TUNNEL LOCATION BY MAGNETOMETER, ACTIVE SEISMIC, AND RADON DECAY METHODS

ALEXANDRIA LABORATORIES REPORT NO.: AL-75-1

W. C. Dean

18 June 1975

Teledyne Geotech

314 Montgomery Street

Alexandria, Virginia 22314

APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED.

Unclassified SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)

REPORT DOCUMENTATION PAGE READ INSTRUCTIONS BEFORE COMPLETING FORM

1 REPORT NUMBER 12. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER

AL-75-1 4. TITLE (and Subtitle) s TYPE OF REPORT & PERIOD COVERED

TUNNEL LOCATION BY MAGNETOMETER, ACTIVE SEISMIC, Technical AND RADON DECAY METHODS

6 PERFORMING ORG. REPORT NUMBER

7. AU THOR(s) 8 CONTRACT OR GRANT NUMBERfs)

w. c. Dean F08606-74-C-0013

9 PERFORMING ORGANIZATION NAME AND ADDRESS 10 PROGRAM ELEMENT. PROJECT TASK

Teledyne Geotech AREA & WORK UNIT NUMBER';

314 Hontgomery Street Alexandria, Virginia 22314

11 CONTROLLING OFFICE NAME AND ADDRESS 12 REPORT DATE

Defense Advanced Research Projects Agency 18 June 1975 Nuclear Monitoring Research Office 13 NUMBE~l/- PAr.<'S

1400 Wilson Blvd.- Arlington, Virginia 22209 ~ 1 4 MONITORING AGENCY NAME B: ADDRES$(1/ different from ControllinR Office) 15. SECURITY CLASS. fof this re(lr·rt;

VELA Seismological Center Unclassified 312 Montgomery Street

Alexandria, Virginia 22314 fts;;_· DECLASSIFICATION DOWN GflADI N-c;--SCHEDULE

16. DISTRIBUTION STATEMENT (of this Report)

APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIHITED. '

\

\

' '·.\ ·, \

17. DISTRIBUTION STATEMENT (of the abstract entered in Block 20, If different from Report) . ' ' '·. \[\

' . ' ,_ \'\.

,J ./ ' ... ~~ ' ' ,>

/ r ' \ 7

·, •' -.• \ . ' ; I

' ''· ·'_)' '· ~

18 SUPPLEMENTARY NOTES < • "' ~·· \ \ \ ~ --\"\ \\\\/'"

\ \ \ \

\) \.'

19 KEY WORDS (Continue on reverse ,•dde if necessary and Identify by block number)

Tunnel location Proton Magnetometer Seismic location Radon detection

20 ABSTRACT (Continue on reverse side If nectlssary and identify by block number)

This project investigated three methods of detecting mines or tunnels in granite: magnetic, seismic, and radioactive decay of radon.

The magnetic, using a proton precession magnetometer, aims to detect tunnels from the magnetic materials (rails, roof bolts, wiring) these mines are expected to contain. This method works for both surface and borehole surveys when the sensor is less than 20 feet from the magnetic source, For deeper rails a surface survey will probably not be successful since the rail

DO FORM 1 JAN 73 1473 EDITION OF 1 NOV 65 IS OBSOLETE

' J Unclassified

SECURITY CLASSIFICATION OF THIS PAGE r"'hen nart~ Entned,

Unclassified SECURITY CLASSIFICATION OF THIS PAGE(When Data Entered)

anomalies are not of one sign but oscillate above and below the mean background level with wavelengths on the order of two rail lengths. Such anomalies cannot be distinguished from weak geologic anomalies.

The active seismic methods aims to detect tunnels by reflection or refraction of P-waves from the tunnel or by reverberations of the tunnel walls. These experiments were unsuccessful due to insufficient high frequency energy being coupled into the rock containing the tunnel. Some evidence of cavity wall reverberations appeared over a coal mine, but the tunnel was shallow (15 feet) and the medium (coal) had a lower seismic velocity than granite.

The radon decay method aims to detect tunnelling activity by the extra release of radon gas freed into the atmosphere. For a tunnelling rate of 16 cubic meters per day, this method will fail due to a poor signal-to-noise ratio relative to all of the other sources of radon in the atmosphere.

• • Unclassified /I SECURITY CLASSIFICATION OF THIS PAGE(When Dntn 1-:ntete{t:

ABSTRACT

This project investigated three methods of detecting mines or tunnels in

granite: magnetic, seismic, and radioactive decay of radon.

The magnetic, using a proton precession m: -... tometer, aims to detect

tunnels from the magnetic materials (rails, rr lts, wiring) these mines

are expected to contain. This method works th surface and borehole

surveys when the sensor is less than 20 feet from the magnetic source. For

deeper rails a surface survey will probably not be successful since the rail

anomalies are not of one sign but oscillate above and below the mean background

level with wavelengths on the order of two rail lengths. Such anomalies cannot

be distinguished from weak geologic anomalies.

The active seismic methods aims to detect tunnels by reflection or

refraction of P-waves from the tunnel or by reverberations of the tunnel walls.

These experiments were unsuccessful due to insufficient high frequency energy

being coupled into the rock containing the tunnel. Some evidence of cavity

wall reverberations appeared over a coal mine, but the tunnel was shallow

(15 feet) and the medium (coal) had a lower seismic velocity than granite.

The radon decay method aims to detect tunnelling activity by the extra

release of radon gas freed into the atmosphere. For a tunnelling rate of 16

cubic meters per day, this method will fail due to a poor signal-to-noise

ratio relative to all of the other sources of radon in the atmosphere.

-ii-

ABSTRACT

EXECUTIVE SUMMARY

INTRODUCTION

MAGNETIC METHODS

TABLE OF CONTENTS

Conclusions on the Magnetic Experiment

ACTIVE SEISMIC METHODS

Conclusions on the Seismic Experiments

ESTIMATE OF THE DETECTABILITY OF MINING ACTIVITIES MONITORING AIRBORNE Rn222

REFERENCES

ACKNOWLEDGEMENT

APPENDIX A

APPENDIX B

-iii-

Page

1

4

5

36

38

52

BY 56

59

60

Figure No.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

LIST OI' FIGURES

Title

Nomogram for estimating magnetic anomalies.

Magnetic survey over rail spur at Ft. Belvoi1, Virginia.

Map of magnetic survey, Alexandria, Virginia.

Magnetic survey over rail spur, Alexandria, Virginia.

Magnetic profiles parallel to the rails, Alexandria, Virginia.

Operator with portable magnetometer over east portal, Carothers Tunnel, B&O Railroad.

Map of magnetic survey over Carothers Tunnel, B&O Railroad.

Magnetic anomalies over Carothers Tunnel, B&O Railroad.

Magnetic survey, upper profiles, over Carothers Tunnel, B&O Railroad.

West portal of Tunnel 19, DRGW Railroad.

Map of magnetic survey over Tunnel 19, DRGW Railroad.

Cross section of Tunnel 19 and magnetic profiles, DRGW Railroad.

Operator with portable magnetometer of upper profiles over Tunnel 19, DRGW Railroad.

Trackside magnetic. survey near Tunnel 19, DRGW Railroad.

9

10

ll

12

13

14

15

17

18

19

')' ·- .l

Magnetic survey over Tunnel 19, DRGW Railroad. 22

Smoothed magnetic profiles over Tunnel 19, DRGH 24 Railroad.

Expected anomalies from rails over Tunnel 19, 25 DRGW Railroad.

-iv-

LIST OF FIGURES (Continued)

Figure No. Title

18 Actual versus theoretical anomaly, profile M, Alexandria survey.

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

Theoretical gradient for 21 profiles over Tunnel 19, DRGW Railroad.

Actual gradient for 21 profiles over Tunnel 19, DRGW Railroad.

Magnetic anomalies from freight train passing through Carothers Tunnel, B&O Railroad.

Special sensor for borehole magnetometer.

Drilling rig over Colorado School of Mines experimental mine.

Cross sectional view of borehole and CSM experimental mine.

Magnetic survey over mine rails at CSM experimental mine.

Borehole magnetometer survey at CSM experimental mine.

Seismic recording truck of Colorado School of Mines.

Dinoseis truck.

Recording units of the Bison hammer seismograph.

Seismic profile over Tunnel 19, DRGW Railroad.

Center set of Dinoseis records taken over Tunnel 19, DRGW Railroad.

Summation record over Tunnel 19 with static and dynamic corrections applied.

Center set of Dinoseis records taken over Tunnel 19, DRGW Railroad.

Conventional stacking of Tunnel 17 data corrected for two-way travel times to tunnel.

Three stacked records over Tunnel 17, assuming a tunnel at the south end, center, and north end respectively.

-v-

Page

28

29

31

32

33

34

35

37

39

41

42

43

44

47

48

49

50

LIST OF FIGURES (Continued)

Figure No. Title P~rL

36 Seismic cross section over Tunnel 17, DRGW Railroad. SJ

37 Typical Bison records over coal mine. 53

38 Map over coal seam with contours showing region of 54 reverberation seismograms.

-vi-

EXECUTIVE SUMMARY

This project investigated three geophysical methods of detecting

underground mines or tunnels. The three methods are magnetic, seismic, and

radioactive decay (of radon).

The magnetic method aims to detect mines from the magnetic anomalies

generated by the rails these mines are all expected to contain. Other

magnetic material in mines might include electrical wiring and structural

material such as roof bolts and steel reinforcement bars in concrete slabs.

