AP PLI CA TION OF 2-D FOR WARD SEIS MIC MOD EL LING FOR IM PROVED IM AG ING OF SUB-SALT ROTLIEGEND STRATA IN POL ISH BASIN

Annales Societatis Geologorum Poloniae (2013), vol. 83: 65 80. AP PLI CA TION OF 2-D FOR WARD SEIS MIC MOD EL LING FOR IM PROVED IM AG ING OF SUB-SALT ROTLIEGEND STRATA IN POL ISH BASIN Pawe³ MARZEC, Magdalena NIEPSUJ, ukasz S ONKA & Kaja PIETSCH AGH Uni ver sity of Sci ence and Tech nol ogy, Fac ulty of Ge ol ogy, Geo phys ics and En vi ron ment Pro tec tion, De part ment of Geo phys ics, Kraków, Po land; emails: marzec@agh.edu.pl, magdalena, niepsuj@gmail.com, lukasslonka@gmail.com, pietsch@agh.edu.pl Marzec, P., Niepsuj, M., S³onka,. & Pietsch, K., 2013. Ap pli ca tion of 2-D for ward seis mic mod el ling for im proved im ag ing of sub-salt Rotliegend strata in Pol ish Basin. Annales Societatis Geologorum Poloniae, 83: 65 80. Abstract: For ward seis mic mod el ling can aid seis mic stud ies of the pre-zechstein strata in ar eas of de vel oped salt tec ton ics, such as the Obrzycko Szamotu³y re gion, NW Pol ish Ba sin. The re sults not only can be used for seis mic in ter pre ta tion, but also can sup port the plan ning of sur vey meth od ol ogy and the workflow of seis mic data pro cessing. This pa per pres ents the re sults of mod el ling that was car ried out, be fore the ac qui si tion of the re gional-scale, seis mic line Obrzycko-1 Zabartowo-1 Zabartowo-2 (Górecki, 2010). An in ter preted, seis mic transect was used to build a ba sic, seis mic-geo log i cal model. The mod el ling was based on seis mic ray the ory. The zero-off set mod el ling (the o ret i cal wave field) for dif fer ent ge om e tries of salt struc tures showed that an in crease in salt thick ness re sulted in a pull-up of re flec tion events, re lated to the sub-salt ho ri zons. The in cor po ra tion of faults and salt over hangs into a model sig nif i cantly com pli cated the seis mic wave field. The re sults of off set mod el ling, pre - sented in this pa per as seis mic ray trac ing and com mon-shot gath ers, proved that (1) the seis mic re sponse of the Rotliegend (Perm ian) for ma tions can be re corded, de spite the pres ence of the over ly ing salt pil lows and diapirs, if off sets sev eral kilo metres long are used, and (2) the com plex con fig u ra tion of seis mic re flec tors (diapirs with salt over hangs, faults) gives rise to com pli cated, seis mic ray paths that may cause dif fi cul ties in com mon-depth-point stack ing and there fore de crease the qual ity of the seis mic re cords. Key words: salt and subsalt struc tures, salt diapirs, seis mic meth ods, seis mic mod el ling, Perm ian, Po land. Manu script re ceived 17 October 2012, ac cepted 10 July 2013 INTRODUCTION Dif fi cul ties in seis mic ex plo ra tion for hy dro car bon re - sources, as so ci ated with the eolian Rotliegend (Perm ian) de pos its, are of ten re lated to the pres ence of the over ly ing Zechstein salt, i.e. salt pil lows and diapirs. Ex ten sive salt tec ton ics gave rise to strong, seis mic re flec tors, char ac ter - ized by com plex ge om e try. As a re sult, the dis tri bu tion of seis mic ray paths is highly non-uni form, which in turn means that seis mic im ag ing does not pro vide a re li able model for the pre-zechstein strata. If this is the case, seis mic mod el ling can sup port seis mic ex plo ra tion in ev ery stage, i.e., seis mic sur vey de sign, data pro cess ing and in ter pre ta - tion (Pietsch et al., 2007a, b, 2008; Kobylarski et al., 2008). For ward seis mic mod el ling, test ing dif fer ent ge om e - tries in the subsurface, was car ried out along the re gionalscale, seis mic test line, Obrzycko-1 Zabartowo-1 Zabartowo-2 (Górecki, 2010), lo cated in the SW part of the Pol ish Ba sin, in the Mid-Pol ish Trough. The ob jec tives of the mod el ling were: (1) test ing of the pro posed sur vey meth od - ol ogy, (2) in ves ti ga tion of the re la tion ship be tween seis mic wave field vari a tions and the na ture of the over ly ing salt struc tures, and (3) in ter pre ta tion of the geo log i cal struc ture of the salt diapir, sam pled by the seis mic transect. The ba sic seis mic and geo log i cal model was cal cu lated based on 2-D seis mic data: the Obrzycko Szamotu³y seis - mic lines, ac quired in 2001 and 2007 by Geofizyka Toruñ POGC Group, and the Murowana Goœlino Klecko seis mic lines, ac quired in 2008, by the Geofizyka Kraków POGC Group (Fig. 1). The seis mic pro files were sit u ated to the SW of the Mid-Pol ish Trough (Fig. 1). In this area, the sys tem of salt diapirs (Drawno Dzwonowo Cz³opa Szamotu³y) was de scribed by Krzywiec (2006a). The seis mic pro file T009 4207 13-1-08K, of the SW NE ori en ta tion (Fig. 1B), was se lected for anal y sis. From the SW to NE, the pro file crosses the Obrzycko Salt Pil low, the Szamotu³y Salt Diapir, and the RogoŸno Salt Pil low (Wag ner, 2012).

