Declassification of Records Relating to Nuclear Weapons Testing and Cleanup Activities in the Marshall Islands Feasibility Study

Declassification of Records Relating to Nuclear

Weapons Testing and Cleanup Activities in the

Marshall Islands

Feasibility Study

August 2022

Executive Summary 8

Background 13

National Defense Authorization Act (NDAA) 13

Summary of Issues and Recommendations 13

U.S. Nuclear Weapons Testing in the Marshall Islands 16

U.S. and Marshall Islands Relations 17

Compacts of Free Association 17

Current concerns of the Marshallese 18

Methodology for the Feasibility Study 20

Appendices 30

Appendix A 30

National Defense Authorization Act of 2020 30

Appendix B 31

Public Interest Declassification Act of 2000, as amended 31

Appendix C 42

Public Interest Declassification Board Member Biographies 42

Appendix D 47

Potentially Responsive Record Collections 47

NARA Holdings and Accessioned Records 47

Federal Records Relating to Nuclear Weapons Testing in the Marshall Islands 47

Section I: Possibly Responsive Records by Operation 47

Section 2: Possibly Relevant Records by Record Group(s) 47

Record Group 19: Records of the Bureau of Ships (1946-66) 47

Record Group 24: Records of the Bureau of Naval Personnel 48

Record Group 38: Records of the Chief of Naval Operations 48

Record Group 48: Records of the Office of the Secretary of the Interior 48

Record Group 52: Records of the Bureau of Medicine and Surgery 49

Record Group 59: General Records of the Department of State 49

Record Group 74: Record of the Bureau of Ordnance (1946-59) 49

Record Group 77: Records of the Army Corps of Engineers 49

Record Group 90: Records of the Public Health Service 49

Record Group 181: Records of Naval Districts and Shore Establishments 49

Record Group 218: Records of the Joint Chiefs of Staff 50

Record Group 220:

Temporary Committees, Commissions, and Boards 50

Record Group 313: Records of Naval Operating Forces 50

Record Group 326: Records of the Atomic Energy Commission 50

National Archives at Riverside 50

Record Group 338: US Army Operational, Tactical, and Support

Organizations (World War II and Thereafter), 1917-1999 50

Record Group 344: Records of the Naval Sea Systems Command (1966 - ) 50

Record Group 359: Records of the Office of Science and Technology 51

(OST) 51

Presidential scientific advisor records in the appropriate Presidential libraries.

Special Assistant to the President for Science and Technology 1957-1972

and Science Advisor to the President, 1973-76. 51

Record Group 374: Records of the Defense Threat Reduction Agency 51

Record Group 385: Records of the Naval Facilities and Engineering

Command 51

Record Group 402: Records of the Bureau of Weapons (1959-66) 51

Record Group 407: Records of the Adjutant General's Office 51

Record Group 412: Records of the Environmental Protection Agency 51

Record Group 434: Records of the Department of Energy 51

Record Group 443: Records of the National Institutes of Health 52

Record Group 494: Records of US Army Forces in the Middle Pacific 52

Record Group 496: Records of GHQ, SW Pacific Area and United States

Army Forces, Pacific (World War II) 52

Record Group 550: Records US Army, Pacific 52

Record Group 593: Records of Strategic Systems Projects Office (SSPO) 52

San Francisco Federal Records Center 52

Presidential Libraries Records Relating to Nuclear Weapons Testing in the Marshall

Islands 53

Dwight D. Eisenhower Presidential Library 53

Herbert Hoover Presidential Library 54

Jimmy Carter Presidential Library 54

The Center for Legislative Archives Records Relating to Nuclear Weapons Testing in the

Marshall Islands 56

Records of the Joint Committee on Atomic Energy 56

Records of the Senate Committee on Energy and Natural Resources 56

Department of Defense 57

U.S. Army 57

US Army Center of Military History, Library at Fort McNair 57

Army Declassification Directorate 58

US Army History and Education Center 58

U.S. Navy 59

Bureau of Medicine and Surgery 59

Naval Sea Systems Command 60

Comma

nder Naval Surface Force 60

Dept of the Navy, US Pacific Fleet 60

Naval History and Heritage Command 60

U.S. Fleet Forces 62

DRMD Declassification Program 62

U.S. Air Force 62

Air Force Historical Research Agency 63

Joint Chief of Staff 63

JS/Joint History Office 63

Defense Technical Information Center 63

Defense Threat Reduction Agency 63

Office of the Secretary of Defense (OSD) 63

Undersecretary of Defense Policy 64

OSD Records Office (WHS/ESD/RDD) 64

Under Secretary of Defense for Personnel and Readiness 66

Department of Energy 67

Headquarters Defense Programs 67

Headquarters DOE Records Management Historians Office 67

Headquarters Environmental, Health, Safety, and Security 68

Headquarters Federal Records Center (WNRC) 68

Brookhaven National Laboratory 69

Lawrence Livermore National Laboratory 69

Los Alamos National Laboratory - Institution Records Management 70

Los Alamos National Laboratory - Weapons Records Services 71

Los Alamos National Laboratory - Weapons Records Services-Project Y Index 72

Nevada National Security Site 73

Office of Scientific Technical Information 73

Sandia National Laboratory 73

Appendix E 75

National Declassification Center Review Process Flowchart 75

Appendix F 77

List of Stakeholders Consulted 77

Bibliography and Resources 79

August 12, 2022

The Honorable Jack Reed

Chairman

U.S. Senate Armed Services Committee

The Honorable James Inhofe

Ranking Member

U.S. Senate Armed Services Committee

The Honorable Adam Smith

Chairman

U.S. House of Representatives Armed Services Committee

The Honorable Mike Rogers

Ranking Member

U.S. House of Representatives Armed Services Committee

The Public Interest Declassification Board (PIDB) is pleased to provide you with its feasibility

study for reviewing and declassifying information related to nuclear weapons, chemical

weapons, and ballistic missile tests conducted by the United States in the Marshall Islands,

including cleanup activities and the storage of waste. This study was conducted pursuant to

section 1685 of the Fiscal Year (FY) 2022 National Defense Authorization Act (NDAA) which the

President signed into law on December 27, 2021.

