Declassified Cold War Nuclear Target Lists, Mapped and Simulated

What a nuclear target list actually is

A target list is not the same thing as a war plan. The American documents in this article are mostly requirements studies: catalogues of everything a command believed it might want to destroy, drawn up to justify how many weapons it should be given. The British documents are the opposite: assessments of what the other side would probably hit, written so that civil-defence planners could decide where to put shelters and regional seats of government. The Warsaw Pact material is a third kind, exercise maps that show how a rehearsed war was expected to unfold.

Three terms come up repeatedly:

• DGZ, designated ground zero. The aiming point for one weapon. A city is not a target; it contains many DGZs, each a power plant, rail yard, airfield or, as we will see, "population".
• BE number. An index number from the Bombing Encyclopedia, the US catalogue of installations worldwide. The 1956 study identifies most targets by BE number, and the Encyclopedia itself is still classified.
• Priority. A rank order for the attack. In 1956 the top priority was Soviet "air power", the bomber bases that could strike America and its allies, ahead of cities and industry.

Weapon assignments, yields and warhead counts are redacted in every American release, so the simulations below let you choose the yield and show the consequences for the whole list at once.

United States, 1956: the SAC Atomic Weapons Requirements Study for 1959

On 22 December 2015 the National Security Archive at George Washington University published the most detailed nuclear target list ever declassified: the Strategic Air Command's Atomic Weapons Requirements Study for 1959, completed in June 1956 under General Curtis LeMay. It runs to more than 800 pages and was released through a Mandatory Declassification Review request that journalist Michael Dobbs had filed in 2006 while researching the Cuban missile crisis.¹

What the study contains

The document has two parts and two kinds of target:¹

• The Air Power list: more than 1,100 airfields across the Soviet bloc, each with a priority number. Numbers one and two were Bykhov and Orsha in Belarus, bases for Tu-16 Badger medium bombers that threatened NATO Europe.
• The "Systematic Destruction" list: more than 1,200 urban-industrial complexes from East Germany to China, also ranked. Moscow was priority one with 179 designated ground zeros; Leningrad priority two with 145.
• Part I was an "unrestricted" allocation of about 3,400 DGZs. Part II was a "restricted" list of 1,209 DGZs sized to 69,000 kilograms of "oralloy equivalent", that is, highly enriched uranium.

The weapons were to be Mark 6 atomic bombs of up to 160 kilotons for the city list, and Mark 15, 27 and 36 thermonuclear bombs for the air-power battle, with yields the Archive summarises as 1.7 to 9 megatons. SAC also asked for a 60-megaton bomb, "essential, not only as a deterrent but also to ensure significant results even with a greatly reduced force in the event of a Soviet surprise attack."¹

"Population" as a target

The detail that made headlines is category code 275. Beside the factories and rail yards, every city on the list carries entries whose category is simply Population. William Burr, who edited the release, wrote that "people as such, not specific industrial activities, were to be destroyed", and that targeting civilians in this way "directly conflicted with the international norms of the day".¹ The study also planned to burst its high-yield weapons at ground level, which maximises blast but produces heavy fallout; its authors claimed, wrongly, that "worldwide contamination is minimized when the surface burst is utilized."

"The requirement to win the Air Battle is paramount to all other considerations."

SAC Atomic Weapons Requirements Study for 1959, introduction (1956)

How the list became data

The Archive published scans without a text layer. In 2016 the Future of Life Institute transcribed the released pages by hand, "spread over several hundred, non-OCR-able pages of smudgy, 60-year-old government documents", and Alex Wellerstein, creator of the NUKEMAP, republished the result as a CSV of 1,154 points: 820 in the Soviet Union, 188 in the rest of the Warsaw Pact and 146 in China and North Korea.³ That file is the backbone of the simulation above. It has coordinates and names but no priorities, so we added the priorities and ground-zero counts that the briefing book documents for 25 targets and detonate those first.

What is still missing is larger than what is known. The weapon assignments for every target are redacted, most of Part II's city list was never released, and the Bombing Encyclopedia that would locate each DGZ inside a city remains classified. The released scans do contain, per city, the priority number and the category codes of each installation; transcribing those 306 pages is the obvious next research project, and this page will be updated when it is done.

What the 1956 plan would do today

Wellerstein ran the list through NUKEMAP in 2016 with 2011 population data and warned that his totals double count people inside overlapping blast zones. Our model samples a 1-kilometre population grid for 2025 and counts each cell once, assigning it to the highest-priority weapon that reaches it. The numbers are still enormous.

Blast Radius 3D estimates, 2025 GHSL population, OTA 1979 casualty rates, burns and fires added, no medical collapse, no nuclear winter. Order-of-magnitude figures. Every number is reproducible in the simulator by picking the yield and burst type.

United States, 1945 to 1950: the first atomic plans

American targeting of the Soviet Union began within weeks of Nagasaki. A chart dated 30 August 1945, "A Strategic Chart of Certain Russian and Manchurian Urban Areas", was prepared for General Groves with an estimate of how many bombs would be needed. Its legible "first priority" cities are Moscow, Sverdlovsk, Omsk, Novosibirsk, Stalinsk, Chelyabinsk, Magnitogorsk, Kazan, Molotov and Gorki, with Leningrad in the second tier.⁴ Most are the Urals and Siberian industrial cities to which Soviet war production had been evacuated in 1941.