66 P. MARZEC ET AL. 2000). At that time, the Pol ish Trough was the depocentre of a sa line ba sin, in which over 1500 m of rock salt were de - pos ited (Wag ner, 1994). The struc ture of this area is dom i - nated now by in tense salt tec ton ics. The pres ence of a thick evaporite com plex at the base of the sed i men tary infill re - sulted in the re gional de tach ment of the Me so zoic sed i men - tary suc ces sion from the sub-zechstein strata and the for ma - tion of faults and folds, as well as pil lows and salt diapirs, ge net i cally re lated to the fault zones (Krzywiec, 2006a; Krzywiec et al., 2006). In tense salt mo bil ity in the Late Cre - ta ceous marked the ini tial stage of re gional in ver sion of the Mid-Pol ish Trough (Wag ner, 1994; Dadlez et al., 1997), which re sulted in the Mid-Pol ish Swell ul ti mately be ing up - lifted at the bound ary of Cre ta ceous and Palaeogene. The up lift re sulted in the in ver sion of lo cal, tec tonic zones, such as the Drawno Cz³opa Szamotu³y Zone (Leszczyñski, 2002; Krzywiec, 2006b). Post-Maastrichtian ero sion of Ju - ras sic to Cre ta ceous strata (Dadlez, 2000) con trib uted to the for ma tion of the pres ently known salt struc tures (Krzywiec 2000, 2009). The in ver sion pro cess re ac ti vated pre-ex ist ing faults, the mo bil ity of which trig gered the growth of salt struc tures and con trolled their ge om e try (Dadlez et al., 1998, Krzywiec 2000, 2009). In the seis mic sec tion ana lysed (Fig. 2), there are three salt struc tures, prob a bly re lated ge net i cally (Wag ner, 2012): (1) the small Obrzycko Salt Pil low, the evo lu tion of which was halted by faults, bor der ing the Szamotu³y Zone; (2) the Szamotu³y Zone, which is part of Me so zoic synsedimentary grabens and half-grabens, de vel oped dur ing the early Keuper to Early Ju ras sic, in re sponse to sub si dence of the Mid-Pol ish Trough (Kwolek, 2000); and (3) the RogoŸno Salt Pil low, lo cated in the NE part of the transect, with a salt ho ri zon, al most en tirely squeezed out in a SW di rec tion. SEISMIC MODELLING Fig. 1. Out line of ge ol ogy of study area. A Map of salt struc - tures within Me so zoic cover of Mid Pol ish Trough (Dadlez and Mark, 1998, mod i fied). B Lo ca tion map for seis mic (1) and bore - hole (2) data. Seis mic data: seis mic transect T0094207 13-1-08K (red line); seis mic pro files, used in struc tural in ter pre ta tion (blue lines); re gional, seis mic pro file Obrzycko-1 Zabartowo-1 Zabartowo-2: AGH28511 (black line). Bore hole data: well with PA mea sure ment (red point); well with v mea sure ment (green point) (Pietsch et al., 2010, 2012) GEOLOGICAL SETTING The study area is lo cated in the NW part of the Permo- Me so zoic Pol ish Ba sin, an east ern seg ment of the Cen tral Eu ro pean Ba sin Sys tem (Ziegler, 1990:, Pha raoh et al., 2010). The Pol ish part of this sys tem, in clud ing the Mid- Pol ish Trough, orig i nated at the be gin ning of Rotliegend sed i men ta tion. Sub si dence ini tially re sulted from con ti nen - tal rift ing dur ing the Early Perm ian, which was fol lowed by a post-rift phase, start ing in the Zechstein (Van Wees et al., The seis mic mod el ling was per formed, us ing Divestco s Outrider soft ware, based on seis mic ray the ory, and cal cu lated 2-D seis mic pro files, both as field re cords and stacked data. The soft ware can be used for ray trac ing and cal cu lat ing the travel times of re flected, compressional (PP) and con verted waves (PSV), as well as di rect (in the case of mod el ling VSP data) and dif fracted waves, us ing the dif - frac tion func tion (al go rithm plac ing source points on seis - mic re flec tors). Mul ti ple waves can not be modelled with soft ware. The full Zoeppritz equa tions were used for cal cu - lat ing re flec tion and con ver sion co ef fi cients. Thus sim u la - tion of am pli tude vari a tion with off set was pos si ble. The mod el ling firstly was done, us ing a zero-off set method that aimed at (1) re solv ing the vari abil ity in seis mic im age of the Rotliegend strata, in re la tion to the Zechstein over bur den, and (2) re li able in ter pre ta tion of the Szamotu³y Salt Diapir struc ture. The fol low ing mod el ling by means of off set meth ods was car ried out to test the po ten tial for im ag - ing the pre-zechstein strata, us ing the sur vey meth od ol ogy, pro posed for the ac qui si tion of the re gional Obrzycko-1 Zabartowo-1 Zabartowo-2 pro file (Górecki, 2010). The struc ture of the Permo-Me so zoic suc ces sion, im - aged by a seis mic pro file, re veals from SW to NE an al most