Over the past three decades, the U.S. Government has declassified substantial volumes of

information, including significant historical information about the 67 nuclear tests conducted in

the Marshall Islands. Declassified and publicly available records date from the earliest test in

1946 to monitoring activities in the 1980s. Although the U.S. government has undertaken a

number of declassification projects specifically related to the U.S. nuclear testing in the Marshall

Islands, department and agency efforts in releasing information about the tests and cleanup

activities have been disjointed, singular, and transactional. There has been no strategic or

“whole-of-government” approach. As a result, declassified information can be difficult to locate

and analyze.

The U.S. Government has also not made all records related to testing in the Marshall Islands

available to the public. Some information remains classified or is not publicly available for other

reasons. Other records are unclassified, or their classification status is unknown, as they are

unprocessed and located in U.S. Government archives or records centers across the nation.

These records are almost always contained within larger collections of records that include a

variety of topics unrelated to weapons testing in the Marshall Islands. As shown in Appendix D

of this study, departments, agencies, and the National Archives and Records Administration

(NARA) identified a multitude of archival collections that may contain potentially relevant

information.

We believe there is new information to be discovered within these unprocessed collections. We

also believe a special project to search for, identify, review for declassification, and make

responsive records available to the public is feasible. However, if undertaken, completing it

successfully will be difficult, costly, and lengthy. Such a project will require dedicated resources,

highly specialized staff with subject-matter expertise, and significant new funding. Critically, if

such a project is to be undertaken, it must be done strategically and in a way that permits

interagency collaboration, integrates declassification and public access review standards, and

enables easy online public access, ideally including access to previously declassified records.

The United States Congress created the PIDB through the Public Interest Declassification Act of

2000, as amended. As authorized by Congress, the PIDB advises the President, other

Executive branch officials, and the Congress on the systematic, thorough, coordinated, and

comprehensive identification, collection, and review for declassification and potential release of

records with archival value and extraordinary public interest. The PIDB is non-partisan and

composed of nine members drawn from the public, five of whom are appointed by the President

and four by the respective leaders of Congress. The PIDB is an independent government entity,

yet it does not have an appropriation and, instead, relies on the Information Security Oversight

Office (ISOO) at NARA for all staff, program, and logistical support. The ISOO staff spent over

600 staff hours supporting the preparation of this study.

This is the first instance of the PIDB conducting a declassification study on a specific topic.

Given the NDAA requirement’s breadth, the PIDB formed its recommendations based on a high-

level plan that included seeking data from the Departments of Defense and Energy and NARA

about classified archival holdings as well as records that were previously declassified. The

PIDB’s plan also included extensive background research, speaking with stakeholders, holding

closed-door investigative meetings, and gaining insight into antiquated records management

and declassification processes.

At the onset, we requested assistance, including staff augmentation, from the Secretaries of

Defense and Energy. They or their predecessor agencies and components were responsible for

the U.S. nuclear weapons testing program in the Marshall Islands and related activities. They

have subject matter expertise and experienced staff who perform records management and

declassification review. They also maintain physical possession of the records in their facilities

and in national laboratories across the country. Importantly, they are also the only two

departments that have declassification authority over information related to nuclear testing. In

accordance with the Atomic Energy Act of 1954, as amended, the Department of Energy is

solely responsible for the classification and declassification of information identified as

“Restricted Data” while the Departments of Defense and Energy jointly are responsible for the

classification and declassification of information identified as “Formerly Restricted Data.”

Between late December 2021 and June 2022, the PIDB and its staff conducted over 40 virtual

and in-person meetings with stakeholders. We met with declassifiers and records managers

from the Departments of Defense and Energy, agency staff supporting the National

Declassification Center, archivists at NARA, and representatives from the Department of State.

We held virtual meetings with representatives from the Marshall Islands and its National Nuclear

Commission to gain perspective and learn about the interests of the Marshallese. We met with

historians to better understand both the history of nuclear weapons testing in the Marshall

Islands and learn about their research methodologies.

In addition to the need for dedicated resources, a special declassification project like this one is

further complicated by the antiquated condition of current declassification policies and

processes. In many ways, the challenges identified in this feasibility study mirror the challenges

the PIDB has previously identified with declassifying and making publicly available historical

records. In previous reports to the President, we expressed the urgent need to reform the

classification and declassification system and offered recommendations for its modernization to

help address these challenges. In our June 2020 report to the President, we offered a road map

to enact reform. We believe that many of the challenges detailed in this feasibility study would

be alleviated or reduced through this modernization and introduction of automated systems. For

instance, investments in and the application of machine learning and artificial intelligence

technologies could support this project, including by identifying potentially responsive records

and assisting declassification and public access reviews.

The PIDB hopes this declassification feasibility study will serve as a call to action for the

government to invest in modernizing the classification and declassification process. The

staggering departmental estimates of the resources required to accomplish greater public

understanding of specific weapons testing—$100-200 million—is far more than we currently

spend annually on all government declassification efforts combined. Action to fix the

classification system is required today. Modernization will not only improve public access to our

nation’s national security history and aid transparency, but it will also support our nation’s 21st

century national security requirements.

Should your schedule permit, we would welcome the opportunity to discuss this feasibility study

in greater detail with you. We look forward to answering your questions and those of the

committee members and staff.

Sincerely,

EZRA COHEN

Chair

Executive Summary

The Public Interest Declassification Board (PIDB) completed a study on the feasibility of

conducting a declassification review of records related to U.S. weapons tests conducted in the

Marshall Islands for the United States Senate and House of Representatives Armed Services

Committees. This study focused on activities concerning the 67 U.S. nuclear weapons tests

conducted between 1946 and 1958. It also included activities related to these tests, including

the cleanup and storage of nuclear waste. The study was performed pursuant to section 1685 of

P.L. 117-81, the National Defense Authorization Act, which President Joseph R. Biden signed

on December 27, 2021.

