James River Flood Control Ass'n v. Watt

MEMORANDUM OPINION DONALD J. PORTER, District Judge. CASE SUMMARY Plaintiffs seek injunctive and declarative relief, claiming violations of the National Environmental Policy Act, 42 U.S.C. § 4321 et seq. (NEPA) and that defendants are constructing a major irrigation project without authorization from Congress. The Court finds the challenged environmental statements to be adequate, and that defendants are still proceeding within the scope of their authorizing legislation. Judgment is therefore entered for defendants. FACTUAL BACKGROUND The earliest version of the Garrison Diversion Unit (GDU) was originally authorized by Congress in the Flood Control Act of December 22, 1944 (58 Stat. 887, 891). Investigations conducted subsequent to 1944 revealed that major portions of the proposed project were impractical, and the project plans underwent substantial revision during the 1950’s. The final report of these revisions was embodied, with various supplements, in House Document 325, 86th Congress, Second Session. Congress, in the Act of August 5, 1965, 89th Congress, First *1287 Session (79 Stat. 433), authorized the GDU “substantially in accordance with the plans set out in the Bureau of Reclamation Report dated November 1962 (revised February 1965), supplemental report to said H.D. 325.” As enacted in 1965, the GDU was projected to ultimately irrigate 1,007,000 acres of land in North Dakota, as well as provide water for municipal, industrial, wildlife, and recreational uses throughout that state. The immediate authorization, though, was for the construction of an Initial Stage 250,-000-acre unit. This Initial Stage development was intended to draw water from the Missouri River in North Dakota at Lake Audubon, a subimpoundment of Lake Sakakawea, transfer this water by way of the McClusky Canal to Lonetree Reservoir, and from there divert this water to several different sections of northern, east-central, and south-eastern North Dakota. Under the 1965 plans, most of the return flows from the irrigated acreage would enter rivers in the Hudson Bay drainage basin, and eventually be carried across the border into Canada. The return flows from the LaMoure-Oakes section of the GDU in southeastern North Dakota, however, would enter the James River, in the Missouri River drainage basin, and be taken down that river into South Dakota. Construction of certain parts of the GDU, particularly the Snake Creek Pumping Plant and the McClusky Canal, was commenced in the late 1960’s and early 1970’s. Although the project was therefore underway prior to the enactment of the National Environmental Policy Act of 1969 (NEPA), the Department of the Interior undertook to prepare environmental impact statements for the project in compliance with NEPA. A final environmental statement on the Initial Stage of the GDU was filed in 1974 (this statement will be hereafter cited as the 1974 FES), and was intended to cover in detail the Principal Supply Works (including the Snake Creek Pumping Plant and the McClusky Canal) as well as the cumulative overall impacts of the 250,000-acre project. At that time, defendants intended to prepare individual environmental statements for three of the particular irrigation areas in the GDU, and a LaMoure and Oakes Section, Draft Environment Statement was completed in 1976 (hereafter cited as 1976 DES) and made available to the public. A final version of the LaMoure and Oakes Section statement was prepared, but never filed. The National Audubon Society 1 filed suit in the District of Columbia in 1976, to enjoin further construction of the GDU, raising issues relative to the 1974 FES, the Fish and Wildlife Coordination Act and the Migratory Bird Treaty Act. 2 In 1977, the United States Department of the Interior stipulated with the Society that the Department would prepare a comprehensive supplementary environmental statement, and the final version of this statement was completed in 1979 (hereafter cited as the 1979 FSES). The FSES makes the statement that it “supplements” the 1974 FES and 1976 DES, “and is intended to substitute for the three planned individual service and environmental statements.” The Canadian Government also took issue with the GDU, and expressed its concerns in a note to the United States in 1969. In the Canadian view, those return flows from the GDU that would ultimately pass into Canada would cause injury to health and property in Canada in contravention of the Boundary Waters Treaty of 1909. In 1974, the United States acknowledged its obligation under the Treaty and stated that no construction on the GDU affecting Canada would be undertaken until it was clear that the obligation would be met. The governments of Canada and the United States referred the matter of GDU impacts on Canada to the International Joint Commission (IJC) in 1975. In 1976, the IJC recommended, inter alia, that a *1288 prototype irrigation test area be built to measure the quality and quantity of return flows. The House of Representatives Committee on Appropriation Report No. 95-1247, June 1, 1978 (accompanying H.R. 12928), stated that the Committee “expected” the Secretary of the Interior to carry out this recommendation for a test, and that construction should proceed of “irrigation and drainage facilities to serve an area not to exceed 5,000 acres in either the LaMoure or West Oakes irrigation area.” In 1979, it was determined that a 5,000-acre area in West Oakes was representative of the lands to be irrigated in the GDU, and the decision was made to use that area for the test site. This West Oakes test site is adjacent to the James River near Oakes, North Dakota, and is to be served by the Oakes Pumping Plant. The water supply will be drawn from the James River, utilizing releases from the Jamestown Reservoir as needed, and the return flows from irrigation will reenter the James River and pass into South Dakota. It is the development of this test site in particular that plaintiffs here seek to block. Plaintiff James River Flood Control Association is composed of individuals who own land along the James River in the northern South Dakota counties of Brown and Spink. Individual plaintiffs are members of the Association. Plaintiffs’ basic contention in their original complaint is that the 1974 FES and the 1979 FSES do not adequately describe or disclose the impacts of the GDU on South Dakota and therefore are in violation of NEPA. Plaintiffs, in an amendment allowed by this Court on May 7, 1982, also allege that, as a result of the discussions with the Canadian Government, the United States Department of Interior has agreed to proceed with a modified GDU plan (proposed by officials of the state of North Dakota) to reroute certain return flows that would otherwise reach Canada into the James River and thus into South Dakota. This modified plan, plaintiffs contend, is not “substantially in accordance” with the plans set forth in House Document 325, and is beyond the scope of the 1965 legislation that authorized the GDU.