Interview with Philip Wymer Allen, August 26, 2004

Nevada Test Site Oral History Project University of Nevada, Las Vegas Interview with Philip Allen August 26, 2004 Las Vegas, Nevada Interview Conducted By Mary Palevsky © 2007 by UNLV Libraries Oral history is a method of collecting historical information through recorded interviews conducted by an interviewer/ researcher with an interviewee/ narrator who possesses firsthand knowledge of historically significant events. The goal is to create an archive which adds relevant material to the existing historical record. Oral history recordings and transcripts are primary source material and do not represent the final, verified, or complete narrative of the events under discussion. Rather, oral history is a spoken remembrance or dialogue, reflecting the interviewee’s memories, points of view and personal opinions about events in response to the interviewer’s specific questions. Oral history interviews document each interviewee’s personal engagement with the history in question. They are unique records, reflecting the particular meaning the interviewee draws from her/ his individual life experience. Produced by: The Nevada Test Site Oral History Project Departments of History and Sociology University of Nevada, Las Vegas, 89154- 5020 Director and Editor Mary Palevsky Principal Investigators Robert Futrell, Dept. of Sociology Andrew Kirk, Dept. of History The material in the Nevada Test Site Oral History Project archive is based upon work supported by the U. S. Dept. of Energy under award number DEFG52- 03NV99203 and the U. S. Dept. of Education under award number P116Z040093. Any opinions, findings, and conclusions or recommendations expressed in these recordings and transcripts are those of project participants— oral history interviewees and/ or oral history interviewers— and do not necessarily reflect the views of the U. S. Department of Energy or the U. S. Department of Education. UNLV Nevada Test Site Oral History Project 1 Interview with Philip Allen August 26, 2004 Conducted by Mary Palevsky Table of Contents Introduction: Weather conditions caused delays for about half of all atmospheric tests at the Nevada Test Site. 1 Mr. Allen describes his role as a member of the Air Force Office of Atomic Testing ( AFOAT- 1) team in charge of tracking nuclear debris during operation Ranger. 3 Mr. Allen shares his reactions to witnessing nuclear tests. 7 Meteorologists worked to predict and analyze fallout patterns for atmospheric nuclear tests. 9 Weather conditions were also important for underground tests, as unexpected venting could create fallout problems. 11 In 1956, the Atomic Energy Commission ( AEC) established a permanent weather station at the Nevada Test Site. Mr. Allen describes how he and other test site personnel established communication networks, coordinated with civilian contractors, and analyzed weather data. 12 Underground testing created concerns for Area 51, as well as communities near Carlsbad, New Mexico. 17 Mr. Allen describes his role in predicting fallout for the Plumbbob series at the Nevada Test Site. 19 Weather conditions caused concern following the Sedan cratering test. 22 The venting of the Baneberry test placed new emphasis on safety procedures, including weather forecasting. 26 Mr. Allen shares his views regarding the need to balance safety concerns with the rewards of potentially risky scientific exploration. 34 Conclusion: Mr. Allen discusses his testimony in the trials surrounding Baneberry’s release of radiation, as well as the new safety procedures implemented after the incident. 35 UNLV Nevada Test Site Oral History Project 1 Interview with Philip W. Allen August 26, 2004 in Las Vegas, NV Conducted by Mary Palevsky [ 00: 00: 00] Begin Track 3, Disc 1. Mary Palevsky: OK. So Phil, you mentioned that there was a correction you wanted to make to our first interview, and why don’t you go ahead and do that? Philip W. Allen: Yes. In talking about the final atmospheric test that was scheduled on October 30, 1958, the last test before the moratorium began, I gave the name as Madison. That is not the correct name. The correct name, because my memory failed me momentarily, the correct name is Adams. Both of them named after presidents. There is another Madison test at the test site. It occurred later. I don’t believe there has ever been an Adams test. The one scheduled on that October 30 was never fired because of bad weather. And the next subject? Yes, I wanted to ask you, just say a little bit what you said off the record, which was about how you were able to research that. That’s interesting to me. You went back? Oh, I looked back through my notes that I made, listing the length of time the tests were delayed. Starting with the first tests in [ Operation] Ranger, I found in the reading room records of the readiness briefings, and in the readiness briefings, by going through and reading what they said and knowing what happened, I was able to determine the length of delays. It turns out that for the atmospheric tests, the tests that were conducted in the atmosphere, 53 out of a total 105 were delayed somewhat on account of weather. Fifty- three out of—? UNLV Nevada Test Site Oral History Project 2 Out of 105 atmospheric tests at the test site. Now, that 105 atmospheric tests is a little higher than the number you usually hear, which is one