)).

Not all testimony supported the proposed project, however. Nancy Wylie of the Georgia Conservancy testified against the reregulation dam, noting that some local governments in the area did not support that alternative and that even the Atlanta City Council had not strongly supported the project. Id. at 2508 (SUPPAR001476) (testimony of Nancy Wylie, Georgia Conservancy). Ms. Wylie also pointed out that neither Atlanta nor Georgia had made any binding financial commitments to fund the project. Id. Other participants in the MAAWRMS characterized the reregulation dam option as the preferred option of the Corps and one that essentially had been foisted on the other study participants. Id. at 2499-2500 (SUPPAR001471-72) (testimony of David Dingle, Chairman, MAAWRMS Citizen’s Task Force). Mr. Dingle supported the reallocation-of-storage alternative, noting that such reallocation would require including water supply as an authorized project purpose. Id. at 2502 (SUPPAR001473). 12

The Chair of the Subcommittee offered his thoughts on the project, asking that Georgia and Atlanta give the Subcommittee a firmer commitment to the proposed reregulation dam. Id. at 2520 (SUP PAR001482) (statement of Rep. Robert Roe). He also predicted what has come to pass:

[WJater resources and the need for water quality and water supply in this Nation is extraordinary.

Power is vital to the Nation, but water is absolutely essential, so that you are facing enormous competition for resources available.... [0]ne of the tragedies of our time is that it takes so long to get anything achieved....

I think it would be a shame to allow this to get away from [Georgia] and 5 or 7 years from today ... that you would not have the natural resources to be able to provide the economic resources required for the project.

Id.

In November 1984, the President’s Office of Management and Budget (“OMB”) declined to support the proposed project. “The plan ... appears to be a desirable project that would go a long way toward meeting water supply demand in the Atlanta area. However, ... non-federal development of the same reregulating dam [is] the most likely alternative to a Federal project for water supply.” Letter from Frederick N. Khedouri, Assoc. Dir., OMB, to Robert K. Dawson, Assistant Sec’y of the Army — Civil Works (Nov. 7, 1984) (SUPPAR036642). The letter emphasized the Administration’s policy of encouraging “non-Federal development of water resources....” Id. In January 1985 the Corps wrote to Congress, stating that it concurred with the OMB’s opinion on the proposed reregulation dam project. Letter from Robert K. Dawson, Assistant Sec’y of the Army — Civil Works, to Sen. Robert T. Stafford, Chairman, S. Comm, on Env’t and Pub. Works, at 2 (Jan. 8, 1985) (ACF010341).

Despite the lack of support from the Administration and the Corps, Congress authorized the construction of the reregulation dam in the Water Resources Development Act of 1986 (“1986 WRDA”), Pub.L. No. 99-662, tit. VI, § 601(a)(1), 100 Stat. 4137, 4140-41. The 1986 WRDA required that the project meet certain criteria, including a general design memorandum and supplemental environmental impact statement prepared and “jointly approved” by the Corps and Georgia. Id. It also provided that the dam could be constructed by Georgia or other local interests “at local cost.” Id. 100 Stat. at 4141. Congress did not appropriate any money to fund any construction costs for the reregulation dam. 13

Shortly thereafter, the Corps determined that reallocation of storage at Lake Lanier was a more feasible alternative than the construction of a reregulation dam. A March 25, 1988, report prepared by the Corps’s South Atlantic Division and entitled “Additional Information Lake Lanier Reregulation Dam,” stated that if the costs of acquiring the land that would be inundated by the dam rose, the reallocation alternative would become the most economic alternative. U.S. Army Corps of Eng’rs, Additional Information, Lake Lanier Reregulation Dam 2 (1988) (SUPPAR016865). The report warned that “[t]he storage to be reallocated under [the second MAAWRMS alternative] is beyond the approval authority of the Chief of En gineers.” Id. The Corps told ARC the same thing:

The Chief of Engineers has the discretionary authority to approve reallocation of storage if the amount does not exceed 50,000 acre-feet, or 15 percent of total usable storage, whichever is lower, and if the reallocation would not have a significant impact on authorized project purposes. Plan B [the second MAAWRMS alternative] would require the reallocation of 202,000 acre-feet of storage to meet the year 2010 peak demand of 103 mgd from the lake and 510 mgd from the river. The reallocation of 202,-000 acre-feet is much greater than the criteria of 50,000 acre-feet. Therefore, the required reallocation is not within the discretionary authority of the Chief of Engineers to approve. It can only be approved by the ASA(CW) [Assistant Secretary of the Army — Civil Works] if impacts are determined to be insignificant. We believe the power losses are significant and expect that Congressional approval would be required for the reallocation.

Letter from C.E. Edgar III, Major General, U.S. Army Corps of Eng’rs, to Harry West, Exec. Dir., ARC, at 5 (Apr. 15,1988) (SUPPAR017113). The Corps estimated that the cost of the storage reallocation would be more than $42 million. Id. at 6 (SUPPAR017114).

The Corps also provided a memorandum with a “chronology” of the Lake Lanier Reregulation Dam noting, among other events, a story in the May 31,1988, edition of the Gwinnett Daily Netus that the Corps was considering supporting the reallocation alternative rather than the reregulation dam alternative. Memorandum, Lake Lanier Reregulation Dam Chronology, to Joseph A. Goode 2 (Aug. 16, 1988) (SUPPAR016869). Several weeks later, the Corps informed Georgia that it would recommend “that the water supply needs be provided through reallocation of storage in Lake Sidney Lanier.” Letter from R.M. Bunker, Major General, U.S. Army Corps of Eng’rs, to J. Leonard Ledbetter, Comm’r, Ga. Dep’t of Natural Res. (Sept. 1, 1988) (SUPPAR016870). The Corps acknowledged the switch in an internal memorandum, characterizing the decision to promote storage reallocation as “a political decision.” Memorandum from James Couey, Chief of Eng’rs, to District Eng’r, U.S. Army Corps of Eng’rs, 1 (Sept. 6, 1988) (SUPPAR017083). This memorandum stated that the “next step” would be “for the Corps to prepare a storage reallocation report to submit through channels to the Secretary of the Army and Congress.” Id.

While the Corps prepared a Post-Authorization Change Report, Georgia prepared legislation to submit to Congress authorizing the reallocation of storage in Lake Lanier. On September 23, 1988, Georgia Governor Joe Frank Harris sent proposed reauthorization legislation to Georgia Senator Sam Nunn. Letter from Joe Frank Harris, Governor, Ga., to Senator Sam Nunn, Ga. (Sept. 23, 1988) (SUPPAR014842). The proposed legislation provided that the Buford project be “modified to provide that the Secretary is authorized to reallocate permanently from hydropower storage to water supply storage up to an additional 300,000 acre-feet for municipal water systems in the State of Georgia, at a total one-time cost not to exceed $29,000,000.” Id. at 2 (SUPPAR014843).

The Corps issued its “Draft Post-Authorization Change Notification Report For The Reallocation of Storage From Hydropower To Water Supply at Lake Lanier, Georgia” (“PAC Report”) in October 1989. Mobile Dist., U.S. Army Corps of Eng’rs, PAC Report (1989) (ACF041152). The PAC Report’s purpose was to recommend that Congress rescind its approval of a reregulation dam and instead approve a reallocation of storage in Lake Lanier to water supply. Id. at 1 (ACF041165). The PAC Report endeavored

to fully evaluate the future water supply demands for the Atlanta region to the year 2010, the storage needed from Lake Lanier to satisfy these projected demands, and to identify the associated impacts to all the project purposes, both upstream and downstream of Buford Dam, of reallocating storage from hydropower to water supply.

Id. at 6 (ACF041170).

According to the PAC Report, Lake Lanier has a total storage capacity of 2,554,-000 acre-feet. Id. at 12 (ACF041176). However, 867,600 acre-feet of that amount is considered “inactive” storage, 637,000 acre-feet is allocated to flood control, and 1,049,400 acre-feet is allocated to conservation storage. Id. The Report calculated that water supply demands in 2010 would require a reallocation of 207,000 acre-feet of storage, and recognized that “congressional approval may be required” for that reallocation. Id. The cost of the reallocation was estimated at $49.3 million. Id. at 21 (ACF041185).

Appendix A to the PAC Report was a draft Water Control Plan for the ACF basin (‘WCP”). Id. app. A, at A-l (ACF041197). The WCP’s objectives included balancing operations to meet the projects’ purposes. Id. app. A, at A-4 (ACF041200). The “purposes cited in the projects’ original authorizations” were “[f]ish and wildlife management, flood control, hydropower and navigation.” Id. In addition, “over the years a variety of activities (industrial and municipal water supply, instream recreation, water quality, etc ...) have become dependent upon the operational patterns of these projects.” Id.

The WCP set forth, apparently for the first time, so-called “action zones” for each of the reservoirs in the basin. Id. app. A, at A-ll (ACF041207). According to the Corps, these action zones “are to be used to determine minimum hydropower generation at each project, as well as the maximum possible assistance to navigation from conservation storage.” Id. The action zones took into consideration other factors, such as the time of year, historical pool levels, and “Resource Impact Levels” or “RILs.” Id. The RILs were the Corps’s attempt to quantify the effect on recreation of the various reservoir operations. Id. app. A, at A-8 (ACF041204). The RILs included: “Initial Impact Level,” defined as “the level where recreation impacts are first observed (i.e., some boat-launching ramps are unusable, most beaches are unusable or minimally usable and navigation hazards begin to surface)”; “Recreation Impact Level,” which is “the level where major impacts to concessionaires and recreation are observed (more ramps are not usable, all beaches are unusable, boats begin having problems maneuvering in and out of marina basin areas, loss of retail business)”; and the final level, Water Access Limited Level,” defined as “all or almost all boat ramps [are] out of service, all swimming beaches [are] unusable, major navigation hazards occur, channels to marinas are impassable and/or wet slips must be relocated, and a majority of private boat docks are unusable.” Id. For Lake Lanier, the Initial Impact Level was pool elevation 1066, Recreation Impact Level occurred at elevation 1063, and Water Access Limited Level was elevation 1060. Id. app. A, at A-9 (ACF041205). The normal elevation of Lake Lanier is 1070.

The “Water Control Guidelines” in the WCP listed objectives for all of the project purposes, including those initially authorized and those subsequently developed. Thus, the WCP outlined management for general hydropower operations, navigation, recreation, and water supply/water quality, among others. Id. app. A, at A-12 to -16 (ACF041202-12). For water supply, management “involves taking water from storage, either directly from the pool or through releases for downstream interests. Of primary concern is that sufficient drinking water is available for urban needs and that agreements to provide instream flow for water quality are not violated.” Id. app. A, at A-15 (ACF041211). “Releases from projects in the system will be the minimum (capacity) release for hydropower or releases needed for basin-wide water quality/water supply, whichever is greater.” Id.

Although the WCP did not fully explain the “action zones” on which it based the ACF basin operations, the Corps uses the action zones “to manage the lakes at the highest level possible for recreation and other purposes that benefit from high lake levels.” Memorandum, U.S. Army Corps of Eng’rs, ACF Drought Conference, at 11 (Sept. 20, 2007) (SUPPAR035773). The Corps describes those action zones as follows:

Zone 1 indicates that releases can be made in support of seasonal navigation [ ]when the channel has been adequately maintained, hydropower releases, and water supply, and water quality releases. If all the lakes are in Zone 1 or above, the river system would operate in a fairly normal manner.

