4410-63-X VESIAC State-of-the-Art Report THE REQUIREMENTS OF A HIGH-SENSITIVITY SEISMOGRAPH STATION DEAN S. CARDER Research Seismologist U. S: Coast and Geodetic Survey October 1963 Acoustics and Seismics Laboratory By4tzwe i Scone aw4 7-e^4^U THE UNIVERSITY OF MICHIGAN Ann Arbor, Michigan

Institute of Science and Technology The University of Michigan NOTICES Sponsorship. The work reported herein was conducted by the Institute of Science and Technology for the Advanced Research Projects Agency of the Office of the Secretary of Defense under Contract SD-78. Contracts and grants to The University of Michigan for the support of sponsored research by the Institute of Science and Technology are administered through the Office of the Vice-President for Research. DDC Availability. Qualified requesters may obtain copies of this document from: Defeilse Documentation Center Cameron Station Alexandria, Virginia Final Disposition. After this document has served its purpose, it may be destroyed. Please do not return it to the Institute of Science and Technology. ii

Institute of Science and Technology The University of Michigan PREFACE VESIAC, the VELA Seismic Information Analysis Center, is an information collection, analysis, and dissemination facility established at the Institute of Science a and Technology of The University of Michigan. The contract is sponsored by the Advanced Research Projects Agency under the Office of the Secretary of Defense. The purpose of VESIAC is to analyze the research information related to the VELA UNIFORM Program of Project VELA and to function as a central facility for this information. The facility will serve all authorized recipients of VELA UNIFORM research information by issuing subject bibliographies with abstracts and special reports as required. In addition, VESIAC will periodically summarize the progress of the research being conducted. VESIAC is under the technical direction of the Acoustics and Seismics Laboratory of the Institute. In its operation VESIAC draws upon members of this laboratory and other members of the Institute and University. iii

Institute of Science and Technology The University of Michigan CONTENTS Notices...................................... ii Preface..................................... iii List of Figures................................. vi List of Tables.................................. vii Abstract..................................... ix 1. Introduction................................. 1 2. Background Noise...................3.......... 3 2.1. Causes of Background Noise 3 2.2. Influence of Geological Environment 10 2.3. Reduction of Noise Effects 14 3. The Seismograph............... 30 3.1. The Seismometer 31 3.2. The Seismograph Recorder 39 3.3. Seismometer-to-Recorder Communications System 41 3.4. Timing Systems 42 4. The Seismograph Station......................43 4.1. Class A Stations 43 4.2. Class B Stations 51 4.3. Class C Stations 51 4.4. Special Purpose Seismograph Stations 52 5. Capabilities of Some Sensitive Seismograph Stations........... 53 6. Summary...............................61 References..................................62 Distribution List.................................65 V

Institute of Science and Technology The University of Michigan FIGURES 1. Wind Noise as a Function of Wind Velocity for the Best and Worst Tank Vaults at WMSO........................... 6 2. Vibrograms Made While the Vibrator Is Situated as in Figure 3 and Brought from Rest to Maximum Safe Speed and Allowed to Coast to Rest..... 7 3. Amplitudes inInches x 10-6 on the Ground near the Hollywood Storage Building while the Vibrator Was Rotating in Resonance with the Building, Period 0.50 Second........................ 8 4. Ground Plot of Amplitudes of Major Axis of Ground Vibrations Representedby Vectors, Forcedby Power-Plant Machinery, Salt Lake City.. 9 5. Section of a Seismogram from the Tucson Station, Showing Strong Noise Generated by a Rock Crusher.................... 9 6. Noise Spectra at Various Sites in the United States.............11 7. The Spectra of Seismic Noise Recorded at 16 Geographic Locations.... 12 8. Pacific Northwest Profile, Showing Average Noise Amplitudes for Long- and Short-Period Records.....................13 9. Central California Seismographic Network, University of California.... 17 10. Seismic Telemeter Equipment for Outstations in the Central California Network................................. 18 11. Telemeter Receiving Console, Central California Network.........19 12. Telemetering System at Tucson..................... 20 13. Results of Ocean Bottom Seismometry................. 24, 25 14. Blue Mountain Array............................27 15. Tryon Three-Dimensional Site........................28 16. Rayleigh Wave Attenuation with Depth....................29 17. Culture Noise Variation with Depth.....................30 18. Benioff Moving-Coil Vertical Seismometer.................31 19. Melton Vertical-Motion Seismometer.................. 33 20. Benioff Variable-Reluctance Horizontal Seismometer............34 21. Benioff Variable-Reluctance Vertical Seismometer............. 35 22. Johnson-Matheson Vertical and Horizontal Seismometer.............36 23. Deep-Well Seismometer, Model 11167, Variable-Reluctance Type..... 37 24. Sprengnether Magnetically Suspended Vertical Seismometer.......39 25. Suggested Seismometer Array for an A-1 Station............... 44 26. Amplitude Response Curves of Variable-Reluctance Short-Period Seismograph Using Galvanometers with Periods of 0.2, 0.5, 1.0 and 2.0 Seconds................................... 47 27. Geotech Attenuation Unit..........................48 28. Blue Mountain Seismological Observatory.................48 vi

Institute of Science and Technology The University of Michigan 29. Minimum Station Facility for World-Wide Seismic Instruments...... 50 30. Lower Limit of P-Wave Reception as a Function of Magnitude and Distance................................... 54 31. P Readings Reported in Distance Range from 50~ to 90~.........56 32. P Readings Reported for Earthquakes of Magnitude from 4.5 to 5.5.. 58 33. P Readings Reported for Earthquakes of Magnitude from 4.5 to 5.0...59 TABLES I. Stations Considered in Sensitivity Study...................55 II. Analysis of Seismic Signals from Three Events...............60 III. Recommended Minimal Distances from Sources of Disturbance to Seismograph Station Short-Period Seismographs.............61 vii

Institute of Science and Technology The University of Michigan ABSTRACT The efficiency of a short-period seismograph is regulated largely by the nature and magnitude of the so-called background noise prevalent at the station. This background noise is more or less continuous and varies from place to place and from time to time. Effects of this noise may be reduced by (1) selecting a station site where the noise is low, either remote from noise sources, in the shadow of noise barriers, or deep underground; or by (2) the use of frequency and wavelength filters. Frequency filters are in large part ineffective because the signals we wish to measure have the same frequency (from 0.5 to 5 cps) as the noise; and wavelength filters require the use of expensive arrays. This report discusses possible sources and methods of propagation of background noise and the selection of optimum sites and methods for reducing the effect of noise. Present and suggested future arrays are included in this discussion. Highly sensitive seismograph stations and instruments in current operation are described, and comparisons are made of the ability of some of the better stations to record magnitude 4 to 5.5 earthquakes. Suggested designs of unmanned and other special-purpose seismic stations are included. ix

THE REQUIREMENTS OF A HIGH-SENSITIVITY SEISMOGRAPH STATION 1 INTRODUCTION A short-period seismograph is defined as one which shows peak sensitivity to vibrations of the earth having periods from 0.2 to 2 seconds with a fall-off toward zero for infinitely long and infinitely short periods. These instruments have great importance in modern seismology because the periods of initial P waves from nearly all earthquakes and large explosions are in this range. Unfortunately, the periods of the ever-present disturbances in the ground are also in this range. The problem therefore is not how to obtain greater magnification from the instruments, as that is now almost unlimited, but how to obtain the greatest effective magnification in relation to these ground disturbances, called background noise. In 1956 I discussed the optimum setting for a high-sensitivity short-period seismograph, using largely the basis of my own experience [1]. Since then much has been done toward seismic background surveys: many new stations have been placed in operation, and the experience thus gained has become available; new instrumentation has been designed; and methods of recording and data processing have been improved. This report discusses an optimum short-period seismograph station from the viewpoint of 1963. During the first six decades of this century, seismometry has advanced from crude mechanical devices to the highly modern moon seismometers which are capable of broadcasting their motion across space to earth-bound recorders. The early inverted-pendulum models, although crude, are not obsolete. Many are still in operation, some at highly rated observatories and others in localities where photographic paper is scarce. Since the record is made by a stylus on a rotating cylinder of smoked paper, magnification is attained through the use of a heavy mass, necessary to overcome friction, and a delicately constructed system of levers. Twenty-ton masses are not uncommon. Inverted pendulums have 2 degrees of freedom and record in two horizontal directions. In Switzerland, the masses are suspended on a heavy spring and recording is in the three cardinal directions. Magnification attained by mechanical seismographs varies from 10 to about 2000, the latter attained with use of a 20-ton mass. These systems record ground motion with periods up to 2 or 3 seconds. Damping is attained by means of a plunger in a confined air space or in a dashpot of low viscosity oil. The mass of a 20-ton seismograph fills the larger part of an ordinary-sized 1

Institute of Science and Technology The University of Michigan room. The one at Zurich, for example, consists of a metal tank filled with discarded World War I armament. In contrast with the 20-ton seismograph is the Wood-Anderson seismograph [2], developed in the earlier 1920's, with its 0.7-gram dynamic mass. Ground periods up to about 0.8 seconds can be magnified about 2600 times; this is possible because the displacement is registered on a rotating cylinder of photographic paper by use of optical levers, which have no friction. The mass consists of a 1-mm diameter copper rod and mirror suspended longitudinally in a near vertical position on a ribbon rolled from a 20-p diameter tungsten wire. The device acts as an accelerometer for ground frequencies less than 1 cps (periods greater than 1 second). It goes into bowstring resonance at about 20 cps and fails as a seismometer for these and greater frequencies [3, 4]. The intermediate-period Wood-Anderson seismograph has a thin bar for a mass and directly records ground periods up to 6 or 7 seconds except that it likewise fails at bowstring resonance. Its magnification is about 500. Electromagnetic flux in the copper vane provides the damping in both cases. Trends toward modern seismometry were under way by 1914, when Prince B. Galitzin developed and described the electromagnetic seismograph that bears his name [5]. A number of modifications of Galitzin's method have been made. One is the Benioff variable-reluctance (VR) seismograph (1932). In the two decades following, various moving-coil (MC) types, including the Benioff, Sprengnether, Wilson-Lamison, Lehner-Griffith, and others, were produced. The modifications mentioned are earth-driven instruments without amplification other than the direct electrical connection between the seismometer and galvanometer, with an intermediate shunt and resistance pad to provide gain control and internal damping. In most areas, the gain is high enough that the background noise limits additional gain. This background noise, its causes, and possible methods of its reduction are subjects of later sections of this report. The Benioff VR and MC and similar instruments have another characteristic in common. They have peak sensitivity to periods near 1 second and are relatively insensitive to the longperiod surface waves that are propagated from most large earthquakes. The fact that P waves from most distant earthquakes also have periods near 1 second was largely responsible for the immediate success of the Benioff instruments. The Benioff VR long-period seismometers developed by Press, Ewing, and others, and a strain seismometer developed by Benioff, drive long-period galvanometers and are sensitive to long earth periods. Discussion of these instruments is outside the scope of this article. As for applied geophysics, the seismic method of exploration geophysics was developed during the 1920's, and by 1930 salt domes in the Gulf of Mexico area had been discovered by that method. During the succeeding decades, instrumentation and practices reached a high state of 2

Institute of Science and Technology The University of Michigan perfection. However, until recently applications of this art to earthquake seismology were not successful, chiefly because exploration geophysicists were interested primarily in arrival times and their gear was tuned to record high-frequency vibrations-not generally prevalent in seismic waves from earthquakes. In the past few years, largely because of the demands brought about by a national emergency, exploration and earthquake seismologists are finding a common meeting ground and are calling on other services, such as the communications services, for assistance. The result is presentday seismology with its greater and greater world coverage with better and better instruments, and with greater and greater emphasis on getting the most from the data on the resulting records. 2 BACKGROUND NOISE In seismology, the term "background noise" applies to all ground unrest identified on the record as continuous or intermittent disturbances not originating from earthquakes. These disturbances, which may vary from hour to hour and from place to place, may result from to and fro bodily translation of the earth particle, as in a bowl of jelly, or from vibrations set up by passing elastic waves. The former are more applicable to spongy ground or saturated fill, and the latter to firm ground and rock. Background noise resulting from traffic, machinery, or wind on buildings, trees, or rocks is usually high in frequency, with periods somewhat less than a second. Such disturbances, troublesome only to short-period instruments or to instruments that have vibrating parts that may be set into resonance by the disturbances, are referred to as "noise." Disturbances caused by natural sources other than vibrations caused by nearby wind are generally known as microseisms, although strictly speaking all ground vibrations not associated with earthquakes, regardless of source, are microseisms. Microseisms (in the restrictive sense) have been put to practical use, as in attempts to apply them to hurricane sensing. More often they are a nuisance and seriously interfere with optimum recording of seismic waves originating from earthquakes. It is background noise, whatever the source, that limits the maximum sensitivity at which a seismograph can be operated. In this discussion two types of background noise which interfere with optimum recording by short-period seismographs will be considered. 2.1. CAUSES OF BACKGROUND NOISE 2.1.1. MICROSEISMS. Microseisms (in the restrictive sense) have periods of 1 to 10 seconds. Longer periods have been noted, but long-period microseisms do not seriously interfere with the operation of a short-period seismograph and will not be considered here. 3

Institute of Science and Technology The University of Michigan The literature abounds with papers on microseisms, their causes and propagation [6-14]. It is generally agreed that they originate from large bodies of water; that they are somehow associated with storms at sea and with fronts moving from sea to land or from land to sea or with other offshore weather conditions; and that they are coherent elastic waves, probably surface waves. 2.1.1.1. The Longuet-Higgens Theory. The mechanism of the origin is subject to considerable controversy. The theory which probably has the widest acceptance was formulated by Longuet-Higgens [9] which for maximum generation of microseisms requires that ocean swells having the same frequency move in opposite directions. Standing waves are thereby produced from which a second-order effect, in phase over a large area, causes fluctuating pressures on the sea bottom having twice the frequency of the individual waves. This theory is strengthened by the observation that microseismic peaks lag behind the closest approach of a storm or the peak intensity of a storm by as much as 48 hours [7]. This lag is the time required for the swells generated by the storm to reach a so-called "generating area" near the continent. Further, the microseismic frequency is often twice that of the oncoming swells. 2.1.1.2. Microseism Propagation Patterns. Determination of the source mechanisms of microseisms, aside from the observation that they are connected with large bodies of water, is more or less academic from the standpoint of optimum seismographic registration. The propagation patterns are not. Microseisms originating on the coast of Norway have been traced by Gutenberg across Eurasia as far as Irkutsk, more than 5,000 km away [10]. In North America, on 31 October 1952, 5-second microseisms throughout the central states, strong enough to almost nullify the usefulness of the short-period seismograph at Fayetteville, Arkansas, were traced to a storm in the Arctic Ocean, 5,000 km north of Fayetteville. Sevensecond microseisms originating off the coast of Labrador have been traced 5,000 km to Pasadena on the West Coast [11]; 9-second microseisms originating off the coast of Alaska have been traced a like distance to the East Coast [12], and, under extreme conditions, as far as Bermuda and Puerto Rico [13], a distance of 6,000 to 7,000 km which includes 1,200 and 2,000 km of oceanic path, respectively. The following observations have thus far been made: (1) 2-second microseisms originating in the Great Lakes have been traced by Lynch 700 km southward into the Carolinas [14]. (2) 4- to 5-second microseisms have been traced as far as 5,000 km, across central and eastern North America, 4

Institute of Science and Technology The University of Michigan (3) Microseisms having periods less than 4 or 5 seconds do not in general travel great distances across the western cordillera. (4) 6 to 10-second microseisms travel from coast to coast. However, attenuation is much greater across the cordillera or the western Atlantic than across the plains. (5) Earth filtering is more effective for the shorter-period microseisms. This may be one explanation for an observation that the period increases with distance. However, if real, this increase probably occurs over the first few hundred kilometers of travel, for microseisms attributed to West Coast or East Coast origins have the same periods in the distance range from 1,000 to 5,000 km [12]. Seismological networks in the United States have been greatly enlarged during the past few years, so that material for more detailed studies of microseismic phenomena is now available. 2.1.2. SHORT-PERIOD DISTURBANCES (PERIODS LESS THAN 2 SECONDS). Background noise in this category is generally local and rarely regional in origin, it may be coherent or apparently noncoherent, and it may have natural or artificial causes. Natural causes are: (1) Surf breaking on a nearby shore or short choppy wave action near the shore. (2) Weather action over small bodies of water. Causative mechanisms here are probably the same as with large bodies of water, but in miniature-or the same as (1) above. Background observations near Duluth, Minnesota, in late October and early November 1952, illustrate this point. A Benioff MC seismograph was temporarily installed on a gabbro sill about 4 km from Lake Superior. Under storm conditions, the background noise was heavy but irregular. During periods of relative quiet between the storms, the background was heavier but more coherent, apparently following the pattern of microseisms from oceanic sources but with shorter periods. Lake Superior is not to be classed as a small body of water, but the point is brought out here for purposes of illustration. As another example, the old Pierce Ferry station which operated for a decade or so in northwestern Arizona was located 1 km, more or less, from an upper arm of Lake Mead, which may be considered a small body of water. The underground was loosely consolidated terrace gravel and silt, but in a remote area. During periods of quiet, the background was relatively low. It was relatively high at the time of the usual afternoon winds. This condition was probably due in part to the wind disturbing the terrace upon which the instrument was located. (3) Eddies produced by wind blowing over nearby objects such as hillocks, rocks, trees, fences, buildings, and other irregularities. Figure 1, taken from a VESIAC report [15], illustrates this point. This effect is greatly reduced if the source of the disturbance 5

Institute of Science and Technology The University of Michigan,._. i.........g 203040 E g ag go gicat$to..0.....st tank i it ( A 20... VELOCITY (mph). FIGURE 1. WIND NOISE AS A FUNCTION OF WIND VELOCITY FOR THEBEST AND WORST TANK VAULTS AT WMO [15] WoBest concrete ~~~~~~~Iet 11vaultAts is streamlined.va (4) Running water, rapids, and waterfalls. This applies likewise to dam spillways. 0 10 20 30 40 VELOCITY (mph) FIGURE 1. WIN]? NOISE AS A FUNCTION OF WIND VELOCITY FOR THE BEST AND WORST TANK VAULTS AT WMO [15] is on alluvium and the seismograph is on bedrock, and if the ground above the seismometer vault is streamlined. (4) Running watabout linearly in a northd waterfalls.ly direction ahis far applies 20 km fromewise t dam spillways. The Bureau of Reclamation briefly operated aal seismograph at a site located on a granite was operated at high sensitivity, but during periods of high water noise with a period cs: 5 to(a) A con resulting fruous 3-cps vibrng wation having amplitudellway, was from 0.2 to 3 was observed nullify the effectiveness of the seismograph. A ground survey indicated that this noise attenuated abound that thlinearly in a northerly direction as far as 20 km from the dam. Artificial sources of background noise include for the most part moving objects such as traffic and vibrating machinery. Man-made objects such as structures vibrating in the wind are considered natural sources. Artificial sources include: (5) Reciprocating power plant and other machinery such as rock crushing, mining, and oil field machinery. This subject was covered in earlier publications [16, 17]. To summarize some important artificial noise characteristics: (a) A continuous 3-cps vibration having amplitudes from 0.2 to 3 p was observed frequency, and that the vibrations attenuated with distance away from active oil fields.