hundred, even. The extra five were tests that were conducted below the surface but expected to vent, such as Sedan, where the explosion actually occurred 600 feet below the surface, but we expected it to vent, and there was a large cloud. So I counted it and several others that had been intended to vent, but were called underground tests. So from a meteorologist’s point of view, they’re not truly underground because you’re getting things happening in the atmosphere? That’s correct. The longest delay on account of weather at the test site was Shasta. The Shasta event occurred in August of 1957, and it took seventeen days to find the right combination of weather and device preparation to get the test off. The longest delay that I’m aware of anyplace on account of weather was the event that occurred much later near Hattiesburg, Mississippi. It was delayed approximately a month, almost entirely because of weather. That’s about all I need to say about that, I think. Great. Great. We’re going to have to go through, I think, afterwards and look at— sometimes there seems to be a— this happened up at Livermore. Sometimes there seems to be a difference between the name that’s listed here and the name that was actually used on the test, as well. Someone at Livermore kept referring to a test— he kept referring to Bravo out in the Pacific as Shrimp, and that was what they called it internally. So I’m wondering—? I can’t answer that question. Yes, I’m wondering about that. OK, so, great. Now, as I mentioned, I had hoped you would be [ 00: 05: 00] able to go back a little bit and talk in some detail about Ranger, because I don’t know— I think there’s more there than we talked about last time. UNLV Nevada Test Site Oral History Project 3 Perhaps so. The weather forecaster for Ranger was a good friend of mine. He was employed by the Air Force Weather Service. He was a civilian. But he and I had taught school at the University of Chicago several years earlier. George Cressman, he made the forecasts using the weather station at Nellis Air Base. The briefings occurred in a temporary building. By temporary, I mean it was left over from World War II and was a wooden building. It was just a block or so from the weather station. Before each test, there was a readiness briefing. He gave the briefings. I did not participate. My function during Ranger was to learn as much as I could about the trajectory made by the clouds after the explosion. So you got to Ranger— remind me of how you actually got out there. You mean the reason for which I went? That’s what I mean. I was working at that time for AFOAT- 1 [ Air Force Office of Atomic Testing], which is the name of the organization that was learning as much as it could about foreign nuclear explosions. It sampled the atmosphere in a number of different ways, and also measured the atmospheric pressure resulting from nuclear tests. A pressure wave would go generally all the way around the world. It also measured seismic waves through the ground. But my function was to be able to track the debris from a nuclear test. Right, and you gave me a lot of detail on that last time. For the Ranger tests, I only remember a couple of the trajectories. But the largest test, I think, made a cloud that held together sufficiently well that it was detected up over Utah and across the Midwest and Illinois, not too far south of Chicago, and out across the Atlantic Coast and was still easily measured as the cloud passed north of Bermuda. We didn’t track it any further than that. Obviously the debris from all of the tests contained very tiny particles that would go clear around UNLV Nevada Test Site Oral History Project 4 the Earth for long periods of time. The particles that were in the lower atmosphere would usually be brought down to the surface in rain. Rain is the major cleanser of the atmosphere, and it took all of the debris out of the troposphere, up to the tropopause, which varies from around twenty thousand feet in the winter to fifty thousand feet in the summer. The other nuclear cloud that I remember from Ranger was from the last one, which was one of the smaller tests. It went southwest, not over Las Vegas but it went over the Spring Mountains. I could watch the cloud. It wasn’t very high and it drifted through one of the passes in the Spring Mountains. I don’t know exactly where because I was looking at it from the air base, Nellis Air Base, and didn’t get a very good view. That’s what I was going to ask you, where you’re positioned for this. You’re out at Nellis? [ 00: 10: 00] I was always at Nellis. I carried with me a clinometer, which is a little device used solely for the purpose of measuring angles, elevations, so that I could measure the heights of the tops of all of the clouds. I wasn’t always correct in my measurement because the aircraft would sometimes report an altitude either higher or lower than I had estimated. So let me understand this. You’re at Nellis in some sort of observation area? Just standing on the street. Just standing on the street, with this instrument. Yes. And you—? Tell me what happened. The instrument has a way of— well, I think it simply has a plumb bob on the bottom of it that holds the instrument straight up, up and down, and then you look through a little telescope, and the angle of the telescope relative to horizontal appears on a dial that you read on the side