Zone 2 indicates that water to support seasonal navigation may be limited. Hydropower generation is supported at a reduced level. Water supply and water quality releases are met. Minimum flow targets are met.

Zone 3 indicates that water to support seasonal navigation may be significantly limited. Hydropower generation is supported at a reduced level. Water supply and water quality releases are met. Minimum flow targets are met.

Zone 4 indicates that navigation is not supported. Hydropower demands will be met at minimum level and may only occur for concurrent uses. Water supply and water quality releases are met. Minimum flow targets are met.

Id. (alterations in original). The WCP’s RILs and “action zones” highlight the shift in operations at Buford from hydropower, flood control, and navigation to water supply and recreation.

The draft WCP was never finalized or adopted, because in 1990 the state of Alabama filed a lawsuit challenging the WCP and various water-supply withdrawal contracts between the Corps and Georgia communities.

C. History of the Litigation

From the time the Buford Dam was constructed and Lake Lanier filled, municipal entities had requested and received permission to withdraw water from the lake. Initially, only Gainesville and Buford, whose water intake structures on the Chattahoochee River had been inundated by Lake Lanier, withdrew water directly from the lake. The withdrawals were relatively small — eight million gallons per day for Gainesville and two million gallons per day for Buford — and amounted to slightly more than 10,000 acre-feet 14 of Lake Lani er’s “conservation” storage; storage that the Corps deemed usable storage, for hydropower or for purposes other than flood control.

1. Water-Supply Contracts

In the 1950s, Gwinnett County asked permission to make withdrawals from the lake. The Corps refused the request at that time, saying that such withdrawals would affect the project’s authorized purposes and that Gwinnett County would have to seek permission from Congress for the withdrawals. F.G. Turner, Ass’t Chief, Eng’g Div., Report on Withdrawal of Domestic Water Supply from Buford Reservoir ¶ 2, at 1 (1955) (SUPPAR005459). Congress ultimately authorized Gwinnett County to use storage space “in an amount not to exceed eleven thousand two hundred acre-feet of water annually,” Pub.L. No. 84-841, 70 Stat. 725 (1956) (amending 1946 RHA), but the Corps and Gwinnett County did not enter into any contracts at that time. 15 As discussed previously, in 1973, without invoking the authority provided by the 1956 statute, the Corps and Gwinnett County contracted for withdrawals of 40 million gallons per day from Lake Lanier pending the completion of the MAAWRMS and the adoption of the study’s recommended plan. Contract Between the U.S. and Gwinnett County, Ga., for Withdrawal of Water from Lake Sidney Lanier at 2 (July 2, 1973) (ACF004025). Forty million gallons per day amounts to almost 37,500 acre-feet of storage using the 1734 yield figure, and more than 43,700 acre-feet using the current 1485 figure. In 1985, Gwinnett County agreed to pay $5.40 per million gallons, or $216 per day, for the 40 million gallons it was allowed to withdraw daily from the lake. Dist. Eng’r, U.S. Army Corps of Eng’r Dist-Mobile, Civil Works Projects Water Supply Contract Status Report — Gwinnett County (1987) (SUPPAR014884). The Corps’s Status Report for this contract noted that the Corps’s South Atlantic Division had approved a supplement to this contract increasing the withdrawals to an annual average rate of 53 million gallons per day. Id. The Status Report also noted that the Gwinnett County contract was an “interim” contract “until 1 July 1989 to allow time for local interests to determine whether they will fund construction of a re-regulation dam downstream of Buford Dam.” Id.

The status reports for the other water-supply contracts similarly note the “interim” nature of the contracts. The ARC’S contract is for 377 million gallons per day, at a charge of $5.79 for each million gallons in excess of 327 million gallons per day. Dist. Eng’r, U.S. Army Corps of Eng’r Dist. — Mobile, Civil Works Projects Water Supply Contract Status Report (1988) (SUPPAR014880); see also Contract, Supplemental Agreement No. 1 to Contract No. DACW01-9-86-145 Between the U.S. and the Atlanta Reg’l Comm’n for Withdrawal of Water from the Chattahoochee River Downstream from Lake Sidney Lanier, Ga., at 1 (June 17, 1986) (ACF011978). Gainesville’s contract, dated May 27, 1987, is for 20 million gallons per day, at a charge of $12.44 for each million gallons in excess of the Congressionally authorized 8 million gallons per day. Contract Between the U.S. and City of Gainesville, Ga. for Withdrawal of Water from Lake Sidney Lanier, at 2-3 (May 28, 1987) (ACF014383). Buford had no new contract aside from the initial authorization of 2 million gallons per day. Dist. Eng’r, U.S. Army Corps of Eng’r Dist.— Mobile, Civil Works Projects Water Supply Contract Status Report — Buford (1988) (SUPPAR014882). The Status Report provided that Buford must enter into a new agreement if it wanted to withdraw more than 2 million gallons per day. Id. The Status Report for Cumming showed two contracts, one from 1978 and another from 1985. Dist. Eng’r, U.S. Army Corps of Eng’r Dist. — Mobile, Civil Works Projects Water Supply Contract Status Report — Cumming (1988) (SUPPAR014883). The 1978 contract allowed withdrawals of 2.5 million gallons per day; the 1985 contract allowed 5 million gallons per day. Id. The Corps charged Cumming $7.88 for each million gallons per day. Id. A contract dated November 16, 1988, allowed Cumming to withdraw 10 million gallons per day. Contract, Supplemental Agreement No. 2 to Contract No. DACW01-977-1096 Between the U.S. and the City of Cumming, Ga. For Withdrawal of Water from Lake Sidney Lanier, Ga., at 3 (Nov. 16,1988) (ACF014401).

All of these “interim” water-supply contracts (save Buford’s and Gainesville’s Congressionally authorized withdrawals of two million and eight million gallons per day, respectively), expired on January 1, 1990. See, e.g., Contract, Supplemental Agreement No. 5 to Contract No. DACW01-9-73-624 Between the U.S. and Gwinnett County, Ga. For Withdrawal of Water from Lake Sidney Lanier, at 3 (July 24, 1989) (ACF004006). However, the municipal entities continue to withdraw water pursuant to these contracts. Alabama and Florida therefore characterize the continuing withdrawals as occurring pursuant to “holdover” contracts. In addition, Alabama and Florida contend that the storage required by the “holdover” contracts has, for all intents and purposes, been reallocated to water-supply storage. They call this “de facto” reallocation.

The Eleventh Circuit Court of Appeals described the Corps’s decisionmaking with respect to the water-supply contracts:

Beginning in the 1970s, in accordance with the Corps’ view that water supply was an appropriate “incidental benefit” of the creation of [Lake Lanier], the Corps entered into interim contracts with local government entities in Georgia to allocate storage capacity in the Lake for local water supply.... As demand for water increased and the local governmental entities desired an assured permanent supply, the Corps in 1989 announced plans to seek congressional approval in accordance with the Water Supply Act of 1958 (“WSA”), 43 U.S.C. § 390b (2003), to enter into permanent water storage contracts with the local governmental bodies, proposing [the PAC Report] for congressional approval.

Alabama v. U.S. Army Corps of Eng’rs, 424 F.3d 1117, 1122 (11th Cir.2005).

2. Pre-MDL Litigation

On June 28, 1990, the state of Alabama filed a lawsuit against the Corps in the United States District Court for the Northern District of Alabama challenging the water-supply contracts and the draft WCP. 16 Shortly after the case was filed, the state of Florida moved to intervene as a plaintiff and the state of Georgia moved to intervene as a defendant. In September 1990, however, before the court ruled on the intervention motions, the parties requested that the court stay the matter pending settlement negotiations. As part of the joint motion to stay, the Corps agreed that it would not “execute any con tracts or agreements which are the subject of the complaint in this action unless expressly agreed to, in writing, by Plaintiff [Alabama] and Florida.” (Ex. A to Docket No. 20 in Case No. 3:07-md-00001 at 2.) The stay also provided that either party could terminate the stay by so notifying the court, the parties, and the proposed intervenors. (Id. at 2-3.) In 1992, the parties negotiated a Memorandum of Agreement (“MOA”) that temporarily resolved the parties’ differences. 17 The stay, however, remained in place.

In 1997, Congress ratified the ACF Compact. Pub.L. No. 105-104, 111 Stat. 2219 (1997). This Compact created an “ACF Basin Commission” composed of the Governors of Florida, Georgia, and Alabama, and a non-voting representative of the federal government, to be appointed by the President. Id. art. VI(b)-(c), 111 Stat. at 2221. The Commission was charged with establishing “an allocation formula for apportioning the surface waters of the ACF Basin among the states of Alabama, Florida and Georgia.” Id. art. VI(q)(12), 111 Stat. at 2222. The Compact did not nullify or otherwise modify any existing water-supply contract, but rather provided that, until a water allocation formula was developed, existing water-supply contracts would be honored and, further, that water-supply providers could increase the amount of water they withdrew from the ACF basin’s waterways “to satisfy reasonable increases in the demand” for such water. Id. art. VII(c), 111 Stat. at 2223-24. The right to use the water pending the allocation formula did not, however, create any permanent or vested rights to the water. Id. art. VII(c), 111 Stat. at 2224. The Compact was to expire on December 31, 1998, but was extended several times. Id. art. VIII(a)(3), 111 Stat. at 2224. The Compact finally expired on August 31, 2003, when the Commission was not able to agree on a water allocation formula. Alabama, 424 F.3d at 1123. The stay of the Alabama case remained in effect during the pendency of the Compact.

In 2001, the Georgia parties filed their own lawsuit against the Corps in the United States District Court for the Northern District of Georgia. Georgia v. U.S. Army Corps of Eng’rs, No. 2:01-CV-26 (N.D. Ga. filed Feb. 7, 2001) (“Georgia I”). This lawsuit challenged the Corps’s denial of Georgia’s water-supply request, which sought a permanent reallocation of storage in Lake Lanier for water supply. (Comp. ¶ 1.) In denying that request, the Corps found that the reallocation Georgia requested would “affect authorized project purposes” and that it “cannot be accommodated without additional Congressional authorization.” Letter from R.L. Brownlee, Ass’t Sec’y of the Army (Civil Works), to Roy E. Barnes, Governor, Ga. (Apr. 15, 2002) (ACF036354).

Meanwhile, the SeFPC filed its own lawsuit against the Corps in the United States District Court for the District of Columbia. Se. Fed. Power Customers v. Caldera, No. 1:00-cv-2975 (D.D.C. filed Dec. 12, 2000). The SeFPC alleged that the Corps’s decision to reallocate water supply to municipal entities in Georgia harmed the SeFPC’s ability to produce power from Buford Dam and increased the cost of that power. The Georgia parties intervened but Alabama and Florida did not intervene. The parties dispute whether Alabama and Florida were informed about the pendency of the case.

