DETERMINING GEOMETRY OF SALT DIAPIRS BY DEPTH MODELING OF SEISMIC VELOCITY PULL-UP

Hemin Koyi
Uppsala University
Uppsala,Sweden

Ole Johan Aasen, Stein Nybakken
Saga Petroleum AS
Harstad, Norway

Kai Hogstad, Bjorn Torudbakken
Saga Petroleum AS
Sandvika, Norway

Introducing knowledge about the overall relief of subsalt horizons, depth modeling of seismic data is here used to constrain the geometry of salt diapirs.

The subsalt horizons will appear as velocity pull-up reflectors in the seismic sections due to the high velocity of salt. Iterative depth modeling based on extension of top salt, the velocity pull-up, and interpreted subsalt geometry give the main geometry of the salt diapir.

The method has been applied on salt diapirs in the Nordkapp basin on the Barents shelf. The results suggest that salt diapirs in this basin have broad overhangs fed by narrower stems. In the absence of wells, the present method offers a simple and time efficient tool in hydrocarbon exploration.

INTRODUCTION

Hydrocarbon traps associated with salt diapirs are known from many sedimentary basins (e.g. Germany, Gulf of Mexico, North Sea).

Salt structures can form and modify potential hydrocarbon traps at a wide variety of depths depending on the geometry and maturity of the individual salt structures. 1 2 3

Defining the salt boundaries and the surrounding traps is generally difficult due to non-optimal detection, preservation, and imaging on conventional acquired and processed seismic data.

The upper boundary of salt diapirs is often well defined on seismic profiles because of the high acoustic impedance contrasts between salt and other sediments. The lower boundary of salt tongues and sills has been detected by reflection seismic data.4 5

These boundaries, however, are gently dipping horizons. Steep contacts of salt diapirs and the shape of diapiric bulbs, on the other hand, are difficult to detect and to interpret from seismic lines.

The boundary between a salt layer and its substrate is often a well defined reflector due to the presence of carbonates and/or anhydrite beds that are frequently present at the initial stages of an evaporate sequences This reflector and any underlying reflectors often show pull-up effect because the seismic velocity of the salt is greater than the velocity of the neighboring younger sediments.

The geometry of velocity pull-up beneath a salt structure is related to the geometry of the structure and the amount of salt the seismic wave travels through. Normally, due to thicker volume of salt in the stem of a diapir, presalt layers beneath the stem suffer more pull-up than those layers that underlay the diapir overhang.

Pull-up can therefore be used to define or at least constrain the shape of salt structures by an iterative seismic modeling procedure. The authors present here a simple depth modeling method for velocity pull-up restoration and salt geometry determination.

MODELING METHOD

The pull-up geometry of reflectors located beneath salt structures is consistent with the geometry of the structures themselves (Fig. 1).

Therefore, the geometry of the pulled up reflector could be used to predict the overall geometry of the salt structure that caused the pull up.

Velocity pull-up restoration is performed by depth conversion of the interpreted reflection seismic profiles. This is normally done by assigning each layer on a seismic profile its interval velocity and calculating the depth of the horizons.

The suggested depth modeling method considers restoration of the velocity pull-up of the base of the salt reflector and other underlying reflectors. A restoration is considered successful when an interpreted pull-up event in the time section is transformed to an assumed reasonable geological position in the depth section.

The advantage of this method is that the stratigraphic sequence underlying salt is generally assumed to be unaffected by salt movement. The depth modeling can be performed by most commercial depth conversion software and once velocity data are loaded, alternative seismic interpretations can be modeled and tested.

It is important to consider the limitations of the method. Due to steeply dipping events and truncations normally associated with salt structures, depth modeling ideally should consider the actual seismic wave paths in addition to velocity data,

Proper depth migration of seismic lines crossing salt structures therefore demands knowledge of structural dips and deformation structures in the adjacent stratigraphic sequence in addition to salt geometry. For complex salt structures 2-D data may create wrong reflector positions and dips that may lead to misinterpretations.

The simple depth modeling method is sensitive to the applied velocity. In principle this effect may be more important close to the salt diapir where local uplift, subsidence, and possible facies changes may occur. It may be necessary to adjust velocity input according to these geological factors. The simple method assumes no velocity difference between equivalent layers on either side of a salt diapir.

Finally, uncertainty is introduced by the seismic interpretation of the base salt and deeper reflectors, influenced by velocity pull-up. Inclined diapirs and presence of several levels of salt overhangs may give unclear velocity pull-up geometries.

If salt diapirs are situated above basin floor faults, the faults will complicate the interpretation of time sections and will also lead to uncertainty of what could be regarded as a reasonable depth restoration.

Diapir overhangs that have elliptical or irregular map view geometry are likely to show different geometries in profile as well. For example, a diapir overhang would be wider on a seismic profile shot parallel to the long axis of its elliptical overhang than perpendicular to it.

NORDKAPP RESTORATION

The simple depth modeling method was applied on salt diapirs in the Nordkapp basin on the Barents shelf.

This basin started its evolution in the late Devonian to early Carboniferous involving rifting and clastic sedimentation. The major precipitation of salt probably occurred in the late Carboniferous. Salt movement took place in the Triassic, Cretaceous, and Tertiary.6

Jensen and Sorensen7 have suggested that the stems of the salt diapirs in the Nordkapp basin are columnar with no overhangs. These authors also neglect the possibilities that these diapirs spread before they were reactivated in Cretaceous times.