of the instrument. I would get the cloud height by triangulation using that angle. And later I would UNLV Nevada Test Site Oral History Project 5 receive the information from the aircraft that were flying through the cloud to collect samples. They would give me the measured height of the cloud. Other aircraft, longer range aircraft, would fly around the cloud, trying not to become contaminated themselves, and try and keep the cloud in sight. I used their locations to track the cloud. I would put the location on my map of the upper winds and if the airplane seemed to be looking in a direction that the cloud was not supposed to go, according to the winds, I would pass that information to the airplane and tell it where it should go to find the cloud. They would track the cloud as long as they could see it visually without going into it. Then, when they could no longer see the cloud visually, they would track by instrument, detecting the radiation level. But they always tried to stay near the edge of the cloud rather than contaminate the airplane by flying through the main part of the cloud. So you were not in those airplanes ever. You were—? I was never in those airplanes. I was always on the ground. I was in the weather station most of the time, where I had the latest upper air wind data, and I would always have a new map plotted with data when it came in. Upper air data are normally taken every six hours by theodolite and following what is called a pilot balloon. What was the first word you said? Theodolite. A theodolite is a device that measures horizontal angle relative to the Earth, relative to straight north, and a small telescope that can see an object in the sky, and by measuring the azimuth angle— the horizontal angle— and the vertical angle— the elevation angle— and knowing the programmed rate of rise, one can determine exactly where the object is. And in the case of measuring upper winds, the pilot balloon is a small balloon, inflated to maybe three or four feet in diameter, either white or dark color so that it can be easily seen as it rises through the UNLV Nevada Test Site Oral History Project 6 atmosphere. A white balloon is used in daylight [ 00: 15: 00] because it can be seen against the blue sky. But it’s not so easy to see it against a cloud. And then there are probably several hundred pilot balloon stations throughout the country, and in addition to that, there are maybe a hundred radio sound stations. A radiosonde consists of a larger balloon which carries a radio transmitter. The transmitter is attached to a pressure device that has an arm on it, and the pressure arm crosses what looks like a miniature piano keyboard, and as it goes from each key to the next, that is a specific pressure, and as it goes up, it crosses a number of those keyboards, and you can tell exactly how high the device is by the number of keys it passes. Each time it passes a key, from one key to the next, it transmits a radio signal of humidity, because the device also carries a little strip that is sensitive to humidity, and electronically they can tell how high the humidity is. It also carries a thermometer and it gives a continuous temperature reading that is interrupted at each pressure reading. The pressure reading is simply the shift from temperature to humidity and back to temperature. And the observer needs to keep track of how many�� well, this is actually printed out on a chart so that the observer can take it later and at his desk, using a number of charts, a number of tables of information, he can get the temperature, humidity, and pressure. And the radio signal is detected by a radio direction finder that gives him the horizontal presentation of which direction it’s going. There are about a hundred of those stations nationwide, and all of that information is available by teletype to all of the major forecasting stations, of which Nellis had one. And so I had all of that information available to me and could use it to see how far away from the actual wind motion the aircraft were getting, and I could tell them then how to find the cloud back where the wind said it should be. But the aircraft gave me the width of the cloud, and when they finally got to flying through it, they could give me the intensity of radiation in those puffs of radiation. UNLV Nevada Test Site Oral History Project 7 What you’re saying is so interesting and raises several questions. I’ll do them one at a time. You’re out at Nellis. So what are you seeing of the actual explosion when it occurs? The explosions were in Frenchman Flat, and all of the Ranger tests were dropped from an airplane. The airplane would go over and— we were too far away, about sixty- five miles from Frenchman Flat. We could not see the airplane at that point. This is also early in the morning. It [ 00: 20: 00] was not yet light. But when the bomb exploded, it would be just below a range of mountains, low mountains, so that we would not see the direct flash from the explosion, but it would not be any more than, oh, two or three minutes before the cloud had risen to where we could see it. The fact that the flash was below the mountains did not prevent us, though, from getting a lot of reflected light from the sky and from any clouds that happened to be in the sky. Usually the sky was almost clear. And by the time the cloud got above the mountains, where we could see