In 2003, before the ACF Compact expired, the Corps, the SeFPC, and the Georgia parties settled the SeFPC’s law suit and also resolved at least some of the issues pending in Georgia’s lawsuit. The Settlement Agreement required the Corps to negotiate interim contracts for the purchase of storage in Lake Lanier with Gwinnett County, Gainesville, and ARC. Settlement Agreement § 3.1, at 4 (SUPPAR024052). Under the terms of the Settlement Agreement, Gwinnett County would purchase 175,000 acre-feet of storage, which would provide a withdrawal of 152.4 million gallons per day from the lake. Id. § 3.1.1(a), at 5 (SUPPAR024053). Gainesville would purchase 20,675 acre-feet, or 18 million gallons per day, also from the lake. Id. § 3.1.1(b), at 5 (SUPPAR024053). ARC would purchase 45,183 acre-feet of storage, which would allow ARC to withdraw 367 million gallons per day from the Chattahoochee. Id. § 3.1.1(c), at 5 (SUPPAR024053). The Corps would calculate a credit to the SeFPC for hydropower benefits foregone, not to exceed the revenues received from the interim contracts. Id. § 4.1, at 13 (SUPPAR024061). The agreement also required the Corps to seek Congressional approval to make the interim contracts permanent, unless a court determined that the Corps was not required to secure Congressional approval for the permanent reallocation of storage. Id. § 3.1.4(a), at 10 (SUPPAR024058); see also Se. Fed. Power Customers, Inc. v. Caldera, 301 F.Supp.2d 26, 33 (D.D.C.2004). The agreement provided for a stay of its provisions pending the Corps’s completion of a NEPA review of the contracts. Settlement Agreement § 5.1.1, at 14 (SUPPAR024062).

In October 2003, the Alabama court enjoined the filing of the Settlement Agreement in the D.C. case, finding that the Corps had violated the terms of the stay in the Alabama case by entering into the Settlement Agreement without first seeking Alabama and Florida’s approval. See Alabama v. U.S. Army Corps, of Eng’rs, 357 F.Supp.2d 1313, 1316 (N.D.Ala.2005) (quoting 1990 Joint Motion to Stay). The court enjoined the Corps from filing or implementing the Settlement Agreement or entering into any new storage or withdrawal contracts affecting the ACF water basin without court approval. Id. at 1320-21. The Corps appealed the decision to the Eleventh Circuit Court of Appeals.

In February 2004, the Southeastern Federal Power Customers court approved the Settlement Agreement subject to the condition that the agreement not be implemented until the preliminary injunction in Alabama was dissolved. Se. Fed. Power Customers, Inc. v. Caldera, 301 F.Supp.2d at 35. The Alabama court refused to modify or vacate the preliminary injunction. Alabama, 357 F.Supp.2d at 1320.

In September 2005, the Eleventh Circuit Court of Appeals vacated the Alabama court’s preliminary injunction. Alabama v. U.S. Army Corps of Eng’rs, 424 F.3d 1117, 1133-36 (11th Cir.2005). The Southeastern Federal Power Customers court then entered final judgment, declaring the Settlement Agreement valid. Mem. & Order at 16, Se. Fed. Power Customers, Inc. v. Caldera, 301 F.Supp.2d 26 (D.D.C.2004). Alabama and Florida appealed that decision to the District of Columbia Circuit Court of Appeals. While the appeal was pending, the D.C. district court stayed the implementation of the Settlement Agreement to allow the Corps to complete the required NEPA processes. Mem. & Order at 1, Se. Fed. Power Customers, Inc. v. Caldera, No. 1:00-cv-2975 (D.D.C. Jan. 20, 2006).

In March 2007, the Judicial Panel on Multidistrict Litigation transferred the Alabama and Georgia cases and two other related cases (Florida v. U.S. Fish & Wildlife Serv., No. 4:06-410 (N.D. Fla. filed Sept. 6, 2006) and Georgia v. U.S. Army Corps of Eng’rs, No. 1:06-1473 (N.D. Ga. filed June 20, 2006) (“Georgia II”)) to this Court for resolution. The Panel did not transfer the Southeastern Federal Power Customers case, because that case was pending before the Circuit Court of Appeals and such transfers exceed the Panel’s authority. Since that time, three more cases have been transferred into the MDL: City of Columbus, Ga. v. U.S. Army Corps of Eng’rs, No. 4:07-125 (M.D. Ga. filed Aug. 13, 2007); City of Apalachicola, Fla. v. U.S. Army Corps of Eng’rs, No. 4:08-23, 2008 WL 460750 (N.D. Fla. filed Jan. 15, 2008); and finally, after the D.C. Circuit remanded the case to the district court, Se. Fed. Power Customers, Inc. v. Caldera, No. 1:00-2975 (D.D.C. filed Dec. 20, 2000).

3. Southeastern Federal Power Customers

As discussed briefly above, the Southeastern Federal Power Customers case and the attempted settlement of that case generated a flurry of litigation. Similarly, the D.C. Circuit’s decision on the legality of the Settlement Agreement in that case has generated much briefing and argument here. At least according to the parties, this Court’s interpretation of and deference to the D.C. Circuit’s opinion will dictate the outcome of the pending Motions.

In Southeastern Federal Power Customers v. Geren, 18 514 F.3d 1316 (D.C.Cir.2008) (“SeFPC”), the D.C. Circuit Court of Appeals considered whether the Corps exceeded its statutory authority by entering into the Settlement Agreement that required the Corps to reallocate some of the storage in Lake Lanier to water supply. The court held that the reallocation accomplished by the terms of the Settlement Agreement violated the requirements of § 301(d) of the WSA, 43 U.S.C. § 390b(d), because that reallocation was a “major operational change on its face.” Id. at 1318. The Corps’s failure to secure the approval of Congress before entering into the Settlement Agreement required, in the D.C. Circuit’s opinion, that the Agreement be set aside. Id.

The D.C. Circuit noted that the reallocations required by the Settlement Agreement amounted to more than twenty-two percent of Lake Lanier’s total conservation storage of 1,049,400 acre-feet, and was nine percent more than the storage space allocated to water supply in 2002. Id. at 1319-20. The court then turned to the statutory requirements for water supply. It noted that the WSA authorizes storage for water supply “ ‘in any reservoir project surveyed, planned, constructed or to be planned ... by the Corps of Engineers ... ’ so long as the costs of construction or modification are adequately shared by the beneficiaries.” Id. at 1321 (quoting WSA § 301(b), 72 Stat. at 319 (codified at 43 U.S.C. § 390b(b))). The court quoted WSA § 301(d), which requires that any modification that “would seriously affect the purposes for which the project was authorized, surveyed, planned, or constructed, or which would involve major structural or operational changes shall be made only upon the approval of Congress____” Id. at 1321-22 (quoting WSA § 301(d), 72 Stat. at 320 (codified at 43 U.S.C. § 390b(d))).

Alabama and Florida argued that the Settlement Agreement’s reallocations of storage constituted a “major operational change” within the meaning of the WSA. The appellees, who were the Georgia parties, the SeFPC, and the Federal Defendants, argued that the Settlement Agreement was not an operational change but merely preserved the status quo of allow ing “ ‘incremental increases in withdrawal amountsId. at 1322 (quoting Appellees’ Br. at 37). The appellees also argued that because the Settlement Agreement provided for temporary contracts of two 10-year periods, the contracts did not require Congressional approval. Id.

The D.C. Circuit rejected the appellees’ arguments:

On its face, then, reallocating more than twenty-two percent (22%, approximately 241,000 acre feet) of Lake Lanier’s storage capacity to local consumption uses constitutes the type of major operational change referenced by the WSA; the reallocation’s limitation to a “temporary” period of twenty years does not change this fact. Even a nine percent (9%, approximately 95,000 acre feet) increase over 2002 levels for twenty years is significant. Appellees’ contrary arguments are unpersuasive.

Id. at 1324 (citation omitted). The court also stated that “the appropriate baseline for measuring the impact of the Agreement’s reallocation of water storage is zero, which was the amount allocated to storage space for water supply when the lake began operation.” Id.

The court concluded:

In other circumstances it is conceivable that the difference between a minor and a major operational change might be an ambiguous matter of degree, where the Court would consider whether [the Corps’s] authoritative interpretation should be accorded deference ... in defining the term “major operational change.” But the Agreement’s reallocation of over twenty-two percent (22%) of Lake Lanier’s storage space does not present that situation. It is large enough to unambiguously constitute the type of major operational change for which section 301(d) of the WSA, 43 U.S.C. § 390b(d), requires prior Congressional approval.

Id. at 1325. The court thus reversed the district court’s approval of the Settlement Agreement and remanded the case. Shortly thereafter, the MDL Panel transferred the Southeastern Federal Power Customers case into this Tri-State Water Rights litigation.

The administrative record is complete and the parties’ Motions for Summary Judgment are fully briefed. 19 The matter is now ripe for the Court’s resolution.

DISCUSSION

A. Standard of Review

The Administrative Procedures Act (“APA”) waives a government agency’s traditional sovereign immunity by providing that “[a] person suffering legal wrong because of agency action, or adversely affected or aggrieved by agency action within the meaning of a relevant statute, is entitled to judicial review thereof.” 5 U.S.C. § 702. The statute also proscribes limits to this general rule. First, an agency action must be final to be reviewable: “A preliminary, procedural, or intermediate agency action ... is subject to review [only] on the review of the final agency action.” Id. § 704. In addition, relief under the APA is limited: a court may “compel agency action unlawfully withheld or unreasonably delayed,” id. § 706(1), and may

hold unlawful and set aside agency action, findings, and conclusions found to be—

(A) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law;

(B) contrary to constitutional right, power, privilege or immunity;

(C) in excess of statutory jurisdiction, authority, or limitations, or short of statutory right; [or]

(D) without observance of procedure required by law....

Id. § 706(2).

Because the agency action at issue here requires review of the agency’s interpretation of a statute — namely the Corps’s determination as to whether the storage reallocations require Congressional approval under the WSA — the Court must engage in a two-step analysis:

First, ... is the question whether Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress. If, however, the court determines Congress has not directly addressed the precise question at issue ... the question for the court is whether the agency’s answer is based on a permissible construction of the statute.

Chevron U.S.A, Inc. v. Natural Res. Def. Council, 467 U.S. 837, 842-3, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984) (footnote call numbers omitted). The Court is not required to set aside the agency’s construction merely because the Court’s interpretation differs from the agency’s. Id. at 843 n. 11, 104 S.Ct. 2778. However, the Court, not the agency, “is the final authority on issues of statutory construction and must reject administrative constructions which are contrary to clear congressional intent.” Id. at 843 n. 9, 104 S.Ct. 2778. Moreover, “a reviewing court ‘must reject administrative constructions ... that are inconsistent with the statutory mandate or that frustrate the policy that Congress sought to implement.’ ” Sierra Club v. Johnson, 541 F.3d 1257, 1265 (11th Cir.2008) (quoting Sec. Indus. Ass’n v. Bd. of Governors of Fed. Reserve Sys., 468 U.S. 137, 143, 104 S.Ct. 2979, 82 L.Ed.2d 107 (1984)).