Institute of Science and Technology The University of Michigan (b) A large building vibrator operating on the thirteenth floor of a monolithic type reinforced concrete warehouse having a rectangular cross section caused the building to go into resonance when the machine period was that of the building. The building was coupled to the ground, which was dry and hard after nine rainless months, by means of skin contact and concrete piers. When the building was being forced into vibration in an E-W direction (along its long axis), resonance occurred at 0o50 second, the natural period of the building. Forced vibrations of the building were transmitted to the nearby ground and were observed by a seismograph. Compressional vibrations died out rapidly to less than 0o2 p 200 meters west of the building (see Figures 2 and 3). 0.50 second 0.50 second A _ ~ ^flwJwlwW I ~"~j 0.005" 0.50 second 0.50 second B fuMww;MMA It@II^Dw' O0.0001" 0.50 second 0.50 second / __ / C N- - - A 0.O0001OO 0.50 second 0.50 second D A, / FIGURE 2. VIBROGRAMS MADE WHILE THE VIBRATOR IS ALIGNED PARALLEL TO THE LONG AXIS OF THE HOLLYWOOD STORAGE COMPANY BUILDING AND BROUGHT FROM REST TO MAXIMUM SAFE SPEED AND ALLOWED TO COAST TO REST. The maxima represent the building in resonance with the vibrator. The minimum between the maxima shows response to highest speed and therefore the greatest imbalance during a given test, but with the building out of resonance. Tests were made (a) on the thirteenth floor of the building; (b) on the ground 112 feet west of the building; (c) on the ground 0.6 mile south of the building; and (d) on the ground 1.2 miles south of the building. A passing pedestrian caused the irregularities of the trace in test d. Shear vibrations having the same amplitude, 0o2 pu, were carried 10 times as far, to nearly 2 km south of the building. After a heavy rain a few days later, forced vibrations in the building were not noted on the nearby ground. (c) Early in 1942, a portion of the downtown area in Salt Lake City was served by power generators driven by two reciprocating steam engines. While one engine was in 7

Institute of Science and Technology The University of Michigan N I Hollywood Storage 9 Building _4 / -- 40 ~.4 0 26 0 - 26 5000 1 4000 4o-10 3000 Scale EW and NS 2000 Feet 1000 8- 8 0 FIGURE 3. AMPLITUDES IN INCHES x 10 ON THE GROUND NEAR THE HOLLYWOOD STORAGE BUILDING WHILE THE VIBRATOR WAS ROTATING IN RESONANCE WITH THE BUILDING, PERIOD 0.50 SECOND operation, vibrations in the nearby ground or in basements were measured. The orbit of ground motion was found to be an ellipse, the major axis of which was always normal to the direction of the source; this indicates transverse or shearing motion (see Figure 4), At a distance of 220 meters from the machine in line with its motion, transverse ground motion having a period of 0.31 second had an amplitude of 0.5 J. At the same distance normal to t his direction the amplitude (parallel to the forcing motion) was 2; this again emphasized the preponderance of shear components in coherent noise transmitted in rigid ground. (d) A coherent 2-cps disturbance was noted on most of the seismograph records [16] during Project COWBOY experiments in Louisiana, December 1959 to March 1960. The noise was believed to originate from pipeline pumps in the area, which had a frequency of 120 rpm. (e) The seismograph at Tucson, Arizona, is disturbed by a rock crusher 1.6 km away. This noise disappears during the noon hour, at night, and on Sundays and holidays (see Figure 5). (6) Vibrations from traffic. This noise is really in the same category as (5). However, noise from traffic is less coherent than that from reciprocating machinery because of the mobile nature of the sources. Trains are probably the greatest offenders; disturbances from them carry to distances of 16 km or more. 8

Institute of Science and Technology The University of Michigan 8.8 N N 4.5 SOUTH TEMPLE STREET 2 Forcing o O W E~ ~ Motion 15 0 C __j 5 ~j FIRST SOUTH STREET! I I FIGURE 4. GROUND PLOT OF AMPLITUDES OF MAJOR AXIS OF GROUND VIBRATIONS REPRESENTED BY VECTORS, FORCED BY POWER-PLANT MACHINERY, SALT LAKE CITY. Numerals represent amplitude in inches x 10-5, forcing motion one machine only. -w 9d,* WI _ ~. ~ "~*~ ~ w__,,1___________-''_P\., t t #^ ^^0, Ir'.,~ ~,,,l,~',,:tz.... e *, '~V,,,'.-:.t _'.. " _ _ _ _~' _ _ _ _ _"~' —. T~ —t." -.-: —.-= _ - -- - -. - - - — _ __ _ FIGURE 5. SECTION OF A SEISMOGRAM FROM THE TUCSON STATION, SHOWING STRONG NOISE GENERATED BY A ROCK CRUSHER _MI 1..I " ~ —.-~ --- -'-P _ _ _ _ _. ta~ t _.. w,, _ _ 9

Institute of Science and Technology The University of Michigan 2.2. INFLUENCE OF GEOLOGIC ENVIRONMENT The geologic environments of the noise source, of the location of the station, and of the intervening area have much to do with the efficiency of a seismograph station. Unweathered granite has generally been considered the best underground for a station. Quartzite, limestone, or basalt, if not underlain by inferior rock, should serve equally well, with certain exceptions which among other things will now be considered. (1) If the noise source and the station are on the same formation, disturbance at the station will be relatively great. If the formation is rock, the noise presumably is transmitted for the most part by shear, as mentioned earlier, or less likely as Rayleigh waves. If the rock is granite, or homogeneous with depth, the shearing motion, if represented as body waves, would be expected to spread out in spherical waves, and attenuation would be relatively large. If the formation is a low-speed layer, however, noise would be channeled in the layer. Further, if there is a low-speed layer near the formation, there may be leakage from that layer and noise attenuation will be relatively low. That is a probable reason why a granite outcrop may be superior to limestone or basalt, unless they are upturned. Note: Noise caused by the impact of falling water over Grand Coulee Dam reached a seismograph on granite 1.6 km south of the dam, probably in the form of Rayleigh or SV waves, because by far the largest amplitudes (of 0.5- to 0.9-second noise) were on the vertical and radial components. A body of water 100 meters deep lay in between. In a relatively non-rigid formation such as unconsolidated or water-saturated alluvium, noise is probably transmitted as Rayleigh waves, by reflection between interfaces, or by other means, but not primarily by shear. Seismic signals recorded by stations on alluvium are relatively large, but noise from local sources is larger, reducing the station efficiency. Where the noise source and the station are on different formations, disturbance at the station may be relatively light, especially if the formations have widely different acoustic properties. For example, if the station is on a basement rock and the noise source is on nearby alluvium, or if the source is on rock and a discontinuity consisting of alluvium, till, or other unconsolidated material lies between, the unconsolidated material will serve as a damper. As an example, the College, Alaska, station lies on residuum of a schist hill. Train noise disturbs the seismograph more when a train crosses an outcrop on the point of a hill about 1.5 km distant than when it passes on valley fill within 0.5 km of the station. A number of surveys have been made in the past few years. Figure 6 is a plot of noise spectra at various sites in the United States made by the Geotechnical Corporation [15]. This study covers the period range from 0.2 to 5.0 seconds, since this includes the range which disturbs a short-period seismograph. Figure 7 shows the results of a similar survey made by The University of Michigan [18]. 10

Institute of Science and Technology The University of Michigan LOCATION.....f~-.. i A St. Louis Bl Florissant A Ann Arbor - _..ii l.. F Ogdensburg (in mine) -._.-. _; ''............ Palisades O.. Chapel Hill.. _.....~..... W r _ )X Spring Hill q)C:^.":^;';^:::l:;:;;";:":^;^: ^:: //^.tt^S^^G^^^ ' " College Station "1 ~~0 0 o i ~~i 4: m~..* Reno - 1111111^ 11ffl 1111111- V Pasadena (d s 4 <........ <=: - ^F E g ] _-........... Tucson MQ)~~,. -'Laramie o Garland: McMinnville ' x., 1 Quehanna. ': OWMSO 101... ci —:___........ 101945t~s [Aj A _',, =United ElectroD, I. [19i, has completed noise correlations wh g c ad graphic environments along two profiles' (a) in the Pacific Northwest from the coast of Washof California to Death Valley. In California, noise levels of periods ranging from 1.25 to 1.5 seconds were related to environment. In the northwest there was a definite correlation with the times. Figure 8 is a representation of average noise amplitudes at three locations along the ~~~~1northwest profile.J. S 11~945 ~ ~ 1 0.2 0. 1 2 5 PERIOD (seconds) FIGURE 6. NOISE SPECTRA AT VARIOUS SITES IN THE UNITED STATES [15] United ElectroDynamics, Inc. [19], has completed noise correlations with geologic and geographic environments along two profiles: (a) in the Pacific Northwest from the coast of Washington to the Wallowa mountains in northeast Oregon; (b) in the Pacific Southwest from the coast of California to Death Valley. In California, noise levels of periods ranging from 1.25 to 1.5 seconds were related to environment. In the northwest there was a definite correlation with the weather: noise came from the ocean during storms at sea, and from the north to west at other times. Figure 8 is a representation of average noise amplitudes at three locations along the 11

Institute of Science and Technology The University of Michigan 1000 /a // M 100/ 8-c ^// / 9-10 H~endricks, Mich, o //~ / 11-Shepherd, Mont. 70/ / / 12-Spokane, Wash. 12 / / 14-Rifle, Colo. / / 15-Bodie, Calif. ~~~~~/ / 1 6-St. George, Utah / / ~ —Brune & Oliver (1959) 12.01.1 1 1 6-WinPERIOD (seconds), La. CORDED AT 16 7-Ft.GEOGRAPHIC LOCATIONS. Data are normalized to I-cps bandwidth [18]. 12

Institute of Science and Technology The University of Michigan 10, 000 1... 10,000 I 1 1 I111111 1 1 111111 1 1 1 11 11 1 100 a) ________ _ ______ p \ __-. _ 100~_- 100~ ~ _ _ - _ -,~ 11 I h T / sH e f 1 O + f @ 0 -. KHA 0 _41 _ PI^~~ I A MPLT U I LOT- I _ _ ~ the _ - T-h4 - - -- - i- -- 1.0 0 PATERSON 0.1 1.0 10 100 0.10 1AT00SON PERIOD (seconds) FIGURE 8. PACIFIC NORTHWEST PROFILE, SHOWING AVERAGE NOISE AMPLITUDES FOR LONG- AND SHORT-PERIOD RECORDS. Markham is on the coast of Washington, Toppenish is in south central Washington, and Paterson is about 90 km ESE of Toppenish [19]. 13 I, ___[//^__ ~ ~ ~ -— ^ — ~^ ^ _ __ ---_- ~ ~ _ -— ENIS ^ ~ ~ ~9 —~/ —/~ -~~~,-' J ~ ~ -~ - — ~ I- ~ / x - ~~ ~ - - ~ ~ ~ _ — o plESN ' I l ______ __ __ y __ _ _ _ _ __ A R HM,_ _ _ ]3

Institute of Science and Technology The University of Michigan 2.3. REDUCTION OF NOISE EFFECTS Reduced noise load can be attained two ways: (a) by choosing a locality where the noise is relatively low, or (b) by the use of filters to reduce undesirable noise while maintaining or augmenting the strength of the signal. These methods presuppose that the noise at the source is not reduced. 2.3.1. LOCALITIES WHERE THE NOISE IS RELATIVELY LOW 2.3.1.1. Geographic Localities. An optimum site for a seismograph station would naturally be one which is remote from noise generators and has a natural filter lying between the chief noise source and the site. This would be a locality on the interior of a continent, "shaded" from the ocean by at least one and preferably two mountain ranges, and remote from lakes, cities, railroads, and other noise sources. The Blue Mountain Seismological Observatory in northeastern Oregon, near Baker, is an example of a quiet 10-element array station; it will be described later. Reservoirs to be formed by present and future dams in the Snake River Canyon are more than 35 km from the site and probably will not cause serious disturbances. The seismograph stations described here can be operated at high gain because of a relatively low background. The instrumentation at these stations is described in a later section. The accompanying initials are the C&GS station abbreviations [20]. Eureka, Nevada (EUR). This station, located in the Great Basin province 600 km from the Pacific Ocean and in the shadow of several mountain ranges and alluvial valleys, has earthquake recording capabilities equal to those of the best stations of the Americas. It may be described as a "shoestring" station because of its low initial cost and because it operates only a shortperiod vertical seismometer. The seismometer, a Benioff MC model, is located a hundred or so meters within the adit of an abandoned mine. It drives a 0.5-second Lehner-Griffith recording galvanometer located in a small building outside the mine. The maximum gain for 1.0-second period ground motion is about 450,000, but this would probably not be attained during autumn and winter, since the 5-second microseisms from trace to trace often overlap. Neither mining operations, traffic on U. S. Highway 50, nor activity in the nearby town 2 to 3 km from the station disturbs the seismograph materially. Albuquerque, New Mexico (ALQ). This station is a unit of the Coast and Geodetic Survey World-Wide Standardized net, described in Section 4.1.2, and is a part of the Survey's seismologicallaboratory. The seismograph vault is a tunnel driven 20 meters into a granite knoll. The city of Albuquerque, the Santa Fe railroad, the airport, and other artificial noise sources are all 20 to 30 km north and west of the station and are for the most part on desert valley fill. 14

Institute of Science and Technology The University of Michigan The instruments apparently are not materially disturbed by any artificial source except occasional blasts on a nearby military base. The short-period instruments operate at a gain of about 400,000 during the summer, and at 200,000 during the winter because of ocean-generated microseisms. This site was chosen after an exhaustive background-noise survey. Its choice was governed in part by the convenience to facilities that a city affords. Sites east of the cordillera were rejected, largely because of the prevalence of a 2-cps background noise. Shillong, India (SHL). The station is located on bedrock in the mountainous area between the Bay of Bengal and the Brahmputra Valley. The city of Shillong is located in a hilly country 2 to 3 km from the station and 100 meters lower in elevation. Industrial noise sources appear to be inconsequential. The entire area is shielded from microseismic sources from the north by the Himalaya Mountains, and the station is not materially disturbed by ocean-generated microseisms except during monsoon seasons and during occasional hurricanes. Until April 1963 gain settings had been somewhat below the microseismic level, so that the true capabilities of the station had not been realized. This condition was alleviated after installation of World-Wide Standardized instruments in April 1963. Tamanrasset, Algeria (TAM). This station is located in the Saharadesert 1700 to 2500 km from the Atlantic ocean and 1400 km from the Mediterranean sea. Culture noise is insignificant. Local winds, however, disturb the short-period seismographs about the same as at any other station. The amplitudes of residual noise having periods of 2.5 to 3.5 seconds are from 10 to 50 times lower than the amplitudes of similar noise at the European stations. Rocard [21] attributes this residual noise to diastrophism. He feels that because it is much lower at Tamanrasset than in Europe the Tamanrasset area is relatively stable geologically. It would seem, however, that the great distance from ocean-generated microseism sources would be a more tenable explanation. Because of its location on the interior of a large continent, this station can be operated at high gain. South Pole, Antarctica (SPA). This station originated as a part of the IGY program, and its continuation is made possible as a result of the IGC. Before its installation, some concern was raised (a concern not shared by this writer) that the station would be ineffective because of the 3,000-meter ice overburden above bedrock. In actuality, bedrock is fairly close to the surface because natural ice is, by definition, a rock. A Benioff MC short-period vertical and two Wilson-Lamison 4-second horizontal seismometers were located in a vault reached by a tunnel 400 meters long. The vertical instrument operated a 0.5-second Lehner-Griffith galvanometer, and the horizontal seismometers operated 4-second galvanometers, all located at a recording station near the living quarters. Early in the history of the station, considerable shut-down time was experienced because of broken galvanometers and other failures caused by environmental 15

Institute of Science and Technology The University of Michigan difficulties. During the past few years, however, operation has been very satisfactory, and culture and wind-induced background noise is remarkably low. Contrary to expectations, oceaninduced microseisms are also quite low, perhaps because there are structural discontinuities between the central plateau and the coast and because the ice cap has high absorption coefficients for short-period surface waves. The South Pole and other Antarctic seismograph stations are of high value because they afford a southern control useful in monitoring earthquakes in the southern hemisphere. A World-Wide Standard seismograph has now been installed at the South Pole station. Quetta, Pakistan (QUE). The site is geographically ideal. The Arabian Sea is 550 km to the south, and in every other direction lies the vast continent of Eurasia with its shielding mountain ranges. The city of Quetta is located near the center of an elevated intermountain valley on valley sediment of unknown depth. The geophysical observatory is located near the edge of this valley, within a few hundred meters of a steep mountain slope. A World-Wide Standard seismograph is installed in a room adjoining the observatory. At the time of writing, the gain of the short-period vertical was about 200,000. From an inspection of available records, ocean-generated microseisms are very low, but high-frequency background noise disturbs the short-period records. It is believed that this noise is largely cultural. According to station descriptions, the short-period seismometers are located near the observatory building. An outpost station on nearby solid rock may materially reduce the noise. 2.3.2. THE OUTPOST STATION AND TELEMETRY. An outpost station is defined as one in which ground motion is sensed or picked up in a seismically favorable but relatively inaccessible locality and recorded at an accessible but seismically unfavorable locality. Several ways of doing this are now in use. (1) The seismometer can be located on a rock outcrop or other favorable environment away from traffic, and the recorder at a more convenient location; there is no amplification between the seismometer and the recording galvanometer. The College, Alaska, outpost station is probably an outstanding example of this. A Benioff MC vertical seismometer is located in an underground pit on a schist hill about 4 km from the main station. The recorder is located at the main station (the C&GS College Magnetic and Seismological Observatory), and the interconnection is a 4- to 4.5-km length of spiral-4 cable, a product of the U. S. Army SignalCorps. The long line in Alaska is possible because atmospheric disturbances such as lightning storms are almost nil in the area. Elsewhere-in Fayetteville, Arkansas, for example-it was found that electrical induction in a 150-meter connecting line was excessive during an electric storm. 16

Institute of Science and Technology The University of Michigan Many other stations with a relatively short seismometer-to-galvanometer connection, Eureka, Nevada, for example, are now in operation; other stations could probably be improved by using this method. (2) When the seismometer site is too remote to be linked to the recording station without amplification, the seismometer signal can be amplified and is telemetered to the recorder over a wire linkage. The possibilities of this method are continent-wide wherever hard-wire telephone service is available. The Berkeley, California, telemetering network (see Figure 9) is a good o CLS o Telemeter Station (Recording at Berkeley) o Station with Local Recording \ Station with Local & Telemeter Recording Scale / PRC r^\ 0 20 40 60 80 100 0o CNC Kilometers \o BKS, BRK SFB fV ~ PAC! ~ MHC o SCC VIT, VIN 0 LLA \ PRS o PRI FIGURE 9. CENTRAL CALIFORNIA SEISMOGRAPHIC NETWORK, UNIVERSITY OF CALIFORNIA. BKS-Berkeley (Strawberry). BRK-Berkeley (Haviland). CLS-Calistoga. CNC Concord. LLA Llanada. MHC Mt. Hamilton. PAC Palo Alto. PRC Point Reyes. PRI-Priest. PRS Paraiso. SCC Santa Cruz. SFB San Francisco. VIN Vineyard. VIT Vineyard telemeter. 17