- DISCUSSION I. SCOPE OF REVIEW The adequacy of environmental statements are “subject to limited judicial review in the federal courts.” Farmland Preservation Association v. Goldschmidt, 611 F.2d 233, 237 (8th Cir.1979). This review involves two functions. First, the court “must determine whether the challenged [statement] is adequate as a statement to satisfy the [procedural] requirements of NEPA.” Id. The test to be employed in determining compliance with these procedural provisions ... is one of good faith objectivity.... The touchstone of [the] inquiry is reason.... While the detailed statement must of course be more than a pro forma ritual ... the discussion of environmental effects and alternative courses of action need not be exhaustive.... The [statement] need contain only sufficient information to permit a reasoned choice of alternatives .... The purpose of NEPA is not to require an objection free document, but rather to give Congress, the responsible agencies, and the public a useful decision-making tool. Minnesota Public Interest Research Group (MPIRG) v. Butz, 541 F.2d 1292, 1300 (8th Cir.1976). See State of Missouri ex rel. Ashcroft v. Department of the Army, 526 F.Supp. 660 (W.D.Mo.1980). Although the duties imposed by NEPA on federal agencies are “ ‘essentially procedural’ ”, Strycker’s Bay Neighborhood Council v. Karlen, 444 U.S. 223, 227, 100 S.Ct. 497, 499, 62 L.Ed.2d 433 (1980), a district court must, once it has satisfied itself that the procedural requirements of NEPA have been met, “go further and engage in a limited substantive review of agency action to determine whether it passes muster under the scope of review provisions of the Administrative Procedure Act.” Farmland Preservation Association, 611 F.2d at 237. The Court must “determine *1289 whether the agency reached its decision after a full, good faith consideration and balancing of environmental factors ... then ... whether the agency’s actual balance of costs and benefits was arbitrary or clearly gave insufficient weight to environmental values.” MPIRG v. Butz, 541 at 1300. “However, it is not the function of the federal court ultimately to pass on the merits of the contemplated action or to substitute its judgment for that of the agency.” Farmland Preservation Association, 611 F.2d at 237. Before addressing the question of whether the 1974 and 1979 statements prepared by defendants are adequate, however, the Court must consider whether the 1976 DES on the LaMoure and Oakes Section can be consulted in this regard. Plaintiffs have contended that the 1976 DES is invalid, since the final version was never filed, citing Appalachian Mountain Club v. Brinegar, 394 F.Supp. 105, 122 (D.N.H.1975) (“Supplemental information, which has not been processed in the same manner as a draft EIS, cannot resurrect a deficient impact statement.”) Yet it is clear that in determining adequacy, this Court is not limited to an examination of the statement in question and nothing else — “the candidate environmental statement, the draft environmental statement, the background studies and the hearing transcript and attached affidavits” are all to be considered. Jackson County, Mo. v. Jones, 571 F.2d 1004, 1008 (8th Cir.1978). In MPIRG v. Butz, supra, for example, the court included in its inquiry the environmental assessment reports (EAR’s) that had been prepared by the agency before the filing of the environmental statement. The court noted that the statement referred “to the EARs which were on file with the [agency] during the publication process,” 541 F.2d at 1036, even though these EARs had evidently not gone through the rigorous process given the environmental statement. The court concluded that “the EIS and EARS were prepared ... with good faith objectivity. They provide enough specific information to permit a reasoned choice of alternatives. They are, therefore, in compliance with the ‘detailed statement’ requirement of NEPA.” Id. This Court concludes, that the 1976 DES can be considered on the question of whether defendants have complied with the requirements of NEPA, particularly since the 1979 statement makes it clear that it was intended to act as a “supplement” to the 1976 DES, and, presumably, anyone who so desired could have commented on any inadequacy in the 1976 statement while the 1979 statement was being processed. II. ADEQUACY OF CHALLENGED IMPACT STATEMENTS A. GDU Project Features. The pertinent GDU project features under the 250,-000 acre initial stage authorized by Congress in 1965 [Plan 1], as described in the environmental statements commence with the pumping of water from reservoirs on the Missouri River, the conveyance of this water by means of the McClusky Canal to Lonetree Reservoir and from Lonetree Reservoir by way of the New Rockford Canal to the James River Feeder Canal. GDU project water will be discharged into the James River itself by means of the feeder canal near New Rockford, North Dakota. The amount of project water entering the James River at this point will average 200 c.f.s. (cubic feet per second), but may reach as high as 450 c.f.s. at peak demand periods. The GDU project water will then travel south for about seventy miles through the upper James River, passing through the Arrow wood National Wildlife Refuge, and