Circuit Judge, concurring in part and dissenting in part.
This case may prove to be one of the most important cases to be decided by the United States courts in this century. The United States Supreme Court, in reversing this court’s previous remand of this case to respondent, the Nuclear Regulatory Commission (NRC or Commission), emphasized the magnitude of the problem it was sending back for our reconsideration:
The Commission itself, in a pamphlet published by its information office, clearly recognizes that these wastes “pose the most severe potential health hazard. . . . ” U.S. Atomic Energy Commission, Radioactive Wastes 12 (1965). Many of these substances must be isolated for anywhere from 600 to hundreds of thousands of years. It is hard to argue that these wastes do not constitute “adverse environmental effects which cannot be avoided should the proposal be implemented,” or that by operating nuclear power plants *45we are not making “irreversible and irretrievable commitments of resources.” 42 U.S.C. §§ 4332(2)(C)(ii), (v).
Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519, 538-39, 98 S.Ct. 1197, 1208-09, 55 L.Ed.2d 460 (1978), rev’g Natural Resources Defense Council v. Nuclear Regulatory Commission, 547 F.2d 633 (D.C.Cir.1976).
In this case we are required to review the continuing effort of the NRC to pit human intelligence against the most primordial force of nature. This force, when involved in its most awful manifestation, exceeds the power of flood, fire, pestilence, earthquake, hurricane and volcano. In this century, it has been demonstrated in this and other countries that this force can be employed for peace and war — for warming a baby’s bottle and for nuclear holocaust.
It is not the function of the judicial branch either to initiate or to halt the development of a nuclear power industry. These national policy choices are constitutionally vested in the President and the Congress of the United States, and they have been made in favor of proceeding.
This court's limited function is to determine from the agency record before us whether or not the Commission’s decision to adopt certain Tables S-3, 10 C.F.R. § 51.-20(e) (1981), was “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.”1 The S-3 Tables instruct the licensing boards, which are delegated the task of licensing all new nuclear power plants, that there will be “zero” release of emissions or releases so small as to be insignificant,2 from the nuclear waste accumulated or to be accumulated from plants now licensed or which may be licensed during the 2,500 centuries it will take for that waste to decay. Thus the Commission seeks to preclude further licensing hearings from considering the hazards from unanticipated, accidental or willful release of emissions beyond those specified in the Tables from the highly toxic and potentially explosive wastes generated by nuclear power plants during temporary storage, reprocessing, transportation and 250,000 years of burial. 46 Fed.Reg. 15154, 15156 (March 4, 1981).
I concur fully with Judge Bazelon’s opinion for the court, except for the cost issue.
I write separately in concurrence to express deep concern for additional problems and uncertainties that I believe to be buried in the naked figures and enigmatic footnotes of the Tables S-3 that this appeal requires us to review. Further, in computing the cost of the proposed nuclear waste disposal system, the NRC has not included the costs of surveillance of and guarding the public from the toxic effects from nuclear waste at all disposal sites for somewhere between 10,000 to 250,000 years. As a consequence, I dissent from the opinion for the court on this issue and state my reasons at the end of this opinion.
My concurrence with Judge Bazelon is based on the reasoning he has ably set forth. But I find what appear to me to be even more substantial reasons to agree with his result. An integral part of the “back end” of the nuclear fuel cycle involves the NRC’s plan for reprocessing spent fuel rods. Reprocessing involves extraction from the spent fuél rods of either the remaining uranium and plutonium (uranium and plutonium recycle) or just the remaining uranium (uranium-only recycle). Both elements can be prepared as fuel usable in light-water reactor power plants. The uranium can be enriched and again fashioned into uranium-based fuel rods; in the alternative, the plutonium can be combined with uranium to form mixed oxide fuel rods. There are many indications in this record that the Commission originally intended to use both uranium and plutonium extracted by reprocessing as supplementary fuel in all nuclear power plants, licensed or to be licensed.
*46On April 7, 1977, a presidential order indefinitely suspended reprocessing of spent fuel. 13 Weekly Comp, of Pres. Doc. 506. A new presidential statement on nuclear energy policy, however, dated October 8, 1981, “lift[ed] the indefinite ban which previous administrations placed on commercial reprocessing activities in the United States.” 17 Weekly Comp, of Pres. Doc. 1102. There is now reason to believe that the Commission will proceed with plans to reprocess spent fuel rods thus extracting both uranium and plutonium.
Table S-3 in its final form purports to summarize all “releases” that may be expected in the entire nuclear fuel cycle, except for radon and technetium, which are irrelevant to our present considerations. Moreover, in a proposed “Narrative Explanation of Table S-3,” the NRC states that “the environmental impacts from reprocessing and related waste management activities are nearly identical for the recycling of uranium and plutonium, and for the recycling of uranium only, as fuel in nuclear power reactors.” 46 Fed.Reg. 15154, 15161 (March 4, 1981).
The proposed narrative avoids discussing the potential environmental impacts of plutonium. The Commission has not, however, disavowed its option to use both uranium and plutonium as reactor fuel — probably in the form of mixed oxide fuel. The NRC may well read any final approval of the S-3 Tables by the federal courts as foreclosing any further public hearings or litigation over the matter.
Absent any clear indication to the contrary from the Commission, I feel prudence requires this court to assume that approval of Tables S-3 would be read by the Commission as judicial sanction for its originally intended — and now presidentially permitted — employment of plutonium as reactor fuel.
Plutonium is the most dangerous substance yet known in the history of the world. Airborne, the tiniest speck of it breathed into a human being will produce lung cancer. Twenty pounds of this powder can be made into a nuclear bomb with the explosive force of the atomic bombs that destroyed Hiroshima and Nagasaki in the fall of 1945. The Commission’s originally contemplated nuclear waste disposal scheme included reprocessing of all spent fuel rods from all nuclear power plants now licensed (71) or in contemplation (190). Such reprocessing, depending on the number of plants in operation, would extract each year 35,000 to 93,500 pounds of plutonium3 in the form of plutonium oxide. The plutonium, in the form of mixed oxide fuel, would then be transported by truck or rail through many heavily populated areas to all nuclear power plants wherever located in the nation and stored until it is consumed as nuclear fuel.
If this scheme is executed according to what I perceive to be NRC’s plan, many private corporations will continually have on hand substantial stocks of the most lethal weapons material ever produced.
Unlike spent fuel rods, plutonium oxide powder can be approached by human beings without fear of radiation. If sealed in an airtight container, it can be transported without harm to the bearer. Also, it is estimated that many thousand people in the United States have the knowledge necessary to fashion a nuclear bomb from this material.4
In the dangerous world in which we live, attempts by foreign or domestic terrorists to divert plutonium oxide to international or domestic blackmail or catastrophic destruction must be anticipated.
*47How NRC contemplates protecting this huge amount of hazardous material from forcible seizure on the highways and rail lines of the United States, and from theft or illegal diversion from storage at between 71 and 190 nuclear power plants is most inadequately dealt with in this record.
Finally, the ancient question, “But who is to guard the guards themselves?” is pertinent. The nuclear weapons of the United States are closely guarded by its armed forces. Their employment is subject to command controls so carefully devised that they terminate in the hands of the President of the United States. Yet it appears that the NRC plans to ship bomb-grade materials freely by road or rail, guarded by civilian employees of the private corporation that secures the reprocessing contract. 10 C.F.R. § 70 (1981). And it appears that the NRC may place supplies of plutonium oxide in the form of mixed oxide fuel with its awesome power on the premises of a total of 71 to 190 nuclear power plants under the control of private corporate officers chosen by and responsible to as many private corporate boards of directors. 10 C.F.R. § 72 (1981).
In the long history contemplated for the nuclear power adventure, we cannot be certain that the raw power placed in so many private hands will never be used against the security of our people or of our form of government.
Any failure to deal with the hazards of this plan may be read by the NRC as validation sub silentio of a plan involving incredible dangers. In its original consideration of using atomic energy for production of electric power, Congress repeatedly required “the protection of the health and safety of the public.” For the reasons which follow, I would not risk the dangers and uncertainties contained in the S-3 Tables without a prior remand of the NRC reprocessing, transportation and storage plans for thorough briefing and argument as to the legality of said plans for the use of plutonium as fuel under the public health and safety requirements of the Atomic Energy Act of 1954. 42 U.S.C. §§ 2012(d), (e), 2133(b), (d), 2232(a) (1976).