We conducted magnetic surveys with a proton precession magnetometer

directly over railroad rails (sensor height eight feet), and over railroad

tunnels where the rails were buried 50 to 100 feet below the magnetometer.

In addition we drilled a borehole which passed within 12 feet of a mine and

ran a magnetic survey with a borehole sensor.

The results show that the rails do exhibit strong magnetic anomalies

(up to 1500 gamma for mainline railroad rails). However the magnetic anomalies

are not constant as we had expected but oscillate above and below the back­

ground magnetic field of the earth with wavelengths comparable to the length

of the rail sections. This oscillation further obscures what was already

expected to be a small anomaly amongst the much larger geologic anomalies

which are ever present. Moreover in the U.S.A. the railroad tunnels invariably

are accompanied by fences and phone lines running over the tunnel parallel

with the rails. These create surface anomalies, which are readily detected

by the magnetometer, but further obscure the weak rail anomalies. Even

without the confusion of parallel fences and phone lines, detection of rails

in a mine by a surface magnetometer survey is doubtful even for mine (rail)

depths of 15 to 50 feet.

In contrast the magnetometer shows strong anomalies, and oftentimes quite

erratic behavior on repeated readings, when the sensor is close (a few inches

to a few feet) to magnetic materials. The borehole survey showed just this

behavior as the sensor passed the roof bolts and reinforcements in the

experimental mine of the Colorado School of Mines. The erratic behavior

-1-

is especially pronounced when there are loops of magnetic Ti>Clt-<:o:: i ,J 1 ,;;-:;r,,-,

(such as wire mesh or barbwire ::ooped around a fencepost). Tile rc:ason is

that the pulsing of the magnetometer creates an electromagnet out of the

magnetic wire loop where the residual effect varies from reading to reading.

Thus borehole magnetic surveys do show some promise of detecting mines

from the magnetic material contained in them although, when successful, the

strongest magnetic anomalies may not be coming from the rails.

The seismic method aims to detect tunnels by reflection or refraction

of P-waves from the tunnel. For this method to work effectively, the seismic

wavelengths in rock should be comparable or shorter than the cross-section

dimensions of the target. Even for railroad tunnels (15 feet by 25 feet)

the wavelengths of seismic energy used in geophysical exploration are much

longer than the tunnel dimensions.

Our field experiments show very little high frequency energy (100 to

500 Hz) coupled into the rocks resulting in no reflection evidence of the

tunnels. This was true when we used Dinoseis, or dynamite caps, or a sledge

hammer (used with the Bison hammer seismograph). The problem at the sites

over the two railroad tunnels in Colorado was due to a low velocity layer on

the surface trapping almost all of the energy preventing any significant

penetration into the granite containing the tunnel.

Watkins (1967) reported on reverberations from underground cavity walls

with a resonant frequency close to that predicted by Biot (1952). We did

not see such reverberations from the railroad tunnels again due to the lack

of energy penetrating the granite. However we did see such reverberations

over a mine in coal (15 feet deep) on a Bison seismic profile. Although the

coal tunnel is smaller (2x2 meters) than railroad tunnels, it is also shallow

(5 meters) and exists in a uniform half space (coal) with a seismic velocity

much less than that of granite.

For any subsequent experiments the field practice should make use of a

multiple geophone system rather than a single sensor and hammer source such

as the Bison. The correlation between traces is better when several traces

are looking at seismic returns from the same source rather than one source per

trace as in the Bison hammer seismograph.

-2-

The radon method aims to detect on-going tunnelling activity by detecting

the release of radon gas freed into the atmosphere. All granite and crystaline

rocks contain some uranium. Radon (Rn222

) is a daughter product of u238 and

will be released to the atmosphere as the crushed rock from the mine is

excavated.

Our study shows that for the amount of tunnelling expected (4 n1eters

per day with a cross section of 2 x 2 meters), it is highly unlikely this

method could detect radon from the tunnelling activity as opposed to all

other sources of radon in the atmosphere, even under the most optimistic

assumptions. With more realistic assumptions there is no hope.

-3-

INTRODUCTION

The objective of this project is to detect underground mines or tunnels

by geophysical methods. The mines are expected to be approximately 2 x 2

meters in cross section, located in crystaline rocks such as granite,

gneisses, or shists in hilly terrain, and with depths of burial of a few

feet to as much as 200 feet or more.

This project has limited its attention to only three methods: detection

of magnetic materials in the mine, reflection or refraction evidence of a

tunnel by seismic exploration techniques, and detection of radon released to

the atmosphere by the excavation activity. Other methods or techniques are

considered outside the scope of this study.

-4-

MAGNETIC METHODS

The tunnels of interest are all expected to have rails. In addition they

probably have electric wiring which may or may not be in use. In some loca­

tions, perhaps where the mines are shallow or in unconsolidated material, they

may have roof bolts or iron reinforcements in concrete walls or ceilings.

The strength of magnetic dipoles decrease as the inverse cube of the -3 distance (r ). However, rails are good approximations to infinitely long

magnetic sources and so would be expected to decrease as the inverse square -2 of the distance (r ). Figure 1 from the Geometries Applications Manual for

Portable Magnetometers (Breiner, 1973) is a nomogram showing the expected

magnetic field strength from various types of objects versus the distance

from the magnetometer. Note that the field strength from most objects vary

as the inverse cube of the distance but that from pipelines vary as the

inverse square.

In Figure 1 Breiner used the cross sectional area, A, of a pipeline as:

A = nDt

where Dis the diameter of pipeline (6") and tis the thickness (1/4"). Such

a pipeline has a volume of 0.1 cubic feet per yard or approximately 50 pounds

per yard. By comparison mainline rails are frequently 135 pounds per yard

and mine rails are approximately 30 pounds per yard.

If we can detect infinitely long magnetic anomalies of 10 gamma, then

we might expect to detect a 6-inch pipeline or a pair of mine rails buried

at a depth of 40 feet. Other expected characteristics of our mine rail

anomalies are that their expected direction is known and that their route

will be roughly horizontal so the anomaly strength, will decrease with the

square of the ground elevation (proportional to h-2).

The equipment we used were proton precession magnetometers manufactured

by Geometries. These instruments can measure the earth's magnetic field

accurately to within one or two gammas. We used both a base station with a

recorder (Model G-826) and a portable unit (Model G-816).

-5-

500

400

300

200

100

50

40

10

~ ~ q.

.<)

5 -:<,

4

y ~

(:'

3

2

FEET 2 4 6 10 20 40 60 80 100 150 200

CENTIMETERS 100 200 400 GOO 800 1000 2000 3000 4000

DISTANCE FROM MAGNETOMETER

Figure 1. Nomogram for estimating magnetic anomalies.

-6-

Magnetometer surveys over rail lines do show magnetic anomalies of 1500

gammas relative to the background field when measured by the portable sensor

eight feet above the surface of the ground. Figure 2 shows two parallel

profiles normal to an east-west rail spur in Fort Belvoir, Virginia.

We ran a more elaborate survey of eight profiles over a rail spur in

Alexandria, Virginia (see map on Figure 3). The anomalies over these profiles

are shown on Figure 4. The interesting feature of these profiles normal to

the rails is that they are not all of the same sign. Profiles run parallel

to the rails as shown on Figure 5 demonstrate the oscillatory behavior of

these anomalies. The half-wavelength of these oscillations correspond roughly

to the length of the rails (33 feet). These anomalies are reproducible even

when measured on different days as indicated by the x's on the center profile.

It is as though one end of the rail is a north pole and the other a south

pole.

The first magnetometer survey over a tunnel we conducted at the Carothers

Tunnel on the main line of the B&O Railroad near Paw Paw, West Virginia.

Figure 6 shows a photograph taken over the East Portal of the Carothers Tunnel.

An operator is shown with the portable magnetometer and the sensor at the end

of the 8 foot aluminum pole. Also shown are the telephone wires running over

the tunnel. In many places these wires are less than 8 feet off the ground.

Figure 7 shows a plan view of the railroad (which is double tracked), the

east tunnel portal, the phone line, and several of the magnetometer profiles

run over the tunnel. Figure 8 shows the magnetic field variations for two

profiles over the rails at a scale of 1000 gammas per inch. The lowest

magnetometer profiles over the tunnel are also plotted on the same figure with

the same vertical scale.

Figure 9 shows the magnetometer profiles at the higher levels with a

vertical scale of 20 gammas per centimeter. The position of the rails under

each profile and the location of the phone line along each profile are clearly

marked.

There is a distinct magnetic anomaly due to the phone line on each

profile. There are other magnetic variations over the rail but it is difficult

-7-

PROFILE A

A - 56000y

PROFILE 8 8 - 56000y

A - 55000y

8 - 55000y

PROFILES 15 FEET APART

-120 -100 -80 -60 -40 -20 0 20

DISTANCE FROM RR CENTER IN FEET

s N

Figure 2. Magnetic survey over rail spur at Ft. Belvoir, Virginia.

-8-

X X X X

PA BUS X

RKING ~ X X X

XX XXX XXX

,1' FE NCE

J K L

I

PROFILES M N

I

0 p Q

~N

RR LINE

j.-20 FT~

- .,......- SENSOR 0: r BASE STATION

THE RR LINE IS ONE BLOCK EAST OF N. FAIRFAX ST., ALEXANDRIA, VA. PROFILE M IS ON THE NORTH EDGE OF WYTHE ST., ALEXANDRIA . PROFILE SEPARATION IS 15 FEET. READINGS EVERY 1 FOOT FROM 0 TO 6 FEET FROM RR,

" 2 FEET " 6 TO 20 " " " 5 FEET 10 TO 100"

10 FEET " 100 TO 130 "

Figure 3. Map of magnetic survey, Alexandria, Virginia.

