Environmental Issues > Nuclear Energy, Nonproliferation, and Disarmament Main Page > All Nuclear Energy, Nonproliferation, and Disarmament Documents

Report page


On 31 July 1997 the U.S. Department of Energy (DOE) announced that its Office of Defense Programs was awarding multi-million dollar, multi-year grants to five universities to establish research "Centers of Excellence" in support of the "Accelerated Strategic Computing Initiative" (ASCI), a key component of DOE's "Stockpile Stewardship and Management Program" (SSMP) for maintaining and improving U.S. nuclear weapons and design expertise. A principle objective of the SSMP is to develop a computer-simulation-based "virtual testing" capability to compensate for the loss of nuclear explosive tests under the Comprehensive Test Ban Treaty (CTBT) now pending before the Senate.

These grants are the initial phase of the DOE nuclear weapons establishment calls its Academic Strategic Alliances Program. The program was first publicly discussed by the DOE in November 1996. Ernest Moniz, then Associate Director of the White House Office of Science and Technology Policy and now Undersecretary of Energy, stated: "The DOE ASCI program is strategically leveraging the administration's scientific and technology instruments at U.S. universities to substantially advance our ability to numerically simulate scientific problems of national significance and unprecedented scale."[2] Forty-eight universities initially sought funding and 20 actually submitted grant proposals for review by the DOE Office of Defense Programs and its nuclear weapons laboratories. The funding level for each center is about $20 million over the next 5 years with the possibility of an additional five years of support -- a total 10-year program budget of $200 million. The money comes from the "Weapons Activities" account of the DOE Defense Programs budget. Table 1.1 lists the principal universities and participating departments, co-participants, and the civilian and nuclear weapons research focus of each center. At least fourteen universities and colleges are now participating in the Academic Strategic Alliances Program, and participation continues to grow as a consequence of DOE's 7 November 1997 request for further research proposals from the academic community and industry.

Table 1.1: Information on the five DOE Academic Strategic Alliances Program Centers of Excellence.

UniversityResearch CenterParticipating DepartmentsCo-ParticipantsCivil Applications Nuclear Weapons Application
California Institute of Technology"A Facility for Simulating the Dynamic Response of Materials" Applied Mathematics, Chemistry, Computer Science, Materials Science, Mechanical Engineering, PhysicsBrown University, Carnegie Institution of Washington, Indiana University, University of Illinois at Urbana-Champaign, University of Southern California, University of Tennessee at KnoxvilleCivilian practices that employ high explosives, material design, and geophysics (e.g oil, gas, mining industries) Improved simulations of high-explosive detonation, ignition, and shock compression of heavy metals, hydrodynamic instabilities
Stanford University"Center for Integrated Turbulence Simulations" Aeronautics and Astronautics, Chemical Engineering, Computer Science, Electrical Engineering, Mechanical EngineeringStanford/NASA Center for Turbulence Research (CTR) Simulation technology suitable for the design of gas turbine engines Compressible flow computations, turbulence, and transport modeling
University of Chicago"Center on Astrophysical Thermonuclear Flashes" Astronomy and Astrophysics, Chemistry, Computer Science, Mathematics, PhysicsArgonne National Laboratory, NASA/Goddard Space Flight Center, Rensselaer Polytechnic InstituteAstrophysical thermonuclear flashesFusion ignition, detonation, turbulent mixing of complex multi-component fluids and other materials
University of Illinois at Urbana-Champaign"Center for Simulation of Advanced Rockets" Aeronautical and Astronautical Engineering, Astronomy, Chemistry, Civil Engineering, Computer Science, Materials Science and Engineering, Mechanical and Industrial Engineering, Nuclear Engineering, Physics, Theoretical and Applied MechanicsNational Center for Supercomputing Applications (NCSA) Detailed, whole-system simulation of solid propellant rockets Aerospace and defense applications such as design safety and reliability testing of new solid rocketsShock physics and quantum chemistry of energetic materials and the aging and damage of components, turbulence
University of Utah"Center for Simulation of Accidental Fires and Explosions" Chemical and Fuels Engineering, Chemistry, Computer Science, Materials Science and Engineering, Mechanical EngineeringBrigham Young University, Thiokol Corporation, Utah State UniversityIndustrial chemical fires, the handling and transport of highly flammable materials, terrorist attacks or car crashesNumerical simulation of accidental fires and explosions involving high explosives, chemistry and physics of high explosive detonations
The data in this table was obtained from the official Academic Strategic Alliances Program website (http://www.llnl.gov/asci-alliances/), university research proposals, and other documentation. The Laboratory for Laser Energetics at the University of Rochester (Rochester, NY) plays a key role in Stockpile Stewardship, but its contributions generally fall within the Inertial Confinement Fusion Program. In June 1997 the "Institute for Shock Physics" was established at Washington State University (Pullman, WA) with Defense Program funds separate from the Academic Strategic Alliances Program.

