Geographical Coordinates and Locations of Various Islands and Regions, Study notes of Statistics

Download Geographical Coordinates and Locations of Various Islands and Regions and more Study notes Statistics in PDF only on Docsity!

UNCLASSIFIED

AD 4 5 45 7 2

DEFENSE DOCUMENTATION CENTER

FOR

SCIENTIFIC AND TECHNICAL INFORMATION

CAMERON STATION ALEXANDRIA. VIRGINIA

UNCLASSIFIED

NOTICE: When govemnent or other drawings, speci- fications or other data are used for any purpose other than In connection with a definitely related government procurement operation, the U. S. Government thereby Incurs no responsibility, nor any obligation whatsoever; and the fact that the Govern- ment may have formulated, furnished, or in any way supplied the said drawings, specifications, or other data is not to be regarded by implication or other- wise as in any manner licensing the holder or any other person or corporation, or conveying any rights or permission to manufacture, use or sell any patented invention that may in any way be related thereto.

>

Request for additional copies by Agencies of the Department of Defense, their contractors and other government agencies should be directed to the:

DEFENSE DOCUMENTATION CENTER (DDC) CAMERON STATION ALEXANDRIA, VIRGINIA 22314

Department of Defense Contractors must be established for DDC services or have their "need-to-know" certified by the cognizant military agency of their project or contract.

All other persons and organizations should apply to the:

U. S. DEPARTMENT OF COMMERCE OFFICE OF TECHNICAL SERVICES WASHINGTON 25, D. C. 20230

►

¥

AFCRL-64-

WORLDWIDE COLLECTION AND EVALUATION

OF EARTHQUAKE DATA

EVALUATION OF 1963 SEISMICITY

By

Ray Fisher, Ralph Guidroz and Research Staff

TEXAS INSTRUMENTS INCORPORATED

P. O. Box 5621 Dallas, Texas

Contract No. AF 19(604)- Date of Contract: 15 May 1961 Contract Expiration Date: 15 November 1964 Project 8652 Task 865207

FINAL REPORT

18 November 1964

Prepared For

AIR FORCE CAMBRIDGE RESEARCH LABORATORIES

OFFICE OF AEROSPACE RESEARCH

UNITED STATES AIR FORCE

BEDFORD, MASSACHUSETTS

WORK SPONSORED BY ADVANCED RESEARCH PROJECTS AGENCY

PROJECT VELA UNIFORM

ARPA Order No. 292, Amendment No. 7, dated 25 March 1963 Project Code No. 3810, Task 2

SYNOPSIS

Texas Instruments Incorporated has been engaged in studies of seismicity and related fields for approximately three years. The purpose of this study was to obtain a quantitative estimate of the low-magnitude seis- micity which constitutes noise in a seismic monitoring system used for the detection and identification of nuclear explosions. This research effort, which began in May, 1961 under Contract AF 19(604)-8517, was sponsored by the Air Force Cambridge Research Laboratories (AFCRL) as part of the Advanced Research Projects Agency (ARPA) Project VELA UNIFORM.

Since the inception of this effort, many contributions have been made toward the understanding of seismicity and seismology in general. In this Final Report little purpose would be served by repeating verbatim what has been published in previous Semiannual or Special Reports; however, for readers interested in such detail, an annotated bibliography is provided in Appendix B.

The primary objective of this report is to evaluate and docu- ment seismic activity that occurred in 1963. The following is a synopsis of major results of this effort.

A. SEISMICITY

1. General

It is apparent from the I960 and 1963 studies that several gaps appeared in the worldwide coverage of earthquakes. The capabilities of some of the standard stations were limited to such extent that there was insufficient areal coverage. In several regions of the earth it is doubtful that many earthquakes of magnitude 4. 5 (mb) were recorded by a sufficient number of stations to permit complete evaluation, while many events of magnitude less than 4 likely went undetected.

Additional factors must be considered in identifying events which have occurred in known seismic areas. Relatively remote, sparsely populated areas where earthquakes of magnitude 3 or 4 may occur should be treated as highly suspicious areas with regard to clandestine testing of nuclear devices. All the northwestern quarter of China is, in this sense, a suspicious area; indeed, it was in this area that the Communist Chinese detonated their first nuclear device. The European and Central Asian seis- mic zones of the USSR are relatively densely populated and are near many stations outside the USSR; however, along the border between the USSR, China and Mongolia lie areas of poorly defined seismicity. An event of mag- nitude less than 4 anywhere in this region may be expected. Relatively few

stations outside the USSR and China or within their political influences are capable of recording events of magnitude 4 within the adjacent Chinese terri- tory. The same may be said of the Northeastern Siberia area within the Arctic seismic belt.

