R D-A128 514 ACOUSTIC RAY TRACING AND THREE-DIMENSIONL PROPAGATION i/i

LOSS REPRESENTATIO..(U) NAVAL OCEANOGRAPHIC OFFICE NSTL

U LAS STATION HS E KHEDOURI ET AL. SEP 77 NCO-TN-3709-55-76

UNCLASSIFIED F/G 20/1 N

Ej

If , _ .* -!s ! -- l - . .. , ,. .. .. -.- • .-. . .. .

I,.

13L2

*140

j 1.8

11111.251

MICROCOPY RESOLUTION TEST CHARTNATIONAL BUREAU OF STANDARDS-1963-A

A.-

SF-ZJRITY CLkAkFhICATIOIN 00 TH1% PA',E. Lefe U i.j

*REPORT DOUAENTAT04. PAG iR$ Jk D ; I

TN4 37010-55- 76

T IrTLE (mw S.Ort I*) t t-' F.ir ~A *3Jst ic R;-iy c g dTr-DeLul1j ..

jin L)S s Rpe sntto e r. t i j i

9 PEFORINGOftAN:IATION NAME~ ANDO ACCR

~~ ~U.S. N.Va1 (Jceancbratrhic ffc.AR&8V'<u': ,rn'

NSTL Station-Ey "-t. Louis. MS 3252.2?lCONTROLLANG OFFICZ NAME AND ADOOIZS [,~ A

ONIVOR14G Ac.EW'4 NAME A AODDRESS(it diII.?wnt Poor C£wotilmj 01 0,, 6) S ' r -e: C.L .

i: 1 L f

I. STWSuTiom IIATie#CIT ()I ihi Rep ort)

*3.0V, r r ibi i re i 1- s sLriiDut i' mi If I

1 7 RI IUT7,SN STA T~M~ 6 IA71. t a~r~ C i 'c 0 t~ 5., S.,

it SPL EMEN, ARY 'iO'ES

CI-i

21, ABSTR~ACT (Foo.In @14 rs"8t~s jiO If r4ScoommV m~d Itivnify by block numbtit).c~i~.tr rjy Irices arnd irtcerisll.ies for f l*x,.i add~rai (~v' I.

W re c ~ided for fIhre co-eet ~ tk-UI .~

I. Li ta2prtAe orcr1,'r. an,~~3OV3 vCJI ; '/EV71 dat.tO \ (

4 ~o ':(-a due to .(w ik d' 1.:. f 1*C!' ( ' L

C-- I , C1S tile r-Alf o -,,, r) re nt isit'

DD17 4r-o o sosiE

S9CUNITY CI.ASSIFICArION OF THIS PAGIE (Wh1e. Z.i4fem

, without a front , it was found that the Gulf Stream Nort!, '! 7 -ouLd etY* increase or decrease propagation loss depending on distance ot the

receiver from the front; the maximum difference was l-2 di;. ITe -outh I :of the Stream had no significant effect on sound intensity.

Three-dimensional propagation loss diagrams show that e~ffcts of th,North Wall on propagation loss persist at all receiver depths from ,asurface to 1000 meters.

.4

N 0102- LF-O 1A-.6601

SECURITY CLASSIFICATION LF THIS PAGErWMefl 0090 91110fd)

.1

NAVOCEANO IN 3700 - 55 -76

ACOUSTIC RAY TRACINGAND

THREE-DIMENSIONAL PROPAGATION LOSSREPRESENTATION IN THE GULF STREAM REGION

by

E. KHEDOURI

and

P. GABORSKI

SEPTEMBER 1977

U. S. NAVAL OCEANOGRAPHIC OFFICEWASHINGTON, D. C. 20373

Approved for pusblic r,,ise;distribution unlimited

l~8

NAVOCEANO TN 3700-55-76

TABLE OF CONTENTS

. ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . .

