Catuneanu Et Al 1997

Interplay of static loads and subduction dynamics in foreland basins: Reciprocal stratigraphies and the “missing” peripheral bulge Octavian Catuneanu Department of Geology, Rhodes University, Grahamstown, South Africa

Christopher Beaumont Paula Waschbusch Department of Oceanography, Dalhousie University, Halifax, Nova Scotia B3H 4J1, Canada

ABSTRACT Foreland basins are created by superimposed mechanisms that flex the lithosphere. In addition to static loads, dynamic loading below the basin by viscous mantle corner flow above a subducting plate may cause long-wavelength subsidence. It is proposed that the interaction of the static and dynamic forces is responsible for the formation and preservation of recently recognized “reciprocal stratigraphies” in retro-foreland regions above subducting slabs. Reciprocal stratigraphies refer to a correlative succession of strata characterized by contrasting stacking patterns that reflect opposite base-level changes between proximal and distal settings. The same interactions may also modify the stratigraphy of the flexural peripheral bulge and mask its presence. An example from the Late Cretaceous of the Western Interior basin, Canada, illustrates the concepts.

INTRODUCTION It is generally agreed (see DeCelles and Giles, 1996, for a review) that foreland basin systems form by the flexural warping of the lithosphere under a combination of supralithospheric and sublithospheric loads. Flexure of the lithosphere under concentrated static loads generates a downwarp proximal to the orogen, the foreland basin, and a low amplitude long-wavelength upwarp, the peripheral bulge or forebulge (Fig. 1). In addition to the static loads, there is increasing evidence that dynamic loading by viscous mantle corner flow coupled to the subducting slab may superimpose long-wavelength subsidence, particularly when there is rapid subduction of a shallowly dipping slab beneath the foreland basin (Mitrovica et al., 1989; Gurnis, 1992; Holt and Stern, 1994). When combined, these loading mechanisms lead to a conceptual model (Fig. 1) in which retro-arc or retro-foreland basins are influenced by both static and dynamic loads. How can the signatures of dynamic and static forcing be recognized? Normally, when orogenic or slab pull static loads increase, the peripheral bulge is raised and eroded. The superposition of a sufficiently longwavelength dynamic subsidence can, however, offset the uplift of the bulge and lead to sedimentation on the peripheral bulge as well as within the foreland basin. Under these circumstances the erosional characteristics of the bulge are suppressed and “missing” from the stratigraphic record, and the bulge may be unrecognizable. We suggest that this same superposition creates a new stratigraphic package, which has recently been recognized and termed reciprocal stratigraphies (Catuneanu et al., 1997). In the present context, reciprocal stratigraphies are characterized by correlative proximal transgressive and distal regressive strata and the converse situation. The accommodation for the reciprocal strata is created by the combined static and dynamic subsidence. The reciprocal properties of the strata reflect the differential subsidence and uplift of the foreland basin and peripheral bulge within the overall long-wavelength dynamic subsidence. The latter provides the accommodation necessary to create and preserve the reciprocal strata. If correct, our interpretation indicates that reciprocal architecture provides a diagnostic tool by which static and dynamic loading may be recognized and separated. We interpret the absence of a recognized peripheral bulge in the Canadian part of the Late Cretaceous–early Cenozoic Western Interior (foreland) basin to be a consequence of superimposed moderate-wavelength static flexure and long-wavelength dynamic subsidence. This inferGeology; December 1997; v. 25; no. 12; p. 1087–1090; 4 figures.

ence is consistent with the reciprocal architecture of the Bearpaw and Cannonball cycles from this region. RECIPROCAL STRATIGRAPHY AND HINGE LINE POSITIONS FROM THE WESTERN INTERIOR BASIN Examples of reciprocal correlation between proximal and distal sequences in a retro-foreland basin are offered by the Bearpaw (late Campanian–Maastrichtian, ca. 77–65 Ma) and Cannonball (Paleocene, ca. 65–55 Ma) cyclothems in the Western Canada basin (the Canadian portion of the Western Interior basin). These cyclothems include coeval marine and nonmarine facies separated by diachronous boundaries reflecting the timetransgressive character of both overall incursions and retreats of the interior seaways.