A. Background

The United States has been engaged in the design, testing and manufacture of nuclear weapons since 1942. Since 1977 the DOE has been the Federal Agency responsible for these activities.[3] Within DOE the Office of Defense Programs, directed since 1993 by Assistant Secretary Victor H. Reis, has management responsibility for research, development, testing, maintenance, production and disassembly of nuclear weapons. The Office of Defense Programs oversees two government-owned but contractor-operated nuclear weapon design laboratories, Los Alamos National Laboratory (LANL) in New Mexico, and Lawrence Livermore National Laboratory (LLNL) in California, which design and test the nuclear explosive portion of nuclear missile warhead and bomb systems, and also conduct a broad range of basic research and technology development in areas applicable to both conventional and nuclear weapon systems. A third nuclear weapons laboratory, Sandia National Laboratories (SNL) with branches in both California and New Mexico, performs design, engineering, and in some cases production of non-nuclear components (e.g., radars, fuzes, batteries, microcircuits, casings) of deliverable nuclear warhead and bomb systems for the military stockpile, develops and operates radiation facilities for simulating nuclear weapon effects and conducting nuclear weapon physics experiments, and conducts warhead reliability and "survivability" (i.e., "crash and burn") testing of nuclear warhead systems and components.

Beginning with the 16 July 1945 Trinity test, nuclear explosive testing served as the primary means of confirming nuclear design computations and judgments, and as the principle basis for "certifying" the nuclear explosive performance of new weapons for entry into the war reserve stockpile. Between 1945 and 31 December 1992 the United States conducted 1149 detonations of nuclear explosive devices, of which all but 35 were explicitly related to the design, performance, safety, or effects of nuclear weapons.[4]

In the United States both the advent of the computer and subsequent advances in supercomputing have historically been propelled by the technical demands of nuclear weapons research. In the 1960s a relationship was established between the national nuclear weapons laboratories and the U.S. computer industry to develop the first generation of supercomputers, culminating in the Cray models. Over three decades explosive nuclear testing and computer modeling became highly interdependent research tools central to the work of designing, producing, and evaluating the rapidly-evolving Cold War U.S. nuclear stockpile. Between 1945 and 1992, the United States utilized about 21 successive generations of supercomputers.

After almost a decade of negotiations the original START treaty was signed on 31 July 1991 in Moscow, reducing nuclear weapons mounted on long range delivery systems by roughly one third, to a common ceiling of 6,000 "accountable" warheads (both achievable and actual U.S. nuclear force loadings were considerably higher). On 27 September 1991 President Bush announced a series of arms control and disarmament initiatives intended to accelerate the adjustment of both the U.S. and Russian nuclear postures to the end of the Cold War, including the removal of nuclear weapons from the arsenals of the U.S. Army, Marine Corps, and the Navy's surface fleet.

The combined effect of these two measures has been to reduce the total stockpile of operational U.S. nuclear weapons by roughly half, from 22,000 in 1990 to about 10,500 today.[5] In an effort to head off a Congressionally imposed test moratorium, President Bush directed in July 1992 that "the purpose of all U.S. nuclear tests of our weapons will henceforth be for the safety and reliability of our deterrent forces," apparently but not explicitly ruling out the use of nuclear explosive tests to certify new weapons and/or future modifications to the stockpile undertaken for the primary purpose of improving military combat effectiveness.[6]

In September 1992, the Congress of the United States established a temporary moratorium on U.S. underground nuclear tests, urged resumption of negotiations on a multilateral test ban treaty, and directed that no U.S. nuclear tests could be conducted after 30 September 1996 unless a foreign state conducts a nuclear test after this date. Presidents Yeltsin and Bush signed the START II treaty on 3 January 1993, providing for the further reduction of warheads mounted on deployed long-range missile and bomber delivery systems to 3500 weapons. In July 1993 President Clinton extended the U.S. nuclear test moratorium and multilateral test ban negotiations began in Geneva in January 1994.