Institute of Science and Technology The University of Michigan example. Ten Benioff VR vertical seismographs are located along the central California fault systems. The ones farthest from Berkeley are 180 km SE and 85 km NNW. All are connected to the University of California's central recording station at Berkeley by communication channels leased from the telephone company. Recording is on a Geotech Develocorder, described in Section 3.2.2, with check recordings on Helicorders. These facilities make possible immediate recognition of earthquakes along central California's fault systems, and on the azimuth of distant earthquakes. Schematic diagrams of the central California telemetering network are shown in Figures 10 and 11. An operating example of an outpost station with which telemetry is used is a station in Tasmania. A 1000-meter high plateau near the center of the island, with moderate to heavy rainfall, provides a natural resource in the form of hydroelectric energyo Because certain power facilities cross ground which may be seismically active, the Tasmanian Hydroelectric Authority cooperated with the University of Tasmania to establish a net of seismograph stations on the island. This net uses a combination of wire and radio linkage in its telemetering system. Two units of a tripartite net located around the center of the island are in line of sight with the top of Mt. Wellington, a 1270-meter high peak near Hobart. The third is in line of sight with VoltageSeismometer Controlled Oscillator Cathode Follower | — \ Twin FM It?, III oe;uplPhoto I --------— ' N ---] Photo Signal Lamp / Tubes Filter M,'.......__) / -I r —V I ~~- Multiplexer G-al - -.-To Tele A^ ^ o T^ Rectifier Line Attenuator Splitter Rectifier Lens Power Supply Calibrator Constant Voltage Timer I I I ITransformer 110v 60 -Input FIGURE 10. SEISMIC TELEMETER EQUIPMENT FOR OUTSTATIONS IN THE CENTRAL CALIFORNIA NETWORK 18

Institute of Science and Technology The University of Michigan FM Signal From Tele Line Local Hi 99AMr Signa Helicorder AM Signal Audio Monitor _Amplifier Line Amplifier Distributor Timing Unit Discriminator Develocorder! ~ ~1. I - ~- i 1 [ Attenuator Switching Develocorder Filter ~ ~ _______________ Unit ~ Amplifier l l Signal Converter | T EIV ~ C lE Coding i ~ 1 ~ Unit WWV Receiver FIGURE 11. TELEMETER RECEIVING CONSOLE, CENTRAL CALIFORNIA NETWORK Mt. Barrow, on which is located a television relay station and which is in line of sight with Mt. Wellington. Seismic signals are telemetered to Mt. Wellington, or are relayed there by radio from Mt. Barrow, and then carried over a land line to the university at Hobart; there they are recorded side by side with signals from a local station. (3) If the seismometer-to-recorder distance is fairly long and the seismometer relatively inaccessible, the communication channel can include a radio link. This system has been in use with varied success in Canada and South Africa. In the United States the method was first successfully used in Tucson, Arizona. The seismograph station there includes a Benioff VR vertical, one of the first to be placed in service. In its early history, the station, located on caliche, a part of the valley fill, was considered the best in the United States. However, a rock crusher located 1.5 km more or less from the station, and other noises associated with the expanding city, rendered weekday daytime short-period registration ineffective. In 1957, a telemetering system, using a radio link, was perfected and placed in service. A Benioff MC vertical seis19

Institute of Science and Technology The University of Michigan mometer was located in a rock outcrop near the summit of Mt. Lemmon, about 10 km northeast of the observatory. The transmitting station with its antenna was about 150 meters distant, and the top of the antenna was on line of site with the receiving antenna at the observatory. As with the wire-linked telemetering system, the signal was amplified at the transmitter by a photoelectric split beam cathode ray follower. A block diagram of the system is reproduced in Figure 12. The new world-wide station is located at a relatively quiet spot 9 or 10 km from the old one; the telemeter has been discontinued. This section has shown that telemetry provides remote recording from a seismically favorable but inaccessible location, and also a means of recording the output from an entire net of stations on a single tape. The subsections to follow describe varieties of telemetering systems for use by really inaccessible outpost stations. Benioff Mc W Damping alva Photo-Electric Subcarrier Motorola 60-Watt - 2Sal Calibration IKC - I ' Generator Recorder Oscillator Recordere Galva- Damping Subcarrier Motorola Sensicon L....... no meterv Bridge Demodulator FM Receiver FIGURE 12. TELEMETERING SYSTEM AT TUCSON 2.3.3. DEEP-HOLE SEISMOMETRY. One means of escaping from noise is to go deep underground. It will be assumed that background noise is surface generated. If it is in the form of Rayleigh waves, by theory the rate of attenuation of vertical motion with depth would be 5 db for the first half wavelength of depth and 10 db for each increment of half a wavelength thereafter. Consequently, attenuation with depth would be greatest for the higher-frequency vibrations and in unconsolidated sediments. This reasoning is prevalent in current literature and panel 20

Institute of Science and Technology The University of Michigan discussions on attenuation of background noise with depth. Little if any thought has been given to the possibility that culture and wind noise may in large part be S waves and that attenuation with depth may be a result of absorption and of inhomogeneities of the rock between the surface and the detector in the bottom of a well. The following subsections describe several experiments with deep-hole seismometry. 2.3.3o1, The Homestake Mine, Lead, South Dakota. A pioneering experiment in deep-hole seismometry was performed by the Coast and Geodetic Survey in this mine. In April 1950 two Benioff MC vertical seismographs were operated side by side on the lowest level, 5,000 feet (1.5 km) beneath the surface, and set to give identical responses. One was then moved to the surface, and both were operated for several weeks. This experiment was repeated on a more elaborate scale in March-June 1953, when one Benioff VR vertical was operated on the 5,000 -foot level and another on the 300-foot level. Signals from both instruments were telemetered over hard wire to a film recorder on the surfaceo Provision was made to filter out noise having frequencies of 5 cps or greater; this gave an element of unreality to signals from blasts in the mine. From the two series of tests, the following observations were made: (a) Ocean-generated microseisms having periods of 4 seconds or greater were not attenuated with depth. (b) Culture-generated noise having periods of 1 second or less caused by traffic and machinery, was attenuated with depth by a factor of at least 10. (c) Noise from operations in the mine disturbed the seismograph on work days. (d) P waves from distant earthquakes having periods of 1 second or less, as recorded on the surface, have an average ratio of 2/1 to those recorded on the 5000-foot level of the mine; this is in accordance with theory. Because of difficulties in operation and the prevalence of machinery-generated noise, the Homestake mine was not considered a suitable site for a permanent station. 2.3.3.2. Exotic Environments 2.3.3.2.1. Deep-Hole Seismographso Deep-hole seismographs have been developed under contract by the Geotechnical Corporation and United ElectroDynamics, Incorporated. Experiments (a) and (b) were conducted by Geotecho (a) Grapevine, Texas. During the summer of 1962 one seismometer was installed at the bottom of a deep well at Grapevine, Texas, depth 3100 meters (10,150 feet); another in a shallow well, depth 23 meters; and the third, a small model Geotech seismometer, at the 21

Institute of Science and Technology The University of Michigan surface for comparison purposes [22]. Available test runs indicated that at 3100 meters a 3- to 4-cps noise was attenuated by a factor of 25 to 1. Early data indicated that the relative strength ratio between the surface signal and the signal at 3100 meters was somewhat less than the theoretical factor of 0.5, more nearly 0.3 to 0.4, and in an extreme case as low as 0.1. This may be because the surface material has lesser seismic wave speed, and hence associated amplitudes must be greater to maintain the same energy content. From this experiment it seems evident that the increase in signal-to-noise ratio with depth is from 2 to 8 times. A later test, in November 1962 [23], indicated that signal strength at depth relative to that at the surface was somewhat more than 0.5 or nearly 1.0. If this finding is valid, the above explanation need not hold; at any rate, the finding introduces the possibility that the low strength of both signal and noise at depth in the early experiment was caused by poor coupling between the seismometer and the borehole. (b) Hobart, Oklahoma. A test run was made in the Prater No. 1 well during the late summer of 1962. The well has about the same depth as that at Grapevine. A nearby well 158 meters (512 feet) deep was tested simultaneously with the seismometer in the deep hole and with one on the surface. It'was found that the S/N in the shallow well was about 3/1. In the bottom of the deep well it was slightly less than this, although at 5,000 feet it was slightly better than 3/1. This test may not be valid; escaping gas in the deep well may have added to the noise level. Tests described here more or less confirm results of earlier tests by the Jersey Production Research Company [24]. (c) Tube noise and casing noise. Experiments were tried at the Hobart well and at some of the 3D wells, described later, on the transmission of noise down the casing. A noise pulse was induced into the casing by striking the top of the casing a sharp blow. It was found that that the speed of the seismic pulse down the hole was that of the tube noise, and not of the casing. It is believed that tube noise may have been partly responsible for the inefficiency of the Hobart wello Use of plugs or baffles in the well might have helped to reduce the noise level. 2o3.3o2.2. Ocean-BottomSeismometry. This form of telemetering has great potential: more than half of the earth's surface is covered by oceanic areas more than 100 km from the nearest land. The Lamont Geological Observatory, the Scripps Institute of Oceanography, and the Texas Instrument Company are currently experimenting with ocean-bottom seismographs [25]. The seismographs have been constructed, and trial experiments have been made. Operational problems include power supply and data recovery. Atomic batteries, although currently expensive, appear to promise an effective power supply. Data can be recovered in several ways. One is to recover the seismograph after a period of operation; another is to telemeter 22

Institute of Science and Technology The University of Michigan seismic signals by wire to a surface recorder on a buoy or to a land-based recorder on an island. Results thus far indicate that the floor of the deep ocean is relatively quiet remote from land, but relatively noisy near land regardless of depth, and that the signal-to-noise ratio is nearly equal to that of land-based seismographs (see Figure 13). Since ocean-bottom seismometry is very costly, it should be combined whenever possible with other oceanographic operations, such as deep sea tide and tsunami research. Furthermore, all possible advantage should be taken of abandoned telegraph cables. 2.3.3,2.3. Moon Seismometers. The Lamont Geological Observatory and the California Institute of Technology have constructed seismometers which will withstand impacts of falling on the moon and function thereafter as vibration detectors with capabilities of telemetering the signals to earth-based recorders. The one attempt thus far to land a seismometer on the moon has failed. These seismometers are relatively small, and they may provide instrumental capability in other inaccessible places, such as ocean bottoms or deep holes. 2.3.4. FILTERS 2.3.4.1. Frequency Filters. Undesirable frequencies maybe reduced or almost eliminated by the proper use of bandpass filters. An electromagnetic earth-driven seismograph is a bandpass filter with gradual cut-offs on the two sides of the band. Electronic filters provide sharper cut-offs. Frequency filters, however, have certain undesirable qualities: (1) They filter out signals that we often wish to investigate. This fault could be overcome by multi-recording a number of frequency bands; but to do so would add to the volume of already voluminous data. Alternatively, the broadest frequency band possible could be recorded on magnetic tape, and desirable frequencies could be investigated on playback. (2) Filters, including electronic ones, distort the signal. In fact, any transfer of a seismic signal from one medium to another-say from the earth to a wiggly line on a film or paper, or from the earth to magnetic tape track and from the tape to a sheet of paper-results in some distortion. A major problem in seismology is to reproduce source conditions, or conditions along a ray, by the study of distorted images on films or sheets of paper. (3) Quite often, the frequency of the signals we wish to study is that of the noise we wish to eliminate or reduce. 2.3.4.2. Wavelength Filters or Arrays. The effectiveness of a wavelength filter is based on an assumption that background noise is surface-generated and is more or less confined to lowspeed surface layers, whereas a seismic wave from an earthquake travels at a higher speed 23

Institute of Science and Technology The University of Michigan, Station 2 100 __ _________/ 19 Sept. 1960 100 10 I I, Station 3 / /,27 Jan. 1961 i Max, / //.. FStation 5 X I/-/ 10 Apr. 1962 3 -1 8 Bermuda (Col LP.Z & ~i. //// / I.Benioff S.PoZ. Ave. t.1 ~ / ~ ' H L.~ - / / '~'' '- f/ / - Station 9 ',,/-, / ~ 28 Sept. 1962 0.01.0001:.1 1 1 10 PERIOD (seconds) FIGURE 13. RESULTS OF OCEAN BOTTOM SEISMOMETRY. (a) Station 2 is about 140 n mi WSW of Bermuda; stations 5 and 9 are about 70 and 60 n mi south of Bermuda; and station 3 is in the Gulf of Mexico about 400 n mi south of New Orleans [25]. 24

Institute of Science and Technology The University of Michigan.~ /: I Ave.:1Mi ^.01 \ / fll. Station 8Station 7 16 May,196210 May, 1962.o0l 1 (^'fSystem Noise Stations 6-8 7/1 Station 7 FIUE1.REUT FOCA OT10 May, 1962 0001..*. *: FIGUE 1O Stations 1, 4 Stations 6-8 and 8 are in the Arctic near 810 north latitude and between 1720 and 173 east longitude. Stations 1 and 4 are in the Atlantic near 720 west longitude and respectively near 35 1/20 and 25 1/20 north latitude [25]. 25

Institute of Science and Technology The University of Michigan and arrives from depth; hence a seismic wave of the same frequency has a greater wavelength than surface-propagated noise. If the seismic signal is a P wave from a regional or distant source, its apparent speed along the surface is from 7.8 to 25 km/sec, whereas the speed of noise is presumably no greater than 3 or 4 km/sec. Linear and cross arrays have been designed with the idea that dominant noise is noncoherent and will pass the array along its length and be phased out while the expected signal reaches one leg of the array broadside. The triangular array is a modification of this. Mr. Leslie F. Bailey, of the Coast and Geodetic Survey, has designed an array which has not been demonstrated but which has possibilities of providing a greater depth of filtering, for noise arising from outside the array within certain frequency ranges. Preliminary work calls for testing an array site for the true phase velocity of the noise to be filtered. Several limited-array observatories are now in operation, most of them under the technical supervision of the Air Force Technical Applications Center (AFTAC) as a part of Project VELA UNIFORM, which is under the overall direction of ARPA (Advanced Research Projects Agency). Two of these stations will be briefly described. They follow the pattern recommended by the Geneva Conference in 1958. 2.3.4.2.1. The Wichita Mountains Seismological Observatory (WMO). This observatory is located on granite basement about 25 km northwest of Lawton, Oklahoma [26]. This 10-element array is arranged in the pattern of an equilateral triangle, 4 elements on each side and one in the center, and the elements 1 km apart. The elements in the array are Johnson-Matheson vertical seismometers. Six of the elements are installed in tank vaults, each imbedded in concrete and buried under 2 feet of earth for insulation. The remaining elements are in concrete vaults which contain also Benioff short-period vertical and horizontal seismometers in addition to Sprengnether and Press-Ewing long- and intermediate-period seismometers. The output from each element of the array recorded is on a Develocorder at 400K gain, together with a WWV trace, a 4-unit summation, a 10-unit summation, N-S short-period (SP) ground motion, EW SP ground motion, and a low gain channel. Broadband and long-period recordings are recorded on another Develocorder; selected recordings are made on 1-inch wide magnetic tape operating at 0.3 ips. 2.3.4.2.2. Blue Mountain Seismological Observatory (BMO). This observatory is located on granite basement about 42 km east of Baker, Oregon, and about 4 km south of Sparta [27]. It is a 10-element array built after the same pattern as the WMO (see Figure 14). Each element is contained in a tank buried under 2 feet of earth, as at WMO. Near one of the elements 26

Institute of Science and Technology The University of Michigan SP Johnson Matheson, Vertical Z1-Z10 Vaults 1-10 SP Benioff, Horizontals Vault 3 (tank farm) __ LP Sprengnether, Invert & 2 Horizontals Vault 3 (tank farm) BB Press-Ening, Vertical Vault 3 (tank farm) BB Sprengnether, Horizontals Vault 3 (tank farm) IB Melton, Vertical Vault 3 (tank farm) / - 7 IB Sprengnether, Horizontals Vault 3 (tank farm) j Vault No, 4 Scale 1 in. = 500 ft \ Mon. Elev. 4105 ~) A I '\ /<V - M,., "\ Vault No.5 - Mon. Elev. = 4082 - \ Vault No. 6fi c 5%. ~-.' i VaultNo. 9 Vault No. 3 Mon. Elev. = 3902 )Mon. Elev. = 3981 / St No 3 [Latitude = N. 44050'56" ~ Sta. N~ o 3 \Longitude = W. 117018'20" \ oJI^^, - I -\..; / /< \'Vault No. 10 X^ /,. I'./Mon. Elev. =3929 \ Vault No. 7 /. o Mon. Elev. 3991 r2 o' / I '3o / ^^^^ \ \ o J <^~/^ Mon. Elev. = 3732 LEGEND & NOTES Unimproved Traveled Roads = ====.= Access Route to Monument Vault No. 9 \ Vault No. 9 M ARRAY Monument, 3/4" Steel Reinforcing Rod Mon. Elev. = 3715hoioa __ 3 _... Set in Concrete, 5 ft in Ground, 1 ft f/r Above Ground, With Sta. No. Stamped \ / outine /f n on Medallion Affixed to Rod. IC \/ T 7CHO/A oVault No. 1ealo Aixat O.> / / SEISMOLOGICAL OBSERVATORY SITE Large ary ai no i eVault No. I ' / ' SITUATE IN SECS. 21.22,27,28 nd 34, TS S.. B.M. 3999.76 Alum. Cap on 1" Pipe Mon. Elev = 3603 R.44 E. WM., BAKER COUNTY, OREGON M ' ~ Surveyed Oct. ElevNv,tl = 60 Monument Elevations Shown P..., Tn. are Ground at Monument FIGURE 14. BLUE MOUNTAIN ARRAY are buried other tank vaults, which contain 2 short-period horizontal and 3 long-period Sprengnether vertical and horizontal seismometers. Signals from the array are recorded on one Develocorder, as outlined for WMO, and those from the long-periods and broadband elements on a second line. Some recordings are also made on magnetic tape operating at 0.3 ips for spectral research purposes. Larger arrays are now in service at the Uinta Basin Seismological Observatory, near Vernal, Utah; the Cumberland Plateau Seismological Observatory, near McMinnville, Tennessee; and the Tonto Forest Seismological Observatory, north of Payson, Arizona. 27

Institute of Science and Technology The University of Michigan 2.3o4.2.3. The Three-Dimension Array, Tryon, Oklahoma. This array, at a station 68 km northeast of Oklahoma and 1.2 km northwest of Tryon, consists of a number of deep holes spaced at selected intervals. It has the advantage of control with depth in addition to monitoring of angles of emergence. An experimental array of this sort is operated by the Century Geophysical Company under contract with the Air Force Office of Scientific Research [28]. Five abandoned oil wells 1600 meters deep, arranged in the plan of a 5-spot, are used for the array (see Figure 15). The inline spacing of adjacent wells is 402 meters. The subsurface lithology is mainly Paleozoic shales and sandstones. Section 3 ecorg D El. 251 X;4V1 8-1 = 9800 f/sec str"Urnts~V1 1-2 10,700 ft/sect ~~~~~~~~~~~EEl. -973 ____ 28,1-3 X > C-1 V1 1-2 = 10,700 ft/sec D-2 FIGURE 15. TRYON THREE-DIMENSIONAL SITE ~~~~~~~28 El. -1551 B-2 V1 2-3 = 14,600 ft/sec D-3 El. 2851 B-3 4e'n- C-3 28