Figure 1. Static and dynamic loads that act on an orogen-foreland basin system. Pro- and retro- refer to the subducting and overriding lithospheres. Static loads act at the surface (the orogen, foreland fold and thrust belt, sediment and water) and beneath the system (e.g., slab pull). The subducting pro-lithosphere induces a dominant viscous corner flow (dashed lines) beneath the retro-lithosphere, which dynamically reduces the pressure and leads to subsidence (vertical arrows) of the retro-lithosphere that is greatest adjacent to the subduction zone. This dynamical loading depends on the viscous drag force and is proportional to the viscosity of mantle involved in the corner flow and the speed of subduction.The horizontal scale of the subsidence (λD in text) is maximized when the slab dip is small. 1087

The marine transgressive-regressive cycles of the Bearpaw strata have been studied through facies analysis at the regional scale of the Western Canada basin (Catuneanu et al., 1997). The analysis included outcrop and core sedimentology of reference sections, gamma ray and sonic log correlations along subsurface profiles (Fig. 2; Catuneanu, 1996), biostratigraphic (particularly ammonite) correlations (using the Caldwell et al., 1993 synthesis), and the use of bentonite beds for correlation and chronostratigraphy. As suggested by the lateral shift between open shelf and shoreface facies, the fining-upward successions are identified as transgressive systems tracts, whereas the coarsening-upward successions are identified as regressive systems tracts. Two distinct types of transgressiveregressive sequences have been recognized within the basin: proximal-type sequences (closer to the orogenic belt) dominated by regressive systems tracts having a transgressive to regressive systems tract ratio of 1/3, and distal-type sequences dominated by transgressive systems tracts having a transgressive to regressive systems tract ratio of 3/1. A reciprocal correlation between the proximal and distal stratigraphies has been documented (Catuneanu et al., 1997): Proximal transgressive systems tracts and distal regressive systems tracts (as well as proximal regressive systems tracts and distal transgressive systems tracts) can be mapped as continuous, genetically related and correlative packages of strata with comparable thicknesses (Fig. 3). These results, anticipated in theoretical com-

Figure 2. Hinge line positions in the Western Interior basin during the late Campanian and earliest Maastrichtian. 1088

