PLANE-WAVE DECOMPOSITION AND RECONSTRUCTION OF SPHERICAL-WAVE SEISMOGRAMS AS A LINEAR INVERSE PROBLEM by JOSE JULIAN CABRERA B.Sc. E n g i n e e r i n g G e o p h y s i c s , I n s t i t u t e * P o l i t e c n i c o N a c i o n a l (Mexico),1981 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES (Department Of G e o p h y s i c s And Astronomy) We accept t h i s t h e s i s as c o n f o r m i n g t o the r e q u i r e d s t a n d a r d THE UNIVERSITY OF BRITISH COLUMBIA December 1983 © Jose J u l i a n C a b r e r a , 1983 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the r e q u i r e m e n t s f o r an advanced degree a t the U n i v e r s i t y of B r i t i s h C o lumbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g of t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head of my Department or by h i s or her r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department of G e o p h y s i c s And Astronomy The U n i v e r s i t y of B r i t i s h Columbia 2075 Wesbrook P l a c e Vancouver, Canada V6T 1W5 Date: 5 December 1983 ABSTRACT t The plane-wave d e c o m p o s i t i o n of the v e r t i c a l d i s p l a c e m e n t component of the s p h e r i c a l - w a v e f i e l d c o r r e s p o n d i n g t o a c o m p r e s s i o n a l p o i n t source i s s o l v e d as a s e t of i n v e r s e problems. The s o l u t i o n u t i l i z e s the power and s t a b i l i t y of the Backus & G i l b e r t ( s m a l l e s t and f l a t t e s t ) m o d e l - c o n s t r u c t i o n t e c h n i q u e s , and a c h i e v e s c o m p u t a t i o n a l e f f i c i e n c y t h r ough the use of a n a l y t i c a l s o l u t i o n s of the i n t e g r a l s which a r e i n v o l v e d . The t h e o r y and a l g o r i t h m s d e v e l o p e d i n t h i s work a l l o w s t a b l e and e f f i c i e n t r e c o n s t r u c t i o n of the s p h e r i c a l - w a v e f i e l d from a r e l a t i v e l y s p a r s e s e t of t h e i r plane-wave components. However, the a l g o r i t h m s do not f o r m a l l y c o n s e r v e the c o r r e c t a m p l i t u d e s of the seismograms. Comparison of the a l g o r i t h m s w i t h d i r e c t i n t e g r a t i o n of the Hankel t r a n s f o r m shows v e r y l i t t l e or no advantage f o r the t r a n s f o r m a t i o n from the t i m e - d i s t a n c e ( t - x ) domain t o the d e l a y time - a n g l e of emergence (r-y) domain i f the seismograms a r e eq u i - s a m p l e d s p a t i a l l y . However, f o r cases where the observed seismograms a r e not e q u a l l y spaced and f o r the t r a n s f o r m a t i o n r-y t o t - x , the proposed schemes a r e s u p e r i o r to the d i r e c t i n t e g r a t i o n of the Hankel t r a n s f o r m . A p p l i c a b i l i t y of the a l g o r i t h m s t o r e f l e c t i o n s e i s m o l o g y i s demonstrated v i a the s o l u t i o n t o the problem of t r a c e i n t e r p o l a t i o n and t h a t of s e p a r a t i o n of c o n v e r t e d S modes from o t h e r modes p r e s e n t e d i n common-source g a t h e r s . In both cases the a p p l i c a t i o n of the a l g o r i t h m s t o a s e t of s y n t h e t i c r e f l e c t i o n seismograms y i e l d s s a t i s f a c t o r y r e s u l t s . i v TABLE OF CONTENTS ABSTRACT i i LIST OF TABLES v i i LIST OF FIGURES v i i i ACKNOWLEDGEMENTS X 1 INTRODUCTION TO THE PLANE-WAVE DECOMPOSITION PROBLEM 1.1 E x p a n s i o n of a Sp h e r i c a l - W a v e F i e l d i n Terms of P l a n e and C y l i n d r i c a l Waves: B a s i c Development 1 1.2 On Plane-Wave D e c o m p o s i t i o n of D i g i t a l Data 15 2 BACKUS AND GILBERT FORMULATION OF THE PROBLEM 2.1 Forward T r a n s f o r m : C o n s t r u c t i o n of the Plane-Wave Components of Sph e r i c a l - W a v e Seismograms 20 2.1.1 S m a l l e s t model c o n s t r u c t i o n 21 2.1.2 F l a t t e s t model c o n s t r u c t i o n 25 V 2.2 I n v e r s e T r a n s f o r m : R e c o n s t r u c t i o n of the S p h e r i c a l -Wave F i e l d from i t s Plane-Wave Components 28 2.2.1 S m a l l e s t model c o n s t r u c t i o n 30 2.2.2 F l a t t e s t model c o n s t r u c t i o n 30 3 EXAMPLES 3.1 I n t r o d u c t o r y Comments 32 3.2 S e p a r a t i o n of C o n v e r t e d S Modes 34 3.3 White Nois e and C o n s t r u c t i o n Q u a l i t y 52 3.4 Angul a r Sampling and Trace I n t e r p o l a t i o n 56 3.4.1 S h a l l o w zone of i n t e r e s t 57 3.4.2 Trace i n t e r p o l a t i o n 61 4 COMPUTATIONAL CONSIDERATIONS 66 5 SUMMARY 69 BIBLIOGRAPHY 71 APPENDIX A X 2 V a l u e and O b s e r v a t i o n a l E r r o r s i n Model C o n s t r u c t i o n 73 APPENDIX B E f f e c t s of the W e i g h t i n g F u n c t i o n Q and S t a n d a r d D e v i a t i o n V a l u e s on Model C o n s t r u c t i o n 76 v i APPENDIX C Inner Product Matrix f o r the Forward Smallest Model C o n s t r u c t i o n 86 APPENDIX D The Forward F l a t t e s t Model: G l o b a l Development 87 APPENDIX E Inner Product M a t r i c e s for the Inverse Smallest and F l a t t e s t Model C o n s t r u c t i o n s 91 v i i LIST OF TABLES I . B a s i c Steps f o r PWD of a Common-Source Gather 17 I i . P r o c e s s i n g Times f o r the PWD A l g o r i t h m s 66 v i i i LIST OF FIGURES 1.1a P r o p a g a t i n g s p h e r i c a l - w a v e f r o n t 4 1.1b Plane waves w i t h d i f f e r e n t p r o p a g a t i o n v e l o c i t i e s ... 5 1.2 Mode e x p a n s i o n i n terms of p l a n e waves 6 1.3 D e f i n i t i o n of the wavenumber components 7 1.4 D i s p l a c e m e n t p o t e n t i a l as a s u p e r p o s i t i o n of p l a n e waves 9 3.1 Model used f o r g e n e r a t i n g the s y n t h e t i c seimograms .. 35 3.2a P-P seismograms 37 3.2b P-S seismograms 38 3.3 PP+PS seismograms 39 3.4 S i g n a l s w i t h l a r g e and s m a l l moveouts 40 3.5 P r e d i c t e d t i m e s f o r PWD components 42 3.6 PWS Hankel t r a n s f o r m 43 3.7 PWS c o n v e r t e d S zone 44 3.8 C o n v e r t e d S modes. Hankel-Hankel 46 3.9 C o n v e r t e d S modes. S m a l l e s t - S m a l l e s t 47 3.10 C o n v e r t e d S modes. F l a t t e s t - F l a t t e s t 48 3.11 C o n v e r t e d S zone. H a n k e l - F l a t t e s t 50 3.12 C o n v e r t e d S zone. S m a l l e s t - S m a l l e s t 51 3.13 PP+PS seismograms. A d d i t i v e random n o i s e 53 3.14 C o n v e r t e d S modes. S m a l l e s t - S m a l l e s t . ( N o i s y seismograms) 54 i x 3.15 Co n v e r t e d S modes. F l a t t e s t - F l a t t e s t ( N o i s y seismograms) 55 3.16 PP+PS seismograms. ( N e a r - o f f s e t r e s a m p l i n g ) 58 3.17 PWS s m a l l e s t model. ( N e a r - o f f s e t r e s a m p l i n g ) 59 3.18 PWS f l a t t e s t model. ( N e a r - o f f s e t r e s a m p l i n g ) 60 3.19 PP+PS seismograms. Uneven Sp a c i n g 63 3.20 I n t e r p o l a t e d S modes. S m a l l e s t - S m a l l e s t 64 3.21 I n t e r p o l a t e d S modes. F l a t t e s t - F l a t t e s t 65 B.1 M o d i f i e d B e s s e l f u n c t i o n s K 0 and K, 77 B.2 The c o n d i t i o n number of the i n n e r p r o d u c t m a t r i x .... 80 B.3 E i g e n v e c t o r s f o r the ( s m a l l e s t ) i n n e r p r o d u c t m a t r i x 81 B.4 The 25th b a s i s f u n c t i o n 82 B.5 E f f e c t of s t a n d a r d d e v i a t i o n s on the b a s i s f u n c t i o n s 85 X ACKNOWLEDGEMENTS I am d e e p l y g r a t e f u l t o my ] /] 1 ~j ^ P (dear, f r i e n d ) Shlomo Levy f o r a l l h i s v a l u a b l e a d v i c e , knowledge and moral support d u r i n g the e n t i r e development of t h i s t h e s i s . I r e g a r d him w i t h p r o f o u n d a f f e c t i o n and look forward t o a p e r d u r a b l e f r i e n d s h i p . I t i s my p l e a s u r e t o warmly acknowledge Dr. Ron Clowes, my s u p e r v i s o r , f o r h i s wise d i r e c t i o n and c o n t i n u o u s encouragement. C e r t a i n l y I have deve l o p e d much r e s p e c t f o r Ron. I t i s a b i t d i f f i c u l t t o e x p r e s s , i n a language o t h e r than my n a t i v e , my g r a t i t u d e t o Dr. Doug Oldenburg. He has p r o v i d e d me w i t h guidance and c o u n s e l l i n g , and shared h i s superb knowledge of I n v e r s e Theory. I hope t o c o n t i n u e working w i t h Doug both as st u d e n t and as "amigo". I w i s h t o acknowledge the generous a d v i c e of Dr. George Bluman of the Department of Mathematics a t UBC, f o r h i s v e r y h e l p f u l (and ca l m i n g ) d i s c u s s i o n s , p a r t i c u l a r l y i n r e l a t i o n t o i n t e g r a l s of a r b i t r a r y and m o d i f i e d B e s s e l f u n c t i o n s . E n t h u s i a s t i c thanks a re a l s o due t o M i c h a e l S h l a x , K e r r y S t i n s o n , Kenneth W h i t t a l l and Ian Jones f o r p r o v i d i n g me w i t h so much i n s i g h t f u l a d v i c e . x i I w ish I c o u l d c o n t i n u e l i s t i n g a l l the e x t r a o r d i n a r y p e o p l e who have c o n t r i b u t e d t o the n i c e environment i n which I have s t u d i e d , but space p r e c l u d e s me from d o i n g so. However, I would p a r t i c u l a r l y l i k e t o mention Lynda F i s k , Mark Lane, Don White and Gemma Jones. My M a s t e r ' s s t u d i e s were s u p p o r t e d by a graduate s c h o l a r s h i p from the S c i e n c e and Technology N a t i o n a l C o u n c i l (CONACYT) of Mexico, and O p e r a t i n g Grant A7707 from the N a t u r a l S c i e n c e s and E n g i n e e r i n g Research C o u n c i l of Canada. LA MEMORI A DE MEMITO 1 1. INTRODUCTION TO