Interview with Alexander G. Smith

Begin Tape 45B

Sullivan

This is talking with Alex Smith at the Haverford AAS meeting, 21st June 1976. Can you just tell me a little bit about your background before you got into radio astronomy?

Smith

I was educated as a physicist. I was always very keen and interested in astronomy. I suppose if the field had looked better I would have gone into it but in those days, which I guess you might say would have been in my late teens when I was selecting careers, and that sort of thing, it was a sort of situation where you could just about enumerate the number of decent jobs in astronomy in the country on the fingers of two hands.

Sullivan

You had come full circle.

Smith

Right, not quite full circle but we’re tending in that direction. So I was always very keenly interested in astronomy and did the usual bit of building and grinding my own mirrors, and building telescopes in my teens and so on and so forth. When push came to shove, I elected to go into physics and went into experimental physics and did my undergraduate work in the experimental physics program at MIT. And worked at the MIT Radiation Lab on radar, of course, during World War II. I graduated in the early stages of the American involvement in the War, and I think there were 17 students in my senior physics class at MIT. And I believe of the 17, 13 went into the Radiation Lab to work on the development of radar. As a matter of fact, I did my senior thesis over there. After the War -

Sullivan

Did you get your PhD there?

Smith

No. I got my bachelor’s degree there in ’43 and then stayed on 3 years during the remainder of the hostilities working at the Radiation Lab.

Sullivan

Do you know anything of Dicke’s work which was actually just after that?

Smith

Oh, I, you know, knew Dicke by name there, but I didn’t really know him. I wasn’t involved with any of that stuff. I was designing and testing magnetrons. We were very tightly compartmentalized. There was this need to know sort of concept to try and keep anybody from knowing any more than they absolutely had to know in order to do their job. As a matter of fact, my wife didn’t even know that I was working on radar. So when the Lab was winding down at the end of the War, Walter Gorde, who was there was going to Duke as a professor and he went around scouting up some graduate students and I was one of the ones he scouted up and made a very attractive offer. So I went down to Duke to do my PhD work in microwave spectroscopy. We did microwave spectroscopy on molecules for a couple of years there. I actually finished up my degree in only a couple of years – my Ph.D.

Sullivan

What year was this?

Smith

’48.

Sullivan

What molecules were you doing?

Smith

Oh, the ClCN - halogen cyanide, mostly is what I worked on. [? And ?] Didn’t do FCN. It wasn’t too practical at that time. ICN.

Sullivan

Did you envision at all this astronomical spectroscopy that has happened in the last 5 years?

Smith

Well in a way I did. I went to, with my Ph.D. clutched in my hand, immediately in ’48, went to the University of Florida where I have been ever since. I started in, as every PhD’s young dream is, as to keep on doing exactly what he did for his doctoral dissertation because it’s one thing he knows he can do or at least he thinks he knows he can do. And he’s not sure he can do anything else. And so I kept on working for several years in microwave spectroscopy at Florida. It eventually got to the point where it seemed like it had become more of a chemist’s game than a physicist’s game, that is, in essence the physicists had demonstrated the principle and developed the equipment and so on. And it became more and more a proposition of trying to dream up molecules whose rotation frequencies were in the right wavelength range and vapor pressure was such that it would work in microwave spectrograph. And as I say it seemed to be edging more toward a chemist’s game than a physicist’s game.

Sullivan

[?]

Smith

Yeah, oh yeah, all kind of boundary conditions that had to be met. So I became that involved in some work in atmosphere optics. This is in the early 50’s when I made the transition. And the Korean War, if you remember started up in that era, and I got involved as a leader actually of a rather large project in VHF electronics. I suppose it is no longer classified information, proximity fuse work, for the Navy and was on that rather intensively for about 3 years. And then got involved in some atmospheric optics work for the Air Force and the Army. This led more or less, you might say, to a renewal of my interest in astronomy. At least I was looking in the right direction.

Sullivan

But you said you had some inkling of the astronomical spectroscopy, how did that - ?

Smith

Ok, I forget the exact date, but I think it was like the middle 1950’s, pretty close around ’55, ’56. I made a proposal, which looking back on it, I think was probably ahead of its time, to do millimeter wave radio astronomy.

Sullivan

Spectral line were you thinking of?