-9-

-N 56000 y

-M 56000y

-L 56000y

-K 56000y

-J 56000y

+40 +30

-0 56000y

-P 56000y

-0 56000 y

+20 -+10 0 -10

DISTANCE FROM RR IN FEET

T 1000

GAMMAS

1 -20

Figure 4. Hagnetic survey over rail spur, Alexandria, Virginia.

-10-

0

p

0

N

M

L

K

J

-30

WB

W4

W-RAIL

RR CENTER

E-RAIL

E3

E5

E10

E16

0

PROFILE LETTER INDICATES EAST OR WEST OF RR CENTER; NUMBER INDICATES DISTANCE IN FEET FROM RR CENTER. NOTE: x'S ARE SAME PROFILE SIX DAYS EARLIER.

20 40 60 80 100 120 140 DISTANCE FROM E- W PROFILE M IN FEET

160

X

l 1000

GAMMAS

T

Figure 5. Magnetic profiles parallel to the rails, Alexandria, Virginia.

-11-

Figure 6. Operator with portable magnetometer over east portal, Carothers Tunnel, B&O Railroad.

-12-

I f-' w I

:::; 300 .... .... z .... z .... 200 .... z Cl : a.

:IE ~ 100 .... .... y z < l­en Cl 0

s

~

EAST PORTAL ~ .. ~ RR

~ PHONE LINE

0

IX

IV VII VIII VI

v Ill

r--•t'"' I'• I '-. ~ ~ """'"

100 200 300 400 500

DISTANCE AlONG TUNNEL EAST-TO-WEST IN FEET

...,

600

TUNNEL HILL CREST

+

700

~w

Figure 7. Map of magnetic survey over Carothers Tunnel, B&O Railroad.

.....__

'EST PORTAL

11/lml

I !--' ~ I

IV .. ~·~···•""'"-"" , ...• --.,..,_ ·-~-·----·-~·---,---~ ....._ ·-~' "'""-

IV - 56000y·

Ill ·· ..... ----·,· ,-,·--··--·~---· ______ , _____ , . .,., .. , ___ ._, ... "'"

Ill - 56000y

A

A - 56000y 8

B - 55000y

II ~ ---------i:i6000y

--·,~--------

-50 j_

-40 -30

I - 56000y

:boo.. N

-20 -10 0 10 20 30 DISTANCE FROM RR CENTER (FEET)

T 1000

GAMMAS

1

40

F i. 1;ure :::l. ~1agnetL: ;momalies over Carothers Tunnel, B&ll Ra ilrodd.

50

D=305'

D=225'

D=115'

D=125'

D=95'

D=75'

Ill D=55'

H=120'

H=114'

H=109'

H=102'

H=98'

IX - 56340y 1 1

1 l

1.1 .ll.

RAIL POSITION UNDER PROFILE

l. l.

VI- 56340y

v - 56340y

IV - 56340y

! Ill - 56340y

t POSITION OF PHONE LINE

H=DEPTH TO RAILS IN FEET l D = DISTANCE FROM EAST PORTAL IN FEET

-100 -80 -60 -40 -20

DISTANCE FROM PHONE LINE IN FEET s ,.....,

0

20 GAMMAS

T 20

Figure 9. Magnetic survey, upper profiles, over Carothers Tunnel, B&O Railroad.

-15-

IX

VIII

VI

v

IV

Ill

t.o attribute these variati.ons directly to the railc:,, ~speclt1ly i:c Jight of

the strange anomalies measured on the two parallel profi]es directly over the

rails (profiles I and II on Figure 8).

A more thorough magnetometer survey was run over Tunnel 19 of the Denver

and Rio Grande Western Railroad near Eldorado Springs, Colorado. This tunnel

(the west portal of which is shown in Figure 10) is in granite, has a single

track, and has much easier access over the tunnel than the one in West

Virginia. Figure 11 shows a map of the railroad, east: portal of the tunnel,

fences, phone lines, and the 21 profiles run. Figure 12 shows a cross

section of the tunnel and the profile elevations along the phone line.

Profile A is 50 feet and profile U is 90 feet above the rails. Figure 13

shows an operator with a portable magnetometer at the upper profiles over

Tunnel 19.

We conducted a track level survey as well. The railroad crosses a high

fill immediately east of Tunnel 19 which provides only limited access to the

north and south of the rails. Figure 14 shows the magnetic anomalies over

the north rail and 10 feet to the north, and over the south rail and 10 feet

to the south. Again the oscillatory character of the magnetic anomalies is

evident and again the oscillation half-wavelength corresponds roughly to the

rail length of 39 feet.

The magnetic data for the 21 profiles over the tunnel are shown in

Figure 15. As was consistent with most of our surveys magnetometer readings

were taken every 5 feet along the north-south profiles. The profile spacing,

east to west, was 8.5 feet. A base station magnetometer was recording at the

top of the mountain 140 feet above the raih:. A constant base station correc­

tion was applied to each half profile (i.e. to those readings taken within

10 minutes of each other).

The magnetic profiles show strong fence and phone line anomalies. Since

they are parallel to the rails, their strong anomalies tend to obscure the

weaker ones which might be expected from the rails.

-16-

Figure 10. West portal of Tunnel 19, DRGW Railroad.

-17-

~ 0:, I Ill

1\1

J r I 1-

H 1-----

G ~ I= L___ ' I

E 1--­D ~­C 1-----B 1----· -A ~-----

RR

FENCE /! f I I

/..,___PHONE LINE

U f- · · ····-·--- ----- _/ / I U T ~ -- { J ! IT

s ~---------- - .J.. / I S

QRl:___ i . f! II~ t~ --- --f- l- ! 1ci

N ---------- i -----· j- : I N M t· - - - .. . - - - . 7 I ~ M

-~~==-==~-~===~=-· ~~ ! I? 1---- -----F / II

--~-~ f I H . --r- . (

I F f E

\

D c

B

SCALE: 1" = 50" - --- \"\ I A

\ ~ EAST '

PORTAL \ !TUNNEL 191\

I::::::::- N

Fi~~ure J 1. ~-l~lp dt magnctiL· sur\·'t'\' l)\'L'l I'ttllilL'l 1'-l, l)l\~,,~., i\,li rl'.~1d.

I 1--'

"' I

50'

PROFILES p

N 0 • • M • L • K • J •

R s T

u • •

0 • • •

G E F

D 8

c • •

A •

• •

H • • •

PROFILE ELEVATIONS ESTIMATED AT PHONE LINE • •

------------TUN-;;E~TOP-;r----- ------- __ _ ------

SCALE: 1" = 20'

EAST PORTAL

RR TRACKS

Figure 12. Cross section of Tunnel 19 and magnetic profiles, DRGW Railroad.

90'

Figure 13. Operator with portable magnetometer of upper profiles over Tunnel 19, DRGW Railroad.

-20-

N + 10'- 56000y

\.

N -56000y

N RAIL (SHOWING JOINT POINT POSITIONS)

S - RAIL

s- 56000y

0 100 200 300

0 !STANCE ALONG RR IN FEET

Figure 14. Trackside magnetic survey near Tunnel 19, DRGW Railroad.

-21-

T 100 GAMMAS

_l_

PROFILE;;S SOUTH FENCE ."' I ~

;;'~/ U \ I

'\\: /

:0\~/ R ~/~~·~ .. /

n ://\~ '-/ : ~/''~. :~~~~ I r~j\~j

I~J• K I \

~\\~~0 J r ~\~

H~~~· G ~\ F ______;-J~ \ E _______J I

D -----J/ c ------J ~Y\-~/" v (\ \. . B ~ I ·vf\\

I~ \ A~ I

READINGS ALONG PROFI~ES EVERY 5 FEET \. PROFILES SEPARATED BY 8.5 FEET

I

-200 -100 0 100 200 DISTANCE FROM SOUTH FENCE IN FEET

Figure 15. Magnetic survey over Tunnel 19, DRGW Railroad.

-22-

We applied several kinds of smoothing. First we smoothed out the fence

and phone line anomalies by hand. We also made a computer interpolation for

each profile as follows:

Profile A +I sin(x-k~x)rr/~x k=-Nak (x-k~x)rr/~x

Trends were removed first to avoid the Gibbs effect at the ends which would

otherwise be discontinuities. These smoothed data are plotted in Figure 16.

We can estimate the rail anomalies at each of the over-tunnel profiles

from a theoretical argument. If the anomalies at rail level vary sinusoidally,

and vary inversely as the square of the rail-magnetometer separation, then

to a first order approximation, the rail magnetic anomalies should be:

~(x,y) M

0 cos ~ L

(1)

where h is the height of the sensor above the rails, x is the distance along

the profile in a north-south direction from the rail center, and y is the

distance in the east-west direction from Profile A. The half-wavelength L,

is equal to the rail length. Figure 17 shows the expected anomalies from

the rails at each of the profiles over the tunnel plotted to the same scale

of the measured data on Figures 15 and 16. Since the rail anomalies are

small, broad and oscillatory, they are readily obscurred by the larger,

sharper, and more continuous geologic anomalies and especially the larger

and sharper phone line and fence anomalies running parallel to the rails.

