Initiation of the Western Interior foreland basin
- *E-mail: miall{at}geology.utoronto.ca.
In this issue of Geology, Fuentes et al. (2009, p. 379) present a new subsidence plot and age data from detrital zircons that help to constrain the date of initiation of the foreland basin in Montana (United States). They offer these data as evidence for the resolution of a controversy regarding the initiation of foreland basin sedimentation in the Western Interior. However, their discussion raises the question of what geological data constitute the definitive evidence of the foreland basin style of sedimentation, and therefore the criteria that should be used to define its commencement.
A useful definition of a foreland basin is that it is a depression that develops adjacent and parallel to a mountain belt. Foreland basins form because the mass created by crustal thickening associated with the evolution of a mountain belt causes the lithosphere to bend, by a process known as lithospheric flexure. At convergent plate margins, foreland basins develop behind the marginal arc (retroarc or retroforeland type; Jordan, 1995) or above the downgoing continental plate adjacent to a collision zone (peripheral or proforeland type; Miall, 1995). The downward flexure is accompanied by the uplift of a forebulge in the basement, typically a few hundred kilometers out from the fold-thrust belt, and beyond that there is a shallow depression called the backbulge basin.
The relationship between crustal loading and basin formation was first articulated by Price (1973) with respect to the foreland basin of southern Alberta, and subsequent modeling of the Alberta Basin by Beaumont (1981) demonstrated the quantitative link between flexural loading, the formation of a crustal depression, and its filling by syntectonic sedimentation. Jordan (1981) demonstrated another key criterion for the definition of the foreland basin style of subsidence and sedimentation: the distinctive isopach pattern of the basin fill, elongated parallel to the mountain front and asymmetric in cross section, with a depocenter located close to the contemporaneous fold-thrust belt, the uplift of which serves as the proximal sediment source. Cross (1986) used this criterion to distinguish foreland basin subsidence from that due to subcrustal mantle loading, which yields a much broader, wider pattern of basinal subsidence.
The Western Interior basin (of which the Alberta basin and the Rocky Mountain basin are parts) was initiated as a retroarc basin as a result of a first-order change in the plate kinematics of Pangea. This supercontinent, assembled by multiple plate convergence and suture between Ordovician and Permian time, began to fragment by rifting in the Triassic, and by early Middle Jurassic time, oceanic crust was forming off what is now the Atlantic coast of the United States. The North American continent began a long process of northwestward drift, which carried the plate some 70° of longitude westward, relative to the Pacific plate, and, some 40° of latitude northward until Paleocene time, followed by a change in trajectory and a subsequent southward drift of ~10° up to the present day (Engebretson et al., 1985).
There are numerous indicators of a first-order change in magmatic, tectonic subsidence and sedimentation patterns associated with this change in plate kinematics. The first appearance of westerly-derived detritus from a rising arc or orogen has long been accepted as providing the timing of initiation of the Western Interior foreland basin. Prior to the establishment of the basin, the western continental margin was miogeoclinal in character, with sediment derived from easterly sources (the craton and the Canadian Shield). Ron Blakey and I (in Miall, 2008), designated the Morrison Formation of the Colorado Plateau area as the unit that first shows the predominance of westerly sediment sources (Blakey, 2008, p. 282). In Alberta, the first appearance of westerly-derived detritus occurs at the transition from the marine Fernie Shale to the deltaic Kootenay Formation (Miall et al., 2008, p. 339; not the same unit as the Kootenai of Montana). These indicators both place the initiation of the foreland basin in the late Jurassic, Kimmeridgian, between ca. 150 and 155 Ma ago.