THE THREE S-3 TABLES
In this case we are now reviewing three variations of Table S-3: an original version, an interim version and a final version. For purposes of this appeal, the differences between the three versions are inconsequential. To illustrate something about the nature of the problem with which this court has been faced, I print the final version of Table S-3 in full:
Table S-3. — Table of Uranium Fuel Cycle Environmental Data1
*48
*49
1 In some cases where no entry appears it is clear from the background documents that the matter was addressed and that, in effect, the Table should be read as if a specific zero entry had been made. However, there are other areas that are not addressed at all in the Table. Table S-3 does not include health effects from the effluents described in the Table, or estimates of releases of Radon-222 from the uranium fuel cycle or estimates ot Technetium-99 released from waste management or reprocessing activities. These issues may be the subject of litigation in the individual licensing proceedings.
Data supporting this table are given in the “Environmental Survey of the Uranium Fuel Cycle,” WASH-1248, April 1974; the “Environmental Survey of the Reprocessing and Waste Management Portion of the LWR Fuel Cycle,” NUREG-0116 (Supp. 1 to WASH — 1248); the “Public Comments and Task Force Responses Regarding the Environmental Survey of the Reprocessing and Waste Management Portions of the *50LWR Fuel Cycle,” NUREG-0216 (Supp. 2 to WASH-1248); and in the record of the final rulemaking pertaining to Uranium Fuel Cycle Impacts from Spent Fuel Reprocessing and Radioactive Waste Management, Docket RM-50-3. The contributions from reprocessing, waste management and transportation of wastes are maximized for either of the two fuel cycles (uranium only and no recycle). The contribution from transportation excludes transportation of cold fuel to a reactor and of irradiated fuel and radioactive wastes from a reactor which are considered in Table S-4 of § 51.20(g). The contributions from the other steps of the fuel cycle are given in columns A-E of Table S-3A of WASH-1248.
2 The contributions to temporarily committed land from reprocessing are not prorated over 30 years, the complete temporary impact accrues regardless of whether the plant services one reactor for one year or 57 reactors for 30 years.
3 Estimated effluents based upon combustion of equivalent coal for power generation.
4 1.2 percent from natural gas use and process.
10 C.F.R. § 51.20(e) (1981).
As of this date the Commission has not finally5 adopted any explanation of Table S-3, except to the degree that such may be found in the footnotes to the Table and in the briefs filed by the Commission.
Turning to the Table itself, the critical language may be found on the fifth from the last line of the Table, which reads:
“TRU and HLW (deep) .... 1.1 X 107 Buried at Federal Repository.”
Translated into possibly more understandable English, “TRU and HLW” mean transuranic and high level waste. “([D]eep)” is the sole stated explanation of the contemplated storage or disposal of such waste under the Commission’s present thinking. The figure 1.1 X 10 to the 7th power refers to a total of 11 million curies per nuclear reactor per year. In a footnote keyed to the Table Heading, the Commission does elucidate further, as follows:
In some cases where no entry appears it is clear from the background documents that the matter was addressed and that, in effect, the Table should be read as if a specific zero entry had been made.
Table S-3 n.l.
This innocent sounding language actually stands for the Commission’s holding that 11 million curies of high level radioactive waste per year for each of the presently licensed reactors will be temporarily stored, reprocessed, transported, and then buried and contained deep underground at some unascertained place in some undetermined *51stratum6 at some undetermined time without any release at all of toxic effluents. The life of presently constructed reactors is estimated to be 30 to 40 years. See Proposed Appendix A, Narrative Explanation of Table S-3, 46 Fed.Reg. 15154, 15162 n.2 (Mar. 4, 1981). This means that within the lifetime of the 70 7 presently licensed nuclear power plants, there will be accumulated for disposal at least 2,100 plant-years of radioactive waste, each plant-year representing by Commission estimate 11 million curies.
If the mathematics involved is not sufficiently impressive, it may be appropriate to point out that a “curie” is defined as “a measure of the number of atoms undergoing radioactive disintegration per unit time and is 37 billion disintegrations per second, or about the rate of decay in 1 gram of natural radium.” C. Fox, Radioactive Wastes 11 n.* (rev. ed. 1969).
If no other nuclear power plant is licensed or brought on line, the task of the Commission will involve containment in one or more federal repositories of radioactive waste sufficient to generate 854,700,000,-000,000,000,000 (854.7 quintillion) disintegrations per second. Since the decay of radioactive waste spontaneously gives off substantial heat, the problem of preventing accidental or deliberate emission is central to any design that may be proposed for the federal repository. The gradual decay of radioactivity will extend, as the Supreme Court pointed out, from 600 to hundreds of thousands of years. Vermont Yankee, 435 U.S. at 538, 98 S.Ct. at 1209. Yet the Commission contends that this mass of energy and toxic waste (plus possibly two and one-half times more, which may be added by additional nuclear power reactor licensing) can and will be contained with “zero” release. This must be read as meaning no impact at all on the human race in its turbulent occupancy of this biosphere for a minimum period of 250,000 years.
THE COMMISSION’S WASTE DISPOSAL PLAN
Licensing of nuclear power reactors began in 1958. Although some 82 licenses have been issued since 1958, several of those plants are no longer operating. At present, 71 reactors are “licensed for operation,” although not all are currently on line.8 The record before us contains published studies of various proposals for handling high-level nuclear waste for over 250,000 years. Many are in conflict. And one important aspect of the Commission’s original plan, the reprocessing of spent fuel rods, had been in suspension from 1977 until October 8, 1981.
Originally, the Commission’s thinking appears to have encompassed consideration of at least the following elements:
1) Containment buildings for spent fuel rods constructed at each nuclear power plant. These buildings are operated as an integral part of the plant and are guarded within the private power company’s perimeter fences. No other temporary, interim, or permanent storage facility for commercial high-level nuclear waste has yet been built.
As a consequence, spent nuclear fuel is currently accumulating in all nuclear power reactor on-site waste storage facilities — in many instances in quantities far in excess of the originally contemplated storage capability. For example, the Vermont Yankee Nuclear Power Station in Vernon, Vermont, a relatively recent entry on the nuclear scene, was designed to store a maximum of 1,000 spent fuel rods. The Vermont Yankee Nuclear Power Corporation has now been given permission by the NRC to store *522,400 such spent fuel assemblies there. It is estimated that the plant can operate only until 1987 under this grant of authority, unless, presumably, some now nonexistent interim or permanent storage facility is provided or authority is granted for the spent fuel rods to be stacked even closer.
2) The NRC’s original assumption was that all spent fuel produced by nuclear power reactors would be reprocessed before it was stored. It was also contemplated that the residue of high level waste, particularly plutonium, would be greatly diminished in volume in the process. Only this residue after such extraction was scheduled for storage in the permanent repository. As will be indicated below, this plan is now at least partially back in place.
In fact, a plant for reprocessing spent fuel was built by private industry at West Valley in New York State. It was small; it produced high levels of radioactive effluent releases. See 44 Fed.Reg. 45362, 45370 (Aug. 2, 1979). The operation proved to be uneconomical and was abandoned. Three other more sophisticated reprocessing plants — one actually constructed in Barn-well, South Carolina, and the others planned for Oak Ridge, Tennessee, and Marion, Illinois — were contemplated as a basic part of the NRC’s nuclear waste disposal plan. This record indicates that thus far no reprocessing of spent nuclear fuel, except in small experimental amounts, has occurred.9
3) The NRC also contemplates an interim storage facility. This was described as a retrievable surface storage facility (RSSF). Scheduled to be built about 1980, the RSSF was thought to be capable of storing high-level waste for up to 100 years. The facility would have been built at ground level or slightly below. The high-level waste was to be placed in canisters capable of containing toxic emissions. The dissemination of heat from the decay processes would be accomplished by bringing containment canisters into contact with a coolant, either water or air. Each canister was then to be stored and subjected to cooling for 100 years in the RSSF before being moved to permanent storage.
No such facility has ever been built.
4)Most Commission documents indicate that all high-level waste ultimately will be placed in a federal repository for permanent disposal. The most frequently cited plan, as indicated earlier, is for deep burial in geologic salt deposits.10 Studies of salt bed deposits have been made, the results of which the Commission staff regards as favorable to permanent containment.