The first step in the Chevron analysis is to determine Congressional intent using the “traditional tools of statutory construction.” Chevron, 467 U.S. at 843 n. 9, 104 S.Ct. 2778. “These tools include examination of the text of the statute, its structure, and its stated purpose.” Miami-Dade County v. U.S. Envtl. Prot. Agency, 529 F.3d 1049, 1063 (11th Cir.2008). If the examination of Congress’s intent does not resolve the matter, the Court then proceeds to the second step, which involves examining the Corps’s construction of the statute. That construction is “deemed reasonable if it is not arbitrary, capricious, or clearly contrary to law.” Ala. Power Co. v. Fed. Energy Regulatory Comm’n. 22 F.3d 270, 272 (11th Cir.1994) (citing Chevron, 467 U.S. at 844, 104 S.Ct. 2778). “Unexplained inconsistency is ... a reason for holding an interpretation to be an arbitrary and capricious change from agency practice under the [APA].” Nat’l Cable & Telecomms. Ass’n v. Brand X Internet Servs., 545 U.S. 967, 981, 125 S.Ct. 2688, 162 L.Ed.2d 820 (2005).

B. Standing

As they have in nearly every motion brought before this Court and other courts involved in litigating the issues in this case, the Georgia parties contest Alabama and Florida’s standing to bring this litigation. 20 See, e.g., Alabama, 424 F.3d at 1130 (holding that Alabama and Florida have standing because “Corps management of Lake Lanier that violates federal law may adversely impact the environment and economy downstream in the ACF Basin, thereby injuring Alabama and Florida”); SeFPC, 514 F.3d at 1322 (holding that Alabama and Florida have standing to assert “major operational change” because they assert “that the proposed reallocation of water storage will result in ‘diminish[ed][ ] flow of water reaching the downstream states’ ” (quoting Appellant’s Br. at 2)).

Standing is both a doctrine reflecting “prudential considerations that are part of judicial self-government” and “an essential and unchanging part of the case- or-controversy requirement of Article III.” Lujan v. Defenders of Wildlife, 504 U.S. 555, 560, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992). The

irreducible constitutional minimum of standing contains three elements. First, the plaintiff must have suffered an injury in fact — an invasion of a legally protected interest which is (a) concrete and particularized ... and (b) actual or imminent, not conjectural or hypothetical. Second, there must be a causal connection between the injury and the conduct complained of.... Third, it must be likely, as opposed to merely speculative, that the injury will be redressed by a favorable decision.

Id. at 560-61, 112 S.Ct. 2130 (citations and internal quotation marks omitted). “The party invoking federal jurisdiction bears the burden of establishing these elements.” Id. at 561, 112 S.Ct. 2130. Moreover, at the summary judgment stage, it is a plaintiffs burden to prove that genuine issues of material fact exist as to whether or not plaintiff can prove standing. See Mize v. Jefferson City Bd. of Educ., 93 F.3d 739, 742 (11th Cir.1996) (concluding that summary judgment is appropriate when “ ‘there is no genuine issue as to any material fact’”) (quoting Fed.R.Civ.P. 56(c)); see also Lujan, 504 U.S. at 561, 112 S.Ct. 2130 (noting that each element of standing must be proved “in the same way as any other matter on which the plaintiff bears the burden of proof, i.e., with the manner and degree of evidence required at the successive stages of the litigation”).

The Georgia parties contend that Alabama and Florida cannot establish any injury in fact, as Lujan requires. They argue that there is no evidence that the Corps’s support of water supply and recreation in Lake Lanier has resulted in any “discernable reduction in flows downstream in Alabama or Florida.” (Ga.’s Mem. in Opp’n to Ala. & Fla.’s Mot. for Partial Summ. J. at 68.) In support of this statement, they cite to an affidavit, a declaration, and a publication that is not part of the administrative record. (Id. (citing Ga.’s Mot. for Summ. J. Factual App. at ¶¶ 2.7-2.9).)

On a motion for summary judgment, such evidence might be sufficient to find no genuine issue of fact as to injury if the opposing party had no evidence to support its claimed standing. Such is not the case here. Alabama and Florida have cited declarations stating the opposite of the declarations and affidavits the Georgia parties cite. (Ala. & Fla.’s Reply Mem. in Supp. of Mot. for Partial Summ. J. at 35 (citing Ala. & Fla.’s Factual App. in Supp. of Mot. for Partial Summ. J. ¶¶ 1132-1226).) It is not the province of the Court, on a motion for summary judgment, to weigh the evidence and determine which evidence to credit. Mize, 93 F.3d at 742.

Alabama and Florida have come forward with evidence sufficient to support their contention that they have suffered harm because of the Corps’s operations in the ACF basin. For example, the Biological Opinion for the Jim Woodruff Dam (“BiOp”) notes that the lower flows in the Apalachicola in the spring and summer are likely due to “a combination of climatic differences ..., higher consumptive uses, as well as reservoir operations.” U.S. Fish & Wildlife Serv., Biological Opinion on the U.S. Army Corps of Engineers, Mobile District, Revised Interim Operating Plan for Jim Woodruff Dam and the Associated Releases to the Apalachicola River 56 (2008). The BiOp states that low flows “are likely among the most stressful natural events faced by riverine biota.” Id. at 57. In other words, according to government documents, low flows in the Apalachicola River are at least to some extent caused by the Corps’s operations in the ACF basin and consumptive uses of the water in the basin, and those low flows cause harm to the creatures that call the Apalachicola home. According to the evidence to which Alabama and Florida cite, low flows harm not only wildlife, but also harm navigation, recreation, water supply, water quality, and industrial and power uses downstream. Even if annually the average flows are reduced by only a small amount, as the Georgia parties argue, the actual variation in flows can wreak havoc on the downstream uses of the water.

Alabama and Florida have standing to bring their claims. Georgia’s Motion on this point is denied.

C. Effect of D.C. Circuit’s decision in SeFPC

Alabama, Florida, and the SeFPC urge this Court to find that the Corps and the Georgia parties are bound under the doctrine of collateral estoppel by the decision of the D.C. Circuit Court of Appeals in SeFPC. In that case, the Court of Appeals held that a reallocation of the magnitude contemplated by the invalidated Settlement Agreement constitutes a major operational change on its face. SeFPC, 514 F.3d at 1318. Not surprisingly, the Georgia parties and the Corps contend that the D.C. Circuit’s holding in SeFPC does not address many of the issues presented by this case, determining conclusively only that the Settlement Agreement was invalid under the WSA. The Georgia parties in particular contend that the D.C. Circuit limited its holding to a determination of the reallocation’s legality under the WSA and did not discuss what the Georgia parties believe is the authority provided by other federal statutes in combination with the WSA for the reallocation Georgia requests.

The law governing issue preclusion is well settled:

The doctrine of collateral estoppel [issue preclusion] bars relitigation of an issue if three requirements are met:

(1) that the issue at stake [is] identical to the one involved in the prior litigation;

(2) that the issue [was] actually litigated in the prior litigation; and

(3) that the determination of the issue in the prior litigation [was] a critical and necessary part of the judgment in that earlier action.

In re Held, 734 F.2d 628, 629 (11th Cir.1984). The Court has previously determined that Florida and Alabama could not relitigate here their claims that the Settlement Agreement in Southeastern Federal Power Customers was invalid. (Mem. & Order 8, October 22, 2007.) Thus, the SeFPC court’s holding that the Settlement Agreement is invalid is binding on all parties to this litigation.

However, Alabama and Florida do not limit their contentions to the validity of the Settlement Agreement. They argue that all of the following determinations from SeFPC are binding in this litigation:

1. No storage for water supply has ever been allocated by Congress at Lake Lanier.

2. The correct “baseline” for measuring the Corps’s proposed and “de facto” reallocations is zero.

3. To determine whether the proposed and “de facto” reallocations constitute major operational change, the Court must evaluate the percentage of conservation storage reallocated.

4. To calculate the percentage of storage reallocated, the Court must compare the amount of reallocated storage to the total conservation storage.

5. The WSA applies to both interim and permanent reallocations of storage.

6. As of 2002, approximately thirteen percent of Lake Lanier’s conservation storage was allocated to water supply.

7. The Corps has never reallocated 95,-000 acre-feet or more in a federal reservoir without seeking Congressional approval.

8. A reallocation of twenty-two percent of Lake Lanier’s conservation storage is a major operational change on its face.

The D.C. Circuit stated all of these things in its opinion in SeFPC. However, it is not the case that all of these statements were “critical” and “necessary” parts of the judgment in SeFPC. Indeed, only two conclusions were necessary to the holding in SeFPC that the Settlement Agreement was invalid. First, the D.C. Circuit concluded that the WSA applied to interim reallocations of storage. SeFPC, 514 F.3d at 1324-25 (“[I]t is unreasonable to believe that Congress intended to deny the Corps authority to make major operational changes without its assent, yet meant for the Corps to be able to use a loophole to allow these changes as long as they are limited to specific time frames, which could theoretically span an infinite period.”). Without this conclusion, the court could not have determined that the Settlement Agreement violated the WSA, because the Settlement Agreement involved temporary reallocations of storage for water supply.

The second conclusion that was critical and necessary to the SeFPC holding is that a reallocation of twenty-two percent of Lake Lanier’s total conservation storage is a major operational change on its face. Id. at 1324. This conclusion is the underpinning of the judgment in SeFPC that the Settlement Agreement is invalid under the WSA.

The remaining determinations are not, however, binding on the parties or on this Court. This is not to say that the D.C. Circuit’s comments about the appropriate “baseline” for evaluating storage reallocations and its calculations regarding storage reallocations are not persuasive authority, for those comments certainly are persuasive. This Court will not, however, blindly accept the SeFPC court’s conclusions; instead, the Court will make its own determination of the evidence and how that evidence affects the legal decisions to be made here.

D. The Water Supply Act of 1958

In 1989, the Corps decided that the WSA did not require it to seek Congressional authorization for the reallocation of significant amounts of Lake Lanier’s storage to water supply. Under the APA and Chevron, this Court must determine whether that decision was arbitrary and capricious. To make that determination, the Court must first examine the statute itself to determine whether Congress has spoken to the precise question at issue before the Court: whether the reallocations undertaken prior to and those proposed by the PAC Report or Georgia’s 2000 water supply request constituted a major operational change or seriously affected the purposes for which the Buford Dam was authorized.

The WSA provides in relevant part:

(a) Declaration of policy

It is ... declared to be the policy of the Congress to recognize the primary responsibilities of the States and local interests in developing water supplies for domestic, municipal, industrial, and other purposes and that the Federal Government should participate and cooperate with States and local interests in developing such water supplies in connection with the construction, maintenance, and operation of Federal navigation, flood control, irrigation, or multiple purpose projects.

WSA § 301(a), 72 Stat. at 319 (codified at 43 U.S.C. § 390b(a)).

In carrying out the policy set forth in this section, it is ... provided that storage may be included in any reservoir project surveyed, planned, constructed or to be planned, surveyed, and/or constructed by the Corps of Engineers or the Bureau of Reclamation to impound water for present or anticipated future demand or need for municipal or industrial water,.... Provided, That the cost of any construction or modification authorized under the provisions of this section shall be determined on the basis that all authorized purposes served by the project shall share equitably in the benefits of multiple purpose construction ....

Id. § 301(b), 72 Stat. at 319 (codified at 43 U.S.C. § 390b(b) (emphasis in original)).

Modifications of a reservoir project heretofore authorized, surveyed, planned, or constructed to include storage [for water supply] which would seriously affect the purposes for which the project was authorized, surveyed, planned, or constructed, or which would involve major structural or operational changes shall be made only upon the approval of Congress....