will enter Jamestown Reservoir, and be stored. Upon a request by the users of GDU project water, the water will be released up to a maximum of 450 c.f.s., into the lower James River (defined in the environmental statements as the 120 river miles below Jamestown Reservoir to the Oakes Pumping Plant, not far north of the boundary between North and South Dakota). The project water would be pumped from the James River both at LaMoure, North Dakota (where 13,350 acres will be irrigated) and further south near Oakes, North Dakota. The water removed from the *1290 James River at Oakes will be used to irrigate 19,660 acres in the West Oakes area near the James River, and will also be pumped into Taayer Reservoir for storage for both the West and the East Oakes (26,-320 acres) areas. Some of the water used for irrigation will not be consumed, and this water, described under the general term of return flows, will drain off the land and back into the James River. In the LaMoure area, for example, Figure 25 in volume two of the 1979 FSES shows a mean annual diversion of water of 9.500 acre-feet for irrigation purposes, with 6.700 acre-feet of return flow re-entering the James River. Some of this return flow, after moving south in the James River, will be again pumped out of the River at Oakes and re-used for irrigation. As Figure 25 shows, there will be a total diversion of 38.500 acre-feet at the Oakes area, but only 1.700 acre-feet of return flow will drain back into the James River. 3 The return flows will naturally follow the course of the James River into South Dakota, and thus arises the source of the plaintiffs’ concerns. B. Plaintiffs’ Contentions. An initial objection that plaintiffs raised, in a pre-trial brief, to the environmental statements is the fact that the statements devote considerable length to the modifications on the James River in North Dakota that will be necessary under the GDU plans and the environmental impacts of the GDU in that state, yet the statements make no reference to any needed river modifications in South Dakota, and contain considerably shorter discussions on environmental impacts to South Dakota. The River does not end at the border between the two states and so, plaintiffs seem to argue, if many pages are necessary to analyze the project in North Dakota, just as many pages should be necessary to describe project impacts on the River in South Dakota. But, as Farmland Preservation Association, supra, recognized, upholding the adequacy of a section of a statement that was only two paragraphs in length, “the paginal length of a ... discussion is not in itself controlling.” 611 F.2d at 239. Plaintiffs’ argument overlooks a crucial difference between the use of the River in the GDU in North Dakota and its use in South Dakota. In North Dakota, from near New Rockford, the point where GDU water is discharged by the James River Feeder Canal into the James River, to near Oakes, where the last GDU pumping station is located, the James River becomes a sort of natural canal for the irrigation water. Below Oakes and into northern South Dakota, the James River ceases to perform this “canal” function, and to the extent the River in South Dakota is part of the GDU at all, the River is merely a drain for return flows. From New Rockford to Oakes, the James River, which currently has flows of about 50 c.f.s., must be modified to ensure its capacity for carrying the maximum 450 c.f.s. flows that are planned with the GDU, and it would accordingly be necessary to describe these modifications in some detail in the environmental statements. Under the GDU plans as set forth in the environmental statements, however, these quite large additional flows in the James River go no further than Oakes, and there is in fact a net decrease of 840 acre-feet in the annual flows reaching South Dakota. Thus, the need for enlarging James River channel capacity in North Dakota is absent in South Dakota, and absent too is the need to devote space in the environmental statements to any such modifications. Because the River flows in South Dakota will be significantly less than in North Dakota, it would initially appear that the space devoted to environmental impacts in South Dakota could also be less than the space devoted to impacts in North Dakota. To determine if the GDU’s impacts on South Dakota are indeed so minimal that they can properly be analyzed in only a few pages, the Court will now turn to the more specific environmental objections asserted by plaintiffs. Plaintiffs have argued that the environmental statements fail to address potential *1291 impacts to the James River in South Dakota as a result of the GDU in several different respects: increased flooding, a raising of the ground water table in land adjacent to the River, salinity problems in soil adjacent to the River and salinity changes in the water of the River itself, and changes in the management of the Sand Lake Wildlife Refuge in northern South Dakota. 4 The flooding argument has its origin in the rather unusual character of the James River as it enters northern South Dakota. 5 The River in North Dakota has a slope of as much as four or five feet per mile, but as it enters South Dakota, the slope is reduced to only inches per mile. In North Dakota, the River is capable of flows of 450 c.f.s. or more, but in northern South Dakota, the channel capacity is reduced to 200 c.f.s., with channel capacity increasing again as the river passes Redfield, South Dakota. These conditions have historically rendered the River particularly prone to flooding in northern South Dakota. It is plaintiffs’ contention that even if there is a net annual decrease in River flows into South Dakota, the operation of the GDU will hold back peak River flows, reducing the size of, but not eliminating, flooding in spring and summer, only to release more water later in the summer and cause a prolonged duration of flooding. Further, plaintiffs contend, Sand Lake will have a tendency to hold some of this water back in the spring and release it over a longer period of time. Plaintiffs also assert that in the late fall and during the winter, when the James River historically has a very low or nonexistent flow, the River channel serves as a drain for ground water in adjacent land. With the GDU, however, the James River will have higher flows during the fall and winter, and this, plaintiffs argue, will block the drainage and cause the ground water table to rise so high (an effect plaintiffs call “seepage”) that adjacent land will be impossible to farm. Further, plaintiffs claim, the raised ground water level will carry with it dissolved salts into the root zone and harm crops. Plaintiffs also claim that the environmental statements inadequately address the impact the GDU will have on the quality of the water flowing in the River into northern South Dakota. 