THE PLANE-WAVE DECOMPOSITION PROBLEM 1.1 E x p a n s i o n of a Spherical-Wave F i e l d i n Terms of P l a n e and C y l i n d r i c a l Waves: B a s i c Development For many y e a r s the study of wave f i e l d s produced by s p h e r i c a l waves t r a v e l l i n g i n s i m p l e e a r t h models has been f a c i l i t a t e d by a n a l y s i s of the p l a n e waves r e p r e s e n t i n g the o r i g i n a l s p h e r i c a l waves. Indeed, f o r a medium c o n s i s t i n g of homogeneous l a y e r s , r e f l e c t i o n , r e f r a c t i o n and mode c o n v e r s i o n a t l a y e r i n t e r f a c e s i s s i m p l e r t o i n v e s t i g a t e u s i n g p l a n e r a t h e r than s p h e r i c a l waves. For example, the method of computing s y n t h e t i c seismograms due t o Fuchs and M u l l e r (1971) i s based on s o l v i n g the wave-propagation problem f o r p l a n e waves and then s u p e r i m p o s i n g the plane-wave s o l u t i o n s t o o b t a i n the s p h e r i c a l -wave f i e l d ( i . e . the s y n t h e t i c seismograms). The r e p r e s e n t a t i o n of a s c a l a r , t i me-harmonic, s p h e r i c a l wave f i e l d (SWF) as a s u p e r p o s i t i o n of p l a n e waves i s w e l l documented i n the s e i s m i c and e l e c t r o m a g n e t i c l i t e r a t u r e ; see f o r example, S t r a t t o n (1941), Bath (1968), Goodman (1968), Born and Wolf (1980), and A k i and R i c h a r d s (1980). L u c i d t r e a t m e n t s of the g e n e r a l t h e o r y a r e g i v e n by B r e k h o v s k i k h (1960), and Devaney and Sherman (1973), and some i n s i g h t f u l a p p l i c a t i o n s a r e 2 p r e s e n t e d by Asby and Wolf (1971), M u l l e r (1971), and T r e i t e l e t a l . (1982). F o l l o w i n g A k i and R i c h a r d s (1980; ch.6) we i n t r o d u c e the Weyl plane-wave and the Sommerfeld c y l i n d r i c a l - w a v e e x p a n s i o n s of a SWF. An o u t l i n e of the m a t h e m a t i c a l s t e p s n e c e s s a r y f o r o b t a i n i n g such r e p r e s e n t a t i o n s f o l l o w s . C o n s i d e r the problem of a p o i n t source at x=0 r a d i a t i n g c o m p r e s s i o n a l waves i n a homogeneous, i s o t r o p i c and unbounded medium. Given t h a t the s o u r c e e x h i b i t s time dependence of the form e x p [ - i c j t ] (u> i s an a r b i t r a r y a n g u l a r f r e q u e n c y ) , compressional-wave p r o p a g a t i o n may be d e s c r i b e d by the ( s c a l a r ) d i s p l a c e m e n t p o t e n t i a l /d2t - V 2 V 2 0 = 4T T V p 2 6 ( x ) e x p t - i w t ] (1-1) where V P r e p r e s e n t s the P-wave v e l o c i t y of the medium. The space-time s o l u t i o n t o (1-1) i s * ( x , t ) = [ 1 / R ] e x p [ i o ; ( R / V P - t ) ] (1-2) w i t h , A A A x = x i + y j + z k R = / x 2 +y 2 + z 2 E q u a t i o n (1-1) may a l s o be s o l v e d u s i n g F o u r i e r t r a n s f o r m methods, i n which case the wavenumber-time s o l u t i o n reads as tf(k,t) = [4irVp 2 / ( k 2 V p 2-£j 2) ] e x p [ - i u t ] (1-3) 3 w i t h k = kx i + k 3 j + k , The r e l a t i o n between (x,t) and (x,t) = [1/R]exp[ i u ( R / V p - t ) ] = [ 1/2TT 2 ] e x p [ - i u t ] / / / [ l / ( k 2 - o ; 2 / V P 2 ) ]exp[ i k - x]dk •too (1-4) -co Equat i o n (1-4) r e p r e s e n t s o s p h e r i c a l wave (x,t) t r a v e l l i n g w i t h a c o n s t a n t v e l o c i t y v p as a s u p e r p o s i t i o n of an i n f i n i t e number of homogeneous p l a n e waves, each of which p r o p a g a t e s w i t h a v e l o c i t y w/k (see F i g u r e s 1.1a and 1.1b). Because 00 and n e g a t i v e f o r z<0 ( n o t i c e t h a t we have e l i m i n a t e d the terms e x p { - i w t ] and e x p l i c i t l y s t a t e d the CJ dependence i n the argument of , each p l a n e wave has an a r b i t r a r y v e l o c i t y V;''=a>/k; , where k;=|kf |. Thus, f o r kbG. So f a r we have d e a l t w i t h an i m p u l s i v e p o i n t source a t x=0. For c ases i n which the p o i n t source has a time dependence F ( t ) w i t h a spectrum F ( C J ) , e q u a t i o n ( 1 - 6 ) i s m o d i f i e d t o 0(o),r,z) = JF(CJ) [ 1 / i k j ]exp[ ik„ z ] k r J 0 ( k r r )dk r ( 1 - 8 ) o Up t o t h i s p o i n t we have c o n s i d e r e d an unbounded homogeneous space. For a medium c o n s i s t i n g of a sequence of homogeneous l a y e r s , the t o t a l d i s p l a c e m e n t p o t e n t i a l a t an o b s e r v a t i o n p o i n t P l o c a t e d a t some depth z i s g i v e n by the 8 c o n t r i b u t i o n from the d i r e c t and r e f l e c t e d p o t e n t i a l s ( r e c a l l t h a t t h e s e p o t e n t i a l s propagate as s p h e r i c a l wave f r o n t s ; see F i g u r e 1.4). In t h i s c a s e , t h e n , the c o r r e s p o n d i n g plane waves ( i n view of 1-5) or c y l i n d r i c a l waves ( i n view of 1-6) a r e weigh t e d by plane-wave r e f l e c t i o n and t r a n s m i s s i o n c o e f f i c i e n t s . A d d i t i o n a l l y , w i t h i n the i t t l a y e r each p l a n e (or c y l i n d r i c a l ) wave i s c h a r a c t e r i z e d by the v e r t i c a l wavenumber k,r>} = y/cj2/V2 ~ k 2 = ucos (7 ; p )/Vrp (1 - 9a) f o r P waves. By s i m i l a r arguments the wavenumber f o r a S wave, k*; = V/CJVV;5 2 - kr2 = ucos(7; s )/V:S (1-9b) may be found. The h o r i z o n t a l wavenumber i s k r= ws i n (y* )/V p, and a s u p e r s c r i p t l e t t e r denotes the wave mode. The d i s p l a c e m e n t p o t e n t i a l a t P ( r , z ) i s then o b t a i n e d from (see F i g u r e 1.4)

° + 01+tf>2 (1-10) where each p o t e n t i a l i s e x p r e s s e d as ( c f . e q u a t i o n 1-8) 0°(cj,r,z)= /F(o)) [ 1/ik; ] e x p [ i k j ; 1 z ] k r J 0 ( k r r ) d k r (1 -11 a) o 0 1 ( c j , r , z ) = /F(o))A[ 1/ik; ] e x p [ i k p l / 1 h 1 + i k P i l (h,-z) ] o k r J 0 ( k r r ) d k r (1-1 l b ) 9 vfv,' Source, vfv* Plr,z) z = 0 O r V,P 7 < vP (C) fe) FIGURE 1.4 Displacement p o t e n t i a l at the ob s e r v a t i o n p o i n t P ( r , z ) as a s u p e r p o s i t i o n of plane waves, (a) The displacement p o t e n t i a l at P i s given by 1 +°, i n (d) the plane wave a f f e c t e d by the r e f l e c t i o n c o e f f i c i e n t A c o n t r i b u t e s to the p o t e n t i a l 0 1, and in (e) the plane wave a f f e c t e d by the t r a n s m i s s i o n c o e f f i c i e n t s B and D and the r e f l e c t i o n c o e f f i c i e n t C c o n t r i b u t e s to 2 ( u , r , z) = fF(u)BCDexp[ i k p y,h , + i k p 2 2 h 2 + i k£ , (h ,-z ) ] [ i / i k * a ] k r J 0 ( k r r ) d k r . (1-1 1c> The v e r t i c a l wavenumbers k^, and k p ; 2 a r e g i v e n by (1-9a) and, because the c o m p r e s s i o n a l source i s l o c a t e d i n the f i r s t -l a y e r , k i = k i ( 1 . A and C a r e plane-wave r e f l e c t i o n c o e f f i c i e n t s , w h i l e B and D are plane-wave t r a n s m i s s i o n c o e f f i c i e n t s . They depend on the e l a s t i c p r o p e r t i e s of the medium and on the ray parameter p ( r e c a l l t h a t p=cjkr =cjsin ( 7 , )/V P ) . From (1-10) and (1-11) the f u n c t i o n d e f i n e d by V ( c j , k r ; z ) = F(co)exp[ i k i , z ] + F (o>) Aexp[ i k\t, h , + i k^ , (h ,-z ) ] + F(w)BCDexp[ i k ; - 1 h , + i k p , 2 2 h 2 + i k P . 1 (h,-z) ] (1-12) w i l l be u n d e r s t o o d as the spectrum of the p l a n e waves d e f i n e d by k r =cjsin {y* )/V p . For a g i v e n a n g u l a r f r e q u e n c y w, V(w,k r;z) g i v e s both the plane-wave c o n t r i b u t i o n from the b o u n d a r i e s between l a y e r s , and the v e r t i c a l phase d e l a y t h a t the (homogeneous) p l a n e waves a c q u i r e d i n each l a y e r . I f k\x i s i m a g i n a r y , an inhomogeneous p l a n e wave w i l l p r opagate h o r i z o n t a l l y i n the i 1 * l a y e r , and the c o n t r i b u t i o n from the bottom i n t e r f a c e t o t h i s l a y e r w i l l have an e x p o n e n t i a l a t t e n u a t i n g term. T h e r e f o r e , w i t h r e f e r e n c e t o F i g u r e 1.4, e q u a t i o n (1-12) g i v e s the response from a d i r e c t p l a n e wave ( f i r s t term . i n the l e f t hand s i d e ) , from a p l a n e wave r e f l e c t e d from the f i r s t 11 boundary (second term) and from a p l a n e wave r e f l e c t e d from the second boundary ( t h i r d t e r m ) . N o t i c e t h a t these p l a n e waves form p a r t of a system of p l a n e waves s h a r i n g the same h o r i z o n t a l wavenumber. In t h i s c o n t e x t , t h e n , V(a>,k r;z) i s viewed as the spectrum of a plane-wave seismogram e q u i v a l e n t l y d e f i n e d by k r, P or 7f, . Some a d d i t i o n a l i n s i g h t i n t o the plane-wave n a t u r e of (1-12) may be seen i n the time domain. B e f o r e i n v e r s e - F o u r i e r t r a n s f o r m i n g t h i s e q u a t i o n , i t i s c o n v e n i e n t t o make the s u b s t i t u t i o n (see 1-9a) k\- = wcos(7, )/V;p i = 1 ,2 Then, we may r e w r i t e (1-12) as V(u,y\;z) = F(u>)exp[ icjzcos ( 7 p , )/V p ] + F(u.)Aexp[iw{h,cos (7 i )/V p + ( h , - z ) c o s ( 7 i )/v'} ] + F(w)BCDexp[ i c j { h ! C O s ( 7 ' ) / V p + 2 h 2 c o s ( 7 f 2 )/Vp2 + ( h 1 - z ) c o s ( 7 e i ) / V ? 1 } ] . (1-13) N o t i c e t h a t we have now used y\ t o c h a r a c t e r i z e V ( & j , k r ; z ) . For homogeneous p l a n e waves ( i . e . f o r 7.p r e a l ) , i n v e r s e F o u r i e r t r a n s f o r m of (1-13) g i v e s 1 2 V C t ^ w - z ) = F ( t - z c o s ( 7 P i )/V\) + A F ( t - { h t C O s O ' J / V j + d ^ - z J c o s f T p / V ' } ) + BCD F ( t - {h,cos(7 f, )/V , , 1+2h 2cos(7 ? 