Id. § 301(d), 72 Stat. at 320 (codified at 43 U.S.C. § 390b(d)).

Thus, the WSA provides that the Corps may set aside storage for water supply in a previously constructed reservoir as long as (1) the beneficiaries of that storage pay a proportionate share of the costs of the project, and (2) the modification does not seriously affect the project’s purposes or constitute a major structural or operational change. There can be no debate that the water-supply users have not paid a proportionate share of the project’s costs, although the record is less clear whether they would be willing to do so were the Court to find that Congressional approval for the requested storage reallocations was not required. The Court will assume for the purposes of the instant Motions that the beneficiaries of the proposed and “de facto” reallocations would pay a proportionate share of the cost of the Buford project.

1. Authorized Project Purpose

The WSA inquiry is academic if water supply was an authorized project purpose of the Buford project, either from the initiation of the project or made so by Congress at some point after the project began. The Georgia parties contend that water supply was always a purpose of the Buford project, as evidenced by the sign at an observation point above Lake Lanier, reproduced on the first page of nearly every one of the Georgia parties’ briefs. This sign states that the “PRIMARY PURPOSES” of Buford Dam are “FLOOD CONTROL — POWER—WATER SUPPLY-INCREASED FLOW FOR NAVIGATION.” (Ga.’s Mem. in Supp. of Summ. J. at 1 (SUPPAR005533).) This sign, however, is not authoritative legislative history, and it is legislative history that the Court must examine to determine whether water supply, in the form of large withdrawals from Lake Lanier itself, was an authorized project purpose.

The legislative history of the Buford project is set forth in detail above and will not be repeated here. It is worth noting that, both before and during construction of Buford Dam, the Corps consistently described the primary purposes of the project as flood control, navigation, and hydropower. See, e.g., F.G. Turner, Ass’t Chief, Eng’g Div., Report on Withdrawal of Domestic Water Supply from Buford Reservoir ¶ 2, at 1 (1955) (SUPPAR005459) (Corps told Gwinnett County that “the primary authorized purposes of the Buford project were flood control, power and low-flow regulation for navigation and other purposes”); 1958 Manual ¶ 85, at 27 (ACF001677) (describing Buford Dam as “a multiple-purpose project with major uses of flood control, flow regulation for navigation, and power”); U.S. Army Corps of Eng’rs, Cost Allocation Studies, Apalachicola, Chattahoochee and Flint Rivers Projects, Basis of All Allocations of Costs for Buford and Jim Wood-ruff Projects Adopted by the Chief of Engineers, app. A, at A-9 (1959) (ACF002116) (allocating costs to the “primary purposes of the Buford project”: navigation, flood control, and power). Others also recognized that the purposes of the project did not include water supply. See, e.g., Civil Functions, Dep’t of the Army Appropriation Bill for 1949: Hearing on H.R. 5524 Before the Sub-comm. of the H. Comm. on Appropriations, 80th Cong. 723 (1948) (statement of Rep. Stephen Pace, Georgia) (SUPPAR026606) (describing the ACF projects as having three purposes: navigation, power, and flood control); Civil Functions, Dep’t of the Army, Appropriations, 1953: Hearings on H.R. 7268 Before the Subcomm. of the H. Comm. on Appropriations, 82d Cong. 1196-97 (1952) (statement of Rep. Davis, Georgia) (SUPPAR02667980) (describing the project as providing flood control, power, and navigation benefits); Public Works Appropriations for 1957: Hearings Before the Subcomm. of the H. Comm. on Appropriations, 84th Cong. 355-57 (1956) (statement of Rep. James Davis, Georgia) (SUPPAR02672022) (discussing flood control, navigation, and power benefits); Proposed Water Resources Development Projects of the U.S. Army Corps of Eng’rs: Hearings Before the Subcomm. on Water Resources of the H. Comm, on Public Works and Transp., 97th Cong. 3251 (1982) (Letter from W.T. Bush, Co-Chairman, Metro Atlanta Water Managers Assoc., Gwinnett County Water & Sewerage Auth., to Sen. Sam Nunn, Ga., at 1 (Aug. 21, 1980) (water supply “not specifically authorized as a purpose” of the Buford project)) (SUPPAR001491).

In the decades after Buford Dam was completed, the Corps continued to describe the project’s purposes as hydropower, flood control, and navigation. See, e.g., Final EIS Statement of Findings (1974) (ACF004338) (Buford Dam’s “[ajuthorized project purposes provide peaking hydroelectric power, flood control, and low flow augmentation”); Drought Contingency Report, Apalachicola, Chattahoochee, and Flint Rivers (A-C-F), Florida and Georgia ¶ 4, at 1-2, in U.S. Army Corps of Eng’rs, Drought Contingency Plan, Apalachicola, Chattahoochee, and Flint Rivers (1982) (ACF008241^2) (stating that costs at Buford project “have been allocated between the three legislatively authorized purposes” of flood control, navigation, and hydropower).

There is also no doubt that both Congress and the Corps anticipated some benefits to water supply from the project. See, e.g., Park Report ¶ 243, at 77 (ACF000160) (describing water supply as “direct benefit” but ascribing no monetary value to water supply benefit); Newman Report ¶ 68, at 27 (ACF000661) (noting that water supply was “incidental benefit! ]” of Buford project); Definite Project Report ¶ 115, at 41 (ACF001486) (describing project’s “principle purposes” as: “to provide flood control; to generate hydroelectric power; to increase the flow for open-river navigation in the Apalachicola River below Jim Woodruff dam; and to assure a sufficient and increased water supply for Atlanta”). As discussed previously, however, the water supply benefit was not from storage for water supply provided by Lake Lanier. Rather, the water supply benefit derived from the regulation of the Chattahoochee River’s flow provided by the dam and the releases for hydropower. Civil Functions, Dep’t of the Army, Appropriations for 1952: Hearings Before the Subcomm. of the H. Comm. on Appropriations, 82d Cong. 120, 121-22 (1951) (statement of Col. Potter, Corps officer) (SUPPAR026656, SUPPAR026657-58) (“[The Buford project does not] furnish[] water directly or furnish! ] storage for that purpose.... [Water supply is] an adjunct to the power supply and flood control. Had we put in some storage purely for water supply, which they would tell us to release at certain intervals, we would then charge them for it.”);. Civil Functions, Dep’t of the Army Appropriations for 1951: Hearings on H.R. 5376 Before the Subcomm. of the H. Comm. on Appropriations, 83d Cong. 503 (1953) (statement of Gen. Chorpening, Corps officer) (SUPPAR026688) (“[The project] will not make available any more water than is now going past Atlanta. It is only going to make it flow by at a more uniform rate.”); Civil Functions, Dep’t of the Army Appropriations, 1955: Hearings on H.R. 8367 Before the Sub-comm. of the S. Comm, on Appropriations, 83d Cong. 325 (1954) (statement of Col. Whipple, Corps officer) (SUPPAR026699) (stating that water supply “is purely an incidental benefit on account of the power releases which does not require any storage to be devoted to that purpose”).

Indeed, from 1955, when the Corps told Gwinnett County that Congressional authorization would be required to accommodate the county’s water-supply request, until at least 1988, when the PAC Report sought Congressional approval for the reallocation of storage in Lake Lanier to water supply, the Corps recognized that allowing water-supply withdrawals from the lake was not an authorized purpose of the project and would require Congress’s approval. Even in 2002, long after this litigation began, Earl Stockdale, the Corps’s Deputy General Counsel, concluded that Georgia’s 2000 water-supply request “would result in serious impacts on other project purposes” so that the Corps could not grant that request “absent legislative authority.” Memorandum from Earl Stockale, Deputy General Counsel, Civil Works & Env’t, to Acting Ass’t Sec’y of the Army for Civil Works 2 (Apr. 15, 2002) [hereinafter “2002 Stockdale Memorandum”] (ACF036355). The 2002 Stockdale Memorandum determined that, even if water supply was a specifically authorized project purpose, the Corps would still lack the authority to grant Georgia’s request without Congressional approval because the Corps did not “have the authority to reorder specifically authorized project purposes without additional Congressional authorization.” Id. at 13 (ACF036367); see also Memorandum from E. Manning Seltzer, General Counsel, U.S. Army Corps of Eng’rs, to Special Assistant to Sec’y of the Army for Civil Functions ¶ 4, at 2 (Jan. 21, 1969) (SUPPAR001361) (“[T]he discretionary authority given the Chief of Engineers to make post-authorization changes in projects extends only to what might be termed engineering changes ... [such as] minor variations in the allocation of storage for the various project purposes----”).

At some point between the 2002 Stock-dale Memorandum and the present Motion, the Corps changed its mind on this important issue. Attached to the Corps’s brief in this matter is a new memorandum from Mr. Stockdale which concludes that the Corps does have the authority to reallocate storage in Lake Lanier to water supply and that Congressional authorization is not required. Memorandum from Earl Stockdale, Chief Counsel, to the Chief of Engineers (Jan. 9, 2009) [hereinafter “2009 Stockdale Memorandum”]. The 2009 Stockdale Memorandum is not part of the administrative record in the case, but the Corps urges the Court to accept the memorandum as part of that record, because it allegedly is an “extra-record document that helps explain complex facts provided in the administrative record and helps explain the Corps’ past and present legal interpretation of its governing statutes and regulations.” (Corps’s Mem. Supp. Mot. for Summ. J. at 3 n.4.) As stated at the hearing on these Motions, however, the Court will not make the 2009 Stockdale Memorandum part of the administrative record in this case. It does not shed any light on the Corps’s decisionmaking with respect to the actions challenged here. Nor does the memorandum explain any complex facts that the Court is unable to understand without such assistance. Moreover, the memorandum is clearly a document prepared for litigation purposes only; large sections of the memorandum appear verbatim in the Corps’s brief with no attribution. The 2009 Stockdale Memorandum does little more than justify the Corps’s current legal position. The merits of that position are for the Court, not the Corps, to decide.

Having thoroughly reviewed the legislative history and the record, the Court comes to the inescapable conclusion that water supply, at least in the form of withdrawals from Lake Lanier, is not an authorized purpose of the Buford project. Therefore, if the Corps’s actions to support water supply constitute “major structural or operational changes” or “seriously affect” the project’s authorized purposes, the Corps was required to seek Congressional approval for those actions and its failure to do so renders the actions illegal. WSA § 301(d), 72 Stat. at 320 (codified at 43 U.S.C. § 390b(d)).

2. Major Operational Change

The Corps’s actions to support water supply in Lake Lanier have taken the form of reallocations of the lake’s storage capacity to water supply. In other words, by committing to allow municipal entities to withdraw a certain amount of water from Lake Lanier, the Corps has either explicitly or effectively allocated some of Lake Lanier’s storage to those withdrawals. Because water supply is not an authorized purpose of Lake Lanier, if any of these reallocations constitute a major structural or operational change or seriously affect the purposes for which the project was authorized, the Corps must seek Congressional approval for the reallocations.