THE REVIEW STANDARD
The statute under which the Supreme Court has directed us to review this case is the Administrative Procedure Act, 5 U.S.C. §§ 551-59, 701-06 (1976). In applicable part the Act says:
To the extent necessary to decision and when presented, the reviewing court shall decide all relevant questions of law, interpret constitutional and statutory provisions, and determine the meaning or applicability of the terms of an agency action. The reviewing court shall —
(1) compel agency action unlawfully withheld or unreasonably delayed; and
(2) 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;
(D) without observance of procedure required by law;
*53(E) unsupported by substantial evidence in a case subject to sections 556 and 557 of this title or otherwise reviewed on the record of an agency hearing provided by statute; or
(F) unwarranted by the facts to the extent that the facts are subject to trial de novo by the reviewing court.
Id. § 706 (emphasis added).
The critical language to which the Supreme Court directed our attention, Vermont Yankee, 435 U.S. at 535 n.14, 98 S.Ct. at 1207 n.14, is found italicized above. The Supreme Court has unanimously interpreted this language as follows:
Under the “arbitrary and capricious” standard the scope of review is a narrow one. A reviewing court must “consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment.. .. Although this inquiry into the facts is to be searching and careful, the ultimate standard of review is a narrow one. The court is not empowered to substitute its judgment for that of the agency.” Citizens to Preserve Overton Park v. Volpe, supra, [401 U.S.], at 416 [91 S.Ct. at 823]. The agency must articulate a “rational connection between the facts found and the choice made.” Burlington Truck Lines v. United States, 371 U.S. 156, 168 [83 S.Ct. 239, 246, 9 L.Ed.2d 207] (1962). While we may not supply a reasoned basis for the agency’s action that the agency itself has not given, SEC v. Chenery Corp., 332 U.S. 194, 196 [67 S.Ct. 1575, 1577, 91 L.Ed. 1995] (1947), we will uphold a decision of less than ideal clarity if the agency’s path may reasonably be discerned. Colorado Interstate Gas Co. v. FPC, 324 U.S. 581, 595 [65 S.Ct. 829, 836, 89 L.Ed. 1206] (1945).
Bowman Transportation, Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 285-86, 95 S.Ct. 438, 441-42, 42 L.Ed.2d 447 (1974).
This court has repeatedly cited and followed Bowman and other Supreme Court authority on the arbitrary and capricious standard:
It is axiomatic that we may not substitute our judgment for that'of the agency. Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814 [823], 28 L.Ed.2d 136 (1971). Yet our review must be “searching and careful,” id., and we must ensure both that the Commission has adequately considered all relevant factors, see id., and that it has demonstrated a “rational connection between the facts found and the choice made,” Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 246, 9 L.Ed.2d 207 (1962).
Home Box Office, Inc. v. FCC, 567 F.2d 9, 35 (D.C.Cir.) (per curiam), cert. denied, 434 U.S. 829, 98 S.Ct. 111, 54 L.Ed.2d 89 (1977). Accord, United States Lines, Inc. v. Federal Maritime Commission, 584 F.2d 519, 526 (D.C.Cir.1978); Ethyl Corp. v. EPA, 541 F.2d 1, 34-36 & n.74 (D.C.Cir.) (en banc), cert. denied sub nom. E. I. duPont de Nemours & Co. v. EPA, 426 U.S. 941, 96 S.Ct. 2662, 49 L.Ed.2d 394 (1976).
“ZERO” RELEASE AT THE FINAL REPOSITORY
On this issue I am in full agreement with Judge Bazelon’s opinion and will add only a few paragraphs in support thereof. I agree that it is arbitrary and capricious for the NRC to close off discussion of the grave problems involved in building a permanent repository for all nuclear waste by categorical determination in Table S-3 that there will be zero releases. As I read this record, neither the Commission nor the Commission staff really defend the zero-release conclusion for any purpose other than to prevent discussion of the issue at licensing hearings. This despite the fact that both Commission and staff recognize that present scientific information strongly suggests that there will be releases of gaseous nuclides during the substantial period of time before the repository is closed. Their studies have shown a probability that transportation and storage canisters designed to contain radiological emissions will corrode and disintegrate in time, releasing their contents. The Commission staff contemplates some escape *54of radionuclides either by gaseous emission or ground water migration. In the proposed Narrative, the Commission staff, at least tentatively, holds that these emissions and migrations will occur “early” in the 250,000-year life of the federal repository and that risks therefrom will be “acceptable.” It is not clear whether the risks become acceptable before or sometime during the first 10,000 years of decay.
The following Table depicts the periods of toxicity of just a few of the various isotopes found in the high-level wastes that may be stored in the federal repository:
Element Half-Life (In years) Period of Toxicity (In years)
Plutonium-239 24,000 250,000 (approx.)
Strontium-90 28 600-1,000
Cesium-137 30 600-1,000
Technetium-99 210,000 2,000,000 (approx.)
The hope and expectation is really that the impact of the releases will be insignificant on human beings.
Even this last conclusion is subject to very significant uncertainty. As yet undetermined amounts of fissionable materials (markedly reduced, however, if reprocessing is employed) will be emplaced in the repository. In* 250,000 years there may be one or more ice ages. See 46 Fed.Reg. 15154, 15172 (Mar. 4, 1981). The probability of container corrosion, the certainty of heat from the quintillions of disintegrations of radioactive wastes, the probable contact with ground water, added to the possibility of geologic disturbance by such forces as glacier, earthquake, or volcanic eruption, all tend to create significant uncertainties. The NRC relies on two studies for its conclusion that the uncertainties involved in waste disposal do not mandate that it alter the zero-release value. 44 Fed.Reg. 45362, 45367 n.19, 45368 & n.24 (Aug. 2, 1979), One study is that of the Interagency Review Group, Report to the President by the Interagency Review Group on Nuclear Waste Management (March 1979) (hereafter IRG Report). The IRG admits, however, that there is a large gap between actual and theoretical capabilities in the technology of nuclear waste disposal. Thus, while the IRG Report may correctly conclude geologic disposal is feasible, it cannot be relied upon to prove that radiological emissions will be zero. The IRG Report concedes as much when it states, “[I]n addition to technical evaluation, a societal judgment that considers the level of risk and the associated uncertainty will be necessary.” Id. at 42.
In the opening pages of the revised Report, the Group states that safe disposal requires a program “whose resolution will clearly require an unprecedented extension of capabilities in rock mechanics, geochemistry, hydrogeology and long-term predictions of seismicity, volcanism, and climate.” Id. at 3. It later summarizes, “The risk assessments performed to date . . . have . . . been based on idealized repository characteristics and are subject to significant uncertainties.” Id. at 45. It then specifically described some of the remaining problems:
The IRG would like to emphasize, however, that risk assessment computations are needed which are site specific in the origin of their data and in their application and only such assessments can truly apply to a specific site. Some uncertainties about the level of risk will always remain both because of the current state of the methodology of risk assessment and because of its inherent limitations.
Uncertainties associated with risk assessment derive from lack of data, lack of experience, inability to identify all release mechanisms for radionuclides, the natural variability in physical properties of geologic media, and inability to predict long-term geologic and climatic processes and social evolution. All of these uncertainties are neither additive nor of equal significance. An important aspect of the research remaining to be done is to understand how each enters into the overall uncertainty of the calculation of risk.
Id. at 46.
It concludes that further study of the problem is necessary to reduce uncertainties:
*55Significant institutional difficulties are involved in: marshalling the resources and programs capable of accurately detailing site suitability criteria and establishment of standards; thoroughly investigating possible sites; accurately assessing site characteristics in light of the technical criteria; carrying out credible analyses of the risks; obtaining agreement on site selection; getting the facility approved and licensed; providing for careful construction and operation of the repository (including safe transportation and handling of the wastes); mitigating accidents and responding to repository failure if that occurs; and providing adequate, long-term monitoring. The level of difficulty of all these problems could increase with the size of the nuclear waste inventory and its rate of growth. Institutions that can cope on a small scale may fail as the demands placed on them multiply. The IRG believes that a more detailed analysis of logistical and other institutional problems which would arise out of attempting to manage wastes on the scale require[d] should be undertaken.
Id. at 88 (emphasis added).
The Commission’s leap from its hopes for safe containment to its conclusion of zero release is, as Judge Bazelon concludes, arbitrary and capricious.
THE PLUTONIUM ISSUE
The Commission’s brief argues that all three of the fuel-cycle rules should be upheld by this court. As previously indicated in this opinion, the Commission has stated that Table S-3 was designed to cover the entire back-end of the fuel cycle. As a consequence, this panel’s affirmance of the validity of Table S-3, if certiorari were denied or the case affirmed by the Supreme Court, could be taken by the Commission as final approval of its original and present plans for reprocessing spent fuel, fabricating uranium and plutonium into mixed oxide fuel therefrom, and shipping this fuel to nuclear reactors all over the nation.
