Only the largest flares could be observed in this way. You can readily detect many smaller flares, however, if you have a radio receiver that can be tuned to very long waves. This part of the radio spectrum is usually quiet during the daylight hours, except, of course, on frequencies used by transmitting stations. During a solar flare the long-wave region crackles with "static" of the sort that we hear on the radio during a thunderstorm. In recent years this upsurge of radio noise has been increasingly used to detect the onset of flares.

David Warshaw of Brooklyn, N.Y., has designed an inexpensive radio receiver expressly for the purpose of monitoring flare activity. When this apparatus became available, a small group of amateurs across the nation volunteered to make a continuous patrol of solar Hares as a contribution to the International Geophysical Year. The group's observations are now published regularly in the Solar-Geophysical Data Bulletin of the National Bureau of Standards and - have been distributed routinely to the IGY World Centers.

"Then, in the summer of 1955, I received a mimeographed circular from the U. S. National Committee for the IGY: 'The Recording of Sudden Enhancements of Atmospherics (SEA) for Purposes of Flare Patrol,' by M. A. Ellison, then astronomer of the Royal Observatory at Edinburgh. The circular explained how X-rays and ultraviolet rays emitted by a flare cause a large increase in the number of free electrons in the ionosphere at heights from 60 to 90 kilometers, and how this effect, in turn, greatly enhances the reflection of very long radio waves that obliquely strike the lowest of the ionosphere's four layers: the so-called D layer. A proportional enhancement is simultaneously observed in the strength of radio signals received from a distant source after one or more reflections from the ionosphere.

"According to Ellison, the effect was first described by the French physicist R. Bureau, who investigated the enhancement of signals by recording the integrated level of tropical thunderstorm atmospherics at many different wavelengths. He found that sudden enhancements of atmospherics, or 'SEAs,' are characteristic of the spectral range from 7,000 to 16,000 meters, being most pronounced at the wavelength of 11,000 meters (27 kilocycles). There is no question, Ellison wrote, 'of an increase in the number of atmospherics at such times; the phenomenon is simply one of improved propagation at these wavelengths. The reflections take place during the daylight hours from a height Of about 75 kilometers, and typical SEAs have never been observed outside the illuminated hemisphere.'

"I gathered from Ellison's lucid description of the apparatus that a station for recording SEAs would be simple and inexpensive to build. Even so, I did not feel competent to undertake the construction. My interests incline mostly toward the analysis of data, and I feel distinctly uncomfortable in the presence of pliers and soldering irons. Our group includes several radio 'hams,' however, and I hoped that I could persuade them to set up a network of SEA stations and turn their recordings over to me.

"At about this time Warshaw, who is a member of our group and an electronics specialist in the communications industry, invited me to accompany him on a tour of a solar observatory on Long Island. During the trip I brought up the subject of recording SEAs. He was immediately fascinated and promised to construct a station as soon as I could supply detailed information to him. I still have the note handed to me by my wife one evening a few weeks later: 'November 8, 1955. David Warshaw telephoned. He is building a 27 k.c. receiver.'

"Dave finished the receiver in about two weeks. Although it performed well, the design proved to be impractical for amateur use. The original circuit called for costly British vacuum tubes and included a number of intricate biasing and voltage-regulating components to keep the indicating meter from fluctuating excessively during heavy signal bursts. After a short trial Warshaw decided to develop a receiver of his own, utilizing transistors. The power and voltage requirements of the transistor are so low that batteries can be substituted for the conventional rectifier, thus eliminating the need for voltage regulation. The design that eventually emerged from Warshaw's basement workshop calls for only three transistors, a solid-state diode, four coils, six resistors and capacitors, a pair of flashlight batteries and a microammeter. The components cost less than $20. The circuit, including the batteries, is housed in a standard apparatus box measuring only three inches wide, four inches deep and five inches long [see illustration below]. It is fully self-contained and provides enough output to trip a buzzer alarm and drive a pen recorder.

"Upon being tested, the new receiver exhibited gratifying sensitivity. Flares of average intensity or above tripped the buzzer, and on several occasions Warshaw found it possible to telephone his industrial colleagues that the cause of a current transmission black-out was ionospheric in origin and not an equipment failure. At this time Warshaw did not own a pen recorder and so was relying on the buzzer to signal the onset of flares. The relay that actuated the buzzer was set to respond to all signals above a predetermined energy level. This arrangement quickly proved to be disappointing. The buzzer responded to all sorts of noise, including interference originating in nearby oil burners, television receivers and electric shavers. SEAs differed from the spurious signals only in their rate of growth and decay, that is, in their 'wave' shape. One way to identify them with reasonable certainty is to make a graph of the signal by plotting signal amplitude against time. The method requires an automatic pen recorder.