Smith

Yeah. You know it was a reflection of my background in microwave astronomy and I’d become as I say, I’d started looking up again.

Sullivan

You had the hydrogen line -

Smith

And putting this together with my interests in microwave spectroscopy my interests in astronomy. I said, “Gee, you know, maybe there’s some future in trying to do this on the sky.”

Sullivan

You knew about the optical CHCN, I suppose?

Smith

Yeah.

Sullivan

And you figured why not? Now who was this proposal to?

Smith

I believe as I recall, it was to the Army for support. And it was a pretty full-blown proposal and got turned down [laughter]

Sullivan

This is about ’55, you think?

Smith

Somewhere in that vicinity.

Sullivan

Gee, do you think you could dig out a copy of that in your archives?

Smith

Oh, I could try.

Sullivan

That would be fascinating to see that. It was about that same time that Townes had an article which you probably know about now, at a symposium in England I think it was, suggesting various microwave lines as possibilities.

Smith

Yeah, of course Townes was one of our competitors in the microwave spectroscopy business.

Sullivan

Yeah, but you’re the first person I’ve heard of that really wanted to get a full-scale program going. So having that turned down you decided to just give it up?

Smith

Well, just about this time, Tom Carr who had done his undergraduate work and his Master’s thesis at the University of Florida, he had started to do graduate work at Chicago and then was stopped because of illness. He went through a period of rather uncertain health. And he started to do some work at Chicago and had to stop because of his health and then somewhat later he tried again at Duke and had to stop again because of his health. And then he was involved in the Bikini tests and then I guess it was roughly after, that he went to what then was Cape Canaveral, I guess it again is Canaveral, and became very heavily involved as a civil service scientist in their radio propagation work down there. And then once more he became very interested in going back to school and finishing up his PhD and he contacted us in what was then just the Department of Physics. It later became Physics and Astronomy at the University of Florida. So he came up – there was some kind of a meeting there of some radio types, and I forget now what it was. I think it was sponsored by the College of Engineering. Anyhow, a lot of radio types running around all over the place. And so he came up for this occasion and we talked with some of those radio types to try to get some advice about what, if anything we could do in the way of radio astronomy that wouldn’t cost a million dollars. And I just at the moment can’t think of his name but there was a Frenchman there who did some early observing. In fact, he wrote a chapter in a book, the University of Chicago series -

Sullivan

Oh yes, the Kuiper series?

Smith

Yeah.

Sullivan

Steinberg, maybe?

Smith

No. Wasn’t Steinberg. He was very active for a short while in radio astronomy, particularly in observing Jupiter. Then he went to the Bureau of Standards and I haven’t heard anything from him since. I don’t know what happened to him but he was a very clever, very genius guy, very full of ideas. And he was the principal one we talked to and he told us about the very recent discovery of the Jupiter radiation and said you know, “Gee, here’s something that’s strong enough that you could observe, with relatively simple antennas, it just seems made to order.” So we said, “Swell.” And we got a few hundred bucks from the Physics Department budget which was a significant fraction of the Physics Department budget in those days.

Sullivan

This was about how long after Burke and Franklin’s - ?

Smith

Less than a year, 8 or 9 months.

Sullivan

Before we get off on this I just have another question about your proposal for millimeter spectroscopy. Can you remember which molecules you were particularly interested in looking at, what frequencies?

Smith

No. I just don’t remember the details.

Sullivan

It would be good to dig that up.

Smith

I’ll have to see if I can’t find that. It would be interesting to me. I haven’t seen that for a long time. Anyhow, we put up – we had a student observatory which I had been particularly responsible, optical observatory. We had gotten a gift of a very fine 8” Clarke, actually the lens was by Clarke, refractor. And had put up a student observatory on campus with the principle (?) essentially the only instrument. So Tom Carr and I, helped very shortly by Colin Barrow who came over from England as a Fulbright student, put up an array of dipole broadside array, I think it was something like 8 quarter wave or half wave dipole. You know the sort of thing you get out of the Radio Amateur’s Handbook – at 18 MHz, it was then megacycles, in order to listen for Jupiter. We got the thing on the air in 1956. But our first list was I guess about the tail end of the Jupiter apparition and we had our first real full apparition in 1957 which by great good fortune was timed when Jupiter was extremely active. We got enough results so that we got one or two papers out of it.

