Kansas Geological Survey, Current Research in Earth Sciences, Bulletin 254, part 1
Prev Page--Results and Interpretations || Next Page--References

Discussion

Stepping back to look at the overall sedimentation of the Ireland Sandstone Member provides the context for the Lone Star exposures and in turn may enhance understanding of the Ireland and the encompassing Amazonia cyclothem (Heckel, 2002; sequence 6 of Feldman et al. 2005, p. 357). Interpretations for the entire sequence using as a vantage point the postage-stamp-scale outcrop (now destroyed) at Lone Star spillway may appear presumptuous, but by standing on the shoulders of prior giant studies, we can see the big picture. The depositional setting of the upper Ireland Sandstone Member (table 1) can be summarized according to the attributes of incised valleys set out by Dalrymple (2006, p. 7).

Table 1--Depositional setting of Ireland Sandstone Member at Lone Star spillway, based on attributes of Dalrymple (2006, p. 7).

A high rate of accommodation for this location on the stable craton (table 1) may be inferred on first principles from high-amplitude oscillations of sea level resulting from contemporaneous continental glaciation in the Southern Hemisphere, the presumed driver of midcontinent cyclicity. At the local scale, the accumulation of more than 15 m of uniform tidal deposits (base of exposure to base of Amazonia Limestone Member, interrupted briefly by coal deposition), a considerable thickness for the midcontinent, indicates that accommodation kept pace with sedimentation. Sedimentation was inferred to be rapid, so slightly circular logic indicates rapid accommodation. As the channel-filling Ireland Sandstone Member is fluvial and the overlying rocks show increasing marine influence (Feldman et al., 2005, p. 357), the entire interval is transgressive. This is true up to the Amazonia Limestone Member, with the exception of the fleeting tryst with the shoreline that resulted in the lower Williamsburg coal. Thus accommodation outstripped sedimentation throughout this interval. The upper part or top of the Amazonia marks maximum flooding and the beginning of a thin highstand systems tract (HST; Feldman et al., 2005, p. 364) that led to prolonged exposure with soil development beneath the upper Williamsburg coal (fig. 3; Joeckel, 1994; Feldman et al., 2005).

Evaluating the sources of sediment (table 1) is more problematic. Tidal processes generally transport marine sediment landward. A marine contribution in this setting, however, would be expected to contain appreciable carbonate, which is absent in the tidal deposits. Feldman et al. (2005, p. 364) note that the HST of their Amazonia sequence, which begins with the Ireland, includes "limestone on upper shelf," [part of the Amazonia] but also considerable shale. On the other hand, the Ireland Sandstone Member and lateral correlatives in the Lawrence Formation include volumes of silt and clay, so the fluvial end of the estuary may have supplied much of the mud to the higher tidal flats. Fluvial supply was apparently high. Marine supply may have been low or may have been moderate.

The size of the sediment varied fairly regularly from cobbles of limestone, shale, coal, and phosphate in the basal valley fill (fig. 3; Bowsher and Jewett, 1943; Feldman et al., 2005, p. 357), through crossbedded, fine- to medium-grained sandstone in the fluvial fill (O'Connor, 1960), up to 50 m thick (Sanders, 1959), grading upward to mudstone with subordinate siltstone in thin laminae and very fine grained sandstone, virtually confined to small channels, in the tidal estuarine deposits.

The Lone Star site was at least 150 km from the limit of incision by the Ireland valley, which probably extended into Woodson and Greenwood counties (fig. 3) and perhaps to the shelf-edge in southernmost Kansas. The Ireland Sandstone Member can be traced at least to Greenwood County (Sanders, 1959), but incision is difficult to establish in the absence of marker beds above the Haskell Limestone Member (fig. 3). The distance to the mouth of the postulated estuary was probably a moving target, as slight fluctuations in sea level or sediment supply could displace the shoreline dramatically in the very flat Pennsylvanian landscape.

This poses the question, were the interfluves of the Ireland drainage network eventually submerged and covered with sediment? This seems likely as the only good marker bed within this sequence, the Amazonia Limestone Member, although discontinuous (fig. 3), can be recognized from central Kansas into Missouri, Nebraska, and Iowa (Feldman et al., 2005, fig. 5). More to the point, though, how extensive are the tidal-estuarine deposits? In principle, it should be possible to distinguish the tidal deposits of the Ireland from the underlying Robbins Shale Member, generally considered "prodeltaic" (Ball, 1985). However, this requires good outcrops and a careful study that has not been undertaken. Moreover, a paleosol, which is to be expected at this sequence boundary and would certainly facilitate a study, has not yet been recognized (Feldman et al., 2005, p. 357).

Despite some uncertainty about drowning of the interfluves, the generally transgressive character of the Amazonia cyclothem, up to the Amazonia Limestone Member near the top, indicates that the Ireland paleovalley was filled and the interfluves were eventually flooded during transgression. This would qualify the Ireland incised valley as "overfilled" in the sense of Garrison and van den Bergh (2006).

Transgression in the Amazonia cyclothem was not quite continuous, however. The paralic coal of the lower Williamsburg represents an interruption when the depositional surface caught up with sea level briefly, perhaps for as little as 1,000 years (Archer et al., 1993, p. 1-20). The distribution of the coal in cross section (fig. 3) strongly suggests that it was confined to the Ireland channel. If this is true, then the channel was not quite filled yet, and the interfluves must have been exposed while the channel axis was a coal swamp, probably at the head of a narrow estuary.