puter models (Jordan and Flemings, 1991), are supported by extensive bentonite seams that demonstrate that the correlative transgressive and regressive strata are of the same age (Fig. 3). The proximal to distal facies change takes place over an area a few kilometers wide that can be defined as a hinge zone. The mid-point of the hinge zone may be taken as the stratigraphic hinge line (distance λH from the orogen, Fig. 4). Similar reciprocal stratigraphies between the proximal and distal sectors of the Western Canada basin have been established for the Maastrichtian-Paleocene nonmarine sequences, partially correlative with or overlying the Bearpaw marine facies (O. Catuneanu and A. R. Sweet, personal commun.). In addition, the interpretation of the coeval transgressive and regressive shorelines indicated by Gill and Cobban (1973), in terms of reciprocal proximal to distal stratigraphies and flexural tectonics, allowed the mapping of the hinge line position on the United States side of the Western Interior basin for the entire Campanian-Maastrichtian interval (Catuneanu, 1996). As an average position for the Campanian-Paleocene interval, the hinge line traces a semielliptical pattern outlining the region of maximum foreland basin subsidence (Fig. 2; Catuneanu, 1996). CONCEPTUAL INTERPRETATION OF RECIPROCAL STRATIGRAPHIES We interpret the synchronous reciprocal transgressive and regressive systems tracts, described previously, to imply simultaneous deepening and shallowing of the marine basin on either side of the hinge line. This behavior is attributed to a combination of static flexure of the lithosphere, which changes sign between the proximal foredeep and the distal peripheral bulge (Fig.1; Beaumont et al., 1988; Jordan and Flemings, 1991) and long-wavelength dynamic subsidence. Static loads (LS, Fig. 4A) tend to be concentrated near the orogen and result in flexure with an exponentially decreasing magnitude. At some distance, λS, static deflections become negligible. A static hinge line (at distance λHS) can be defined where negative static deflections (basin subsidence) of a horizontal reference surface change to positive static deflections of much lower amplitude (peripheral bulge uplift), and the converse occurs when static loads decrease. The distance λHS depends on the flexural properties of the lithosphere and the distribution of static loads. Dynamic loads (LD, Fig. 4A) are independent of static loads. They act on a length scale determined by mantle corner flow (Fig. 1) and cause tectonic subsidence of a horizontal reference surface over a distance λD. In many circumstances, particularly for small subduction dip, λD > λHS (Mitrovica et al., 1989). When static and dynamic loads combine to cause subsidence and λD > λHS, the tectonic deflection of the horizontal reference surface may look like the solid line (Fig. 4B). The adjacent dashed lines (Fig. 4B) show the conceptual change to the deflection when the static load is increased, LS(I), or decreased, LS(D), while LD remains constant. Correspondingly, the change in deflection when the dynamic load is increased, LD(I), or decreased, LD(D), while LS remains constant is shown by the dashed lines (Fig. 4C). The sedimentary response can be complex. However, the general effect is the creation of reciprocal increases and decreases in accommodation space on either side of λHS when LD is constant and LS varies. Corresponding reciprocal transgressive and regressive systems tracts will be created if sedimentation rates are smaller than the rate of change of accommodation space (Fig. 4B situation). This mechanism provides a simple explanation for the creation of reciprocal stratigraphies. An increase in LD provides an equally simple explanation for their preservation. The role of the long-wavelength dynamic load is twofold: to create accommodation space above the static peripheral bulge so that deposition, not erosion, will occur, and to cause additional subsidence necessary to preserve reciprocal sequences against later static uplift of the peripheral bulge. While LD remains constant, as assumed in Figure 4B, λH ≈ λHS and the hinge line position measures the normal static width of the foreland basin. The distal GEOLOGY, December 1997

Figure 3. Diagrammatic illustration of reciprocal correlation between transgressive and regressive systems tracts across foreland basin hinge line. A, B, C: proximaltype sequences, 1, 2, 3: distal-type sequences. I, D: stages of inferred increasing (I) and decreasing (D) static load in orogenic belt: I: proximal subsidence correlated to distal uplift; D: proximal uplift correlated to distal subsidence. Distal ravinement surface is superimposed on subaerial unconformity (sequence boundary), assuming that nonmarine portion of the transgressive systems tract has been eroded during marine transgression. In relationship to foreland basin flexural profile, coarsest sediments are supplied to distal sector at end of orogenic unloading stages. Average thickness of third-order sequence is 25 m and corresponding time span is 0.75 m.y. Vertical scale represents both time and thickness.

regressive zone marks the peripheral bulge, which is apparently “missing” because it is not uplifted and eroded. The mechanism also provides a probable explanation for the creation and preservation of “back bulge” basins (DeCelles and Giles, 1996; Giles and Dickinson, 1995) or an “eastern platform” (Kauffman and Caldwell, 1993) (Fig. 4D) when dynamic subsidence extends beyond the static peripheral bulge. More specifically, Figure 4 (D and E) illustrates the simplest explanation of Bearpaw reciprocal stratigraphies, the basin response to the tectonics type shown in Figure 4B. Long-wavelength dynamic subsidence has created marine accommodation space everywhere except the most distal craton. Immediately adjacent to the orogen this space has been overfilled by sediment. As LS increases (Fig. 4D), proximal transgression (transgressive systems tracts) will be coeval with forebulge regression (regressive systems tracts) if the sedimentation rate is less than the rate of change of accommodation space. The converse occurs when LS decreases (Fig. 4E). Other stratigraphic responses can be envisaged. When LS and LD change simultaneously, the positions of the transgressive and regressive systems tracts will depend on the rates of these changes and the sedimentation rates. The distance λH is unlikely to coincide with λHS, and reciprocal stratigraphies are improbable when dynamic subsidence exceeds static uplift of the peripheral bulge. Equivalent reciprocal stratigraphies can be created in nonmarine environments (Catuneanu et al., 1995). Their style depends on the effect of dynamic and static loading on nonmarine (e.g., fluvial) accommodation space. It is notable that neither changes in the amplitude of LD nor eustatic sea-level changes cause reciprocal stratigraphies when acting alone. However, increases in one or both of these quantities are necessary in order to preserve reciprocal strata. For example, the downward dynamic deflection (Fig. 1) is amplified during intervals when subduction rate increases. A change in slab dip will also modulate the dynamical deflection. For example, the overall effect of a decrease in slab dip is to increase distal accommodation space at the expense of that proximal to the orogen. GEOLOGY, December 1997