Sullivan

Right. I was just noticing your first paper I think is ’57 Nature -

Smith

Right.

Sullivan

And what were your results at that time?

Smith

We detected copious quantities of Jupiter radiation so that we were able to plot histograms of the current probability vs. central meridian longitude. And, of course, confirmed the fact that there were certain apparent source regions as people had come to call the 3 primary regions from which, or at least central meridian longitudes, which seemed to generate more Jupiter noise when turned toward the earth. And we also got a good measure of the rotation period and which put together really with the earlier results gave us a better handle on the rotation period than had been possible because we had simply extended the time baseline and that always give you -

Sullivan

You mean Shane’s results?

Smith

Yeah, Shane’s results. We then proposed, I guess one of our earlier results, was proposing a system 3 longitude system based on this new 1957.0 how it came to be designated because 0 meridian was on the central meridian and on New Year’s January 1, 1957. And proposed this longitude system with the rotational period that we had determined.

Sullivan

Is that the one that has come to be known as System 3?

Smith

Yeah. The radio system as System 3 of course has been revised. In fact, it just very recently has been revised again to a 1965.0 system. Jim Warwick is the principal (?) detected the latest revision.

Sullivan

I noticed here that you talked about this System 3 at an APS meeting and then published it in Ap.J. Apparently you felt like you should talk more to the physicists than the astronomers.

Smith

Well, we were physicists [laughter]

Sullivan

Yeah, that’s how you thought of yourself.

Smith

Yeah, we were members of the Physical Society at that point I guess rather than the Astronomical Society. I don’t really remember when I joined the Astronomical Society. It’s like so many people, of course, who came into astronomy through a physics background. That was where we’d been giving papers, the natural thing to do.

Sullivan

That’s also the same paper in which you in the abstract mention a sporadic radiation from Saturn. What can you tell me about that?

Smith

Of course, it’s a very natural extrapolation since the planets – or at least Saturn is I guess one would say, superficially the planet that would most closely resemble is Jupiter and if you know all these fireworks coming from Jupiter it doesn’t take an awfully lot of ingenuity to say, well, you know, maybe Saturn does it too. And so we early on started listening to Saturn and also to Uranus for obviously the same reason. On a number of occasions, we thought that we possibly had heard bursts coming from Saturn. And when we put together these occurrences they even gave a sensible rotation period. But the radiation was always so weak and so infrequent, nothing like the strong persistent noise storm which one frequently gets from Jupiter, that we were never willing to stake our reputation on it and say with any degree of certainty that we had heard noise. We simply reported it as a possibility. But we never made any positive claim that we had heard noise from Saturn. Now I think the Yale people were in exactly the same boat. Harlan Smith and Jim Douglas at Yale. They had the same sort of thing and unfortunately none of our observations coincided, the Yale and Florida observations. It remained in that status essentially indefinitely.

Sullivan

Did anyone try and make a confirming program or anything like that?

Smith

Well, we did. We observed Saturn in more recent years with more powerful equipment and sort of had, I guess you might say, the uncomfortable result that the more powerful the equipment was the less Saturn we seem to receive. (laughter) Of course, finally last year I guess it was, Brown at Goddard apparently succeeded in detecting Saturn bursts in the satellite records. And since the radiation bursts as he reported them seemed to peak in the vicinity of 1 MHz, it seemed quite possible that they may not really be observable from the ground. We’ve got an antenna now which was built in part to listen to Saturn and in probably in larger part to back up the RA 2 frequencies. 6.55 MHz, and we have done a fair amount of Saturn listening with that antenna without getting much satisfaction. And it may simply be that the Saturn radiation is at sufficiently low frequencies so that it just doesn’t essentially penetrate the ionosphere in detectable amounts.

Sullivan

What about the milieu at that time in the late 50s? It seems to me like people were claiming comets for giving off low frequency radiation and Venus had bursts. I think it was Kraus that was claiming that. There were a lot of these things – just a little bit too much enthusiasm?

Smith

Kraus had fitted actually the Jupiter radiation – the Venus radiation to an interferometer pattern. You see, we didn’t even have that comfort because we were not using interferometry. Well, we had interferometry but we were not using them, I guess you might say routinely on Jupiter. Basically, we didn’t feel it was the way to go.