6 Plaintiffs’ expert conceded at trial that the projected increase in salinity in the James River water at the North Dakota-South Dakota border was “relatively insignificant”, but asserted that the salinity levels will be compounded just within South Dakota by the operation of Sand Lake. Plaintiffs argue that during the summer, the heavy plant growth at Sand Lake will cause the loss of significant quantities of water by means of transportation through leaves, leaving salts behind in increased concentrations. Likewise, in the winter, plaintiffs claim, the water in Sand Lake and throughout the River will freeze pure, concentrating the salts under the ice in the flowing water. Plaintiffs also complain of other detriments in water quality for James River users in South Dakota, such as increased hardness, possible laxative effects, and increased salinity for irrigators. Finally, plaintiffs allege that nutrients in the return flows from North Dakota could stimulate algae growth in Sand Lake, altering the chemical balance of the water and possibly becoming toxic to wildlife. C. The Statements. It is defendants’ position that the environmental statements clearly show that the GDU, as described in *1292 those statements, will have no significant impact on South Dakota, primarily because of the extreme variability of the James River in its existing setting, including the effect that Sand Lake has on that setting. This view has prevailed throughout all the environmental statements prepared by the defendants on the GDU, commencing with the 1974 FES. The 1974 FES is the most general of the statements, at least insofar as the James River is concerned, and the data it utilizes was revised later in the 1970’s by minor changes in GDU plans and information gathered by more advanced study methods. The principal references to the River in the 1974 FES are to the entire River system, describing the usual flows, chemical quality, and major physical features of the River. The FES states that return flows to the River from the GDU will average an annual 3,600 acre-feet, an increase of about 5% and amounting to an increase in the monthly average flow by about 1 c.f.s. to 15 c.f.s. The FES also states that water quality will be somewhat benefited by the GDU, mainly because the increase in water flowing in the River will dilute pollutants, although TDS will be increased by about 10%. On the subject of flooding along the James River, the FES states that water for the LaMoure and Oakes areas in the early irrigation season will be drawn from spring runoff naturally in the River to provide a small reduction of the flood wave. The potential for seasonal floods resulting from rainstorms during the irrigation season would be increased a small amount by irrigation releases from Jamestown Reservoir. Problems of this type would occur when a rainstorm below the dam produced enough runoff to cause flooding when combined with irrigation water in the river. Although releases would be suspended as soon as flooding was evident, the travel time necessary for water to reach the Oakes Pumping Plant would not allow sufficient control to completely alleviate the solution. 7 Occurrences of this nature would be very infrequent. The 1976 DES contained a much more detailed description of'the- James River, including several pages on South Dakota. The DES makes observations about the River’s “extremely flat slope” in the northern portions of South Dakota, and describes the course of the channel and the adjoining landscape. The frequency and extent of flooding is noted, as well as the tendency of releases from Pipestem and Jamestown Reservoirs to prolong flooding while reducing the magnitude of floods. The DES states that the natural flow of the James River “is irregular and varies considerably in quantity and quality.” The average discharge at Columbia, South Dakota from 1945 to 1973 was 103 c.f.s., reaching a maximum of 5,420 c.f.s. in 1950, and annual flows varying from 884 acre-feet in 1961 to 379,600 acre-feet in 1950. As for chemical quality, the DES said it is “relatively good but fluctuates widely depending upon the flow in the river”; TDS concentrations of up to 1,400 mg/1 being present in low flow periods, and as low as 200 mg/1 during high flows. The DES also devoted one and one-half pages to a description of Sand Lake and its vegetation, waterfowl, fish and other wildlife. The 1976 DES concludes that the GDU “will not significantly increase flows or alter water quality in the James River” as it enters South Dakota. The DES, like the 1974 FES, uses the calculation that there will be an average increase of 3,600 acre-feet to the James River in South Dakota, including periods when high natural flows are diverted for irrigation, and periods of low natural flows when additional project water is discharged from the GDU. Although the DES stated that median water quality at the North Dakota-South Dakota border would be somewhat degraded, the average level would still be an improvement over current levels “now experienced dur *1293 ing summer and fall low-flow periods and the change should have no appreciable effect on the aquatic ecosystem or man’s use of the water.” As for Sand Lake, the DES said that pool levels will not be appreciably higher and the volume of water will not have a significant impact on refuge management. The slightly higher average level of total dissolved solids could allow more salt tolerant plant species ... to