Figure 4. Conceptual effects of static (LS) and dynamic (LD) tectonic loads on retro-foreland basins and some sedimentary consequences. A:Tectonic deflection of horizontal reference surface by loads acting independently for case λD > λHS. B: Combined deflection when LS and LD act together (solid line) and effect (dashed lines) of increases, LS(I), and decreases, LS(D), in static load while LD is constant. C: Sensitivity of same combined deflection (solid line, B) to increases, LD(I), and decreases, LD(D), in dynamic load while LS is constant (dashed lines). D and E: Topographicbathymetric profiles illustrating likely stratigraphic response of basin to tectonics illustrated in B. Dynamic deflection has created marine accommodation across most of basin. D: Increased static load increases foredeep accommodation, reduces peripheral bulge accommodation, and leaves back-bulge accommodation unchanged. If sedimentation rates are less than rate of change of accommodation, proximal transgression (transgressive systems tracts) is coeval with peripheral bulge regression (regressive systems tracts). E:The converse of D.

SPECULATIONS REGARDING THE WESTERN INTERIOR BASIN AND CONDITIONS FAVORING THE CREATION OF RECIPROCAL STRATIGRAPHIES On the largest scale, the eastward-dipping subducting plate beneath western North America has been interpreted to have had a time-varying subduction dip angle, largely based on sweeping magmatic patterns (Coney and Reynolds, 1977; Armstrong and Ward, 1991, 1993; Constenius, 1996) and varying subduction velocity (see Ward, 1995, for a review). If this interpretation is correct, the changes in dip angle should be recorded in the foreland basin stratigraphy as relative changes in the length scale, λD, of dynamic loading. The magmatic patterns suggest that moderate to steeply dipping 1089

subduction occurred during most of the Cretaceous period. Magmatism migrated eastward during latest Cretaceous to early Eocene time (ca. 75–54 Ma), extending to a maximum ~1000 km east of the trench. This migration is interpreted to result from a shallowing of the subduction dip. Earlymiddle Miocene magmatism rapidly migrated westward, interpreted as a return to a steep subduction dip angle. What are the predicted consequences of such net changes in slab dip angle (steep-shallow-steep) for the creation and preservation of reciprocal stratigraphies? On the basis of λD estimates from Mitrovica et al. (1989, Fig. 13), steep slabs (dip > 45°) have λD ≤ 500 km, so that λD ≈ λHS for the Western Interior basin when the position of the supracrustal static loads east of the trench is taken into account. Under these circumstances, the width of a basin created by dynamic loads is similar to that of a basin created by static loads and the potential to distinguish these mechanisms on the basis of length scales is poor, particularly when the effects of other factors (e.g., sealevel fluctuations, loading history, sediment supply, local heterogeneities) are considered. The stratigraphy would most likely be interpreted as that resulting from simple static orogenic loading, but the load would be incorrectly estimated. One diagnostic feature is that a normal peripheral bulge should be recognizable when λD ≤ λHS. Shallow slabs (having dips of 30° or 20°) have λD ~ 1000 km or approaching 2000 km, respectively (Mitrovica et al. 1989, Fig. 13). Therefore, λD > λHS for the Western Interior basin, and the potential to distinguish dynamic and static effects on the basis of reciprocal stratigraphies increases. In addition, the net change in accommodation space following a slab dip decrease is mainly one of increased wavelength and can produce a distal stratigraphic signature that is similar to a first-order sea-level rise. The converse occurs when subduction steepens. The implications with respect to the Western Interior basin are as follows. (1) Creation of reciprocal stratigraphies is favored when λD > λHS (inferred to be approximately the 75–54 Ma interval) and peripheral bulges may be “missing” (i.e., unrecognized) for this interval. (2) When λD ≤ λHS, reciprocal stratigraphies are not favored (i.e., before ca. 75 Ma), but peripheral bulges are. (3) Preservation of reciprocal stratigraphies is favored when λD > λHS and LD increases (i.e., times of accelerating subduction, inferred to be the 75–55 Ma interval; see Ward [1995] for a review). These predictions are in accord with the observations. Moreover, it is unlikely that eustatic sealevel rise is the primary preserver of reciprocal stratigraphies in the Western Interior basin during latest Cretaceous and early Cenozoic time because the sediment thickness variation across the basin requires a persistent longwavelength component in accommodation space (Mitrovica et al., 1989).