Sullivan

Why was that?

Smith

I think probably several reasons. One was that we felt that the interferometer pattern was sort of restrictive in terms of observing the actual morphology. You know, you’re super-imposing the fringe pattern on top of the morphology of the Jupiter noise storm and we were really more interested in mapping the Jupiter noise storm without pollution of the interferometer pattern. I guess this was one reason. And another reason was we didn’t think we needed it because the mode of operation that we developed early on was to always work with an operator. A live operator to identify the noise and mark it on the records, eliminate interference if possible if it came along. And our experience was that we could identify genuine Jupiter noise with essentially 100% accuracy with the live operator technique. I think this was confirmed to a certain extent, I’m a little hesitant to say this, by Jim Douglas’ experience at Yale, where they did try to operate with interferometers and as I understood the situation, essentially got so little data that in order to complete his doctoral dissertation, Jim Douglas had to write us for our data. So this made us think we were really on the right track.

Sullivan

So you are saying that Jupiter was in view so much less time with the small fringes of the interferometer that you lost a lot of data?

Smith

Right. And when you combine the highly sporadic nature of the Jupiter radiation with what I guess you could term the sporadic nature of an interferometer pattern you require double coincidence in order to get your data [laughter].

Sullivan

And in fact, what your interests were in seems to be time resolution which the single antenna could do as well. Rather than angular resolution assuming it all came from Jupiter.

Smith

Yeah. Assuming you could identify your noise – it’s sufficiently distinctive that we felt we could do it with essentially 100% accuracy. And after we got our Chile station in operation we had a way of confirming this. It’s sort of amusing that we would have instances where we would have Jupiter noise maybe, let’s say, a few centimeters high above the baseline in the face of thunderstorm static that was penning the pen at full scale and the observer would mark in the log, “Gee, I’m almost sure I hear Jupiter under all this thunderstorm noise.” And later when we’d compare this record with the Chile record sure enough this was right in the middle of a Jupiter noise storm. It’s surprising that once you’ve listened to it there’s virtually no possibility of confusing the Jupiter noise with terrestrial static.

Sullivan

Just from listening though?

Smith

Yes, just from listening. You can well believe we spent a lot of time trying to think of circuitry that would do this but we never found anything that would do it as well as graduate student. It’s rather odd, the thing which is most apt to be confused with it orally is station. What you probably heard, if you done much radio listening, a station trying to fade in from a distance and going, “sh sh sh sh sh sh sh sh.” That sounds like Jupiter. If all you can do is listen there are many times when your station starts to fade and create interference when you wouldn’t be sure it was Jupiter. You see, there you’ve got a hand on the situation so all you do is grab hold of a knob on the set and turn it a little. And if the station goes away and Jupiter doesn’t. So this, of course, is what the observers are trained to do.

Sullivan

Can you give an example as you’ve already done, the main difference between what Jupiter sounds like and what thunderstorm - ?

Smith

Well, of course, everybody has heard thunderstorms on the radio. You get these very sharp crashes. The Jupiter noise, well there are really two types that observers distinguish now, the L bursts, the “L” standing for long and the S bursts, obvious “S” stands for short. The L bursts, the best analogy that I know of I think is one that Colin (Barrow?) was responsible for at least the first time I recall hearing of it was waves breaking on a beach. Ocean waves breaking on a beach – just sort of slow roaring rumbling sound – sh…. That sort of thing. Nothing at all like the sharp crash that you get from terrestrial static. And it’s almost impossible vocally to emulate that S bursts but the best analogy there that I’ve heard anyone – maybe I even thought of this one – was hail falling on a tin roof. It would occur to many Floridians. But yeah, that’s just a very sharp rattling kind of sound, again not at all likely to be confused. I guess the L bursts are too long. The wave form is not steep enough to confuse with terrestrial static and I guess with the S bursts the wave front is too steep. They’re both in different time domains from the terrestrial static.

Sullivan

Ok, we got off on this when I brought up the question of -

Smith

Of Saturn.

Sullivan

Well, of Venus.

Smith

Of Venus, yes.

Sullivan

People were going wild it seems with these low frequency detections of solar system noise.