Kamchatka must not be overlooked as a potential test site; while closer than the other areas to good stations outside the USSR, it is still quite distant from sensitive stations in Japan, Alaska and Canada. The high recurrence frequency of earthquakes in the magnitude-4 range com- pounds the problem of identification of nuclear explosions since criteria for discrimination must be applied to a large number of events.

2. Magnitude

The following observations are apparent from the study of magnitudes:

a. Magnitudes as presently calculated require several corrections which are not very well determined. For example, empirical studies indi- cate a large variation of amplitude with azimuth.

b. Magnitudes determined from P-waves are influenced by instru- ment type. For teleseisms, short-period instruments generally yield magnitudes about one unit less than long-period instruments.

c. Magnitudes calculated for events at shadow-zone distances fre- quently are much too high in comparison with those calculated for events at teleseismic distances. The implication is that the A curve has values too high for some regions of the earth.

d. The apparently anomalous relationship between the m. scale, recently adopted by the USC&GS and AFTAC, and the widely used Richter scale, M, may be explained by the different rules for measuring P-amplitudes by instrument band-pass characteristics and by possible shadow-zone effects. At approximately 16 degrees A and greater the relationship between m^ and M becomes approximately M = m^ + 0. 8.

B. NOISE STUDIES

1. Visual Methods

Digitized noise spectra were compared with visual evaluation of the noise. It was concluded that a visual evaluation will yield a measure of the average peak amplitudes as seen by the observer and cannot be directly related to the amplitude or energy density content of the noise record.

limited number of stations may be practical. Before this can be accomplished, however, much more research must be conducted to resolve problems con- cerning regional travel times and phase attenuations.

WORLDWIDE SEISMICITY, 1963

SECTION I

DEFINITION AND SUMMARY

OBJECTIVES

Annual numbers of earthquakes which might be confused with nuclear explosions are of great importance to VELA UNIFORM. Providing additional data for estimating such numbers is the primary objective of this study.

Prior to the publication of the report on worldwide seismicity in I960, little was known of seismic activity below M = 6. 0 on a worldwide

basis. That report succeeded in extending knowledge of seismicity down to M = 5.0. However, the I960 report showed, as was generally known, that

seismic activity is subject to sometimes extreme variations from year to year. Hence, the apparent necessity for a more extended seismicity study resulted in this 1963 study.

Using nearly the same methods developed and standardized in the I960 study, the following objectives were set:

• Documentation of seismic activity during 1963
• Comparison with seismicity in I960 and other years
• Definition of areas having unusual activity in 1963
• Estimation of seismic activity in magnitude ranges and geographical areas of primary interest to VELA UNIFORM In addition to studies performed in the I960 effort, magnitudes on the USC&GS scale (m ) were calculated. Thus, 1963 seismicity data are

available for the same time period, same events and from the same suite of stations based upon both the old and the new magnitude scales, and the results are compared.

It was also necessary to ascertain the capabilities of the suite of stations to record seismic activity. These capabilities were determined from the calculation of theoretical limits of perceptibility and the results obtained were used in estimating the lowest magnitude for which worldwide data may be considered complete.

u I I

8s Is

s « S oo u a; 2-

<t 00 4t (^0) &z °£z it

I I I J

< Q Z 0 h < D u < u a D

hl-H

ü <

o P u 2 12

3

**: A

• 2**

<|H <

t

|a

i 9

<IH <

o < u^ a

ao X P

Li.

<|H

<IH <

<IH

- 4 2"

i 'S- (^) 2 -S " «' 8? ■ » _ • o g B.

4

<i^

tv^ if (^1) _lt_* - ^ Oo o •'

- 2 o 2 1 u S •• - u _ > >. ^ • • • 0 ■

<|H <|H

■ 0

eo ou a

2 ii o " 0 Ö »00 , O (_. < X

2 E S- ^a < S' « Sa — i3 '

2 2

w n

•^» E * >S 3 "

fl S ge (^) *> I-. S 8 - • Ü O" 4 v ¥8X1 ■? o 3 5 X U ■ 9

o

(^1) rf 0 s <

< H

S|H

< O

üU^00 - U ■ o^ a«^ (-•^ O"' J^ B^1 B 0 B 3 5 J r-^ *****^ "•.

< H

0 o 3 • r-

accumulated for i-ach method of magnitude determination are:

ALPHA

BETA GAMMA

SIGMA

ETA

Average difference between calcu- lated and published magnitudes Standard deviation of ALPHA

Maximum algebraic value of ALPHA Minimum algebraic value of ALPHA Number of times each method has been compared to published values

The values of ALPHA are corrections which are added to each magnitude for ^ > 16 except for m. In the case of m , the uncorrected b b value is retained and another m is derived which has the correction applied at all distances. Weights basea upon these statistics are assigned each method. In general, these weights are unity except where the statistics show general unreliability of the method. Also, surface-wave magnitudes computed from short-period instrument data are weighted less due to the decreased reliability of instrument response data at longer periods.