Since it is the oscillatory behavior of the rail anomalies which most

effectively hides them in the geologic anomalies, we can ask whether some

other measure of the magnetic field would give an effect more constant in

value along the axis of the rails. One possiblity is the gradient. The

magnetometer measures the vector sum of the earth's field and that from the

rails. Thus the effect from the rails could be a rotating vector rather

than one that goes to zero. If the form of the rail anomaly is given by

equation (1), then the gradient will be:

-23-

PROFILES

~ /+ u +

T + /(;\, /"/~ /- + \ ~-----"

s + ;r~\/ ~//, R + J/\\~J +

0 + ~\_ +

p + ;; ~~v~~~/ +

0 + /~/\¢~~ +

N + ~·'-/-"'\ ~I +

M + I J"\~::; +

L + y\~~~ +

K + +

J + +

I + +

H + +

G + ~ +

____r~---\fi~\ F + +

E + Yr~ \ ~/"-·~

+

0 + +

c + ~~-,~ +

T B + ~ \ +

100 GAMMAS ~~'v· --~~j\\ _1_ A + +

+ \ +

-200 -100 0 100 200 DISTANCE FROM SOUTH FENCE IN FEET

Figure 16. Smoothed magnetic profiles over Tunnel 19, DRGW Railroad.

-24-

T 100 GAMMAS

_L

PROFILES u T

s

R

Q

p

0

N

M

L

K

J

H

G

F

E

0

c

8

A

-200 ~

-100 0 100

DISTANCE FROM RAILROAD CENTER IN FEET S ,..... ,..._ N

Figure 17. Expected anomalies from rails over Tunnel 19, DRGW Railroad.

-25-

'Vcp(x,y)

The absolute value

lv¢1 h2

In this form when

or when x +.!::.+ n-

M 0

-i h2 + 2 h2 X

of the gradient

M 2x 0

2 h2 + + X X

L 2 2 - h

1/2

'TT

(2)

2x ~. 'TT sin ~ cos L -J -

2 L L + X

will be

2 2.2!Z

2 . 2~

1/2 'TT

cos L + - Sln 1 2 L

the coefficients of the sine and cosine terms are equal and the sinusodial

variation of the absolute value of the gradient, <lv¢1), disappears.

To test whether empirical data yields reasonable values for h and L,

we consider profile M from the Alexandria survey shown in Figure 18. The

theoretical curve gives the fits the data approximately when h = 6.5 feet,

a fairly close match to the sensor height of 8 feet, and L = 21 feet (found

by averaging distances between zero crossings), which is the same order of

the rail length (33 feet).

A theoretical gradient computed for the 21 profiles over Tunnel 19 of

the DRG-RR is shown in Figure 19. Here the height, h, is the sensor height

over the rails and rail length (L) is 39 feet.

The actual gradient of the data over Tunnel 19 with the phone line and

fence anomalies smoothed out is shown on Figure 20. As shown on Figure 19

the expected gradient is still oscillatory as was the original field.

Moreover, the gradient of the actual data does not show an anomaly trend

parallel with the rails which can be readily attributed to the rails.

-26-

I N -....J I

W,.......... DISTANCE IN FEET (y) t::==..... E -40 -30 -20 -10 0 10 20 30

-CENTER OF TRACKS IS AT y = 0.

THEORETICAL

-- I "'· j_./

DATA

T 1400y

Figure 18. Actual versus theoretical anomaly, profile M, Alexandria survey.

PROFILES ---u +

T +

s +

R +

Q +

p +

0 +

N +

M +

L +

K + +

J +

I +

H +

G +

F +

E +

0+

C+

T a+ 1 GAMMA/FOOT

A~l-===~~=---~?~----=~==--~1 _i_

-200 -100 0 100 200 DISTANCE FROM CENTER OF RAILS IN FEET

Figure 19. Theoretical gradient for 21 profiles over Tunnel 19, DRGW Railroad,

-28-

SOUTH FENCE

PROFILES 1 PHONE LINE

' + u +

+ T +

+ s +

+ R +

+ Q +

+ p +

+ 0 +

+ N +

+ M +

+ L +

+ K +

+ J +

+ + + H +

+ G +

+ F +

T+ E +

10 GAMMAS/FOOT + D +

l_ c + +

I I

+ B I +

I A +

-200 -100 0 100 200 -- -- -DISTANCE FROM SOUTH FENCE AT PROFILE A IN FEET

Figure 20. Actual gradient for 21 profiles over Tunnel 19, DRGW Railroad.

-29-

One type of magnetic anomalies is easy to detect. When a train is

moving through the tunnel, erratic readings of several hundred gamma are

easily detected on a magnetometer. Figure 21 shows a time series of such

readings taken 10 seconds apart. This behavior is always apparent even when

the sensor is 150 feet above the rails. Similar changes (20 to 100 gamma)

in the background magnetic field were noticed when busses were parked or

removed from a lot 100 feet from the base station during the Alexandria

survey. Heavy equipment moving through a mine might be expected to have a

similar effect.

The final test was a borehole magnetic survey conducted at the experi­

mental mine of the Colorado School of Mines near Idaho Springs, Colorado.

Geometries built a special sensor for borehole use. It has a 2.5 inch

outside diameter and is approximately two feet in length. It is hermetically

sealed to a 600 foot cable. Figure 22 shows a photograph of the special

sensor. Because of the extra length of the cable the base station magneto­

meter with a higher current capacity than the portable unit, must be used

to drive the sensor.

Geometries reports that this special sensor is not as good in cancelling

stray electric fields as their standard sensor (4.5 inch outside diameter).

However, we noticed no difference in the response of the sensor \vhen the

mine lights were on or off during our borehole tests.

We drilled a 34 foot borehole some 30 feet from the entrance of the CSM

experimental mine. Figure 23 shows a photograph of the entrance to the

experimental mine for the drilling operation underway over the mine. A map

and cross sectional view is shown on Figure 24. The borehole was in granite

and passed within 12 feet of the edge of the mine. The borehole was 15 feet

from the center of the rails at its closest point.

We ran a survey over the rails outside the CS~1 Mine with the portable

magnetometer. Figure 25 shows the magnetic profile along the rails. These

mine rails weigh 28 pounds per yard. The rail anomalies measure 1000 gamma

above the background field.

-30-

I w f-' I

- 57000y

APPROACHING

+

t ENGINE AT PORTAL

- 56000y t t CABOOSE LEAVES TUNNEL

TRAIN IN TUNNEL

I I I I I I I I I I I I I I I I 0 60 120 180 240 300

TIME IN SECONDS (ONE READING EVERY 10 SECONDS)

Figure 21. Magnetic anomalies from freight train passing through Carothers Tunnel, B&O Railroad.

Figure 22. Special sensor for borehole magnetomeLer,

·-32-

Figure 23. Drilling rig over Colorado School of Mines experimental mine.

-33-

MINE

RAILS

PROOF BOLTS AND

WIRE MESH )

0'

- -5'

- -10'

29'

- -25'

- -34'

Figure 24. Cross sectional view of borehole and CSM Experimental mine.

-34-

r 58000y

4-- ANOMALY DUE TO CAR DUMPER

Cl ...... ~ ....

I c.J w - 56000y V1 .... I 1.1,1

z c.:l C(

:IE

54000y

I I I I I I I I I I I 0 20 40 60 80 100 120 140 160 180 200

DISTANCE ALONG MINE RAILS IN FEET

Figure 25. Magnetic survey over mine rails at CSM Experimental mine.

The main anomaly (3000 gammas) is due to a mine car dumper, a 2 foot high

by 6 foot long iron hump along the inside rail designed to dump the cars over

the edge of the slag pile.

The borehole survey is shown in Figure 26. Repeated surveys all show

the same results. The main anomaly is not due to the rails. Rather it occurs

at the mine roof level and is due to the proximity of roof bolts, wire mesh,

and steel reinforcements. The readings at this level are always erratic

which is indicative of iron wire loops. The probable mechanism is that the

impulse field from the magnetometer sensor magnetizes these iron loops and

the loop residual magnetism varies from reading to reading.

Conclusions on the Magnetic Experiment

As a result of our experiments with the proton magnetometers we arrive

at the following conclusions:

1. The rail magnetic anomalies are the size expected but they oscillate.

2. The oscillation half-wavelength seems to correlate with the length

of the rail sections.

3. These oscillations further obscure already weak anomalies from the

rails in surveys over the railroad tunnels.

4. We tried computing gradients of the magnetic field in an effort to

enhance an anomaly .of one sign parallel to the rails. However the gradient

does not help to detect the rails appreciably.

5. Moving trains cause erratic readings of 1000 gammas or more and

are easy to detect on magnetometers even when they are placed 150 feet above

the rails.

6. Vehicles (buses, trucks) moved into or out of the vicinity of the

magnetometers, 100 feet or more away, give similar (20 gammas to 100 gammas

or more) changes to the background field. Such changes in the background

(vehicle removed from vicinity within 10 to 30 seconds) are easy to detect.

7. Nearby wires and loops of magnetic material show up easily (a few 100

to 1000 gammas or more) often by erratic behavior or repeated readings.

8. Borehole surveys show promise, but surface magnetometer surveys do

not, especially when the sensor to rail depth of burial is greater than 20

to 30 feet.

-36-

1-... ... I.&.

z a: Q Cl)

z ... Cl)

I.&.

Q

:::c: 1-g, ... Q

0 SURFACE

• X

"' 5 •X

J ~

\ 10 4(

/ K

I 15 )(

I •X (

20 .c

/ X •

25

30

35

55000

...............

)(

( ,. \ k

\

55200

X

55400 55600

•

• SENSOR DESCENDING

X SENSOR ASCENDING

READINGS 30 SECONDS APART

MINE LIGHTS OFF

55800 56000

MAGNETIC FIELD IN GAMMAS

56200

Figure 26. Borehole magnetometer survey at CSM Experimental mine.