A. Presidential Concern About Reprocessing
The lethal character of plutonium and its potential for use in nuclear bombs evoked expressions of concern from two successive Presidents, Gerald Ford and Jimmy Carter, about the reprocessing of nuclear waste and the separation of bomb-grade plutonium. This concern resulted in April 1977 in a Presidential statement, issued by President Carter, suspending all NRC plans for reprocessing spent nuclear fuel. This statement reads:
There is no dilemma today more difficult to resolve than that connected with the use of nuclear power. Many countries see nuclear power as the only real opportunity, at least in this century, to reduce the dependence of their economic well-being on foreign oil — an energy source of uncertain availability, growing price, and ultimate exhaustion. The U.S., by contrast, has a major domestic energy source — coal—but its use is not without penalties, and our plans also call for the use of nuclear power as a share in our energy production.
The benefits of nuclear power are thus very real and practical. But a serious risk accompanies worldwide use of nuclear power — the risk that components of the nuclear power process will be turned to providing atomic weapons.
We took an important step in reducing the risk of expanding possession of atomic weapons through the Non-Proliferation Treaty, whereby more than 100 nations have agreed not to develop such explosives. But we must go further. The U.S. is deeply concerned about the consequences for all nations of a further spread of nuclear weapons or explosive capabilities. We believe that these risks would be vastly increased by the further spread of sensitive technologies which entail direct access to plutonium, highly en-, riched uranium, or other weapons usable material. The question I have had under review from my first day in office is how can that be accomplished without forgoing the tangible benefits of nuclear power.
*56We are now completing an extremely thorough review of all the issues that bear on the use of nuclear power. We have concluded that the serious consequences of proliferation and direct implications for peace and security — as well as strong scientific and economic evidence— require
—a major change in U.S. domestic nuclear energy policies and programs; and
—a concerted effort among all nations to find better answers to the problems and risks accompanying the increased use of nuclear power.
I am announcing today some of my decisions resulting from that review.
First, we will defer indefinitely the commercial reprocessing and recycling of the plutonium produced in the U.S. nuclear power programs. From our own experience, we have concluded that a viable and economic nuclear power program can be sustained without such reprocessing and recycling. The plant at Barnwell, South Carolina, will receive neither Federal encouragement nor funding for its completion as a reprocessing facility.
Second, we will restructure the U.S. breeder reactor program to give greater priority to alternative designs of the breeder and to defer the date when breeder reactors would be put into commercial use.
Third, we will redirect funding of U.S. nuclear research and development programs to accelerate our research into alternative nuclear fuel cycles which do not involve direct access to materials usable in nuclear weapons.
Fourth, we will increase U.S. production capacity for enriched uranium to provide adequate and timely supply of nuclear fuels for domestic and foreign needs.
13 Weekly Comp, of Pres. Doc. 506 (1977) (emphasis added).
The Commission recognized the impact of the Presidential order in a “Memorandum of Decision” issued December 23, 1977. In that memorandum, the Commission announced its decision:
(1) to terminate the GESMO proceeding;
(2) to terminate the proceedings on pending or future plutonium recycle-related license applications, except for—
(a) proceedings on licenses for the fabrication or use of small quantities of mixed oxide fuel for experimental purposes, and
(b) those portions of proceedings which involve only spent fuel storage, disposal of existing waste, or decontamination or decommissioning of existing plants;
(3) to reexamine the above matters after the completion of the ongoing domestic and international studies of alternative fuel cycles, now expected to take about 2 years;
(4) to publish the draft safeguards supplement to the GESMO documents as a staff technical report;
(5) as a consequence of the above decisions, to withdraw the November 1975 ’ policy statement on mixed oxide fuel, 40 Fed.Reg. 53056; and
(6) to reserve for decision, if it arises, the question whether a facility such as the Allied-General Nuclear Services (AGNS) Nuclear Fuels Plant at Barn-well, South Carolina, may be licensed for experimental and feasibility purposes on a non-commercial basis to investigate processes which support the nation’s non-proliferation objectives.
This memorandum provides the reasons for the December 23 decision.
BACKGROUND
The use of mixed oxide fuel has been before the Commission and its predecessor, the AEC, for more than a decade. In 1960, Nuclear Fuel Services (NFS) began a small reprocessing plant at West Valley, New York, which operated from 1966 through 1971. Construction of the AGNS Plant at Barnwell began in 1970 and parts of the plant are now essentially complete. AGNS’ application for an operating license is currently before the Commission. In 1973, Westinghouse Electric Corporation (Westinghouse) re*57quested a construction authorization letter for a mixed oxide fuel fabrication plant near Anderson, South Carolina. Finally, Exxon Nuclear Company, Inc., (Exxon) is currently seeking permission to construct a reprocessing plant at Oak Ridge, Tennessee.
The health, safety, and environmental impacts of the wide-scale use of mixed oxide fuel were evaluated in the draft Generic Environmental Statement on Mixed Oxide Fuel, published in August 1974. The draft GESMO prompted many public comments, including a January 1975 letter from the President’s Council on Environmental Quality which stressed the need to consider the safeguards aspects of wide-scale plutonium recycle. In May 1975, the Commission announced its provisional intention to supplement GES-MO with an analysis of safeguards and to limit interim licensing of recycle-related activities to experimental purposes. 40 Fed.Reg. 20142 (May 8, 1975). Over 200 public comments were received in response. In November 1975, the Commission published a policy statement which announced that safeguards alternatives would be a part of the GESMO decision, provided for hearings on the GESMO documents, and stated criteria under which interim licensing of nonexperimental recycle-related activities would be considered. 40 Fed.Reg. 53056 (November 14, 1975), corrected 40 Fed.Reg. 59497 (December 24, 1975). The United States Court of Appeals for the Second Circuit affirmed the Commission’s hearing procedures but held that interim licensing of recycle-related activities on a commercial scale violated the National Environmental Policy Act (NEPA). Natural Resources Defense Council, Inc. v. Nuclear Regulatory Commission, 539 F.2d 824 (1976), vacated and remanded to determine mootness, 434 U.S. 1030 [98 S.Ct. 759, 54 L.Ed.2d 777] (January 16, 1978).
The final impact statement on health, safety, and the environment was published in August 1976 and associated public hearings began that November. The hearings progressed through questioning of the NRC staff on its testimony and the filing of written testimony by all partid-, pants and proposed questions on that testimony.
Congress and other parts of the Federal Government, members of the public, and experts in the national security field have continued to express concerns related to the nuclear weapons proliferation risks of plutonium recycle technology as it is presently conceived. That technology produces separated plutonium, which can be used in the production of nuclear explosives. The concern is basically that the international spread of plutonium recycle technology for commercial nuclear power production creates a risk that nonnuclear weapons states might turn plutonium from the commercial fuel cycle to the production of nuclear explosives. See generally Senate Committee on Government Operations, 94th Cong., 1st. Sess., Peaceful Nuclear Exports and Weapons Proliferation — A Compendium (Comm. Print 1975). Moreover, a decision by the United States to proceed with commercial plutonium recycle domestically would undermine U.S. efforts to restrain premature international resort to plutonium. The risk led the Commission in late summer 1976 to direct its staff to begin an analysis for GESMO for international proliferation risks and safeguards.
On October 28, 1976, President Ford discussed these risks in his Statement on Nuclear Policy, and stated that the nation “should pursue reprocessing and recycling in the future only if they are found to be consistent with our international [nonproliferation] objectives" (p. 4).
President Carter expressed his concern over the proliferation risks of plutonium recycle and the potential availability of other weapons-grade materials in the nuclear fuel cycle in his April 7, 1977, Statement on Nuclear Power Policy. As part of his response to these risks, the President stated that the Administration’s policy would be to defer indefinitely domestic plutonium recycle and to initiate a multinational evaluation of alternative *58 fuel cycles in order to promote the government’s international nonproliferation goals. The GESMO hearings were indefinitely postponed by the GESMO Hearing Board, and on May 8 the Commission announced that it intended to reassess “the future course and scope of GESMO, the review of recycle-related license applications, and the matter of interim licensing.” 42 Fed.Reg. 22964 (May 5, 1977). Public comments were received in June.