Institute of Science and Technology The University of Michigan Deep-hole seismometers were assembled by attaching preamplifiers to Hall-Sears 2-cps seismometers. These assemblies were clamped to the well casings at 3 levels: the A-level at -368 meters (considering the surface zero), the B-level at -768 meters, and the C-level at -1168 meters. Maximum drilled depth could not be used because of the difficulty of removing a plug in one of the holes. Outputs from the 15-element array, together with summation traces and a trace from a small model Benioff vertical at the surface, are recorded on a 15-channel oscillograph which operates periodically at 0.5 to 1.0 ips, and continuously on magnetic tape having a transport speed of 0o36 ips (9.1 mm/sec). A brief summary of test results from this experiment indicates that noise from the top of the holes is not transmitted down the holes unless it is of the pulse type over an individual hole; that surface waves from local blasts definitely identified as Rayleigh waves do attenuate with depth according to theory (see Figure 16); but that culture-generated noise, e.g. from the passing of a train 106 km to the south, does not materially attenuate with depth (see Figure 17). Noise p R "I 'A"l Level- 500' i n;Ii III"B'ITevel- 1800' U1. I P RAYLEIGH WAVE ATTENUATION WITH DEPTH "A"l Level Center Hole Bn EvenlollLa EventA1 ~I r i 29

Institute of Science and Technology The University of Michigan Level B.'' ' '' " - ' ' ' ^ 1: ' '\ Level C " ' i, Train Nise ': r ' ir I Raw Data 15 Trace Composite Benioff Level A Level B Level C ~ '.' ' I ' '.Train Nose I Time Corrected for Arrivals From the South 15 Trace Composite ' ' Benioff FIGURE 17. CULTURE NOISE VARIATION WITH DEPTH in the array has frequencies from 2 to 10 cps and is apparently body-wave energy. This, in part, verifies earlier experiments, where it was observed that, in a rigid medium, preponderant cultural noise is in the form of shear, although the earlier experiments did not establish that it was in the form of body waves. 3 disturbances in th e earth. The four parts which comprise the modern seismograph are: 1. The seismometer or detector, which picks up ground motion. 30

Institute of Science and Technology The University of Michigan 2. The recorder, which makes possible a permanent record of ground motion. 3. The communications system between the seismometer and the recorder. 4o The time reference, to a world standard, for timing receipt of seismic signals. Items I to 3 are essential parts of a seismograph by definition, and item 4 is essential if the instrument is to be of value to the World-Wide network. 3.1. THE SEISMOMETER The most commonly used seismometer consists of a dynamic mass suitably suspended, to which is attached a transducer. Modern systems are equipped to transfer mechanical energy created by differential motion between the ground and mass into electrical energy. 3.1.1. THE MOVING-COIL TYPES. The Benioff moving-coil (MC) seismometer (see Figure 18) was designed in the late 1930's and is now in limited service on four continents, FIGURE 18. BENIOFF MOVING-COIL VERTICAL SEISMOMETER 31

Institute of Science and Technology The University of Michigan including Antarctica. In the vertical-motion instrument, the 45-kg mass, a horizontal softsteel cylinder, is hinged to the dust-tight supporting case and is suspended vertically by a series of arcuate springs [29]. The cylindrical transducer coil is attached to the end of a multiplying arm and is suspended vertically in a powerful magnetic field formed between the core and the outer ring of a doughnut-shaped Alnico magnet. The coil fits over the core and clears both core and ring. The horizontal-motion pickup unit is a vertical unit rotated 900 so that the cylindrical mass is in a near vertical position; the arcuate springs are removed, and the case designed to fit. The free period of the seismometers is from 1.0 to 1.4 seconds. The impedance of the external circuit is such as to critically damp the system. Most other MC seismometers are similar to the one just described except that the masses are usually lighter, and the vertical-motion units are suspended by coiled isoelastic, zerolength springs. The lesser masses of these units afford greater portability without appreciable loss of sensitivity. The Lehner-Griffith MC seismometer, for example, has a voltage output almost equal to that of the Benioff. Wilson-Lamison [30] and Sprengnether shortperiod seismometers are other MC seismometers available at relatively low cost. In the Melton vertical-motion seismometer the mass is suspended by a horizontal coiled spring (see Figure 19), one end of which is attached to an arm of the fixed support [31]. The period is adjustable from 0.5 to 7.0 seconds. A seismometer in general use in Russia [32] is similar to the Melton one except that the support end of the spring is adjustable vertically as well as forward and backward horizontally. Very-low-frequency 2-cps exploration-type seismometers are now coming into limited use in earthquake seismology, but electronic amplification is necessary if high gains are to be attained. 3.1.2. THE VARIABLE RELUCTANCE TYPE. The Benioff VR seismometer [33] is in wide use on every continent. The transducer system consists of coils having iron cores as a part of the pendulum mass and fixed magnets as a part of the frame [34]. The seismometer in most general use has a 100-kg cylindrical mass. The mass on the horizontal motion seismometers is suspended at each end on three steel strips placed 1200 apart with one strip vertical (see Figure 20). The mass is in two parts, with the transducer system in the middle. The mass of the vertical-motion unit (see Figure 21) is suspended from two oppositely wound coiled springs attached end to end and passing through the tubular center, the lower end attached to the bottom of the mass. The natural period is about 0.5 second, but the magnetic system reduces the restoring force to 1/4 that of the mechanical system, and in practice the period is 1.0 second. The so-called "Baby Benioff," designed and constructed by the Geotechnical Corporation, is in limited use. The transducer system virtually constitutes the dynamic mass. 32

Institute of Science and Technology The University of Michigan FIGURE 19. MELTON VERTICAL-MOTION SEISMOMETER 33

Institute of Science and Technology The University of Michigan FIGURE 20. BENIOFF VARIABLE-RELUCTANCE HORIZONTAL SEISMOMETER 34

Institute of Science and Technology The University of Michigan FIGURE 21. BENIOFF VARIABLE-RELUCTANCE VERTICAL SEISMOMETER. The spring system in late models consists of two oppositely wound springs placed end to end and attached near the center of the mass. 35

Institute of Science and Technology The University of Michigan 3.1o3o MOVING MAGNET SEISMOMETERS. In this type, the transducer coil is attached to the support and the magnet is a part of the dynamic mass. The two Willmore seismometers, designed by P. Willmore and manufactured by Hilgar and Watts, are in widespread service, especially in the British Commonwealth. Both are relatively small and inexpensive. Both have an advantage over models described heretofore in that with a simple internal change they can be operated as either horizontal or vertical instruments. Both are equipped with watertight cases and can be buried without protection, thus eliminating seismometer vault expense. The Mark I has a period of 1 cps; the period of the Mark II can be adjusted from 1/3 second to 3 seconds, which makes it a versatile research instrument. Both are available with either low-impedance coils, for use with galvanometers, or high-impedance coils, for use with amplifiers. The Mark I has given excellent service in detecting Pwaves. The Johnson-Matheson seismometer (see Figure 22), designed in the laboratory of the National Bureau of Standards [31], is currently being used in array seismology. It is somewhat FRAME \ CALIBRATOR CASTING\ / COIL SUPPORT LEVER SUSPENDED MAGNET STRUCTURE MAIN COIL SUPPORT MASS CLAMP ACTUATOR It FIGURE 22. JOHNSON-MATHESON VERTICAL SEISMOMETER 36

Institute of Science and Technology The University of Michigan larger than the Willmore, and its period is 1.25 seconds. With a watertight case, it can be operated under several meters of water. If the borehole is large enough, a 1-cps borehole seismometer recently designed and constructed by the National Geophysical Corporation has considerable promise. As of this writing, results of field tests are not available. 3.1.4. THE BOREHOLE SEISMOMETER. In an effort to reduce the effects of background noise by going deep underground, VELA UNIFORM requirements brought forth the borehole seismometer. High-frequency assemblies have been in general use in exploration for several decades. However, requirements for a 1-cps unit have made the problem somewhat more difficult. The apparatus must function in a hole 10,000 feet deep at temperatures to 1500C, where corrosion may be great and where the hole may be inclined by as much as 100. Provisions for recalibration and recentering must be made, and the response must be equal to that of surface instruments. One down-hole seismometer is equipped with moving coil transducers. To obtain proper damping, large magnets were required which, for a 100-kg mass, necessitated constructing the mass in several sections with a transducer for each section. A second seismometer (Figure 23) has a VR transducer which functions with smaller magnets so that only one mass is required. Centering and Main Spring Assembly Mass Electrodynamic / Transducer ~\ Calibrator / / A ^O-Rings Cable Head and Cable-Head Adapter Delta Rods Delta Rods FIGURE 23. DEEP-WELL SEISMOMETER, MODEL 11167, VARIABLE-RELUCTANCE TYPE 3.1.5. VERTICAL SEISMOMETER SUSPENSION SYSTEMS 3.1.5.1. Spring-Mounted Systemso Heretofore, the dynamic mass of a vertical-motion seismometer has been held in place by a spring or a series of springs. To avoid the effects of 37

Institute of Science and Technology The University of Michigan temperature changes, and elastic hysteresis, and other troubles, considerable care in the design and selection of the spring is necessary. In addition, unless the spring is properly designed, the seismometer is subject to a troublesome second-order effect which results in a complicated response to ground motion. Now we have the zero-length spring, in which the length of the spring is proportional to the load; and a temperature-compensated spring, made up of two metals which have different coefficients of expansion and are constructed so that effects from temperature changes are compensated. Until lately a fairly uniform temperature in the seismometer vault was necessary; but the recently developed temperature-compensated systems gives greater temperature tolerance. 3.1.5.2. A Magnetic Suspension System. A novelty consisting of a bar magnet suspended with no apparent support except that it is guided by an upright frame has been available for some time. Suspension, of course, is achieved by magnetic repulsion. The idea was not applied to a seismometer system until recently. The so-called "Springless Sprengnether," a design of the Sprengnether Instrument Company, is probably the first seismometer to which this principle was applied. This design uses magnetic attraction, not repulsion. Figure 24 shows a schematic diagram of this instrument. R and Q are two ring-shaped multipolar magnets, with the poles facing each other as shown. R is a part of the dynamic system, free to rotate about its axis, which is normal to the plane of the paper. Q is rigidly attached to the frame, except that it can be rotated manually in the direction of the arcuate arrow. The mass M and its boom make up the rest of the dynamic system. The forked section of the boom straddles Q and is rigidly attached to the two sides of R. The dynamic system is supported by two sets of crossed Cardan hinges placed along the axis of R. To bring the system into suspension Q is rotated (counterclockwise in the diagram) until the magnetic attraction between the poles exactly balances the gravity moment of the dynamic system. At the time of writing, this instrument is in the prototype stage. It has been satisfactorily proved in trial runs along with other high-sensitivity seismographs. Tests through great ranges of temperature remain. Theoretically temperature changes within operational limits should not require vertical adjustments; this is the instrument's great potential advantage. Its limitation is that the mass moment cannot exceed the magnetic attraction. Seismometer periods of about 1 second have been reached. Other springless vertical seismometers have been considered. In one the mass rests on a stack of piezoelectric plates, and in another the detector is a deep-hole vertical strain seismometer responsive to short periods. These systems have not yet been used in earthquake seismology. 38

Institute of Science and Technology The University of Michigan S N0I FIGURE 24. SPRENGNETHER MAGNETICALLY SUSPENDED VERTICAL SEISMOMETER 3~2. THE SEISMOGRAPH RECORDER 3.2.1. THE TRANSLATING DRUM TYPE. This recorder consists of a rotating drum containing a sheet of record paper or film which translates as it rotates so that an entire span of unattended operation, usually 24 hours, is contained on a single sheet. For recording shortperiod waves, the peripheral speed of the drum is 60 mm/min on paper, or 15 mm/min on film. The effective length of the record is 90 cm, with about 2 cm for overlap. If paper is used, the sheet is 30 cm wide; if film 35 mm. Three-component operation requires three drums, which may or may not be on the same shaft. Film recorders in common use have four film channels, two drums of two channels each or four drums of one channel each, operating on a common shaft. The fourth channel, if used, records automatic time signals to calibrate the local chronometer. 39

Institute of Science and Technology The University of Michigan Film recorders using 70-mm wide strips, designed for 72 hours of unattended operation, are in service in a number of relatively inaccessible stations in southeastern Australia. The "stylus" on recorders using film or photosensitive paper is a sharply focused point of light reflected from a galvanometer mirror. Focus is super-sharp on the film recorders, because the record must be enlarged eight to ten times without losing definition, Other record sizes and drum speeds are in use, but the ones mentioned are the most general. Ink and heat styluses, described later, and delicate mechanical styluses writing on smoked paper, are widely used for visual recording. 3~2.2. THE MOVING-TAPE TYPE. In this case a considerable number of traces from different detectors are recorded side by side. Most exploration seismographs record on rapidly moving photo-sensitive paper or magnetic tape, and operate for a few seconds only. If recording is continuous, the tape speed must be slow for economic operation. Several ink-recording seismographs using paper tape were in operation a few years ago, but they have become obsolescent because of the uncertainty of ink recording and because of storage and processing difficulties, However, two types of moving tape recorders have great potential usefulness. (a) The Develocorder, developed by the Geotechnical Corporation, permits recording of 20 traces side by side on continuously moving 16-mm film. The sensitive side of the film is automatically developed, fixed, and washed very soon after exposure, and within eleven or twelve minutes is available for viewing on a ground glass screen which is a part of the recorder. This permits side by side viewing of traces from an array or from a multi-unit telemeter system without loss of time. (b) Magnetic tape recording is probably the most versatile, but is relatively expensive in first cost, operation, and storage. The best systems record a frequency modulated carrier wave. Recording resolution is improving with increasing technical know-how. As of 1963, frequencies from DC to 50 cps can be resolved on tape moving at 0.3 ips which uses a 10.5-inch roll in 24 hours of operation. Other resolvable maximum frequencies are in proportion to the tape speed. Magnetic tape offers the advantages that it can be played back at any desired speed, and desirable signals can be transferred to another tape, the originals erased, and the tape reused. A desired record may be transferred to a loop and subjected to detailed frequency and other analyses, and it can be reproduced as a visual record which is almost identical to a simultaneous record from direct recording. 40

Institute of Science and Technology The University of Michigan 3o2.3. HOT WIRE RECORDING. The Helicorder, produced by the Geotechnical Corporation, is in wide use. The stylus is the end of a hot resistance loop gliding on the surface of heatsensitive paper wrapped about a standard drum. The traces are sharp and resolution high. Performance is somewhat superior to ink recording and almost equal to photographic recording. Cost of operation is about the same as the latter. However, the records are easily mutilated, and are subject to deterioration with hard use or storage. 3.2.4~ LONG DURATION RECORDING. The Autocorder, a 60-day recorder, was developed by the Sprengnether Instrument Companyo Outwardly, it has nearly the same appearance as a standard drum recorder. It will produce 24 hours of record at 60 mm/min, as does a standard short-period seismograph recorder. At the end of a recording period, recording ceases; the drum returns to the starting position; the exposed paper is conveyed to the interior of the drum; a new supply is drawn into place from a roll of unused paper, also on the interior; and a new recording day starts. The changeover process requires about one minute, and loss of record because of the width of the silt through which the paper is drawn does not exceed 1 mm. Duration of unattended service is in inverse ratio to the paper speed. A 200-foot roll will give 60 days of unattended operation at 60 mm/min. Processing a 200-foot roll of unmarked photographic paper without confusing the dates and/or losing accurate time control is another matter. Magnetic tape recorders operating at 0.02 ips will store 30 days of record having frequencies of 5 cps or less on one 14-inch roll of tape. 3.2o5. DRIVING MECHANISMS. Uniformity of recorder operation is highly important. Mechanical drives governed by a conical pendulum with weights providing the motivation power are being used in Europe and elsewhere. Synchronous motors are in most common usage. They have the disadvantage that their speed varies with the frequency, which is somewhat erratic in isolated areas provided with diesel electric power. Electronic drives of more recent adoption are the most reliable. These drives may be a part of the time standard mechanism. 3.3. SEISMOMETER-TO-RECORDER COMMUNICATIONS SYSTEM In direct recording mechanical seismographs, motion is communicated to the recording drum by a mechanical lever ending in a stylus. In a photomechanical system it is communicated by an optical lever ending in a sharply focused beam of light. In electromagnetic types, motion is communicated through an electrical connection of resistance networks to the galvanometer. Electronic systems are various: a direct or indirect feed to an amplifier at one end, and to galvanometers or record heads at the other; filters, modulators, demodulators, etc. are included in the circuit as necessary. 41

Institute of Science and Technology The University of Michigan 3.4. TIMING SYSTEMS An automatic seismograph time marker has two parts: the chronometer with its programmer, and the time-marking radio for calibrating the chronometer. The chronometer may be an accurately adjusted pendulum clock with a built-in programmer; it may be a marine chronometer, manually or electrically wound, and equipped with minute or half-minute marking contacts, and an external programmer; or it may be an electronic chronometer. Electronic clocks are frequency regulated by a tuning fork in a temperature-regulated oven; or by a crystal in a temperature-regulated oven together with frequency dividers which terminate in the frequency of a synchronous motor driving the clock. Electronic clocks have the advantage that they are highly accurate, 5 parts or less in 107 and they can frequency control the driving mechanism of the recorder. They have the disadvantage that they are initially expensive, and their operation is relatively complicated. The last is not really a disadvantage. Just as repair of an electronic clock requires an electronic expert, repair of a mechanical clock generally requires a chronometer expert; and chronometer experts are scarcer. The chronometer programs displacements of the trace at fixed time intervals, usually once each minute, and an additional mark, or no mark at all, once each hour. Other intervals are used, 1 each 30 seconds or 1 each 10 seconds, with no mark at the minute. For efficient operation, film recorders require additional programming. On Geotechnical film recorders, time marks are programmed once each 10 seconds with no mark at the minute, and longer marks at other programmed intervals. These interval marks are recognized, if the background permits, by relatively long displacements of the trace. Early in the 1940's, the Coast and Geodetic Survey [29] adopted a method of marking by brightening the source lamp for a second or two at programmed intervals; this results in an easily recognized darkened trace. The adopted intervals follow a definite code which permits fixing the accurate time of a signal seen on the screen of a viewer within a few seconds after it is observed. This system is believed superior to others, although there is a risk of shortened life of the source lamps. For accurate time control, radio marks from a time broadcasting system should be automatically placed on the record at least four times daily if the chronometer is mechanical, or at least once each day if it is electronic. In fact, on some electronic clocks, the chronometer output can be synchronized with WWV or other broadcast signals. 42

Institute of Science and Technology The University of Michigan 4 THE SEISMOGRAPH STATION This section, partly a summary of earlier sections, discusses the selection of station sites, the problem of background noise and how to reduce it or minimize its effects, and instruments in general use today. Earlier papers [1, 36] emphasized the importance of adequate seismographic coverage, with emphasis on operation by universities. This emphasis is being retained here except for a few specialized cases beyond the facilities of most universities. These will be discussed first. For any seismograph station, the selection of instruments depends of course on available facilities in relation to background noise and available funds. A highly sensitive short-period seismograph cannot be operated efficiently in a locality subject to high background noise. However, if the high background is only intermittent, resulting from traffic during the day or heavy surf during winter months or stormy periods, the highly sensitive equipment can be used for recording during the quiet periods, and equipment more suitable to the area can be used during noisy periods. In the following discussion, classes A to C refer to stations equipped with electromechanical seismographs which record short-period seismic waves. Class A stations are considered optimum and will be discussed first. In the discussion To is the seismometer period and T the galvanometer period. The galvanometer is normally driven by earth power through a resistance and voltage divider control, and recording is photographic. However, electronic amplification is needed for driving ink or hot-wire recorders, for telemeter systems, and for actuating recording heads of magnetic tape recorders. 4.1o CLASS A STATIONS Stations in this class have peak sensitivity in the period range 0.05 to lo5 seconds. Effective peak magnification for a 1-second wave is 200,000 or more. 4.1.1. A-l-THE ARRAY STATIONo There are intermountain areas where background noise in the 1-cps range is 2 mpL or less. The Blue Mountain Observatory in northeastern Oregon is an example. The array suggested here is more elaborate than that at Blue Mountain, and should be little more expensive. The array suggested by L, F. Bailey for an A-1 station is illustrated in Figure 25. It is a series of 19-element arrays arranged in the form of concentric hexagons stepped down by octaves. Each array of the series is illustrated in the insert. It consists of 12 elements on the periphery, 43