CONCLUSIONS Reciprocal stratigraphies have recently been identified in the Bearpaw and Cannonball (77–55 Ma) cycles of the Canadian part of the Western Interior basin. Reciprocal stratigraphies consist of correlative proximal transgressive and distal regressive facies, and vice versa, and the interface separating them, termed the hinge line, can be mapped in space and time. Retro-foreland basins, including the Western Interior basin, are subject to static and dynamic (mantle corner flow) loads, amplitudes LS and LD, which cause subsidence with respective wavelengths λHS and λD. The sedimentary response to these combined tectonic loads, when λD > λHS, is a likely cause of reciprocal stratigraphies. The most simple combinations are: LS varies while LD is constant (which creates reciprocal stratigraphies and disguises the peripheral bulge), and LD increases while LS is constant (which preserves these stratigraphies). Our tectonic explanation of the Bearpaw and Cannonball reciprocal stratigraphies is consistent with the inferred shallow dip (λD > λHS) and acceleration (LD increasing) of the subducted plate beneath western North America at that time, ca. 75–55 Ma. If the proposed relationship between reciprocal stratigraphies and the tectonic model is confirmed, mapping such strata has the potential to identify and separate static and dynamic effects in foreland basin systems and to place constraints on paleo-subducting slabs. 1090

ACKNOWLEDGMENTS Catuneanu acknowledges financial support from Natural Sciences and Engineering Research Council (NSERC) research grant to Andrew D. Miall, University of Toronto, and from Rhodes University. Beaumont was supported by an NSERC research grant and the Inco Fellowship of the Canadian Institute for Advanced Research. Waschbusch was supported by an NSERC International Postdoctoral Fellowship and a Killam Trust Postdoctoral Fellowship during part of the time this research was undertaken. We thank Susan Ellis and Russell Korsch for comments, Timothy Lawton and Teresa Jordan for reviews that improved the manuscript, and Earle Kauffman, who more than a decade ago impressed upon Beaumont that there is more to the Western Interior basin than static flexure. REFERENCES CITED Armstrong, R. L., and Ward, P., 1991, Evolving geographic patterns of Cenozoic magmatism in the North American Cordillera: The temporal and spatial association of magmatism and metamorphic core complexes: Journal of Geophysical Research, v. 96, p. 13201–13224. Armstrong, R. L., and Ward, P. L., 1993, Late Triassic to earliest Eocene magmatism in the North American Cordillera: Implications for the Western Interior basin, in Caldwell, W. G. E., and Kauffman, E. G., eds., Evolution of the Western Interior basin: Geological Association of Canada Special Paper 39, p. 49–72. Beaumont, C., Quinlan, G., and Hamilton, J., 1988, Orogeny and stratigraphy: Numerical models of the Palaeozoic in the eastern interior of North America: Tectonics, v. 7, p. 389–416. Caldwell, W. G. E., Diner, R., Eicher, D. L., Fowler, S. P., North, B. R., Stelck, C. R., and von Holdt, W. L., 1993, Foraminiferal biostratigraphy of the Cretaceous marine cyclothems, in Caldwell, W. G. E., and Kauffman, E. G., eds., Evolution of the Western Interior