it, or the fireball got above the mountains, it had cooled off enough that you could look at it with the naked eye. So you could actually see the fireball rising. We could see the fireball rising after the first couple of minutes. And there were articles in the newspaper in Las Vegas about being able to read newspapers by the light from the detonation. Those were true, but most of the light was reflected. It was not direct from the fireball. OK. But you could look right at the fireball without protection at that point. By the time it got over the mountains, yes. It would be terribly bright. Well, that’s the obvious human question, is what— you had seen war. What was your reaction? How do you sort of gauge the size? You tell me. In human terms of what’s actually going on. UNLV Nevada Test Site Oral History Project 8 The only part of the war that I saw was in daylight, and these occurred at night. There’s no comparison. I never saw anything to compare with these. To have such bright light over such a great distance was just amazing and interesting. I don’t know what else I could add. I guess what’s interesting sort of to the public and to me and to history is this is this weapon— you’re a scientist so you’re obviously focused in on a lot of the scientific things, but I’m wondering, do you make a connection with war when you see something like that? Do you think about that aspect of it? I did not. To me, this was a physics experiment. And an unusual one, yes, but in those days, I didn’t get excited very easily and I was more interested in what was happening in the device and in the atmosphere. I really did not connect any of these with Hiroshima or Nagasaki or war, although the connection was obvious. Well, I think what’s interesting is that people have a variety of reactions, and I don’t find them to be uniform, at least in my small experience so far. So that’s interesting. After World War II, although as a meteorologist I did not see a lot of the actual fighting, all that I saw directly myself was the invasion of southern France when I had made an early morning forecast and then climbed into an airplane and flew over the invasion point, and I saw plumes of cloud near the ground, far below the airplane, where fighting was occurring. By 1951 when these [ 00: 25: 00] tests occurred, Russia [ USSR] had detonated their first device and we were just making sure that we were ahead of them, that we had more knowledge about nuclear weapons and were in the process of producing a deterrent. And so my interest was, 99 percent at least, scientific. And 1 percent? I would pretty much ignore that. UNLV Nevada Test Site Oral History Project 9 OK, we’ll just let that go. Statistically, it’s insignificant. But that leads me to my next question, which is great, which is when you’re doing your diagnostics and you’re looking at the cloud and how far it travels, explain to me how that helps you on the ground, and correct me, the things that come into my mind are, is it to understand how we detect these tests? Is it to understand danger to populations? What are you trying to understand in addition to the actual sort of pure science of how these things move? How are you applying it? In the back of my mind, as I was watching the clouds, there was the information, of course, that the clouds were radioactive and therefore dangerous. I knew that people had been moved out of the paths of the clouds so no one was being exposed to the clouds while they were of maximum intensity. I supposed that if I had been more of an environmentalist, I would have been worried about the small animals, the chipmunks and foxes and so forth, rabbits— in those days there were a lot of jackrabbits in Nevada— but I didn’t. I would hear the readings come back from the Public Health people who were tracking the fallout pattern on the ground, so that I knew where the cloud was passing on the ground. The ground position of the cloud was not always directly under the upper part of the cloud because winds at the surface are frequently somewhat different from the winds up above the surface. And at that point, I would not keep track of the surface positions of the cloud. My job had to do with the long range transport. That referred more to the clouds from, oh, five thousand feet above the surface on up. And how would that data be used by the people that were going to use it? For future tests or—? That’s what I’m trying to understand also. During Ranger no one was working on the problem of connecting the path of the cloud with the fallout, the intensity of fallout. That was done later, though. Starting in [ Operation] Upshot- [ 00: 30: 00] Knothole, which I believe occurred in 1953. The Weather Bureau section that I was UNLV Nevada Test Site Oral History Project 10 attached to when I was on loan to AFOAT- 1, called the Special Projects Section, was under Dr. Lester Machta. And he started the investigation that really tied it together, the entire cloud from top to bottom to the amount of fallout that had been measured. And he did so using the data back through previous tests, including Ranger. All of that wind data had been preserved, of course, and all of the fallout measurements had been preserved. They first had been preserved, oh, by the Weather Bureau and by other organizations. The intensity of fallout had been preserved by the Public Health Service. It took them