Institute of Science and Technology The University of Michigan, --- - -- -, / \ / \ \ /\ / \ / \ / -— 800 Meters- - Access Road Power Line - FIGURE 25. SUGGESTED SEISMOMETER ARRAY FOR AN A-i STATION 44 v^ \!,..... \ 1^-800 Meters~^I //~~~~~~~^ Acs Road \ Power Line ~ ~~~~ — FIUR 25\UGSE ESOEE ARYFRA -TTO 44~~~~

Institute of Science and Technology The University of Michigan six elements at mid distance to the center, and one element at the center. The larger illustration shows four such arrays nesting downwards so that each smaller array has half the diameter of the next larger one. The dimensions of the array depend on the wave speeds of local background noise. Under average conditions, suggested dimensions are: for the diameters of the outer and consecutively smaller inside arrays, 3.2 km, 1.6 km, 0.8 km, and 0.4 km; and for the separation of the elements in the respective arrays, 800 meters, 400 meters, 200 meters, and 100 meters. This makes a total of 55 elements, since many of the elements are common to two or more of the arrays. This permits wavelength filtering in four octaves. In theory, it will selectively reject noise (Rayleigh or SV waves) having average velocities in the 1- to 10-cps band. In practice, much depends on the spacing, on the nature of the subsurface media, and on the ability to provide duplicate instrumental characteristics including uniform gain settings. Selective weighting or increase in gain settings of the elements toward the center should increase the efficiency of the array. Each element should consist of a 1-cps small-model vertical seismometer buried several feet deep, preferably in a small canister. Johnson-Matheson, National 1-cps, Willmore, or similar seismometers of other makes are suggested. Sprengnether's springless model, if proved, should not need to be buried more than about a foot deep. Preamplifying signals at each seismometer station should obviate the necessity for expensive wire connecting the elements with the central recording facility; this would offset some of the cost of preamplifiers. If funds permit, a class A-1 station should include an array of three or four deep-hole stations for the purpose of determining the emergence angles of incoming signals. The class A-1 station suggested here includes an array of seismometers arranged in a pattern believed to be the optimum consistent with reasonable available funds. It also includes other seismometers, and the necessary recording facilities. This means not that a recording channel should be provided for the output of each element in the array, but that, by appropriate selectivity and weighting, selected elements and combinations of elements can be recorded. Magnetic tape recording is recommended for this purpose. Instrumentation should therefore include: Seismometers: Array seismometers (with preamplifiers); one set of three-component short-period; one set of three-component long-period. Deep-hole seismometers in an array if possible, for experimental comparative recording. Recorders: Conventional film or paper recorders, one drum for each seismometer not of the array. Develocorders, rate not exceeding 15 mm/min, preferably two for recording from the array. 45

Institute of Science and Technology The University of Michigan Magnetic tape recorders, rate 0.3 or 0.6 ips, with at least 19 channels for recording from the array and three more channels from the high-frequency seismometers. Magnetic tape recorder, 0.02 ips, for experimental long-range recording from long-period and short-period seismometers. Two or three Helicorders for sampling from the array and/or individual seismometers. One ink recorder for continuous visual operation. One 12-channel visicorder for reproducing direct or filtered signals from the magnetic tapes. Communications The array: Preamplification at each unit will materially reduce the cost of wire leading from elements of the array. This reduction can be accomplished by radiating power lines outward from the center, and, where practicable, by returning the seismometer outputs over the same facility. Preamplifiers in common use are virtually the first stage of a telemetering system~ The galvanometer and galvanometer bridge. Lehner-Griffith or Geotech galvanometers are recommended. Manufacturers of short-period seismographs usually supply 5-cps galvanometers. If the seismometer has a period of 1.0 second, the seismometer-galvanometer connection in general use has a peak magnification for periods of about 0.3 second (see Figure 26). This combination seems to give the best results on oceanic islands and perhaps in certain coastal areas where 1-second microseisms are prevalent over shorter-period ones, especially if the station site is fairly free from culture noise. If the galvanometer has a period of 0.5 second (2 cps), the seismograph is less sensitive to culture noise or to disturbances caused by wind blowing on nearby rocks, trees, or buildings. In a conventional seismograph station, a 0.5- to 1.0-second galvanometer is recommended. For arrays, oceanic islands, and certain coastal areas, 0.2-second galvanometers are recommended. The seismometer-to-galvanometer coupling circuit is shown in Figure 27. Gain is adjusted by varying the shunt resistances; the series resistances must then be adjusted if the damping balance is to be maintained. Circuits provided by the Geotechnical Corporation do this automatically. Timing. A crystal-controlled electronic chronometer (time or frequency standard) is recommended. This standard should control the frequency of all recorder drives. Frequency standards supplied by some manufacturers have a built-in radio for calibrating with a world time standard. 46

Institute of Science and Technology The University of Michigan T = 1.0 sec 0.1 S T = 0.2 sec g ---- T = 1.0 sec S T = 0.5 sec ___ USING GAL MET = 1.0 sec p; T 1.0 sec 21 May 60/RCS/223-118 T = 1.0 sec g.01 T =1.0sec SEIS GALVO T = 2.0 sec Critically Critically Damped Damped 7453 0.1 1 10 PERIOD IN SECONDS FIGURE 26. AMPLITUDE RESPONSE CURVES OF VARIABLE-RELUCTANCE SHORT-PERIOD SEISMOGRAPH USING GALVANOMETERS WITH PERIODS OF 0.2, 0.5, 1.0 AND 2.0 SECONDS 47

Institute of Science and Technology The University of Michigan db attenuation mag. trim O I. 5^ 70) 70QL 79-1 79- -— 11:62. 5A 0n 0 l Of 1 meter Geotech Geotech Geotech UED Horizontal Seismometer, Geotech Galvo Galvanometer Photographic Shake Table Model 1101 Control, Model 8438 Model 4100-281 Recorder Model 6845 Ser. #208 Serial #40 Serial #2579 Model DR-273 20 microns "A" coil config. fog = 1. 33 cps p-p fos = 1.0 cps CDRX = 79 ohms CDRX = 120 ohms FIGURE 27. GEOTECH ATTENUATION UNIT Seismometer vaults. Housing for the array seismometershas been described: a buried watertight canister just large enough to hold the seismometer and preamplifier. See Class A-2 for other seismometer housing. Office and recorder housing. The Blue Mountain Station (see Figure 28) serves as an example of housing requirements of an array station. The outbuilding is for storage of supplies and equipment. The main building is provided with (1) a dark room with galvanometer and recorder piers; (2) a photographic dark room for processing photo paper; (3) a room provided with various visual recorders; (4) an instrument panel for control of circuits from the array, which includes a magnetic tape recorder (two or more are needed for the array proposed here); (5) office space; (6) a wash room; and (7) temperature control apparatus. FIGURE 28. BLUE MOUNTAIN SEISMOLOGICAL OBSERVATORY 48

Institute of Science and Technology The University of Michigan 4.1.2. CLASS A-2 STATIONS. These are first class, non-array stations. A Coast and Geodetic Survey World-Wide Standard seismograph station, which adequately represents a station in this category, has the following facilities: Seismometers. Benioff (Geotech) short-period VR-3 components; TO = 1.0 second, T = 0.75 g -9 second; external critical damping resistance, 80 ohms; galvanometer sensitivity, 10 amperes per mm trace deflection at 1 meter. Sprengnether long-period, three-component; TO = 30 seconds,T = 100 seconds. Recorders. Photosensitive paper on 30- x 90-cm drums, one drum for each galvanometer; paper speeds 60 mm/min for short-periods and 30 mm/min or 15 mm/min for long-periods. Time and Frequency Standard. Crystal controlled with built-in radio programmed to receive time signals at 6-hour intervals. Power supply. 28-volt Nickel-Cadmium (Nicad) batteries, with trickle charger. Driving motors are frequency controlled by the time standard. Housing. Figure 29 is a housing plan recommended for a world standard seismograph. For better operation, the seismometer vault should be isolated 100 meters more or less from the recorder building, and on competent bedrock if at all possible. The long-period instrument should be temperature protected with styrofoam covers whose tops are provided with anticonvection heat control. Discussion: Specification for a World-Wide Standard seismograph, outside of housing, background level, and underground conditions, are unvarying. Otherwise the seismograph would not be standard. The only variable is the gain, which must be adjusted according to the background noise level. Recommended instrumentation for non-standard Class A-2 stations includes: Seismom eters. Three-component sets of either Benioff MC or VR; Lehner-Griffith MC; Johnson-Matheson; Willmore; or National 1-cps. One "Springless Sprengnether" vertical Three long-period seismometers with photo paper recorder and 90- to 100-second galvanometers. Recorders. Paper recorders as for standard stations. Film recorders for SP seismographs. Develocorder if telemetering from other stations is employed. 49

CC \ 6 inches nches 3 inch Conduit - ). ~1 e' 2 inch Gooseneck-.l X 1" Triple Drum Recorder i LP-H Pier LP-Z Galvanometers ouble 0\ I I 0 00oors i 0 0 0 I / L II ^ /P-H " Recorder Room Triple Drum Recorder SP-H SPH Pier Seismometer Vault 6 inches - 6 -- 8 feet 1 inches inches FIG RE 29. MI IMU S ATI N ACI IT FO W RLD WI E S IS IC NS RUM NT _ Console I <D 6 inches -61 ~ 9 feet~16 F incs inches inches FIGURE 29. MINIMUM STATION FACILITY FOR WORLD-WIDE SEISMIC INSTRUMENTS Q

Institute of Science and Technology The University of Michigan Advantages of paper and film recording: Paper-The entire record can be scanned quickly and read with greater speed. Reading requires no projector or enlarging apparatus, and results in less eye strain. Film-~If the standard 35-mm size is used, the initial cost is one-fourth that of paper, and storage requirements are one-sixth; duplication and development of records is relatively simple. 4.1.3. CLASS A-3 STATIONS. This includes a station whose site meets the specifications for a Class A station, but which has facilities for only one to three seismometers. The housing should be large enough to include a three-seismometer operation. However, if available funds permit only one seismometer, it should be a short-period vertical, with paper recording facilities. If the station is to be a part of the seismological community, accurate timing, to within 0.1 to 0.2 second, is necessary. A precision pendulum clock or spring-driven chronometer is satisfactory, provided that it is adjusted properly and radio calibrated at least once, and preferably four times, daily. Visual time calibration is recommended only if accompanied by automatic recording of radio time marks on the record. If funds permit operation of three seismometers, one short-period vertical, as described above, and two intermediate-period horizontal seismometers (To = 4 to 8 seconds, T = 2 to 4 seconds) are recommended. In addition to the S waves of most earthquakes, the intermediateperiod instruments are sensitive to ocean-generated microseisms. Recommended instruments are the Wilson-Lamison horizontal, the Wood-Anderson intermediate-period horizontal (which is direct recording), or the Sprengnether intermediate-period horizontal. Stations now using this combination include the South Pole (until January 1963) and Marie Byrd Stations in Antarctica, and the station in Fayetteville, Arkansas. 4.2. CLASS B STATIONS These stations have peak sensitivity in the period range from 0.5 to 2.0 seconds, and peak magnification up to 150,000. They have the same characteristics as Class A stations except that background noise lowers their peak sensitivity. Instruments: (1) Same as A-2 or A-3 except that the short-period instruments have lower magnification. (2) Wilson-Lamison MC seismometer, an inelegant and low-cost instrument. It uses the same galvanometer as is used in Class A stations: TO = 1.0 second, T = 0.5 second. 4.3. CLASS C STATIONS These stations have peak response to periods 0.5 to 2.0 seconds, and peak magnification up to 25,000. 51

Institute of Science and Technology The University of Michigan Instruments: (1) Same as under A2, A3, or B, except at lower magnification. (2) Wilson-Lamison vertical or vertical and horizontal MC seismometers. (3) Sprengnether MC short-period vertical or vertical and horizontal seismometers. These can be operated in an industrial area if the foundation of the building where they are installed is satisfactory. Damping is provided by a metal plate moving in a magnetic field. TO = 2.0 second or more; T is chosen to match the seismometer. 4.4. SPECIAL PURPOSE SEISMOGRAPH STATIONS 4.4.1. THE LISTENING POST STATION. This is a proposed station adapted to a seismic area, especially along an active fault. Its purpose is to monitor the pattern of fault activity before, during, and after earthquakes. Instrumentation should be that for a Class A-2 or Class B station; also provision should be made for continuous recording on magnetic tape of ground frequencies from 0.1 to 100 cps in four dynamic ranges of the lower frequencies. This includes output on six channels from a strong motion seismograph, three channels for acceleration and three for ground displacement. In the high magnification range, frequencies of 1 or 2 cps and higher are to be taken from deep enough underground to avoid local wind or culture noise. Additional pickup of microtremors along the fault with telemetering to the main station should be provided. In addition, capability for intermittent recording of all sounds in the audible range and higher should be available. 4.4.2. THE UNMANNED SEISMIC OBSERVATORY. A posible treaty on a moratorium of nuclear explosions may include the installation of a number of portable seismographs in unfriendly territory, with a provision for an inspection of these stations at limited intervals, say once a month. Several versions have been suggested: (1) A limited number of "Geneva" stations, in enemy territory, each surrounded by outposts or satellite stations equipped to telemeter seismic signals, continuously or on demand, to the parent station, the outstation to be in the form of a tamper proof box ("tamper proof," as used here, means only that any tampering whatever will leave telltale marks). (2) The same as in (1) except that signals from the outpost are recorded locally on a longduration recorder. (3) The same as in (2) except that there is no parent station; only the outpost stations on unfriendly territory are permitted, with a provision that the records may be changed at stated intervals, say once a month. 52

Institute of Science and Technology The University of Michigan An early proposal was that the outpost seismometers record on, say, 6-hour loops, and that information on the loops be broadcast on demand by telemeter. Six seismometers were to be provided, three components each of long- and short-period ones. A recent suggestion provides recording the output from six seismometers on magnetic tape traveling at the rate of 0.02 ips. This permits 30 days of record on a 14-inch roll of tape. If this method is to be successful, accurate unattended time control is necessary. But if presentday long- and short-period seismometers, a time standard, 14-inch magnetic tape transport, and the necessary batteries are all to be contained in a box, that box could hardly be classed as portable. However, if the seismometers are condensed to the size of Willmores or the moon seismometers, and if the electronics are likewise condensed, which is within reason, the longest dimension of the black box need not exceed 100 cm, which is the longest dimension of the tape transport. Down-hole long-period seismometers are in the design stage. If they are developed, then there will be some hope of a six-component 30-day seismograph which can be housed in a fairly small container. If specifications are less demanding, the components of a 60-day portable seismograph are now available. If the short-period vertical is sufficient, recommended instrumentation includes (1) a magnetically suspended seismometer, or some other seismometer with a temperaturecompensated suspension; (2) a 60-day Autocorder driven by an electronic time standard modified to compact dimensions; and (3) a 28-volt "Nicad" battery, with a windcharger nearby to supply power. Each additional seismometer requires an additional recorder and more power, but not an additional time standard. 5 CAPABILITIES OF SOME SENSITIVE SEISMOGRAPH STATIONS* Detection capabilities of certain of the well known sensitive seismograph stations will now be reviewed. Since we are for the most part interested in earthquake magnitudes of 4.5 to 5.5, discussion here will be limited to events in this range. However, before proceeding further, it should be pointed out that the large number of methods of assigning magnitude has led to considerable confusion. Magnitudes reported by regular observatories are generally derived from surface-wave amplitudes. VELA UNIFORM participants have used the so-called uniform magnitude, which in practice is derived mainly from the maximum amplitude in the first few cycles of the P wave and should more properly be called body-wave magnitude. A cross check of *This section was prepared by J. N. Jordan and D. W. Gordon, geophysicsts, U. S. Coast and Geodetic Survey. 53

Institute of Science and Technology The University of Michigan a large number of earthquakes has revealed that magnitudes reported by independent observatories are often 0.5 to 1.0 units higher than the assigned uniform magnitudes reported by VELA UNIFORM participants. The determination of the smallest earthquake which would be detected by a given station at a given epicentral distance is very complex. An approximate answer can be obtained by comparing the expected amplitudes extrapolated from the depth-distance (Q) tables with minimum readable amplitudes compatible with the response curve of the particular instrument. This answer can then be checked by examining the station bulletins for reading from earthquakes of known magnitude. However, cross checking reveals that the apparent sensitivity of a given station varies with azimuth to the epicenter as well as with distance. A statistical analysis of this problem has been performed at the Wichita Mountains Observatory (WMO), a highly sensitive Class A-1 array station. As Figure 30 shows, WMO detected 6 5 - College, theoretical QA~~~~~~ ~~~curve ~*~ *^~~~~- \College, observed capability / B / f**^ ~~ ^.**^ / capability 200 30~ 400 500 600 700 800 900 1000 1000 2000 4000 6000 8000 10,000 12,000 DISTANCE (km) FIGURE 30. LOWER LIMIT OF P-WAVE RECEPTION AS A FUNCTION OF MAGNITUDE AND DISTANCE 54

Institute of Science and Technology The University of Michigan 90% of the earthquakes with unified magnitudes falling above the dashed line. The curve is fairly flat between about 500 and 900, and WMO can detect 90% of magnitude 4.0 earthquakes in this range. Over its entire length the WMO curve is roughly parallel to the theoretical lower limit of earthquake detectability for College, Alaska, which is shown by the solid curve in the figure. The theoretical limit was based on Q values for shallow earthquakes, the instrumental response curve, and a signal-to-noise ratio of 2 to 1. The dotted line in the figure represents College observations of Kamchatka earthquakes between 1953 and 1960. To obtain a statistical appreciation of the relative capability of sensitive network stations, about 130 earthquakes were selected for study from the Bulletin Central International Seismologique (BCIS) for 1959. The earthquakes were evenly distributed in the magnitude range from 4.5 to 5.5, and widely scattered geographically. The list of station readings for each earthquake was examined for P-readings from ten well known sensitive stations. These stations and pertinent data relative to their instrumentation are listed in Table 1. The scatter of the magnitudes quoted in BCIS for a specific earthquake was often as great as one unit; in many cases the magnitude was determined by only one station. Another troublesome aspect of the data is that the earthquakes TABLE I. STATIONS CONSIDERED IN SENSITIVITY STUDY Station Seismometer TO T V 0 g m (seconds) (seconds) (km) College, Alaska Benioff MC 1.5 0.5 425 Eureka, Nevada Benioff MC 1.1 0.6 425 South Pole, Benioff MC 1.6 0.5 100 Antarctica Charters Towers, Benioff VR 1.0 0.2 120 Australia Lwiro, Congo Benioff VR 1.0 0.2 100 Matsushiro, Benioff VR 1.0 0.2 110 Japan Kiruna, Sweden Grenet-Coulomb 1.4 0.7 11 Tamanrassett, Grenet-Coulomb 1.8 0.8 47 Algeria Resolute, Canada Willmore 1.0 0.3 300 Pruhonice, SVSN 1.0 1.5 36 Czechoslovakia All Seismographs are short-period vertical TO = Seismometer period T = Galvanometer period V = Maximum magnification x 103 m 55