The PIDB believes it would be feasible to conduct a topical-based declassification project, but

would encourage Congress to consider whether such a project merits the resources required.

To be successful, such a project will need to address several significant challenges and will

require dedicated new resources. Based on data provided by Executive branch agencies, the

project will cost between $100-200 million. It will require approximately 100 new Full Time

Equivalent (FTE) staff who are fully cleared at the Top Secret and Q levels, and who are trained

to identify and review technical nuclear weapons data. Given the large volume of unprocessed

records to search—estimated to be over 40-80 million pages scattered in various facilities

around the country—and the cumbersome manual processes associated with searching for and

then reviewing records on a word-for-word basis, the PIDB estimates that it will take at least four

to six years to complete.

Alternatively, the government could scale back the project. Taking this approach would reduce

the volume of records to search and review, reduce costs, and require fewer FTE staff. Working

with stakeholders, the Government would prioritize the search of archival collections deemed

most likely to contain relevant information and information that is historically significant. There

would be fewer records to search and to review for public access. However, staff would still

require the appropriate security clearances and would also still require extensive technical

training. Although the overall time needed to complete a scaled-down project would be reduced,

it is our assessment that such a project will still take between three to four years to complete.

A third option would be for the government to focus on digitization of suspected records and the

subsequent application of machine learning technologies to support either the search for

relevant records or public access review—or both. If combined with the digitization of previously

declassified records, that additional data could be used to support technology-assisted review.

Digitization costs, according to NARA and the departments, are not insignificant. Estimates we

received were based on previous projects and varied in price from just over $1.00 per page up

to over $3.00 per page, depending on requirements. These costs likely preclude digitization of

entire unprocessed collections, although some key unprocessed collections identified in a

prioritization plan—for instance, the U.S. Congress’ Joint Committee on Atomic Energy—

should be digitized in their entirety, given their historical significance and likelihood of containing

important relevant information.

Completing a Declassification and Public Access Review Project: What It Would Take

Ultimately the U.S. Congress will make the determination whether to dedicate funds for a

special project to search for relevant records and review them for public access. The three main

Executive branch entities—the Departments of Defense and Energy and NARA—will need to

develop and agree on a comprehensive project plan with an end goal of making relevant

records available and accessible to the public. In addition to dedicated funding and new

resources, to be successful, a declassification project of this scope and size will need to

address challenges the PIDB identified during the course of its research and in its interviews

with stakeholders.

Addressing these challenges form the basis of the PIDB’s recommendations:

● Sustained executive-level leadership, direction, and support. Senior-level officials from

participating agencies must oversee the project and remain actively engaged throughout

its duration. This continued engagement serves several purposes, including reinforcing

its importance. Senior-level officials also can resolve interagency disputes, provide

direction, monitor progress, and hold staff accountable.

● A senior-level official from the Department of Energy to serve as project leader. The

Department of Energy has experience leading several declassification projects related to

nuclear weapons testing, including projects concerning nuclear weapons testing in the

Marshall Islands. Importantly, it is solely responsible for identifying, classifying, and

declassifying technical nuclear weapons information under the Atomic Energy Act of

1954, as amended, and leads the Historical Records Restricted Data Recognition and

Historical Records Restricted Data Reviewers certification and training courses.

Recently, the Department of Energy has been testing advanced technologies to support

the identification of Restricted Data (RD) information in digital data.

● Interagency agreement on the project plan. The Departments of Defense and Energy

and NARA need to agree on all elements of the project, including its scope, terminology

for conducting searches, timelines, declassification review and public access review

standards, multi-agency review standards, scanning and redaction standards, and equity

review referral policies and practices. Additionally, all participating departments must

have access to secure communication systems and software for redacting sensitive

information.

● Include unclassified records in the project. The NDAA limited the scope of the PIDB’s

study to classified records, meaning unclassified and previously declassified records fell

outside the scope of Congress’ request. In fact, as the PIDB gathered information, it

learned that many unprocessed records are not classified, including some that may be

relevant to the request. These unclassified records are likely very important to the

Marshallese people, including information on health, safety, environmental testing, and

cleanup activities.

● Interagency agreement on declassification guidance for Restricted Data (RD) and

Formerly Restricted Data (FRD). Typically, once identified, records containing RD and

FRD information are simply excluded from public release. The project leader should

convene an interagency group of technical subject matter experts, historians, and

declassifiers to develop a comprehensive and detailed declassification guide for all

participating agencies to use for the project. At the outset, this group should engage with

stakeholders and researchers to learn about their interests. This special RD/FRD

declassification guide should seek to declassify and disclose as much information as

possible, including evaluating whether the public interest in disclosure outweighs the

need for restriction. It should receive approval from interagency senior leaders and its

contents incorporated into training reviewers. The declassification guide should be

sufficiently specific and detailed that it could support a project to use advanced

technology to support automated review.

● Interagency agreement on declassification guidance for national security restrictions.

Participating agencies should develop a comprehensive and detailed topical

declassification guide to review national security information. Like the process used in

creating the RD/FRD declassification guide, it should take stakeholder interest into

account, evaluate the public interest in disclosure, and be sufficiently detailed to permit

its use for automated review. Once approved by interagency senior leaders, reviewers

from all participating agencies should receive authorization to use it.