-37-

ACTIVE SEISHIC METHODS

An active seismic experiment is one which uses dynamite or some other

source of compressional waves with a geophone and recording system listening

to the response of the medium. Such a system is in contrast to a passive

seismic system which detects and locates whatever is going on, be it earth­

quakes, quarry blasting, or tunneling activity. This study is concerned

only with the detection of tunnels by an active seismic system. The objective

is to test whether seismic exploration techniques can detect an underground

tunnel by reflection or refraction evidence.

Cook (1965) shows that underground cavities can be detected by seismic

reflections when the cavity dimensions (diameter of the tunnel) are the same

order of magnitude or larger than the wavelength of the reflected seismic

energy. Thus for tunnels in granite with P-wave velocities of 16000 feet/second

wavelengths of the order of the tunnel diameter (25 feet) would require

appreciable seismic energy at 640 Hz.

f v A

16000 ft/sec 25 ft/sec

640 Hz.

Watkins, et al. (1967) demonstrated another method of seismic detection

of near-surface cavities. This method, which detects oscillations of the

cavity walls on the geophones directly over the cavity, is based upon a

theoretical development by Biot (1952). According to Biot 1 s group velocity

curves, we have for a cylindrical hole in an infinite solid:

f = v

s 1.55D

where V is the shear-wave velocity, D is the diameter of the cylinder, and s

f the resonant frequency of the cavity wall oscillations. Our field experi-

ments were designed to test both methods for detecting tunnels.

We used a standard exploration seismic system owned by the Colorado

School of ~1ines and built by SIE. Figure 27 shows a photograph of the instru­

ment recording truck and the geophone implacements along the road. We used

-38-

Figure 27. Seismic recording truck of Colorado School of Mines.

-39-

both 6 geophones per channel and a single geophone per channel, 24 channels,

digital and analog tape recording, geophones of the Hall-Sears Junior style.

(See Appendix A for their characteristics.)

Sources used included dynamite caps and the Dinoseis, a truck-mounted

cylinder filled with propane and oxygen and exploded while in contact with

the ground. Figure 28 shows a photograph of the Dinoseis truck.

Other equipment and sources included the Bison hammer seismograph,

(Bison Manufacturing Company, Minneapolis, Minnesota). A sledge hammer blow

on the ground is recorded by a single geophone onto an oscilloscope with a

digital memory. Subsequent blows can be added (stacked) to the original

permitting shot-array procedures. In addition to the oscilloscope a paper

recorder can provide a permanent record of any desired seismogram showing on

the oscilloscope. Figure 29 shows a photograph of the Bison oscilloscope

and recording units.

The primary sites chosen were two tunnels of the DRGW Railroad near

Eldorado Springs, Colorado. They are single track, in granite, and possess

fairly easy equipment access over the tunnel. The road is 250 feet above

the Tunnel-17 and 120 feet above Tunnel 19. Figure 30 shows the seismic

profile over Tunnel 19; the setup over Tunnel 17 is similar. Railroad

tunnels are larger than the mines we are seeking. However, we expected con­

siderable difficulty in seeing the tunnels seismically because the seismic

wavelengths are so long compared to the cross section dimensions of any

tunnel. If the seismic method will not work on a railroad tunnel, we would

not expect it to work on a mine. Other advantages of the railroad tunnels

are that they exist singly and are well mapped. In contrast most of the mines

in Colorado exist in multiple shafts and drifts and are poorly mapped making a

controlled experiment difficult.

There are four Dinoseis records taken at each of the 24 geophone stations

plus six more beyond each end of the geophone spread. The four common shot­

point records were added. Figure 31 shows a few of these seismograms where

the Dinoseis sauce is near the center of the spread. The complete set of

these plots are reproduced by the Phoenix computer of the Seismograph Service

Corporation in Denver.

-40-

Figure 28. Dinoseis truck.

-41-

Figure 29. Recording units of the Bison hammer seismograph.

-42-

I ..,. w I

24 23

s ..e:::1.

11 10 9 8 7

120' GEOPHONE SPACING 20'

25' ~N

__L

Figure 30. Seismic profile over Tunnel 19, DRGW Railroad.

t::I'Tl :::0 1-'· CJOQ ::;:;c .., ::oro Pl 1-'·W 1-'1-' .., . 0 Pl P..CJ • (1)

::J rt (1) ..,

~ .;.i rt •

rt Pl ?' (1)

::J

0 < (1) .., >-3 c ::J ::J (1)

1-'

1-' -o

~~

t

. t·· . -~ ~ ttl ~ 11 f ~l~

··••••t! ·• Jtllt ....

..n1l!···'*+H~

1 1 t rlt l

.. ...

.. w

.. "'

.. 0

-l II l II t II

HI· 1l)!JlJ 1Jll_-~-_'ll-!1]·1llUIU .. J~-~--- -j.•··j·tti1·11l1!_

111 · jl···· ·•.1 ·r1r.~~ 0 • ),/ · I i ' (\ ', loJol\ '!I I )\ ... • · 't/.( 1 1 1 1

'"' )rp ' f/,' .lh tr."f.; ~ -:JII).

~ I-' 1 J ~ ij-\' , , l (, ' {-~~....-.)<) ( ( < , •rB. -~-. 1-~~ 1')\ . <.. !"" . ._.,,tJ ! ,)/ .... ~~,., o o t\ -· tr~:- )-:,;., ..• .l;tJ oil +- ({Hr+ .· -.·-·--- ~\(tt_t-1·-tfh·~-l····--- ..... ·f"~+h:- r i:i )-·1·-t_- -~. .- . ~-. \~_~, --,\u","ir~v_---. - ~~-----2~. U .~~~·~;•:;l:~i·~~:·(~ !{\(·qJ·J ~;~\~:~~~ ,'\}'}; .t~,-''jtl 1 !.~r~ )'\1) (l·:.~~~Hb )'"n;<:.~ ·. J}; g;o~ rl )i_hk(rt·~:=:)\.11U~~ \1.1,~)) '_.~~~~\\~~~(· .. , <:(dtL~~?,..-!t~\\\~: HJi{\( ;,~-~~~!{/ r.))(\(J-' · _ ·:~ t-'• 0. I-' • .., 0 Pl (") o.ro . ~

rt ro .., rn

I ro ~ rt v 0 I H1

t:1 t-'• ~ 0

~ ... . .... T ll r .l I lrl I' I I ' rl I lj II n I 'I ' ~ -. . ..... l.. . .. .. .. r ... ···)j . ... .. ' ). ·I) ........ ··r . .. . . \1111 . ... . 1-- --

; ·· ·••· · •· . •• ·· ·· ·• · ·· ···· · · ·· ·· ···• •· ·· · ·· ·· · .. ·· · ·· ··. •••· •• .. •• .. ·· 1 , ·· · ·· ·· 1 .... · lr~~~ ri rl 1111111111111111111 111 1111111111 J 111111111111 llllllllllllllllllllll 11111111111111111111111 mil rrmnn1nrnrr1

The Dinoseis uses two time breaks, one from the recording truck triggering

a second from the Dinoseis truch which actually fires the shot. The programs

in the Phoenix computer respond to the first time break whereas the first

arrivals are consistant with the second one. Since there can be 10 to 20

milliseconds variation between the two, static time corrections had to be

entered into the Phoenix computer to align the four common shot point records.

As a result of this adjustment the Phoenix computer has aligned the second

time break at the 0.1 second mark which becomes the time origin for these

plots.

In addition a time correction for weathering and the one-way travel time

to the top of the tunnel was applied to each of the 23 different shot-point

records along the spread. Then these 23 records with both static and dynamic

corrections were summed together yielding the single record playout shown on

Figure 32.

The objective of summing the 23 corrected records was to suppress coherent

noise and enhance any diffraction patterns emanating from the tunnel. If the

diffraction effect from the tunnel is present, it is not readily apparent.

Similar computations were applied to the data from the Tunnel 17 profile.

Figure 33 shows the center records with no corrections applied. The remaining

seismograms are shown in the Appendix. Again no reflection evidence of the

tunnel is apparent.

Two kinds of stacking procedures were applied to the Tunnel 17 data.

The first shown on Figure 34 is conventional where corrections for two-way

travel time are applied. The second on Figure 35 shows three records with

the static and dynamic corrections applied as for Tunnel 19. The center

record has been corrected to a one-way travel time where the tunnel is. The

other two records have been corrected to the top of a tunnel under one or the

other end of the profile where, of course, there is no tunnel. If this

stacking operation were helping, the center seismogram would look appreciably

different than the other two. It does not.

A major problem arises from the presence of a low velocity layer.

Figure 36 shows the geophone spread and two layer cross section over Tunnel 17.

-46-

HlD TF= 1

.... I);

1)··.

a .

.. ... . t1;

..... I);

.. 1) •

.... a.

. . r··· () .

Figure 32. Summation record over Tunnel 19 with static and dynamic corrections applied.

-47-

(1) ;:l rt (1)

" 'JJ ro rt

u H1

t::! f-'· ;:l 0 CJl

1(1) .p. f-'· OJUJ

I " (1) (l 0

" 0.. [f)

rt Ill ;>;" (1) ;:l

0 < (1)

" >-] c ;:l ;:l Q I-'

I-' \.0

"IlL

• ! 1;

.. I. • I . ~ ·, ~

.. ,· ..... ' ... i .. ! .