I. INTRODUCTION ...... ..................... ..

II. METHOD ........... ........................ 1

III. SOUND VELOCITY PROFILES, RAY TRACES AND PROPAGATION

LOSS . . . . . . . . . . . . . . . . . . . . . . . . 3

IV. 3-D PROPAGATION LOSS REPRESENTATION .... ......... 5

V. CONCLUSIONS .......... .. ................... 7

ACKNOWLEDGEMENTS ........... ...................... 9

REFERENCES ............ ......................... 10

FIGURES

Figure 1 - Surface Thermal Structure and Location of AXBTProfiles ...... .................... . .11

Figure 2 - Sound Velocity Profiles for Section A ..... . 12

Figure 3 - Sound Velocity Profiles for Section B ..... 12

Figure 4 - Sound Velocity Profiles for Section C ..... . 13

Figure 5 - Ray Trace and Propagation Loss for Section A 14

Figure 6 - Ray Trace and Propagation Loss for Section B 15

Figure 7 - Ray Trace and Propagation Loss for Section C. 6 ISlope to Sargasso Water ... ............ . 16

IFigure 8 - Ray Trace and Propagation Loss for Section C. A

Sargasso to Slope Water. .............. 17 F-e.~,.

%1 Figure 9 - 3-D Propagation Loss for Section A ......... .. 18 D-i c -

Figure 10 - 3-D Propagation Loss for Section B. .. ....... 19 st. to

Figure 11 - 3-D) Propagation Loss for Section C. Slope to BSargasso Water ..... ................. .. 20

Figure 12- 3-D Propagation Loss for Section C.Sargasso toA -al._,bljt, e.

Slope Water ...... .................. . 21 • *, an-

NAVOCE.ANO TIN 3700-55-76

ABSTRACT

Acoustic ray traces and intelisities for lixed and variable rect,iver

depths were calculated for three cross sections in the Gulf Stream region

using AXBT temperature profiles and historical .ound velocity data. Two

sections were in the Sargasso Sea and one crossed the Gulf Stream. it is

shown that for shallow source and receiver depths, a 4 dB difference in

intensity is found in the Sargasso Sea due to sonic layer depth differences.

When sound intensities across the Gulf Stream were compared to intensities

for similar situation without a front, it was found that the Gulf Stream

North Wall could either increase or decrease propagation loss depending on

distance of the receiver from the front; the maximum difference was 18 dB.

The South Wall of the Stream had no significant effect on sound int-.isity.

Three-dimensional propagation loss diagrams show that effects of the

North Wall on propagation loss persist at all receiver depths from sea

surface to 1000 meters.

..

- ---'- ,--

I. INTRODUCTION

Effects of oceanic fronts and eddies on acoustic sound propagation based

on measured or calculated values have been documented in several publications

(Gemmill, 1974; Gemmill and Khedouri, 1974; Levenson and Doblar, 1976).

There is little doubt that a strong front such as the Gulf Stream has a

significant effect on sound propagation. A front can either increase or

decrease transmission loss depending on the distance of the source or receiver

from the front, and the magnitude of transmission loss depends on factors

such as bottom topography, bearing angle of sound transmission to the front

and the depth of the source or receiver. Direct measurement of sound intensity

across a front at various range and depth intervals is impractical. However,

computer simulation and three-dimensional representation of propagation loss

can be used to determine intensity variation for a given cross section as a

function of range and depth.

The objective of this paper is to show theoretical sound paths and

propagation loss values for three cross sections in the Gulf Stream region

using methods which are adaptable to operational use. Figure 1 shows the

location of the cross sections and the surface thermal structure of the area.

Sections A and B are in Sargasso water and section C crosses the Gulf Stream.

The data for this report consist of airborne expendable bathythermograph

(AXBT) temperature profiles taken during February 1976, combined with deep

historical sound velocity data.

II. METHOD

In frontal regions where horizontal temperature gradients exist along a

given cross section, a multiple profile model (range dependent environment)

must be used for ray tracing and propagation loss calculations in order to

I

reflect the sound velocity changes along the cross section. NRL GRASS model

(Cornyn, 1973) was used for all the acoustic calculations in this report.

This is a multiple profile and variable bottom depth experimental model

used primarily for research purposes. The model was significantly modified

* I by the authors to make it more suitable for operational use.

The NRI. version of GRASS requires either sound velocity vs depth, or

salinity and temperature vs depth as input. The program automatically

extrapolates sound velocity profiles to the ocean floor. This extrapolation

can result in erroneous sound velocity values when the bottom is considerably

deeper than the deepest available data point and consequently, in deep

waters, relatively deep sound velocity data are required. This type of

data, however, is not readily available operationally. One of the

modifications made to GRASS was to combine it with the ICAPS merge program

(Ranesen and Tucker, 1975). This program merges temperatures from AXBT

traces or similar near-surface temperature data, with mean seasonal temperature

and salinity data from historical files for a given geographic area (5 degree

square), to create sound velocity profiles from sea surface to the ocean floor.