APPLICATION OF 2-D FORWARD SEISMIC MODELLING 67 F ig. 2. I nterpreted seismic transect T0094207 13-1-08K after migration (Pietsch e t al., 2010, 2012). Seismic boundaries: M top Miocene, K3st top Upper Cretaceous, K3a Santonian, K t Turonian, Kc Cenomanian, K1 Lower Cretaceous; J3 Malm, J2 Dogger, J1 Lias; T3 Rhaetian, Tk3G Upper Keuper (Upper Gypsum Series), Tk3D Lower Keuper (Lower Gyp - sum Series), T2 Muschelkalk, T1 Roethian, Tp2 Middle Buntsandstein, Tp1 Lower Buntsandstein, P3 Zechstein, A3 Leine anhydrite, A2 Stassfurt anhydrite, P1 Rotliegend (Saxonian), Pa Rotliegend (Autunian); C top Carboniferous flat-par al lel con fig u ra tion of beds, with the small Obrzycko Salt Pil low, the com plex Szamotu³y Zone (mostly re verse faults and large squeeze out of salt), and fi nally the large RogoŸno Salt Pil low (Fig. 2). The ba sic, seis micgeo log i cal model was built, us ing the interpreted, seis mic transect (Fig. 2), ex cept for the Szamotu³y Zone (Pietsch et al., 2010, 2012). The dis rupted lay out of seis mic bound aries, observed in this zone, pre cludes any de fin i tive, geo - logical interpretation of the seismic record. For this rea son, sev eral mod els for the Szamotu³y Zone were pro posed and a layer-block struc ture of the Szamotu³y Zone was as sumed in the ba - sic model (Fig. 3). For the suc ces sion start ing from the top of the Up per Keuper (Tk3G) and con tin u ing to the top of the Zechstein (P3), the ve loc ity model was con structed, us ing a depth-de pend ent velocity gra di ent (Wilk, 2010; Pietsch et al., 2012). Seismic velocities for the remaining, geological successions, i.e. the Cretaceous, Jurassic, Rheatian and Perm ian (P2 and P1), were de ter mined from seismic inversion (Pietsch et al., 2010, 2012). Barely vis i ble bound aries be low the Zechstain beds have been in ter preted, us ing available, geological information (Bujak, 2010). The ba sic, seis mic-geo log i cal model in a depth do main, which was cal cu lated on the ba sis of the data de scribed above, is pre sented in Fig ure 4. ZERO-OFFSET MODELLING The aim of zero-off set mod el ling was mainly to de scribe the vari abil ity in a seis mic wave field, in re la tion to dif fer ent mod els of the Zechstein salt struc tures. Dur ing the sub se quent it er a tions, changes to the seis mic-geo log i cal model were in tro duced, which re sulted in mod - els rang ing from a strat i fied (Fig. 4A) to a diapiric salt struc ture (Figs 5A 8A). A the o ret i cal wave field was cal cu lated, us ing Ricker-type wave lets of 26 Hz fre quency (best ap prox i mat - ing re corded sig nal) with 20 m dis tances be - tween traces (Pietsch et al., 2010). Cal cu la tions uti lised the dif frac tion method (ex plod ing re - flector algorithm). The synthetic seismic wave field sim u lates the stacked, unmigrated, seis mic sec tion with wave dif frac tions. To con struct the theoretical profiles, the finite-difference poststack mi gra tion was used. The pa ram e ters for FD mi gra tion were: max i mum dips of 45 65 and an RMS ve loc ity field in the range 50 100% (S³onka, 2013). The initial, seismic, geological model (Fig. 4A), im aged in the depth do main, is based on the as sump tion that the inter-fault zone is as so - ci ated with the strat i fied ar chi tec ture of both the

68 P. MARZEC ET AL. Fig. 3. Part of seis mic transect, in which layer-block struc ture Szamotu³y Zone (model I) was es tab lished. Seis mic bound aries as in Fig. 2 Rotliegend suc ces sion (model I) and the over ly ing strata (in clud ing salt). The syn thetic, seis mic wave field, cal cu - lated for this model, is shown in Fig ure 4B. The fol low ing mod els as sume that a salt diapir grew be - tween mar ginal, Palaeozoic-rooted faults of the Szamotu³y Zone. In model II (Fig. 5A), the rock salt pen e trates the Lower Keuper and the Bunter Sand stone strata in the SW and NE parts of the Szamotu³y Zone, re spec tively. The pos - tu lated salt struc ture cor re lates in the ana lysed transect with do mains of un clear or partly miss ing re flec tions (Fig. 2). The wave field, cal cu lated for this model, us ing the dif frac - tion method and af ter the mi gra tion pro ce dure, is shown in pan els B and C re spec tively of Fig ure 5. The same out line is used for mod els, shown in Fig ures 6 8. Model III as sumes a salt diapir, ris ing up to the Mid dle Ju ras sic in the SW part of the Szamotu³y Zone (Fig. 6A). In model IV, a salt diapir pen e trates the Up per Ju ras sic and has two over hangs (Fig. 7A). The top of the dome is cut by small faults, which ter mi nate within the salt body. A gyp - sum-anhydrite cap rock was sim u lated by in creas ing the seismic velocity, associated with the uppermost part of the salt diapir. The last model ana lysed (model V) was con structed on the ba sis of the petrophysical pa ram e ters, es ti mated for the ini tial seis mic-geo log i cal model (Fig. 3A), al though the ge - om e try of seis mic re flec tors in the inter-fault zone was mod - i fied. A large salt diapir with two over hangs was tested in this model with a shape, re sem bling that of the Szamotu³y Salt Diapir as in ter preted by Krzywiec (2006a), on the seis - mic line, lo cated to the NW of the transect ana lysed (Fig. 8A). A com par i son of the syn thetic, seis mic re cords, gen er - ated for the mod els ana lysed (Figs 4B, 5B, 6B, 7B, 8B), shows the dif fer ence be tween the seis mic wave fields, as so - ci ated with the as sumed salt struc ture. This dif fer ence is re - lated to the higher ve loc ity con trast be tween lat er ally ad ja - cent lithologies. In the ini tial, strat i fied model I, the pat tern of re flec tion events re lated to the pre-zechstein strata is gen er ally sim i lar in the depth (Fig. 4A) and time (Fig. 4B) sec tions. If a salt struc ture does not pierce the top Muschelkalk (Mid dle Tri as sic), it causes no sig nif i cant dis tor - tion in the im ag ing of the pre-zechstein bound aries. This is re lated to the small ve loc ity con trast be tween the salt bod ies and the Lower and Mid dle Tri as sic for ma tions (Fig. 4A). In the sub se quent mod els, as sum ing the de vel op ment of a salt diapir (Figs 5A, 6A, 7A, 8A), the in crease in the thick ness of the salt re sults in a clear pull-up of the seis mic re flec tion events, the top Rotliegend and Car bon if er ous, be low the salt body (Figs 5B, 6B, 7B, 8B). This is as so ci ated with much higher, seis mic ve loc i ties in a salt body, com pared to the sur round ing strata. In such a case, an an ti cli nal pat tern of re - flec tion events is dis played in the time sec tion for the Rotliegend suc ces sion. The dif frac tion en ergy of waves, orig i - nat ing at points where the salt diapir pierces the over ly ing seis mic re flec tors, is also in creased, ow ing to higher ve loc - ity con trasts. This re sults in higher en ergy dis si pa tion and also in re duc tion of the en ergy of waves, reach ing the pre- Zechstein strata. A com par i son of the syn thetic, seis mic wave fields, cal cu lated for model IV (Fig. 7A, B) and model V (Fig. 8A, B), proves an im por tant role of faults in gen er at - ing strong dif frac tion waves. Since the mod elled salt diapirs were as sumed to be iso tro pic bod ies, there are no dif frac tion waves in the syn thetic, seis mic sec tions that can orig i nate within them. The theoretical wave fields, calculated for various, seis - mic-geo log i cal mod els, af ter wards were used in the geo log - i cal in ter pre ta tion of the Szamotu³y Salt Diapir. The wave fields, cal cu lated for the mod els with large salt diapirs (Figs 7, 8), are sig nif i cantly dif fer ent from the wave field re - corded (Fig. 2). In the the o ret i cal field, the zones with out re - flec tions reach the top of the Ju ras sic for ma tion (Figs. 7, 8). How ever, in the seis mic data the ex tent of these zones is smaller. A seis mic re cord from the Szamotu³y Zone (Fig. 2) in di cates the pos si ble pres ence of a small, dou ble salt diapir. Fig ures 5A, C, and 6A, C show seis mic-geo log i cal mod els and cal cu lated wave fields for the pos si ble salt struc ture. The seis mic bound aries from both mod els, re duced to the time do main, were over laid onto the reg is tered wave field (Fig. 9A, B). Model II (Fig. 9A) shows a better match of the bound aries with the reg is tered events, com pared to model III. For ex am ple, the mag ni tude of the pull-up ef fect within the pre-zechstein strata in the mod elled and reg is tered fields is com pa ra ble (Fig. 9A). In the case of model III, the mod elled pull-up ef fect is sig nif i cantly larger than in the re - corded field (Fig. 9B). A higher pull-up ef fect is caused by the lat eral gra di ent in crease of av er age ve loc ity, when the salt diapir pierced for ma tions with lower ve loc i ties (Fig. 6A).