2)/V2 + ( h , - z ) c o s ( 7 ? 1 ) / V , } ] . (1-14) For a g i v e n e a r t h model, t h i s e q u a t i o n demonstrates t h a t the p l a n e wave seismogram V( t , 7 ? , ; z ) i s g i v e n by the source f u n c t i o n F ( t ) a p p e a r i n g a t d e l a y times dependent on the c o s i n e s of the a n g l e s of p r o p a g a t i o n of p l a n e waves w i t h i n each l a y e r . F u r t h e r , t h i s source f u n c t i o n i s weighted by plane-wave r e f l e c t i o n and t r a n s m i s s i o n c o e f f i c i e n t s . Thus, the (homogeneous) plane-wave response f o r the d i r e c t p o t e n t i a l <£°(t,r;z) c o r r e s p o n d i n g t o the a n g l e i s S i m i l a r l y , the plane-wave response f o r the d i s p l a c e m e n t 0 ' ( t , r ; z ) i s g i v e n by N o t i c e t h a t we have used "plane-wave response" t o d e s c r i b e those time f u n c t i o n s whose s p e c t r a c o n s t i t u t e complex w e i g h t s f o r p l a n e or c y l i n d r i c a l waves (see e q u a t i o n s 1-11). I f the o b s e r v a t i o n p o i n t P ( r , z ) i s a t the s u r f a c e z=0, the plane-wave response f o r the d i r e c t p o t e n t i a l i s s i m p l y the g 0(t,7 r,.;z) = F ( t - t d ) = F ( t - z c o s ( V i ) / V ^ ) (1-15a) g M t , 7 i ; z ) = A F ( t - t A ) = A F ( t - {h,cos(Vi )/Vi + ( h , - z ) c o s ( 7 F , )/V ?,}). (1-15b) 1 3 s o u r c e f u n c t i o n w i t h no d e l a y t i m e , i . e . g°(t,7^;z) = F ( t ) f o r a l l a n g l e s 7 ? , . Summarizing and g e n e r a l i z i n g the r e s u l t s , f o r a sequence of homogeneous l a y e r s w i t h a c o m p r e s s i o n a l s o u r c e F ( t ) p l a c e d a t (r=0,z=0), o b s e r v a t i o n p o i n t P(r,z>0) and g i v e n a n g u l a r f r e q u e n c y o>, the r e p r e s e n t a t i o n of the t o t a l d i s p l a c e m e n t p o t e n t i a l at P i n terms of a s u p e r p o s i t i o n of c y l i n d r i c a l waves i s g i v e n .by (see 1-10 t o 1-13) * ( u , r , z ) = JV(w,k r ;z) [ 1 / i k\ ]k r J 0 ( k r r ) d k r . (1-16) o where V((w,r ,z)} = J 3 / 3 z { V ( u , k r ; z ) } [ 1 / i k ; ] k r J 0 ( k r r ) d k r (1-17) o In e q u a t i o n (1-13) i t i s seen t h a t the z-dependence of V(w,k r;z) i s g i v e n o n l y by the terms e x p [ i i c j z c o s ( 7 ^ )/V p] = exp[±ik pz]. Hence (1-17) g i v e s S(o>,r,z) = /U(cj,k r ; z ) [ i k ; ][\/ik\ ] k r J 0 ( k r r )dk r (1-18) o and because of the ± s i g n s i n the argument of e x p f i k ^ z ] , U(o>,k r;z) and V ( c j , k r ; z ) may d i f f e r o n l y i n the s i g n a s s o c i a t e d t o each of t h e i r terms. I f the o b s e r v a t i o n p o i n t i s l o c a t e d a t the s u r f a c e z=0, e q u a t i o n (1-18) becomes S(w,r) = /U(aj,k r ) J 0 ( k r r ) k r d k r o (1-19) 1 5 Plane-wave d e c o m p o s i t i o n (PWD) i s u n d e r s t o o d as the problem of computing the plane-wave seismograms U(u>,kr) from the s p h e r i c a l - w a v e o b s e r v a t i o n s S ( o , r ) . M u l l e r (1971) r e c o g n i z e d e q u a t i o n (1-19) as a z e r o - o r d e r Hankel t r a n s f o r m and c o n s e q u e n t l y , p r e s e n t e d i t s f o r m a l i n v e r s i o n as U(cj,k r) = fs(cj, r ) J 0 ( k r r ) r d r (1-20) o p r o v i d i n g the b a s i c f o r m u l a t i o n f o r PWD. 1.2 On Plane-Wave D e c o m p o s i t i o n of D i g i t a l Data The i n t r o d u c t i o n of s l a n t s t a c k i n g ( S c h u l t z and C l a e r b o u t , 1978) and plane-wave d e c o m p o s i t i o n ( T r e i t e l e t a l . 1982) i n t o the r e a l m of r e f l e c t i o n s e i s m o l o g y has r e v e a l e d a l a r g e number of p o s s i b l e a p p l i c a t i o n s i n e x p l o r a t i o n s e i s m o l o g y . In a r e c e n t p u b l i c a t i o n , T r e i t e l e t a l . , (1982) have shown the r e l a t i o n s h i p s between s l a n t s t a c k i n g and plane-wave d e c o m p o s i t i o n . They p o i n t e d out t h a t a l t h o u g h e q u a t i o n (1-20) i s r e s t r i c t e d t o the v e r t i c a l component of c o m p r e s s i o n a l waves r e c o r d e d over l a t e r a l l y homogeneous media, t h i s f o r m a l i s m seems t o p r o v i d e ' an a c c e p t a b l e a p p r o x i m a t i o n of a l a r g e r c l a s s of e a r t h models, i n p a r t i c u l a r those c o n s i s t i n g of d i p p i n g l a y e r s . The g e n e r a l procedure f o l l o w e d t o p e r f o r m the plane-wave d e c o m p o s i t i o n of s p h e r i c a l - w a v e seismograms r e s u l t i n g from a 1 6 common-source g a t h e r i s i l l u s t r a t e d i n Table I . The a p p l i c a t i o n of PWD i n r e a l s e i s m i c work r e q u i r e s the d i s c r e t i z a t i o n of e q u a t i o n ( 1 - 2 0 ) . The form suggested by T r e i t e l e t a l . (1982) i s , U(w,k r) = ArLSdd.r. ) J 0 ( k r r - ) r f (1-21) In t h i s form, e q u a t i o n (1-21) may be a p p l i e d t o a s e t of seismograms S(o>,r; ) found a t e v e n l y spaced i n t e r v a l s Ar = r ;,, - r ; . In t h i s c a s e , the' i n t e g r a t i o n increment Ar i s f a c t o r e d out of the summation and i n t r o d u c e d l a t e r as a form of g l o b a l s c a l i n g . When uneven seismogram s p a c i n g i s e n c o u n t e r e d , we observe t h a t the f o r m a l use of e q u a t i o n (1-21) w i t h v a r i a b l e Ar f causes d e t e r i o r a t i o n of the d e c o m p o s i t i o n due t o i n a p p r o p r i a t e w e i g h t i n g of c e r t a i n terms i n the summation. In t h i s case i t might be b e t t e r t o e v a l u a t e (1-20) u s i n g an a p p r o p r i a t e n u m e r i c a l i n t e g r a t i o n scheme (e . g . i n t e r p o l a t i n g S(,kr) i s c o n s t r u c t e d . Henry e t a l . ' s s o l u t i o n i s e f f i c i e n t and i s a p p l i e d d i r e c t l y t o the f o r w a r d t r a n s f o r m a t i o n from t i m e - o f f s e t domain t o the d e l a y time - a n g l e of emergence domain ( i . e . from t - x t o r-y; because we a r e c o n c e r n e d w i t h one d i m e n s i o n a l e a r t h models, we make no d i s t i n c t i o n between the space v a r i a b l e s x and r ) . The same approach may be used f o r the i n v e r s e t r a n s f o r m a t i o n TABLE I 1 7 BASIC STEPS FOR PWD OF A COMMON-SOURCE GATHER COMMON-SOURCE GATHER Input of N s e i s m i c t r a c e s S ( t , r ; ) . FORWARD FOURIER TRANSFORM Temporal F o u r i e r t r a n s f o r m a t i o n of each t r a c e i n the common-source g a t h e r . C a l c u l a t i o n of S ( u u , r . - ) . PLANE-WAVE DECOMPOSITION At each a n g u l a r f r e q u e n c y u>, compu t a t i o n of U(UJ, k r ) (or e q u i v a l e n t l y of U(LU,JT)) f o r M d i f f e r e n t a n g l e s of emergence, INVERSE FOURIER TRANSFORM ( w i t h r e s p e c t to<^) From U ( L u , k r ) , c o m p u t a t i o n of M plane-wave seimograms U ( ^ , T ) . 18 ( i . e . from the r-y domain t o the t-x domain). S i n c e f o r some a p p l i c a t i o n s PWD may ser v e as a f i l t e r i n g o p e r a t i o n , our o b j e c t i v e i s t o modify Henry et a l . ' s s o l u t i o n and produce a more f l e x i b l e and s t a b l e a l g o r i t h m which w i l l p e r f o r m both the fo r w a r d ( t - x t o r-y) and the i n v e r s e (r-y t o t- x ) t r a n s f o r m a t i o n s . Rather than use the i n n e r p r o d u c t g i v e n by Henry et a l . , we i n t r o d u c e an e x p l i c i t w e i g h t i n g f u n c t i o n i n t o ( 1 - 1 9 ) OO S ( u , r ) = /[U(w,k r )Q" 1 ] [ Q J 0 U R r ) k r ]dk r , o f o r the f o r w a r d t r a n s f o r m a t i o n , and i n t o (1-20) U ( u , k r ) = / [ S ( u , r ) Q - 1 ] [ Q J 0 U . r ) r ] d r o f o r the i n v e r s e t r a n s f o r m a t i o n . T h i s i s an example of a l i n e a r q u e l l i n g (Backus, 1970), and a l l o w s the use of the u s u a l d e f i n i t i o n of the i n n e r p r o d u c t of two f u n c t i o n s . W i t h t h i s , i n a d d i t i o n t o f i n d the s m a l l e s t model s o l u t i o n s f o r U(co,k r)Q" 1 and f o r S ( c j , r ) Q " ' , we f i n d the f l a t t e s t model s o l u t i o n s . F i n a l l y , the s o l u t i o n s of the f o r w a r d and i n v e r s e t r a n s f o r m a t i o n s a r e found s u b j e c t t o the x 2 c r i t e r i o n (see Appendix A ) , so t h a t o b s e r v a t i o n a l e r r o r s a r e a c c o u n t e d f o r . The a l g o r i t h m s a re a p p l i e d t o the problems of s e p a r a t i n g c o n v e r t e d S modes from o t h e r modes i n a common-source g a t h e r , 19 and of t r a c e i n t e r p o l a t i o n . In the f i r s t problem we w i l l f o l l o w Tatham et a l . (1983), and t r a n s f o r m the common-source g a t h e r ( t - x domain) t o the plane-wave domain (r-y domain). For reasons which w i l l be o u t l i n e d l a t e r , c e r t a i n c o n v e r t e d modes w i l l occupy a d i s t i n c t p o r t i o n of the plane-wave domain. I n v e r s e t r a n s f o r m a t i o n of o n l y t h i s p o r t i o n back t o t-x space w i l l y i e l d t he common-source g a t h e r (CSG) r e p r e s e n t a t i o n of the S modes p r e s e n t i n the chosen r-y zone. The second problem i s s o l v e d by u t i l i z i n g the a l g o r i t h m s t o c o n s t r u c t a d d i t i o n a l seismograms a t o f f s e t s not r e p r e s e n t e d i n the o r i g i n a l CSG. I t i s i m p o r t a n t t o emphasize t h a t a l t h o u g h the use of ( 1 - 1 9 ) and (1-20) are r e s t r i c t e d t o the r e c o r d e d c o m p r e s s i o n a l waves a t the s u r f a c e z=0, we r e a l i z e t h a t k i n e m a t i c a l l y t h e s e e q u a t i o n s a r e s t i l l s a t i s f a c t o r y f o r o b t a i n i n g the plane-wave s i g n a t u r e of r e c o r d e d S waves. The reason f o r t h i s i s u n d e r s t a n d a b l e from the d i s c u s s i o n d e v e l o p i n g ( 1 - 1 2 ) . In the case of S p l a n e waves, V(u>,k r;z) has v e r t i c a l p r o p a g a t i o n terms of t he form exp[ ic j z c o s ( 7 s , )/V* ] and hence, the d e l a y times a r e s t i l l governed by c o s i n e f u n c t i o n s . In Chapter 3 we w i l l r e t u r n t o t h i s m a t t e r . 