become more dominant in refuge waters. The value of the refuge to waterfowl and other wildlife will not be diminished. In other sections of the DES, more detail is presented on analysis of streamflows and monthly water quality parameters, and the DES observes that flooding in South Dakota will be substantially reduced in some years, that water quality will be stabilized, if slightly degraded, but that the water “will still be suitable for domestic uses without treatment for increased salinity.” The DES states that the GDU will decrease peak nitrate concentrations slightly in the James River, although the increase in nutrients in the River might increase the frequency and/or duration of algal blooms (growth) in areas downstream of the GDU, but that “coliform counts will be reduced and algae buildup and associated effects will tend to be dissipated.” Finally, the DES repeats the 1974 FES statements about the infrequency of possible aggravated flooding in periods of rainstorms during the irrigation season. In preparing the 1979 FSES, a reduction was made in the amount of water being discharged into the James River by the GDU (certain wildlife areas no longer taking delivery of project water); the annual James River flows reaching South Dakota would now be an average of 840 acre-feet less than historic flows, rather than the annual 3,600 acre-feet increase projected in the 1974 and 1976 statements. Defendants also used a more sophisticated method of analysis than before. A 21-year (1953-1973) historic record of the James River was replicated three times to predict water quantity and quality changes over a 63-year period. This method, which the evidence at trial indicated was the state-of-the-art method available at the time of the studies, first presented the first 21 years as the historic period, then proceeded from the concept that as irrigation development proceeds, water quality and quantity changes will occur in the return flows. Irrigated acreage will increase over the period of construction, and there will be a natural initial leaching of salts from the irrigated soils. Thus, the salinity of the return flows will gradually increase and peak in about 20 years from the beginning of irrigation, and then decrease and reach equilibrium in the last 20 years of the 63-year period. 8 The results of the FSES study on the James River at the North Dakota-South Dakota border indicated that median monthly flows would increase during eight months, providing water in low-flow periods in the late summer, fall and winter. Figure 22 of the FSES shows that the median flows will range from 40 c.f.s. to 60 c.f.s. in the months of September to January to 30 c.f.s. in February. Monthly median flows would decrease in the normally high flow months of April through July, ranging from between 20 c.f.s. in April, May, and July, up to 60 c.f.s. in June. 9 The FSES states that The primary impact from project return flows ... might be to extend the duration of spring floods by a short time... Any incremental flood damage which might be attributable to return flows would generally be unidentifia *1294 ble.... In the smaller channels of the streams of the study area affected by return flows, flooding might be increased significantly. Median monthly TDS concentrations would increase during all months, with the largest increases occurring from November through February, “when natural flows are at their lowest and return flows will exert the greatest influence on the stream.” The FSES stated that sulfate and hardness concentrations would increase in all months. According to the FSES, the projected “levels of salinity are commonly found in water used by man in the area; these levels are not expected to affect existing uses.” Also, in its analysis of GDU effects on water in the James River, the FSES recognizes that the actual downstream impacts would be affected considerably by the operation of Sand Lake. The potential river operations possible are being studied. The low river gradient of the James River and the current inability to alter floodflows in South Dakota by operation of North Dakota facilities points out the potential natural buffer involved. The FSES contains various other discussions of projected changes due to the GDU or nitrate-nitrogen concentrations, phosphates, pesticide levels, dissolved oxygen, stream temperature, sulfate and hardness; all findings supported the FSES conclusions that: Return flows ... would cause overall increases in median concentrations of TDS... . Chemical water quality would be stabilized and severe stagnation eliminated from several reaches ... which now flow only intermittently. Return flow accruals might reduce water quality parameter concentrations during historically high periods and increase these concentrations during historically low periods. These changes would not affect overall usability ... significant increases in sulfate concentrations and hardness would be the primary quality impact.... An increase in cost to treat this water for domestic and municipal service would result. Changes would occur in levels of herbicides, insecticides, dissolved oxygen, temperature, phosphate, or biochemical oxygen demand but are not expected to be widespread or major in regard to man’s use. Minor increases in depth and duration of flooding in downstream reaches ... would occur. 