The announcement

Van Allen’s quiet disclosure of our discovery to IGY program officials in Washington in April, combined with the setting of a date for a public release on 1 May, set in motion a feverish effort in our laboratory to prepare the material in a written form that would be easily understandable and thoroughly convincing.

Most of the rest of the month of April 1958 is a blur in my memory! On the first workday after my return to the Iowa campus, i. e., on the same day that Van Allen was calling Washington to reveal our initial Explorer findings, I left for Huntsville on a several days’ session for Juno II planning. In spite of the exciting revelations from the Explorer data, I had to remain heavily involved in designing our instrument for that new mission. Even that effort was soon trumped by the evolving Argus project, as detailed in the next chapter.

The two new projects had the effect of precluding my attendance at the history­making gathering in Washington, where Van Allen made the trapped radiation an­nouncement on 1 May 1958.

In preparing for the public announcement of the new discovery, Van Allen, Ernie Ray, Carl McIlwain, several other students, and I were engaged during the second half of April in working up the Explorer I and III data and preparing the written paper. Ernie Ray took the lead in the data-processing and plotting efforts, while Van, with help from Carl, Ernie, and Joe Kasper, developed the physical interpretation.

The group’s efforts during that period can best be summarized as (1) refining the plot of the GM counter response to high-intensity radiation; (2) further organizing the Explorer I data and plotting them in a suitable form; (3) plotting Explorer III

OPENING SPACE RESEARCH

playback data from additional passes, to the extent that tapes were received in time; (4) developing and organizing a cogent physical explanation for the newly observed phenomenon; and (5) preparing the written paper.

Step 1: The Geiger-Miiller counter performance As soon as I arrived back in Iowa City with the spare Explorer I payload, Carl exposed it to his X-ray beam. He hardened the beam (that is, eliminated the lower-energy particles) by placing a three-eighth inch thick brass absorber between the X-ray tube and the GM counter. During his runs, he varied the X-ray tube voltage between 50 and 90 kilovolts and the distance between the X-ray tube and GM counter over a substantial range. Additionally, he used various lead shields between the source and counter. Those three techniques resulted in the production of a very wide range in the X-ray flux at the location of the counter.

Carl produced the semilog plot shown in Figure 12.5 to portray the effect of the Explorer III onboard encoding and recording arrangement.36 Since the linear scale on the vertical axis extended only to 350 counts per second, the entire upper portion (up to an observed actual rate of more than 1400 counts per second) is cut off. The color coding described in the figure caption was used in all subsequent plots of the rates from the satellite’s GM counter.

The flat portion of the curve at 128 counts per second reflects the limitation in transmitted data rate resulting from the encoding associated with the onboard tape recorder in Explorer III. That encoding had been designed to accommodate a rate substantially greater than the maximum rate expected from cosmic rays but was, of course, completely inadequate for the unexpected intense radiation.

Interpreting Figure 12.5 in terms of the flight data, the transmitted rate accurately tracked the GM counter’s incident flux rate at values from zero to 128 counts per second. For rates increasing above 128 events per second, the onboard tape recorder encoding circuits limited the reading to that value. That persisted to a counting rate that would have been about 15,000 counts per second in a zero-dead time instrument. As the flux increased still further, the GM counter pulses piled up at the scaler input, as described earlier, so that more and more of the resulting pulses were incapable of triggering the scaler. In that region, the scaler output rate dropped off, as seen by the sharp decline between 15,000 and 30,000 counts per second near the right of the figure. With still higher incident fluxes, none of the pulses was large enough to trigger the scaler, and its output remained constant, appearing to signify a zero counting rate.

Step 2: Organizing the Explorer I data Most of the earliest work on the Explorer I data was with recordings made at the JPL Microlock stations in California (the first data recordings to arrive in Iowa City). But by 12 April, data were beginning to trickle in from Minitrack stations in South America.

We prepared many different exploratory plots in order to arrive at the most un­derstandable way of summarizing and presenting the Explorer I data.37 From those,

CHAPTER 12 • DISCOVERY OF THE TRAPPED RADIATION FIGURE 12.5 The response ofthe GM counter in the Explorer I spare payload upon exposure to high-intensity X-rays in the laboratory. The pencil trace above the flat region shows the rela­tionship between the incident flux and the observed counting rate from the bare GM counter as measured by a fast laboratory instrument. That curve would have peaked at a value of about 1400 ifthey-axis ofthe plot had been extended that far. The truncation at 128 actual counts per second shows the limiting effect ofthe scaler and encoding circuits. The original plot was color coded, with the ascending portion at the left being blue, the flat top green, and red for the zero level on the right. (Courtesy ofthe University Archives, Papers of James A. Van Allen, Department of Special Collections, University of Iowa Libraries.)

several formats were selected for inclusion in our announcement paper. The first was a simple plot showing the counting rate as a function of height for data received from 25 passes over the California stations.38 It convincingly conveyed a picture, even at the high latitude of those stations, of a rising intensity as the altitude increased, from counting rates of about 25 to over 100 counts per second.