"Warshaw and another member of our group, George Warren of West Chester, Pa., tried to build recorders, but after some weeks they were forced to conclude that such instruments are beyond the resources of most amateurs. Thus the summer of 1956 found us at an impasse. We could neither build a recorder nor afford to buy a commercial instrument. Without demonstrating the effectiveness of Warshaw's receiver, on the other hand, we could not hope to borrow recorders from some interested government group. Then, one day in August, Warshaw located a second-hand recorder within our means-and the next day he picked up a second one! By September he had the first instrument in working order.

"Our first AAVSO Solar Division SEAs Station was now in operation. Up to this point I had acted as little more than a kibitzer. But now I began the analysis of Dave's recordings. I shall never forget the discouragement of those early pen scribbles. The hieroglyphics traced by the 27-kilocycle noise seemed utterly meaningless. The easily distinguished SEAs that stood out so clearly. in the published data simply did not appear! Warshaw's station was located in Brooklyn at the intersection of Atlantic Avenue and Flatbush Avenue, where ignition noise is emitted by passing auto. mobiles day and night. He is within half a block of the subway and close to a school that teaches electric welding! Disturbances from these sources appeared at first to mask all other effects.

"Warshaw nonetheless persisted. Finally, on October 4, our first 'no-doubt-about-it' SEA appeared on the record. In fact, the tracing for this date displayed three clear flares. These were immediately confirmed by Robert Lee of the High Altitude Observatory of the University of Colorado.

"In the meantime Walter A. Feibelman, a physicist of Pittsburgh, Pa., became interested in the project and built a station after Warshaw's design. In May, 1957, he submitted some of the clearest SEAs ever recorded by our patrol. Our second station was in operation. That fall another member, Val Isham of Powell, Ohio, assisted by G. H. Pieterson, put a station into operation. (He had made a previous attempt at Columbus, Ohio, but was frustrated by local noise.)

"The records submitted by these three stations compared favorably to those of professional observatories. Warshaw's design was vindicated. After examining these results Walter Orr Roberts, then chairman of the U. S. IGY Solar Activity Panel, advised us that the Bureau of Standards would assign four Brown recorders to the project for a period of a year. Ellison, who had served as our consultant from the beginning, urged us to set up at least two stations in high latitudes. There, he thought, SEAs of greater amplitude would be detected.

"I do not know what difficulties professionals experience when establishing a network of observatories, but we had plenty. The matter of putting your finger on competent amateurs willing to stick with the job around the clock for a year is no easy chore. There is also the matter of equipment maintenance. Not all competent observers are good technicians. We could not afford to finance a roving trouble-shooter, although we soon had enough work to keep one- busy. Some instruments were damaged in transit. Others failed in use. In addition, there was always the problem of tracking down sources of local noise and taking corrective action.

"The recruiting job was handled by correspondence, and after many exchanges we finally selected four locations for the recorders. It turned out that the chart drive of the Brown recorders ran too fast for our purpose. So we modified the instruments for a chart speed of one inch per hour. The recorders were then equipped with 27-kilocycle receivers built by Warshaw and were shipped.

"The new observers included Ralph Buckstaff of Oshkosh, Wis., who operates one of the world's most complete amateur astronomical observatories. Farther west we recruited Walter Scott Houston of Manhattan, Kan., another leading amateur astronomer. Our northernmost station was set up by Franklin Loehde in Edmonton, Alberta (54 degrees North). On the West Coast we had Robert Evans of Victoria, B.C.

"Despite the best efforts of all concerned, only one of the new stations contributed substantially to the flare-patrol program. In one case a trolley line created so much disturbance that records could not be deciphered. I was the Edmonton station. A grave illness terminated the activity at Victoria. Limited success was achieved in Kansas, but power supplied by the rural electric line (for operating the recorder) proved to be so erratic that system records could not be kept. The station maintained by Buckstaff was productive.

"While these installations were on trial C. H. Hossfield of Ramsey, N.J. assembled an apparatus of his own though situated well within the metropolitan area of New York City, this station has made the clearest records submitted by our patrol. For some reason Ramsey is relatively free of man-made electrical disturbances.