Alabama, Florida, and the SeFPC challenge several of the Corps’s water-supply reallocations. The first are what Alabama and Florida call “de facto” reallocations. Prior to the initiation of this lawsuit in 1990, the Corps had entered into water-supply contracts with several Georgia entities: ARC, Gwinnett County, Gainesville, Buford, and Cumming. While all of these contracts expired in 1989 or 1990, there is no dispute that the Corps continues to allow these entities to withdraw water pursuant to the contracts today. In fact, the amount of water these entities presently withdraw far exceeds the amount they were entitled to under the so-called “holdover” contracts. With the exception of the withdrawal amounts approved by Congress in the 1950s, Alabama and Florida contend that all of the “holdover” contracts require an illegal reallocation of storage to water supply. The Corps maintains that these contracts were interim only and that no permanent reallocations were intended or accomplished under the pre-1990 contracts. However, the Corps is bound by the D.C. Circuit’s determination in SeFPC that interim contracts are subject to the strictures of the WSA. Thus, the “de facto” reallocations accomplished by the “holdover” contracts must be evaluated under the WSA.

The PAC Report endeavored to make permanent these “de facto” reallocations and some additional reallocations of storage. The PAC Report’s reallocations are the second type of reallocations at issue.

The final reallocations that must be evaluated under the WSA are the reallocations requested by Georgia in the 2000 water-supply request. These reallocations are the largest of the three reallocations at issue. Thus, if the Court determines that either the “de facto” reallocations or the PAC Report’s reallocations violate the WSA, then the water-supply request’s reallocations likewise violate the WSA.

a. “De facto ” reallocations

Starting in the early 1970s, the Corps began allowing municipalities surrounding Lake Lanier to withdraw water directly from the lake. Two of these municipalities, Buford and Gainesville, had a preexisting right to withdraw some water from the lake because their previous water intake structures on the Chattahoochee River were inundated by Lake Lanier. The Corps recognized that it could not unilaterally determine that the remaining municipalities were allowed to withdraw large amounts of water from the lake, and thus characterized the various contracts as “interim.” For example, the Corps told Gwinnett County in 1973 that it could withdraw water from the reservoir, pending the completion of the MAAWRMS and the changes in the project that the Corps expected to result from that study.

By 1990, contracts were in place for reallocations that would allow 85 million gallons per day to be withdrawn from Lake Lanier and 50 million gallons per day to be withdrawn from the Chattahoochee River. 21 In 2006, the average daily withdrawals from the lake totaled 141 million gallons per day. Report, Water Withdrawals — Lake Sidney Lanier (Buford Reservoir), Georgia — Chattahoochee River — ACF Basin 1-8 (no date) (ACF044236^3). The ARC’s average daily withdrawal was 316 million gallons per day. 22 Id. at 9.

Under normal operations, the contracted-for withdrawal amounts equal approximately 86,200 acre-feet of storage for the lake withdrawals, and approximately 50,-700 acre-feet for the excess river withdrawals. (See supra n. 14.) In 2006, the actual withdrawals required 143,000 acre-feet of storage for the withdrawals from the lake. Although the ARC did not require the additional 50 million gallons per day in its contract with the Corps, by virtue of the Corps’s commitment to provide that amount (should ARC need it), the additional 50 million gallons per day, or 50,700 acre-feet, were nevertheless held in water-supply storage and were unavailable for other uses. Thus, the average daily total amount of storage in Lake Lanier dedicated to water supply was 193,700 acre-feet under normal conditions, or 18.5% of Lake Lanier’s conservation storage of 1,049,400 acre-feet. If the Corps’s “critical yield” calculations are used, however, the amount of storage dedicated to water supply rises to an average of slightly more than 208,800 acre-feet. The Corps generally calculates storage requirements using critical yield, as opposed to normal operations. PAC Report, app. C, at C-l (ACF041299) (stating that to determine storage-yield relationship, the Corps selects a severe drought period “during which the project will be expected to provide a ‘firm’ yield”); see also Steven R. Cone, Team Leader, Planning & Pol’y Div., U.S. Army Corps of Eng’rs, Summary of “Technical Data” on Impacts of GA Request for WS at Lake Lanier ¶ 1 (2002), in 2002 Stockdale Memorandum, enclosure (SUPPAR005092) (using “critical period” yield for storage calculations).

According to the Corps’s storage calculation method, the reallocation accomplished by virtue of the “holdover” contracts is 208,100 acre-feet, or 19.8% of Lake Lanier’s conservation storage. This calculation assumes that 327 million gallons per day in the ARC’s river withdrawals are indeed “incidental” to the power operations at the dam, a point that the parties vigorously dispute. It also assumes that the “baseline” for operations in Lake Lanier is zero storage for water supply. The D.C. Circuit concluded that zero storage was the correct baseline but, as discussed above, neither the Court nor the parties are bound by that conclusion.

The base of operations at Buford Dam was to provide 600 cfs of flow past Atlanta and to allow Buford and Gainesville to withdraw a total of 10 million gallons per day from Lake Lanier. 1958 Manual app. B, at B-13 (ACF001796) (providing for flows of 600 cfs to Atlanta). In 1975, the Corps, Atlanta, and Georgia Power Company agreed that “existing practices” allowed an average annual downstream withdrawal of 230 million gallons per day. (Corps’s Mem. Supp. Mot. Summ. J. at 31) (citing Letter from Edwin C. Reiser, Col., U.S. Army Corps of Eng’rs, to Leonard Ledbetter, Dir., Ga. Dep’t of Natural Res. (July 21, 1975) (SUPPAR036976-77).) In 1979, the same parties determined that “an annual average of 266 million gallons per day ... could be withdrawn from flows that occur incidentally as a result of project operations.” (Id. at 47 n. 34 (referencing the 1979 Modified Interim Plan, described in MAAWRMS at 8 (ACF015500), but not included in the administrative record).) 23 Withdrawals of 266 million gal- Ions per day would, however, require operational changes. (Id. at 32 (citing Letter from Kenneth E. McIntyre, Brigadier Gen., U.S. Army Corps of Eng’rs, to Leonard Ledbetter, Dir., Ga. Envt’l Protection Div. (Apr. 27, 1979) (SUPPAR03699737002)).)

Thus, the Corps determined in 1975 that the “baseline” for operations was 230 million gallons per day downstream, plus the 10 million gallons per day Gainesville and Buford were Congressionally authorized to withdraw from the lake. This “baseline” amounted to slightly less than 224,700 acre-feet of storage, using the 1734 cfs yield figure. The 1979 baseline of 266 million gallons per day for downstream withdrawals required storage of almost 258,400 acre-feet. In 2006, the Corps allowed an average of 141 million gallons of water to be withdrawn daily from the lake, and committed to 377 million gallons per day for the ARC’S use downstream. See Report, Water Withdrawals- — -Lake Sidney Lanier (Buford Reservoir), Georgia— Chattahoochee River — ACF Basin 1-9 (Corps document listing total withdrawals from 1987 through September 2007) (ACF044236-44). These commitments amount to almost 485,000 acre-feet of storage using the 1734 cfs yield (and 566,300 acre-feet using the more current 1485 cfs yield figure). This is 226,600 acre-feet more than the “base” operations the Corps described in 1979 and 260,300 acre-feet more than the 1975 base operations. 24 Whichever baseline is used, the difference is more than 21.5% of Lake Lanier’s total conservation storage. Thus, without any Congressional authorization, the Corps has reallocated nearly a quarter of Lake Lanier’s conservation storage to support water supply.

That this reallocation is a major operational change is self-evident. The D.C. Circuit held that a reallocation of twenty-two percent of Lake Lanier’s conservation storage was a major operational change “on its face” and, as discussed previously, the parties are bound by this holding. The WSA requires the Corps to seek Congress’s authorization before effecting any major changes to project purposes. The Corps failed to do so and thus the so-called “de facto” reallocations violate the WSA.

b. PAC Report

The PAC Report recommended that Congress approve a reallocation of 207,000 acre-feet of storage in Lake Lanier to support water supply. PAC Report at 12 (ACF041176). Under the Corps’s calculations, this amounts to 19.7% of the total conservation storage in Lake Lanier.

The PAC Report assumed that by 2010, water-supply withdrawals from Lake Lanier would reach 151 million gallons per day. Id. app. C, at C-2 (ACF041300). The projected downstream needs were 378 million gallons per day. Id. Using a “firm yield” figure of 1734 cfs from the 1939— 1942 drought, 25 151 million gallons per day of lake withdrawals requires 141,700 acre-feet of storage. Id. 26

To calculate the storage required for downstream withdrawals, the Corps assumed that 200 million gallons per day were available for withdrawal downstream during the 66-hour off-peak (weekend) generation period, as a result of the smaller turbine’s releases of 600 cfs during this period. The Corps calculated that the storage necessary to accommodate the extra 178 million gallons per day, or 275 cfs, necessary for downstream water supply during this off-peak period was 65,225 acre-feet. Id. app. C., at C-4 (ACF041302). To achieve this number, the Corps did not perform the usual calculation. 27 Rather, the Corps assumed that 378 million gallons per day could be accommodated by existing operations during peak generation periods. According to the Corps, the only withdrawal that would require a reallocation of storage was the 178 million gallons per day in non-incidental withdrawals that occurred during the off-peak generation period of 66 hours, or 2.75 days. Thus, the Corps multiplied the cfs required for 178 million gallons per day (calculated as 178 x 1.547) by 2.75 to give a “dsf’ figure. 28 The dsf were then divided by 7 days to give a daily cfs rate of 108 cfs. The Corps then performed the usual calculation (108/1734 x 1,049,400) to determine that 178 million gallons per day of off-peak withdrawals would require storage of only 65,225 acre-feet.

The assumption that 378 million gallons per day is available downstream as incidental to the peak operation of the dam is, however, far greater than any assumption the Corps has ever made regarding “incidental” operation of the project. If 378 million gallons per day is “incidentally” available for 4.25 days every week, with 200 million gallons per day available for 2.75 days, the average daily “incidental” benefit is more than 308 million gallons per day, which is 68 million gallons per day more than the 1975 “baseline” average and 42 million gallons per day more than the Corps’s 1979 assumptions. The Corps does not explain this large discrepancy.

Using instead the “baseline” average of 230 million gallons per day available incidentally to downstream users, the PAC Report’s reallocations are much greater. To accommodate the projected need of 378 million gallons per day minus the incidentally available 230 million gallons per day would require an average of 148 million gallons per day, or 138,500 acre-feet of storage using the 1734 cfs yield figure. Using the more recent critical yield figure of 1485 cfs, the PAC Report’s reallocations for downstream use is almost 161,800 acre-feet. When added to the acknowledged 141,000 acre-feet necessary to support in-lake withdrawals, the total reallocation requested by the PAC Report is 279,500 acre-feet, or 302,800 acre-feet using current yield figures. The percent of storage reallocated under the PAC Report is 26.6% to almost 28.8% of Lake Lanier’s total conservation storage.