The plans that followed grew with the stockpile, but only their totals have been declassified:⁵

Dropshot was published as a book in 1978 and its target systems are public: petroleum, electric power, steel and "weapons of mass destruction" facilities. Its list of 100 cities has never appeared as a list, which is why there is no Dropshot simulation here.

SIOP-62 and after: what is still secret

From 1960 the targeting of all US strategic forces was merged into the Single Integrated Operational Plan. Documents released by the National Security Archive in 2004 and 2007 give the shape of SIOP-62 but not its targets: a full pre-emptive strike of more than 3,200 weapons against 1,060 DGZs, including 151 urban-industrial complexes, or an alert force of 1,706 weapons against 725 DGZs. Of the 1,043 DGZs in the plan, 706 were in the Soviet Union; the 337 in China and Eastern Europe are still classified.⁶

The few allocations that have leaked are telling. Leningrad was to receive 9 weapons on 4 targets, Moscow 23 weapons on 6 target complexes, and Kaliningrad 18 weapons on 7 target areas. Eisenhower, briefed on the plan, said it "frighten[ed] the devil out of me". Marine Commandant David Shoup objected that "any plan that kills millions of Chinese when it isn't even their war is not a good plan."⁶ Later SIOPs ran to tens of thousands of potential targets, and none of them has been released with names attached.

Soviet Union and Warsaw Pact: Seven Days to the River Rhine

No Soviet target list of Western or American cities has ever been declassified; Moscow's archives on the subject stay shut. What has come out came from the Pact's junior members after 1989.

On 25 November 2005 Poland's defence minister, Radosław Sikorski, stood beside a fading map in Warsaw and released 1,700 Warsaw Pact documents. The map, from a 1979 exercise called Seven Days to the River Rhine, showed red mushroom clouds over Bonn, Frankfurt, Cologne, Stuttgart, Munich, Hamburg and Brussels, and blue NATO clouds over Warsaw and Prague. France and Britain were left untouched. The exercise assumed NATO struck first, destroying Polish forces on the Vistula, and estimated 2 million Polish civilian dead.⁷

"The objective of the exercise on this map is to take over most of western Europe – all of Germany, Belgium and Denmark."

Radosław Sikorski, Polish defence minister, 25 November 2005

Two earlier documents fill in the south. The Czechoslovak People's Army war plan of October 1964, published by the Parallel History Project in 2000, sends the Czechoslovak front toward Nuremberg, Stuttgart and Munich "immediately after the nuclear strike" and to Lyon within nine days.⁸ A Hungarian command-staff war game of May 1965 names Vienna, Munich, Verona and Vicenza as cities to be "completely destroyed"; Vienna, a neutral capital, was to take two 500-kiloton weapons, the Italian cities one each.⁹ Denmark's first two targets in the 1979 plan were Roskilde and Esbjerg. The whole Rhine attack was to use about 7.5 megatons.

A caveat: the map has never been published as data. Our 16 points are the cities that the Guardian, Reuters and the Parallel History Project reported from the documents, not a tracing of the map, and the yields for Germany, Belgium and Denmark are our assumption of 500 kilotons, the figure the Hungarian document gives for Austria and Italy. East German plans, often rumoured, have not surfaced in the Bundesarchiv.

United Kingdom, 1972: 106 probable targets

Britain never published its own targets in the Soviet Union, but in June 2014 the National Archives released, and the Guardian reported, the list Whitehall drew up of what the Soviet Union would probably hit in Britain. Circulated top secret by Air Commodore Brian Stanbridge on 2 May 1972 and approved by the Cabinet Office, it named 106 places: 38 towns, cities and centres of government, 37 UK and US air bases, 25 control, communications and radar sites and 6 naval sites.¹⁰

The Joint Intelligence Committee's annex assumed an opening salvo of 150 land-based missiles plus submarine-launched ones, and warned that the Soviets "are unlikely to be inhibited by the question of overkill." The stated yields were:¹⁰

• London: two to four airbursts of up to 5 megatons each.
• Glasgow, Birmingham and Manchester: one or two airbursts of up to 5 megatons.
• Swansea and Leicester: 3 megatons, which Home Office scientists called excessive when "one megaton on each target would be sufficient to cause almost complete destruction".
• Saxa Vord, the radar on Unst in Shetland: 3 megatons.

Oxford, Brighton and Aberdeen were absent, and so, the researcher Brian Burnell noted, were power stations and other infrastructure: the planners cared most about whether the V-bombers would survive to retaliate. The underlying file, HO 322/785, is still listed by the National Archives as retained by the department.¹¹

United Kingdom, 1980 to 1982: Square Leg and Hard Rock

The British government also rehearsed the attack. Exercise Square Leg, held from 11 to 25 September 1980, assumed about 130 to 150 weapons of 500 kilotons to 3 megatons, roughly 205 to 280 megatons in all, with the proportion of ground bursts raised deliberately to create fallout problems for the regional controllers. Inner London was not hit directly but was destroyed by strikes on its outskirts. Independent analysts Openshaw and Steadman estimated 29 million dead, 53 percent of the population; the official figure available by 1982 was 8.5 million blast deaths and 2.5 million radiation deaths.¹² The bomb plot survives only in Duncan Campbell's New Statesman reporting and has not been transcribed, so it is not simulated here yet.