This modification made it possible to use AXBTs or similar readily available

temperature data as input to the program. A variation of the above technique

is to merge the AXET temperatures with historical temperatures and salinities

for a particular water mass (Fisher, 1976). This method uses the near-surface

temperature characteristics to select the proper water mass. The method gives

superior results in frontal regions because oceanic fronts are not constrained

* by fixed geographical boundaries. The water mass concept was used in this

report for determining sound velocity profiles from AXBT temperature profiles.

2

-'- -r 'V.-v - -vj. '- - - - S - -r-. -

Another modification made to GRASS was the use of three-dimensional plots

to display propagation loss vs range vs receiver depth. This modification

allows the propagation loss at all receiver depths (1-1000 mn) to be displayed

in one diagram which shows the changes in propagation loss as receiver depth

is varied.

III. SOUND VELOCITY PROFILES, RAY TRACES AND PROPAGATION LOSS

Sound velocity profiles used for ray tracing and propagation loss curves

for sections A, B, and C are shown in figures 2, 3 and 4 respectively.

Profiles used for sections A and B are in Sargasso Water. Profiles used for

Section C are in Slope, Gulf Stream, and Sargasso waters.

Ray traces and transmission loss curves for sections A and B are shown

in figures 5 and 6. In both cases a 290 Hz source at 137 mn depth was used

wit an 88 in deep receiver. Comparison of ray traces shows that in section A

most of the transmission Is through bottom bounce and deep sound channel,

whereas in section B, considerable amount of transmission is through the

near-surface duct. Inspection of the sound velocity profiles used in these

sections (figures 2 and 3) reveals that the only difference is in the near-

surface sound velocity and in the sonic layer depth (SLD). In section A,

the SLD was 240 mn at the source and 105 in at the receiver, whereas in section

B, the SLD was 200 m both at the source and the receiver, resulting in well

* defined surface ducting. This surface duct transmission resulted in 4 dB

higher sound level at 50 nmi(92.6 kin) range.

Ray traces for section C, crossing the Stream from Slope to Sargasso

Water, with a source at 137 m depth are shown in figure 7. Sound velocity

profiles used for this section (figure 4) show that in addition to near-

3

surface sound velocity diffences, the depth of the deep sound channel axis

(DSC) deepens from 700 m in Slope Water to about 1200 m in the Gulf Stream

and 1400 mn in the Sargasso Sea. En addition, sound velocity in the upper

layer of the sound channel is higher in the Stream and Sargasso than in the

Slope Water. Deepening of the sound channel and the increasing sound velocity

across the Stream are reflected in the ray traces, where it can be seen that

after the rays cross the North Wall of the stream, they are channeled downward

into the deep sound channel. Similar results were shown by Germnill (1974).

Propagation loss curves in figure 7 show the intensities for a 290 Hz

source at 137 mn and a. receiver depth of 88 m. The solid line shows the

intensities from the eight profiles used in the ray trace, and the dashed line

shows the intensities using only the first Slope Water profile. The difference

between the two curves, therefore, represents the difference caused by the

fronts. When the North Wall is crossed, the sound rays become out of phase

and the convergence zones which are typical for a single water mass are

destroyed. The resultant effect on intensities is seen by the significant

difference between the two curves when the North Wall is crossed. The

maximum difference of 11 dB occurs at 90 mud (166.7 kin) range.

Figure 8 shows the ray traces and propagation loss curves for Sargasso

to Slope transmission across the Stream using sound velocity profiles shown

in figure 4. Ray traces were calculated for a 137 mn source depth. When

the South Wall of the Stream is crossed the surface bounce propagation is

terminated. This is because the SLD is considerably shallower in the Stream

than it is in Sargasso Sea as shown in the sound velocity profiles. When

* sound travels across the North Wall, convergence zones become reinforced with

significant effect on propagation loss values. Propagation loss curves in

figure 8 were calculated for a 290 Hz source at 137 mn and a receiver depth

4

of 88 m. The solid line represents the propagation loss calculated from

1 6. the eight profiles used in the ray traces and the dashed line shows the loss

for identical conditions using the first Sargasso profile. The difference

between the curves, therefore, is caused by the fronts. It can be seen

that the South Wall of the Stream has only a minimal effect (4 dB) on

propagation; the North Wall of the Stream, however, has a very significant

effect which is caused by reinforcement of convergence zones. For example,

at 130 nmi (241 kin) range, transmission is decreased by 14 dB, but at 135 ni

range, transmission is improved by 8 dB because of the front.