,::· ·---·······---

••• 1 •• L

--- •. ·-----1

., ...• ··: ··:··

- ··1---,-··i····'--··---·--·-·-j-··'--·--

:···i : ••• 1 ••• ,. , . r·

-1·-

, !-..... ,·

' ,: ' ' ;, ~ ~ ~

a · ·~ ':1, '~~· b; ·,~~ .:~ ~~ · ·'·b~ .:,,

Figure 34. Conventional stacking of Tunnel 17 data corrected for two-way travel times to tunnel. -49-

. :_,

-. ,·

.0

.6

'I

,;3 ........... +····~

.9

-t. ,..,t, -::rr i'l I

S4..,-/J uu;~J:. ""'-f"'o -1~~-ee.-di{fe ,.._,..-r r•-1! 1 dd;i: C ~,,.. .. ·c

l ' .

Figure 35. Three stacked records over Tunnel 17, assuming a tunnel at the south end, center, and north end respectively.

-50-

j

. ' ~ ( r ! 'It

1

r' /)! . ·' ( ••

I\ I I

. ·...:

·,r)

.... '7

8

g

7600

...I

~ 7500 1.1.1 ...I

cc 1.1.1 en 1.1.1

> 0 a:l cc z 7400 0

I­cc > 1.1.1 ...I 1.1.1

3 5 7 9 11 13 15 17 19 21 23

............... - v1 = 3350 ft./sec ·~ ..

-200

..._. ............. .

-100 0

TUNNEL 17 DRGW-RR

100

DISTANCE FROM PROFILE CENTER (FEET)

v2 = 15500 ft./sec

200

Figure 36. Seismic cross section over Tunnel 17, DRGW Railroad.

-51-

300

An analysis of first breaks shows that the upper layer has a velocity of

3350 feet/second whereas the lower medium with the tunnel has a velocity of

15,500 feet/second. Since the sine of the critical angle is v1;v

2, all

energy more than 12.5° from the vertical will be totally reflected. Thus

less than 14% (ratio of 12.5°/90°) of the energy from the source will pene­

trate the lower layer. The seismograms are showing mainly the energy trapped

in the upper layer.

Other negative evidence comes from Biot's (1952) relation for the

resonance of the cavity walls. With a shear velocity of 3800 meters/second

in granite and a tunnel diameter of 8 meters, the expected resonnant frequency,

f, of the tunnel walls would be:

f v

s 1. 55D

3800 m/sec 12 .. 4 m

307 Hz

which is significantly higher than the 50 Hz energy on the records.

Better evidence of tunnel reverberations showed up with the Bison equip­

ment over a coal mine. A coal seam at Walden, Colorado was being strip

mined by the Kerr Coal Company when one piece of equipment fell into an old

coal mine which was not known to be there. Bison records show the differences

over solid coal and over cavities (see Figure 37). The response over solid

coal is impulsive; the response over the stope reverberates for several cycles

longer.

We contoured the map in t11VO levels, cross hatching the area where

reverberation records were recorded and leaving blank the area where impulsive

records were recorded (see Figure 38). We postulate that the reverberations

lie over the route followed by the old coal mine. As of this writing the coal

strip mining has not proceeded far enough to verify the precise route of the

mine.

Conclusions on the Seismic Experiments

As a result of our seismic experiments we concluded:

1. No reflection evidence of the railroad tunnels showed up on our

records, whether the sources were Dinoseis, dynamite caps, or Bison hammer.

2. The stacking of two dozen seismograms to enhance diffraction patterns

from the top of the tunnel was not successful.

-52-

IMPULSIVE. OVER SOLID COAL

RINGING. OVER CAVITY

0

Figure 37. Typical Bison records over coal mine.

-53-

I IMPULSIVE\

SCALE: It---~ 10'

Figure 38. Map over coal seam with contours showing region of reverberation

seismograms.

-54-

3. No evidence of wall reverberations as reported by Watkins (1967)

appeared on the seismograms over the railroad tunnels.

4. The primary problem was a low velocity layer over the granite which

trapped most of the seismic energy. Hence very little energy penetrated the

granite layer which contained the tunnel.

5. Some evidence of tunnel wall reverberations appeared over a coal

mine. At this site although the tunnel is smaller, there was no low velocity

layer (coal outcropped and was being strip mined) and the P-wave velocity is

much lower than that for granite.

6. Seismic field experiments to detect tunnels by reverberations of the

walls should be conducted using multichannel (24) seismic systems rather than

single channel systems like the Bison. Neighboring traces from single channel

systems are often difficult to correlate.

-55-

ESTIMATION OF THE DETECTABILITY OF MINING ACTIVITIES BY

MONITORING AIRBORNE Rn222

The detection of a source of radioactive gas for given conditions of

atmospheric stabiilty depends primarily on the source strength (Q), the wind

velocity (u), and the limit of detection (x).

For a m1n1ng operation which advances 4 meters per day with a cross

section of 4m2

, Q is calculated as follows for granite based on the parameters

in Table 1. Metamorphic rock is assumed to be the same composition as granite

and basalt is omitted because it averages five times less uranium than granite.

Volume of rock mined, v

Uranium content for granite, u g

TABLE

16

25

4

1

3 m /day

X 10-6

X 10-6

6 or 16 x 10 cc

g U/g rock (upper range)

g U/g rock (average)

The specific activity of U, U s o.332 ~ci/g u238

~Ci 2.2 x 10 6 disintegrations per minute (dpm)

The density of granite, ~ 2.8.

S . R 222 · d h of u238 only d h h. h t · 1nce n 1s a aug ter an t e 1g es concentrat1on

that can exist in the activity in equilibrium with the parent isotope, the

Rn222

activity in a rock will be equal to the u238 activity of 0.332 ~Ci/g U.

Since U = 25 x 10-6 g U/g rock the total Rn222 activity = 8.3 x 10-6 ~Ci/g

g rock if the upper range of U in granite is chosen. The amount of rock mined

6 6 per day is Vp = 16 x 10 cc x 2.8 g/cc = 45 x 10 g. The source strength, Q,

is 8.3 x 10-6 vCi/g rock x 45 x 10 6 g rock/day or 375 ~Ci/day if we assume

that all of the available Rn222

escapes from the rock material.

Although the detection limit of Rn222 is very low in the laboratory, the

practical limit of detectability in the natural environment is limited by the

natural background which is omnipresent and quite variable depending mainly

on wind velocity, atmospheric stability, barometric pressure, and uranium

-56-

content of the terrain. -6 3 3 A concentration of x = 10 ~Ci/m (2.2 dpm/m ) would

probably not be detectable above background, but let us for purposes of

illustration take this level as the minimum detectable level (MDL). The

ground level (2 m) concentration of a-active aerosols (regarded as represen­

tative of Rn222

concentration) was found by Kirichenko (1970) to vary from

2 x 10-4 to 6 x 10-4 ~Ci/m3 and a typical value given by Rankama and Sahama

(1949) is 10-4 ~Ci/m3 , so the value of x = 10-6 ~Ci/m3 for background is

probably too low by a factor of 100 even if a signal to noise ratio of 1:1

could be accepted.

The areas over which a given source of radioactive gas is detectable have

been calculated for a variety of combinations of atmospheric stability, wind

velocity, detection limit, and source strength. The relation between detection

limit, source strength and wind velocity reduces to a single parameter

(x/Q) ~which has the units of ~Ci/m2 • For x = 10-6 ~Ci/m3 , Q = 375 ~Ci/d and a wind velocity of 1 mph (.44 m/sec):

X u = Q

10-6 ~Ci/m3 375 ~Ci/d (.44 m/sec) (86,400 sec/day)

The larger this quantity, the smaller will be the area of detection.

Therefore, an increase in wind velocity or detection limit, or a decrease in

source strength will decrease detectability.

The area of detectability calculated for moderately stable conditions - -4 -2 and (x/Q)u = 1 x 10 m gives a maximum range of detection of 1 km for a

stationary sampler directly downwind from the source.

Since we have taken the most optimistic parameters, i.e., moderately

stable atmospheric conditions, wind velocity of 1 mph, a signal to noise ratio 222 of less than 1, total release of all Rn in the rock, and a granite with

the upper range of uranium content, we can conclude that it is highly unlikely

that mining activities of the type contemplated can be detected at useful

d . b . . i b Rn 222 1stances y mon1tor1ng a r orne •

-57-

If the calculations were carried out using what might be termed a

"reasonable" set of parameters rather than the "conservative" values used v,1e

would have

and

u 4 wg/g so Q = 60 wCi/d g

-4 -6 3 x 10 wei rather than 10 wCi/m

u 5 mph rather than 1 mph.

(x/Q)u would be a factor of 3000 larger and the area of detectability would

actually be extremely small.

-58-

REFERENCES

Biot, M. A., 1952, "Propagation of elastic waves in a cylindrical bore

containing a fluid", Jour. Appl. Physics, Vol. 23, pp. 997-1005.

Breiner, s., 1973, "Applications manual for portable magnetometers" printed

by Geometries, Inc., 914 Industrial Ave., Palo Alto Calif. 94303, USA.

Cook, John c., Aug. 1965, "Seismic mapping of underground cavities using

reflection amplitudes"; Geophysics, Vol. 30, No. 4, pp. 527-538.

Godson, R. H. and Watkins, J. S., "Seismic resonance investigation of a near­

surface cavity in Anchor Reservoir, Wyoming", Bull. Assoc. of Engr 1 g.

Geologists, Vol. V, No. 1, pp. 27-36.

Kirichenko, L. V., 1970, "Radon Exhalation from Vast Areas according to Vertical

Distribution of its short-lived decay products", Jour. Geophys. Res.,

Vol. 75, No. 18, pp. 3639-3649.