several months. By the next test series a year or so later, it was possible for the meteorologists, when they predicted the winds for the test, to give that information to Dr. Machta’s people, and they could add the information that they would get from the laboratories about the contents of the device and the explosive energy, and putting all of that together, come up with a predicted fallout pattern. And those were usually pretty good. The errors were either a result of inability to predict accurately the upper winds or, once in while, the explosive energy would depart from that that was predicted. Right. Right. OK, so that’s helpful. That helps me understand that. So really you’re describing this whole new problem that you have to face and there are various factors. Yes. It was also during this time that, starting with Ranger and the next two or three test series, when the Sandia meteorological people began working on the connection between explosive power and blast through the atmosphere. And they began to predict whether windows would be broken in downwind areas. The blast would be focused by a combination of the vertical temperature structure and the vertical wind structure. Sometimes the focusing effect would be much greater than at other times, and when it was great, tests have been delayed in order to avoid damage to buildings and consequently to people in the ( usually) downwind areas. Upwind focusing did not [ 00: 35: 00] occur; the focusing occurred always downwind. UNLV Nevada Test Site Oral History Project 11 So, great. That gives me— since Ranger was my question, I think that— unless there’s anything you want to add about it. There are probably other things to be said, but I can’t think of them at the time. Great. You’ve given me a lot. Thank you. I might move to the subject of the need for weather support when the tests are held underground. The main reason for considering weather when the test is underground is that there’s never a total 100 percent guarantee that the radiation will be contained underground. When testing first went underground, there were fairly frequent small ventings. Not enough to be detected off the Nevada Test Site [ NTS], but enough that people that were onsite would have to be kept out of the contaminated area. And the contaminated area, of course, depended on the wind direction. So we always predicted winds. We also predicted clouds and rainfall because there were always aircraft operations, even with underground tests, and if there were accidental venting, we did not want that to occur in rain. Also, there were other activities that were hampered by having rain in the area. So there was always a need for weather support, even though the test was underground. Right. So you’re saying that consideration was always made, weather- wise, once we go underground. Yes. Before every test, as long as testing occurred, whether atmospheric or underground or in tunnels, there were readiness briefings beforehand, and those readiness briefings included a weather briefing in which everything was discussed that could possibly affect the safety of people and success of measuring the results of the test. Right. So just chronologically now, we’ve moved from you having been at Ranger to not being at the test site, then you came back in. UNLV Nevada Test Site Oral History Project 12 Yes. I worked for AFOAT- 1, detecting foreign nuclear explosions, from 1949, before the first Russian test, until completely through 1952. Then for three years, from 1952 through about May of 1956, I worked at the National Meteorological Center outside of Washington, at Suitland, Maryland. Then in 1956, there was— well, during all of that period, the weather support for the [ 00: 40: 00] tests here in Nevada was handled by the Air [ Force] Weather Service mobile unit, stationed somewhere in Oklahoma. The tests during that period were in short series, short being from an individual test up to a series of them, twelve or thirteen, over a period of two or three months. But the entire weather activity was handled on a mobile basis by the Air Force. In 1956, the decision was made by the AEC [ Atomic Energy Commission] to set up a permanent weather facility at the test site. And they asked the Weather Bureau to do that because they did not want the military— the military was sufficiently tightly connected to the testing program that it was considered potentially biased in safety considerations, and they wanted that potential bias to be eliminated, so they invited the civilian weather service, the Weather Bureau. I got involved because I was friendly with the people working for Dr. Machta who were still connected with the AEC testing program and still connected with AFOAT- 1. But I personally was working in the other part of the Weather Bureau, the meteorological center. A couple of the fellows from Dr. Machta’s office were visiting in the building where I was one day, and they stopped to talk, and in the course of conversation they said that the Weather Bureau had been invited to set up a permanent weather research facility at the Nevada Test Site. And upon hearing that, I volunteered the information that if the Weather Bureau did that, I would be interested in participating. They took that information back to Dr. Machta, and the next thing I knew, I had a phone call from one of Dr. Machta’s administrators that I was friendly