● Interagency agreement on other non-classification restrictions. The PIDB learned that

there are records and information withheld from the public for reasons other than

national security or Atomic Energy Act requirements. Typically, these records are

unclassified but contain information that, if publicly released, could violate the privacy of

a person. However, each agency has its own definitions which reviewers then apply

routinely without conducting research. This project should have common and agreed-

upon definitions for privacy and other non-national security restrictions that also include

examples to guide reviewers. Working with stakeholders, agencies should seek the

narrowest definitions, and consider possible substituted text that informs users of

content. The creation and use of these types of declassification guides and standard

criteria for non-national security restrictions ensures consistency of declassification and

public access reviews across agencies and maximizes the information that can be made

available to the public.

● Agreement on processing non-textual records. Agencies identified several forms of non-

textual records that could potentially contain responsive information. They are listed in

Appendix D and include motion picture film, audio recordings, still pictures, and

microfilm. There must be agreement on whether to include these formats in the project,

and if so, how to search and review them for public access. While some formats do not

present difficult challenges, others do. Microfilm is a particularly vexing challenge as

each reel might contain thousands of pages from hundreds of documents on a variety of

topics. Isolating, reviewing, and redacting individual documents is even more

complicated. Regardless, all decisions about the treatment of non-textual records should

be carefully documented, described, and be included in the public record.

● Agreement on overcoming logistical challenges associated with the physical locations of

records. This report highlights challenges related to the physical locations of potentially

responsive records. Executive Order 13526 established the National Declassification

Center (NDC) at NARA in College Park, Maryland with the express purpose to facilitate

the review of records containing multiple-agency equities, including the Department of

Energy’s authority to review information designated as RD and FRD. At present, it

remains the only entity focused on the review of these multi-equity records. Yet, as

detailed in Appendix D of this report, many important unprocessed collections are

physically located in other facilities that lack any ability to conduct or lead a multiple-

agency equity declassification review process. These facilities include the Department of

Energy sponsored labs in New Mexico, California, and New York, a Defense Threat

Reduction Agency facility in New Mexico, and the Center for Legislative Archives (CLA)

and the Naval History and Heritage Command (NHHC) in Washington, D.C. None have

processes that permit the declassification review of multi-equity records. Successful

completion of this project must include solutions that permit the review of these

important unprocessed collections, including:

o developing and agreeing to a plan that permits a thorough search for responsive

records; and

o deciding how best to conduct a declassification and public access review on

these records. One option to consider would be to digitize them—or the entire

collection, depending on its historical significance.

● Access to secure communications networks. While several facilities have either Secret

and/or Top Secret communications networks, not all do. Specifically, the NDC and the

NHHC only have Secret network communications, while the CLA does not have a

secure communications capability. The accreditation process for obtaining a Top Secret

network and terminal is both costly and, more importantly, time-consuming.

● Digitize the records. Agencies should evaluate the benefits and costs of digitization for

this project. While digitizing all records identified in Appendix D is both cost-prohibitive

and not desirable, there is great value in digitizing relevant records. Both the NHHC and

the CLA have unprocessed collections that should be considered for digitization.

Agencies should also consider the potential benefits to digitizing a few complete series

or entire collections that contain relevant information and are also historically significant.

Digitization will also improve public access to the records by making them available

online instead of only in research rooms or by request. It also enables the possible use

of advanced technologies to support declassification and public access review.

● Develop and use metadata standards for digitized records. It is critical that this project

develops and uses metadata standards to support any digitization effort. These

standards will support online access, increase discoverability, and provide users with

important contextual information, including archival description and arrangement

information. Researchers can track these records back to their original locations and

learn more about the contents and value of those series and collections.

● Consider the use of advanced technologies. This project offers an opportunity to test

machine learning and artificial intelligence technologies and learn how they can support

more effective and efficient declassification and public access review. The Department

of Energy is currently testing the use of this technology for identifying nuclear weapons

design information classified as RD. It should consider expanding this test to include

evaluating records for this project. The declassification guide created for this project

should be sufficiently detailed and precise to support advanced technology use. The

Department of Energy should also incorporate previously declassified records to improve

accuracy.

● An integrated public access plan that facilitates use. One challenge identified by

researchers is the confusion of where to research and find relevant declassified records.

Currently, there is no integrated point to access declassified and publicly available

records. They are scattered or grouped on multiple government websites, including

several different sites operated by the Department of Energy, its affiliated national

laboratories, the Defense Threat Reduction Agency, the Nuclear Testing Archive, and

NARA. Other websites contain broken links that are no longer active. Still others note

that online documentation is incomplete, and accessing specific records requires a

written request. Responsive records are provided at a later date. To be successful, this

project must include a strategic plan that enables easy public access. This plan must

include all participating agencies, and, ideally, should include previously declassified

records.

Considering a Cost/Benefit Analysis and Timeliness for Completion

In evaluating our recommendations, agencies must naturally weigh the costs to fund this project

and consider the additional resources required. As currently written in the FY 2022 NDAA,

Congress asked the PIDB to study the feasibility for declassifying information “relating to

nuclear weapons, chemical weapons, or ballistic missile tests conducted by the United States in

the Marshall Islands.”

The Department of Energy conducted several declassification projects over the past 30 years to

search for and review records relating to nuclear weapons testing in the Marshall Islands

These projects have resulted in the declassification of millions of pages of government records.

Department of Energy declassifiers informed the PIDB that almost all information has been

declassified—with only specific information detailing nuclear weapons design information still

classified as RD. Still, fully documenting the 67 nuclear weapons tests conducted in the

Marshall Islands between 1946 and 1958 has proven elusive as each project identifies some

new information.

While the Departments of Defense and Energy have informed the PIDB that the U.S. did not

conduct any tests of chemical or biological weapons in the Marshall Islands, the Department of

Defense and its components have a long history of using the Marshall Islands for rocket

launches and testing missiles, including ballistic missiles. This testing began in 1946 and

continues today at the 750,000-square-mile Reagan Ballistic Missile Defense Site. Including

information about all the missile tests conducted over the past 75 years would significantly

increase the scope of a declassification project. The Department of Defense informed the PIDB

that searching for potentially responsive records would include between 100 and 200 million

records and would take a decade or longer to complete. Most of the information in the historical

records, even if unprocessed or not yet accessioned to NARA, has been automatically

declassified.