Rankama, K. and Sahama, T. G., Geochemistry, the University of Chicago Press,

Chicago, 1949.

Watkins, J. s., Godson, R. H., and Watson, K., 1967, "Seismic detection of

near-surface cavities'', Professional Paper 559-A, U. S. Geological

Survey, pp. 1-12.

....59-

ACKNOWLEDGEMENT

The author gratefully acknowledges the help received in conducting this

study and writing the report. Dr. Donald Schutz of Teledyne Isotopes

conducted the study and wrote the section on Radon decay. Dr. Phillip Romig

of the Colorado School of Mines (CSM) directed the seismic field efforts and

acted as chief consultant throughout the project on the seismic and magnetic

methods. Mr. Jim Callaway of CSM did the weathering and derived the seismic

cross section as well as conducted the data processing at SSC's Phoenix

Computer. Mr. Steven Peterson of CSM measured the frequency responses of

the Hall-Sears geophones.

-60-

APPENDIX A

A-1

APPENDIX A

Response Curves for Geophones (Hall-Sears-Junior type) on the Shaking Table

Apparatus

1. Audio oscillator and power amplifier (GRC 1308-A)

2. Oscilloscope (Tektronix 561, with plug-in units 63 and 67)

3. Colorado School of Mines shaking table

4. Oscilloscope with transducer and strain gauge preamplifier

(Tektronix 531, with plug-in unit G)

5. One string of geophones (Hall-Sears-junior) with 6 geophones

per string.

Procedure

The shaking table was operated at a constant displacement for frequencies from 20- to 800-Hz using the apparatus described above and portrayed in Figure 2. The movement of the shake table is monitored by the Tektronix 531 oscilloscope. The voltage output from the geophones is measured by the Tektronix 561 oscilloscope. The peak-to-peak voltage output from the string of geophones was recorded at 20-Hz intervals.

Results

The frequency response of this geophone is shown on Figure A-1.

Harmonic distortion is observed for these geophones as follows: 1. From 280- to 340-Hz and from 500- to 600-Hz for a 1/8-micron

displacement.

2. From 500- to 600-Hz for a 1/4-micron displacement.

3. From 360- to 400-Hz for a 1/2-micron displacement.

The logarithmic frequency response is nonlinear above 300-Hz.

Preceding page blank A-2

100~----~--~--~~~~~~----~--~--~~~~~~----~

80

60

40

• • • 20 ••• •• • • •• • • • • • ' •• 10 •• •

Cl) 8 • 1- • ..... • Q 6 •• > ..... 5 •• ..... • :::E 4 •

• ..... • Cl) 3 z

Q • ~ Cl)

2 • ..... a: ::.:: • ct ..... ~ •

1.0

.8

.6

• . 4

.2

.1------_.--~--~_._.~~~------~----~----~~,._ ______ _ 10 20 40 60 80 100 200 400 600 1000 2000

800 FREQUENCY (Hz)

Figure A-1. Frequency response of the Hall-Sears junior geophone.

A-3

APPENDIX B

B-1

_;\PPENDIX B

Seismograms from north-south profiles over the Denver and Rio Grande

Western Railroad Tunnel 19 and Tunnel 17 near Eldorado Springs, Colorado.

Source: 4 Dinoseis shots at each geophone station

Geophones: Hall-Sears-junior

Geophone Spacing: 20 feet

Seismograph: SIE Digital System owned by Colorado School of Mines

Playout: from Phoenix computer of Seismograph Service Corporation,

Denver, Colorado.

Preceding page blank B-2

~8~!~~NTR" P•eT RUNNING SYSOOB:'f'S

•••• PH~ENIX LINE HEADER

XC

J0B ID N0. LHiE N0. NO 0F SAI1PLES SAMPLING PATE CHANNELS/GR0UP INPUT CREATED

INPUT REE• N0

H2 H

500 2

24 A00815

1905

f0RT~RN IU DN393944 REU.H STAPT G~IN <08) -~,.,

0

. ---------+----- --- ---·

~~~-----.----t

--.----------- -----

-""--~----~~---- -- ----

Figure point. at 0.1

B-1. Tunnel 17, DRGW RR- Sum of 4 Dinoseis records at each shot Static corrections added to align 1st breaks. Data time origin

seconds. B-3

'Tj f-'·

[]0 c 'i (1)

b:l I

f-'

(') 0 ::J rt f-'· ::J c (1) 0.. . H c ::J ::J (1) f-'

f-' -...J ~

b:l t:J I ~ _p.. 2

~ I

Cll c s 0 1-1'1

_p..

t:J f-'· :J 0 [JJ (1)

f-'· [JJ

'i (1) (')

0 'i 0.. [JJ

i

... _ ..

I I) f

! :

I II' I'

I ! J l

iII , I,

'' l it I i I i'

t·· l

' l i lj

i I ' it'' :I

i i I

i,: I' I I I I I :I I,

1 ; l I :

i i

!

I I II j 1 1 Ht

. ·.I i ! : : it: I l

I' I

i :lt I I I I

! ! j

; i t' II + L I

: I

11

1 II I'

II I t I'

iII 1 lj' . ! ' ' .

~ ( r I ,

I \ I\ 1 [, j II I t •·' I I

II II !T :I t:

i

~ ! ' ' ~ . I f I

i :I I i' ! j; II: i I!!! i I

l i -: ~ i . : ! f l I i J i .: j

! <; t ' ! ! I l !

I 1 j I I

1 I I

:;

! i

'

11 ·. ! I , I

\! l

I

i 1

.. '"%j ro

f-'· (]Q c t"i ro l:d I ·--···· I I I I 1 1 I I' T I 1 T I 1 , T T -- ,--,- "T" .,-.....,..-_.,. -'1 -.--...---.--.--...-....--... --.. _.., --+-+-- - •-+--4 -- -> -~---+ -o--1 + +--I -+ -+--1 -- 1--4 -6- ..... _. --------+--+·-<·-+-+·--- -~ ....... -.. .................... ~--1----·

1-'

(l

0 :::: rt f-'· :::: c ro p..

H c :::: :::: ro 1-'

1-' -....! ~

l:d I::!

I :;o

\.J1 ~

~ I

en c s 0 ...., ~

I::! f-'· :::: 0 en ro f-'• en t"i ro (l 0 t"i p.. u~ .

Figure B-1 continued. Tunnel 17, DRGW RR- Sum of 4 Dinoseis records.

B-6

•••• PH0ENIX ~I~E J0B ID N8. LINE Nl!l, N0 0F SAMPLES SAMPLING RATE CHANNELS/6R:8UP INPUT CREATED

HEf'IOEft ••• 14

1 500

2 24

SVS2.00A

INPUT REEL N0 1099

XC FI!IRTRAN IU ON393944 REU.H START GAIN (08) -33

c

3 • • •

'. ·: ~--+-

-. '

Figure point. at 0.1

B-2. Tunnel 19, DRGW RR- Sum of 4 Dinoseis records at each shot Static corrections added to align 1st breaks. Data time origin.

seconds. B-7

'"7j f-'·

(JQ

c '"' (D

tJj I

N

n 0 :J rt f-'· :J c (D p. . >--3 c :J !:l (D

f-'

f-' '-0 tJj ~

I CIJ t::::J

~ 0 ::8

~ I

(/)

c s 0 H;

.j::--

t::::J f-'· :J 0 Ul (D

f-'· Ul

'"' (D

n 0 '1 p. Ul .

1 1 Ji 1· t 1 i. 11 .. l! 1 ~. (0 1 f j' · . (·.. ." ... '·. 1 · ··· ·· ··1··.··1-·1-1 .. t r r.·.trmTrf. J 1 1 11 1 11 1 11 1 1111 1111 1 ., -d I Ill' I . . II .. . . II ........ .

' j llll,t ' .·. .... ' ''·'' ~ • ( l. '.· ''.· .~ . • ..,.· \' · ... :· ~ \). ' ' ~·). . ' • 1 I .... ' • ) ~( ' I ... . . . . '. ' '·. ' -·~-

{~ .· \ ... ~. .~. ·. ' ... · ... · . · ..... • ·. · .. • ) ~. ' \~ {! J r . J( ~ J ~ . t \ j ... ·. . ; :'!J'.· .. ·.·. ·~~.'') .• . . . ) .•. ·.·• ; . , ,~>)::, .. · • ), ~ ".,. t· ... ,. jl\ .... '{?· '<lj i ' ,, ... ·-~-~~·· ''\ .. \ \ . ' -': ....

\ ... ( ). ~t!. . . ' p i I \ ( j I~ ! ·~· ..,. t II 1( } (~ ··. ' •\ ' . ? l l ' t' ~·. ·~· ..... .... f ·. , " I · , ( \ ~ I · 1 I · · ,. ' .1 .. ·.• ' · · ·· · '< • '· ·- • ·"' •

. ~~ t .. }\1\·l.· ~·~.·.t.+~~.~.~ ·; \). i~ttl.s ·xtf.ltJt j\~L \ ~~~~~ ~ '."f;p ,~:· .. ,Jr:. t ?t 1t. {~.). . )dt\~)(i~t!t'~: !(~-tH~· \i/))1( 1:~~}1!Y [{).t\pq I ih{L\' I'~~ )h ;(~\ ~~ {1 ),J f\c r~> ~t\) / '! ( ~. . . ~- •.. .... I ,. , , I '.. :k . c, I t , ·~· ) ,, ,. J, , ·t• . . . .. .. t .. .. :·, .. ..,... .

qli')'@!Jfu~~~~~ 111 1 (\ n H 1\! :v1i !l,~li!h /m)ll\ilA)I \t~IH 1(1 )~li IW?m~ ~~H L f \)\, "' 'i r· '\ 1.1 (\\'\I 'I I; I I}'· 1. IT I ,,.~\I; nhr.r r \I j'('(l \r;t'"~(t·"j''\1 ,.l\1'1"11'/ rltf1·rl· ·.·}\( "cf.-·\y~·.t+L</1 : I f I ' ( I r I r I \ I' I I ' . I \ I I I I I \ ' }( I 1 '\ ) J I [ t ) I ' I I ' I I I . I. . • \ )J j I It~+ I . \ I I I ' i \ ) )) q \? /.( / .I\ j !I J? \ q 1 ! .~ \.! l 1/\ I \ \

1

1 /.\ ( } I ( l ~~~) fJ ' i )r' \.) ~I~.))~\'.] ( I I (, ) L1

) ) ~ 1 .. · . ~ I,. I\ ll \) i 1

I) ( ( 1 . ) ll ) , \1 i 11 1 \. I,

111l ·' 1 ) il ( 1 ) ~ \ : ,\ r 1 ) P ) 1, 11 \ \ n 1 1 11~ j . l, J ~ . 1 . 1 \ 11! ()\ 1L I L 11 1l ,,rl,, tl 'l,t JLJ rJ),1!t 1\IH l1 , 1}\i \I )IJIII\1) 11111 \ · J \ ) ,[it ill

i)\l\lll)j

11

~ ~~J1

1 j W~\lllll\ J/11

( ~~~)/lJ) lli 1\11~111//\) \l)i. f f!

\ I I 1[11 11

(i\) ljrll 11

lr)l/1j( /ll

11(llllf H\ f)jll! lliJlJJIJII.-l! I' I II I ill i !I I . . . hL I llhll f I 11

.l/