IV. 3-D PROPAGATION LOSS REPRESENTATION

Three-dimensional propagation loss diagrams serve primarily as visual

aids to show the variation of intensity at several receiver depths. Exact

values of propagation loss, for any receiver depth and range, however, can

* also be determined from the diagrams.

The following example, using figure 9, will illustrate how to read the

propagation loss value at any point. To determine the propagation loss

value at 40 nmi (74 kcm) range and 200 meter receiver depth, the following

procedure can be used:

1) Find the propagation loss vs depth line at 40 nini range (see arrow in

figure 9).

2) Find the propagation loss vs range line at receiver depth of 200 mn.

3) Find the intersection of the two lines.

4) Draw a line from this intersection parallel to the range axis and

equal in length to 40 nmi on the range axis scale. The resulting propagation

loss value of -88 dB can be read on the propagation loss plane. In cases of

5

complex sound velocity structure, parts of the propagation lass surface

may be hidden and it may not be possible to read some points from the

diagrams. In these cases, a new viewpoint can be selected to display

hidden regions of interest.

The 3-D propagation loss diagrams for sections A and B are shown in

figures 9 and 10 respectively. The two diagrams are similar except for

the near-surface values, caused by the near-surface sound duct as described

in Section III. In both cross sections, which are in the Sargasso Sea, there

is a significant decrease in propagation loss between the su-'ace receiver and

a receiver at 25 m; below 25 m propagation loss vs depth is arly constant.

(The abrupt change at 25 m has no physical significance; it caused by

computational procedures where receiver depth was increment- . 25 m intervals.

The 3-D plots of propagation loss across the Stream for section C, from

Slope to Sargasso and from Sargasso to Slope, are shown in figures 11 and 12

respectively. When sound travels from cold Slope Water to warmer Gulf Stream

and Sargasso Water the convergence zones become weaker as sound is deflected

into the deep sound channel (see the ray traces in figure 7). The effects of

weaker convergence zones are reflected in the 3-D plot of propagation loss

in figure 11, and it can be seen that the effects of the front persist at all

receiver depths from sea surface to 1000 m. When sound travels from

Sargasso to Slope Water, the convergence zones are reinforced as can be

seen in the ray trace in figure 8 and the resultant 3-D propagation loss

plot in figure 12. Effects of the front are again evident at all receiver

d depths.

Note that reciprocity of sound pressure level exists because the ocean

system is linear with respect to the pressure levels encountered in acoustic

surveillance. Therefore, identical propagation loss curves would be obtained

if the source and receiver locations were interchanged. In the case of 3-D

6

diagrams, this means that identical propagation loss surfaces would be

obtained for variable source levels and a single receiver depth as for single

source depth and variable receiver depth. The 3-D diagrams in figures 9 to

12, therefoi.e, can also be used for receiver depth of 137 m and variable

* source levels from 1-1000 m. In order for reciprocity to strictly hold, all

possible rays must be used in intensity calculations. In the above figures

an envelope of rays from +200 to -2Oo from the horizontal at 10 increments

was used, and rays were terminated after 5 bottom bounces and 10 surface

I bounces. This has little effect on propagation loss values and reciprority,

but provides considerably clearer ray diagrams

V. CONCLUSIONS

1) For shallow source and receiver depths,* there was a 4 dB difference

in intensity at 50 nmi (92.6 kmn) range between the two cross sections in

the Sargasso Sea. The difference was attributed to a deeper SLD in one of

the sound velocity profiles which resulted in r.ows being trapped in the

surface duct.

2) When sound is propagated from Slope Water across the Gulf Stream into

Sargasso Sea, recurrent convergence zones which are'typical for a single

water mass are destroyed. When sound is propagated from the Sargasso Sea,

across the Stream, into the Slope Water, the convergence zones are reinforced.

This has a significant effect on propagation loss values.

3) When propagation loss for shallow stource and receiver depths calculated

for a single Slope Water profile is compared to propagation loss values

calculated from multiple profiles across the Gulf Stream under similar

conditions, a maximum of 11 dB difference occurs. When propagation loss is

calculated from a single Sargasso Sea profile and compared with propagation

loss values using multiple profiles across the Gulf Stream, a maximum difference

7

of 18 dB results. In both cases the difference is caused by the Gulf Stream

North Wall.