The government must be realistic about what it can accomplish. Accordingly, designing the

project with a proper and realistic scope is important. It should be transparent about how it

develops and designs the project. This must include search policies and terms, metadata

requirements, scanning and digitization processes, and public access and declassification

review guides. The explanation should also include specific information about what information

remains unavailable or restricted. Once the project is complete and records are posted online,

an agency must be designated to maintain and update the website, adding newly available

records once restrictions lapse.

Between 1993 and 1997, the Department of Energy searched over 3.2 million cubic feet of records to

identify records related to nuclear weapons testing, and related activities as part of Secretary Hazel

O’Leary’s Openness Initiative. By the end of 1997, it had declassified over 10 million pages, including

records related to human radiation experiments and records related to nuclear weapons testing in the

Marshall Islands. Later, in the early to mid-2000s it identified additional records for declassification.

Beginning in 2012, the Department of Energy began a new round of searches, declassifying records later

that year and in subsequent years through 2015. Many of these records were provided to the Marshallese

at different points in time—on paper, disks and CDs. Other declassified records were posted on several

different websites, including the Department’s Office of Scientific and Technical Information website,

OpenNet; Lawrence Livermore National Lab’s website; and the Nuclear Testing Archive, to name a few.

Background

National Defense Authorization Act (NDAA)

The FY 2022 National Defense Authorization Act (NDAA), signed by President Joseph R. Biden

on December 27, 2021, included a section that mandated the Public Interest Declassification

Board (PIDB) conduct a feasibility study on the declassification of information “relating to

nuclear weapons, chemical weapons, or ballistic missile tests conducted by the United States in

the Marshall Islands, including with respect to cleanup activities and the storage of waste

relating to such tests,” (section 1685).

Summary of Issues and Recommendations

The PIDB identified many challenges that must be addressed as part of a declassification

project related to nuclear weapons testing in the Marshall Islands. The records that may be

responsive are in the legal custody of the Department of Defense, Department of Energy, and

the National Archives and Records Administration (NARA).

NARA holds records that may be responsive to the study. Archivists identified series that are

restricted for national security or privacy. Included in this study are unrestricted records that may

be responsive. These records may be either unprocessed or not fully described. An additional

challenge affecting the search is that the records are located in various locations throughout the

country. Records are held at the Washington, D.C. and College Park, Maryland facilities, the

regional archival facilities of Seattle and San Francisco, and the Hoover, Eisenhower, and

Carter Presidential Libraries. The National Declassification Center (NDC), established in

accordance with Executive Order 13536 section 3.7, coordinates the review of classified

documents in the custody of the National Archives. The process to conduct the review is

detailed in Appendix E.

though section 3.7 (e) of the executive order provided the authority for agency heads to

establish centralized facilities to conduct internal declassification reviews, agencies never

implemented internal declassification centers. Currently, there is no established process for

agencies to conduct internal reviews for records in their holdings, nor is there a process to

review records with interagency equities. Coordination among the agencies with equities for the

responsive records will be a major challenge requiring resources.

Challenges to equity reviews will present a problem across agencies. Staff at the Naval History

and Heritage Command noted that there are pending referrals to some of the responsive

records to the Department of Energy (DOE), Office of the Secretary of Defense (OSD), Joint

Chief of Staff (JCS), and the United States Air Force (USAF). Although some organizations

have conducted referral reviews, it is unclear if these reviews are complete. They further noted

that they have no evidence that DOE has reviewed any of the material referred to them over the

last 20 years. Similarly, the Defense Threat Reduction Information and Analysis Center has

historical records that have been partially declassified. These records are not publicly available

since other equity-owning agencies have not reviewed them. The largest equity-owning agency

with unreviewed equities is the Department of Energy.

Any future declassification project will require additional resources including staffing,

technology, and funding.

Department of Energy. The Department of Energy provided cost figures for their current

declassification program. It costs $5.50 per page for pass/fail reviews for a collection with 15%

of the collection containing classified information. This cost includes the support contractor at $5

per page (which includes scanning the documents for review) and federal staff to inspect the

work at $0.50. The cost increases to $7.50 a page for a collection containing 100% classified

documents. It costs about $32 per page for redaction review at the sentence level and to

provide a sanitized version of the document.

For the feasibility study, DOE estimated that the agency has 3.8 million pages of classified

Marshall Islands records to review (1.8 million DOE pages and 2 million referral pages). DOE’s

survey was done at the document level or box level and this estimate is based upon an average

of 10 pages per document or for boxes, at 2,500 pages per box. DOE would require additional

staff for such a project. If DOE hires an additional 18 new contractors and 2 new federal staff,

they estimate it would cost $134 million (at $35 a page) and take 23.5 years to redact 3.8 million

pages of classified Marshall Islands documents.

Department of Defense. The Department of Defense provided a rough estimate to perform a

Marshall Islands search, declassification review, and public release process. DoD would require

an estimated additional funding allocation of $60‒80 million dollars. The funds would support a

staffing increase for program managers, records management personnel, digital archivists,

records research analysts, and declassification personnel. They would also require technology

investments. They anticipate classified records would have to be transported to the Washington,

D.C. area for digitization, review, declassification, and public release processing. DoD would

also require funding to either improve or build declassification collaboration and workflow

systems. This would be a multi-year project, conservatively estimated to take five years.

The PIDB also identified the following areas that would have to be addressed for any future

undertaking to declassify records relating to weapons testing in the Marshall Islands. These can

be grouped into three categories: structural, organizational, and policy challenges.

Structural challenges:

● Overcoming poor record-keeping practices;

● Searching large quantities of record series—including non-textual records such as film—

for individual responsive records and devising a way to identify each, including their

original archival file location;

● Using technology and secure systems to process and review records; and

● Addressing unclassified records identified in the search of unprocessed series of

records.