APPLICATION OF 2-D FORWARD SEISMIC MODELLING 69 F ig. 4. S eismic-geological depth model (I) along seismic transect ( A ) ; zero-offset synthetic section (calculated in version: diffraction ) ( B ) Seismic boundaries as in Fig. 2

70 P. MARZEC ET AL. F ig. 5. Seismic-geological depth mode l (II), with salt jammed in inter-fault zone (A ) ; zero-offset synthetic section, calcu - l ated in version: diffraction ( B ); seismi c p rofile after migration FD ( C ). Seismi c boundaries as in Fig. 2 F ig. 6. Seismic-geological depth mode l ( III), with two-part salt diapir ( A ). B a nd C as in Fig. 5, seismic boundaries as in Fig. 2 F ig. 7. Seismic-geological depth mode l (IV) with salt diapir with a corniche and o verhang ( A ). B a nd C as in Fig. 5, seismi c boundaries as in Fig. 2 F ig. 8. Seismic-geological model (V ) w ith Szamotu³y Salt Diapir ( A ). B a nd C as in Fig. 5, seismic boundaries as in Fig. 2

APPLICATION OF 2-D FORWARD SEISMIC MODELLING 71 Fig. 9. Struc tural in ter pre ta tion of Szamotu³y Zone, us ing seis mic bound aries dis tri bu tion for model II (Fig. 5A) (A) and model III (Fig. 6A) (B). Seis mic bound aries as in Fig. 2 OFFSET MODELLING The main ob jec tive of the off set mod el ling was to as - sess the po ten tial for im ag ing the pre-zechstein strata bound aries with the sur vey meth od ol ogy, orig i nally proposed for the ac qui si tion of the re gional-scale test pro file Obrzycko-1 Zabartowo-1 Zabartowo-2 (Górecki, 2010). The re sults of off set mod el ling are pre sented as seis mic ray paths, gen er ated for the set of source points, and com monshot gath ers, which pres ent the cal cu lated travel-time curves, convolved with the source sig nal. The re la tion ship be - tween the am pli tude and the an gle of in ci dence of seis mic rays was in cluded in the cal cu la tions. The re sults can be used, both for de sign ing the sur vey meth od ol ogy and for plan ning and op ti mis ing the seis mic data pro cess ing. Off set mod el ling was car ried out for the same seis - mic-geo log i cal mod els as de scribed above, i.e., rang ing from the strat i fied model (Fig. 4A) to the model, re sem bling the Szamotu³y Salt Diapir (model V, Fig. 8A). The ray paths for model I (Fig. 4A) were gen er ated for two sources at PS 32000 and PS 17000. Point PS 32000 was lo cated above the salt pil low, while point PS 17000 lied above the fault zone. Con se quently, they were lo cated in the po si tions with the great est im pact on the seis mic im age of the pre-zechstein strata. The dis tri bu tion of seis mic ray paths, gen er ated at point PS 32000, is pre sented in Fig ure 10A, while Fig ure 10B shows part of a seis mic-geo log i cal model, with the seismic reflectors identified by seismic rays. The analysis of both fig ures proves that if the pro posed, seis mic spreads of 4500-0-4500 [m] and 6500-0-6500 [m] are used for ac qui si - tion of the re gional-scale, seis mic line (Górecki, 2010), the ma jor ity of seis mic ray paths, re flected from the pre-zechstein in ter faces, should be re corded. The syn thetic com monshot gather, gen er ated for the pro posed seis mic spreads, is shown in Fig ure 10C. The the o ret i cal travel-time curves, marked with colours, which cor re spond to the colours in di - cat ing the re spec tive seis mic re flec tors in the mod els, show that the Zechstein and Rotliegend re flec tors also can be im - aged, us ing a seis mic spread with lon ger off sets (ex ten sions of the travel-time curves, es ti mated for the seis mic spread 4500-0-4500 [m]). The ar range ment of ray paths, gen er ated at point PS 17000, lo cated above the fault zone, is shown in Fig ure 11A, while the syn thetic com mon-shot gather for this source is pre sented in Fig ure 11B. A sim i lar ap proach was used for mod el ling of the salt dome, which re sem bles the Szamotu³y Salt Diapir (model V in Fig. 8A). The seis mic ray paths, gen er ated for the source lo cated above the top part of the salt diapir (PS 17000), are shown in Fig ure 12A, while the com mon-shot gather, cal cu - lated for this point, is pre sented in Fig ure 12B. Ad di tion ally, ray paths were cal cu lated for model IV (Fig. 8A), i.e., the large salt diapir with two over hangs (Fig. 13). The pat tern of seis mic ray paths, as so ci ated with such a com plex seis - mic-geo log i cal model, as well as the cal cu lated travel-time curves, in di cate po ten tial prob lems, as so ci ated with the in - ter pre ta tion of deeper re flec tors. The anal y sis of the cal cu - lated re cords, which was car ried out in or der to eval u ate the po ten tial for im ag ing of the pre-zechstein strata, shows that the least re li able re sults can be re corded with re ceiv ers, which are ad ja cent to a source point, lo cated above the salt diapir. This is be cause the ray paths, which are ver ti cal, en - coun ter steeply in clined, seis mic bound aries (fault planes,

