Johnston Island had already been established as a launch site for United States high-altitude nuclear tests, rather than the other locations in the Pacific Proving Grounds.

His opposition was motivated by fears that the flash from the nighttime high-altitude detonations might blind civilians who were living on nearby islands.

The nuclear explosions also release heavier debris ions, which also carry an electrical charge, and which also travel in a spiral along the Earth's magnetic field lines.

"[7] Some of the charged particles traveling along the Earth's magnetic field lines cause auroras and other geophysical phenomena in the conjugate areas.

Other charged particles are reflected back along the magnetic field lines, where they can persist for long periods of time (up to several months or longer), forming artificial radiation belts.

[8] According to the Operation Fishbowl planning document of November 1961, "Since much valuable data can be obtained from time and spectrum resolved photography, this dictates that the test be performed at nighttime when auroral photographic conditions are best.

The first planned test of Operation Fishbowl was on June 2, 1962, when a nuclear warhead was launched from Johnston Island on a Thor missile just after midnight.

No nuclear detonation occurred and no data was obtained, but subsequent investigation found that the Thor was actually following the proper flight trajectory.

Navy Explosive Ordnance Disposal and Underwater Demolition Team swimmers recovered approximately 250 pieces of the missile assembly during the next two weeks.

The Starfish Prime electromagnetic pulse also made those effects known to the public by causing electrical damage in Hawaii, about 1,445 kilometres (900 mi) away from the detonation point, knocking out about 300 streetlights, setting off numerous burglar alarms and damaging a telephone company microwave link[10] (the detonation time was nine seconds after 11 p.m. in Hawaii).

In addition, a large number of rocket-borne instruments were launched from a firing area at Barking Sands, Kauai, in the Hawaiian Islands.

[12] A very large number of United States military ships and aircraft were operating in support of Starfish Prime in the Johnston Island area and across the nearby North Pacific region, including the primary instrumentation ship USAS American Mariner providing measurements conducted by personnel provided by RCA Service Company and Barnes Engineering Company.

An interesting side effect was that the Royal New Zealand Air Force was aided in anti-submarine maneuvers by the light from the bomb.

[15] Telstar I lasted the longest of the satellites that were clearly damaged by the Starfish Prime radiation, with its complete failure occurring on February 21, 1963.

[4]On July 25, 1962, a second attempt was made to launch the Bluegill device, but ended in disaster when the Thor suffered a stuck valve preventing the flow of LOX to the combustion chamber.

With the Thor engulfed in flames, the Range Safety Officer sent the destruct command, which split the rocket and ruptured both fuel tanks, completely destroying the missile and badly damaging the launch pad.

The radioactive contamination on Johnston Island was determined to be a major problem, and it was necessary to decontaminate the entire area before the badly damaged launch pad could be rebuilt.

According to the Operation Dominic I report, "The enforced pause allowed DOD to replan the remainder of the Fishbowl series.

[9]: 241 On October 19, 1962, at about 90 minutes before midnight (local Johnston Island time), an XM-33 Strypi rocket launched a low-yield nuclear warhead which detonated successfully at an altitude of 147 kilometres (91 mi).

It was reported that, "Observers on Johnston Island saw a green and blue circular region surrounded by a blood-red ring formed overhead that faded in less than one minute.

"[9]: 241 The fourth attempt at the Bluegill test was launched on a Thor missile on October 25, 1962 (Johnston Island time).

One of the official reports for the project stated that, for the altitudes planned for the Bluegill, Kingfish and Checkmate tests, "the thermal-pulse durations are of the same order of magnitude or shorter than the natural blink period which, for the average person, is about 150 milliseconds.

"[8] Two cases of retinal damage did occur with military personnel on Johnston Island during the Bluegill Triple Prime test.

[17] As with the other Fishbowl tests, a number of small rockets with various scientific instrumentation were launched from Johnson Island to monitor the effects of the high-altitude explosion.

[9]: 247 According to the official report, at the time of the Kingfish detonation, "Johnston Island observers saw a yellow-white, luminous circle with intense purple streamers for the first minute.

At Oahu a bright flash was observed and after about 10 seconds a great white ball appeared to rise slowly out of the sea and was visible for about 9 minutes.

Prompt gamma ray output measurements on these later tests were also carefully obtained so that a new theory of the mechanism for high-altitude EMP could be developed and confirmed.

[19] As of the beginning of 2011, the EMP waveforms and prompt gamma radiation outputs for Bluegill Triple Prime and Kingfish remain classified.

An unclassified report confirms that these measurements were successfully made and that a subsequent theory (which is the one now used) was developed which describes the mechanism by which the high-altitude EMP is generated.

According to a Sandia National Laboratories report, EMP generated during the Operation Fishbowl tests caused "input circuit troubles in radio receivers during the Starfish and Checkmate bursts; the triggering of surge arresters on an airplane with a trailing-wire antenna during Starfish, Checkmate, and Bluegill; and the Oahu streetlight incident.