As Explorer I data from locations closer to the equator were examined, the picture began to come into sharper focus. One figure included in the discovery paper showed the geographic latitudes and heights for 12 cases. The counting rates were normal in four of those cases (about 30 counts per second), no counts were observed in seven cases, and one case showed a rate in transition.39 In summary, counting rates below about 400 miles height were normal, regardless of latitude, while those above about 1200 miles appeared to be zero.

OPENING SPACE RESEARCH

The third plot of Explorer I data included in the paper showed the results of seven high-altitude passes over the Quito, Ecuador, station and one over Antofagasta, Chile. It showed that four high-altitude passes occurring between about 0 degrees and 20 degrees south latitude produced counting rates that appeared to be essentially zero. Four other passes farther from the equator showed rates that were in transition between normal and zero counting rates.40

Steps 3,4, and 5: The Explorer III data and preparing the paper Turning to the Ex­plorer III data, the first task was to complete and refine the preliminary plot of data from the 28 March San Diego Explorer III readout that Van Allen had drawn in his hotel room. The improved full plot, included here as Figure 12.6, shows the counting rate as a function of time for the complete recorder dump.41 Although this particular plot is undated, it was probably produced at around the time of our pivotal meeting on Saturday, 12 April.

An examination of the final packing lists for the tapes produced for us at NRL revealed that copies of readouts of the onboard recorder were copied at NRL onto five reels of magnetic tape.42 The first reel, dated 4 April, contained data from nine onboard recorder readouts that took place on 27 and 28 March. The list for the second reel, dated 7 April, lists 24 data readouts that took place on 28 March-5 April. List three, dated 9 April, records 24 additional readouts. Packing list four, dated 14 April, lists 27 readouts. The fifth and final list, dated 17 April, lists 17 readouts. It can reasonably be assumed that the tape reels arrived in Iowa City shortly after the dates on those lists. The data on the first reel were of largely unsuccessful readout attempts on the first two days of the satellite life and were not useful. But reels two and three contained a gold mine of information.

Many exploratory data plots were tried. Some of them turned out to be of little use, but two turned out to be invaluable. One was a set of plots of radiation intensity as a function of height above the ground and geographic latitude for nine orbits on 28 through 31 March.43

Those nine charts were merged for the next plot, which highlighted the regions of high radiation intensity as a function of position over the globe. The original chart was color coded but was redrawn in black-and-white form for the disclosure paper.44 The information from that figure, overlaid on a world map, appears here as Figure 12.7. The oddly shaped contour marked “300” over South America and the South Atlantic shows the region where the magnetic field at the Earth’s surface is a minimum, and therefore, where the region of trapped radiation dips closest to the Earth. The figure shows the strong correlation between the highest-intensity radiation readings from the flight data (the bold segments) and the field minimum. The displacement of the bold segments somewhat to the east of the region of minimum magnetic field resulted from the fact that the satellite’s altitude was increasing during those segments and therefore climbing deeper into the radiation region.

CHAPTER 12 • DISCOVERY OF THE TRAPPED RADIATION

FIGURE 12.6 Final plot of the San Diego station readout of Explorer III on 28 March. Since the onboard recorder playback occurred in reverse order, the right end of this plot represents the be­ginning of the recording interval, while the left end is the end of the recording and the time of the data readout. The horizontal axis scale at the top of the plot shows the actual universal time of recording. (Courtesy of the University Archives, Papers of James A. Van Allen, Department of Special Collections, University of Iowa Libraries.)

This information provided one of the most convincing arguments that the satellite was observing a region of high-intensity radiation.

Our results were first prepared as a Department of Physics research report.45 Al­though dated 1 May 1958, it was actually completed well before 28 April, since it

had reached Washington and was being distributed on that date, marked “Release May 1.”

Our report’s opening included a brief, simple summarizing statement:

[We] conclude that [the blanking of the G. M. counter] is not due to equipment malfunction, but is caused by a blanking of the Geiger tube by an intense radiation field. We estimate that if the Geiger tube had had zero dead time, it would on these occasions have been producing at least 35,000 counts/sec.

It continued with a discussion of the instrument and a more complete summary of the preliminary observations. A third major section contained an interpretation of the data that served to justify the claim that the zero counting rate resulted from exposure to very intense radiation. The concluding section, dealing with implications, dwelt on three major points:

The particles are unlikely to have as much as several BeV of energy each. They must initially be associated with plasmas that seriously perturb the magnetic field at an Earth radius or so.

CHAPTER 12 • DISCOVERY OF THE TRAPPED RADIATION

The energy loss in the residual atmosphere above 1000 km may contribute significantly, if not dominantly, to the heating of the high atmosphere.

There are obvious biological implications of the results.

Throughout this chapter, the singular term region of high intensity radiation, Van Allen Belt, or some variation has been used to describe the discovery. This emphasized the fact that, at first, we thought that we were dealing with a more or less homogeneous phenomenon located in a single large region around the Earth. A little later, after examining the data from Explorer IV and Pioneer 3 (as described in the following chapters), it became clear that there were two distinct regions. Soviet data substantiated that finding. After that, the collective term Van Allen Radiation Belts (plural) was widely used.