On May 5, Marcus Rowden, then Chairman of the Commission, wrote the President requesting his views on these matters. Stuart Eizenstat, Assistant to the President for Domestic Affairs and Policy, responded for President Carter on October 4. The letter (reprinted at 42 Fed. Reg. 57186 (November 1, 1977)) states that “the President believes that his nonproliferation initiatives would be assisted both domestically and internationally if the Commission were to terminate the GESMO proceedings,” “terminate] ... staff reviews and hearings relating to recycle activities . . . den[y] . . . interim licensing of fuel cycle facilities, den[y] . . . interim licensing for use of mixed oxide fuel in reactors, except in small quantities for experimental purposes,” and publish the staff’s safeguards supplement.
In re Mixed Oxide Fuel, 7 N.R.C. 711, 714-16 (1978) (emphasis added).
The Commission also recognized that the President’s request was due substantial deference and added that there were strong indications of Congressional agreement:
In addition, it is significant that Congress as a body has not taken any action disagreeing with the President’s position on plutonium recycle in light water reactors.6 Over a year has passed since the President made the deferral of plutonium recycle a matter of national policy and no substantial Congressional opposition has appeared. Indeed, the Congress has in several instances supported the actions the President has taken to implement the Administration’s policy. The Department of Energy Act of 1978 — Civilian Applications, Pub.L.No. 95-288, 92 Stat. 47,7 provides $13 million for various activities at the Barnwell Nuclear Fuels Plants related to alternative fuel cycle technologies and the nation’s non-proliferation objectives, but “none of the authorized funds may be used for operations of the plant to process spent fuel from reactors.”8 Sections 101(20), 106.9 Congress has also explicitly supported the President’s alternative fuel cycle studies in Section 105 of the Nuclear Non-Proliferation Act of 1978:
The President shall take immediate initiatives to invite all nuclear supplier and recipient nations to reevaluate all aspects of the nuclear fuel cycle, with emphasis on alternatives to an economy based on the separation of pure plutonium or the presence of high enriched uranium, methods to deal with spent fuel storage, and methods to improve the safeguards for existing nuclear technology .... 10
In addition, Congress has strongly supported the non-proliferation goals of the Administration’s policy. Section 2 of the Nuclear Non-Proliferation Act of 1978 states the following policy:
The Congress finds and declares that the proliferation of nuclear explosive devices or of the direct capability to manufacture or otherwise acquire such devices poses a grave threat to the security interests of the United States and to continued international progress toward world peace and development. Recent events emphasize the urgency of this threat and the imperative need to increase the effectiveness of international safeguards and controls on peaceful nuclear activities to prevent proliferation. Accordingly, it is the policy of the United States to—
(a) actively pursue through international initiatives . . . the establishment of more effective international controls over the transfer and use of nuclear materials and equipment and nuclear technology for peaceful purposes in order to prevent proliferation....
*59Moreover, Section 3 states that
It is the purpose of this Act to promote the policies set forth above by—
(a) establishing a more effective framework for international cooperation ... to ensure that the worldwide development of peaceful nuclear activities and the export by any nation of nuclear materials and equipment and nuclear technology intended for use in peaceful nuclear activities do not contribute to proliferation . . .
Finally, Section 2(c) strongly endorses the Treaty on the Non-Proliferation of Nuclear Weapons, July 1, 1968, T.I.A.S. No. 6839, the sole function of which is to halt the proliferation of nuclear weapons.11
7 N.R.C. at 719-20.
It is appropriate at this point to take judicial notice of a diametrically opposing point of view expressed by President Reagan:
Nuclear Energy Policy
Statement Announcing a Series of Policy Initiatives. October 8, 1981
A more abundant, affordable, and secure energy future for all Americans is a critical element of this administration’s economic recovery program. While homeowners and business firms have shown remarkable ingenuity and resourcefulness in meeting their energy needs at lower cost through conservation, it is evident that sustained economic growth over the decades ahead will require additional energy supplies. This is particularly true of electricity, which will supply an increasing share of our energy.
If we are to meet this need for new energy supplies, we must move rapidly to eliminate unnecessary government barriers to efficient utilization of our abundant, economical resources of coal and uranium. It is equally vital that the utilities — investor-owned, public, and coops — be able to develop new generating capacity that will permit them to supply their customers at the lowest cost, be it coal, nuclear, hydro, or new technologies such as fuel cells.
One of the best potential sources of new electrical energy supplies in the coming decades is nuclear power. The U.S. has developed a strong technological base in the production of electricity from nuclear energy. Unfortunately, the Federal Government has created a regulatory environment that is forcing many utilities to rule out nuclear power as a source of new generating capacity, even when their consumers may face unnecessarily high electric rates as a result. Nuclear power has become entangled in a morass of regulations that do not enhance safety but that do cause extensive licensing delays and economic uncertainty. Government has also failed in meeting its responsibility to work with industry to develop an acceptable system for commercial waste disposal, which has further hampered nuclear power development.
To correct present government deficiencies and to enable nuclear power to make its essential contribution to our future energy needs, I am announcing today a series of policy initiatives:
(1) lam directing the Secretary of Energy to give immediate priority attention to recommending improvements in the nuclear regulatory and licensing process. I anticipate that the Chairman of the Nuclear Regulatory Commission will take steps to facilitate the licensing of plants under construction and those awaiting licenses. Consistent with public health and safety, we must remove unnecessary obstacles to deployment of the current generation of nuclear power reactors. The time involved to proceed from the planning stage to an operating license for new nuclear powerplants has more than doubled since the mid-1970’s and is presently some 10-14 years. This process must be streamlined, with the objective of shortening the time involved to 6-8 years, as is typical in some other countries.
(2) I am directing that government agencies proceed with the demonstration of breeder reactor technology, including completion of the Clinch River Breeder Reactor. This is essential to ensure our *60preparedness for longer-term nuclear power needs.
(3) I am lifting the indefinite ban which previous administrations placed on commercial reprocessing activities in the United States. In addition, we will pursue consistent, long-term policies concerning reprocessing of spent fuel from nuclear power reactors and eliminate regulatory impediments to commercial interest in this technology, while ensuring adequate safeguards.
It is important that the private sector take the lead in developing commercial reprocessing services. Thus I am also requesting the Director of the Office of Science and Technology Policy, working with the Secretary of Energy, to undertake a study of the feasibility of obtaining economical plutonium supplies for the Department of Energy by means of a competitive procurement. By encouraging private firms to supply fuel for the breeder program at a cost that does not exceed that of government-produced plutonium, we may be able to provide a stable market for private sector reprocessing and simultaneously reduce the funding needs of the U. S. breeder démonstration program.
(4) I am instructing the Secretary of Energy, working closely with industry and State governments, to proceed swiftly toward deployment of means of storing and disposing of commercial, high-level radioactive waste. We must take steps now to accomplish this objective and demonstrate to the public that problems associated with management of nuclear waste can be resolved.
(5) I recognize that some of the problems besetting the nuclear option are of a deep-seated nature and may not be quickly resolved. Therefore, I am directing the Secretary of Energy and the Director of the Office of Science and Technology Policy to meet with representatives from the universities, private industry, and the utilities, and requesting them to report to me on the obstacles which stand in the way of increased use of nuclear energy and the steps needed to overcome them in order to assure the continued availability of nuclear power to meet America’s future energy needs, not later than September 30, 1982.
Eliminating the regulatory problems that have burdened nuclear power will be of little use if the utility sector cannot raise the capital necessary to fund construction of new generating facilities. We have already taken significant steps to improve the climate for capital formation with the passage of my program for economic recovery. The tax bill contains substantial incentives designed to attract new capital into industry.
Safe commercial nuclear power can help meet America’s future energy needs. The policies and actions that I am announcing today will permit a revitalization of the U. S. industry’s efforts to develop nuclear power. In this way, native American genius, not arbitrary Federal policy, will be free to provide for our energy future.
17 Weekly Comp, of Pres. Doc. 1101-02 (Oct. 12, 1981).
With this recent Presidential debate in mind, I turn to the possible consequences for this case of President Reagan’s announcement on October 8, 1981, which removed all restrictions on nuclear fuel reprocessing.
B. The NRC Options
It is, of course, obvious that the NRC has many decisions yet to make in relation to the back-end of the nuclear fuel cycle. It has yet to license a private corporation to build a retrievable surface storage facility, if, indeed, such will ever be built. It has yet to choose a site for its proposed long-term repository. And it has yet to license a commercial reprocessing facility, although it now appears that such will be licensed. Finally, the NRC has not decided whether the reactor fuel to be produced from the reprocessing of spent fuel will be fabricated from 1) uranium only into enriched uranium fuel, 2) uranium and plutonium into mixed oxide fuel, or 3) some other combination of elements not presently described.