72 P. MARZEC ET AL. Fig. 10. Off set mod el ling: part of ba sic, seis mic-geo log i cal model I (RogoŸno Salt Pil low), PS 32000. A Seis mic ray tra jec to ries; B Light ing bound aries seis mic rays; C Seis mic re cord. Hodographs of re flec tion waves, marked with colours: Ju ras sic bound aries blue; Tri as sic bound aries vi o let; Zechstein bound aries or ange; in side Rotliegend bound ary brown, and bot tom of Car bon if er ous black

APPLICATION OF 2-D FORWARD SEISMIC MODELLING 73 Fig. 11. Off set mod el ling: part of ba sic, seis mic-geo log i cal model I (Szamotu³y Zone), Ps 17000. A Seis mic ray tra jec to ries; B Seis - mic re cord. Colours marked as in Fig. 10C slopes of the dome), which they reach at a large an gle. Thus, they may be re fracted eas ily (large ve loc ity con trasts), with - out be ing trans mit ted into the un der ly ing strata. The al go - rithm sim u lates this phe nom e non by omit ting such ray paths. The ray paths, gen er ated for lon ger off sets, even lon - ger than the range of seis mic spread, in di cate the po ten tial for im ag ing of the deep re flec tions (see Figs 11B, 12B, 13). The anal y sis of seis mic il lu mi na tion of the deeper re - flec tors with waves, gen er ated in the cen tral part of the de - formed zone (PS 17000), was car ried out for each of the seis mic-geo log i cal mod els. It shows that the poor est cov er - age of the bot tom re flec tors in the mod els is as so ci ated with the strat i fied model (model I; Fig. 11A). This re sults from the pres ence of a large num ber of faults, which transect all strati graphic units. Be cause a seis mic sig nal is re flected from the fault planes, it does not reach the deep est re flec - tors. Slightly better seis mic il lu mi na tion of the deep re flec - tors is ap par ent in model IV, which as sumes that the salt diapir is as so ci ated with two over hangs (Fig. 13). In this model, the deep, seis mic re flec tors are much better im aged,

74 P. MARZEC ET AL. Fig. 12. Off set mod el ling: Szamotu³y Salt Diapir model (V), PS 17000. A Seis mic ray tra jec to ries; B Seis mic re cord. Colours marked as in Fig. 10C par tic u larly on the NE side of the salt diapir. The lack of re - flec tions on the SW side of the salt struc ture is mainly due to the pres ence of a large salt over hang, in trud ing into the Tri - as sic. Be cause of the large ve loc ity con trast, the ray paths in ci dent on the top of the over hang, are de flected and reach the ground sur face be yond the ex tent of the seis mic spread. The best, seis mic il lu mi na tion of the deep re flec tors is un - doubt edly in model V, which re sem bles the Szamotu³y Salt Diapir (Fig. 12A). This is clearly be cause there are no large faults in the model, which would oth er wise dis perse the seis - mic rays. How ever, the pres ence of the salt over hang casts a shadow on the Lower Zechstein and deeper re flec tors. However, not only the seismic illumination of reflectors is im por tant for the ef fec tive ness of the seis mic method. The ge om e try of ray paths is also im por tant. The anal y sis of the ray paths, reach ing the seis mic spread, in di cates that the ray tra jec to ries are in many cases very com plex. Be cause of this, com mon-depth-point (CDP) stack ing def i nitely will be more prob lem atic. Cor rect data stack ing will re quire an in - di vid ual ap proach to all pro cess ing steps, in or der to con -

APPLICATION OF 2-D FORWARD SEISMIC MODELLING 75 Fig. 13. Off set mod el ling: salt diapir, with two over hangs model (IV), PS 17000: Seis mic ray tra jec to ries struct seis mic time pro files. Pro cess ing al go rithms, such as DMO and Kirchhoff pre-stack depth mi gra tion (PreSDM) or, better yet, the most mod ern meth ods such as wave equa - tion migration (Farmer et al.,1996), re verse time mi gra tion (Jones, 2008) and beam migration (Farmer, 2006) will have to be applied. Some prob lems that may oc cur, when pro cess ing data from ar eas of halokinesis, are il lus trated by the com par i son of the seis mic ray paths, gen er ated at points above both sides of the Szamotu³y Salt Diapir, with the ray paths, sim u - lat ing CDP gather data. The mod el ling uti lised the pro gram Outrider, which does not sim u late CDP gath ers, al though it can be used to sim u late dif frac tion ray paths. Thus, dif frac - tion rays were sim u lated, in or der to ap prox i mate the CDP gath ers. Fig ures 14 and 15 show the seis mic ray paths, gen - er ated on both sides of the salt diapirs, at points PS 15000 (SW side of the salt struc ture) and PS 20000 (NE side). In both cases, the salt over hang formed a dis tinct ob sta cle for the prop a ga tion of seis mic waves. The dis tri bu tion of seis - mic rays, ob tained in the way dis cussed above, ob vi ously dif fers from the real ray paths. Outrider can not be used to sim u late rays, re flected mul ti ple times, but only rays re - flected a num ber of times from the steep bound aries of the salt diapir and the sur round ing strata on its re turn path to the sur face. Such prob lems are not ex pected in ar eas, un der lain by salt pil lows. Sim u la tions of seis mic ray prop a ga tion from point PS 30000 (Fig. 16A), de scribed by co or di nate 30000 and lo cated on top of the Rotliegend (dif frac tion rays, Fig. 16B), showed a fairly reg u lar dis tri bu tion of both the in ci - dent and dif fracted rays, which ap prox i mate the CDP gather. In such a case, CDP stack ing should not cause any dif fi cul ties. The dis tor tions, which are ap par ent around the point, de scribed by co or di nate 25000, re sulted from the pres ence of faults which reach the Lower Zechstein. These con clu sions are con firmed by mod el ling, done for points PS 5000, 10000, 25000 and 30000 (Pietsch et al., 2010). An other prob lem may be as so ci ated with a sig nal of in - suf fi cient en ergy, reach ing the pre-zechstein strata. Ow ing to the high pos i tive re flec tion co ef fi cient (ap prox i mately 0.3) of the shal low ero sional bound ary un der ly ing the Ter - tiary succession, the critical angle associated with this re - flec tor is low (ap prox i mately 30 ). This means that only a small part of the wave front can be trans mit ted deeper. This is ap par ent in the com mon-shot gather data, com puted for points PS 17 000 (Figs 11B, 12B) and PS 32000 (Fig. 10C). The fig ures pres ent a very big dif fer ence be tween the range of max i mum off set of waves, re flected at an gles lower than a crit i cal an gle from the Ter tiary/ju ras sic (light blue traveltime curve) bound ary, and the range of max i mum off set of waves, re flected from the Upper/Middle Jurassic reflector (dark blue travel-time curve). SUMMARY Zero-off set mod el ling and off set seis mic mod el ling, which tested dif fer ent ge om e tries of the subsurface, were car ried out, us ing seis mic ray the ory. Such mod el ling can be very help ful in as sess ing the sur vey meth od ol ogy for the pro posed re gional-scale seis mic test line, Obrzycko-1 Za-