Organizational challenges:

● Determining how to address and/or include records from previous declassification

projects, Freedom of Information Act requests, Mandatory Declassification Review

requests, or other public access requests, including those conducted by the Defense

Threat Reduction Agency and the Department of Energy.

Policy challenges:

● Creating a comprehensive list of search terms;

● Adopting policy boundaries on records to include in the project;

● Agreeing to standards that permit declassification of classified national security

information, including the boundaries of information that would otherwise remain

classified;

● Developing declassification requirements for information that is cross-classified as

“Restricted Data” in accordance with the Atomic Energy Act of 1954, as amended, and

processes that include quality control/quality assurance reviews in accordance with

section 3161 of the 1999 National Defense Authorization Act (Kyl-Lott Amendment);

● Adopting standards that sufficiently address privacy concerns, especially those that

involve health information, that may permit declassification but limit or prevent public

release;

● Evaluating all applicable statutes in devising criteria for declassification and for public

release of information;

● Developing processes to permit multi-agency declassification review of records that

have not been accessioned to the NARA or are not included in NDC processes; and

● Deciding on how to release all declassified records to the public and ensure its

continued public access.

This list, while incomplete, is illustrative of the many challenges that need to be addressed for

this project to be successful. Completion of such a large topical project will require new and

dedicated resources. They include millions of dollars for secure communications capabilities at

various facilities across the nation where classified records are currently stored, project

management tools, redaction software, and web design and hosting. The project will require

staff to perform searches and digitize responsive records, conduct public access and

declassification reviews, and design and manage the public-facing website.

Staff assigned to the project will require Top Secret and “Q” level security clearances. Currently,

it costs $5,410 for a Tier 5 (TS/SCI) clearance while an expedited clearance costs $5,845. Staff

will require specialized training in identifying scientific and technical information related to

nuclear weapons design, in making declassification and public access determinations using

criteria established for the project, and in using software and project management applications.

Given the geographic locations of the records across the U.S., staff will either have to travel to

those locations, or new staff hired at those locations.

U.S. Nuclear Weapons Testing in the Marshall Islands

At the onset of World War II, the Marshall Islands were occupied and under the control of the Empire

of Japan. The United States captured the Marshall Islands in 1944, administering it as an occupied

enemy territory until 1947. The Marshall Islands, along with Micronesia, the Northern Mariana

Islands, and Palau, became part of the U.S.-administered United Nations Trust Territory of the

Pacific Islands. At this time, the U.S. established the Pacific Proving Grounds and conducted

nuclear testing on various sites in the Marshall Islands and Pacific Ocean.

Between 1946 and 1958, the United States conducted 67 atmospheric atomic and

thermonuclear tests over the Marshall Islands. The tests were conducted on Bikini Atoll and on

or near Enewetak Atoll. The Castle Bravo test, conducted on March 1, 1954, on Bikini Atoll, was

the largest test and the detonation exceeded the expected yield. Bravo, a thermonuclear

weapon, was the largest weapon detonated by the United States. Scientists calculated an

expected 6 megaton explosive yield; however, they miscalculated, and the detonation was

significantly larger than expected at fifteen megatons. It was a thousand times more powerful

than the atomic weapon dropped on Hiroshima. The Marshallese on Bikini and Enewetak had

been evacuated prior to the test. Radioactive fallout impacted the inhabited atolls of Rongerik,

Rongelap, and Utirik and necessitated an emergency evacuation. According to reports, shortly

before the scheduled test the winds shifted direction. This shift was within the forecast and

therefore the test was deployed as scheduled. A 2013 Defense Threat Reduction Agency study

concluded that the lower-altitude winds did not contribute to the spread of fallout, rather the

radioactive material from the high-altitude portion of the Bravo cloud spread the fallout to the

inhabited atolls of Rongelap and Rongerik. Calculations had predicted that the fallout would

extend to the stratosphere, instead because of the size of the detonation, it was a one-hundred-

mile cloud which dispersed the radioactive material over a much larger distance.

A few months later, those from Utirik were permitted to return to their homes while those from

Rongelap would only be allowed to return home three years after the test, in 1957. By the 1980s,

although the island was declared safe for human habitation, the inhabitants asked to be

relocated and in 1985 Greenpeace agreed to relocate the Marshallese to other islands and

atolls.

The U.S. continued to use Bikini Atoll as a testing ground until 1958. After conducting radiation

monitoring and mitigation efforts and although there was still some radiation detected, the

former inhabitants returned to Bikini in 1969. Their food sources from the atoll were restricted

because of contamination and they would be required to undergo health monitoring. In 1974

testing discovered greater radiation and residents had higher readings than expected. In 1978

the U.S. removed the inhabitants and declared the area off-limits for at least 60 years.

U.S. and Marshall Islands Relations

Compacts of Free Association

In 1983 the United States, the Marshall Islands, and the Federated States of Micronesia signed

a Compact of Free Association. The United States provided economic assistance, defense, and

other services and benefits. In exchange, the U.S. was granted certain operating rights. The

Compact was signed by President Reagan in 1986 and that year the Republic of the Marshall

Islands declared independence.

The U.S. has continued to maintain a missile test facility, the Ronald Reagan Ballistic Missile

Defense Test Site, which is operated by the Army. The U.S. has operated this facility since post

World War II, initially as a Naval Station. The facility is located on Kwajalein Island and consists

of the test facility and a community for U.S. Army personnel and civilian contractors. The area is

leased by the U.S. and the Marshallese are excluded from living in this community although

they may work on the island.

In 2003 the Compact was amended and expires in FY 2024. Currently, there are negotiations

underway to amend the Compact. The Department of Interior (DOI) administers the grant funds

that are disbursed to the Marshallese affected by the nuclear testing, including from the fallout.