In September 1994, the Clinton Administration announced the results of a lengthy Nuclear Posture Review, including the determination that, while it did not foresee a military requirement for "new-design nuclear warhead production," nuclear forces would remain a "cornerstone" of the U.S. defense posture for the foreseeable future, and thus DOE was directed to "maintain capability to design, fabricate, and certify new warheads," President Clinton signed the Comprehensive Test Ban Treaty on 24 September 1996.

These various arms control agreements and initiatives required DOE to "reconfigure" and "downsize" its complex for maintaining the U.S. nuclear weapons stockpile while still meeting the Pentagon's reduced but nonetheless evolving "military requirements" for nuclear weapons. In May 1995 the Department published a brief report, entitled "Stockpile Stewardship and Management Program," that outlined significantly revised strategies for carrying out DOE's nuclear weapons missions. This new program plan stated that "the primary goal of the Stockpile Stewardship and Management Program (SSMP) is to provide confidence in the safety, security and reliability of the U.S. stockpile to ensure the effectiveness of the U.S. nuclear deterrent while simultaneously supporting U.S. arms-control and nonproliferation policy."[7] Pending the elimination of nuclear arms, few dispute the need for a program to ensure the safety and security of a reduced nuclear stockpile. A less robust consensus supports the view that a high level of confidence should be maintained in the "military effectiveness" of prospective U.S. nuclear attacks, or in the "reliability" (i.e. achievement of nominal explosive performance) of discrete nuclear devices (as opposed to a sufficiently destructive capability in the overall force). And perhaps even less agreement surrounds the issue of whether DOE's chosen approach to the SSMP actually "supports" rather than hinders the achievement of U.S. arms control and non-proliferation policy goals.

B. Stockpile Stewardship and Nuclear Deterrence Strategy

Now fully evolved at an annual cost of $4.5 billion, the SSMP is implementing the DOE's revised strategies for continuing the research, design, development, testing, modification, production, and maintenance of U.S. nuclear weapons without reliance on additional underground nuclear explosive tests. At this level of expenditure, as shown in Figure 1.1, the SSMP significantly exceeds the $3.7 billion average level of annual expenditure on comparable activities throughout the Cold War period.

Figure 1.1: U.S. DOE Nuclear Weapons Activities -- Annual Budget Authority. (Constant FY1997 dollars in millions). 1948-1995: Weapons Research, Development, Testing and Production; 1996-2002: Science Based "Stockpile Stewardship" and "Stockpile Management." Source: Brookings Nuclear Weapons Cost Study and DOE.

Thus there is an unavoidable mismatch between the "military requirements" of the prevailing nuclear posture -- such as retaining or developing the ability to "defeat" deeply buried nuclear/biological/chemical targets -- and the limitations that a ban on testing would ostensibly impose on development of new or modified weapons to meet these requirements. Rather than rescind the underlying U.S. nuclear strategy, the Clinton Administration has sought to have it both ways, trumpeting the restrictive effect of the CTBT on the nuclear weapon programs of other nations while developing a massive 15 year, $67 billion "Stockpile Stewardship and Management Program" to "mitigate the impact" of the CTBT on the U.S. nuclear force posture. In the view of many critics both at home and abroad, such an elevated level of funding for nuclear design code development, warhead "replacement," and "stockpile management" is simply not commensurate with the reduced role that the United States claims to seek for nuclear weapons in world affairs.

Key members of the nuclear weapons establishment charged with carrying out the SSM program are not themselves persuaded that this program will be able to meet the requirements of the prevailing nuclear strategy, which assumes a continuing ability to make the kinds of nuclear weapon modifications and improvements that previously required "certification" by underground tests to establish or regain confidence in nuclear explosive performance. A principal objective of the "Science-Based Stockpile Stewardship" (SBSS) component of the overall SSM program is to greatly improve the capability of the nuclear weapons laboratories to predict accurately the explosive behavior of nuclear weapons through the use of enhanced computer simulations.