76 P. MARZEC ET AL. Fig. 14. Seis mic ray tra jec to ries, Szamotu³y Salt Diapir model (V), PS 15000. A shot re cord sim u la tion B dif frac tion point (CDP gather sim u la tion) bartowo-1 Zabartowo-2 (Górecki, 2010), de sign ing the ap - proach for data pro cess ing and de vel op ing cri te ria for the seismic interpretation of recorded wave field. The con clu sions with re gards to the sur vey meth od ol - ogy are mainly based on the anal y sis of seis mic ray paths, cal cu lated for the seis mic-geo log i cal mod els, which show the sub se quent stages in the de vel op ment of the salt struc - ture. The dis tri bu tion of ray paths and the seis mic il lu mi na - tion of re flec tors show that it should be pos si ble to re cord a seis mic sig nal, re flected from the Rotliegend strata, both when salt pil lows over lie them and when sim ple-shaped salt diapirs are present. Significant limitations are expected,

APPLICATION OF 2-D FORWARD SEISMIC MODELLING 77 Fig. 15. Seis mic ray tra jec to ries, Szamotu³y Salt Diapir model (V), PS 20000. A shot re cord sim u la tion; B dif frac tion point (CDP gather sim u la tion) where salt diapirs with large over hangs or fault zones are pres ent above the strata. In such a sit u a tion, due to the sig nal being reflected from the additional, seismic interfaces, which are usu ally as so ci ated with strong con trasts and are steeply in clined, the ray paths of the re flected waves reach the ground sur face far be yond the sur vey stretch, while ray paths, associated with the refraction waves, are very easily refracted. The re sults of the anal y sis of seis mic ray paths, seis mic re cords and syn thetic wave fields also in di cate that the pro - cess ing of re corded, seis mic data may be prob lem atic. The non-uni form dis tri bu tion of ray paths, which re sults in the

78 P. MARZEC ET AL. Fig. 16. Seis mic ray tra jec to ries, Salt Pil low RogoŸno, PS 32000. A shot re cord sim u la tion; B dif frac tion point (CDP gather sim u la - tion) non-hy per bolic shape of travel-time curves, in the curve dis con ti nu ity, ap par ent on the in di vid ual re cords, as well as in a com plex wave field (such as a large num ber of dif - fracted waves), ob served in tec tonic zones, in di cates that it will be nec es sary to ap ply the sur vey meth od ol ogy and pro - cess ing al go rithms, de vel oped for the seis mic ex plo ra tion of com plex salt struc tures (Hale, 1991; Hill et al., 1991; Cavalca and Lailly, 2005; Farmer and Jones, 2006; Buur and Kuhnel, 2008). The syn thetic wave fields, cal cu lated for the sub se - quent, seis mic-geo log i cal mod els, rang ing from the strat i - fied model to a salt diapir, re sem bling the Szamotu³y Salt Diapir, in di cate the fol low ing conclusions: the pres ence of small salt pil lows does not sig nif i -