Using these weights and corrections, magnitudes for each event at each station are averaged in the fourth stage as follows:

M

yr (^) pp

M (^) SH

n LR

mean of corrected short- and long-period maximum P-magnitudes, or if P not recorded,

mean of correct short- and long-period PP magnitudes. mean of long- and short-period SH mag- nitudes, weighted average of corrected surface wave magnitudes.

The values of the USC£GS unified magnitude (nrv ) are kept separate from b these magnitudes throughout the program. The final magnitudes calculated for each event at each station consist of TJ , (the mean of Kf , ~ and "NT S P S M LR m, and correc.sd m,.

),

5.0 -Ms-. 6.0 1000

6. 0 ^ M • - 7.0 158 7.0 « M < • 7-3/4 8

Ms 7-3/4^0

Shallow Intermediate Deep (h J 70 km) (70 km < h " 300 km) (h > 300 km)

Total 1166

239

26 1 1_ 267

OS

11 1 o 107

These data give relationships between number and magnitude as follows:

(1) Shallow —log N = 8.00 - 0. 96 M, (2) Intermediate — log N = 8. 14 - 1.11 M , and (3) Deep — log N = 6. 70 - 0. 92 M (^) 3.

The circum-Pacific belt accounted for about 76, 85 and 93 percent of shallow, intermediate and deep focus activity (M 5. 0) in 1963.

The remainder of the deep focus shocks were located in the Sunda arc. Most of the activity outside the circum-Pacific belt was located in the Alpide, although the oceanic ridges and rises were rather well defined by shallow seismic activity.

The southern Kurile Islands were the most active single area in 1963, although the Kermadec-Tonga Islands area had the greatest amount of intermediate and deep-focus shocks. Besides the Kurile-Kamchatka area, activity in the USSR was located east of the Caspian Sea along the border with Iran, in the Caucusus Mountains and in Central Asia near the Hindu Kush.

Activity in southern Chile was lower than in 1960 by a consider- able margin, after less than three years following the onset of the great earthquake swarm of I960.

D, PRINCIPAL CONCLUSIONS

In 1963 the numbers of shallow earthquakes M 6.0 were about the same as the mean annual numbers given by Gutengerg and Richter (1954) while more shocks in the range 5. 0 ^ M < 6, 0 we/e recorded than predicted by Gutenberg and Richter. In comparison to A96O, seismic activity was quite low, mostly due to the reduction in activity in southern Chile.

(^10) ^

However, in most regions fewer shallow shocks were observed in 1963 than in I960. Two exceptions were the Kurile and Aleutian Islands, particularly the former.

Activity at depths deeper than normal was considerably increased over I960. Part of this increase may be only apparent since hypocenter determination has been improved over that in I960 due to the greater number of high-quality stations and use of electronic computer capabilities in processing seismic data.

Analysis of 1963 seismicity, plus reference to the I960 study and other works in seismicity, enable a worldwide system of fault traces to be sketched. These tectonic features outline continental and oceanic stable areas.

Comparison of statistics based on M (^) a and m (^) D shows apparent inconsistencies from region to region, related to the proximity of seismo- graph stations to active zones. Somewhere between 15 and 20 degrees dis- tance, the relationship between m (^) b and M (^) S becomes approximately M (^) S = mb

• 0.8. At distances less than about 15 degrees, a different relationship (one very near Richter's m = 0. 63 M + 2. 5) seems to hold.

Lack of consistent results in relating the two scales and in comparing m data from region to region, compounds the problem of esti- mating the numbers of shallow shocks which might be confused with nuclear explosions in the USSR and China. Estimates of annual numbers of expected shallow earthquakes 4.0 ^ M "' 5.0 can be made, however. These esti- 3 mates are: Kuriles and Kamchatka — 1500; other USSR activity — 50; China — 100. The estimated worldwide shallow earthquakes (4.0 < M «"5.0) was 9000.

E. RECOMMENDATIONS

Seismicity studies should be continued, with emphasis placed on the accumulation of statistics on smaller shocks (M < 6.0). The work of Gutenberg and Richter (1954) essentially established the trends in the geographical distribution of seismic activity. However, their studies were based on data for large shocks only, and activity in the magnitude range of interest to VELA UNIFORM was inferred from statistics of much larger magnitudes. The studies of I960 and 1963 seismicity extended the range of knowledge down to M =5.0 and showed that the estimates of Gutenberg and Richter for annual numbers of earthquakes in the range 5. 0 -- M < 6. 0 were probably too low. Only the accumulation of more statistics can substantiate this conclusion.