20 2. BACKUS AND GILBERT FORMULATION OF THE PROBLEM 2.1 Forward Transform (t-x to r-y): C o n s t r u c t i o n of the Plane-Wave Components of Spherical-Wave Seismograms To apply the Ba c k u s - G i l b e r t (E-G) theory to the problem of plane-wave decompositon, we use equation (1-19) at s p e c i f i e d o f f s e t r- and angular frequency u, that i s S(w,r,- ) = /U(cj,k r ) J 0 ( k r r ; ) k r d k r ( 2 - 1 ) o We can now solve ( 2 - 1 ) f o r U(cj,k r) as a set of i n v e r s e problems each of which corresponds to a given angular frequency co. In order to expedite the f o l l o w i n g p r e s e n t a t i o n we introduce the terminology and n o t a t i o n s to be used throughout the remainder of t h i s work: (a) S(cj,r : ), the temporal F o u r i e r - t r a n s f o r m e d elements of the spherical-wave seismograms, at a given angular frequency CJ and o f f s e t r- , are termed 'observations' and are denoted by e?. (b) U(w,k r), the temporal F o u r i e r transform (FT) of the plane-wave seismograms at a given angular frequency u>, i s termed the 'model' and i s denoted by m; m i s a continuous f u n c t i o n of the h o r i z o n t a l wave number k r.. 21 (c) J 0 ( k r r ; ) k r , the z e r o - o r d e r B e s s e l f u n c t i o n s m u l t i p l i e d by the h o r i z o n t a l wave number, a r e termed ' k e r n e l s ' and a r e denoted by G;. They are c o n t i n u o u s f u n c t i o n s of the h o r i z o n t a l wave number k r . (d) The i n n e r product of the f u n c t i o n s f ( k ) and g(k) i s CO denoted by ( i . e . , = / f ( k ) g ( k ) d k ). o In the f o l l o w i n g two s e c t i o n s we w i l l o u t l i n e the B-G s o l u t i o n t o problems of the form of e q u a t i o n (2-1) t o show how s m a l l e s t and f l a t t e s t models can be computed. F u r t h e r t r e a t m e n t of the proce d u r e i s found i n P a r k e r (1977), and Oldenburg and Samson (1979). 2.1.1 S m a l l e s t model c o n s t r u c t i o n ( f o r w a r d t r a n s f o r m ) C o n s i d e r the problem, e? = i = 1 , . . . ,N (2-2) where N i s the number of o b s e r v a t i o n s . Assume t h a t the g i v e n o b s e r v a t i o n s a r e c o n t a m i n a t e d by e r r o r s {5e ;} w i t h z e r o mean and e s t i m a t e d s t a n d a r d d e v i a t i o n a,-, i . e . e? = et±6e,- , e* b e i n g the t r u e d a t a . T h e r e f o r e the e q u a t i o n t o be s o l v e d i s , e*±5e,- = which upon d i v i s i o n by a- becomes, [e;*±6e; ]/o; = 22 e? = (2-3) e° a r e our new o b s e r v a t i o n s w i t h a s s o c i a t e d e r r o r s of u n i t v a r i a n c e , and G ; a r e the new s c a l e d k e r n e l s . G i v e n N o b s e r v a t i o n s e? and the f u n c t i o n a l form ( 2 - 3 ) , we would l i k e t o o b t a i n a c o n t i n u o u s model m. T h i s problem i s always underdetermined and admits i n f i n i t e l y many s o l u t i o n s but a s p e c i f i c model i s o b t a i n e d by m i n i m i z i n g some norm of the model and u s i n g the d a t a e q u a t i o n s as c o n s t r a i n t s . The s m a l l e s t model c o r r e s p o n d s t o the requirement t h a t the L 2 norm of the c o n s t r u c t e d model w i l l be s m a l l e r than t h a t of any o t h e r p e r m i s s i b l e s o l u t i o n ( i . e . a l l those s a t i s f y i n g ( 2 - 3 ) ) . G i v e n t h i s r e q u i r e m e n t , the s o l u t i o n t o (2-3) can be e x p r e s s e d as a l i n e a r c o m b i n a t i o n of the k e r n e l s , t h a t i s (Oldenburg and Samson, 1979), The c o e f f i c i e n t s a f a r e o b t a i n e d by s u b s t i t u t i n g (2-4) i n t o ( 2 - 3 ) , c h a n g i n g the o r d e r of summation and i n t e g r a t i o n and s o l v i n g the system, m = I a : G ; (2-4) e° = Ta (2-5a) t h a t i s , a = T- 1e° (2-5b) where a i s the v e c t o r of k e r n e l c o e f f i c i e n t s , e° i s the v e c t o r of o b s e r v a t i o n s , and 23 T~1 i s the i n v e r s e of the (NxN) i n n e r p r o d u c t m a t r i x T d e f i n e d by, r,-= (2-6) The f o r m a l s o l u t i o n g i v e n i n e q u a t i o n s (2-4) t o (2-6) cannot y i e l d a p h y s i c a l s o l u t i o n i f the k e r n e l s G,- are not square i n t e g r a b l e . The c u r r e n t problem of PWD i s an example of t h i s o c c u r r e n c e . A way to c i r c u m v e n t t h i s i s t o use the q u e l l i n g o p e r a t i o n (Backus,1970) which i s e s s e n t i a l l y a mapping of the k e r n e l s i n t o a H i l b e r t space. The method we use i s named " q u e l l i n g by m u l t i p l i c a t i o n " (Backus,1970) i n which we lo o k f o r a w e i g h t i n g f u n c t i o n Q such t h a t G? = G,-Q i s i n L2(0,°°) f o r a l l i . Once such a Q i s s p e c i f i e d we r e w r i t e (2-3) a s , e? = = (2-7) and c o n t i n u e t o f i n d the s m a l l e s t model m* as o u t l i n e d i n e q u a t i o n s (2-4) t o ( 2 - 6 ) . S u b s e q u e n t l y we "unweight" m* and o b t a i n the d e s i r e d model. The f i n a l s o l u t i o n i s then g i v e n by, m = La; Q 2G ; (2-8) An i m p o r t a n t c o n s i d e r a t i o n i n the c h o i c e of the f u n c t i o n Q i s the ease w i t h which the e v a l u a t i o n of the i n n e r p r o d u c t r;j-= can procee d . The e f f i c i e n c y of the c o n s t r u c t i o n a l g o r i t h m i n c r e a s e s g r e a t l y i f an a n a l y t i c a l e x p r e s s i o n r e p r e s e n t i n g the elements of the i n n e r p r o d u c t m a t r i x " i s found. A l s o , the w e i g h t i n g f u n c t i o n Q s h o u l d l e a d t o an e f f i c i e n t 24 c o n s t r u c t i o n of the i n v e r s e t r a n s f o r m ( i . e . from 7 - 7 t o t - x ) . Indeed, s i n c e many i n v e r s e s o l u t i o n s are r e q u i r e d (one f o r each f r e q u e n c y ) , n u m e r i c a l e f f i c i e n c y i s g a i n e d i f Q i s chosen so t h a t o n l y a s i n g l e m a t r i x s p e c t r a l d e c o m p o s i t i o n i s r e q u i r e d . I f t h i s o b j e c t i v e can be a c h i e v e d , the i n t r o d u c t i o n of the x 2 c r i t e r i o n (Appendix A) f o r n o i s y d a t a does not decrease the a l g o r i t h m ' s e f f i c i e n c y . Our s o l u t i o n t o the s m a l l e s t - m o d e l f o r w a r d c o n s t r u c t i o n i n v o l v e s the w e i g h t i n g f u n c t i o n Q = k r " °-5Kc0-5 (k r b) , where K 0 i s a m o d i f i e d B e s s e l f u n c t i o n of z e r o o r d e r , and b i s an a r b i t r a r y p o s i t i v e r e a l number whose r o l e i s demonstrated i n Appendix B. Wit h the above c h o i c e of w e i g h t i n g the c o n s t r u c t i o n proceeds w i t h t he f o l l o w i n g i d e n t i f i c a t i o n s , nr = m/Q = k r °-5K0-°-s ( k r b ) U ( c j , k r ) (2-9a) and Gr = GrQ = k r °-5K00-5 (k r b) J 0 ( k r r : )/o; (2-9b) Hence, from e q u a t i o n (2-8) the frequ e n c y " r e p r e s e n t a t i o n of the plane-wave seismograms i s g i v e n by, U( i s d e s c r i b e d i n Appendix C, whereas d e t e r m i n a t i o n of the c o e f f i c i e n t s a ; t o o b t a i n p r o p e r x 2 v a l u e i s d i s c u s s e d i n Appendix A. 2.1.2 F l a t t e s t model c o n s t r u c t i o n ( f o r w a r d t r a n s f o r m ) C o n s i d e r the problem o u t l i n e d i n the p r e v i o u s s e c t i o n , i . e . g i v e n N o b s e r v a t i o n s c o r r e s p o n d i n g t o N f u n c t i o n a l r e l a t i o n s e° = , f i n d a model m which s a t i s f i e s these r e l a t i o n s - In t h i s s e c t i o n , we s e a r c h f o r the model which e x h i b i t s the l e a s t amount of change w i t h r e s p e c t t o the independent v a r i a b l e . The c o n s t r u c t i o n of t h i s type of model (commonly r e f e r r e d t o as the f l a t t e s t model) i s a c h i e v e d by the m i n i m i z a t i o n of the norm ||m'||, w i t h m' b e i n g the d e r i v a t i v e of the model. In the problem of PWD, the f l a t t e s t model r e q u i r e s t h a t the F o u r i e r t r a n s f o r m of the PWS e x h i b i t s the l e a s t amount of v a r i a t i o n w i t h r e s p e c t t o k r a t each a n g u l a r f r e q u e n c y o>, and hence i t may y i e l d b e t t e r l a t e r a l c o n t i n u i t y i n terms of both a m p l i t u d e and time d e l a y . In o r d e r t o c o n s t r u c t the f l a t t e s t model we i n t e g r a t e the r . h . s . of e q u a t i o n (2-3) by p a r t s t o o b t a i n , OO e? = H,- m| - (2-1 1 ) o where, H ; ( k r ) = /G;(u)du 2 6 Presuming we can evaluate the term R;m| we s u b s t r a c t i t from the o l e f t hand s i d e to get, e,t = - (2-12) Using the technique o u t l i n e d i n s e c t i o n 2.1.1 above we proceed to f i n d the smallest m' model. If the new kernels H; are not square i n t e g r a b l e , we intr o d u c e a weighting f u n c t i o n Q and then s o l v e e;t = <-H;Q,m'/Q> to ob t a i n m'. The s o l u t i o n to t h i s problem i s given by ( c f . equation 2 - 8 ) , m' = -Z/3- Q2H,- (2-13a) I - r where the 0,- are obtained from, I = r - ' l t (2-13b) and, r..3 = (2-13c) Taking the i n d e f i n i t e i n t e g r a l of equation (2-l3a) we o b t a i n , m(k r) = -Z/3j JQ 2 (u)R,- (u)du + C (2-14) Two important c o n s i d e r a t i o n s should be emphasized at t h i s po i n t : (a) the c o n s t r u c t i o n of the f l a t t e s t model n e c e s s i t a t e s the a d d i t i o n a l knowledge of a (boundary) value of m, from which the constant C i s found. 