APPLICATION OF 2-D FORWARD SEISMIC MODELLING 79 cantly af fect the im age of the un der ly ing strata. The dis tri - bu tion of seis mic re flec tion events on time pro files re sem - bles the dis tri bu tion of seis mic re flec tors in the depth do - main; when a large salt pil low is pres ent, cor rect re cord ing of re flected waves should not be prob lem atic. How ever, in such a case, a re verse ar range ment of seis mic events in time sec tions and seis mic re flec tors in depth sec tions should be ex pected. Flat-ly ing or synclinal geologic bound aries may ap pear as anticlinal seis mic events in time pro files. The pull-up ef fect will in crease with in creas ing thick ness of salt strata and with in creas ing seis mic velo- city, associated with them; the pres ence of salt diapirs sig nif i cantly com pli cates the seis mic im age of sub-salt struc tures. This is due to the shape of the salt diapir and the wave ve loc ity con trast be - tween a salt body and the sur round ing strata. The pres ence of an ir reg u larly shaped salt diapir and as so ci ated faults re - sults in large num bers of ad di tional seis mic in ter faces. In such a sit u a tion, the seis mic sig nal is re flected in all di rec - tions, which significantly complicates accurate, seismic data pro cess ing and de creases the qual ity of the sig nal, re - flected from the Rotliegend strata. The ve loc ity con trast is par tic u larly ap par ent, when a salt diapir pen e trates the Muschelkalk strata, resulting in a clear increase in the pull-up effect; the mod el ling ex er cise per formed dem on strates the ef fec tive ness of multi-op tional mod el ling in the geo log i cal interpretation of complex salt structures. The mod el ling, us ing a method, based on seis mic ray the ory gives sim pli fied ray paths and syn thetic wave fields. Nev er the less, the mod el ling, pre sented here, shows the dif - fi cul ties which may be en coun tered dur ing seis mic ex plo ra - tion of the pre-zechstein suc ces sions in ar eas of salt tec ton - ics. The mod el ling in di cates that both high-ve loc ity con - trasts in zones, where salt diapirs reach the Cre ta ceous and Ter tiary strata, and steep in cli na tion of seis mic re flec tors, as so ci ated with large-dis place ment faults and the walls of diapirs, af fect the seismic image of the pre-zechstein successions. The main con clu sion of the pres ent study is that the ar - chi tec ture of the eolian Rotliegend strata can not be ac cu - rately re cog nised with out a good un der stand ing of the struc - ture of the Zechstein suc ces sion. In or der to gain a re li able pic ture of the Zechstein suc ces sion, it is nec es sary to adopt ad vanced meth ods of seis mic data pro cess ing, suited to the im ag ing of steeply in clined, subsurface fea tures, fre quently cut by faults. Equally im por tant is the con struc tion of a re li - able ve loc ity model, since the model de ter mines the ef fec - tive ness of the pro cess ing pro ce dures (such as mi gra tion) and is also nec es sary for de vel op ing us ing seis mic mod el - ling the cri te ria for the seismic interpretation of recorded wave field. Acknowledgements The pa per was com pleted as part of re search grant No. MNiSW WND-POIG.01.01.02.00.122/09 (Im prove ment of the ef - fective ness of seis mic sur vey for pros pect ing and ex plo ration for natu ral gas de pos its in Rotliegend for ma tions), su pervised by Wojciech Górecki (AGH Uni ver sity of Sci ence and Tech nol ogy, Faculty of Geology Geophysics and Environment Protection, Kraków, Po land,). We are greatly in debted to re view ers, Marek Grad (Uni ver sity of War saw), Yuriy Maystrenko (NGU Trondheim) and an anon y mous re viewer for their com ments, which helped to im prove our work. REFERENCES Bujak, A., 2010. Model sedymentologiczno-facjalny utworów dolnego permu w rejonie Rokietnica-Golêczewo-Objezierze- Obrzycko. In: Górecki, W. (ed.), Im prove ment of the Ef fec - tive ness of Seis mic Sur vey for Prospection and Ex plo ra tion for Nat u ral Gas De pos its in Rotliegendes For ma tions. Re - search Pro gram: MNiSW WND-POIG.01.01.02.00.122/09. Ar chive Katedra Surowców Energetycznych WGGiOŒ AGH, Kraków. [Un pub lished re port; in Pol ish]. Buur, J. & Kuhnel, T., 2008. Salt in ter pre ta tion en abled by re - verse-time mi gra tion. Geophysics, 73: 211 216. Cavalca, M. & Lailly, P., 2005. Pris matic re flec tions for the de lin - ea tion of salt bod ies. In: 75 th Annual International Meeting, SEG, Expanded Abstracts. Society of Exploration Geophysicists, pp. 2550 2553. Dadlez, R., 2000. Pom er a nian Caledonides (NW) Po land, fifty years of con tro ver sies: a re view and new con cept. Geological Quarterly, 55: 221 236. Dadlez, R., JóŸwiak, W. & M³ynarski, S., 1997. Sub si dence and in ver sion in the west ern part of Pol ish Ba sin data from seis - mic ve loc i ties. Geological Quarterly, 41: 197 208. Dadlez, R., Marek, S. & Pokorski, J. (eds), 1998. Palaeogeo - graphi cal At las of the Epicontinental Perm ian and Me so zoic in Po land (1:2500000). Pol ish Geo log i cal In sti tute, Warszawa. Dadlez, R. & Mark, S., 1998. Ma jor faults, salt-and non-salt anticlines. In: Dadlez, R., Marek, S. & Pokorski, J. (eds.), Palaeogeographical At las of the Epicontinental Perm ian and Me - so zoic in Po land (1: 2 500 000). Pol ish Geo log i cal In sti tute, Warszawa. Farmer, P., 2006. Back to the fu ture: New ad vances in re verse time migration provide sub-salt imaging solutions. Hart s E&P, 79: 63 65. Farmer, P., Miller, D., Pieprzak, A., Rutledge, J. & Woods, R., 1996. Ex plor ing the subsalt. Oil field Re view, 8: 50 64. Farmer, P. & Jones I. F., 2006. Ap pli ca tion of re verse time mi gra - tion to com plex im ag ing prob lems. First Break, 24: 65 73. Górecki, W. (ed.), 2010. Im prove ment of the Ef fec tive ness of Seis - mic Sur vey for Prospection and Ex plo ra tion for Nat u ral Gas Deposits in Rotliegendes Formations. Re search Pro gram: MNiSW WND-POIG.01.01.02.00.122/09. Ar chive Katedra Surowców Energetycznych WGGiOŒ AGH, Kraków. [Un - pub lished re port; in Pol ish]. Hale, I. D., 1991. Migration of Seismic Turning Waves. European Patent Application, no. EP19910304335, publication no. EP0513448 A1. The Hague, 20 pp. Hill, N. R., Wat son, T. H., Hassler, M. H. & Sisemore, L. K., 1991. Salt-flank im ag ing us ing Gaussi an beam mi gra tion. 61 st Annual International Meeting, SEG, Expanded Abstract. Society of Exploration Geophysicists, Houston, pp. 1178 1180. Jones, I. F., 2008. A mod el ling study of pre-pro cess ing con sid er - ations for re verse-time mi gra tion. Geophysics, 73: 99 106. Kobylarski, M., Pietsch, K. & Kowalczuk, J., 2008. PP and PS mod el ling as a tool for compressional and shear ve loc ity