4) The effects of the North Wall on propagation loss are evident at all

receiver depths from sea surface to 1000 meters.

* ..

!8

I

ACKN OWLEDGEMENTS

The authors acknowledge the efforts of Mr. Mitchell K. Shank, Jr., for

his administrative and technical guidance; Mr. Rudolph Michlik for his

assistance in developing the 3-D plots; Mr. Glen Voorheis, Mr. Randy Doblar,

and Ms. Kay Collins for their help in preparation of the manuscript.

- .9

i9

REFERENCES

Cornyn, John J.,(1973) GRASS: A Digital Computer Ray-Tracing andTransmission Loss Predicted Model. Vol. I, Overall Description,NRL Report 7642.

Cornyn, John J., (1973) GRASS: A Digital Computer Ray Tracing andTransmission Loss Prediction Model. Vol. II, Users Manual, NRLReport 7642.

Fisher A., (1976) Selection of Water Mass History from BathythermogramCharacteristics. U.S. Naval Oceanographic Office TN 3700-51-76.

Ge-,nill, W.H. and E. Khedouri (1974) A Note on Sound Ray Tracing Througha Gulf Stream Eddy in the Sargasso Sea. U.S. Naval OceanographicOffice TN 6150-21-74.

Gemmill, W.H., (1974) A Note on Sound Ray Tracing Across the Gulf Stream.U.S. Naval Oceanographic Office TN 6150-27-74.

Hanssen, G.L. and W.B. Tucker, III. (1975) Interim Integrated CarrierAnti-Submarine Warfare Prediction System (ICAPS) manual, (CP 642-BComputer System) (U), U.S. Naval Oceanographic Office, ReferencePublication 10, Washington, 100 pp, (CONFIDENTIAL).

Levenson, C. and R.A. Doblar (1976) Long-range Acoustic PropagationThrough the Gulf Stream, Acoustical Society of America, 59 (5), 1134-1141.

1

p °,0

38N

A LOCATION OF AXBT PROFILES

19

S 20

211

36 1

20 Setion0

211

~~. 21.22d

VELOCITY (m/sec) VELOCITY (m/sec)

1400 1500 1600 1400 1500 1600b•.:oI I

0- 0-

1000 1 1000 2

Sargasso Sargasso

SLD = 240 m SLO =105, m

2000 DSC =1400 m 2000 =1400m

3000 o 3000

4000 4000

5000 - 5000 -

Figure 2 - Sound Velocity Profiles for Section A

VELOCITY (m/sec) VELOCITY (m/sec)

1400 1500 1600 1400 1500 1600

0 0

1000 3 1000 4

Sargasso Sargasso

SLD = 200 m SLD = 200 m2000 DSC =140 m 2000 DSC = 1400 m

• E

' 3000 -' 3000

4000 4000

5000 - 5000

Figure 3 - Sound Velocity Profiles for Section B

12

o 0L44

0 .

o

Lu>

0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0o 0 0 0 0 0 0 0 0

(wu) HIM3 H1430)

0 0

000 C-4-0

Lu4

0 0.

o 00

o 0 0 0 0 0 0 0 0

- C, C.) v &M - Of (l) fln

(w14Id3U (W) Hid3a

0o 0 Eo 000 0 '

CL 0

0- 00

00

00 0 0 0 0 0 0 0 0 0 0 >0 0 0 0 0 0 0 0 0 0 41o 0 0 0 0 0 0 0 0 0

C1 C- 4 ~ n C4d m~ le "

Mw H~dU( HIM~

0 E 0 E>E 0

00

1411

o~ 0 00 0 0

o 0 0 0 0 0 0 0 0o 0 0 0 0 0 0 0 0 0

- C4 M -W n - 04d ( ~ S

()H~d3O (W) Hld~a

13

RAY TRACE (SARGASSO)

1000

92000

3500

4000

0 5 10 Is 20 25 30 35 40 4S soRANGE (nmi)

PROPAGATION LOSS (SARGASSO)

.55. I

.60 SOURCE DEPTH =137 mn-65 -RECEIVER DEPTH = as m

.~- .0 -FREQUENCY 290 Hz

-75

0

-90

-95

.100

0 5 10 Is 20 25 30 35 40 45 soRANGE (nmti3

Figure 5 -Ray Trace and Propagation Loss for Section A

NAVOCEANO TN 3100-55-76

RAY TRACE (SARGASSO)