10 D. Procedural NEPA Compliance. Having briefly summarized the relevant portions of the GDU environmental statements, the Court now comes to the ultimate question of whether these statements, taken together, are sufficient under the requirements of NEPA. The “burden is upon the plaintiffs to establish by a preponderance of the evidence that the [statement] was inadequate.” Sierra Club v. Froehlke, 534 F.2d 1289, 1300 (8th Cir.1976). Giving full consideration to all the evidence presented, the Court cannot find that plaintiffs have upheld this burden. Plaintiffs’ “contentions boil down to questioning the degree of detail rather than the lack of it... [w]hile ... additional facts may have been useful ... [this Court must] conclude that the detail presented was sufficient to uphold the [statement].” Sierra Club v. Morton, 510 F.2d 813, 825 (5th Cir.1975). Plaintiffs have presented nothing to cast doubt upon defendants’ conclusion that the average annual change in water quantity in the James River in South Dakota due to the GDU is so small as to be virtually unmeasurable. The environmental statements acknowledge the altered nature of the River in South Dakota, and that the GDU will produce minor “increases in depth and duration of flooding.” The statements also all indicate that there will be some adverse impacts on water quality, such as increases in hardness and salinity, yet these increases will constitute little significant change from the quality of the James River as it is now. The most substantial allegation plaintiffs *1295 raise is defendants’ supposed failure to consider how Sand Lake might compound environmental impacts. Plaintiffs’ expert conceded, however, that there has never been a comprehensive scientific analysis of the hydrology of Sand Lake. The 1979 PSES recognized, in the section cited earlier, that Sand Lake does have substantial impacts on the James River, and that studies are being conducted. A similar problem was present in Environmental Defense Fund, Inc. v. Corps of Engineers, 325 F.Supp. 749 (E.D. Ark.1971)., aff’d. 470 F.2d 289 (8th Cir.1972), where there was a demonstrated need for studies on stream life in a dam project area, but the Court found that it was sufficient if the environmental statement simply indicated that there was such a need. “The decision makers can then determine whether to proceed without such a study or to postpone the project while such a study is being undertaken.” 325 F.Supp. at 760. See also Sierra Club v. Froehlke, supra, 534 F.2d at 1296. The statements point out the likelihood of algae growth that plaintiffs raise as a possible impact to Sand Lake, stating that its effects will tend to be dissipated. It is true that the statements do not conduct a study of GDU-caused changes in the chemical balance of Sand Lake, but plaintiffs’ expert was himself doubtful that such a study could be conducted. It is also true that the statements do not address the possibility that increased River flows might adversely impact the ground water table, and thus cause salinity and “seepage” problems in the soil adjacent to the River in South Dakota, but in view of the small differences that the project will apparently cause in water quantity in the River (well within the channel capacity of the James River in South Dakota and in fact amounting to a small yearly decrease in quantity), it hardly seems unreasonable on the part of defendants to fail to investigate this possibility. “An impact statement must be particularly thorough when the environmental consequences of federal action are great.... However, an impact statement need not discuss remote and highly speculative consequences.” Warm Springs Dam Task Force v. Gribble, 621 F.2d 1017, 1036 (9th Cir.1980); Environmental Defense Fund, Inc. v. Hoffman, 566 F.2d 1060,1067 (8th Cir.1977). This Court must observe that defendants’ tendency to stop their analyses at the North Dakota-South Dakota border suggests a certain insensitivity to the legitimate concerns of South Dakotans. But even though it would have been desirable for there to have been more explicit recognition of South Dakota impacts of the FDU, the Court cannot declare the statements inadequate on this basis. As the Court said in Environmental Defense Fund, Inc. v. Corps of Engineers, 470 F.2d 289, 297 (8th Cir.1972), it is doubtful that any agency, however objective, however sincere, however well-staffed, and however well-financed, would come up with a perfect environmental impact statement in connection with any major project. Further studies, evaluations and analyses by experts are almost certain to reveal inadequacies or deficiencies. But even such deficiencies and inadequacies, discovered after the fact, can be brought to the attention of the decision-makers, including, ultimately, the President and the Congress itself. This Court must also repeat the observation made in the case of United Family Farmers, Inc. v. Kleppe, 418 F.Supp. 591, 601 (D.S.D.1976), another case dealing with the adverse impacts on the James River of an irrigation project, that “NEPA was not intended to prohibit those projects resulting in environmental degradation. The purpose of NEPA was to insure that environmental degradation would occur only after such effects were brought into the decision-making process.” The environmental statements prepared by defendants have fulfilled these goals, and have adequately put the decision-makers on notice of the hazards to South Dakota if the GDU project is completed. E. Substantive NEPA Review. Turning now to the second function of this Court’s review under NEPA, the Court must determine “whether the agency acted *1296 within the scope of its authority.” Environmental Defense Fund, Inc. v. Corps of Engineers, supra, 470 F.2d at 300. Given the long history of Congressional authorization and continued appropriations for the GDU, sometimes even in the face of periods of opposition by at least one Administration, the Court can hardly do otherwise than to find that the decision to proceed with Plan 1 in the 1979 FSES was indeed within the scope of defendants’ authority. The inquiry does not end there, however, for the Court is to next determine “whether the decision reached was arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.... [The Court] must decide if the agency failed to consider all relevant factors in reaching its decision, or if the decision itself represented a clear error in judgment.” Id. The Congressional authorization and continuing appropriations for the GDU are, of course, a factor to be taken into account, but such Congressional action does not in itself, as defendants have suggested, decide this question. “NEPA requires that construction projects be completed in accordance with its substantive provisions. An appropriation act cannot serve as a vehicle to change that requirement.” Environmental Defense Fund, Inc. v. Froehlke, 473 F.2d 346, 353-54 (8th Cir.1972). Considering, however, the low level of