Smith

The point being that John Kraus felt that he had this radiation from Venus. He was fairly positive about the identification because it perfectly well matched his interferometer pattern. But apparently, you know in the light of hindsight which is always the best kind, it must have been terrestrial interference as he rather soon recognized himself. And we always felt that basically the problem was that he did his, as I recall the situation, he did his observing in the evening at an eastern elongation of Venus and we tried to confirm it. We always did our observing at a western elongation just before dawn in morning, which from a physiological point of view is less desirable. But from an ionospheric point of view it’s vastly quieter. That’s, of course, the quietest time at decameter wavelengths, just before dawn and of course in the late afternoon is the noisiest time.

Sullivan

You tried to confirm these Venus - ?

Smith

Oh yes, we put in a lot of time observing Venus as a result of Kraus’ observations.

Sullivan

Did you ever publish not finding it?

Smith

Yes, I believe so. Surely we did.

Sullivan

I didn’t find it in the abstracts anyway.

Smith

Yeah. Surely we did. We had a paper in which we summarize a number of, I think, it was in Vol. 134 of Ap.J.

Sullivan

Ok, I only went up to 1960 so this is getting later.

Smith

Yeah, this would probably have been one of our papers in the early 60s in which we summarize observations of several planets in one paper. And I’m sure that we reported the negative results on Venus.

Sullivan

What about comets? It seems like a lot of people were enthusiastic over – there was one particular comet, ’58 or so -

Smith

Was it (?) the one that had the spike?

Sullivan

No, that was ’65 or so I think.

Smith

No. This was ’57.

Sullivan

Ok, I’ve got them mixed up. Anyway, there was a comet (?) there were several reports but they disagreed with each other as to the nature of it.

Smith

We tried to observe – we did some observing of a comet, passage of a comet, and I don’t remember what year it was. It conceivably could have – no, I think it must have been later. Yeah, it was later than that because I remember we had moved our observatory from the original site when we had this campaign on the comet. Of course, we had an interferometer operation at that time, we never had any interferometer at the old site. We observed – this initial site with the student observatory we only observed there about 2 apparitions then we moved. We very quickly ran out of room for antennas. Then we moved to another site still on campus but on the fringe of campus on agriculture land and we had built an interferometer there, 22 MHz interferometer, by the time we tried to do these comet observations. And so I don’t remember exactly what year it was but I do remember that we, among others, tried to make observations of this comet and came up absolutely nothing.

Sullivan

Do you think that there were these various (?) probably over-enthusiastic interpretations simply because people were not appreciating the problems of noise interference?

Smith

Probably so because, of course, of all the radio astronomical bands it’s the noisiest from a number of points of view the harder – I guess most of us think of the decametric spectrum lies right across the ordinary short-wave band where of course there is a tremendous amount of activity. There’s also bands in there where people use all kinds of infernal machines like (?) machines and plywood machines where they use electromagnetic feeding for gluing the plywood and so on. So there’s an awful lot of manmade interference and power line leakage, leaking neon signs. I hate to tell you how many hours we spent driving up and down highways with portable radios looking for which motel it was whose neon sign was leaking and causing interference at the radio observatory. We’d notice it particularly every time there was little fog or a little mist that would incur the breakdown of these signs, and sometimes essentially put us out of business. Since we couldn’t observe we’d jump in a car and drive up and down the road looking for the sign.

Sullivan

And what was your procedure when you found it?

Smith

Ask then to fix it.

Sullivan

And would they usually fix it?

Smith

Astonishingly cooperative and very apologetic. We never ran into anybody who was the least bit ugly about it.

Sullivan

Didn’t know my sign can do that!

Smith

No, glad you told us. We’ll do something about it. And they did. Amazing how cooperative people were.

Sullivan

I find in the AJ in 1960 University of Florida radio observatory report which apparently is the first time that you felt like you were an astronomical observatory and should put a report in.

Smith

Amazing. That was the first time (?) subscribed to the journal. We didn’t know there was such a thing as a report. [laughter]

Sullivan

But all of this that you had done in the late 50’s was planetary radio astronomy? In the comet perhaps?

Smith

Yes. Yeah. The comet was probably early 60’s.

Sullivan

Now what about a paper in Nature in ’59 by Smith, Carr and Perkins on a solar burst you received at night?