80 P. MARZEC ET AL. model con struc tion. Kwartalnik AGH, Geologia, 34: 285 300. [In Pol ish, Eng lish sum mary]. Krzywiec, P., 2000. On mech a nisms of the Mid-Pol ish Trough in - ver sion re sults of seis mic data in ter pre ta tion. Biuletyn Pañstwowego Instytutu Geologicznego, 393: 135 166. [In Pol ish, Eng lish sum mary]. Krzywiec, P., 2006a. Struc tural in ver sion of the Pom er a nian and Kuiavian seg ments of the Mid-Pol ish Trough lat eral vari a - tions in tim ing and struc tural style. Geo log i cal Quarterly, 50: 151 168. Krzywiec, P., 2006b. Tri as sic Ju ras sic evo lu tion of the NW (Pom er a nian) seg ment of the Mid-Pol ish Trough base ment tec ton ics vs. sed i men tary pat terns. Geo log i cal Quarterly, 51: 139 150. Krzywiec, P., 2009. Ge om e try and evo lu tion of se lected salt struc - tures in the Pol ish Low lands in the light of seis mic data. Przegl¹d Geologiczny, 57: 812 818. [In Pol ish, Eng lish ab - stract]. Krzywiec, P., Wybraniec, S. & Petecki, Z., 2006. Budowa tektoniczna pod³o a bruzdy œródpolskiej w oparciu o wyniki analizy danych sejsmiki refleksyjnej oraz grawimetrii i magnetometrii. In: Krzywiec, P. & Jarosiñski, M. (eds.), Struktura litosfery w centralnej i pó³nocnej Polsce obszar objêty PO- LONAISE'97. Prace Pañstwowego Instytutu Geolgicznego, 188: 107 130. [In Pol ish]. Kwolek, K., 2000. The age of tec tonic move ments in the Poznañ Kalisz dis lo ca tion zone, Fore-Sudetic Monocline. Przegl¹d Geologiczny, 48: 804 814. [In Pol ish, Eng lish ab stract]. Leszczyñski, K., 2000. The Late Up per Cre ta ceous sed i men ta tion and sub si dence south-west of the K³odawa Salt Diapir, cen - tral Po land. Geo log i cal Quar terly, 44: 167 174. Pha raoh, T. C., Dusar, M., Geluk, M. C., Kockel, F., Krawczyk, C. M., Krzywiec, P., Scheck-Wenderoth, M., Thybo, H., Vejb k, O. V. & van Wees, J. D., 2010. Tec tonic evo lu tion. In: Doornenbal, J. C. & Stevenson, A. G. (eds.), Pe tro leum Geo - log i cal At las of the South ern Perm ian Ba sin Area. EAGE Pub li ca tions B.V., Houten, pp. 25 57. Pietsch, K., Marzec, P., Kobylarski, M., Danek, T., Leœniak, A., Tatarata, A., & Gruszczyk, E., 2007a. Iden ti fi ca tion of seis - mic anom a lies caused by gas sat u ra tion on the ba sis of the o - ret i cal P and PS wavefield in the Carpathian Foredeep, SE Po land. Acta Geophysica, 55: 191 208. Pietsch, K., Kobylarski, M. & Ur ban, A., 2007b. Seis mic mod el - ling a sup port tool for struc tural in ter pre ta tion of seis mic data from the area of the Outer Carpathians be tween Babia Góra and Wadowice. Kwartalnik AGH, Geologia, 33: 183 196. [In Pol ish, Eng lish sum mary]. Pietsch, K., Nawieœniak, A., Kobylarski, M. & Tatarata, A., 2008. Can seis mic wave at ten u a tion be a source of in for ma tion about gas sat u ra tion de gree of res er voir lay ers? A mod el ling case study. Przegl¹d Geologiczny, 56: 545 552. [In Pol ish, English summary]. Pietsch, K., Marzec, P. & Niepsuj, M., 2010. Sejsmiczne modelowania strukturalne. In: Górecki W. (ed.), Improvement of the Ef fec tive ness of Seis mic Sur vey for Prospection and Ex plo ra - tion for Nat u ral Gas De pos its in Rotliegendes For ma tions. Re search Pro gram: MNiSW WND-POIG.01.01.02.00.122/ 09. Ar chive Katedra Surowców Energetycznych WGGiOŒ AGH, Kraków. [Un pub lished re port; in Pol ish]. Pietsch, K., Marzec, P., Niepsuj, M. & Krzywiec, P., 2012. The in - fluence of seismic velocity distribution on the depth imaging of the sub-zechstein ho ri zons in ar eas af fected by salt tec ton - ics: a case study of NW Po land. Annales Societatis Geologorum Poloniae, 82: 263 277. S³onka,., 2013. Anal y sis of post-stack FD mi gra tion on 2D syn - thetic sec tion cal cu lated for geo log i cal model of salt tec ton - ics. 75th EAGE Conference & Exhibition incorporating SPE EUROPEC 2013, Ex tended Ab stract. European Association of Geoscientists & En gi neers, Lon don, doi: 10.3997/2214-4609.20131105. http://www.earthdoc.org/publication/ publicationdetails/?publication=68824 [10.06.2013]. Van Wees, J. D., Stephenson, R. A., Ziegler, P. A., Bauer, U., McCann, T., Dadlez, R., Gaupp R., Narkiewicz, M., Bit zer, F. & Scheck, M. 1997. On the or i gin of the South ern Perm ian Ba sin of Cen tral Eu rope. In: EUPOPROBE TESZ Meet ing, Terra Nos tra, 97(11): 153 157. Wilk, M., 2010. Opracowanie dwuwymiarowych modeli prêdkoœci w rejonie otworu Golce-1, In: Górecki, W. (ed.), Im - prove ment of the Ef fec tive ness of Seis mic Sur vey for Prospection and Ex plo ra tion for Nat u ral Gas De pos its in Rotliegendes Formations. Re search Pro gram: MNiSW WND-POIG. 01.01.02.00.122/09. Ar chive Katedra Surowców Energetycznych WGGiOŒ AGH, Krakówe. [Un pub lished re port; in Pol ish]. Wag ner, R., 1994. Stra tig ra phy and evo lu tion of the Zechstein Ba - sin in Pol ish Low land. Prace Pañstwowego Instytutu Geologicznego, 146: 1 71. Wag ner, R., 2012. Podsumowanie badawcze i opracowanie koñcowe budowy strukturalno-tektonicznej wzd³u profilu Obrzycko-Zabartowo, In: Górecki, W. (ed.), Improvement of the Ef fec tive ness of Seis mic Sur vey for Prospection and Ex plo ra - tion for Nat u ral Gas De pos its in Rotliegendes For ma tions. Re search Pro gram: MNiSW WND-POIG.01.01.02.00.122/ 09. Ar chive Katedra Surowców Energetycznych WGGiOŒ AGH, Kraków. [Un pub lished re port; in Pol ish]. Ziegler, P. A., 1990. Geological Atlas of Western and Central Eu - rope, 2nd edi tion. Shell Internationale Pe tro leum Maatschappij B.V. and Geo log i cal So ci ety Pub lish ing House, Bath, 239 pp.