adverse impacts on South Dakota that have been demonstrated, impacts that have clearly been considered in good faith in the formulation of defendants’ decision to proceed, and balancing this against the decades of Congressional support for the GDU and the very obvious benefits that will be the result of the GDU in North Dakota, this Court can hardly find that defendants’ “decision was an arbitrary and capricious one,” Jackson County, Missouri v. Jones, 571 F.2d 1004, 1015 (8th Cir.1978), or “gave insufficient weight to environmental values.” MPIRG v. Butz, supra, 541 at 1300. Plaintiffs’ first claim is therefore dismissed. III. UNAUTHORIZED PROJECT CLAIM The second of plaintiffs’ claims is, in essence, that even if the 250,000-acre Initial Stage plan (described in the 1979 FSES as Plan 1) is in compliance with NEPA, such compliance is now irrelevant because defendants are in the process of constructing another version of the GDU which has not been authorized by Congress. As noted earlier, the government of Canada has consistently, since 1969, expressed its opposition to the GDU because of possible environmental damage from return flows entering Canada by way of the rivers in the Hudson Bay drainage basin in North Dakota. Lengthy consultations have taken place between the governments of the United States and Canada, but this issue has yet to be resolved. 11 In an attempt to accommodate Canadian concerns, and to get at least a portion of the GDU built while negotiations continued, officials of the State of North Dakota formulated a proposal called “Phased Development” of the GDU. Under Phase I of this plan, only those areas of the GDU (about 85,460 acres) that could be kept separate of the Hudson Bay drainage basin would be developed. Phase I would make several changes in the design of the GDU from the way the GDU was previously described; most notably, these alterations include a rerouting of return flows of the New Rockford area (20,935 acres) away from the Sheyenne River (in the Hudson Bay basin) and into the James River (which is in the Missouri River basin) and construction of the emergency outlet of Lonetree Reservoir so that emergency releases would enter the James River *1297 rather than the Sheyenne River. 12 Phase II would entail the development of GDU irrigation in the Hudson Bay drainage area. It is plaintiffs’ position that defendants have in fact adopted the North Dakota Plan, and are proceeding to construct Phase I. This, plaintiffs allege, is neither in compliance with the 1965 authorizing Act, which directed construction of the GDU “substantially in accordance” with the Bureau of Reclamation report of November, 1962 (revised February, 1965) supplemental to House Document 325, nor is Phase I in compliance with NEPA because the impact of additional return flows from the New Rockford area entering the James River has never been addressed in the environmental statements. Plaintiffs have presented a considerable number of governmental memos and correspondence that strongly suggest that defendants have in fact decided to develop the 85,000-acre Phase I. For example, a memo bearing the date March 31, 1982, from the Commissioner of the Bureau of Reclamation to the Bureau’s Associate Solicitor, states that “Reclamation’s present construction program is to implement the 85,000-acre first phase of the project consisting of features, including the Oakes Pumping Plant and West Oakes Irrigation Area, that would not affect Canada.” Several letters, dated in April, 1982 from the Assistant Commissioner of the Bureau of Reclamation to private citizens in North Dakota, state that “we will be awarding contracts this spring on the Oakes Pumping Plant on the James River and on other features of the 85,000-acre first phase later in the year.” Also, the Diversion Digest of March, 1982 (a publication of intervenor-defendant Garrison Diversion Conservancy District) states “we are pushing forward with the construction of Phase I of the project — the phase that will permit irrigation of 85,000 acres all within the Missouri River watershed.” Against this evidence, and a mass of other, somewhat more equivocal, indications of defendants’ approval of Phase I, must be placed a letter of December 28, 1981 from the Assistant Secretary for Land and Water Resources of the Department of the Interior to Senator Abdnor of South Dakota, stating that the “fact is that the Secretary is still closely examining the 85,-000-acre first phase proposal, but as yet has not made a formal decision on the proposal.” A number, of other documents, including a letter dated April 13, 1982 from the Governor of North Dakota to Secretary Watt, also suggest no final decision by federal defendants on Phase I has been made. Further, there have been repeated assertions under oath by the project manager of the GDU (at trial) and the Assistant Secretary mentioned above (by affidavit) that no final decision has been made by the Department of the Interior. In the Court’s view, the question is a close one, but the Court is unable to find that plaintiffs have met their burden on this issue, particularly in view of the fact that “[t]he presumption of regularity supports the official acts of public officers and, in the absence of clear evidence to the contrary, courts presume that they have properly discharged their official duties.” United States v. Chemical Foundation, 272 U.S. 1, 14-15, 47 S.Ct. 1, 6, 71 L.Ed. 131 (1926). This Court cannot now find that defendants have in fact decided to proceed with the 85,000-acre Phase I for the GDU *1298 rather than the plans described in the environmental statements. 