Its successor, Exercise Hard Rock, planned for 1982, was cut from 105 to 54 targets and under 50 megatons after criticism, and then cancelled when 20 county councils refused to take part. Campbell published its full attack pattern: 54 weapons on 51 named places, eight of them air bursts, on Runcorn, Walsall, Leicester, Honington, Aylesbury, Brentwood, Hastings and Fawley, and the rest ground bursts from Lossiemouth and Glasgow down to Plymouth and Falmouth.¹³

Targets inside the United States

There is no Soviet list of American targets in the public domain, so every "nuclear target map" of the United States is an American guess. The most authoritative is FEMA's Nuclear Attack Planning Base 1990, a 500-page study from 1987 released in 2005 "for its academic and historical value only". It placed 72 percent of Americans, 175 million people, inside areas at risk from blast of 0.5 psi or more, and published county-by-county risk classes, but not the weapon laydown it assumed.¹⁴

The map of black dots and triangles that circulates every few years as a "FEMA target map" comes from a 2015 television feature that combined a Natural Resources Defense Council scenario with civil-defence data. FEMA has said it "does not, and has not, released any type of formal map of potential nuclear targets".¹⁵

France, China and the other nuclear powers

We found no declassified target list with named cities for France, China, Israel, India, Pakistan or North Korea, and none for Britain's own targets in the Soviet Union beyond the well-documented "Moscow criterion" that shaped the Polaris and Chevaline programmes. Russian archives have released nothing on Soviet targeting of the West. If that changes, this page will too.

How the simulation works and what it leaves out

The simulator on this site normally animates a single detonation in three dimensions over real buildings. A plan with a thousand weapons needs a different approach, so the multi-strike mode treats each target as a point on the globe whose blast, thermal and fireball radii are drawn to scale by the map engine, and plays the attack in priority order with the flashes, rings and fallout plumes of every weapon animated together. Clicking any target opens it in the full 3D view.

• Effects. Fireball, 5 psi and 1 psi blast, third-degree burn and 500 rem radii from the scaling laws in Glasstone and Dolan, The Effects of Nuclear Weapons (1977), the same physics as the single-shot simulator.
• Casualties. People in each 1 km cell of the 2025 Global Human Settlement population grid, multiplied by the blast-zone rates of the US Office of Technology Assessment (1979), with burns and fires added and, for surface bursts, the first-week fallout dose. Cells covered by more than one weapon are counted once, for the first weapon that reaches them.
• Fallout. Idealised Chapter IX plumes with one wind for the whole theatre. Real weather on the day would move every plume; Wellerstein's 2016 experiment with live weather showed Copenhagen inside the fallout on some days and clear on others.
• Not modelled. Multiple weapons on one DGZ, missiles that fail, fires that merge into firestorms, the collapse of medical care, long-term cancers and nuclear winter. Every one of these would raise the toll.
• Population. Today's, not 1956's or 1972's. The point is what these plans would mean now, not a historical reconstruction.

The target data is public domain where it comes from US government records; the 1956 coordinates are credited to the Future of Life Institute and Alex Wellerstein, who digitised and published them. All lists are in plain CSV in the site's repository and corrections are welcome.

Frequently asked questions

Has the United States ever declassified a nuclear target list?

Yes, once in this detail: the 1956 SAC study released in 2015, with more than 1,100 airfields and 1,200 cities ranked by priority. Weapon assignments remain redacted, and no SIOP target list has been released with names.

Did the Soviet Union publish its nuclear target list?

No. The nearest things are Warsaw Pact exercise documents released by Poland, the Czech Republic and Hungary, which name Bonn, Frankfurt, Cologne, Stuttgart, Munich, Hamburg, Brussels, Vienna, Verona, Vicenza, Padua, Roskilde and Esbjerg.

Which British cities were on the Soviet target list?

According to Britain's own 1972 assessment, 24 towns and cities from Glasgow to Maidstone, 14 centres of government including Central London, and 68 military sites. London was expected to receive two to four airbursts of up to 5 megatons.

How many people would the 1956 plan kill today?

Roughly 126 million dead and 161 million injured with one 1.7-megaton air burst on each of the 1,154 digitised targets, counting overlaps once; 47 million dead with 160-kiloton bombs; 209 million with 9-megaton bombs.

Is there a real Soviet map of targets in the United States?

No. US civil-defence studies such as FEMA's NAPB-90 estimated risk by county, and the viral "FEMA target map" is not an official document.

Can I load my own target list?

Not yet through the interface. Each list is a small CSV and JSON file; the repository explains the format, and the single-detonation mode works anywhere on Earth.

Sources