27 (b) the c h o i c e of the w e i g h t i n g f u n c t i o n Q i s now burdened by the a d d i t i o n a l e v a l u a t i o n of the i n d e f i n i t e i n t e g r a l JH.Q 2. In the s o l u t i o n t o the f o r w a r d - t r a n s f o r m f l a t t e s t - m o d e l c o n s t r u c t i o n we make the f o l l o w i n g i d e n t i f i c a t i o n s : (a) G ; = k r J 0 ( k r r : )/ot . (b) The new k e r n e l s -H;= - k r J , ( k r r , ) / [ o ; r ; ] (see Appendix D). (c) L i m i t i n g r ; t o be g r e a t e r than z e r o , we have H ;m|=0. On o the o t h e r hand, assuming a band l i m i t e d s o u r c e f u n c t i o n , kr-*<» i m p l i e s i n f i n i t e l y - a t t e n u a t e d inhomogeneous waves. Hence we c o n s i d e r H;m| =0 (see Appendix D), which means t h a t the new o b s e r v a t i o n s e-f a r e the same as e f . (d) The w e i g h t i n g f u n c t i o n Q we have chosen i s K,°-5(krb) where i s a m o d i f i e d B e s s e l f u n c t i o n of f i r s t o r d e r and b i s an a r b i t r a r y p o s i t i v e r e a l number (see Appendix B ) . We n o t i c e t h a t because k r=wsin ( 7)/V, a d i f f e r e n t w e i g h t i n g f u n c t i o n Q i s used f o r each a n g u l a r f r e q u e n c y u>. (e) The c o n s t a n t of i n t e g r a t i o n C i s e q u a l t o zer o (see Appendix D). F o l l o w i n g the development i n Appendix D, t h e plane-wave seismograms a t a g i v e n a n g u l a r f r e q u e n c y a re t h e r e f o r e g i v e n by, U ( u , k r ) = -L0; JQ2H,- = |[0j /o; r , ] { b k r J , (k r r ; ) K 0 ( k r b ) + r. k r J 0 (k rr,- )K, ( k r b ) } / [ r ; 2 + b 2 ] (2-15) 28 2.2 I n v e r s e T ransform (r-y t o t - x ) : R e c o n s t r u c t i o n of the S p herical-Wave F i e l d from i t s Plane-Wave Components The i n v e r s e problem t o be s o l v e d i s e x p r e s s e d i n e q u a t i o n ( 1 - 2 0 ) . k r has been r e p l a c e d by r as the independent v a r i a b l e , and the model and o b s e r v a t i o n s have t r a d e d p l a c e s so t h a t the former now r e p r e s e n t s the t e m p o r a l FT of the s p h e r i c a l - w a v e seismograms S(w,r) whereas the l a t t e r c o n s i s t s of the FT of the plane-wave seismograms U(&>, k r ).. Indeed, t h e r e i s no d i f f e r e n c e between c o n s t r u c t i n g plane-wave and s p h e r i c a l - w a v e seismograms from each o t h e r i n the way f o r m u l a t e d i n the p r e v i o u s s e c t i o n . However, because we have f o r m u l a t e d the problem i n terms of the h o r i z o n t a l wavenumber k r r a t h e r than i n terms of the ray parameter p, the i n n e r p r o d u c t m a t r i x f o r the i n v e r s e t r a n s f o r m a t i o n e x p l i c i t l y depends on the a n g u l a r f r e q u e n c y (see Appendix E ) . To see t h i s and compare the form of the i n n e r p r o d u c t m a t r i c e s f o r the f o r w a r d and i n v e r s e t r a n s f o r m a t i o n s , l e t us c o n s i d e r the f o l l o w i n g i n t e g r a l s r;F- = j G ; ( k r r ; )G- (k rr_; ) Q 2 ( k r b ) d k r (2-16a) o and T-j = ; G ; ( k r . r ) G j ( k f j r ) Q 2 ( b r ) d r . (2-16b) r;F- i s an element of the i n n e r p r o duct m a t r i x f o r the f o r w a r d t r a n s f o r m a t i o n w h i l e r,* i s an element f o r the i n v e r s e t r a n s f o r m a t i o n . In terms of the ray parameter p, (2-16a) and (2-!6b) r e a d as 29 f oo r.- = /G; (upr ; )Gj (cjprj )Q 2 (upb)dup (2-17a) o and r?i = o/G; (up. r)G- (up^rjQMbrJdr. ( 2 - l 7 b ) These integrals have the same form. I f in ( 2 - l 7 a ) we set b=r c =positive constant, and in (2-17b) b=cjpc , with p„=positive constant we find rf- = /G; (cjpr ; )G0 (cjprj )Q 2 ( u p r c )dwp (2-18a) and r?j = 1/CJ /G; (£jp; r ) G j ( c j p J r ) Q 2 ( u p c r ) d c j r . ( 2 - l 8 b ) T h e r e f o r e , integration of ( 2 - l 8 a ) with respect to cjp and ( 2 - l 8 b ) with respect to ur w i l l give matrices r* and r A which can have m u l t i p l i c a t i v e factors dependent on u. In both transformations, spectral decomposition of a single inner product matrix i s done only once. For the inverse transformation the d e f i n i t i o n of b=a>pc in the argument of Q means that, as in the forward problem, a di f f e r e n t weighting function is used at each angular frequency CO. 30 2.2.1 S m a l l e s t model c o n s t r u c t i o n ( i n v e r s e t r a n s f o r m ) The proc e d u r e here i s p a r a l l e l t o t h a t o u t l i n e d i n the s e c t i o n d e a l i n g w i t h the c o r r e s p o n d i n g f o r w a r d t r a n s f o r m . I d e n t i f y i n g , G; = r J 0 ( r k r ; )/a ; Q = r-°-5K0°-s ( r b ) and u s i n g e q u a t i o n ( 2 - 8 ) , the s m a l l e s t model f o r the s p h e r i c a l -wave seismograms a t a g i v e n u> i s , S(w,r) = Z [ a ; / c ; ] K 0 ( rb) J 0 ( r k r ; ) w i t h b = w s i n ( c ) / V (2-19) where M i s the number of plane-wave seismograms. 2.2.2 F l a t t e s t model c o n s t r u c t i o n ( i n v e r s e t r a n s f o r m ) The development here i s s i m i l a r t o t h a t of the c o r r e s p o n d i n g f o r w a r d t r a n s f o r m . At the st a g e of the s o l u t i o n of the s m a l l e s t model m', the s u b s t i t u t i o n b = c j s i n ( c ) / V i s made (Appendix E ) . Assuming S (a>, r=°°) =0, the i n t e g r a t i o n c o n s t a n t C i n e q u a t i o n (2-14) i s z e r o . Hence the freq u e n c y r e p r e s e n t a t i o n of the v e r t i c a l component of the s p h e r i c a l - w a v e seismograms i s g i v e n by, 31 b = S ( u , r ) = Z[/3,/a, k r ; ] { b r J , ( k r ; r ) K 0 ( r b ) i - t + k r ; r J 0 ( k r ; r ) K 1 ( r b ) } / [ k 2+b 2 ]} us i n ( c ) / V ( 2 - 2 0 ) where M i s the number of plane-wave seismograms, 32 3. EXAMPLES 3.1 I n t r o d u c t o r y Comments In the f o r w a r d t r a n s f o r m our g o a l i s t o o b t a i n plane-wave seismograms U(r,y) from s p h e r i c a l - w a v e seimograms S ( t , x ) . To a c h i e v e t h i s , we f o l l o w the s t e p s i l l u s t r a t e d i n T a b l e I of s e c t i o n 1.2 (p. 17). C o n v e r s e l y , i n the i n v e r s e t r a n s f o r m our o b j e c t i v e i s t o compute the v e r t i c a l - d i s p l a c e m e n t s p h e r i c a l - w a v e seismograms S ( t , x ) from plane-wave seismograms U{r,y). To a c h i e v e t h i s we s t i l l f o l l o w those s t e p s g i v e n i n T a b l e I though kee p i n g i n mind t h a t the i n p u t d a t a are plane-wave seismograms U ( r , 7 ) . We r e f e r t o f o r m a t i o n of the plane-wave or s p h e r i c a l -wave seismograms v i a the d i s c r e t e form of e q u a t i o n (1-20) or e q u a t i o n (1—19) as the Hankel t r a n s f o r m , and c o n s t r u c t i o n i n v o l v i n g the B-G i n v e r s i o n as the s m a l l e s t or f l a t t e s t model. In t he examples t o f o l l o w we w i l l o b t a i n plane-wave seismograms f o r v a l u e s of the a n g l e of emergence y between 0° and 90° ( t h a t i s t o say 0>1 ( i . e . f o r h i g h f r e q u e n c i e s and/or l a r g e o f f s e t s ; A k i and R i c h a r d s , 1980, c h . 6 ) . In (a) c l o s e r e c e i v e r s c o n t r i b u t e t o form plane-wave s i g n a l a t s m a l l a n g l e s of emergence whereas f a r r e c e i v e r s do i t f o r l a r g e emergence a n g l e s . N o t i c e the d i f f e r e n c e s i n d e n s i t y of a n g u l a r i n f o r m a t i o n f o r s m a l l and l a r g e o f f s e t s . In (b) the s p h e r i c a l - w a v e s i g n a l from a deep i n t e r f a c e has s m a l l moveout throughout the r e c e i v e r s and hence, i t s plane-wave si-gna-1 w i l l be o b s e r v e d a t s m a l l a n g l e s of emergence. 41 of the o b s e r v e d modes i n the plane-wave domain, we d i s p l a y i n F i g u r e 3.5 the t h e o r e t i c a l t r a j e c t o r i e s of the modes which a r e e x p e c t e d f o r the g i v e n model. To s t a r t t h i s example, we show the Hankel plane-wave seismograms i n F i g u r e 3.6. The c o n v e r t e d S modes PPSS, PSSP, PSSS and PS a r e seen q u i t e c l e a r l y i n the a r e a 0.2S 2. 2 m n 2. 4 2. 6 2. 8 3. 0 3. 2 3. 4 3 . 6 3 . 8 4. 0 FIGURE 3.21 C o n v e r t e d S-mode t - x seismograms o b t a i n e d by ( i ) PWD of F i g u r e 3.19 u s i n g the f o r w a r d s m a l l e s t - m o d e l a l g o r i t h m , and ( i i ) r e c o n s t r u c t i o n of the t - x seismograms u s i n g the i n v e r s e f l a t t e s t - m o d e l a l g o r i t h m . The seismograms a r e n o r m a l i z e d by t r a c e . 66 4. COMPUTATIONAL CONSIDERATIONS The CPU t i m e s (on an Amdahl 470 V/8 computer) f o r the PWD a l g o r i t h m s are p r e s e n t e d i n Table I I below. The seismograms used to g e n e r a t e t h i s t a b l e were d i g i t i z e d a t 8 ms and were l i m i t e d t o the band . 