"By the end of 1957 Justin Ruhge, a physicist of China Lake, Calif., had built a Warshaw flare-recorder and set it up for automatic operation in an isolated desert shack. His station, the southernmost in our group, records SEAs of the highest amplitude as well as those with the greatest change in signal level between night and day. Recently Stanislaus Scharlach, a Dominican priest who is a student of physics and spectroscopy in Oakland, Calif., completed a station and joined the patrol. Other late recruits include Harvey Hepworth of Blauvelt, N.Y., Hans Arber of Manila in the Philippines and Pierre R. Gouin, director of the geophysical observatory of University College at Addis Ababa in Ethiopia. All these observers send their records to me, and after analysis I forward the results to the Bureau of Standards and the AAVSO.

"What do SEA recordings show? Typically a 24-hour graph displays high-amplitude signals during the night, followed by a low-level trace beginning at sunrise. This characteristic is explained by the formation of the D layer at sunrise. The long waves emitted by ever present thunderstorms in the tropics reach the higher ionospheric layers by an unimpeded path at night and are reflected to the surface without significant loss of energy. During the day, however, the waves must traverse the lower and lightly ionized D layer. Because of its light ionization, the D layer acts not as a reflector of electromagnetic energy at long wavelengths but as a partial absorber. This loss of energy to the D layer accounts for the precipitous dip in the recordings with the approach of sunrise. SEAs appear on the graphs during the hours of daylight, because X-rays and ultraviolet rays emitted by solar flares impinge on the D layer and increase the intensity of its ionization. The D layer then functions as a reflector, and the amplitude of recorded signals increases as shown in the accompanying graphs [Figure 2].

"An exceptionally clear SEA is depicted in the graph made by Warshaw on July 16, 1959 [Figure 3 ]. Note the sharp rise at the right side of the graph, marking the onset of the flare, and the characteristic jagged decline at the left side. This SEA was caused by the large flare shown in the accompanying spectroheliogram, made by Ben Parmenter of Spokane, Wash. Most SEA recordings take this general shape, but the magnitude and detail recorded vary with the location of the station. This is illustrated by the graphs of a flare recorded simultaneously on March 17, 1960, at Oshkosh, Wis., Oakland, Calif., and Ramsey, N.J. [Figure 4].

"Proficiency in distinguishing SEAs from other disturbances comes with practice. The observer will acquire the knack by the time he has analyzed his first half-mile of graph paper. Then the recordings can become exciting. When you spot a big flare, you can confidently predict a black-out in short-wave radio communications within 26 hours, the interval required for the cloud of electrically charged particles ejected by the flare to reach the earth from the sun. Their arrival is often signaled by auroral displays in high latitudes, the onset of magnetic storms and the more or less severe disruption of electrical communications. Unfortunately there appears to be no direct relationship between the size of a flare and the amplitude of the resulting SEA. Large flares are often accompanied by small SEAs, and the opposite is also true.

"By the spring of 1957 our stations were functioning routinely. As I have mentioned, my task was to find some meaningful regularity in the seemingly random traces. One gross pattern, immediately evident in 80 per cent of the records, was the diurnal variation in signal intensity. The graph dips at sunrise, and rises with the approach of sunset.

"A close examination of the sunrise traces disclosed an interesting and equally regular pattern first observed in one fortuitous sequence of recordings that spanned 14 days. These disclosed that the signal level does not drop abruptly with the approach of dawn. Instead, the amplitude wavers for about 90 minutes, during which time four characteristic elements are observed. Some 45 minutes before the local sunrise the curve has a pronounced dip. About 10 minutes later a typical hump appears. I have termed this the precursor hump. The hump gradually slopes off with the approach of sunrise. The signal intensity then falls to its daytime level. Finally, about 50 minutes after sunrise, occasional recordings show a 'post-sunrise hump' of varying duration and intensity. To the best of our knowledge this detailed sunrise pattern had not been recognized prior to the analysis of our data.

"What are the implications of the pattern? Although the question is still open, Warshaw has suggested one interesting explanation. The phenomenon could be accounted for by a detached zone of low ionization that forms at an altitude of about 50 miles on the dark side of the sunrise terminator (the line between the illuminated and dark side of the earth). This zone might be described as a detached portion of the D layer. Signals reflected from a higher layer would be partially absorbed by the detached zone during their downward transit, an effect that would account for the predawn dip. Then, some 20 minutes before sunrise, the earth's rotation would advance the detached zone three or four degrees and would again establish a clear path for the reflections. The precursor hump would now appear, as the signal intensity approached its nighttime value. Warshaw suggests that the atmosphere in the detached zone may be ionized to about the normal intensity of the D layer by ultraviolet rays that proceed through the atmosphere on the lighted side of the terminator and impinge on the detached zone at the optimum angle for creating ionization. The accompanying chart [below] depicts the suggested mechanism together with an idealized graph of the sunrise pattern.