WZhether the Court uses the Corps’s calculations of a 19.7% reallocation or its own calculations, however, is of no moment to the WSA analysis. As the Corps itself acknowledged when sending the PAC Report to Georgia’s Senator Nunn, the reallocations recommended by the PAC Report would require Congressional authorization under the WSA. Letter from Louis J. Martinez, Lt. Col., U.S. Army Corps of Eng’rs, to Sen. Sam Nunn, Ga., at 2 (Dec. 29,1989) (SUPPAR011719). Before the Corps can implement any of the recommendations in the PAC Report, it must secure Congress’s approval to do so.

c. Georgia’s 2000 Water-Supply Request

In May 2000, Georgia Governor Roy E. Barnes sent a formal request to the Corps to allow withdrawals from Lake Lanier of up to 297 million gallons per day by 2030, and to provide sufficient releases from the dam to allow downstream withdrawals of 408 million gallons per day by 2030. Letter from Roy E. Barns, Governor, Ga., to Joseph W. Westphal, Asst. Sec’y of the Army for Civil Works, at 1 (May 16, 2000) (ACF042582). The Corps denied the request, stating that the requested withdrawals would require a reallocation of 370,930 acre-feet of storage, or more than thirty-four percent of the total conservation storage in Lake Lanier. 29 2002 Stock-dale Memorandum at 9 (SUPPAR001050).

Given that the D.C. Circuit in SeFPC determined that a reallocation of twenty-two percent of Lake Lanier’s conservation storage was a major operational change that required Congressional approval, there can be no doubt that Georgia’s request to reallocate thirty-four percent of Lake Lanier’s conservation storage likewise requires Congressional authorization.

3. “Seriously Affect” Project Purposes

The Corps contends that any storage reallocation to accommodate existing water-supply needs will have an insignificant impact on the project’s authorized purposes of hydropower generation and downstream navigation. According to the Corps, the reallocations will cause only a one percent reduction in hydropower generation. (Corps’s Mem. Supp. Mot. Summ. J. at 60.) However, as discussed above, the Corps’s calculations of the storage required to meet current needs are suspect: according to the Corps, existing needs require only 122,714 acre-feet of storage for in-lake withdrawals and no storage for downstream withdrawals, because those withdrawals are within the 327 million gallons per day of alleged incidental benefit from operation of the dam. The Corps uses the wrong baseline, however, assuming not only that 327 million gallons per day are available downstream, but also assuming that the “baseline” for in-lake withdrawals is considerably higher than the 10 million gallons per day allowed by the 1950s contracts.

As noted above in footnote 23, the Corps determined in 1986 that 327 million gallons per day were available incidental to hydro-power generation at Buford Dam. The Corps’s conclusion was not, however, that more water was somehow going through the turbines to allow for the increased downstream withdrawals. Rather, the Corps determined that allowing down stream withdrawals of 327 million gallons per day would not seriously affect the hydropower benefits. In other words, the Corps determined in 1986 that 327 million gallons per day for downstream withdrawals was “the point at which the Lake Lanier project authority ends.” (Corps’s Mem. Supp. Mot. Summ. J. at 33.) Because the Corps has not sufficiently supported its conclusions with respect to the 327 million gallons per day figure, the Court has used an earlier Corps determination that 230 million gallons per day is available as truly incidental to power generation at the dam.

Not only has the Corps failed sufficiently to support the 327 million gallons per day figure, but its incremental increases of the alleged water-supply benefit incidental to hydropower illustrate a fundamental problem with the Corps’s arguments regarding when its authority under the WSA ends. To take the Corps’s arguments to their logical conclusion, the Corps may allow small changes in operations year after year, without seeking any Congressional approval for those changes. Thus, if hydropower is affected only one percent this year, another one percent next year, and so on, the Corps would argue that no Congressional authorization is required. But if the cumulative effect on hydropower throughout the years adds up to twenty percent, then the question becomes at what point Congress must be consulted. As the D.C. Circuit stated, “it is unreasonable to believe that Congress intended to deny the Corps authority to make major operation changes without its assent, yet meant for the Corps to be able to use a loophole to allow these changes” to occur incrementally, rather than all at once. SeFPC, 514 F.3d at 1324-25. The Court must evaluate the cumulative effect of all of the changes in operations at Lake Lanier. In doing so, the Court has determined that 327 million gallons per day are not available as incidental to the operations of Buford Dam as Congress, the Corps, and the hydropower interests envisioned. Rather, as the Corps determined in 1975, 230 million gallons per day are available as a result of the normal operation of the Buford Dam.

In the original Cost Allocation Studies for the Buford project, the Corps computed the available power benefits from Buford Dam as 170,000,000 kilowatt hours (“kwh”), or 170,000 megawatt hours (“mwh”). U.S. Army Corps of Eng’rs, Cost Allocation Studies, Apalachicola, Chattahoochee and Flint Rivers Projects, Basis of All Allocations of Costs for Buford and Jim Woodruff Projects Adopted by the Chief of Engineers 19 (1959) (ACF002101). According to the SeFPC, one way the harm to hydropower can be calculated is by comparing the actual annual generation to the benefits the Corps believed would be available from the project. Only four times since 1994 has the Buford Dam generated 170,000 mwh or more; and in five different years, power generation has fallen below 100,000 mwh. According to the SeFPC, the total value of the loss of hydropower benefits at Buford Dam is more than 60,000 mwh, which is worth $59 million. Now the Corps and the Georgia parties take issue with the SeFPC’s calculation of its damages. However, in the 2002 Stockdale Memorandum the Corps stated that the expected loss of hydropower benefits from the reallocations Georgia requested were more than 95 mwh per day, or a $3 million annual reduction in benefits. 2002 Stockdale Memorandum at 9 (SUPPAR001050).

Another way to look at the harm to hydropower is in the change from peak operations to non-peak operations. From the beginning of the Buford project, the purpose of weekend release was to support water supply. Thus, the generation figures demonstrate that releases were much lower on weekends for the first decades of the project’s operation. See, e.g., Report, 24 Hour — Actual Generation at Buford, 1960 Water Year 3 (SUPPAR026251) (showing weekend generation figures that are hundreds of mwh lower than weekday figures). In 1989, only nine percent of the energy generated by the Buford project was generated on Saturdays and Sundays. By 2007, however, weekend energy generation constituted nineteen percent of the total power generated by the dam. (Ala. & Fla.’s Factual App. ¶ 750 (citing Report, 24 Hour — Actual Generation at Buford 1-51 (SUPPAR026249-99)).) Because non-peak power is much less valuable than peak power, the harm to hydropower from this change in operations is obvious. 30

The SeFPC argues that, if the Court orders the Corps to put in place the “crediting mechanism” described by the Southeastern Federal Power Customers Settlement Agreement, the serious effect on hydropower will be remedied. (E.g., SeFPC’s Resp. Mem. to Mot. for Summ. J. and Opp’n filed by Corps at 1.) It is far from clear that Congress intended that the Corps could sidestep the Congressional-authorization requirement of the WSA by merely paying off the interests seriously affected. Such a remedy is, in the Court’s opinion, for Congress to consider when it evaluates the proposed changes in the project’s operation.

The Corps’s decision to support water supply has seriously affected the purposes for which the Buford project was originally authorized. The Corps is therefore in violation of the WSA.

E. Combined Authorities

The Georgia parties claim that the WSA, 1944 FCA, 1946 RHA, the 1956 statute that allowed the Corps to contract with Gwinnett County for water-supply withdrawals, and the Corps’s contracts with Gainesville and Buford (the “relocation contracts”), taken together, establish that water supply is an authorized purpose of the Buford project. The Court has addressed the legislative history of the Buford project, including the 1946 RHA, the relocation contracts, and the Gwinnett County water-supply request and resulting Congressional enactment. See supra pp. 1310-22. Contrary to the Georgia parties’ argument, taken together the relevant statutes and legislative history point to only one conclusion: water supply, in the form of withdrawals from Lake Lanier and large-scale withdrawals from the Chattahoochee River, was not an authorized purpose of the Buford project. The Georgia parties’ argument that a combination of authorities allows the water-supply withdrawals is without merit.

F. Remaining Claims

The parties claim that the Corps’s operations of the Buford project violate NEPA, the 1944 FCA, the CZMA, and other statutes, and that the various manuals, plans, and other methods through which the Corps operates the Buford project also violate federal law. Because the Court has determined that the Corps must seek Congressional authorization before it can reallocate storage in Lake Lanier to water supply, the parties’ remaining Phase 1 claims regarding the Corps’s operations and the plans for those operations are moot. See Envtl. Def. Fund, Inc. v. Alexander, 467 F.Supp. 885, 888 (N.D.Miss.1979) (noting that, if a project is not legally authorized, “all other issues are mooted until such time as proper authorization may be obtained from Congress”).

G. Operations Going Forward

The Court recognizes that it will take time to secure the required Congressional authorization for the changes to the operation of the Buford project. In addition, the municipal entities that withdraw water from Lake Lanier and the Chattahoochee River cannot suddenly end their reliance on that water merely because a federal court has determined that the Corps failed to comply with its statutory obligations. Thus, the Court will stay Phase 1 of this litigation for three years, to allow the parties to obtain Congress’s approval for the operational changes the water-supply providers request. During the stay, the parties may continue to operate at current water-supply withdrawal levels but should not increase those withdrawals absent the agreement of all other parties to this matter. The Court does not believe that a stay of Phase 2 is warranted at this time, and therefore will consider the Phase 2 claims in accordance with the most recent scheduling order.

At the end of three years, absent Congressional authorization or some other resolution of this dispute, the terms of this Order will take effect. For Atlanta and the communities surrounding Lake Lanier, this means that the operation of Buford Dam will return to the “baseline” operation of the mid-1970s. Thus, the required off-peak flow will be 600 cfs and only Gainesville and Buford will be allowed to withdraw water from the lake. The Court recognizes that this is a draconian result. It is, however,- the only result that recognizes how far the operation of the Buford project has strayed from the original authorization.

As the Court stated at the hearing, the slow pace at which the Corps operates has only served to further complicate and provoke this already complicated and inflammatory case. It is beyond comprehension that the current operating manual for the Buford Dam is more than 50 years old. Certainly, the pendency of this litigation has made the Corps’s completion of plans and manuals more difficult. However, the states and municipalities that rely on the ACF basin for water cannot determine how the operation of the project will affect their interests if they do not understand how the Corps intends to operate the project. The uncertainty created by the Corps’s alarmingly slow pace only adds to the frustration of all parties involved in this litigation. The Court encourages the Corps to complete its plans for the ACF basin as quickly as possible, to allow the parties and Congress to analyze more effectively the future of this vital resource.

The blame for the current situation cannot be placed solely on the Corps’s shoulders, -however. Too often, state, local, and even national government actors do not consider the long-term consequences of their decisions. Local governments allow unchecked growth because it increases tax revenue, but these same governments do not sufficiently plan for the resources such unchecked growth will require. Nor do individual citizens consider frequently enough their consumption of our scarce resources, absent a crisis situation such as that experienced in the ACF basin in the last few years. The problems faced in the ACF basin will continue to be repeated throughout this country, as the population grows and more undeveloped land is developed. Only by cooperating, planning, and conserving can we avoid the situations that gave rise to this litigation.

CONCLUSION

As we all learned in grade school, the separation of powers is fundamental to our federal government: a power reserved to one branch may not be usurped by anoth er. This litigation presents a case study in the need for this tripartite federal system. Congress authorized and paid for the Buford Dam, and gave the Corps authority to operate the dam. Congress specified, however, that the Corps’s authority was not without limits. If the Corps believes that it must operate the project in a manner contrary to Congress’s initial authorization of the project, it must so inform Congress and secure Congress’s permission to do so. Congress has made no exceptions for situations such as the present, when the need for the change is great: the WSA does not provide that “changes shall be made only upon the approval of Congress unless it is inconvenient to do so.” Congress reserved to itself the power to change the purposes for federal projects such as the Buford Dam project. The executive branch simply may not circumvent that authority. Congressional approval of the reallocation of storage in Lake Lanier is required.