*61The NRC is seeking judicial sanction for the conclusion in the S-3 Tables that any releases that may be produced as a result of the exercise of any of these options can lawfully be termed “zero” or can lawfully be described as “umbrellaed” by the insignificant release figures set forth in the S-3 and S-3A Tables. I see no way by which we can in advance give such sanction consistent with the Administrative Procedure Act and the health and safety requirements of the Atomic Energy Act.
1. The Uranium-only Recycle Option. The NRC may decide to adopt a uranium-only recycle plan. This plan contemplates the reprocessing of spent fuel rods and the production of highly enriched uranium. Nonetheless, the extraction and storage of the plutonium remains a problem. The NRC described the option when it promulgated the final version of Table S-3 in August 1979:
The fuel cycle was to be taken to include . . . reprocessing of spent fuel for purposes other than recycle of plutonium, with follow-on interim and/or long-term storage or disposal of plutonium and wastes from reprocessing, with plutonium either separated from or included with the wastes.
44 Fed.Reg. 45362, 45366 (Aug. 2, 1979) (footnote omitted).
The NRC also envisions that
[t]he uranium would be converted to uranium hexafluoride for recycling at an enrichment plant. The plutonium, still containing about five percent of the fission products to deter diversion, would be converted to plutonium oxide and packaged for disposal in a Federal waste repository. The high-level liquid waste (HLLW) containing the bulk of the fission products, would be stored up to five years in tanks and then calcined and formed into glass for repository disposal.
Id. at 45369.
This option would, of course, reduce the hazards involved in extracting, storing and transporting plutonium dioxide. This option, however, raises by many degrees the uncertainties involved in the ultimate disposal of the wastes from the spent fuel rods. The remaining plutonium must be handled and guarded at the retrievable surface storage facility. Also, it must be shipped to the long-term waste disposal site and handled and guarded there until the repository is sealed.
Once in the repository, the heat and bulk of the decaying plutonium would increase the thermal stresses on the geologic medium that constitutes the repository. U. S. Nuclear Regulatory Commission, Environmental Survey of the Reprocessing and Waste Management Portions of the LWR Fuel Cycle, (NUREG-0116) 4-103 (Oct. 1976). Finally, the existence of large amounts of plutonium buried underground would add to the chance that shifting geologic conditions would cause the inadvertent formation of a critical mass and consequent nuclear explosion. Id. at 4-103, 4-104.
Continuation of the no-recycle option would, of course, raise somewhat similar problems.
2. The Uranium-Plutonium Recycle Option. The possibility that the NRC might read an affirmance of Table S-3 as justification to proceed with the recycling of plutonium and the fabrication of mixed oxide fuel is particularly troubling. Initially, it is important to note that the NRC has deliberately excluded the risks of nuclear fuel diversion from Table S-3 because it believes those risks are negligible, not because it believes these risks are beyond the scope of this rulemaking. This has been evident throughout the course of the proceeding. For example, when the NRC published its assessment of the risks of the back-end of the light water reactor fuel cycle, NUREG0116, supra, public commentators questioned the exclusion of the risks of accident and sabotage from the interim version of Table S-3. The NRC answered that it excluded those risks because it thought they were negligible:
Comment (CEC B-139): It is unclear whether sabotage impacts have been included in the revised Table S-3.
Response: The risks of sabotage are judged to be negligible and therefore no *62entry is given in Table 2.10 in NUREG0116. This judgment is based on the required protective measures of physical security, intrinsic protective features of containment systems required by regulations, and assessment of radiological consequences of given successful acts of sabotage. The NRC is conducting consequence studies of postulated sabotage events in all fuel-cycle facilities to assess current regulations and possible regulatory improvements as part of its continuing review of the adequacy of present safeguard measures.
U. S. Nuclear Regulatory Commission, Public Comments and Task Force Responses Regarding the Environmental Survey of the Reprocessing and Waste Management Portions of the LWR Fuel Cycle 3-96 (March 1977) (NUREG-0216).11 After additional hearings on the proposed rule, the NRC Hearing Board expressed the same conclusion:
As a result of its analyses, the staff concluded that the impacts from accidents would be only a few per cent of the impacts from normal operation and did not include them in the values in Table S-3....
After considering the impacts of sabotage, the staff concluded that, overall, those impacts would be small in comparison with the impacts of normal operation and need not be explicitly included in Table S-3. This conclusion was based on the fact that acts of sabotage tend to resemble accidents. Measures provided to mitigate the effects of accidents should also mitigate the effects of sabotage. Moreover, the Commission now requires that substantial measures be taken to protect nuclear facilities from sabotage, including measures to protect against groups of highly motivated, well-trained saboteurs. We find the staff’s conclusion reasonable.
Conclusions and Recommendations of the Hearing Board Regarding the Environmental Effects of the Uranium Fuel Cycle, Docket No. RM 50-3 44-45 (October 26, 1978); 4 J.A.S. 1466. The NRC adopted the recommendations of the Hearing Board when it promulgated the final version of Table S-3. See 44 Fed.Reg. 45362, 45367 (Aug. 2, 1979).12
Second, the nuclear regulatory agencies of this country have always contemplated that plutonium recycling would be a part of any fuel reprocessing regime. At 10 C.F.R. § 70.4(m) (1981), which deals with the licensing of reprocessing facilities, we find this definition:
“Special nuclear material” means (1) plutonium, uranium 233, uranium enriched in the isotope 233 or in the isotope 235, and any other material which the Commission, pursuant to the provisions of section 51 of the act, determines to be special nuclear material ....
And, the NRC’s new regulations dealing with the transportation of nuclear materials also envision the transportation of plutoni*63um.13 Section 73.2(aa) defines “[strategic special nuclear material” thusly:
“Strategic special nuclear material” means uranium-235 (contained in uranium enriched to 20 percent or more in the U-235 isotope), uranium-233, or plutonium.
10 C.F.R. § 73.2(aa) (1981).
Accordingly, the NRC could read our affirmance of Table S-3 to support the conclusion that plutonium recycling can be accomplished with zero release and without endangering the public. The NRC believes that “the environmental impacts from reprocessing and related management activities are nearly identical for the recycling of uranium and plutonium and for the recycling of uranium only, as fuel in nuclear power reactors.” 46 Fed.Reg. 15154, 15161 (Mar. 4, 1981).14
The section of this opinion which follows concerns possible employment of the plutonium option. The NRC may avoid this option because it is clearly the “worst case” in terms of hazards. Or it may adopt it because it appears to be the best option in terms of the economics of the nuclear power industry.
C. The Plutonium Hazards
I fear that when the history of this century is written that the greatest debacle of our nation will be seen ... to be . . . our creation of vast armadas of plutonium, whose safe containment will represent a major precondition for human survival, not for a few decades or hundreds of years, but for thousands of years more than human civilization has so far existed.
James D. Watson,
Nobel Laureate, Medicine
On March 13,1975, Senator Stewart Symington addressed the United States Senate on “Controlling the Cancer of Nuclear Proliferation.” His concerns extended well beyond the boundaries of this country and the scope of this case. The following paragraphs, however, serve to illustrate the nature of the problems generated by the recent Presidential decision to lift the prohibition against the NRC’s plan to reprocess nuclear waste so as to produce plutonium:
Only 10 to 20 pounds of plutonium is necessary for making an atomic bomb with the same destructive power as the one dropped on Hiroshima. Yet today, close to 90,000 pounds of plutonium have been produced by commercial nuclear power plants — not including amounts used for the nuclear weapons programs of the Big Powers.
Five years from now some half-million pounds of plutonium may be commercially available throughout the world. As the amount of plutonium grows internationally so does the risk of its diversion from peaceful uses to weapons-development programs.
121 Cong.Rec. 6437 (1975).
In present posture, this case requires us to compare the NRC’s flat assurance of zero or insignificant release of effluents as stated in Table S-3 with the incredible risks involved in transporting large quantities of plutonium all over the United States.
Plutonium is one of the most deadly substances known to humankind. It is toxic to human beings if a microscopic particle is breathed into the lungs.
A small quantity of Plutonium-239 smoke deposited in the lung — about one ten-thousandths of an ounce, would kill a person through radiological destruction of lung tissue. A quantity smaller than one one-millionth of an ounce would give rise to a substantial risk of lung cancer . . . Plutonium-239 ... is at least 20,000 times *64more toxic than cobra venom or potassium cyanide and ten times more toxic than heroin or modern nerve gas.