There are two interconnected grant programs. One is administered through Section 177 of the

Compact of Free Association and is focused on healthcare programs for the Marshallese living

on the four atolls impacted by the nuclear testing. The other grant program is broader and

provides ad hoc grants. The Government of the Marshall Islands requests funding for certain

projects that align with five development areas as defined in the Compact. These projects are

wholly separate from those grants associated with those focused on communities affected by

the nuclear tests and related activities.

The Compact allows Marshallese to migrate to the U.S. without requiring visas and permits

them to obtain employment. There are large Marshallese communities in Arkansas, Oklahoma,

Kansas, and Missouri. Although the relocated Marshallese pay income tax, they are not eligible

for other types of federal assistance such as Social Security, Medicare, or Medicaid. According

For historical background see: Holly M. Barker, Bravo for the Marshallese: Regaining Control in a Post-

Nuclear, Post-Colonial World (Cengage Learning, 2012); Adam Horowitz and Richard Einhorn, Nuclear

Savage: the islands experiment of Secret Project 4.1( [San Francisco, Calif.]: Video Project, 2012);

Thomas Kunkle and Byron Ristvet, Castle Bravo: Fifty Years of Legend and Lore: A Guide to Off-Site

Radiation Exposures (Fort Belvoir, VA: Defense Threat Reduction Information Center, 2013, DTRIAC SR-

12-001); Walter Pincus, Blown to Hell: America’s Deadly Betrayal of the Marshall Islanders. Diversion

Books, 2021.

to a 2020 report by the Government Accountability Office, more than 94,000 compact migrants

from Micronesia, the Marshall Islands, and Palau, live and work in the United States and the

territories.

Current concerns of the Marshallese

Long-term health concerns

The Department of Energy administers the Special Medical Care Program which provides

medical examinations and treatment for potential radiation-related cancers. This program serves

individuals who were exposed to radiation at the Rongelap and Uitrik Atolls during the Castle

Bravo test. At the time of the test there were 253 people on the atolls and according to the

FY2020 DOE report to Congress, Republic of the Marshall Islands Special Medical Care and

Logistics Program Activities, FY 2019, there were 91 eligible participants at the beginning of

FY2020. At the April 20 PIDB executive session, a DOE representative noted that there are

currently 74 eligible participants.

The Government of the Marshall Islands contends that the contamination from the fallout and

nuclear waste has affected all aspects of their lives. According to the Marshallese, the nuclear

fallout has greatly impacted their ability to rely on agriculture and fishing, resulting in a negative

effect on their diet and health. This has changed their diets and their overall health. This has

impacted generations of Marshallese, especially those who were forced to relocate.

The Marshallese have had a long history of distrust when it comes to their health and the

information concerning their healthcare. For many years they were not given access to their

own medical records, and many believed that the U.S. Government medical testing program

was for research purposes instead of medical monitoring. After the Bravo test, the U.S. began a

biomedical project within Operation Castle. This project was numbered 4.1 and suspicion of this

project stems from many factors. The final report as well as the addenda on Project 4.1 have

been declassified and are available on the Defense Threat Reduction Agency’s website, U.S.

Nuclear Test Radiation Exposure Reports. Unfortunately, it appears that the lab records from

Project 4.1 were not retained by the Department of Energy.

In their 1995 report, the Advisory Committee on Human Radiation Experiments, an independent

temporary advisory committee established by President Bill Clinton, concluded that there was

no evidence that either the fallout exposure or the medical tests were conducted for research

purposes. They noted there was a tension between data gathering and patient care since the

same physician was responsible for both. Additionally, the Advisory Committee noted that

insensitivity to cultural differences, failure to include the Marshallese in the planning and

implementation of the program, the failure to obtain consent for the medical treatments, and the

lack of openness led to the Marshallese’s suspicion and distrust of the healthcare provided by

the United States.

Runit Dome

Constructed in the late 1970s, Runit Dome is a containment structure on Runit Island, located

on Enewetak Atoll. From 1947 to 1958, the U.S. conducted 42 tests on the atoll. Runit Island is

uninhabited although the Marshallese returned to Enewetak Atoll in 1980 after a cleanup and

rehabilitation program. Runit Dome contains over 100,000 cubic yards of radioactively

contaminated soil, metal, concrete and testing debris which was encapsulated in concrete inside

the unlined nuclear test crater created by the Cactus nuclear test.

During the 1970’s, the Defense Nuclear Agency (DNA), the predecessor agency of the Defense

Threat Reduction Agency (DTRA) was the lead agency responsible for the cleanup and the

construction of the Runit Dome. Over 6,000 military service members from the Department of

Defense were involved in the cleanup and construction. Cleanup activities included clearing

contaminated vegetation, removing contaminated soil and debris, demolishing and removing

uncontaminated buildings and debris, recovering and disposing of unexploded ordnance, and

preparing the atoll for resettlement. These service members were exposed to radiation and

claim that they did not have sufficient training or protective safety equipment to complete their

work.

The United States was required to monitor the containment structure under the United Nations’

Trusteeship of the Marshall Islands. The first Compact required an agreement for the U.S. to

provide services and assistance, with the Government of the Marshall Islands assuming

responsibility and control of the Runit Dome. In 2012, through the Insular Areas Act of 2011,

containment monitoring and reporting was assigned to the Secretary of Energy.

The Government of the Marshall Islands is concerned that rising sea levels associated with

climate change will threaten the structural integrity of the dome. According to the Department of

Energy’s 2020 Report on the Status of the Runit Dome in the Marshall Islands, the dome is not

in danger of collapse or failure; however, there may be risk from the flow of contaminated

groundwater from beneath the dome into the marine environment. The rising sea levels in the

area, along with storm surges, may affect groundwater beneath the dome.