basin: Geological Association of Canada Special Paper 39, p. 477–520. Catuneanu, O., 1996, Reciprocal architecture of Bearpaw and post-Bearpaw sequences, Late Cretaceous–early Tertiary, Western Canada basin [Ph.D. thesis]: Toronto, Canada, University of Toronto, 301 p. Catuneanu, O., Sweet, A. R., Lerbekmo, J. F., and Braman, D. R., 1995, Palynological support in understanding third and fourth order stratigraphic cycles in the uppermost Cretaceous/lowermost Tertiary, Western Canada Basin: An example of reciprocal proximal and distal stratigraphies, in Bell, J. S., Bird, T. D., Hillier, T. L., and Greener, P. L., eds., Proceedings of the Oil and Gas Forum ’95, Energy from Sediments: Geological Survey of Canada Open File 3058, p. 17–22. Catuneanu, O., Sweet, A. R., and Miall, A. D., 1997, Reciprocal architecture of Bearpaw T-R sequences, uppermost Cretaceous, Western Canada sedimentary basin: Bulletin of Canadian Petroleum Geology, v. 45, p. 75–94. Coney, P. J., and Reynolds, S. J., 1977, Cordilleran Benioff Zones: Nature, v. 270, p. 403–406. Constenius, K. N., 1996, Late Paleogene extensional collapse of the Cordilleran foreland fold and thrust belt: Geological Society of America Bulletin, v. 108, p. 20–39. DeCelles, P. G., and Giles, K. A., 1996, Foreland basin systems: Basin Research, v. 8, p. 105–123. Giles, K. A., and Dickinson, W. R., 1995, The interplay of eustasy and lithospheric flexure in forming stratigraphic sequences in foreland settings: An example from the Antler foreland, Nevada and Utah, in Dorobek, S. L., and Ross, G. M., eds., Stratigraphic evolution of foreland basins: SEPM (Society for Sedimentary Geology) Special Publication 52, p. 187–211. Gill, J. R., and Cobban, W. R., 1973, Stratigraphy and geological history of the Montana Group and equivalent rocks, Montana, Wyoming, and North and South Dakota: U.S. Geological Survey Professional Paper 776, 73 p. Gurnis, M., 1992, Rapid continental subsidence following the initiation and evolution of subduction: Science, v. 255, p. 1556–1558. Holt, W. E., and Stern, T. A., 1994, Subduction, platform subsidence and foreland thrust loading: The late Tertiary development of Taranaki basin, New Zealand: Tectonics, v. 13, p. 1068–1092. Jordan, T. E., and Flemings, P. B., 1991, Large-scale stratigraphic architecture, eustatic variation, and unsteady tectonism: A theoretical evaluation: Journal of Geophysical Research, v. 96, p. 6681–6699. Kauffman, E. G., and Caldwell, W. G. E., 1993, The Western Interior basin in space and time, in Caldwell, W. G. E., and Kauffman, E. G., eds., Evolution of the Western Interior basin: Geological Association of Canada Special Paper 39, p. 1–30. Mitrovica, J. X., Beaumont, C., and Jarvis, G. T., 1989, Tilting of continental interiors by the dynamical effects of subduction: Tectonics, v. 8, p. 1079–1094. Ward, P. L., 1995, Subduction cycles under western North America during the Mesozoic and Cenozoic eras, in Miller, D. M., and Busby, C., eds., Jurassic magmatism and tectonics of the North American Cordillera: Geological Society of America Special Paper 299, p. 1–45. Manuscript received April 7, 1997 Revised manuscript received August 26, 1997 Manuscript accepted September 18, 1997

Printed in U.S.A.

GEOLOGY, December 1997