Institute of Science and Technology The University of Michigan which are assigned magnitudes are not evenly distributed geographically. For instance, the magnitude of earthquakes to South America is seldom reported for events of less than magnitude 6. This problem should be alleviated in the near future when a program being planned by the Coast and Goedetic Survey becomes operational. This program calls for routine publishing of magnitudes determined at the standard stations of the World-Wide net, in conjunction with the present reports of arrival times. Figure 31 shows the results of the BCIS search for eight stations that have reported a reasonably adequate amount of data. To minimize the effect of the variation of sensitivity with distance, the graph is limited to data for the distance range from 50~ to 90~. The figure indicates the expected general correlation between sensitivity and instrument magnification. It must be noted that the percentages of positive results depend not only on the mechanical capability of a) % b 100 o0 k~ ~76 a I 80 44 29 0 o0 44C d 660 48 o.. o 21 51 CY 40 45 53 20 0 3,?,_________ 50 100 300 350 400 450 APPROXIMATE MAXIMUM INSTRUMENT MAGNIFICATION x 103 FIGURE 31. P READINGS REPORTED IN DISTANCE RANGE FROM 50 TO 90 DEGREES. 44 Total No. of Events Considered Magnitude Range 4 1/2-5 1/2 Magnitude Range 4 1/2-5 56

Institute of Science and Technology The University of Michigan the seismographs but also on the efficiency of the reporting system. The results indicate that stations of the quality of Eureka, College, and Resolute report P readings for about 60%o of the earthquakes between magnitudes 4.5 and 5, for epicentral distances between 500 and 900. On the other hand, stations operating at about 100,000 magnification detect only about 20% to 40% of such earthquakes. Figure 32, a frequency polygon with 100 increments, shows the percentage of reported readings for magnitude 4.5 to 5.5 earthquakes over the entire P range. Similarly, Figure 33 shows the percentages reported in the magnitude range from 4.5 to 5. The dashed line represents the combined data from Eureka, College, and Resolute; the solid line represents the combined data from the other seven selected stations, which operate at considerably smaller magnifications. The extreme maximum and minimum, shown by the dashed line, are probably strongly weighted by anomalous reception of earthquake data at certain distances or from specific areas. The minimum between 300 and 400 represents data from California and Nevada events recorded at College and Resolute; the high peak between 600 and 700 depends almost entirely on College and Resolute readings. However, it is interesting that the combined data from the seven stations also show a zone of maximum sensitivity between 600 and 700. Table 2 contains a similar analysis of P-wave data from three nuclear explosions, all having a unified magnitude of approximately 5. In general, the table indicates positive results. Five of the stations recorded all of the events, and of the 21 possible readings in the P range, only two were missed. Unfortunately, no data were available from Tamanrasset. However, this station probably recorded at least two of the three events. The data presented here apparently indicate that reception is better for nuclear tests than for earthquakes. This may be due in part to the different method of computing magnitudes; i.e., earthquake magnitudes quoted in the BCIS are generally determined by surface-wave amplitudes, and the magnitudes given for the explosions depend largely on measurements of the P wave alone. Conclusions in this section are based on incomplete data. As expected, however, a station's detection capability is roughly parallel to the sensitivity of the seismograph, which in turn is governed by the level of background noise or on the station's ability to reduce that noise. The problem is further complicated by evident lateral and vertical inhomogeneities deep within the earth. 57

Co^~~~~~~~~~~~~~~~~~~ -~~~~~_1-. CD 0 cn CD ~~5' % 177 9 D 100 2 " Indicates Size -T 80le- e52,53 Eu a R l51 of Sample r 23 o 0 29 \ 60- 38 W~~~. 6054 42 D 40 E — 61 87 " 20- 43 38 100 200 300 400 500 600 700 800 900 1000 1050 2000 4000 6000 8000 10,000 12,000 DISTANCE (km) C FIGURE 32. P READINGS REPORTED FOR EARTHQUAKES OF MAGNITUDE FROM 4.5 TO 5.5. Kiruna, Matsushiro, Pruhonice, Tamanrasset, Lwiro, Charters Towers, South Pole College, Eureka, Resolute - o

C o 159 5~~~~~~~~~~~~~~~~~~~~~~5 % 9 -lOO 10 10 / 4 80 2 \ Indicates Size of Sample Frl ~~~~~21 4 200 SV1 ~ ~ ~ ~ ~ ~ ~ ~ 9 (\150 S 60 240 40 FIGURE 33. P READINGS REPORTED FOR EARTHQUAKES OF MAGNITUDE FROM 4.5 TO 5.0\ Kiruna, Matsushiro, Pruhonice, Tamanrasset, Lwiro, Charters Towers, South Pole College, Eureka, Resolute nn ^ - \ / ^ - - ^ ^ \ _______ I I ~ _______ I I _______ I I I _______ I _______ I _______ I ^ ^ o IQO 20^ 30^ 40^ 50^ 600 ^0 800 90^ 100^ 1050~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.

TABLE I ANALYSIS OF SEISMIC SIGNALS FROM THREE EVENTS C 0 en 0. TABLE II. ANALYSIS OF SEISMIC SIGNALS FROM THREE EVENTS 5 -SAHARA - May 1, 1962 SEMIPALATINSK - Fle,. 2, 1962 HAYMAKER - June 27, 1962 Distance Signal Ground Ground Distance Signal Ground Ground Distance Signal Ground Ground (degrees) Amplitude Period (degrees) Amplitude Period (degrees) Amplitude Period (ma) (seconds) (mW) (seconds) (ma) (seconds) College 89 + 2 0.7 60 + 76 0.8 34 + 38 0.9 Charters 144 + (PKP) 93 108 Towers Eureka 96 + 6 0.7 90 + 34 0.6 3 + Off record Kiruna 45 + 125 0.8 33 + 580 0.8 78 + 27 0.6 Lwiro 35 + ** 67 + * 132 Matsushiro 104 48 + * 80 + * * Pruhonice 27 + * 40 + 28 1.0 83 +* * Resolute 69 + 5 0.9 62 + * 39 +* * South Pole 114 ** * 140 + (PKP) 127 + (PKP) - Tamanrassett 2 ** * 61 ** * 99 ** * C c + Readable signal - No signal C * No information available Note: Ground amplitudes are measured 1/2 peak to peak. Co 0^

Institute of Science and Technology The University of Michigan 6 SUMMARY In summary, Table 3 will serve as a guide toward the selection of a site for a highly sensitive short-period seismograph station. If the locality of the station is limited to the physical environment near the/institution operating the station, the table will assist in the selection of the site and the most suitable instrumentation. TABLE III. RECOMMENDED MINIMAL DISTANCES FROM SOURCES OF DISTURBANCE TO SEISMOGRAPH STATION- SHORT-PERIOD SEISMOGRAPHS Seismometer on Hard Massive Seismometer on Hard Clay Source of Disturbances Rock, Granite, Ouartizite, etc. Hardpan, etc. A B C A B C km km km km km km 1. Oceans-a. with coastal mountain systems as Pacific North and South America 300 50 1 300 50 1 b. with broad central and coastal plains as eastern North and South America 1000 200 10 1000 200 20 2. Inland seas, bays, and large lakes-Great Lakes a 150 25 1 150 25 1 b 500 100 5 500 100 5 3. High waterfalls, cataracts or excessive flow over c 40 10 5 50 15 5 large dams, e.g.,Niagara, Grand Coulee Dam d 60 15 5 100 25 10 4. Transcontinental oil or gas pipe lines, etc. c 20 10 5 30 15 5 d 100 30 10 100 30 10 5. Small lakes c 20 10 1 20 10 1 d 50 15 1 50 15 5 6. Reciprocatingpowerplantmachinery,rockcrushers, c 15 3 1 20 5 2 heavy machinery, etc. d 25 5 1 40 15 3 7. Low waterfalls, rapids of a large river, inter- c 5 2 0 15 5 1 mittent flow over large dams d 15 3 1 25 8 2 8. Railways, if frequent operation c 6 3 1 10 5 1 d 15 5 1 20 10 1 9. Airports and airways having heavy traffic 6 3 1 6 3 1 meters 10. Non-reciprocating power plant machinery, c 2 0.5 100 10 4 1 balanced industrial machinery d 4 1 200 15 6 1 11. Busyhighway-nearly continuous traffic or meters meters mechanized farm area 1 300 100 6 1 500 meters 12. Graded country roads, high buildings 300 200 50 2 1 500 13. Low buildings and high trees, wind charger for meters meters seismograph batteries if coupled to ground 100 30 10 300 100 50 14. High fence, low trees, large rocks, high bushes 50 25 5 100 50 10 15. Wind charger for seismograph batteries if decoupled from ground 30 15 5 60 30 10 a.-as in la b.-as in lb c.-Source and seismometer on widely different formations, or that mountain ranges or alluvial valleys intervene. d.-Source and seismometer on same formation and with no intervening alluvial valley or mountain range. A-Gain for 1 cps 200,000 or more; B —Gain 50,000 to 150,000; C-Less than 25,000 61

Institute of Science and Technology The University of Michigan The distances in the table are based largely on the experience of the writer or on material from available literature, and are subject to modification according to local needs or to regional physical conditions. The table applies to two types of subsurface conditions: (a) granite, massive limestone, quartzite, or basement complex, and (b) hardpan, caliche, or firm clay in an alluvial area. Less firm alluvial material is not considered; subsurface rock such as shale, volcanics, and loosely consolidated sandstone is intermediate between (a) and (b). An additional subdivision is made on the basis of homogeneity or inhomogeneity of the formations between the noise source and the station. REFERENCES 1. D. S. Carder, The Seismograph and the Seismograph Station, U. S. Coast and Geodetic Survey Publication, September 1956. 2. J. A. Anderson and H. O. Wood, "Description and Theory of the Torsion Seismometer," Bull. Seis. Soc. Am., March 1925, Vol. 15, pp. 1-78. 3. H. E. McComb and Frank Wenner, "Shaking-Table Investigations of Teleseismic Seismometers," Bull. Seis. Soc. Am., 1936, Vol. 26, pp. 291-316. 4. D. S. Carder, "Response of a Torsion Seismometer to Sustained Vibrations," Bull. Seis. Soc. Am., October 1940, Vol. 30, pp. 343-351. 5. B. Galitzen, "Vorlesungen Uber Seismometrie," (Translated into German by O. Hecker)- Leipzig und Berlin, DruckandVerlag von B. G. Teubrier, 1914. 6. B. Gutenberg and F. Andrews, Bibliography on Microseisms, Second Edition, Parts I (1952) and II (1956), California Institute of Technology, Div. of Earth Sci. Contrib. 602, 759. This is a complete bibliography up to 1955. 7. E. W. Kammer and J. E. D. Dinger, Results of Field Experiments During 1951 on Microseisms as Related to Storms at Sea, Naval Research Laboratory, Mechanics Div., 1952. 8. D. S. Carder and R. E. Eppley, The Microseismic Program of the U. S. Navy, a Terminal Report, U. S. Coast and Geodetic Survey, 1959. 9. M. S. Longuet-Higgins, "A Theory of the Origin of Microseisms," Phil. Trans. Royal Soc., London (A), 1950, Volume 243, pp. 1-35. 10. B. Gutenberg,,"Untersuchungen uiber die Bodenunruhe mit Perioden von 45-105 in Europa," Veroff. Zentr. Int. Seism. Assn., Strassburg, 1921, pp. 1-106. 11. D. S. Carder, "The Continent and Ocean Floor as Transmitting Media of Microseims," Earthq. Notes, 1951, Vol. 22, p. 26. 12. B. Gutenberg, "Microseisms in North America," Bull. Seis. Soc. Am., 1931, Vol. 21, pp. 1-24. 13. D. S. Carder, "Transmission of Microseisms Across North America and the Western North Atlantic," Trans. Am. Geoph. Union, 1955, Vol. 36, pp. 83.8-842. 14. Joseph Lynch, "The Great Lakes, a Source of Two-Second Frontal Microseisms," Trans. Am. Geoph. Union, 1952, Vol. 33, pp. 432-434; Earthquake Notes, 1953, Vol. 24, pp. 17-18. 62

Institute of Science and Technology The University of Michigan 15. Robert C. Shopland and Ernest Stephens, Jr., "Site Selection-The Noise Problem," Problems in Seismic Background Noise, Vesiac Advisory Report No. 4410-32-X, Institute of Science and Technology, The University of Michigan, Ann Arbor, Mich., October 1962, pp. 1-15. 16. D. S. Carder and M. H. Gilmore, "Ground Vibrations," Bull Seis. Soc. Am., 1945, Vol. 35, pp. 13-26. 17. D. S. Carder and W. V. Mickey, Seismic Ground Effects from Coupled and Decoupled Shots in Salt, Project COWBOY, Final Report, U. S. Atomic Energy Commission, 1960. 18. Gordon E. Frantti, "Short-Period Seismic Noise," in Problems in Seismic Background Noise, VESIAC Advisory Report No. 4410-32-X, Institute of Science and Technology, The University of Michigan, Ann Arbor, Mich., October 1962, pp. 16-23. 19. Field Study of Variation in Characteristics of Seismic Noise and Signals with Geologic and Geographic Environment, Semi-annual Progress Report, United ElectroDynamics Inc., VT/078-23, 1962. 20. Seismograph Station Abbreviations, U. S. Coast and Geodetic Survey, June 1962 or later. 21. Y. Rocard, "Various Characteristics and Some Possible Causes of Seismic Ground Noise," in Problems in Seismic Background Noise, VESIAC Advisory Report No. 4410-32-X, Institute of Science and Technology, The University of Michigan, Ann Arbor,'Mich., October 1962, pp. 85-93. 22. Monthly Report No. 15, Project VT/1139, The Geotechnical Corporation, Garland, Texas, September 1962. 23. Monthly Report No. 18, Project VT/1139, The Geotechnical Corporation, Garland, Texas, December 1962. 24. D. R. Van Sandt, Detection of Distant Explosions by Seismic Measurements in Deep Boreholes, Report AF 19(604)-8346, Jersey Production Research Co., July 1962. 25. D. Prentiss and J. Ewing, "The Seismic Motion of the Deep Ocean Floor, " Bull. Seis. Soc. Am., July 1963, Vol. 53, pp. 765-781. 26. Wichita Mountains Seismological Observatory-Phase III, The Geotechnical Corporation, Technical Report No. 62-8, May 1962. 27. Monthly Report No. 15, Project VT/1124, Texas Instruments, Inc., Houston, Texas, August-September 1962. 28. Technical Summary Report on Study of a Three-Dimensional Seismic Detection System, Century Geophysical Corporation, for Air Force Office of Scientific Research, 1962. 29. D. S. Carder, "Seismic Investigations in the Boulder Dam Area, 1940-1944, and the Influence of Reservoir Loading on Local Earthquake Activity," Bull. Seis. Soc. Am., 1945, Vol. 35, pp. 175-192. 30. R. M. Wilson and A. E. Lamison, A Simple Short-Period Vertical Seismometer, U. S. Coast and Geodetic Survey, 1951. 31. Ben S. Melton, and D. P. Johnson, "Inertial Seismograph Design-Limitations in Principle and Practice," Proc. IRE, 1962, Vol. 50, pp. 2328-2339. 63

Institute of Science and Technology The University of Michigan 32. "SKM-3 Seismographs of High Sensitivity," Institute of Earth Physics, Soviet Academy of Sciences, 1962, Translated for the Institute of Science and Technology, The University of Michigan, Ann Arbor, Mich., 33. H. Benioff, "A New Vertical Seismograph," Bull. Seis. Soc. Am., 1932, Vol. 22, p. 155. 34. James J. Devlin, S.J., "The Electromechnical Transducer in the New Benioff Seismograph," Bull. Seis. Soc. Am., 1938, Vol. 28, p. 255. 35. Deep-Borehole Seismic Research, VESIAC Special Report 4410-16-X, Institute of Science and Technology, The University of Michigan, Ann Arbor, Mich., March 1962. 36. Frank Neumann, "The University Seismograph Problem," Earthquake Notes, 1945, Vol. 19, p. 175. 64

Institute of Science and Technology The University of Michigan DISTRIBUTION LIST Advanced Research Projects Agency Air Force Office of Aerospace Research Air Force Systems Command Nuclear Test Detection Office Headquarters Andrews Air Force Base The Pentagon Washington 25, D. C. (2) Washington 25, D. C. (15) Washington 25, D. C. (4) ATTN: Major Philip J. Crossman (RROS) Air Force Systems Command Advanced Research Projects Agency Andrews Air Force Base The Pentagon Air Forceers sWashington 25, D. C. Headquarters Washington 25, D. C. Washington 25, D. C. (2) ATTN: Major J. C. Stokes, USAF ATTN: Dr. Charles C. Bates ATTN: Lt. Col. James A. Fava (RROS) HQ USAF (AFTAC) Advanced Research Projects Agency AVELA Seismological Center The Pentagon Electronic Systems Division Washington 25, D. C. (15) Electronic Systems Division Washington 25, D. C. L. G. Hanscom Field ATTN: TD-1, Lt. Col. Ridenour ATTN: Mr. Russell Beard Bedford, Massachusetts University of Alaska Advanced Research Projects Agency System 477-L (ESH) Geophysical Institute The Pentagon e - eCollege, Alaska Washington 25, D. C. Air Force Cambridge Research Laboratories ATTN: Dr. Edward Berg ATTN: Mr. R. Black Headquarters L. G. Hanscom Field Alberta, University of Advanced Research Projects Agency Bedford, Massachusetts (5) Calgary Campus The Pentagon Calgary Alberta Canada Washington 25, D. C. ATTN: Major Robert A. Gray USAF ATTN: Prof. K. Vozoff ATTN: Mr. William Bolton VIA: Office of the Secretary of Defense Air Force, Department of the Advanced Research Projects Agency Directorate of Research and Technology Advanced Research Projects Agency The Pentagon DCS/D Pentagon Washington 25, D. C. Washington 25, D. C. ATTN: Mr. D. Clements ATTN: Major E. C. Lowry ATTN: J. R. Loftis, Jr. Advanced Research Projects Agency Air Force Office of Aerospace Research Allied Research Associates, Inc. The Pentagon Europea43 Leon Street Washington 25, D. C. The Shell Building Boston, Massachusetts ATTN: Mr. Robert Frosch 47 Rue Cantersteen American Geological Institute Brussels, Belgium Brussels, Belgium 2101 Constitution Ave., N. W. Advanced Research Projects Agency VIA: Office of the Secretary of Defense Washington, D. C. The Pentagon Advanced Research Projects Agency Washington 25, D. C. The Pentagon ATTN: Mr. Robert C. Stephenson ATTN: Lt. Col. Robert Harris Washington 25, D. C. ALPENS ATTN: J. R. Loftis, Jr. 10 Rue Claude Bernard Advanced Research Projects Agency rance ~~~~~~~~~~~~~~~~The Pentagon ~Paris V, France Washington 25, D.C. Force, DepartmeATTN: Prof. Y. Rocard Office of the Assistant Secretary of the Air Force ATTN: Dr. Robert L. Sproull (Research and Development) VIA: Office of the Secretary Washington 25, D. C. Advanced Research Projects Agency Advanced Research Projects Agency The Pentagon The Pentagon M F J Washington 25, D. C. Washington 25, D. C. (10) Air Force, Department of the ATTN: J. R. Loftis, Jr. ATTN: TIO Office of the Deputy Chief of Staff Development American Museum of Natural History Aerospace Corporation Washington 25, D. C. Central Park West at 79th Street P. 0. Box 95085 New York 24, New York Los Angeles 45, California (2) ATTN: Lt. Col. G. T. Grottle AFDSD/MS ATTN: Mr. George H. Goodwin, Jr., Librarian ATTN: Mr. Carlton M. Beyer Air Force Office of Aerospace Research American University Aerospace Corporation Commander Department of Earth Science 2400 East El Segundo Boulevard Tempo D Washington, D. C., 20016 El Segundo, California 6th and Constitution Ave., S. W. ATTN: Professor Paul Bauer ATTN: Dr. Byron P. Leonard Washington 25, D. C. (4)' Arctic Institute of North America AGI Technical Publications Air Force Office of Scientific Research 3458 Redpath Street 2101 Constitution Ave., N. W. Executive Director a S Washington 25, D. C. Washington 25, D. C. Montreal 25, P. Q., Canada ATTN: Mr. Martin Russell ATTN: SRYG. John C. Reed Executive Director Air Force, U. S., Headquarters Air Force Office of Scientific Research VIA: Office of the Secretary of Defense Aeronautical Chart & Information Center Washington 25, D. C. (4) Advanced Research Projects Agency Second and Arsenal ATTN:SRPG The Pentagon St. Louis 18, Missouri Lt. Col. William J. Best Washington 25, D. C. ATTN: ACDEG-4 USAF ATTN: J. R. Loftis, Jr. 65