with, asking me if what I had said was true and telling me that I was the first person that had shown any UNLV Nevada Test Site Oral History Project 13 interest in being involved. As a matter of fact, he said that he doubted that anybody in the Weather Bureau would want to go to such a place as Las Vegas, Nevada. But my wife had been with me during the series in 1953. She had stayed in Las Vegas with the kids and had liked the climate so well and the casual living that she was ready to go in a minute. So the Weather Bureau confirmed that I would be in charge, and another fellow came with me. In April of 1956, we surveyed the test site and the surrounding areas, looking for suitable places for weather stations, and made our initial plans, started procuring equipment, and locating people. We consulted with the weather training programs at UCLA [ University of California, Los Angeles] and University of Utah and other places and advertised the positions. [ 00: 45: 00] That was in April, and by June, when I reported for duty in Las Vegas, we had identified four other meteorologists besides me and six or eight meteorological technicians and an electronics technician, and we started the office with those people during the month of June 1956. At that time, we were located in the— the AEC gave us the office building right next to them. They were located at, I think it was 1231 South Main Street, and we were at 1229 South Main Street. That was interesting to me, what you just said. I think we have some of this from the first conversation, but what interested me this time, so it really was mobile up till then. There weren’t permanent weather stations out at the test site until you came. That’s correct. That’s correct. One of the first things we did was to make a daily weather forecast, and set up our own teletype network. We had stations at Tonopah and a couple of places on the test site, to start with. We connected in with Nellis Air Base and with McCarran Airport, both of which had weather stations. And several other scientific organizations at the test site tied into that teletype network, and we would every day put a forecast for the test site. That UNLV Nevada Test Site Oral History Project 14 was the first regular forecasting for the test site, and there were more users than we had anticipated, of course. Interesting. This just seems like such a trivial question, but I’m curious. You come to do this work. Just give me a really brief sense of how that worked. You set up weather stations. Do you tell the AEC? Do you work directly with the contractor? We work with the AEC in the sense that AEC is providing the funding, and they instructed me on how to work with the contractors. That was primarily with Reynolds Electric[ al and Engineering] Company, REECo, and I can remember one of the first people that I talked with in Reynolds Electric Company was Davy Crockett. He was the father of the Crockett lawyer who still practices in Las Vegas. There were several other people in Reynolds Electric Company that we gradually became acquainted with and worked with. They obtained our teletype circuits, which are Weather Bureau circuits but handled by the telephone company, of course, as all communications were in those days. We set up two teletypes. The weather data came in two channels by teletype, so that we needed two teletypes, and a third channel by telephone called a facsimile. People are very familiar with the term “ facsimile” these days, but in those days, “ facsimile” referred primarily to the transmission of weather data from the National [ 00: 50: 00] Meteorological Center where I had worked to the forecasting centers. The telephone signal is received at the forecasting centers by a machine that produces weather maps. In those days, it was a big machine about thirty inches across and two feet high, and it produced weather charts that were of about the size of fifteen by twenty inches. And we received those. We here in Las Vegas would get all of the charts that were produced by the National Meteorological Center in Washington, and we would use them for our basic meteorological analysis and forecasting. But we would prepare our own charts for Nevada and Utah, Arizona, and California so that we UNLV Nevada Test Site Oral History Project 15 could insert any additional information that we have, insert such things as mountain ranges and lakes and so forth, that would affect the weather, and have it in sufficient detail that we could forecast the weather to the required amount of detail here. Now that raises another interesting question, for those who will not even comprehend what you’re talking about, because everything is done on computer now. So explain to me, when you say “ insert the mountain ranges,” do you have some calculating method or…? No, not at that time. That came later, with computers. At that time, it was the maps that we used. The maps were designed specially for meteorological use. They would have circles for towns, and each circle would have a designating series of letters beside it to indicate which town it was. Each circle would be a town that had a weather reporting station in it and would be on the teletype network so that we would get their weather data. The weather data would come in coded form, and we had subprofessionals decode the teletype data and plot it on the weather charts so that for each