More recent work has, on the other hand, shown that the diapirs in the Nordkapp basin may have had sufficient time from Upper Triassic through Jurassic to spread and form overhangs, whether or not their stems are narrow.8 9

Salt structures with broad overhangs are reported from other areas (e.g. San Felipe Barbers Hill, and Bethel domes, Texas,1 Hainesville diapir,10 Wienhausen and Hanigsen salt diapirs in Germany,11 and Bethel diapir.12 In the absence of wells through the diapirs in the Nordkapp basin, the velocity pull-up effect was used to constrain the geometry of the diapirs. The actual base of the salt has not by certainty been identified on seismic data in the Nordkapp basin.

Below the stems of some of the diapirs, however, deep seismic reflectors have been recognized (Fig. 2). It is assumed that in depth, the general dip trend of the deep reflectors observed in the area adjacent to the diapir also continue below the salt diapir.

This picture is, however, complicated by a fault that offsets the deep reflectors on the right side of the diapir (Fig. 2).

In order to prepare the line for depth modeling, stratigraphic correlation was made to wells at the margin of the Nordkapp basin through a net of seismic lines. The resulting interpretation is shown in Fig. 2. Selection of velocity data for the stratigraphic units was based on the available well data in the region.

The seismic subdivision of the stratigraphy, the velocity pull-up interpretation, and three alternative interpretations of the salt geometry were digitalized (Figs. 3a, 4a, and 5a). The time sections were then depth converted, and for each model details of the salt geometry were iteratively changed to improve the depth model with respect to the assumed positions of the deep reflectors (Figs. 3b, 4b, and 5b).

RESULTS

Depth conversion of the diapir with the columnar geometry (Fig. 3a) causes the underlying layers to become unrealistically troughs beneath the column-like diapirs (Fig. 3b).

The cone shaped diapir (Fig. 4a) also causes unrealistic relief in the underlying reflectors (Fig. 4b). On the other hand, the broad diapir overhang fed by a narrower stem (Fig. 5a) is giving the most satisfactory result on the depth converted profile (Fig. 5b).

The reflectors beneath the salt diapir possess their sub-horizontal geometry in consistence with the overall geometry of the other reflectors. This is also supported by the evolution history of the Nordkapp basin, where thin sequences were deposited during the Upper Triassic to Lower Cretaceous.

During this period the salt diapirs extruded at or near the surface and formed broad overhangs before they were reactivated due to further burial by Cretaceous and Tertiary sediments.

EVALUATION, CONCLUSIONS

The application of the suggested depth modeling method on salt structures in the Nordkapp basin showed that the method is easy and time efficient in practical use.

The authors also suggest that the method is a powerful tool when integrated with seismic and geological interpretation. The method may be used to distinguish or even exclude main types or classes of diapirs that cannot be consistent with the seismic data.

If more detailed resolution of the salt related geometries is required, more sophisticated methods in the data acquisition and migration routines should be considered.

With respect to the Nordkapp basin the method has indicated that salt diapirs with considerable overhangs exist. This result is interesting for oil and gas exploration because these types of salt structures may form favorable hydrocarbon traps.

ACKNOWLEDGMENTS

The authors would like to thank Christopher Talbot and David Worsely for reading and commenting on this manuscript and Saga Petroleum AS for permission to publish the work. Hanne Skulstad drafted the figures. The Norwegian Petroleum Directorate gave permission to publish the seismic example.

REFERENCES

Halbouty, M.T., Salt domes - Gulf region, U.S. and Mexico, second edition, Gulf Publishing Co., Houston, 1979, 561 p.
Woodberg, H.O., Murray, I.B., and Osborne, R.E., Diapirs and their relations to hydrocarbon accumulations, in Facts and principles of world petroleum occurrence, CSPG Memoir 6 (ed. Miall), 1980, pp. 119-142.
Jenyon, M. K., Salt tectonics, Elsevier Applied Science Ltd., 1986, 191 p.
Simmons, G., Study suggests salt traps in deep gulf, AAPG Explorer, Vol. 12, 1991, p. 4.
Nelson, T.H., and Fairchild, L.H., Emplacement and evolution of salt sills in northern Gulf of Mexico, AAPG Bull., Vol. 73, p. 393.
Gabrielsen, R.H., Faerseth, R.B., Jensen, L.N., Kalheim, J.E., and Riis, F., Structural elements of the Norwegian continental shelf, Part 1: The Barents Sea Region, NPD Bull. 6, 1990, 33 p.
Jensen, L.N., and Sorensen, K., The tectonic framework and halokinesis of the Nordkapp basin, Barents Sea, in Arctic Geology and Petroleum Potential, Conference Volume (Vorren, T., ed.), in press.
Talbot, C.J., Koyi, H., and Clark, J., Multiple halokinesis in the Nordkapp basin, in Arctic Geology and Petroleum Potential, Conference Volume (Vorren, T., ed.), in press.
Koyi, H., Talbot, C.J., and Torudbakken, Modelling diapiric structures in the Nordkapp basin, Norwegian Barents Sea, in International meeting on Arctic Margins, abstracts and program.
Loocke, J.E., Growth history of the Hainesville salt dome, Wood County, Tex., master's thesis, the University of Texas at Austin, 1978, 95 p.
Richter-Burnberg, G., Salt tectonic, interior structures of salt bodies, Bull. Cent. Rech. Explor.-Prod. Elf Aquitaine, 1980, pp. 373-393.
Wood, D.H., and Giles, A.B., Hydrocarbon accumulation patterns in the East Texas salt dome province, the University of Texas at Austin, Bureau of Economic Geology, Geological Circular 8256, 1982, 36 p.