Ford Foundation Energy Policy Project, A Time to Choose 210 (1974).
After reprocessing, plutonium is reduced to oxide powders for shipment to fuel fabrication plants:
NRC regulations prohibit the shipment of plutonium compounds, apart from small quantities, in liquid form. Consequently, the plutonium is converted into the solid dioxide [Pu02] prior to shipment. Plutonium Oxalate is generally precipitated from the nitrate solution; the oxalate is separated, dried and calcined to form the dioxide. The product is ground and screened to yield a powder of the desired size.
S. Glasstone & W. Jordan, Nuclear Power and Its Environmental Effects 234 (1980).
This powder would be extremely dangerous if dispersed into the air as it could be by accident, theft or sabotage:
Both in storage and in transit, separated plutonium requires the most careful . . . measures . .. against theft by non-state actors. Owing to the serious inhalation hazard of powdered Pu02, currently the form in which most nonmilitary stocks are held, equally strict measures for protection of public health and safety are . . . required. These precautions ... are to be applied from the time the material emerges from the reprocessing plant — at least until ... it is fabricated into reload fuel assemblies, thereby rendering accounting easier and theft more difficult.
G. I. Rochlin, Plutonium, Power and Politics 90 (1979).
Indeed, even the fabricated mixed oxide fuel would be an attractive target for terrorists. The Nuclear Energy Policy Study Group explained why:
Shipments of fresh mixed oxide fuel for the yearly reloading of a single reactor would contain enough plutonium for fifty nuclear weapons. Separation of plutonium ... would be much easier than from irradiated fuel, involving only simple chemical operations and little radiation hazard.
Nuclear Energy Policy Study Group, Nuclear Power: Issues & Choices 331 (1977) (hereafter NEPSG).
The Study Group pointed out the general, worldwide increase in terrorism over the last decade:
The past few years have seen an upsurge in the size, sophistication and capabilities of terrorist groups around the world. Terrorist activities offer strongly motivated political or dissident groups a way to dramatize their causes and influence their adversaries .. . the worldwide development of civilian nuclear power provides additional opportunities for terrorists — whether revolutionaries, nationalists, dissidents or criminally motivated— to employ nuclear energy as a weapon... .
Terrorists might choose the nuclear industry as a target to exploit the mystique that surrounds nuclear weapons. The threat of nuclear terrorism may be used to extort money, secure the release of prisoners or publicize a particular cause.
Id. at 301.
The simplest form of sabotage would probably consist of blowing up a truck that was transporting plutonium from the reprocessing plant to a fuel fabrication plant or nuclear reactor. A high explosive charge fracturing the shipping container and discharging the oxide powder into the air could occasion disaster to any area downwind.
In addition, plutonium is highly explosive and extremely powerful. It has been estimated that 10 pounds of plutonium fashioned into a bomb could kill 100,000 people in a densely populated city. It has also been estimated by one knowledgeable expert that many thousands of persons in the United States have the knowledge required to design a workable bomb.
With only presently-licensed nuclear reactors taken into account, the NRC’s plan for reprocessing nuclear waste would produce 200-300 kilograms (440-660 lbs.) of plutoni*65um from each of 71 reactors, or over 35,000 pounds of this hazardous substance.
On this score the Nuclear Energy Policy Group Study said:
[T]he weakest link in security would be transportation .... The [1973] regulatory changes [to reduce the possibility of employee diversion] did not eliminate the possibility that a small armed group could successfully hold up a truck and hijack its load of nuclear materials .... Transport procedures are not vastly different than those employed by armored car companies and other carriers that ship high-value cargo. Some criminal and terrorist groups have already demonstrated capabilities to defeat such precautions.
Id. at 304.
Lovins, Lovins & Ross summarized the present problems of preventing plutonium diversion:
Political arrangements for safeguards must rest on technical measures for materials accounting and for physical security. The former measures are so imprecise and post hoc that they cannot, even in principle, provide reasonable assurance that many bombs’ worth of plutonium per year are not being removed from a good-sized reprocessing plant. Primary reliance must therefore be placed on physical security measures to limit access . . . These measures must forestall well equipped groups perhaps including senior insiders acting in concert with the host government or a faction of it. Even modestly effective measures would be costly, fallible and intrusive.
Lovins, Lovins & Ross, Nuclear Power and Nuclear Bombs 58 Foreign Affairs 1137, 1145 (Summer 1980).
The Nuclear Energy Policy Study Group summed up, “If plutonium and highly enriched uranium become increasingly available, the route to national or subnational explosives capabilities will become easier and quicker and safeguards will lose much of their ability to provide reassurance.” NEPSG, supra at 294.
Commentary on these hazards could be extended indefinitely, since the literature concerning the hazards of plutonium is almost endless.
If this were a wholly peaceful world, or if ours were a wholly peaceful country, the NRC’s plan to recycle plutonium and fabricate and burn it as a fuel would be subject to much less objection. In fact, however, the United States confronts many hostile powers, some with vast wealth and the consequent ability to train and arm desperadoes or to bribe and corrupt personnel connected with either private or government aspects of the nuclear cycle. In addition, past history (and in particular the last two decades) teaches that domestic tranquility is a constitutional goal rather than a reality in our land. Organized crime as we know it in our country is capable of hijacking a truck with armed guards, and terrorist groups capable of stopping and robbing Brinks trucks of millions of dollars have demonstrated a somewhat like capability. See NEPSG, supra at 304.
The routine use of plutonium for nuclear power under the Commission’s present plans will call for a private corporation to reprocess the spent fuel rods from all nuclear reactors, thus producing plutonium. By Commission regulation, that corporation is required to organize a private army larger than has ever existed in this country and to deploy it at the reprocessing plant and on U. S. highways all over the land. See 10 C.F.R. § 73.25 (1981). Nevertheless, under the Commission’s present plans, it will be impossible to guarantee the security of plutonium shipments from an. intelligently organized surprise attack by substantial forces on a lonely stretch of highway in a remote area. Under the Commission’s present plans, the ultimate reliance is upon a call to local law enforcement. See 10 C.F.R. § 73.26(c)(2)(iii)(A), (i)(6) (1981). In the event of attack at a lonely milepost, the call might produce one deputy sheriff. Protecting plutonium shipments is a task appropriate for the military forces of the United States. No such authorization has been sought from or considered by Congress.
*66I assume that the theft by stealth or force of sufficient plutonium to fabricate a bomb • and its subsequent employment by threats or fact of explosion would constitute a “release.” In my view, the threat of such a “release” of plutonium if it is fabricated, stored and shipped all over the nation, is anything but “insignificant.” The NCR’s finding to the contrary constitutes “a clear error of judgment.” See Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 823, 28 L.Ed.2d 136 (1971). As indicated earlier, I therefore join in holding the affected portions of Table S-3 to be arbitrary and capricious.
The fundamental law that launched this nation on the road to peaceful use of nuclear power was the Atomic Energy Act of 1946, eh. 724, 60 Stat. 755 (1946). That Act was superseded by the Atomic Energy Act of 1954, ch. 1073, 68 Stat. 921 (1954) (codified in 42 U.S.C. § 2011 et seq.). The 1954 Act repeatedly requires the “protect[ion] of the health and safety of the public.” 42 U.S.C. §§ 2012(d), (e), 2133(b), 2232(a) (1976); see id. § 2133(d).
After much acquaintance with this case, I have come to believe that Table S-3, inasmuch as it requires licensing boards to accept that only insignificant releases will result from transporting plutonium all over the nation to 71 (potentially 190) reactors is “otherwise not in accordance with law.”
COST
Appellants New York and Wisconsin contend that the figure of 10% of construction costs per nuclear reactor determined by the NRC as the cost of nuclear waste disposal is arbitrary and capricious. I agree.
As I understand this record, the 10% of cost (building and operating) per nuclear reactor represents the Commission’s determination of the economic cost of all nuclear waste disposal. Thus, it underpins the Commission’s adoption of the S-3 Tables and is properly before this panel for decision at this time.
This record indicates that the NRC now assumes that all nuclear waste is to be under the guardianship of the federal government practically in perpetuity and that all nuclear waste costs above the 10% referred to above for each reactor will be borne by the United States. Under ordinary standards of accountability (absent subsidy determined by law), appropriate charges for maintaining facilities owned and operated by government for private industry would be determined and charged back to the private industry making use of them. The cost of maintenance, surveillance and guarding the public from toxic waste at all such facilities will continue from somewhere between 10,000 and 250,-000 years. The currently estimated life of each reactor is 30 years. The nuclear power industry may not survive its problems. Its waste will, for all practical purposes, survive forever. This policy decision potentially involves an open-ended subsidy of great dimension. Constitutionally, it can only be made as a matter of legislation by Congress and the President. It certainly should not be made sub silentio by an administrative agency and routinely endorsed by the federal courts.