Smith

The Type III solar bursts are like the Jupiter bursts. If you’ve ever heard one, you always recognize it as it’s hardly anything else you can confuse it with because it has this “shhhh” build up and then this beautiful exponential follow-up shhhh. I forget who was observing, don’t remember whether it was I or who it was, anyhow heard this humongous solar burst at 3:00 in the morning or some time like that. Unmistakably a solar burst. And it seemed passing strange, and since the sun was many hours from rising. So as I recall we corresponded with some solar observatories around the other side of the earth and sure enough there was this big burst at that time. We thought this was a curious enough propagation phenomenon so we wrote up a little note and published it in Nature calling attention to the fact that by golly such things can happen.

Sullivan

I’ve never heard of anything. Have you ever seen anyone else report such a thing?

Smith

No, I haven’t but I suspect people probably were. You know it’s the same sort of thing we used to hear reports when I was working on radar during the War about people receiving 10-cm radar all the way across the Indian Ocean by some kind of peculiar ducting effect. Oh, incidentally, along the same line, I very well remember on a number of occasions listening to Jupiter noise storms which continued after Jupiter had set. One I remember following until it was something like 10 or 12 degrees below the horizon. It just kept right on. Peculiar, obviously some peculiar ducting effect in the ionosphere.

Sullivan

What about the paper Carr, Smith and Bollhagen in Phys. Rev. Letters in 1960 in which you correlate the Jupiter bursts with the earth’s geomagnetic activity and thereby deduce that the solar wind essentially that is influencing these things? Were you the first to establish that would you say?

Smith

Well, I think about the same time, as I recall, Jim Warwick was doing some work along similar lines. Probably nearly simultaneously. It’s one of those strange things like geophysical correlations are apt to be. And we worked intensively on this since that time. As a matter of fact, we’ve got two doctoral students, this is not their sole thesis work but it’s in each case a fairly significant part of their thesis work, doing it right now. Looking for correlation between the Jovian decametric emission and geophysical events or solar events. We’re getting a little more sophisticated now in the sense that we’re trying to tie the thing together with models of the solar wind and propagation times of solar wind out to Jupiter and all that kind of stuff. Making use of the spacecraft observations (?) But it’s one of those strange things because you can find periods in the data when you seemingly get striking correlations. And you can find other periods where it doesn’t look like anything.

Sullivan

Just a strong burst and yet there’s no correlation.

Smith

And we think, one of these two doctoral students that I’m talking about, may have come up with the answer to this in that he investigated a period in which there was very stable interplanetary magnetic field, very stable solar sector field, and found the correlation during this period of very unusually stable interplanetary magnetic field pattern. Investigated another period more or less as a control, didn’t find the correlation so maybe this is the answer. That if you have a stable interplanetary field then the situation is well enough organized so that you can find a correlation. But it’s a kind of frustrating thing. At this particular moment in time, I wouldn’t state my reputation on the fact there is a definite correlation but we’ve certainly been suspicious enough, had enough positive results over the years that we’re still working on it and trying to demonstrate the cause and effect of the relationship.

Sullivan

Was there any other correlations in those early days of solar activity and Jovian activity?

Smith

There’s this long term variation in the mean level of the Jupiter activity as no small effect. In fact, if you average the probability of occurrence of the Jupiter noise over all the longitudes of Jupiter to wipe out the source effect so that you just kind of get – do this over an apparition so that you just kind of get an apparition wide mean rate of radiation independent of what longitude Jupiter happens to be turned toward the earth you find that between Jupiter maximum and Jupiter minimum there is about an order of magnitude difference in the probability of occurrence. As I said a little while ago we were lucky in a sense in that we had our first real full apparition in 1957 which was a good one but it was on the declining leg, very sharply declining leg of this. And ’58 and ’59 we got so little radiation that Tom Carr stroked his chin and said, “My God, we’re studying a dying comet.” He was particularly concerned (?) It came back and in the early 60’s, around ’63, just had tremendous rate of radiation. Then we went through another minimum and now once more it’s back up. Last year was as big as the ’63 minimum. So there seems to be a long term cyclic change in Jupiter’s mean rate of emissions or at least in our mean rate of receiving Jupiter’s (?)

End Tape 45B