The Court is sympathetic to the plight of the Corps, which is faced with competing and legitimate claims to a finite resource. Neither the Corps nor the Court can make more water. However, as the D.C. Circuit remarked, “Congress envisioned that changed circumstances or ‘difficult situations’ might arise and specified that any solution involving ‘major operational ... changes’ required its prior authorization.” SeFPC, 514 F.3d at 1325 (citations omitted). The Corps’s failure to seek Congressional authorization for the changes it has wrought in the operation of Buford Dam and Lake Lanier is an abuse of discretion and contrary to the clear intent of the Water Supply Act. As such, the Corps’s actions must be set aside.

Accordingly, IT IS HEREBY ORDERED that:

1.Alabama and Florida’s Motion for Summary Judgment (Docket No. 191) is GRANTED in part and DENIED in part;

2. The Georgia parties’ Motion for Summary Judgment (Docket No. 195) is DENIED;

3. The SeFPC’s Motion for Summary Judgment (Docket No. 238 in Civ. No. 3:08-640) is GRANTED in part and DENIED in part;

4. The Corps’s Motion for Summary Judgment (Docket No. 227) is DENIED;

5. APC’s Motion for Summary Judgment (Docket No. 86 in Civ. No. 3:07-249) is DENIED;

6. Columbus and Columbus Water Works’ Motion for Partial Summary Judgment (Docket No. 22 in Civ. No. 3:07-1033) is DENIED;

7. Apalachicola’s Motion for Summary Judgment (Docket No. 190) is GRANTED in part and DENIED in part; and

8. The claims raised in Phase 1 of this litigation are hereby STAYED for a period of three (3) years.

1 . The Court will refer to Atlanta, Columbus, Gainesville, Gwinnett County, DeKalb County, Fulton County, the ARC, the Cobb County-Marietta Water Authority, and the Lake Lanier Association collectively as “the Georgia parties.”

2 . The Court will refer to the Corps and the Corps's officers collectively as the “Federal Defendants.” The United States Fish and Wildlife Service ("USFWS”) and a USFWS official also are defendants in one of the member cases (3:07-250), but the claims against USFWS are not at issue in this phase of the litigation.

3 . Congressional inquiries into the uses for the Apalachicola and Chattahoochee Rivers began even earlier than 1925, but most of these inquiries sought only to examine the rivers’ usefulness for navigation. See George W. Sherk, Buford Dam and Lake Lanier: Statutory Perspectives and Limitations 11-34 (2000) [hereinafter “Sherk, Buford Dam ’’].

4 . The voluminous administrative record in this matter is divided into the original record and the supplemental record. The Corps has consecutively stamped each portion of the record, with the original record bearing the prefix “ACF” and the supplemental record having the prefix "SUPPAR.” When possible, the Court will endeavor to cite not only to the document itself, but to its place in the administrative record.

5 . The Newman Report also refers to the Lanier site as the Buford site. See, e.g., Newman Report ¶ 69, at 27 (ACF000661).

6 . The City of Buford executed a similar contract because its waterworks facilities were also inundated by the waters of Lake Lanier. See Contract between U.S. Army Corps of Eng'rs & City of Buford, Ga. (Dec. 19, 1955) (ACF014450-56) (allowing Buford to withdraw two million gallons per day from the reservoir).

7 . Appendix A to the 1958 Manual is the regulation manual for the Jim Woodruff Reservoir on the Apalachicola River. (ACF001722.)

8 . Million gallons per day is often abbreviated "mgd" or "MGD.”

9 . The MAAWRMS characterized recreation as an "authorized project purpose under general legislation, the 1944 Flood Control Act and the Federal Water Project Recreation Act of 1965.” MAAWRMS at 11-40 (SUPPAR002020). Because of the Court’s resolu tion of the water-supply issue, it is not necessary to reach the issue of whether recreation is indeed an authorized project purpose.

10 . Although the analysis was "based on the assumption that [water-supply] withdrawals are part of a base condition,” the MAAWRMS also assumed that “a separate contract would be entered into for such demands.” Id. app. C, at C-58 (SUPPAR002643).

11 . The MAAWRMS does not explain why the second alternative plan assumes that approximately 10,000 acre-feet at Lake Lanier were allocated to water-supply storage in 1980, but the third alternative assumes that Lake Lanier had no storage allocated to water supply in 1980.

12 . Others also recognized that water supply was "not specifically authorized as a purpose” of the Buford project. Id. at 3251 (SUPPAR001491) (Letter from W.T. Bush, co-chairman, Gwinnett County Water & Sewerage Auth. to Sen. Sam Nunn, Ga. (Aug. 21, 1980)).

13 . In the design memorandum Congress required when it authorized the reregulation dam, the Corps determined that the most economical alternative was no longer a reregulation dam, but was instead reallocation of storage at Lake Lanier. Memorandum from Ralph V. Locurcio, Colonel, U.S. Army Corps of Eng’rs, to Commander, S. Atl. Div. (Oct. 13, 1988) (SUPPAR035867). This memorandum recommended that the Corps prepare “a Post-Authorization Change Report recommending reallocation of storage in Lake Lanier ... for submittal to Congress for authorization.” Id. ¶ 3.

14 . One million gallons per day is equal to 1.547 cfs of flow. See MAAWRMS at III-6 (SUPPAR002078). According to the Corps, during normal operations, 1600 cfs equal one acre-foot of storage. The conservation storage of Lake Lanier is 1,049,400 acre-feet. Therefore, the total storage used between Gainesville and Buford is 10,146 acre-feet. (10 mgd x 1.547 cfs 4- 1600 = .00966875. And .00966875 x 1,049,400 = 10,146 acre-feet.) Under the conditions present in the 1986-1988 drought, during which 1485 cfs equaled one acre-foot, the storage necessary for 10 million gallons per day would be 10,-932 acre-feet. The WCP uses a different figure for storage in which 1734 cfs equals one acre-foot. See WCP app. C, at C-4 (ACF041302). Using this figure, 10 million gallons per day is equal to 9,362 acre-feet.

15 . 11,200 acre-feet of storage would provide approximately 10.2 million gallons per day of water, using 1485 cfs as the yield figure. Assuming 1734 as the yield, 11,200 acre-feet provides almost 12 million gallons per day.

16 . Originally, the lawsuit challenged not only the Corps's operations in the ACF basin, but also the operations in the Alabama-Coosa-Tallapoose ("ACT”) basin. The claims involving the ACT basin are no longer part of this case.

17 . According to the Court of Appeals, the MOA required the Corps to abandon the PAC Report. Alabama, 424 F.3d at 1123.

18 . Peter Geren was the Secretary of the Army at the time the D.C. Circuit considered the SeFPC case. He was preceded in that position by Louis Caldera.

19 . None of the parties addresses any statute of limitations issues in their extensive briefing on these Motions, although some of the contracts Alabama and Florida challenge were first executed in the 1970s. However, due to the "renewing” nature of the contracts and the PAC Report’s acknowledgment in 1989 that the Corps was attempting to create a new scheme for the allocation of storage in Lake Lanier, the Court would find that Alabama and Florida’s claims are within the statute of limitations in any event.

20 . The Corps has not challenged Alabama and Florida's standing.

21 . ARC’s contract with the Corps provided for 50 million gallons per day in addition to what the Corps considered the "incidental benefit” from releases for power of 327 million gallons per day. As discussed infra, the Corps's conclusion that 327 million gallons per day is available incidentally to power operations is not supported by the record.

22 . The Georgia parties argue that the Court must take into account return flows, which are water the municipal entities return to the lake and the river in the form of highly treated wastewater. According to the Georgia parties, "[o]mitting return flows is a major omission, and error, because storage utilization is a function of net, and not gross, withdrawals of water.” (Ga.’s Mem. in Opp'n to SeFPC's Mot. at 41.) However, none of the municipal entities is required to return any water to Lake Lanier or the Chattahoochee River, but the Corps is required by the various water-supply contracts to allow the entities to withdraw a certain amount of water from the lake and river. The Court must evaluate the Corps’s obligations, independent of any voluntary return flows, because regardless of the return flows the Corps's obligations remain the same.

23 . In 1986, the Corps revised this number to the 327 million gallons per day figure it uses today. There is no explanation in the record as to how the incidental benefits of regular power operations at the dam would increase from 230 million gallons per day in 1975 to 327 million gallons per day in 1986. Even the Corps appears to recognize that 327 million gallons per day is at best an estimate, stating that it “expects that further analysis ... would validate the Corps' 1986 determination that up to 327 mgd could be provided on an annual average basis from flows that occur incidentally as a result of project operations.” Id. at 47 n. 34. The Court must rely on the data that is supported by the record, however, not data that the Corps expects, at some point in the future, to be borne out by “further analysis.”

24 . Adding the 11,200 acre-feet Congress allocated to Gwinnett County in 1956 increases the 1975 baseline to 235,900 acre-feet, and the 1979 baseline to 269,600 acre-feet. As noted above, however, Gwinnett County did not begin to withdraw water from Lake Lanier until sometime in the 1970s, pursuant to contracts that did not purport to be based on the 1956 legislation and which allowed far greater withdrawals than Congress envisioned. The inclusion of Gwinnett County's original authorization does not, however, significantly change any of the Court's calculations. Moreover, Gwinnett County's Congressionally authorized use of 11,200 acre-feet of storage expired in 2006.

25 . The Corps recognized that the 1986-1988 drought would likely result in a lower "firm yield” than the 1939 drought, and estimated that the new yield figure would be 1455 cfs. Id. Actual yield from the 1986 drought has been set at 1485 cfs. See supra n. 14.

26 . The Court’s own calculation of the storage required for 151 million gallons per day yields a slightly different figure of 141,370 acre-feet.

27 . The usual storage calculation would have been 178 mgd x 1.547 cfs/1734 x 1,049,400. Under this formula, 178 million gallons per day requires 166,600 acre-feet of storage.

28 .The Corps nowhere defines this term.

29 . The 2002 Stockdale Memorandum stated that the total conservation storage in Lake Lanier is 1,087,600 acre-feet. 2002 Stockdale Memorandum at 8 (SUPPAR001049). The Corps uses this figure throughout its briefing on the instant Motions. From the time of Buford Dam's construction, the Corps has calculated the conservation storage as 1,049,-400 acre-feet. It appears that the 1,087,600 acre-feet figure is in fact a seasonal variation — during the summer months the Corps increases the conservation pool from elevation 1070 to 1071. See Apalachicola Basin Reservoir Regulation Manual app. B, at B41(ACF018475). Because the larger storage amount is a short-term variation from the usual conservation storage figure, the Court has used the well-documented, historical storage amount in its calculations. However, the use of the larger storage amount would not significantly change the calculations.

30 . That hydropower has been harmed is relevant in determining whether the Corps’s operation of the project to support water supply has seriously affected the Congressionally authorized purposes. However, this does not mean that the SeFPC has any monetary claim for lost hydropower benefits. As the Court has made clear in previous rulings, the Court will not consider arguments regarding remedies at this time.