13 The question yet remains of whether the defendants have so committed themselves to the Phase I plan as to require additional NEPA analysis of the additional return flows accruing to the James River. The leading case on this point is Kleppe v. Sierra Club, 427 U.S. 390, 96 S.Ct. 2718, 49 L.Ed.2d 576 (1976), as it has been explained in this circuit by Monarch Chemical Works, Inc. v. Thone, 604 F.2d 1083 (8th Cir.1979). As Thone said, “ ‘the moment at which an agency must have a final statement ready “is the time at which it makes a recommendation or report on a proposal for federal action.” ’... The mere ‘contemplation of a project and the accompanying study thereof does not necessarily result in a proposal for major federal action’... The administrative decisionmaking process must be of sufficient maturity before a judicially enforceable duty to prepare an environmental assessment of a proposed project exists ... To require the preparation of an environmental assessment at an earlier point would be to invite judicial involvement in the day-today decision-making process of an agency, and to require the preparation of unnecessary impact statements.” 604 F.2d at 1090. (emphasis in original). Here, again, although the Court views it as a close question, the Court cannot say that defendants have gone beyond the “mere contemplation and accompanying study thereof” of the North Dakota proposal. 14 The Court does note, however, that defendants have conceded that if the North Dakota proposal is adopted, an environmental review of the altered plans would have to be conducted. See May 27, 1982 Affidavit of Garrey E. Carruthers, Assistant Secretary, Land and Water Resources, U.S. Department of the Interior, paragraph 5. If the officials of any federal agency attempt to proceed with a project without NEPA compliance, the courts will of course be open to review that agency action and to order such remedy as may be appropriate. But, at least, as matters stand now, 15 there is nothing before this Court to require its exercise of judicial power. Plaintiffs’ complaint, including both the first and second claims, is therefore dismissed. The foregoing represents the findings of fact and conclusions of law of this court. Footnotes [1] The Society also appears as amicus here. [2] It appears that the issue of the adequacy of the 1974 FES has subsequently been deleted from the Society’s complaint in the District of Columbia. [3] This is partly because the return flow drainage from the East Oakes area under this plan will accrue to the Wild Rice River, which is in a different drainage basin than the James River. [4] Plaintiffs’ complaint also alleged that the reduction of flows to South Dakota with the GDU would impair ability to irrigate from the James River. Plaintiffs, upon whom is placed the burden of proof, presented no significant evidence in support of this allegation. [5] See this Court’s rather brief description of the River in Oahe Conservancy Sub-District v. Alexander, 493 F.Supp. 1294, 1298-99 (D.S.D. 1980). [6] The term “salinity” is used to indicate the degree of total dissolved solids (TDS) present in water, which are the chemicals that are left when a particular quantity of water is evaporated. The general term “water quality” relates to the levels of specific substances present in water, including TDS, various chemicals, and bacteria. [7] The 1976 DES states that “[w]ater travel time from Jamestown Dam to Oakes Pumping Plant will be about 100 hours.” [8] Plaintiffs contend that this study shows only the effects of the GDU after a period of about 50 years has elapsed, and ignores the short-term adverse impacts of the GDU before the system has reached equilibrium. The Court is satisfied, however, that the 1979 FSES, particularly in Table 37, does clearly show GDU impacts throughout the entire period of the project’s operation. [9] These projected flows contrast with the historic months of high median flows, which range from a high of 110 c.f.s. in April and 80 c.f.s. in June, and the historic months of low median flows of 10 c.f.s. to 20 c.f.s. in the months of September through February. It might again be noted that the capacity of the James River in northern South Dakota is 200 c.f.s. [10] Similar conclusions were reached in the analysis of each alternative GDU plan described in the FSES, each of which included a discussion of GDU effects on the James River at the North Dakota-South Dakota border. [11] The 250,000 acre GDU authorized by Congress in 1965 contemplates 203,890 acres of irrigation which are within the Hudson Bay watershed and will ultimately drain into Canada. It is reasonable to conclude that the United States has in effect made a commitment to Canada not to construct any part of the GDU which will drain into Canada, unless Canada consents. The court reaches this conclusion from the evidence in the record, but notes also that House Joint Resolution 631, enacted December 21, 1982 provides that the four million dollars appropriated for GDU shall not be used for construction of features of the GDU affecting waters flowing into Canada. [12] It should, however, be noted that the 1979 FSES does discuss the possibility of the emergency outlet to the James River at Lonetree Reservoir. This outlet would have a 2,000 c.f.s. capacity and, in the words of the FSES, the possibility for its use “is extremely remote and would only occur in the event of natural disaster or emergency repair of a dam . . . with the full release of 2,000 ft3s, serious flooding would occur along the river valley from south of Harvey to Jamestown Reservoir.” In a December 18, 1980 memo, which also considered the possibility of flooding in South Dakota if the outlet was used, defendants concluded that, largely because the probability of outlet use (as well as use of an increased capacity in the James River Feeder Canal to aid in dam evacuation) was “so small as to be incalculable”, the alterations in emergency methods were not a “substantial change” in the GDU and no additional environmental analysis was required. This Court has no basis to disturb this finding. See Warm Springs Dam Task Force v. Gribble, 621 F.2d 1017, 1026-25 (9th Cir.1980). [13] Thus, the court does not reach in this case the question of whether the 85,000 acre “Phase I” plan may be constructed without additional authority from Congress. [14] There is thus, as yet, no “substantial change” in the GDU, as under 40 C.F.R. § 1502.9(c)(1), that would require a supplementary environmental statement. [15] The record includes exhibits 1 through 208, and the transcript of testimony. The last filing in the record before the court is a post-trial rebuttal brief filed August 9.