Previous Declassification Projects

In 1993, at the request of the Marshall Islands Government, the Department of Energy began a

search and retrieval effort of documents related to the U.S. nuclear testing program in the

Marshall Islands. In 2001, the agency believed that they had largely completed the task after

making publicly available over 14,000 Marshall Islands records. The information was provided

as hardcopies and on CD-ROMs. The agency set-up a website, OpenNet database, as part of

their openness initiative, which contained a third set of the information. Over the years, records

or bibliographic citations have been added to the OpenNet database. The database is updated

regularly as information is declassified.

After 2001 there were additional releases of information to the Marshall Islands Government. In

2005, the agency released declassified documents although the volume is not known. In 2008,

the agency released an additional 12,000 records that had been declassified. In 2015, another

set of CDs was provided, although this may have been a replacement set since the initial set

was found to contain personally identifiable information (PII). According to the agency, the

information they have on the releases is not well documented; for instance, they do not have a

comprehensive list of records although they do know that a “master index” was provided in

2016. The OpenNet database is comprehensive, for example, the agency noted that all of the

records from the Nuclear Testing Archive have been uploaded to the database; however,

researchers mentioned that searching can be difficult and it appears, in some cases, links have

been broken.

Methodology for the Feasibility Study

The PIDB conducted a high-level study rather than a record-by-record assessment. This study

focused on developing a more complete understanding of the status of potentially responsive

records created by federal agencies, departments, organizations, units, entities, and national

laboratories that participated in any aspect of nuclear weapons testing, environmental and

human radiation testing, containment, storage, and cleanup activities between 1946 through

1989.

The PIDB gathered information from departments and agencies about various classified records

that may be responsive and on records and records series that are:

● scheduled as permanent records;

● scheduled as temporary records;

● still in active use by the agency; and

● unscheduled.

As part of the study, the PIDB held meetings with stakeholders, including agency

representatives from the Departments of Defense and Energy, congressional staff,

congressional committee staff, State Department officials, representatives from the Republic of

the Marshall Islands, historians, researchers, records managers, and archivists at the National

Archives. A list of stakeholders is included in Appendix F.

The NDAA required that the Secretary of Defense and Secretary of Energy each provide

assistance to the PIDB in conducting the study. As part of the process, the PIDB sent formal

letters outlining the requirements of the feasibility study, identified the potential responsive

components, and the types of information that should be included in the response. The letters

are reproduced below.

Permanent record means any Federal record that has been determined by NARA to have sufficient

value to warrant its preservation in the National Archives of the United States, even while it remains in

agency custody. Permanent records are those for which the disposition is permanent on SF 115, Request

for Records Disposition Authority, approved by NARA on or after May 14, 1973. The term also includes all

records accessioned by NARA into the National Archives of the United States.

Temporary record means any Federal record that has been determined by the Archivist of the United

States to have insufficient value (on the basis of current standards) to warrant its preservation by the

National Archives and Records Administration.

Unscheduled records are Federal records whose final disposition has not been approved by NARA on a

SF 115, Request for Records Disposition Authority. Such records must be treated as permanent until a

final disposition is approved (36 CFR § 1220.18).

CHAIR

Ezra Cohen

MEMBERS

Paul-Noel Chretien

Laura DeBonis

Michael Lawrence

Benjamin Powell

EXECUTIVE

SECRETARY

Mark Bradley

MEMBERS

Trey Gowdy

Alissa Starzak

John Tierney

December 28, 2021

The Honorable

Lloyd J. Austin III

Secretary of Defense

1000 Defense Pentagon

Washington, DC 20301-1000

Dear Mr. Secretary,

Yesterday, President Biden signed the FY 2022 National Defense Authorization Act (the

NDAA) into law. The NDAA mandates in Section 1685 that the Public Interest Declassification

Board (the Board) “conduct a study on the feasibility of carrying out the declassification review

of nuclear, chemical, and ballistic missile tests conducted by the United States in the Marshall

Islands,” including cleanup activities and waste storage. This provision also aligns with 50

U.S.C. 3355a which requires the Board “to review and make recommendations to the President

in a timely manner with respect to any congressional request… to declassify certain records, to

evaluate the proper classification of certain records, or to reconsider a declination to declassify

certain records.”

On behalf of the Board and pursuant to the NDAA, I request assistance from Department of

Defense to complete this study as mandated by law.

Section 1685(d) specifies that the Departments of Defense and Energy “shall each provide to the

Board such assistance as the Board requests” to conduct the study. Given the technical nature of

the subject and the short timeframe for completion, the Board requests that three subject matter

experts from the Department of Defense be assigned and detailed to the Board for a period of 90

to 120 days. The subject matter experts must hold all appropriate security clearances, including

and active clearances for evaluating Formerly Restricted Data and Restricted Data. This request

also aligns with 50 U.S.C. 3355a which permits Federal government employees to be detailed to

the Board without interruption or loss of civil, military, or foreign service status or privilege.

Section 1685(b) requires the Board to complete its study and then submit a final report to the

Secretaries of Defense and Energy and the congressional Armed Services committees by March

27, 2022. As dictated by the NDAA, the report must address:

• The feasibility of carrying out the declassification review;

• The resources required to carry out the declassification review;

• A timeline to complete the declassification review; and

• Any other issues the Board determines relevant.

I have attached background information on the Board, its statutory authorities and

responsibilities, and biographical information on our members. The point of contact for this

support request is Mr. Mark Bradley, the Board’s Executive Secretary. He can be reached via

email at [email protected].

Very respectfully,

Ezra Cohen

Chair, Public Interest Declassification Board

Attachment

CHAIR

Ezra Cohen

MEMBERS

Paul-Noel Chretien

Laura DeBonis

Michael Lawrence

Benjamin Powell

EXECUTIVE

SECRETARY

Mark Bradley

MEMBERS

Trey Gowdy

Alissa Starzak

John Tierney

December 28, 2021

The Honorable

Jennifer M. Granholm

Secretary of Energy

1000 Independence Avenue, SW

Washington, DC 20585

Dear Madam Secretary,