Elsewhere, rhythmically laminated sequences very similar to the sequence at Lone Star Lake spillway have commonly been interpreted as tide-dominated deposits. Kvale et al. (1989) studied laminated siltstones within the Mansfield Formation of Orange County, Indiana. The stacked, coarsening-upward siltstone laminations in that study were interpreted as deposits from a tide-influenced environment in which neap- and spring-tide variations were recorded as variations in thickness of lamination. Other examples from many similar deposits (cf. Klein, 1970; Weimer et al., 1982) are the Elatine Formation (late Precambrian) in Australia (Williams, 1988; Sonett et al., 1988) and the Francis Creek Shale Member of the Carbondale Formation (Middle Pennsylvanian) of northern Illinois (Kuecher et al., 1990). The Brazil Formation (Lower Pennsylvanian) in Indiana (Kvale and Archer, 1990) has channel-fill sequences similar to those observed in the Lawrence Formation at the Lone Star spillway.

The recognition of sedimentary structures and architecture within the Tonganoxie Sandstone Member of the Douglas Group (fig. 2) indicative of transitions from fluvial to tidal estuarine to marine deposition from northeast to southwest across Kansas (Archer et al., 1993, 1994; Feldman et al., 1995) has popularized the estuarine model for deposition of sand and shale intervals in the midcontinent. It is unlikely that most midcontinent units would fit into the dimensions of any estuary, but the occurrence of the study interval near the top of the Ireland Sandstone Member incised-valley fill greatly strengthens the case for an estuarine setting. The Ireland Sandstone Member is strikingly similar to the underlying Tonganoxie Sandstone Member in many respects (Sanders, 1959; O'Connor, 1960; Ball, 1964; Archer et al., 1994), but it has not received the extensive regional study necessary to delineate similar facies trends. This is in part because the trend of the Ireland paleovalley in outcrop is roughly westward (Sanders, 1959; O'Connor, 1960), severely limiting lateral exposure in the NNE-SSW-trending outcrop belt.

Despite the indications of tidal activity and the superposition of a coal bed and eventually a regional soil marking a sequence boundary near the top of the Lawrence Formation (Joeckel, 1994; Feldman et al., 2005), no indications of exposure such as desiccation cracks, raindrop impressions, or root casts were observed within the study interval of the Lawrence Formation. The only exception is the interval of blocky, gleyed mudstone with "extensive rooting" (Archer et al., 1993, p. 1-22) beneath the lower Williamsburg coal.

Conclusions

Exposures of the upper part of the Ireland Sandstone Member of the Lawrence Formation at the Lone Star spillway consist of interlaminated shale, siltstone, and sandstone with pinstripe, lenticular, and minor flaser lamination. Channel-form sandstone and mudstone lenses are prominent in the upper part of the study interval. Bedding patterns, sedimentary structures, bipolar paleocurrents, and dominance of mud indicate deposition in a tide-dominated upper estuary. Modern depositional systems comparable to the Lone Star Lake site in lithology and sedimentary structures are found in large estuaries. The numerous deformational features were imposed on soft sediment rapidly deposited with inverse density gradients. These interpretations support and somewhat amplify those of Archer and West (1991), Archer (1993), Lanier (1993), and Feldman et al. (2005) for this part of the Lawrence Formation.

Acknowledgments--We thank classmates in sedimentology who measured detailed sections at the Lone Star spillway: Jeremy Littlejohn, Jon Holmgren, Ryan Pearson, Matt Brookshier, Glenn Newell, Jeff Zuehlke, John Keller, Lisa Armatas, Victoria G. Christensen, and Doug Linger. Sections measured by the 1992 class: Matthew Briney, Peter Cattaneo, Yenli Choong, Merritt Forman, Richard Godsil, Monica Hochanadel, Terrence Huettl, Sheila Kortlucke, Daria Sander, Gregory Siek, and Robert Younger were integrated, or in some cases served as the primary source, in fig. 5. Thanks to Steve Franklin, laboratory instructor, photogenic scale, and enthusiast, and Brian Macy, who helped scan initial photographs, for their involvement in this project. Greg Seik helped generate interest in the project but had to move on before it was completed. Alex Martinez and Scott Beaty labored heroically to meld the outcrop photographs into a single panorama, but were frustrated by 1995 technology. It remained for John Charlton (KGS) to merge the photographic overlaps and produce the panorama using Adobe Photoshop® CS3, extended version. Jennifer Sims (KGS) produced the final figures. All this support is gratefully acknowledged.

Helpful reviews of an early version of the manuscript were generously provided by Allen W. Archer, Kansas State University, and Howard Feldman, then of the Kansas Geological Survey. The final manuscript benefited from reviews by George deVries Klein, Sed-Strat Geoscience Consultants, Inc., Sugar Land, Texas, and Howard Feldman, now with ExxonMobil Exploration Co.

Prev Page--Results and Interpretations || Next Page--References

Kansas Geological Survey
Placed online July 16, 2008
http://www.kgs.ku.edu/Current/2008/Enos/03_results.html
email:webadmin@kgs.ku.edu