5-25 Hz (the number of frequency samples i n t h i s band i s e q u i v a l e n t t o those c o n t a i n e d i n the band 10-50 Hz on data w i t h 4 ms s a m p l i n g i n t e r v a l ) . F u r t h e r m o r e , a l l c a l c u l a t i o n s ( w i t h the e x c e p t i o n of B e s s e l f u n c t i o n c o m p u t a t i o n s ) were made i n double p r e c i s i o n . TABLE I I TIME INPUT OUTPUT CPU TIMES SAMPLES TRACES TRACES HANKEL SMALLEST FLATTEST 1 28 256 512 . 40 40 40 45 45 45 2.416s 4.768s 9.349s 3 .556s 6.770s 12.965s 5.342s 10.192s 19.909s The f o l l o w i n g p o i n t s a r e emphasized: ( i ) Most of the p r o c e s s i n g time i s e l a p s e d i n B e s s e l f u n c t i o n c o m p u t a t i o n s . For t h e s e , we have used p o l y n o m i a l a p p r o x i m a t i o n s w i t h e r r o r s of 0 ( 1 0 " 8 ) g i v e n i n Abramowitz and Stegun (1970). E f f i c i e n c y of t h e s e c o m p u t a t i o n s can be i n c r e a s e d by: (a) d e c r e a s i n g the a c c u r a c y of the a p p r o x i m a t i o n s which w i l l reduce p r o c e s s i n g t i m e s , though i t 67 may y i e l d some d e g r a d a t i o n of the o u t p u t , and (b) u s i n g the d e r i v a t i v e r e l a t i o n between J 0 and J , i n f l a t t e s t model c a l c u l a t i o n s . From the summation r e p r e s e n t e d by e q u a t i o n (2-10) f o r the s m a l l e s t or t h a t of e q u a t i o n (2-15) f o r the f l a t t e s t model, i t i s deduced t h a t i n c r e a s i n g the number of i n p u t or output t r a c e s causes a l i n e a r i n c r e a s e i n the number of B e s s e l f u n c t i o n c o m p u t a t i o n s . S i m i l a r l y , an i n c r e a s e i n the number of time or frequency samples r e s u l t s i n a l i n e a r i n c r e a s e i n the number of B e s s e l f u n c t i o n c o m p u t a t i o n s . ( i i ) The CPU time f o r the s i n g u l a r v a l u e d e c o m p o s i t i o n (SVD) of the i n n e r p r o d u c t m a t r i x T behaves l i k e the cube of the number of i n p u t t r a c e s . Hence, depending on the r a t i o between the number of time and freq u e n c y samples t o the number of i n p u t t r a c e s , a s i g n i f i c a n t p e r c e n t a g e of the t o t a l run time may be spent i n SVD. However, when p r o c e s s i n g a l a r g e number of CSG's w i t h f i x e d geometry and s t a n d a r d d e v i a t i o n e s t i m a t e s , SVD i s ex e c u t e d once. In t h i s c a s e , the t o t a l s m a l l e s t - m o d e l CPU time i s comparable t o t h a t of the Hankel a l g o r i t h m . ( i i i ) The memory r e q u i r e m e n t s of the B-G a l g o r i t h m s i n c l u d e one a r r a y of s i z e NPTSxNTRACE and two of s i z e NTRACExNTRACE, where NTRACE i s the number of e i t h e r i n p u t or o u t p u t t r a c e s w h i c h e v e r i s l a r g e r , and NPTS i s the number of samples per t r a c e . In c o n t r a s t , the Hankel a l g o r i t h m needs o n l y one a r r a y of s i z e NPTSxNTRACE. 68 ( i v ) The CPU run t i m e s and memory re q u i r e m e n t s c o r r e s p o n d i n g t o the i n v e r s e t r a n s f o r m a l g o r i t h m s are e q u i v a l e n t t o those of the f o r w a r d t r a n s f o r m . 69 5. SUMMARY E f f i c i e n t a l g o r i t h m s f o r the d e c o m p o s i t i o n of a s p h e r i c a l -wave f i e l d i n t o i t s plane-wave components have been p r e s e n t e d . A l s o , i t has been shown t h a t t h e s e a l g o r i t h m s a l l o w the r e c o n s t r u c t i o n of the s p h e r i c a l - w a v e f i e l d from a r e l a t i v e l y s p a r s e sample of i t s plane-wave components. The p r a c t i c a l v i a b i l i t y of the proposed a l g o r i t h m s has been demonstrated u s i n g the problem of s e p a r a t i o n of c o n v e r t e d S modes from o t h e r modes i n a common-source g a t h e r , and the problem of t r a c e i n t e r p o l a t i o n . The f o l l o w i n g p o i n t s s h o u l d be n o t e d : ( i ) The plane-wave seismograms a r e o b t a i n e d by u s i n g the Backus & G i l b e r t c o n s t r u c t i o n t e c h n i q u e s , s u b j e c t e d t o the requirement of w e i g h t e d s m a l l e s t or f l a t t e s t model. ( i i ) The c o n s t r u c t i o n schemes a l l o w the h a n d l i n g of e r r o r s i n the d a t a and hence, p e r m i t a c e r t a i n c o n t r o l on the model s t r u c t u r e p r o v i d e d by the b a s i s f u n c t i o n s . C a u t i o n s h o u l d be e x e r c i s e d i n a s s i g n i n g the s t a n d a r d d e v i a t i o n v a l u e s t o u n n o r m a l i z e d o b s e r v a t i o n s . ( i i i ) The proposed a l g o r i t h m s a r e not l i m i t e d t o e v e n l y spaced d a t a and c o n s e q u e n t l y , a l l o w the d e s i g n of an a p p r o p r i a t e geophone a r r a y which s h o u l d produce a more f a i t h f u l r e p r e s e n t a t i o n of the plane-wave components. 70 ( i v ) N u m e r i c a l s t a b i l i t y i s g a i n e d by proper use of the b v a l u e a p p e a r i n g i n the arguments of the w e i g h t i n g f u n c t i o n s . Large b v a l u e s decrease the degree of l i n e a r independence of the k e r n e l s and seem t o be a p p r o p r i a t e f o r the problem of t r a c e i n t e r p o l a t i o n . (v) Dynamic a s p e c t s ( e.g: t r u e a m p l i t u d e s ) of the forwa r d and i n v e r s e c o n t r u c t e d models are not f o r m a l l y handled by the a l g o r i t h m s as d e v e l o p e d . ( v i ) For a g i v e n a n g l e of emergence 7 , homogeneous p l a n e waves c o r r e s p o n d i n g to h i g h f r e q u e n c i e s may be s i g n i f i c a n t l y a t t e n u a t e d by the w e i g h t i n g f u n c t i o n Q. T h i s f u n c t i o n t h e n , i s viewed as a p o t e n t i a l a l i a s i n g s u p p r e s s o r . ( v i i ) For a g i v e n a n g u l a r frequency C J , l a r g e wavenumber components are s e v e r e l y a t t e n u a t e d by the w e i g h t i n g f u n c t i o n Q. Indeed, depending on the b v a l u e chosen, inhomogeneous waves a s s o c i a t e d w i t h wavenumbers l a r g e r than a c e r t a i n v a l u e a re p r a c t i c a l l y e x c l u d e d from the d e c o m p o s i t i o n . T h i s e f f e c t i s analogous t o f o r m u l a t i n g PWD as an i n v e r s e problem w i t h f i n i t e l i m i t s of i n t e g r a t i o n . 71 BIBLIOGRAPHY Abramowitz, M. , and Stegun, I . , 1970, Handbook of M a t h e m a t i c a l F u n c t i o n s : New York, Dover P u b l i c a t i o n s I n c . , 1046 p. A k i , K., and R i c h a r d s , P. G., 1980, Q u a n t i t a t i v e Seismology. Theory and Methods: San F r a n c i s c o , W. H. Freeman and Co., v. I , 557 p. Asby, R., and Wolf, E., 1971, Evanescent waves and the' e l e c t r o m a g n e t i c f i e l d of a moving charged p a r t i c l e : J . Opt. Soc. Am., v. 61, p. 52-59. Backus, G., 1970, I n f e r e n c e from inadequate and i n a c c u r a t e d a t a , I I : P r o c . Nat. Acad, of S c i . , v. 65,- p. 281 -287. Bath, M., 1968, M a t h e m a t i c a l A s p e c t s of Seismology: Amsterdam, E l s e v i e r P u b l i s h i n g Co., 415 p. Born, M., and Wolf, E., 1980, P r i n c i p l e s of O p t i c s : T o r o n t o , Pergamon P r e s s . , 808 p. B r e k h o v s k i k h , L. M., 1960, Waves i n Lay e r e d Media: New York, Academic P r e s s . , 561 p. C a r t e r , W. H., 1975, B a n d - l i m i t e d a n g u l a r - s p e c t r u m a p p r o x i m a t i o n t o a s p h e r i c a l s c a l a r wave f i e l d : J . Opt. Soc. Am., v. 65, p. 1054-1058. Devaney, A. J . , and Sherman, G. C , 1973, Plane-wave r e p r e s e n t a t i o n s f o r s c a l a r wave f i e l d s : SIAM Rev., v.15 , p. 765-786. Fuchs, K., 1968, The r e f l e c t i o n of s p h e r i c a l waves from t r a n s i t i o n zones w i t h a r b i t r a r y depth-dependent e l a s t i c moduli and d e n s i t y : J . Phys. of the E a r t h , v.16, p. 27-41. Fuchs, K., and M u l l e r , G., 1971, Computation of s y n t h e t i c seismograms w i t h the r e f l e c t i v i t y method and c o m p a r i s i o n w i t h o b s e r v a t i o n s : Geophys. J . R. A s t r . S o c , v. 23, p. 417-433. Goodman, J . W. , 1968, I n t r o d u c t i o n t o F o u r i e r O p t i c s : San F r a n c i s c o , M c G r a w - H i l l Co., 287 p. G r a d s h t e y n , I . , and R y z h i k , I . , 1980, T a b l e of I n t e g r a l s , S e r i e s and P r o d u c t s : T o r o n t o , Academic P r e s s , 1160 p. Henry, M., O r c u t t , J . , and P a r k e r , R., 1980, A new method f o r s l a n t s t a c k i n g r e f r a c t i o n d a t a : Geophys. Res. L e t t . , v. 7, p. 1073-1076. 72 M u l l e r , G., 1971, D i r e c t i n v e r s i o n of s e i s m i c o b s e r v a t i o n s : Z e i t s c h r i f t f u r Geophysik, v. 37, p. 225-235. Oldenburg, D. W, and Samson, J . C, 1979, I n v e r s i o n of i n t e r f e r o m e t r i c d a t a from c y l i n d r i c a l l y symmetric r e f r a c t i o n l e s s plasmas: J . Opt. Soc. Am., v. 69, p. 927-941. P a r k e r , R. L., 1977, U n d e r s t a n d i n g i n v e r s e t h e o r y : Ann. Rev. E a r t h P l a n e t . S c i . , v. 5, p. 35-64. Ryu, J . V., 1982, D e c o m p o s i t i o n (DECOM) approach a p p l i e d t o wave f i e l d a n a l y s i s w i t h s e i s m i c r e f l e c t i o n r e c o r d s : G e o p h y s i c s , v. 47, p. 869-883. S c h u l t z , P. S., and C l a e r b o u t , J . F., 1978, V e l o c i t y e s t i m a t i o n and downward c o n t i n u a t i o n by wavefront s y n t h e s i s : G e o p h y s i c s , v. 43, p. 691-714. S t r a t t o n , J . A., 1941, E l e c t r o m a g n e t i c Theory: New York, McGraw-H i l l Co., 615 p. Tatham, R. H., Goolsbee, D. V., M a r s e l , W. F., and N e l s o n , H. R., 1983, S e i s m i c shear-wave o b s e r v a t i o n s i n a p h y s i c a l model ex p e r i m e n t : G e o p h y s i c s , v. 48, p. 688-701. T r e i t e l , S., G u t o w s k i , P., and Wagner, D., 1982, Plane-wave d e c o m p o s i t i o n of seismograms: G e o p h y s i c s , v. 47, p. 1375-1401. 