The overall stewardship program strategy accords a central -- indeed critical -- role to computer modeling. Successful development of advanced computational capabilities has become indispensable to the success of the technical approach adopted for the overall program. ASCI was developed as a focal point within Stockpile Stewardship for the purpose of generating the quantum leaps in supercomputing hardware, systems software, software development tools, data archiving, and applications software that are needed to realize a "virtual testing" capability for the nuclear weapons program.[8]

Both SBSS and ASCI are fundamental paradigm shifts from past U.S. nuclear weapons work. For SBSS as a whole, the shift is toward much heavier reliance on computer simulation -- buttressed by laboratory experiments to obtain more accurate modeling data and validate computer code predictions -- for the performance of DOE's historical nuclear weapons missions. For ASCI, the paradigm shift involves the use of a relatively new category of supercomputers which employ Massively Parallel Processing (MPP). Much of the motivation for the Academic Strategic Alliances Program and related industry collaboration with Defense Programs stems from uncertainties introduced by the technology of massively parallel processing and the associated complexities of the high-resolution three-dimensional physics models enabled by the new "tera-scale" MPP architectures.[9]

The Academic Strategic Alliances Program also represents a degree of reliance on non-government academic scientists for assistance in the design and development of U.S. nuclear weapons not seen since the original wartime Manhattan Project, an ironic development in light of the disappearance of the East-West geopolitical confrontation that spawned (and allegedly justified) the terrible risks of national and even global annihilation during the Cold War era of massive countervailing nuclear threats. Within the secretive confines of the U.S. national security establishment, however, a fervid attachment to the doctrines and force postures of nuclear deterrence persists even as its raison d'etre -- Soviet communist expansionism -- has faded into history.

While clinging to what remains of the old "Soviet threat" in Russia's aging arsenal, the vast American state-sponsored apparatus of nuclear deterrence -- now shorn of some of its cold war excess -- has increasingly shifted its focus to the more diffuse mission of deterring an indeterminate category of so-called "rogue states" from threatening to use not just nuclear weapons -- which they have yet to acquire -- but any and all chemical, biological, or radiological "Weapons of Mass Destruction" (WMD). However, not only does this new policy contradict previous "negative security assurances" delimiting the circumstances under which the United States would defend itself, or its allies, with nuclear weapons, but it undermines the moral and political legitimacy of U.S. nonproliferation efforts. By what right, other countries ask, do the United States and a few other nuclear weapon states arrogate to themselves alone the privilege of insulating themselves from WMD threats by threatening prompt (and possibly preemptive) nuclear retaliation? Moreover, as retired Air Force General George Lee Butler, commander of U.S. strategic nuclear forces during the Bush Administration, has trenchantly observed, the political and moral underpinnings of this policy are flawed -- "There are no rogue nations," says Butler, "only rogue leaders."[10]

Likewise, the very legality of the revised U.S. nuclear posture is now in question, following a 1996 advisory opinion of the World Court in the Hague that concluded "the threat or use of nuclear weapons would generally be contrary to the rules of international law," while enumerating but one possible exception, based on the fact that the current state of international law does not provide a sufficient basis to "conclude definitively whether the threat or use of nuclear weapons would be lawful or unlawful in an extreme circumstance of self-defense, in which the very survival of the State would be at stake (emphasis added)."[11] Since no nation in the Pentagon's (current) gallery of "rogue states" now poses, or is likely to pose, a threat to the very survival of the United States (or to any of its allies covered by a nuclear security guarantee), one is lead to the conclusion that virtually all of the WMD "contingencies" for which the Pentagon has developed plans for possible employment of nuclear weapons, if implemented, would result in U.S. violations of international law.

An intriguing legal question, which can only be touched upon here, is whether current U.S. force planning and doctrine for contingent use of nuclear weapons against prospective WMD-armed opponents itself constitutes an unlawful "threat" to employ nuclear weapons, given that no threat to the "very survival" of the United States or a nuclear alliance state is present. An often cited exception to this view is the case of Israel, surrounded by hostile states with WMD arsenals and more numerous conventional forces, but Israel has its own nuclear force of medium range missiles and aircraft to deter threats to "the very survival of the state."