I dissent from my colleagues’ treatment of the economic issue.' I would remand the issue concerning the economic feasibility of the S-3 Tables to the Commission for further consideration. Public discussion of these costs at licensing hearings should not be terminated until much more knowledge about them has been developed.
Outline
I. Analysis_____________________________ 517
II. Background__________________________ 518
A. The Challenged Rules_______________ 518
B. Vermont Yankee I__________________ 519
C. Vermont Yankee II_________________ 520
III. Our Task on Remand____________________ 524
A. The Agency’s Decision ______________ 524
B. Contemporaneous Explanation________ 525
1. The Commission’s Purpose________ 525
2. The Commission’s Methodology____528
3. The Commission’s Conclusions______529
C. The Appropriate Standard of Review___530
D. The Standard of Review Applied to the Agency’s Decision__________________ 533
1. “Scope of Authority" ____________ 533
2. “Observance of Procedure" _______535
3. “Reasoned Decisionmaking"_______537
IV. Living With Uncertainty _______________540
V. Conclusion___________________________ 545
Honorable George Clifton Edwards, Jr., Chief Judge, U. S. Court of Appeals for the Sixth Circuit, sitting by designation.
. 5 U.S.C. § 706(2)(A) (1976).
. I choose the term “insignificant” to describe the impact of the many release figures set forth in the three S-3 Tables because the NRC’s staff has equated them with the impact on human health of “a few puffs on a cigarette [or] a few sips of wine.” 46 Fed.Reg. 15154, 15167 (March 4, 1981).
. Each 1,000 MWe Light Water Reactor produces 200-250 kilograms of plutonium a year. Speth, Tamplin & Cochran, Plutonium Recycle: The Fateful Step, Bulletin of Atomic Scientists 15, 16 (November 1974). Two hundred-250 kgs. times 2.2 lbs/kg. equals 440-550 lbs. per reactor per year. If we assume a figure of 70 operating reactors, 70 times 440-550 equals 30,800-38,500. If we assume 170 projected reactors, 170 times 440-550 equals 74,800-93,-500.
. J. McPhee, The Curve of Binding Energy 124 (1974) (figure from interviews with physicist Theodore Taylor). See H. Feiveson & T. Taylor, Nuclear Proliferation 136-144 (1977); T. Taylor & M. Willrich, Nuclear Theft; Risks & Safeguards 12-20 (1974).
. We, of course, are aware that on March 4, 1981, long after this case had been submitted to this court and after respondent had made Table S-3 effective in licensing of additional nuclear power reactors, the Commission did release, by printing in the Federal Register, a proposed “Narrative." 46 Fed.Reg. 15154 (Mar. 4, 1981). I understand this term to represent a statement in explanation of the terse and difficult language of Table S-3. Because the Narrative, at present, is proposed and has not been finally adopted by the NRC, it is not ripe for judicial review. However, the Narrative represents the fullest expression of the NRC’s most recent findings and thinking concerning the matters comprehended by Table S-3. For this reason, it has been deemed part of the record and considered.
I am unable to understand why this “explanation” (incomplete as it is) was not made available long ago and in ample time for our .present decision. 1 have proposed to my colleagues on this panel the delay of the release of our decision pending completion of rulemaking on the Narrative, and, understandably, they feel such delay would be inadvisable from the point of view of their court. It seems likely to me, however, that with the issuance of the opinions of this case and the anticipated application for writ of certiorari in the Supreme Court, the Commission will proceed to complete its rulemaking with such modifications of the Narrative, as it deems advisable and will then argue that the newly adopted Narrative requires another remand of the case on a claim that the subsequently adopted Narrative has rendered this panel’s action moot.
Additionally, we are informed that the NRC has initiated a “waste confidence” proceeding which may update Table S-3. See 44 Fed.Reg. 45362, 45363 (Aug. 2, 1979).
Appreciating as I do the complexities and hazards with which respondent is required to deal, I am still compelled to wonder whether the “moving target” aspects of this case have all been required by uncontrollable circumstances or have been in part contrived to frustrate effective judicial review. At oral argument of this case, when a question concerning the “moving target” issue was posed by the author of this opinion, the response was, “Judge, we have able lawyers.”
. Public information furnished by telephone by the NRC, August 26, 1981.
. This case and opinion pertain, of course, to commercial nuclear power. The result in this case, therefore, would not apply to the licensing, construction and operation of military and experimental reprocessing plants.
. Other proposed plans for high-level waste disposal include a) deep burial in geologic granite formations, b) burial in clay beds in the bottom of the Pacific Ocean, and c) dispatching canisters of high-level waste into -permanent orbit around other planets.
. Table 2.10 in NUREG-0116 became part of the basis for Table S-3a in the “Proposed Narrative” of March 1981. See 46 Fed.Reg. 15154 (March 4. 1981).
. I recognize that the closely related Table S-4 allegedly measures the impacts of transportation “of cold fuel to a reactor and of irradiated fuel and radioactive wastes from a reactor.” 10 C.F.R. § 51.20(e) n.l. Thus, it may be contended that the risk of plutonium diversion during transportation is not before us for review. Table S^4, however, measures the risk of transporting cold fuel to and radioactive wastes from the reactor, not the larger risks of transporting Pu02 from reprocessing plants to fuel fabrication plants.
Additionally, the relation of S-3 and S-4 is exceedingly unclear. Table S-3 also contains an entry for transportation effects, and Table S — 4 purports to measure “Normal Conditions of Transport” and “Accidents in Transport,” not the risks of intentional diversion of nuclear fuel. The NRC has never attempted to clarify this interrelationship. The risks of plutonium recycling are too great to leave a loophole through which the NRC’s “able lawyers” could evade review of these hazards.
Finally, the material from NUREG-0216 and the Hearing Board Recommendations, supra, indicates that the NRC did not include the risk of diversion in Table S-3 because it thought the risks insignificant, not because it thought the problem outside the scope of Table S-3.
. The NRC promulgated these regulations in November 1979, well after they terminated the GESMO proceeding. See 44 Fed.Reg. 68187 (Nov. 28, 1979).
. Also instructive is an alternative version of S-3 which the NRC promulgated during the pendency of the GESMO proceeding. This version of Table S-3 purported to define the impacts of a mixed oxide, fuel cycle. Like its successor, this version of Table S-3 neglected the risk of accidental or intentional diversion of plutonium. See 41 Fed.Reg. 40506, 40509-510 (Sept. 20, 1976).
. 444 F.2d 841 (D.C.Cir. 1970), cert. denied, 403 U.S. 923, 91 S.Ct. 2229, 29 L.Ed.2d 701 (1971).
. Id. at 851-53. See Pikes Peak Broadcasting Co. v. FCC, 422 F.2d 671, 682 (D.C.Cir.), cert. denied, 395 U.S. 979, 89 S.Ct. 2134, 23 L.Ed.2d 767 (1969). See also infra pp. 532-533.
. NRDC v. NRC, 547 F.2d 633, 644 (D.C.Cir. 1976) (emphasis added) [hereinafter Vermont Yankee I maj. op.].
. Vermont Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519, 98 S.Ct. 1197, 55 L.Ed.2d 460 (1978) [hereinafter Vermont Yankee II].
. I express here only my disagreement with the analysis found in sections IV.A.-C of Judge Bazelon’s opinion [hereinafter Bazelon op.] and the disposition found on page 69.
I agree with the analysis under Section IV.D. of Judge Bazelon’s opinion, affirming the Commission’s decisions regarding economic feasibility on the ground that they were not arbitrary and capricious. I question only Judge Bazelon’s decision not to apply the same form of the arbitrary and capricious test to the other issues in this case. See infra p. 540.
. Bazelon op. at 468-475.
. NRDC v. NRC, 547 F.2d 633, 658 (D.C.Cir.1976) (Tamm, J., concurring in result) [hereinafter Vermont Yankee I concurrence].
. Licensing and Regulatory Policy and Procedures for Environmental Protection, 10 C.F.R. Part 51 (1980), 44 Fed.Reg. 45,362, 45,369 (1979) (emphasis added) [hereinafter Final Rule Statement Basis and Purpose].
. 42 U.S.C. § 4332(2)(C) (1976).