73 APPENDIX A X 2 V a l u e and O b s e r v a t i o n a l E r r o r s i n Model C o n s t r u c t i o n When s o l v i n g problems which a re a s s o c i a t e d w i t h i n a c c u r a t e o b s e r v a t i o n s , i t i s u n d e s i r a b l e t o c o n s t r u c t models which reproduce these d a t a e x a c t l y . In t h i s c a s e , i t i s common t o r e q u i r e the c a l c u l a t e d o b s e r v a t i o n s t o f i t the d a t a i n a manner c o n s i s t e n t w i t h the o b s e r v a t i o n a l e r r o r s . In t h i s appendix, we o u t l i n e the s t e p s r e q u i r e d i n the c o n s t r u c t i o n of models w i t h c a l c u l a t e d o b s e r v a t i o n s which a r e r e l a t e d t o the observed d a t a by an ex p e c t e d x 2 v a l u e of a p p r o x i m a t e l y N. F i r s t l y , from e q u a t i o n (2-5a) we have, Ta = 1° E x p r e s s i n g T i n terms of i t s s p e c t r a l components and s o l v i n g f o r a we g e t , a = r- 1t° = RA" 1R Te° (A-1) where R i s an (NxN) m a t r i x whose columns a re the e i g e n v e c t o r s of T, A i s an (NxN) d i a g o n a l m a t r i x whose d i a g o n a l c o n s i s t s of the e i g e n v a l u e s of T a r r a n g e d i n d e c r e a s i n g o r d e r , and R T i s the t r a n s p o s e of R. - M i s f i t t i n g the o b s e r v a t i o n s e° i s r e a d i l y a c h i e v e d by winnowing, say, the K s m a l l e s t e i g e n v a l u e s w i t h t h e i r a s s o c i a t e d e i g e n v e c t o r s , t h a t i s , t r u n c a t i n g m a t r i c e s R, A and R T t o s i z e 74 (NxM), (MxM) and (MxN) r e s p e c t i v e l y , with M=N-K. The c o e f f i c i e n t s ac c o n s t r u c t e d from the truncated set of s p e c t r a l components are, a? = R ^ R j i 0 (A-2)' Using a c we can compute the c a l c u l a t e d o b s e r v a t i o n s , i . e . 1° = T a c (A-3) Secondly, c o n s i d e r the x 2 value d e f i n e d by X 2 = Z ( e T - e P ) 2 = ||A?|| 2 (A -4 ) with A? = e'-e° and E ( x 2 } - N f o r N>5. The l e n g t h of the vecto r Ae i s not changed upon r o t a t i o n . T herefore we can p r o j e c t e 1 and e° onto the e i g e n v e c t o r s of R, i . e . X 2 = ||R Tt" - R T e ° | | 2 o r X 2 = £(e? - S ? ) 2 (A-5) F i n a l l y , from (A-3) and (A-2) we w r i t e , e 6 = R A R T R M A : 1 R j t ° P r e m u l t i p l y i n g t h i s equation by R T and w r i t i n g down the n o t a t i o n f o r r o t a t e d o b s e r v a t i o n s we o b t a i n , ec = AR TR wA;'e 0 , (A-6) A from which we r e a l i z e that ef = e° f o r i=1,...,M and ef = 0 f o r i=M+1,...,N. Hence (A-5) g i v e s , X 2 =z(i?)2 75 For complex data we use, X 2 = Ze^e?*" (A-7) where * i n d i c a t e s the complex conjugate. S t a r t i n g with M=N-1 we form the summation given in equation ( A - 7 ) , and keep adding terms u n t i l t h i s summation y i e l d s the c l o s e s t value to N (number of o b s e r v a t i o n s ) . The f i n a l index M g i v e s the number of eigenvalues and e i g e n v e c t o r s to be r e t a i n e d in the c a l c u l a t i o n of a c . These c o e f f i c e n t s , when used i n the c o n s t r u c t i o n , y i e l d a model which s a t i s f i e s the o b s e r v a t i o n s in a manner c o n s i s t e n t with o b s e r v a t i o n a l e r r o r s . 7 6 APPENDIX B E f f e c t s of the Weighting Function Q and Standard D e v i a t i o n Values on Model C o n s t r u c t i o n A f t e r i n t r o d u c i n g the standard d e v i a t i o n values o- and weighting f u n c t i o n Q, our o r i g i n a l problem e,- = has been m o d i f i e d t o, e ; = (B-1) where, e; = e; / O ; G* = G ;Q/a; m+ = Q"1m Q"1 = 1/Q with Q=k;°- 5K 0 0- 5 (k r b) f o r the (forward) s m a l l e s t problem Q=K,°' 5(k rb) f o r the (forward) f l a t t e s t problem and b an a r b i t r a r y p o s i t i v e r e a l number. In t h i s appendix we h i g h l i g h t the e f f e c t s of Q with a given b v a l u e , and of a- on the c o n s t r u c t e d model m=U(cj,k r). 1. E f f e c t s of Q on the s i z e of m. The r o l e of the constant b i n a t t e n u a t i n g l a r g e h o r i z o n t a l wavenumber components i s p o r t r a y e d in F i g u r e B.1. T h i s f i g u r e shows the p l o t s of the m o d i f i e d B e s s e l f u n c t i o n s K 0 ( r ( 7 ) ) and 77 K , ( r ( 7 ) ) , where the argument i s d e f i n e d by r ( 7 ) = b w s i n ( 7 ) / V with 0 . 1 ° < 7 < 9 0 ° , cj=407rrad/s, V=1500 m/s , and b assumes the values 1 and 10. C l e a r d i f f e r e n c e s on a t t e n u a t i o n r a t e imposed by these f u n c t i o n s are observed, such that the s i z e of the c o n s t r u c t e d model w i l l be s i g n i f i c a n t l y a f f e c t e d . In what f o l l o w s we d i s c u s s the s m a l l e s t model problem, whose weighting f u n c t i o n i n v o l v e s K 0 . 0 . 0 9 0 . 0 0 . 0 9 0 . 0 A X I S I N D E G R E E S P X I S I N D E G R E E S ( a ) ( b ) FIGURE B.I M o d i f i e d B e s s e l f u n c t i o n s (a) K 0, and (b) K, f o r two d i f f e r e n t b values i n the argument r ( 7 ) (r( 7 )=bwsin( y ) / V , tj=407rrad/s and V=1500m/s). Large b values s e v e r e l y attenuate l a r g e h o r i z o n t a l wavenumbers and decrease the l i n e a r independence of the k e r n e l s . The m i n i m i z a t i o n of ||m*|| r e q u i r e s that (Backus, 1970), | |Q- 'm| |,kr) i s c o n f i n e d to those whose high wavenumbers are s t r o n g l y attenuated. T h i s model a t t e n u a t i o n i s more severe f o r weighting f u n c t i o n s Q with l a r g e b va l u e s . Homogeneous plane waves have h o r i z o n t a l wavenumbers r e s t r i c t e d between 0 and C J / V ( i . e . angles of emergence between 0° and 90°). For t y p i c a l e x p l o r a t i o n seismic work, C J / V i s smaller than u n i t y and hence, with small b va l u e s , these wave components are not s e v e r e l y attenuated by Q"1 in the model m. On the other hand, inhomogeneous waves have wavenumbers between u / V and °° ( i . e . complex angles between 90° and 90°-i<»: Brekhovskikh, 1960). When k r>u ) / V but i s "reasonably s m a l l " , these waves are s t i l l c o n t r o l l e d by the data equation (B-1). However, as k r becomes l a r g e r the inhomogeneous waves are i n c r e a s i n g l y attenuated by the requirement s p e c i f i e d i n (B-2) so that t h e i r amplitudes w i l l decrease e x p o n e n t i a l l y . I t i s c l e a r that the " t r a n s i t i o n " value of kr at which c o n s t r a i n t (B-2) predominates (B-1) depends on the chosen b value, that i s to say | |k r°- 5(k rb)°- 2 5exp(k rb/2)U(cj,k r) | | has to be kept f i n i t e as k r becomes l a r g e (k r-*•»). The preceding d i s c u s s i o n a p p l i e s to the forward f l a t t e s t problem as w e l l . In t h i s case, the weighting f u n c t i o n Q a f f e c t s the d e r i v a t i v e of m, or e q u i v a l e n t l y , the rate of change of the 79 c o n t r i b u t i o n of the homogeneous and inhomogeneous plane waves at a given angular frequency GJ. 2. E f f e c t s of Q on the inner product matrix T. The elements of the inner product matrix T for the (forward) s m a l l e s t and f l a t t e s t problems are i n v e r s e l y p r o p o r t i o n a l to [ (r ; 2+b 2 + r-2) 2 - 4 r - 2 r J 2 ], where r ; and rj are the o f f s e t s corresponding to the i ' t h and j ' t h geophone l o c a t i o n s such that r- T2 2>. . .>ruiJ . For b<T 2 2>...>C„. T h i s behaviour of the d i a g o n a l elements of T provi d e s a n a t u r a l o r d e r i n g f o r the b a s i s f u n c t i o n s such that the long-wavelength s t r u c t u r e of the model i s c o n t r o l l e d by the b a s i s f u n c t i o n s a s s o c i a t e d with the l a r g e s t eigenvalues, whereas the f i n e s t r u c t u r e o r i g i n a t e s from those b a s i s f u n c t i o n s a s s o c i a t e d . with the sm a l l e s t e i g e n v a l u e s . D i v i d i n g each element of T by a- <7j the r e l a t i o n s , >T 2 2> . . . >rufJ do not n e c e s s a r i l y h o l d . In p a r t i c u l a r , f o r the case of small b value and l a r g e l y v a r y i n g standard d e v i a t i o n s , we do not expect that the eigenvalues of T arranged i n d e c r e a s i n g order w i l l correspond to i t s d i a g o n a l elements i n t h e i r o r i g i n a l order. The new order r e l a t i o n s w i l l depend on the r e l a t i o n s h i p s between the standard d e v i a t i o n s . For example, a very small o 2 value c o u l d make r v > J the l a r g e s t d i a g o n a l element and consequently, the f i r s t ordered eigenvalue w i l l e s s e n t i a l l y correspond to t h i s element. But the most important r e s u l t of t h i s r e o r d e r i n g of eig e n v a l u e s versus d i a g o n a l elements of T i s the consequent r e o r d e r i n g of the b a s i s f u n c t i o n s i>; . Hence, i n the above example, the l a s t weighted k e r n e l ( a s s o c i a t e d with YKtl) c o u l d become the f i r s t b a s i s f u n c t i o n and consequently small-wavelength model s t r u c t u r e w i l l stem from T h i s i s i l l u s t r a t e d i n F i g u r e B.5 where we have p l o t t e d on Panel (a) some b a s i s f u n c t i o n s f o r the case a,=1, whereas on Panel (b) we have d i s p l a y e d the same b a s i s f u n c t i o n s with O,=10% of the maximum (amplitude) s p e c t r a l value of the 84 i ' t h t r a c e of the data presented i n Fi g u r e 3. We summarize by s t a t i n g that in the process of winnowing b a s i s f u n c t i o n s , the standard d e v i a t i o n values w i l l play a major r o l e in determining the type of inf o r m a t i o n to be included in the c o n s t r u c t e d model. Thus a p p l i c a t i o n of a-, v a l u e s to a given PWD problem should be e x e r c i s e d with c a u t i o n . 