Institute of Science and Technology The University of Michigan Armour Research Foundation Atomic Energy Establishment Observatoire de Geophysique Illinois Institute of Technology Apollo Pier Road College Jean-de-Brebeuf Technology Center Bombay 1, India 3200 Chemin Ste-Catherine Chicago 16, Illinois ATTN: Dr. H. M. Iyer Montreal 26, Canada ATTN: Mr. Andrew Unger ATTN: Prof. M. Buist S J VIA: Office of the Secretary of Defense Arms Control and Disarmament Agency Advanced Research Projects Agency Director Science and Technology Bureau The Pentagon VIA: Office of the Secretary of Defense iRoom 5486, State Department Washington 25, D. C. Advanced Research Projects Agency Washington 25, D. C. ATTN: J. R. Loftis he Pentagon Washington 25 D. C ATTN: Dr. H. R. Meyers Roland F. Beers, Inc. ATTN: J. R. Loftis, Jr. Roland F. Beers, Inc. Army P. 0. Box 23 Signal Research and Development Laboratory, U. S. Alexandria, Virginia Professor Dr. Hans Closs Fort Monmouth, New Jersey Bundesanstalt fur Bodenforschung ATTN: Roland F. Beers Wiesenstrasse 1, Hannover 3 ATTN: Commanding Officer West Germany West Germany Army, Department of the Bell Telephone Laboratories S e Ata Office of the Chief Murray Hill, New Jersey VIA S. Emassy Research and Development ATTN: Dr. Bruce P. Bogert Bonn, Germany OCS Washington 25, D. C. VIA: Office of the Secretary of Defense Washington 25, D.~ C. Bell Telephone Laboratories A ceReearch Projects Agenc ATTN: Brig. Gen. David C. Lewis Room IA-218 The Pentagon Murray Hill, New Jersey Washington 25, D. C. Washington 25, D. C. Army Engineer Research and Development ATTN: Dr. John W. Tukey Laboratories, U. S. Laboratories, U. S. " ~ '" -~ ATTN: J. R. Loftis, Jr. Fort Belvoir, Virginia Bell Telephone Laboratories ATTN: Mine Detection Branch Whippany, New Jersey Bureau of Mines, U.LS. Applied Physics Research Laboratory ATTN: Dr. R. A. Walker College Park, Maryland Army Research Office CM Duke Station ATTN: Dr. Leonard Obert Durham, North Carolina Bell Telephone Laboratories Western Electric Company Bureau of Ships, U. S. Atlantic Refining Company Whippany, New Jersey Department of the Navy 4500 West Mockingbird Lane ATTN: C. F. Weibush Washington 25, D. C. Dallas, Texas ATTN: Code 689D ATTN: Helen McKenzie Dr. F. G. Blake, Jr. Librarian Bureau of Ships, U. S. c/o Mrs. F. G. Blake Code 689D Atomic Energy Commission, U. S. 226 Lawrence Street Washington 25, D. C. Albuquerque Operations Office New Haven 11, Conn. A M.. P. 0. Box 5400 ATTN: Mr. A. H. Sobel P. O. Box 5400 Albuquerque, New Mexico Blue Mountains Seismological Observatory California Institute of Technology ATTN: Mr. R. TT M-vler The Geotechnical Corporation California Institute of Technology ATTN: Mr. R. E. Miller The Geotechnical Corporation Seismological Laboratory Baker, Oregon 220 North San Rafael Avenue Atomic Energy Commission, U. S. ATTN: Station Manager Pasadena, California Director Division of Military Applications ATTN: Dr. Hugo Benioff Group Captain John Rowlands, RAF Washington 25, D. C. (2) Washington 25, D. C. (2) Atomic Coordinating Office ATTN: Brig. Gen. Austin W. Betts Room 427 California Institute of Technology British Defence Staffs Division of Engineering Atomic Energy Commission, U. S. British Embassy Pasadena, California Division of Military Applications 3100 Massachusetts Avenue, N. W. ATTN: Dr. George W. Housner Chairman Washington 8, D. C. (5) Washington 25, D. C. California Institute of Technology VIA: Office of the Secretary of Defense, ARPA Seismological Laboratory ATTN: Mr. Don Gale The pentao-n Seismological Laboratory ATTN: Mr. Don Gale The Pentagon 220 North San Rafael Avenue Washington~25~ D. C. 220 North San Rafael Avenue Atomic Energy Commission,.. Pasadena, California Division of Peaceful Nuclear Explosions ATTN: J. R. Loftis, Jr. ATTN: Dr. Frank Press Chairman Washington 25, D. C. British Petroleum Company Ltd. Jet Propulsion Laboratories ATTN: Mr. John Kelly Exploration Division California Institute of Technology BP Research Center 4800 Oak Grove Drive Atomic Energy Commission Chertsey Road Pasadena California General Managers Office Sunbury-on-Thames, Middlesex, England Germantown, Maryland (2) ATTN: Mr. Raoul Choate ATTN: Manager ATTN: Dr. S. G. English VIA: Office of the Secretary of Defense California Institute of Technology Atomic Energy Commission Advanced Research Projects Agency Seismological Laboratory Technical Information Service Extension The Pentagon 220 N. San Rafael Avenue Oakridge, Tennessee (2) Washington 25, D. C. Pasadena, California ATTN: Hugh Voress ATTN: J. R. Loftis, Jr. ATTN: Dr. L. Knopoff 66

Institute of Science and Technology The University of Michigan California Research Corporation Carnegie Institution of Washington Cornell University 200 Bush Street Department of Terrestrial Magnetism Research and Advanced Studies San Francisco, California 5241 Broadbranch Road, N. W. 303 Day Hall ATTN: Mr. R. S. Faull Washington 15, D. C. Ithaca, New York ATTN: Dr. John J. Steinhart ATTN: Dr. Frank A. Long California Research Laboratory Vice President P. 0. Box 446 Century Geophysical Corporation La Habra, California P. O. Box C Cumberland Plateau Seismological Observatory Admiral Station McMinnville, Tennessee ATTN: Dr. P. Edward Byerly, Jr. Tulsa 15 Oklahoma ATTN: Station Manager ATTN: R. A. Broding California, University of A Cawrinc Radiation Latyoratorf Data Analysis and Technique Development Center Lawrence Radiation La2oratory Century Geophysical Corporation P. O. Box 334 Livermore, California (2) P. O. Box C Alexandria, Virginia (6) ATTN: Dr. Roland F. Herbst Admiral Station Dr. Charles E. Violet Tulsa 15, Oklahoma ATTN: Mr. F. B. Coker ATTN: Daniel Hearn Defense Atomic Support Agency California, University of Department of Defense Lawrence Radiation Laboratory Century Geophysical Corporation Chief Livermore, California P. 0. Box C Washington 25, D. C. Admiral Station ATTN: Dr. Kenneth M. Watson Tulsa 15 Oklahoma ATTN: Major J. Dickson Tulsa 15, Oklahoma ATTN: J. E. A. T. Woodburn Defense Atomic Support Agency California, University of Field Command Seismograph Station Commerce, Department of Sandia Base Berkeley 4, California Coast & Geodetic Survey, U. S. Albuquerque, New Mexico Geophysics Division ATTN: Prof. P. Byerly Washington 25 Div n ATTN: Lt. Col. Conrad R. Peterson CalifATTN: Dr. Dean S. Carder Defense Atomic Support Agency California, University of Lawrence Radiation Laboratory Headquarters Technical Information Division Commerce, Department of Washington 25, D. C. P. 0. Box 808 Coast & Geodetic Survey, U. S. ATTN: John G. Lewis Livermore, California Geophysics Division ATTN: Clovis G. Craig - Washington 25, D. C. Defense Atomic Support Agency ATTN: Mr. J. Jordan Weapons Effects and Tests Headquarters California, University of e Commerce, Department of Field Command La Jolla Laboratories Coast & Geodetic Survey, U. S. Sandia Base La Jolla, California Geophysics Division Albuquerque, New Mexico ATTN: Dr. Walter Munk Washington 25, D. C. ATTN: FCWT ATTN: Mr. Leonard Murphy VELA UNIFORM Programs Officer California, University of Seismological Laboratory Commerce, Department of Defense, Department of Berkeley 4, California Coast & Geodetic Survey, U. S. Office of the Assistant Secretary of Defense Camp Mercury, Nevada for International Security Affairs ATTN: Dr. D. Tocher ATTN: T. H. Pearce Washington, D. C. ATTN: Cdr. E. C. Cline Cambridge University Commerce, Department of Room 4E829 Department of Geodesy & Geophysics ologal er Seismological Data Center Madingley Rise, Madingley Road U. S. Coast and Geodetic Survey Defense, Department of Cambridge, England 25 D (2) Office of the Assistant to the Secretary of ATTN: Sir Edward Bollard AT:DrcoDefense of Atomic Energy Dr. Maurice Hill ATTN: Director Washington 25, D. C. Dr. Maurice Hill VIA: Office of the Secretary of Defense Commerce, Department of ATTN: Dr. Gerald Johnston VIA: Office of the Secretary of Defense Coast & Geodetic Survey, U. S. Advanced Research Projects Agency Geodetic S, Advanced Research Projects Agency Seismological Laboratory Defense Documentation Center The Pentagon Sandia Base Cameron Station Washington 25, D. C. Sandia Base Albuquerque, New Mexico Alexandria, Virginia (20) ATTN: J. R. Loftis, Jr. ATTN: Technical Director Defense, Department of Office of the Director of Defense Canadian Defese Research Sf Commerce, Department of Research & Engineering W2450 Massachusetts Aenue, N. W. National Bureau of Standards, U. S. The Pentagon Washington 8, D. C. (5) Washington 25, D. C. Washington 25, D. C. ATTN: Dr. C. E. Hubley ATTN: Mr. Harry Matheson ATTN: Director, Office of Atomic Programs VIA: Office of the Secretary of Defense Advanced Research Projects Agency Continental Oil Company Director of Research and Development The Pentagon P. O. Drawer 293 Office of the Secretary of the Army Washington 25, D. C. Ponca City, Oklahoma Washington 25, D. C. ATTN: Mr. J. R. Loftis, Jr. ATTN: Dr. John M. Crawford ATTN: Lt. Col. William D. Sydnor 67

Institute of Science and Technology The University of Michigan Dresser Electronics European Research Office (9851 DU) Hughes Aircraft Company SIE Division U. S. Department of the Army Communications Division P. 0. Box 22187 Frankfurt/Main, Germany P. 0. Box 90902 Houston 27, Texas (2) APO 757, U. S. Forces (5) Airport Station ATTN: Mrs. Mildred Beaman VIA: Office of the Secretary of Defense Los Angeles 45, California Librarian Advanced Research Projects Agency Humle Oil and Research Las. The Pentagon Th~e ~P~entagon ~Geophysics Department Dresser Electronics Washington 25, D. CP.Geo. Box 2180 SIE Division ATTN: J. R. Loftis, Jr. Houston 1, Texas P. 0. Box 22187 Houston 27, Texas General Atronics Corporation ATTN: Dr. Lynn Howell ATTN: Mr. H d A. Bd One Balla Avenue ATTN: Mr. How ard A. Bond Balla-Cynwyd Pennsylvania Hunting Geophysical Services, Inc. Vice-Pres., Military Engineering v 10 Rockefeller Plaza ATTN: Mr. Joseph T. Underwood, III New York 20, N. Y. Dunbar Laboratory General Dynamics Corporation ATTN: Mr. R. H. Stebbins, Gen. Mgr. Oxford Street Electric Boat Division linois University of Cambridge 28, Massachusetts Groton, Connecticut Department of Mining ATTN: Dr. H. Pollock ATTN: Mr. J. V. Harrington Urbana, Illinois General Electric Company ATTN: Dr. Adrian Scheidegger Dupont Company 735 State Street Interior, Department of Eastern Laboratory Santa Barbara, California Bureau of Mines, U. S. Gibbstown, New Jersey ATTN: Mr Finn Dyolf Bronner Washington 25, D. C. (5) ATTN: Mr. A. B. Andrews Defense Electronics Division ATTN: James Hill Explosives Department Geophysical and Polar Research Center Interior, Department of Edgerton, Germeshausen, and Grier, Inc. 6021 South Highland Road Branch of Crustal Studies Director of Research Madison 6, Wisconsin U. S. Geological Survey 160 Brookline Drive ATTN: Prof. George Woollard 7580 W. 1th Street Boston, Massachusetts Director Lakewood 15, Colorado ATTN: Clyde Dobbie Geophysical Service, Inc. ATTN: Dr. George Keller Geophysical Service, Inc. P. O. Box 35084 Interior, Department of Edgerton, Germeshausen, and Grier, Inc. Dallas 35, Texas Branch of Crustal Studies 160 Brookline Drive SaU. S. Geological Survey Boston, Massachusetts r.uner7580 W. 16th Street ATTN: Dr. Raymond C. O'Rourke Georgia Institute of Technology Lakewood 15, Colorado (5) Engineering Experiment Station ATTN- Louis C. Pakiser Atlanta, Georgia Edgerton, Germeshausen, and Grier, Inc. Interior, Department of 160 Brookline Avenue ATTN: Mr. Frederick Bellinger, Chief Geolo Surve Materials Sciences Division Geological Survey Boston, Massachusetts DivisioDenver Federal Center ATTN: F. T. Strabala Georgia Institute of Technology Denver 25, Colorado Engineering Experimental Station ATTN: Dr. H. Roach Electro-Mechanics Company Atlanta, Georgia P. O0. Box 802 AT M.JhE.HsInterior, Department of PA Box 802 TATTN: Mr. John E. Husted Office of the Science Advisor to the Secretary Austin 64, Texas Research Scientist, Head Washington 25, D. C. ATTN: Dr. Fred H. Morris Geotechnical Corporation ATTN: Dr. Calhoun WMSO Engineer Research and Development Laboratory P. 0. Box 2071 Interior, Department of the Mines Detection Branch Lawton, Oklahoma (2) U. S. Geological Survey Fort Belvoir, Virginia Room 4215, GSA Building ATTN: Station Director Washington 25, D. C. (2) ATTN: Mr. S. E. Dwornik Geotechnical Corporation ATTN: Dr. William Thurston Engineering-Physics Company Dr. Wayne Hall 5515 Randolph Street.. Box 28277 Interior, Department of Rockville, Maryland Dallas 28, Texas (19) Geological Survey ATTN: Mr. Vincent Cushing ATTN: Dr. W. Heroy, Jr. Branch of Military Geology Washington 25, D. C. (3) Harvard University Harvard University ATTN: Mrs. Mildred P. White The Superintendent Department of Geology The Observatory Cambridge, Massachusetts Itek Laboratories Eskdalemuir ATTN: Prof. L. Don Leet Lexington 73, Massachusetts Langholm, Dumfriesshire, Scotland A NProf. Lee ATTN: Mr. Robert Fleming VIA: Office of the Secretary of Defense Hawaii, University of Advanced Research Projects Agency Hawaii Institute of Geophysics Jersey Production Research Company The Pentagon Honolulu 14, Hawaii 1133 North Lewis Avenue Washington 25, D. C. ATTN: Dr. Doak Cox Tulsa 10, Oklahoma ATTN: J. R. Loftis Department of Earth Sciences ATTN: Mr. P. Jacobson 68

Institute of Science and Technology The University of Michigan John Carroll University Massachusetts Institute of Technology National Bureau of Standards Cleveland 18, Ohio Department of Electrical Engineering Boulder Laboratories ATTN: Rev. H. F. Birkenhaer, Cambridge, Massachusetts Boulder, Colorado ATTN: Rev. H. F. Birkenhauer, S. J. ATTN: Prof. J. Ruina ATTN: Mrs. Victoria S. Baker, Librarian John Carroll University Cleveland 18, Ohio Massachusetts Institute of Technology National Engineering Science Company ATTN: Dr. Edwad W r Lincoln Laboratory 1711 South Fair Oaks Avenue Lexington 73, Massachusetts (2) Pasadena, California Jordskelstasjonen ATTN: Dr. Paul Green ATTN: Dr. N. D. Baratynski University of Bergen Bergen, Norway National Engineering Science Company Massachusetts Institute of Technology 1711 South Fair Oaks VIA: Office of the Economic Counselor Lincoln Laboratory oun ai Oa U. S. Embassy Lexington 73, Massachusetts Pasadena, alornia Drammesveien, Oslo, Norway ATTN: Lirarian ATTN: Dr. J. F. Hook ATTN: Librarian VIA: Office of the Secretary of Defense Advanced Research Projects Agency Massachusetts Institute of Technology National Science Foundation The Pentagon Department of Earth Sciences 1951 Constitution Avenue, N. W. Washington 25, D. C. Cambridge 39, Massachusetts Washington, D. C. ATTN: J. R. Loftis, Jr. ATTN: Prof. S. M. Simpson ATTN: Dr. Roy E. Hanson LaCoste and Romberg National Science Foundation Austin, Texas Melpar, Inc. Office of Antarctic Programs Technical Information Center uice oi Antarctic programs ATTN: Mr. Lucian J. B. LaCoste 3000 Technical Information Bouleva rd 1951 Constitution Avenue, N. W. ATTN: Mr. Lucian J. B. LaCoste 3000 Washington 25, D. C. Falls Church Virginia Washington 25, D. C. Lamont Geological Observatory Falls Chu Virginia Columbia University ATTN: P. D.Vachon Information Officer Palisades, New Yorkrector ATTN: Dr. Maurice Ewing Melpar, Inc.National Security Agency Applied Science Division Fort George G. Meade Maryland Lamont Geological Observatory 11 Galen Street G. Meade, Maryland Columbia University Watertown 72, Massachusetts ATTN: C3/TDL (Room 2C087) Palisades, New York ATTN: Mrs. Lorraine Nazzaro Miss Jean Creswell ATTN: Dr. K. L. Hunkins Librarian U. S. Naval Radiological Defense Laboratory amo n t Geolial Observatry Meteorological Office (M. O. 14) San Francisco 24, California Palisades New York (2) London Road, Bracknell ATTN: Mr. E. R. Bowman Berkshire, England Head, Library Branch ATTN: The Director-General ATTN:~ Libraryt~ A~ Thei~ Director-General~~ ^U.S. Naval Radiological Defense Laboratory Lamont Geological Observatory VIA: Office of the Secretary of Defense San Francisco 24, California Columbia University Advanced Research Projects Agency Palisades, New York The Pentagon ATTN: Mr. Ken Sinclair ATTN: Dr. Jack E. Oliver Washington 25, D. C. Naval Research Laboratory Lebel, Jean ATTN: J. R. Loftis Washington 25, D. C. CLES CLES ATTN: The Librarian 6 Avenue Daniel Leseur Minnesota, University of Paris 7a, France Institute of Technology VIA: Oice of te Secr y of D e School of Mines and Metallurgy Navy Department VIA: Office of the Secretary of Defense Minneapolis 14 Minnesota Bureau of Ships Advanced Research Projects Agency Minneapolis 14, Minnesota Bureau of Ships Advanced Research Projects Agency Washington 25 D C The Pentagon ATTN: Prof. Harold M. Mooney Washington 25, D. C. Geophysics ATTN: Code 362 B ATTN: J. R. Loftis Minneapolis Honeywell Company Navy, Department of the Heiland Division Office of the Deputy Chief of Naval Operations Los Alamos Scientific LaboratoryDenver Colorado (Development) Los Alamos, New Mexico Washington 25, D. C. ATTN: Mr. John Paul ATTN: Dr. Conrad L. Longmire Director of Magnetic Tape Division ATTN: Mr. W. Magnitizky Op-07T Mandrel Industries, Inc. Electro-Technical Labs Division The MITRE Corporation 5234 Glenmont Drive LabsDivisionDept. D-17 Navy, Director of the 5234 Glenmont Drie Box 208 Office of the Deputy Chief of Naval Operations P. stox 36,3060 xBedford, Massachusetts (Development) Houston 36, Texas e a ATTN: Dr. R. N. Ostreim ATTN: Dr. John C. Morganstein Washington 25, D. C. Mandrel Industries, Inc. MOHOLE Project Navy, Department of Research Division National Academy of Sciences Office of Naval Research 1025 S. Shepard 2101 Constitution Avenue, N. W. Washington 25, D. C. (5) Houston 19, Texas Washington 25, D. C. ATTN: James W. Winchester ATTN: Dr. Lewis M. Mott-Smith ATTN: Executive Secretary Code 418 69