For the present, U.S. policymakers take shelter in the common law notion of "belligerent reprisal" to reconcile U.S. nuclear weapons employment policy and planning against non-nuclear states with the non-use obligations established by U.S. negative security assurances and the requirements of international humanitarian law. Government lawyers reason that in the wake of actual WMD use by a state, U.S. nuclear retaliation would be justified in order to forestall future illegal WMD attacks by that state. Whatever legal merit there may be in this reasoning, it clearly does not apply to all instances of the use of WMD for belligerent reprisal. For example, the Chemical Weapons Convention that entered into force in 1996 contains an absolute prohibition on the use of chemical weapons under any circumstances. In other words, chemical weapons have become an inadmissable instrument of warfare, even for the execution of what would otherwise be regarded as possibly justifiable "belligerent reprisals." Many millions of people worldwide feel that nuclear weapons belong in the same legal category and thus seek multilateral negotiations on an international convention abolishing nuclear weapons.

C. The Role of the Academic Strategic Alliances Program Centers

The five Academic Strategic Alliances Program "Centers of Excellence" are intended to enable replacement of the "integrated system demonstrations" represented by underground nuclear explosions with detailed computer simulations as the locus for future confidence in nuclear weapon system performance. The Alliances Program Centers are intended to develop and incorporate in microcosm the methodologies to be employed in the overall Stewardship Program -- tackling disparate problems involving the simulation of complex physical phenomena and technologies, using software written for and run on weapons laboratory ASCI supercomputers within the decade allotted to bring the stewardship simulation effort to fruition.

In addition, it is likely that some Alliance Program Centers will be called upon to solve physical and computational problems involving nuclear weapons that are of immediate and direct concern to the ongoing Stockpile Stewardship and Management Program. The Centers are also intended to train and recruit the next generation of nuclear weapon designers from among today's students, and more generally to foster greater interaction between DOE Defense Programs and the U.S. academic community, and create a market niche beyond the nuclear weapons program for massively parallel processing.

The tension between open academic research and the secrecy of nuclear weapons work is evident in the "Short-Term ASCI Alliances Platform Access Policy,"[12] in which the Department of Energy has chosen to deny foreign nationals (students, postdoctoral associates, and professors) accounts on ASCI computers, or access to ASCI computer software or technical data which currently falls under export control restrictions. To the extent that these limitations effectively shut out foreign students from meaningfully participating in the core Alliances Program research, longstanding academic traditions of openness and non-discrimination are compromised. To the extent that these limitations do not shut out foreign nationals, the Academic Strategic Alliances Program could provide citizens from non-nuclear weapon states with many of the scientific skills and even some of the critical data needed for nuclear weapons design and engineering. Article I of the Nuclear Nonproliferation Treaty (NPT) binds each nuclear weapon State Party to the Treaty "not in any way to assist, encourage, or induce any non-nuclear weapon State to manufacture or otherwise acquire nuclear weapons or other nuclear explosive devices, . . .(emphasis added)"[13]

An alternative to massive investments in new "science-based" capability for nuclear weapons simulation has been suggested by NRDC and other critics of the current program, but DOE has suppressed detailed consideration of any alternative to its proposed stewardship program in the public documentation required under the National Environmental Policy Act. This alternative would acknowledge that in the absence of nuclear weapons testing and new-design weapon engineering and production, DOE's nuclear weapons design expertise will inevitably decline. The appropriate task is to acknowledge that this decline is inevitable, plan for it, and develop a strategy now to assure than a reliable and safe residual stockpile will be maintained in spite of it, if required.

Far from developing a $4.5 billion dollar annual program to "offset" the limitations imposed by a test ban on new knowledge about nuclear weapons performance -- in order to "confidently" make changes in weapons without resort to testing -- the alternative paradigm would accept these limitations as the ground reality of the new political and strategic environment. Maximum effort would be placed in the near term on gaining the assurance now, before DOE's accumulated nuclear design expertise erodes through further downsizing and retirements, that three or four of the weapon designs slated for retention in the enduring stockpile are placed in a configuration that supports future remanufacture with confidence. That is, the future production configurations of such weapons must be thoroughly examined and where, necessary, respecified, based on the qualification of replacement processes and materials that DOE will guarantee to make available for remanufacture in the future.

This process should be carried out now while most of the designers and production engineers for these weapons are still available to DOE, rather than relying on an elaborate high risk program to train a whole new generation of weapons designers on surrogate facilities who will then be expected, in DOE's scenario, to make major future modifications to the U.S. nuclear stockpile based on the insights provided by new three-dimensional codes and experimental capabilities that do not yet exist, will cost billions to develop, and may never work as planned.