85 Panel a 1 3 H E R 7 Z Panel b 0. 3 E ' 0 2 t _ j _ _ L _ i _ ( _ L i — i — L — i O . U E - O S t - 0 . 2 E - 0 3L_i_i—I I 1 1 — J — I — I — I - 0 . 5 E » 0 3 J — i — I — I — I — I — i — J — I — u 2 . 9 0 . 2 . 9 0 . p p f - . n i , , , , . i i i i i 0 . 2 E « 0 3 i — i — i — i — i — i 0 . 3 E - > 0 3 L _ i — j i i i i i i i r 0 . 2 E * 0 3 i ^ - i i — I — i — i — I — i — i — t 25 - 0 . 3 E • 0 3J1 I I I I I—I—1_ 2 . 0 . 2 E » 0 3 i i I i — i — i — i — i - 0 . 2 E - » 0 3 J 25 _i I - 0 . 3 E * 0 3 1 I 1 I I—I—I—I—1— 9 0 . 2 . 9 0 . 0 . U E * 0 3 i i I i i i i — i — I — L _ - 0 . 1 E * 0 3 J -9 0 . 2 . ( H O R I Z O N T A L R X E S I N D E G R E E S ! FIGURE B.5 Ba s i s f u n c t i o n s 1, 13, 25 and 35 corresponding to the forward smallest-model c o n s t r u c t i o n . Standard d e v i a t i o n s are set to (a) 1 f o r a l l input t r a c e s , and (b) percentages of the maximum s p e c t r a l amplitude of each t r a c e (see t e x t f o r d e t a i l s ) . N o t i c e that the r e - o r d e r i n g of the b a s i s f u n c t i o n s i s such that i n (a) the f i r s t b a s i s f u n c t i o n c o n t r i b u t e s with long-wavelength model s t r u c t u r e , whereas i n (b) t h i s b a s i s f u n c t i o n g i v e s short-wavelength s t r u c t u r e . 86 APPENDIX C Inner Product Matrix f o r the Forward Smallest Model C o n s t r u c t i o n The c o n s t r u c t i o n of the forward s m a l l e s t model r e q u i r e s the e v a l u a t i o n of the elements of the inner product matrix X. Using the k e r n e l s from equation (2-9b) we have, r:. = = [l/a,-o ;] f k r K 0 ( k r b ) J 0 ( k r r ; ) J 0 ( k T rj )dk r (C-1) 0 and from Gradshteyn and Ryzhik, 1980, equation 6.578.15 we obta i n , r;- = [ l / a ; a - ] / [ (r?+b 2 + r i 2) 2-4r i 2r j 2]°- 5 (C-2) 87 APPENDIX D The Forward F l a t t e s t Model: G l o b a l Development T h i s appendix d e s c r i b e s the s o l u t i o n s to the set of problems which are encountered in the c o n s t r u c t i o n of the f l a t t e s t - m o d e l forward transform. In p a r t i c u l a r , the f o l l o w i n g problems are undertaken: 1. Computation of the f l a t t e s t model k e r n e l s H;. 2. E v a l u a t i o n of the term H; m°f . 3. C a l c u l a t i o n of the inner product K r 4. S o l u t i o n to the i n d e f i n i t e i n t e g r a l JQ 2H;. 5. E v a l u a t i o n of the i n t e g r a t i o n constant o f equation (2-14). 1. Computation of the f l a t t e s t model k e r n e l s H ;. With the i d e n t i f i c a t i o n , = k r J 0 iK r r )/o: we e v a l u a t e the i n d e f i n i t e i n t e g r a l , H; = [ ] / k ' J 0 ( k ' r ; )dk' Upon change of v a r i a b l e r ' = k'r ; we get, H; = [ \ / a . r , 2 ] / r ' J 0 ( r ' ) d r ' and from Gradshteyn and Ryzhik, 1980, equation 5.56.2 we o b t a i n , H ; = [ l / a ; r ; ] k r J , ( k r r ; ) (D-1) 88 2. E v a l u a t i o n of the term H ;mj . Using (D-1) and the asymptotic behaviour of J , ( k r r r ) f o r l a r g e argument, the f i r s t term of the r.h.s. of equation (2-11) fo r kr-»°° behaves as, U(oj,kr-°°) l i m {[ 1/ff, r,- ] k r [ 2 A k r r , ] ^ } = U(u,k r») l i m {2k r , , 2/ff, 7r1'2r;3/2} (D-2) T h i s e x p r e s s i o n goes to zero provided U(a>,kr—°°) goes to zero f a s t e r than k r " 1 / 2 f o r a l l f r e q u e n c i e s (because the modified B e s s e l f u n c t i o n s K 0(x) and K,(x) are not d e f i n e d at x=0 we exclude both o>=0 and 7=0°). R e c a l l i n g that k r=wsin (y)/V, and c o n s i d e r i n g a b a n d - l i m i t e d source, kr-*°° represents i n f i n i t e l y -a t t e n u a t e d inhomogeneous waves. In the present work we set • H; mT = 0 Then we w i l l be concerned with models U(w,k r) which go to zero f a s t e r than k r~ 1' 2 as kr-»°°. 3. C a l c u l a t i o n of the inner product matrix. With Hj as given i n (D-1) and Q = K,°- 5(k rb), the e n t r i e s of the inner product matrix T are given by, r;- = = [ 1/r- r- o.o- ]|kr2K, ( k r b ) J , ( k r r . ) J , ( k r r j )dk r and from Gradshteyn and Ryzhik, 1980, equation 6.578.15 we have, ri-=[\/oios] 4 b / [ ( r 2 + b 2 + r . 2 ) 2 - 4 r 2 r 2 ] 1 - 5 (D-3) 89 K r 4. E v a l u a t i o n of the i n d e f i n i t e i n t e g r a l JQ 2H-. The s o l u t i o n to the i n d e f i n i t e i n t e g r a l appearing i n the l . h . s . of equation (2-15) can be d e r i v e d from the f o l l o w i n g ( i n d e f i n i t e ) i n t e g r a l given i n Gradshteyn and Ryzhik, 1980, equation 5.54.1, J y Z p ( a y ) B P ( c y ) d y = {cyZ ?(ay )B*_, (cy) - ay Z?_, (ay ) B P (cy)}/[ a 2 - c 2 ] (D-4) where Zv and B ? are any B e s s e l f u n c t i o n of order p, and a and c are c o n s t a n t s . For p=1 we i d e n t i f y , Z, = J , B, = H,"' H , ( 1 ) i s the f i r s t Hankel f u n c t i o n of order one. (D-4) now reads as, J y J , ( a y ) H , ' 1 ' ( c y ) d y = {cyJ,(ay)H 0< 1 ' ( c y ) - a y J 0 ( a y ) H , ( 1 ' ( c y ) } / [ a 2 - c 2 ] (D-5) but from Gradshteyn and Ryzhik, 1980, equations 8.407.1 and 8.407.2 we have, H 0 ( 1 ) ( i b y ) = -[2/TT] i K 0 ( b y ) (D-6a) H , ( 1 » ( i b y ) = -[2/TT] K,(by) • (D-6b) Using (D-6) and i d e n t i f y i n g c=ib, (D-5) transforms to, -[2/TT] J y J , (ay)K, (by)dy = { ( i b ) y J , ( a y ) [ - 2 / i r i K 0 ( b y ) ] - ayJ 0(ay)[-2/w K,(by)]} / [ a 2 - ( i b ) 2 ] 90 that i s to say, f y j , ( a y ) K , ( b y ) d y = - {byJ,(ay)K 0(by) + a y J 0 ( a y ) K , ( b y ) } / [ a 2 + b 2 ] (D-7) from which a f t e r making y = k r and a = r-( we o b t a i n the r.h.s. of equation (2-15). 5. The i n t e g r a t i o n constant of equation (2-14). We show that with the c o n d i t i o n U(u, kr-»°°) = 0 the constant of i n t e g r a t i o n C i s equal to zero. From equations (2-14), (D-1) and (D-7) (with y = k r and a = r,- ) we see t h a t , C = -Z[0,-/a ;rj ] [ l / ( r ? + b 2 ) ] { 1 im{ bk r J , ( r r k r ) K 0 (bk r )} + l i m { r ; k r J 0 ( r r k r)K, (bk r)} } +U(co,k r^») (D-8) Asymptotic expansions f o r l a r g e arguments of the i n v o l v e d B e s s e l f u n c t i o n s and the m o d i f i e d B e s s e l f u n c t i o n s read as (Abramowitz and Stegun, 1970), J 0 ( r ( . k r ) ~ [2/7rr ; k r ] 0 - 5 {cos(r,.k r) + 0 ( | r - k r | - 1 )} J , ( r ; k r) ~ [2/jrr f k r ] °-5 { s i n ( r ; kr ) + 0( | r. kr | " 1 ) } K 0 ( b k r ) ~ [7r/2bk r] 0- 5 e x p ( - b k r ) { l + 0 ( | b k r | - 1 ) } K,(bk r) ~ [7r/2bk r]°- 5 e x p ( - b k r ) { l + 0 ( | b k r | - 1 ) } T h e r e f o r e , the l e a d i n g behaviours of the l i m i t i n g e x p r e s s i o n s i n (D-8) are given by, l i m {bk r [2/irv, k r ] °-5 [ 7r/2bk r ] °-5 exp(-bk r)} + l i m {r f k r [ 2 / 7 r r ; k r ] ° - 5 [ 7 r / 2 b k r ] 0 - 5 exp(-bk r)} = 0 (D-9) that i s to say, C = 0 91 APPENDIX E Inner Product M a t r i c e s f o r the I n v e r s e S m a l l e s t and F l a t t e s t Model C o n s t r u c t i o n s The c o m p u t a t i o n of the i n n e r p r o duct m a t r i c e s f o r the i n v e r s e s m a l l e s t and f l a t t e s t problems p a r a l l e l s t h a t made f o r the f o r w a r d problems. Indeed, by r e p l a c i n g r ; by k r ; as the parameter and kr by r as the independent v a r i a b l e a p p e a r i n g i n the f o r m u l a t i o n of the f o r w a r d problems, one r e a d i l y o b t a i n s the i n n e r p r o d u c t m a t r i c e s f o r the i n v e r s e problems. In p a r t i c u l a r , the e n t r i e s of the T m a t r i x f o r the s m a l l e s t i n v e r s e t r a n s f o r m are g i v e n by, rVj(ej) = [Wo; 0 j ] / [ ( k 2 + b 2 + k 2 ) 2 - 4k r 2 k 2 . ] 0 - 5 (E-1) A f t e r the s u b s t i t u t i o n s k r (=cjsin (7,- ) / V and b=ojpc =a>sin (c ) / V , where c i s a c o n s t a n t such t h a t s i n ( c ) > 0 , and V i s the s u r f a c e P-wave v e l o c i t y , (E-1) y i e l d s r ; j(cj) = V 2 / C J 2 {[ I/a; c j j ] / [ ( s i n 2 (7,. ) + s i n 2 ( c ) + s i n 2 ( 7 j ) 2 -4 s i n 2 ( 7 ; ) s i n 2 ( 7 i ) ] 0 - 5 } . We w r i t e rr(u) = V 2 / G J 2 F • (E -2) where, r r = [ 1/of a; ]/[ ( s i n 2 (7, ) + s i n 2 ( c ) + s i n 2 (7. ) 2 -4 s i n 2 (7- )-sin 2 (75 ) ] 0 - 5 92 Since the frequency-dependent term i n (E-2) i s f a c t o r e d out of the inner product matrix, the inverse r~ 1 (a>) = C J 2 / V 2 T" 1 i s c a l c u l a t e d with minimal computational e f f o r t . S i m i l a r l y , the e n t r i e s of the inner product matrix for the in v e r s e f l a t t e s t problem are given by, r,.(«) = [ l/o ; o: ]4b/[ ( k 2 + b 2 + k 2 ) 2 - 4 k r 2 k 2 ] 1 - 5 (E-3) Upon the s u b s t i t u t i o n s b = c j s i n ( c ) / V and kr- =u>sin(7r ) / V we o b t a i n from ( E - 3 ) , rv(cj) = V 5 / u 5 r ; j where T ; j = [ i/o; O j ] 4 s i n ( c ) / [ ( s i n 2 (7,- ) + s i n 2 ( c ) + s i n 2 (y± ) -4 s i n 2 (7 r ) s i n 2 (7^ ) ] '-5.