~~~· ~-pi ~~~~~! I·

~. I \ l I i I ' Ill\ \I\ ) I .. ( ) 11 I I ( I I ) . . . l I ' I . ; I I \ . 1

L ! 1

1 ! lrl I! 111!\ · HI I \ Ill Jll I

1-:tj 1-'·

(lG (: t1 ro 0::1 I

N

(l

0 ::J rt 1-'· ::J (: ro 0.. . H § ::J ro I-'

I-' \0 ~

tJ 0::1 :::0 I

~ \0

~ I

(Jl

~ 0 H1

~

tJ 1-'· ::J 0 1:/l ro 1-'· 1:/l

t1 ro (l

0 t1 0.. 1:/l .

'.

.. .. "'

_·.1\_.-·-·~·~-~- -~( __ ·· __ ·_··_ · .. _ ---~ .... · ·_. ··.·· "'-'·~·~---··············•• _•_--_(·_.···,._·-~· . ........ _~., .... _.-- .. -....... _·-•_· ...... ·. ~n , . \ J . ~ . ~ . . . ~ . . ) ,. ~,. .. . , . . ,. • . . ~F-~ 'I. . · .. ·• . - -~ . ..... . . j' .. '• ( . . / • ' ). . -~ . .. -~ /' ) . -~ . .. --"""ii .... - l ..•. ·-.~ +'_-_ .. - .· ... -'.~- •. . J.~-)· ·'• . ~---.~_··. -------.. --.Jl1'1····)· ···. ,tJ __ :)o ...... __ - ... _. --. -. ' • . .. r 1 -- .... ~- t r• ... (. " . ,.). . .. . .. ~{·x~\ . ) ',!~~~-.......... 1-.a.l ··. \ .... ~~· '"r I\' ) '"'" ~ .. 1f"' N I\~'-;~ .. ·- r~. _J..... . . . .-~~l~~j~t.;r.~r)(\'Y~ lni\1it~--~>At Nh\\)1·_ -____ -- _ ·_----.~ ~~~r(Nr.~~""u·.

{ (I ( \t~·( rK( ... t . \I[ \I) l 'Ul-~ y·u 'J r ! \ . · ..... ·. . . .. .. . .. . . ' . l jl~~(llWH_ ·, H( ll, ~I I! ~~1 ~'~'';h)~ tl! J~ < 1\ I ~J\1m. ~L,~{HJ i ntrm 6·"""\lll

I i!\(ldi 11)i 1 I(! I I \ i I . I ! i ( ~ \ j I I I - · H_-- \ + -- +· L -- + -- - + 1 I l I i I /I r I n d \I ·

1111 (/\\''\

11 Ill ·~ \ 1·~~-· ! ((l!\)(_"(

11

jll[fi .

1-.

{ I I p I I / I -~ ) ' ( I r ) ) l ( f , \

Ji!r• •••••.••..•••• + ••••. I.... I ·t)hj· ··l' .. . ......• (1 ...... -r .. Jf I !jl1rj· ·rtrnr mnnmTillr ...... ..

1··! r ---- -{ l------ ----- -- -------- f ----- - rl ~~~ r J J 1 --- L.... . ..................... ..

..................... T 11- ---- ------ -~- -- _-- ----- lilt! I --~ I lj --- -- ruuJHIJL ... :--·--·--HH I I II l ·- .. .. lit l I l . I lJ lll UJIII L ... lllfll!ll!lllfllllllll ll!IIIIIIIIIJIIIII!illll 11!{ llillllllllllllllll I ffllfltllilllllfll!l

-~-+---~- -~-~- ............. -~~-· ---- ------~--

Figure B-2 continued. Tunnel 19, DRGW RR- Sum of 4 Dinoseis records.

B-10

:::---

Figure B-2 continued. Tunnel 19, DRGW RR- Sum of 4 Dinoseis records.

B-11

td I

I-' N

>'%j f-'·

(]Q

c ., ro td I

N

() 0 ::s rt f-'· ::s c ro p_. . H c ::s ::s ro I-'

I-' \0 ~

t:J

2 ~

Ul

§ 0 H)

..,.. t:J f-'· ::s 0 CfJ

ro f-'· CfJ

., ro ()

0 ., p_. CfJ .

"'

1

1

. ·.r..

1-. .rn l

l 1' 1

f-. -t- .. 1': ·_ -. -.- -- -- -_ - l \

ro rv ...

I .... \\l[f\1 l . t '~!~llt/J'f\'[1 - , 1 •t' \ \ ~: -~( I ... •- '' 1 1''\jl ' ,)~ \)IJ> \ [1 , ,. ,. · 1 11 \ I. --.~<'\ 1 •\t,.t (J I · 2 ,~·~:l~\q':~ -) ~-q ~~ t~H {, (~ '•rli U jl ·. ~~~;..-(r~lt.--~rt?J\(H!U1-. '~t Jf ) ' \ ( ) \ II II ' I ' " .. { \ ( ) L• I ) I 1., I ljl I ) \' ( t . .. I I II \ \II ' ( I I \ I I 1\

.·~;~-~~ ~.{ 'l·,· :;,~~~ 1)1 t ~ r\1 !' ~ UH~~~ l t \ ~~ \! ::~: H h II':: ll' ? ~ lr :~ ~t-~~~: ~) nt~~ ~! !,1 \1' H ll I

I I --~~ \( ~-- l ( [ ~ \ I - ' I t I ! ( ' ~ ~ i t I I, I i I \ I ' I I I I \ ( ) ' l I ' I ! .. '

.3 ) I~~ )t)~- }_I,\ P>)(~~· >~s ~- )I~\ 11 r i~\ ~ 11\(f tl 'll\ t I'\ ------t 1 J_j_ll_iJ{. {-1~-(~-;.L\~·-rt-~-t -1.---'\ ~ 1 , 1 q r~, .. )\1 1 }r~)-,i~ (t\ 1 \' )il)l\1 \1' •,: 1 t'J (li\1,111 1

11 1111

', ) I t \ !\ ...._< t ) \ I ( I l ) ~ I I I ' ? I \ (1 \ ( \ ' I \ I, II I I i l . I \ I ' I 1 '~ I ! ~ \ ) I I i I ' \ \, i'

) I ( P\ 1 ~~ I ~.1) \~J ~ ~: ·; ~ ~~- ~>>: ~~ ; 1 !1 r 1 / 11 { ~: u 1 ~: ~~ q 11 :- ~ ~~ 1 1.ll :· 1 r l J\ (. ;d ,: : w !. 1

1·: !: ~:~~ n ~~ ~~

/•1 \)\)ft J ('\) 11

]1

1 t(llj\ \ 11'

11 '1! 1 1''11! 11

11/\jf' \II" . 4 l ll \\ ( \ ', 1 I \ l, l \ \ \ t t t I 1 \ I \ \ j • ) \ t I \ ~ . 1 . Ill 1 I 1 I 1 ' I \ .. + ~) .J q L)

1

1 1il1' ~~~W )ril ll l'lr \11 llli!f'llllt~l~(~j~ lltW. llj1il(flt{!l1

1

1

1

l\

1

1i

I } I ! \ \ j I I \ \ ·I I \ .. I \ I( I t I l ,_< l I I l I t .

, , , I , I t l f I' ! II : \ l r )\Ill I I Ti I P tnn , lll

1 •'

! . '

' z ' ' 7 •. \ ) \ l. \I

, . '"~~~~~~11·. 1 · ....•• · 1 lUtJHl L 11! . !lr"," 11 ~ . l l

• ~ ) l l I \ • "

llltlll , II 1, t rr: II I , •

.7

,9

IIIIII!Tit ··~1ll1l··~·~--~· ;~······