50a-I

SOURCE

no 2500A

2500

4000 w

0 s 10 is 20 2S 30 35 40 45 soRANGE (nm4;

PROPAGATION LOSS (SARGASSO)

-60 -SOURCE DEPTH 137 mn.65 -RECEIVER DEPTH =88 m

.70

-75

0-6.0

0

-90

-95

0 s 10 15 20 25 30 3S 40 4S soRANGE (nmi)

Figure 6 -Ray Trace and Propagation Loss for Section B

Nlk/U(CLtliU IF) Ji1UU-D:-/ti

RAY TRACE FROM SLOPE TO SARGASSO0 WATER

0I

2500

1'y

3500

N ORT H RANGE (nm.3) SOUTHSLOPE WALL STREAM WALL SARGASSO

PROPAGATION LOSS FROM SLOPE TO SARGASSO WATER

-60 I I

-65

-70 SOURCE DEPTH 137 m

JS5 RECEIVER DEPTH = 88 mFREQUENCY = 290 Hz

-80.- ---0 SINGLE PROFILE1 .65 *-- MUTPEPOIE

o-90

-100

-105

.110

-115 -

r -- 120 -

0 10 20 30 401 SO 60 70 40 90 1001 110 120 130 140 150

*NORTH RANGE (nhl) SOUTHSLOPE WALT, STREAM "ALL SARGASSO

Figure 7 -Ray Trace and Propagation Loss for Section C.Slope to Sargasso Water

* 16

......

RAY TRACE FROM SARGASSO TO SLOPE WATER

300

3000

-S

-0

SORE ETH=13

.0

-60-6S9

-90111TIL

.125

2.130

.130

-140 -

0 10 20 30 40 s0 60 70 s0 90 100 110 120 130 140 I50

RANGE (,smi)

SARGASSO SOUTH STREAM NORTH SLOPE

WALL WALL

Figure 8 -Ray Trace and Propagation Loss for Section C.Sargasso to Slope Wqater

17

SOURCE DEPTH =137 m

- -FREQUENCY =290 Hz

-. 60

-10

0~o'

-10Figue 9 0 %- rpgto os o eto

S. . ...

S - - -- .-. - - - .t.2 l NO

NAVUCEAI'U ITN J/UU-'-rb-

SOURCE DEPTH =137.1 :.FREQUENCY =290 Hz

0 60<

0* -7,I-90

-flu

*11

0t

Fiur 0 - PopgtinLos o Sci.

i~V'JLt~Li Ld JiU ) .j tb

SOURCE DEPTH =137m

- FREQUENCY =290 Hz

.4o

010

Figure 11-3-D Propagation Loss for Section C.Slope to Sargasso'a t er

20

SOURCE DEPTH =137m

-FREQUENCY 290 Hz

4r

"14

21

DISTRIBUTION LIST

!- .DATE: GERIAL"DAE: September 1977 R NAVOCEANO TN 3700-55-76

. SUBJECT: Acoustic Ray Tracing and Three-Dimensional Propagation Loss

Representation in the Gulf Stream Region

-TECHNICAL NOTE -EVALUATiON REPORT3700-55-76

PROGRESS/STATUS REPORT OTHER

SYSTEMS ANALYSIS REPORT

-LETTER

CLASSIFICATION: UNCLASSIFIED .. "NUMBER OF COPIES MADE: 50

COPY COPYNO. (Internal) NO.

1 Code 00 212 Code O1 22 __-

3 Code 02. 23-.4 Code 011 __24 _

5 Code 0111 (6 copies) 25 _

6 Code 1600,7 Code 30008 Code 3004

-" "* 9 Code 300510 Code 3400 ....

11 Code 3440 , _ _.... ...* 12 Code 3500 _

13- Code 3700: 14 'Code 3/1U

-- 15 ' Code 3 7zu' 16

171- 18 o L

19 "20 j_ _ _ _ _ _ __ _ _ _ _

•RE(APIS:

l.I

~Ik~4 .

4,. 4

~- ~ ~ ~

-~-J~i 4 '

1~'.-~ -- ~ ,'1.~

~ ,~

N '*~*'

1

't~!4~ K>

4- L I