"No explanation has been advanced for the occasional post-sunrise hump. Its existence has been clearly established, however. Ellison has observed it on a number of occasions. On this side of the Atlantic the effect appears most clearly in the records of our station in China Lake, Calif. If you should decide to undertake the interpretation of flare recordings, you will discover that the post-sunrise hump can be vexing in the extreme. It resembles nothing so much as a clear SEA recording.

"Usually the graph begins a gradual climb in the afternoon. (The band becomes noisier.) This is to be expected, because the number of local thunderstorms, in addition to those in the tropics, increases during the latter half of the day. The signal pattern thus reveals its meteorological origin. Our records show that relatively quiet a.m. periods exceed quiet p.m. periods by a ratio of 10 to one. In spite of this there is a strange excess of SEAs during the p.m. hours. Roughly 1.6 times as many SEAs-about 62 per cent of the total-are recorded in the afternoon.

"Nighttime fluctuations are particularly great; frequently pseudo-SEAs appear. We even have some examples of pseudo-SEAs recorded at night that coincide with the onset of flares. As Ellison has pointed out, however, one cannot properly classify such patterns as SEAs.

"We have adopted the classification system for reporting SEA data which was proposed by J. Virginia Lincoln, Chief of the Radio Warning Services Section of the Bureau of Standards, and which was described in Bureau of Standards Report 5540 (November, 1957). According to this system, the amplitude of SEAs ranges from 1 - (lowest in amplitude) to 3 + (highest). Further, we rate the certainty with which the identity of a SEA is established. This scale ranges from 5 (definite) to 0 (questionable). We have recorded approximately 2,000 SEAs to date. Of these, five of our stations submitted 1,060, almost 13 SEAs per month per station.

"What good are our recordings? For one thing, they correlate satisfactorily with those made by the official observatories. We catch some SEAs that are missed by the official stations, and vice versa. In a letter extending the loan of the Bureau of Standards recorders Walter Orr Roberts has written: 'Miss Lincoln is very pleased with the data she has been receiving from your people and we feel that this program is making an important contribution to the observing effort.'

"If you decide to go in for flare recording you should not find Warshaw's receiver particularly difficult to duplicate. The parts are available from most dealers in radio supplies and are interconnected as shown by the accompanying circuit diagram [above]. All resistors are of 1/2-watt capacity and should be within 10 per cent of the value specified, except as has been indicated in the diagram.

"Warshaw suggests the use of No. 14 solid-copper plastic-covered wire for the ground bus, an arrangement that provides a solid mechanical support for the coils. The coils designated L₁ and L₂ must be mounted side by side and spaced approximately half an inch apart. The remaining coils are mounted at right angles to L₁ and L₂ and perpendicular to each other. In other words, if L₁ and L₂ are mounted vertically, one of the two remaining coils must be mounted east and west, and the other north and south. This arrangement minimizes magnetic coupling in the coil assembly and prevents the receiver from going into uncontrolled oscillation. The coils contain ferromagnetic slugs, and the slug of L₁ is adjustable for tuning the receiver. For operation at 27 kilocycles the screw supporting the L₁ slug is first run home and then backed out two full turns. When wiring the receiver, it should be recalled that transistors are easily damaged by heat. Do not solder transistor leads. Cut them to a length of about half an inch, and make the circuit connection either by means of binding posts or by spring clamps salvaged from a miniature vacuum-tube socket. The current drain on each flashlight cell approximates only 600 microamperes, hence the batteries should last their normal shelf-life. The output of the receiver is sufficient to drive a 200-microampere pen movement to full scale when signals of maximum amplitude impinge on an L-type antenna 100 feet or more in length."

Bibliography

ASTRONOMY. Robert Horace Baker. D. Van Nostrand and Co., Inc., 1959.

Suppliers and Organizations

The Society for Amateur Scientists
5600 Post Road, #114-341
East Greenwich, RI 02818
Phone: 1-401-823-7800

Internet: http://www.sas.org/