How does this differ from the data provided by Fuentes et al. (2009), and is the timing of initiation of the basin controversial? Their detrital zircon ages and subsidence plot suggest initiation of foreland basin subsidence at ca. 170 Ma before present. They suggest that the succession they examined in Montana was deposited in the backbulge of the foreland basin, a zone that subsequently migrated eastward as a result of crustal shortening through the Cretaceous (DeCelles, 2004), and implying that the contemporaneous foredeep would have been further west and had subsequently been uplifted and eliminated by Sevier tectonism. DeCelles and Currie (1996) similarly suggested that the Morrison Formation of Utah was deposited in the backbulge basin, but their subsidence plot indicates that subsidence commenced earlier, at ca. 170 Ma ago, with deposition of the underlying Preuss and Twin Peak formations. These units are contemporaneous with the Carmel Formation, which is the first unit to show (in northern Arizona) a “strong shift in fluvial paleocurrents relative to older Jurassic and Triassic rocks” (Blakey, 2008, Fig. 31, p. 282), and is contemporaneous with marine carbonates that define a north-south belt through central Utah and southeast Idaho in the Utah-Idaho Trough (Blakey, 2008, Fig. 31). Is this a foreland basin (Bjerrum and Dorsey, 1995), or the product of dynamic subsidence (Lawton, 1994)? Could the Utah-Idaho Trough be the precursor of the Western Interior foreland basin? Subsidence patterns developed by Bjerrum and Dorsey (1995) suggest the commencement of subsidence of the trough in the early Middle Jurassic (Bajocian; ca. 170 Ma ago), consistent with the timing proposed by Fuentes et al. (2009). However, the sedimentation patterns are different, consisting in part of marine carbonates, and showing only limited evidence of westerly sediment sources (actually southwesterly, and only in northern Arizona). It may, therefore, be in part a question of semantics as to what defines the initiation of a foreland basin: subsidence style, sedimentary facies, or isopach patterns?
Is there a discrepancy between the age of initiation of the basin in Montana (ca. 170 Ma before present according to Fuentes et al., [2009]) and nearby Alberta (the base of the Kootenay Formation, with an age of ca. 152 Ma before present according to Poulton et al. [1994])? Ross et al. (2005) reported sandstone petrofacies and detrital zircon ages from the foreland basin clastic wedge of southwest Alberta very similar to those of Fuentes et al. They did not recover zircons from the Kootenay Formation. Their zircons from the much younger Dalhousie Sandstone (Aptian) have a similar distribution to those recovered from the Morrison and Kootenai formations of Montana by Fuentes et al., including a cluster in the 150–170 Ma age range. These indicate the age of magmatic cooling, but not necessarily the age of detrital deposition. Uplift, unroofing, erosion, and transportation could cause a significant lag between zircon crystallization and subsequent sedimentation.
The temporal link between load-induced subsidence and sedimentation is a complex one. As various researchers, including Heller, Paola, and Burbank, have debated (see Miall, 1996, Sect. 11.3.6), the traditional genetic link assumed to exist between tectonism and clastic-wedge pro-gradation (going back to Sloss, 1962) may be complicated by variations in flexural rigidity of the crust, trajectories and rates of crustal shortening, the consequent rates of subsidence, and variations in source-area geology and climate, all of which influence sediment supply, relative sea level, and regional paleogeography (see summary in Miall, 1996, Sect. 11.3.6).
The best evidence for mid-Jurassic initiation of foreland basin subsidence in Montana is the subsidence plot of Fuentes et al. (2009), a mode of analysis that has not yet been applied to the succession in Alberta. Clearly it should be. Subsidence analysis is a sensitive indicator of the relationship between sedimentation and tectonics (Allen and Allen, 2005). Given that the original definition of a foreland basin, going back to Price (1973), places emphasis on the relationship between tectonic loading and subsidence, this should perhaps be accepted as the primary indicator of basin initiation. In no way, however, does this negate the valuable stratigraphic and other work cited by Fuentes et al. (2009) as contributing to the controversial question about the age of initiation of the basin.
Not discussed in that paper (Fuentes et al., 2009) are reasons why the initiation of the basin could, in fact, be diachronous. Although the primary cause of foreland basin initiation is the change along the western continental margin from an extensional to a contractional regime, in detail the nature of the tectonic regime along the leading edge of the continent, from Mexico to northern British Columbia, varied considerably, reflecting regional variations in preexisting structural conditions. In Yukon Territory, the Whitehorse Trough, a forearc basin of Lower to Middle Jurassic age, contains the first clear evidence of a contractional regime along that part of the continental margin (Ricketts, 2008). In central British Columbia, contraction of the Nicola Arc (which initiated uplift and erosion of the first clastic sediment source for the foreland basin in Alberta) began with delamination of the backarc basin in the early Middle Jurassic (Price and Monger, 2003). Arc magmatism in the Sierra Nevada of California commenced in the Triassic (DeCelles, 2004), but detrital evidence of this activity did not reach Utah until the Middle Jurassic. Given this along-strike variation in the initial timing and style of contractional tectonism, and the time required for this to result in the establishment of a crustal load (because of local variations in crustal heterogeneity), it would be surprising if the establishment of the foreland basin was, in fact, of exactly the same age along its full 3000 km length.
- © 2009 Geological Society of America