11

SECTION II

PRESENTATION OF DATA

A. WORLDWIDE RESULTS

The 1963 seismicity was, as is usual, concentrated in the circum-Pacific belt. Of those earthquakes for which M 2 5.0, 76 per cent of the shallow, 85 per cent of the intermediate and 93 per cent of the deep focus shocks were located in the circum-Pacific seismic belt. Other seismic activity occurred in the Alpide zone (principally in the Eastern Mediterranean, Hindu Kush to the western Himalayas, southern Sumatra and eastern Java, and the eastern extremity of the Sunda arc), in the Caribbean loop and along most of the oceanic ridges and rises.

Figures 1 through 5 illustrate the geographical distribution of seismic acitvity in 1963. Figures 1 and 2 show plots of epicenters for which M 2 4. 0. Epicenters in Figure 1 are only for shallow and normal focus shocks (h £ 70 km) while epicenters in Figure 2 are for both inter- mediate (70 km < h« 300 km) and deep-focus (h > 300 km) earthquakes. Figures 3 and 4 parallel Figures 1 and 2 in content, except that in these figures, epicenters are shown for which m 24.0. Figure 5 is a seismicity map which shows annual numbers of earthquakes per unit area in the magni- tude range 5. 0 <: M < 6. 0. The unit area chosen is approximately the size

of a 5-degree grid, or 3 x 10 square kilometers.

Several features of 1963 seismicity, as illustrated in Figures 1 through 5, are of particular interest. These are:

(1) The earthquake swarm in the Kurile Islands which began 12 October. Several minor swarms occurred earlier in the year, such as in June and July, and may have been hearlding the onset of the major activity. (2) The clear delineation of oceanic ridges and rises by seismic activity. The mid-Atlantic ridge is well out- lined by earthquake epicenters, as are the ridge sys- tems in the Indian Ocean and the Southeastern Pacific. (3) The return of activity in southern Chile to near that level existing before the great earthquake swarm of

(4) Somev/hat increased activity along the southern USSR border in the vicinity of the Caucasus Mountains and in the Kopet-Dag region.

13

(5) Greatly increased intermediate depth activity in the Banda Sea-Molucca Passage area. The only "class a" (M ■ 7-3/4) earthquake recorded in 1963 (magnitude 8.2), coordinates 6. 8 S; 129. E; depth, 80 km; date, 4 November; 01:17: GCT was located in this area.

A more detailed analysis of the geographical distribution of seismic activity as pictured in the first five figures, plus consultation of similar data for I960 seismic activity, reveals a clear pattern of stable areas surrounded by zones of varying seismicity levels. By tracing these seismic zones and extrapolating some of their trends, the major tectonic features of the earth or global zones of maximum stress may be defined. Beginning <U a point just off the east coast of Kamchatka at about 65 degrees north, a fault system may be traced through the Komandorskie Islands along the Aleutian Island arc extending up the Alaska peninsula and the Aleu- tian range to about the Arctic Circle north of Fairbanks (this may extend on to the northern coast of Alaska) where a sharp turn to the southeast may be traced. From this point, the tectonic system trends along the coast and then just off the Alexander archipelago to the Queen Charlotte Islands, where a lesser zone branches off to Vancouver Island ( and possibly connects to the activity in Montana-Wyoming-Utah area across Idaho), while the main branch parallels the Washington and Oregon coasts, some 400 or 500 kilometers off shore, to about 45 degrees latitude north, where it curves back to the Cali- fornia coast at about the point where the San Andreas fault disappears into the ocean. After reaching the coast, the fault systenn follows along the coast ranges into southern California where an intersection is made with a fault systenn trending along the strike of the Sierra Nevadas (this may be connected with the branch through Vancouver Island, thus forming a loop about a small stable area in western Oregon) and continuing down to and through the Gulf of California, then south along the 110th west meridian to the Revilla Gigedo Islands, where another turn (approximately 90 degrees) in the trace leads to an intersection with the Jalisco, Mexico coast.

Just off the coast at this point another branch occurs. The main branch continues along the coasts of Mexico and Central America while the other branch trends generally south to about the equator and then curves to the southwest toward Easter Island. The main branch continues along the Central America coast to the Peninsula de Azuero, where it leaves the Panama coast and continues in a southerly direction to about 5 degrees north latitude and turns sharply to the east. From this point, yet another branch extends westward to the Galapagos Islands (this branch may connect with the Easter Island-to-Revilla Gigedo Islands branch).

14