Institute of Science and Technology The University of Michigan Navy Electronics Laboratory, U. S. PETTY Geophysical Engineering Co. Rice University Director P. 0. Drawer 2061 Houston 1, Texas San Diego 52, California San Antonio 6, Texas ATTN: Prof. J. Cl. DeBremaecker ATTN: Code 2350 ATTN: J. O. Parr, Jr. Dept. of Geology Navy Electronics Laboratory, U. S. Phillips Petroleum Company Sandia Corporation Point Loma Laboratory Bartelsville, Oklahoma Division 7000 San Diego, California Sandia Base ATTN: Mr. Harold L. Mendenhall Albuquerque New Mexico ATTN: Charles Johnson District Geophysicist ATTN: Mr. G. A. Fowler Director Planetary Sciences, Inc. Navy Electronics Laboratory 501 Washington Street Sandia Corporation San Diego 52, California P. O. Box 561 Sandia Base ATTN: Mr. T. McMillian Santa Clara, California Albuquerque, New Mexico ATTN: Dr. William Adams ATTN: William R. Perret Navy Oceanographic Office, U. S. The Hydrographer Prengle, Dukler, and Crump, Inc. Saskatchewan, University of Washington 25, D. C. 5417 Crawford Street Saskatoon Houston 4, Texas Saskatchewan, Canada Navy Radiological Defense Lab., U. S. Saskatchewan, Canada Director ATTN: Dr. Clark Goodman ATTN: Dr. James Mawdsley San Francisco 24, California Princeton University VIA: Office of the Secretary of Defense ATTN: Dr. Eugene Cooper Physics Department Advanced Research Projects Agency Princeton, New Jersey The Pentagon Navy Radiological Defense Laboratory Printon nton Hunters Point ATTN: Dr. Val L. Fitch Washington 25, D. C. Building 815 ATTN: J. R. Loftis, Jr. San Francisco, California Pure Oil Company ATTN: Dr. L. Gevantman Research Laboratories Seismograph Service Corporation Crystal Lake, Illinois P. O. Box 1590 New Mexico, University of ATTN: Dr. Ira Cram, Jr. Tulsa, Oklahoma Albuquerque, New Mexico Radio Corporation of America ATTN: Dr. James E. Hawkins ATTN: Dr. H. L. Walter David Sarnoff Research Center Director of Research Princeton New Jersey Seismological Laboratory Uppsala, Sweden North American Aviation (S+ID) ATTN: Dr. D. S. McCoy ATTN: Dr. Markus B&th 12214 Lakewood Boulevard Downey, California Rand Corporation, The VIA: Office of the Secretary of Defense 1700 Main Street Advanced Research Projects Agency ATTN: Mr. Robert Fowler Santa Monica, California The Pentagon Dept. 197-330 ATTN: Dr. Richard Latter Washington 25, D. C. ATTN: J. R. Loftis, Jr. Nortronics Raytheon Company Geophysics Research Group Missile and Systems Division 77 A Street Bedford, Massachusetts Shell Development Lab. Bedford, Massachusetts Needham Heights 94, Massachusetts P. 0. Box 481 ATTN: Mr. C. C. Abt Houston 1, Texas ATTN: Dr. Paul R. Miles Rensselaer Polytechnic Institute ATTN: Dr. Sidney Kaufman The Ohio State University Research Foundation Troy, New York 1314 Kinnear Road Sinclair Research Inc. Columbus 12, Ohio ATTN: Mr. R. B. Finch P. O. Box 3006, Whittier Station Director of Researchlsa Oklahoma ATTN: Dr. Robert C..Stephenson, Executive Director Rensselaer Polytechnic Institute ATTN: Mr. John Bemrose Troy, New York Oklahoma, The University of South Carolina, University of Research Institute ATTN: Dr. Samuel Katz Columbia, South Carolina Norman, Oklahoma Research Institute of National Defense ATTN: Professor 0. Scheutte ATTN: Dr. Norman Ricker Stockholm 80, Sweden Head, Dept. of Physics Oregon State University ATTN: Dr. Ulf A. EricssonSouthern Methoist University Department of Oceanography VIA: Office of the Science Attache Department of teology Corvallis, Oregon U. S. Embassy Dallas Texas ATTN: Dr. P. Dehlinger Stockolm, SwedenATTN Dr. Euge T. Herrin ATTN: Dr. Eug ne T. Herrin VIA: Office of the Secretary of Defense Oregon State University Advanced Research Projects Agency Southwest Researh Institute Department of Oceanography The Pentagon 8500 Culebra Road Corvallis, Oregon Washington 25, D. C. San Antonio 5, Texas ATTN: Dr. Joe Berg ATTN: J. R. Loftis, Jr. Space-General Corporation Pennsylvania State University University of Rhode Island 777 Flower Street University Park, Pennsylvania Kingston, Rhode Island Glendale 1, California ATTN: Dr. B. F. Howell, Jr. ATTN: Professor Kane ATTN: Dr. Glenn Brown 70

Institute of Science and Technology The University of Michigan Space Technology Laboratories, Inc. St. Louis University Toronto, University of One Space Park The Institute of Technology Department of Physics Redondo Beach, California 3621 Olive Street Toronto 5, Canada ATTN: Mr. R. Whitmer St. Louis 8, Missouri ATTN: Prof. J. T. Wilson ATTN: Dr. R. R. Heinrich ATTN: Dr. R. R. Heinrich VIA: Office of the Secretary of Defense Special Assistant to the President for Science and Technology St. Louis University Advanced Research Projects Agency Science and Technology St. Louis University T Office of the Executive Office Building Department of Geophysics The Pntagon The White House 3621 Olive Street Washington 25, D. C. Washington 25, D. C. St. Louis 8, Missouri ATTN: J. R. Lftis, Jr. ATTN: Mr. Spurgeon M. Keeny, Jr. ATTN: Dr. William Stauder, S. J. Texas Instruments, Inc. Sperry Rand Research Center Strategic Air Command P. 0. Box 35084 Sudbury, Massachusetts (OAWS) Dallas 35, Texas (7) ATTN: Mr. Alan Steeves Headquarters ATTN R. A. Arnett Librarian Offutt Air Force Base, Nebraska (2) W. F. Sprengnether Instrument Co., Inc. Sydney, University of Uinta Basin Seismological Observatory 4567 Swan Avenue Department of Mathematics Vernal, Utah St. Louis 10, Missouri Sydney, Australia ATTN: Station Manager ATTN: Mr. R. F. Hautly ATTN: Dr. Keith Bullen Stanford Research Institute VIA: Office of the Secretary of Defense United Earth Sciences Division Palo Alto, California Advanced Research Projects Agency P. O. Box 334 The Pentagon Alexandria, Virginia ~ATTN: Dr. Allan Peterson Washington 25, D. C. ATTN: E. A. Flinn Stanford Research Institute ATTN: J. R. Loftis, Jr. Menlo Park, California United Earth Sciences Division Technical Operations, Inc. P. O Bo 334 ATTN: Dr. Thomas C. Poulter Central Research Office Alex, Director Poulter Laboratories Alexandria, Virginia Stanford Research Institute Burlington, Massachusetts ATTN: Mr. J. Griffin Menlo Park, California ATTN: Mr. C. Fain ATTN: L. M. Swift United Earth Sciences Division ATTN: L. M. Swift Tenneco Oil and Gas Co. P. O. Box 334 Stanford University P. 0. Box 18 Alexandria, Virginia High Energy Physics Laboratory Houston 1, Texas ATTN: Ted Winston Stanford, California ATTN: Lynn Ervin ATTN: Dr. Wolfgang K. H. Panofsky United ElectroDynamics, Inc. A N Dr. WolfgTexaco, Inc. 200 Allendale Road Stanford Research Institute Research and Technical Department Pasadena, California (2) Building 108 P.. Box 509 ATTN: Mrs. Dorothy A. Al Menlo Park, California Beacon, New York ATTN: Dr. Robert B. Vaile, Jr. ATTN: Dr. L. C. Roess United Kingdom Atomic Energy Authority Director, Physics Division Graduate Research Center Blacknest, Brimpton, Near Reading Stanford Research Institute P. O. Box 8478 Berkshire, England Menlo Park, California Dallas 5, Texas ATTN: Dr. Eric Carpenter ATTN: Mr. E. C. Wood ATTN: Dr. Anton L. Hales VIA: Office of the Secretary of Defense Manager, Geophysics Dept. Advanced Research Projects Agency Gravity Meter Exploration Co. The Pentagon State, Dept. of 3621 W. Alabama Avenue Washington 25, D. C. Arms Control and Disarmament Agency Houston 27, Texas ATTN J Loftis ATTN: J. R. Loftis, Jr. Washington 25, D. C. (2) ATTN: Dr. Nelson Steenland ATTN: Bureau of International Relations Gulf Research and Development Co. Virgnia Polytechnic Institute State, Dept. of Geophysical Research Division Department of Geology Arms Control and Disarmament Agency P. 0. Box 2038 Blacksburg, Virginia Washington 25, D. C. Pittsburgh 30, Pennsylvania ATTN: Professor Charles Sears ATTN: Research Reference Staff ATTN: Dr. T. J. O'Donnell Director University of Utah State, Department of the Department of Geophysics Office of International Scientific Affairs Texas, University of Mines Building Washington 25, D. C. Austin 12, Texas Salt Lake City, Utah ATTN: Dr. A. Rollefson ATTN: Prof. W. T. Muehlberger ATTN: Professor Ken Cook Dept. of Geology St. Louis University University of Wisconsin The Institute of Technology Tonto Forest Seismological Observatory Geophysical and Polar Research Center 3621 Olive Street P. O. Box 337 6021 South Highland Road St. Louis 8, Missouri Payson, Arizona Madison 6, Wisconsin ATTN: Dr. Carl Kisslinger ATTN: Mr. Allen M. Rugg, Jr. ATTN: Director 71

Institute of Science and Technology The University of Michigan University of Witwatersrand Waterways Experiment Station Western Geophysical Corporation Bernard Price Institute of Geophysical Research Jackson, Mississippi 933 N. LaBrea Street Johannesburg, South Africa ATTN: Mr. James Platty Los Angeles 38, California ATTN: Director ATTN: Mr. Carl H. Savit VIA: Office of the Secretary of Defense Advanced Research Projects Agency Waterways Experiment Station, U. S. Weston Observatory The Pentagon Vicksburg, Mississippi Weston 93, Massachusetts Washington 25, D. C. ATTN: Librarian ATTN: Rev. Daniel Linehan, S.J. ATTN: J. R. Loftis, Jr.avier University Xavier University U. S. Geological Survey Weizmann Institute of Science Cincinnatti, Ohio U. S. Geological Survey 345 Middlefield Road ot O e x 2 ATTN: Rev. E. A. Bradley, S. J. Menlo Park, California Rehovoth, Israel ATTN: Librarian ATTN: Prof. C. L. Pekeris Assistant Chief Geologist for Engineering Geology VIA: Office of the Secretary of Defense U. S. Geological Survey Advanced Research Projects Agency GSA Building The Pentagon Washington 25, D. C. Washington 25, D. C. ATTN: V. R. Wilmarth ATTN: J. R. Loftis, Jr. 72

+ -F - AD Div 2/10 UNCLASSIFIED AD Div 2/10 UNCLASSIFIED Inst. of Science and Technology, U. of Mich., Ann Arbor I. VESIAC Inst. of Science and Technology, U. of Mich., Ann Arbor I. VESIAC THE REQUIREMENTS OF A HIGH-SENSITIVITY SEIS- II. Carder, Dean S. THE REQUIREMENTS OF A HIGH-SENSITIVITY SEIS- II. CarderDeanS. MOGRAPH STATION by Dean S. Carder. A VESIAC III. Advanced Research MOGRAPH STATION by Dean S. Carder. A VESIAC III. Advanced Research State-of-the-Art Report. Oct. 63. 64 p. incl. tables, Projects Agency State-of-the-Art Report. Oct. 63. 64 p. incl. tables, Projects Agency 36 refs. IV. Contract SD-78 36 refs. IV. Contract SD-78 (Rept. 4410-63-X) (Rept. 4410-63-X) (Contract SD-78) Unclassified Report (Contract SD-78) Unclassified Report Defense Defense Documentation Center Documentation Center UNCLASSIFIED UNCLASSIFIED + + + AD Div 2/10 UNCLASSIFIED AD Div 2/10 UNCLASSIFIED Inst. of Science and Technology, U. of Mich., Ann Arbor I. VESIAC Inst. of Science and Technology, U. of Mich., Ann Arbor I. VESIAC THE REQUIREMENTS OF A HIGH-SENSITIVITY SEIS- II. Carder, Dean S. THE REQUIREMENTS OF A HIGH-SENSITIVITY SEIS- II. CarderDeanS. MOGRAPH STATION by Dean S. Carder. A VESIAC III. Advanced Research MOGRAPH STATION by Dean S. Carder. A VESIAC III. Advanced Research State-of-the-Art Report. Oct. 63. 64 p. incl. tables, Projects Agency State-of-the-Art Report. Oct. 63. 64 p. incl. tables, Projects Agency 36 refs. IV. Contract SD-78 36 refs. IV. Contract SD-78 (Rept. 4410-63-X) (Rept. 4410-63-X) (Contract SD-78) Unclassified Report (Contract SD-78) Unclassified Report Defense Defense Documentation Center Documentation Center UNCLASSIFIED UNCLASSIFIED 4-+ +

AD UNCLASSIFIED AD UNCLASSIFIED DESCRIPTORS DESCRIPTORS Seismological stations Seismological stations Earthquakes Earthquakes Nuclear explosions Nuclear explosions Microseisms Microseisms Seismographs Seismographs Recording systems Recording systems Seismic waves Seismic waves UNCLASSIFIED UNCLASSIFIED + AD UNCLASSIFIED AD UNCLASSIFIED DESCRIPTORS DESCRIPTORS Seismological stations Seismological stations Earthquakes Earthquakes Nuclear explosions Nuclear explosions Microseisms Microseisms Seismographs Seismographs Recording systems Recording systems Seismic waves Seismic waves UNCLASSIFIED UNCLASSIFIED

+ + -F AD Div 2/10 UNCLASSIFIED AD Div 2/10 UNCLASSIFIED Inst. of Science and Technology, U. of Mich., Ann Arbor I. VESIAC Inst. of Science and Technology, U. of Mich., Ann Arbor I. VESIAC THE REQUIREMENTS OF A HIGH-SENSITIVITY SEIS- II. Carder, Dean S. THE REQUIREMENTS OF A HIGH-SENSITIVITY SEIS- II. Carder, Dean S. MOGRAPH STATION by Dean S. Carder. A VESIAC III. Advanced Research MOGRAPH STATION by Dean S. Carder. A VESIAC III. Advanced Research State-of-the-Art Report. Oct. 63. 64 p. incl. tables, Projects Agency State-of-the-Art Report. Oct. 63. 64 p. incl. tables, Projects Agency 36 refs. IV. Contract SD-78 36 refs. IV. Contract SD-78 (Rept. 4410-63-X) (Rept. 4410-63-X) (Contract SD-78) Unclassified Report (Contract SD-78) Unclassified Report Defense Defense Documentation Center Documentation Center UNCLASSIFIED UNCLASSIFIED + + + AD Div 2/10 UNCLASSIFIED AD Div 2/10 UNCLASSIFIED Inst. of Science and Technology, U. of Mich., Ann Arbor I. VESIAC Inst. of Science and Technology, U. of Mich., Ann Arbor I. VESIAC THE REQUIREMENTS OF A HIGH-SENSITIVITY SEIS- II. Carder, Dean S. THE REQUIREMENTS OF A HIGH-SENSITIVITY SEIS- II. Carder, Dean S. MOGRAPH STATION by Dean S. Carder. A VESIAC III. Advanced Research MOGRAPH STATION by Dean S. Carder. A VESIAC III. Advanced Research State-of-the-Art Report. Oct. 63. 64 p. incl. tables, Projects Agency State-of-the-Art Report. Oct. 63. 64 p. incl. tables, Projects Agency 36 refs. IV. Contract SD-78 36 refs. IV. Contract SD-78 (Rept. 4410-63-X) (Rept. 4410-63-X) (Contract SD-78) Unclassified Report (Contract SD-78) Unclassified Report Defense Defense Documentation Center Documentation Center UNCLASSIFIED UNCLASSIFIED + +h +

AD UNCLASSIFIED AD UNCLASSIFIED DESCRIPTORS DESCRIPTORS Seismological stations Seismological stations Earthquakes Earthquakes Nuclear explosions Nuclear explosions Microseisms Microseisms Seismographs Seismographs Recording systems Recording systems Seismic waves Seismic waves 0-c UNCLASSIFIED UNCLASSIFIED 01- r\) a AD UNCLASSIFIED AD UNCLASSIFIED DESCRIPTORS DESCRIPTORS (o Seismological stations Seismological stations Earthquakes Earthquakes Nuclear explosions Nuclear explosions Microseisms Microseisms Seismographs Seismographs Recording systems Recording systems Seismic waves Seismic waves UNCLASSIFIED UNCLASSIFIED