Not only would a more modest and technically conservative approach result in the ability to sustain confidence in the stockpile without explosive testing at less cost for an indefinite period, but it is far more consistent with the purposes of the Comprehensive Test Ban Treaty and the Nuclear Nonproliferation Treaty, which seek an end to the further technological development of nuclear weapons and the further proliferation of nuclear weapons capabilities. One political price, and a major drawback, in DOE's "new knowledge" approach to nuclear weapons stewardship is that much of the resulting science and technology will become available, overtly or covertly, to other nations, aggravating the proliferation of nuclear weapons design expertise.

Even worse, determined pursuit of the stewardship approach may result in new compact means of initiating explosive releases of energy in uncontrolled materials, such as hydrogen and lithium, creating a path to nuclear weapons that will evade entirely the existing superstructure of proliferation restraint based on controlling access to significant quantities of fissile materials, such as plutonium and highly-enriched uranium.[14]

The fundamental purpose of this report is to better inform policy-makers and scientists within the national security community, and the faculty, students and other members of the affected university communities, about DOE's new reliance on academic institutions to provide critical support to the development of advanced simulation tools for maintaining, modifying, and modernizing the U.S. nuclear weapons stockpile.


2. Kramer, David, "DOE Seeks Universities' Help in High Performance Computing," Federal Technology Report, November 21, 1996, p. 9.

3. Prior to the establishment of DOE in 1977, these functions were the responsibility of the Manhattan Engineering District (1942-46), the Atomic Energy Commission (1947-74), and the Energy Research and Development Administration (1975-77).

4. United States Nuclear Tests: July 1956 through September 1992, DOE/NV-209 (Rev.14) December 1994, p. viii. These 1149 detonations of discrete nuclear explosive devices were distributed among, and are often reported as 1054 "tests," including the 24 tests at the Nevada Test Site conducted jointly with the United Kingdom, but this number does not include the two wartime uses of nuclear weapons by the U.S. at Hiroshima and Nagasaki, Japan. No doubt the 35 "Peaceful Nuclear Explosions" (PNEs) also added to the stock of nuclear weapon design knowledge.

5. William M. Arkin, Robert S. Norris, Joshua Handler, Taking Stock: Worldwide Nuclear Deployments 1998, Natural Resources Defense Council, Washington, D.C., January 1998.

6. Letter from National Security Advisor Brent Scowcroft, Secretary of Defense Richard B. Cheney, and Secretary of Energy James D. Watkins to Senator J. Bennett Johnston, July 10, 1992.

7. DOE, "Stockpile Stewardship and Management Program," 1995, p. 3.

8. Hardware refers to the supercomputers themselves (such as Sandia National Laboratory's "Teraflops"); associated data storage and retrieval technology; high-speed, secure communication between the nuclear weapons laboratories; and so forth.Applications software are actual problem-solving codes.Systems software refers to the more basic computer codes such as operating systems, compilers, and utilities.

9. "tera-scale" computer systems are those in which basic characteristics such as speed¾floating-point operations per second¾are measured in the trillions.

10. R. Jeffrey Smith, "The General's Conscience," Washington Post Magazine,December 7, 1997, p. 44.

11. International Court of Justice, The Hague, Communiqué No.96/23, 8 July 1996, p. 2.A thorough and insightful analysis of the possible ramifications of this decision is contained in John Burroughs, The (Il)legality of Threat or Use of Nuclear Weapons: A Guide to the Historic Opinion of the International Court of Justice, Lit Verlag, Munster, 1997.

12. "Short-Term ASCI Alliances Platform Access Policy," http://www.llnl.gov/asci-alliances/9711policy.html (Modified on 11/11/97).

13. Arms Control and Disarmament Agreements: Texts and Histories of Negotiations, USACDA, Washington, D.C., 1982, p. 92.

14. For example, the very high current pulse and strong magnetic field of a high-explosive magnetic flux compression generator might be used to compress a plasma on its axis, causing a powerful burst of x-rays to ignite a deuterium-tritium target. Such novel concepts are being explored in the in the Stockpile Stewardship Program.

Back to contents page

last revised 1/22/1998

Sign up for NRDC's online newsletter

See the latest issue >

NRDC Gets Top Ratings from the Charity Watchdogs

Charity Navigator awards NRDC its 4-star top rating.
Worth magazine named NRDC one of America's 100 best charities.
NRDC meets the highest standards of the Wise Giving Alliance of the Better Business Bureau.

Donate now >

Share | |
Find NRDC on