COLLABORATIVE RESEARCH: INTEGRATED STRATIGRAPHIC STUDY OF CAMBRIAN-ORDOVICIAN INNER SHELF FACIES OF THE WESTERN UNITED STATES

Paul M. Myrow, Department of Geology, The Colorado College

Robert L. Ripperdan, Department of Geology, University of Puerto Rico, Mayagüez

John F. Taylor, Geoscience Department, Indiana University of Pennsylvania

PROJECT SUMMARY: Extinctions through the Cambrian-Ordovician boundary interval set the stage for the Ordovician radiation, one of the most important diversification events in the history of marine invertebrates. Stratigraphic data from carbonate platform and off-platform strata in the Great Basin and elsewhere have been used to propose process-response models that invoke "eustatic events" as a forcing mechanism for the extinctions. However, sedimentological interpretations that form the basis of these models are (at best) inconclusive. They rely on ambiguous lithologic information and/or lack the precision of correlation necessary to provide a rigorous test of the linkage between extinctions and paleoceanographic events. Also lacking are detailed data from nearshore and shoreline environments, settings where facies are highly responsive to relative sea level change.

For these reasons, an integrated biostratigraphic, chemostratigraphic, and sequence stratigraphic analysis of Upper Cambrian and Lower Ordovician rocks in Colorado (previous NSF grant to Myrow) was conducted over the last three years. Despite the abundance of stratigraphic gaps in many sections, and the sparsely fossiliferous character of these inner shelf facies, this multidimensional study produced some of the most complete sections yet documented for the Cambrian-Ordovician "inner detrital belt". Additionally, trilobite faunas and lithofacies documented under the previous grant more tightly constrain the position of the Transcontinental Arch, which separated the eastern and western (modern coordinates) shelves of the continent at that time.

The proposed research will extend this study into coeval rocks in Wyoming/Montana and the southwestern United States, which offer sections from slightly more distal inner shelf settings on the west and east sides of the Transcontinental Arch, respectively. Continued use of multiple correlation tools, and utilization of the varied data in Graphic Correlation, will result in unprecedented precision of correlation of measured sections with one another, as well as with standard Cambrian-Ordovician successions in Utah and Oklahoma. This precise, integrated stratigraphic framework will be used to monitor sea level behavior (recorded in the vertical succession of lithofacies and sequence boundaries), paleoceanographic events (reflected in the isotope stratigraphy), and bioevents (represented by extinction horizons and intervals of adaptive radiation).This will allow us to test null hypotheses that horizons of extinction are marked by isotopic anomalies and/or lithofacies shifts, suggesting that paleoceanographic events forced the faunal changes. Comparison of relative sea level curves from opposite sides of the arch will test the validity of "eustatic events" proposed in previous studies of Cambrian-Ordovician strata on various continents. The precise chronostratigraphic framework developed in this study will also provide necessary temporal constraints for future chemostratigraphic, paleomagnetic, and paleobiologic studies of this important statigraphic interval.

Myrow and Taylor will continue to involve undergraduates at Colorado College and Indiana University of Pennsylvania as junior collaborators in all components of the research; Ripperdan and his students at the University of Puerto Rico, Mayagüez will follow this model. Twelve students (9 from CC, 3 from IUP) benefited in this way from the previous NSF grant. A minimum of 15 students will participate over the term of the 3-year grant described in the present proposal. Greater participation by IUP undergraduates not only increases the total number of students involved, but extends the opportunity for research involvement to another extremely important population -- education majors preparing for a career in secondary science education. The participation of undergraduates from the University of Puerto Rico, Mayagüez and the exchange of students between the PI host institutions provides a much-needed opportunity for students to experience scientific research in new intellectual, geographical and cultural environments.

STATEMENT OF PROBLEM

The numerous extinctions that affected shallow marine faunas on the tropical shelves surrounding Laurentia in the Cambrian and Early Ordovician have been the focus of many detailed biostratigraphic, evolutionary, and paleoecologic studies. The extinction horizons have proven to be ideal for subdivision of the Cambrian System into well-defined series and stages (Ludvigsen and Westrop, 1985; Palmer, 1998) that allow precise correlation of shallow marine facies throughout Laurentian North America. From a macroevolutionary perspective, the intervals between extinction horizons (biostratigraphic units known as biomeres) are intriguing as well-preserved archives that offer valuable insight into the patterns and processes of biotic recovery following the crises that periodically decimated invertebrate faunas, particularly (but not exclusively) the trilobites. Recent compression of the Cambrian time scale (Bowring et al., 1993), particularly for the Middle and Late Cambrian which may be only 20 m.y. in duration (Landing et al., 1998), means that the extinction and radiation events of the Late Cambrian were even more rapid than previously thought. (***Careful here; the temporal compression induced by the Bowring date noted above might be for only the Early and Middle Cambrian. See also references by Davidek and another Landing paper lead-authored by (I think) one of Sam’s students. References can be found in Doug Erwin’s latest GSA Today cover article on evolutionary rates) ..Paul, address this and remove this comment. Cambrian biological experimentations were fast-paced, and the transition into more stable, long-standing Phanerozoic faunas, which is accomplished by the end-Cambrian extinction and subsequent Ordovician radiation, is one of the most dramatic shifts in the history of the marine biosphere.

Many aspects of the extinctions,the intervening adaptive radiations, and the physical record of both remain a matter of extensive and healthy controversy that extends across a broad range of specialties in both the geological and biological sciences. A variety of extinction mechanisms have been offered (Stitt, 1977; Palmer, 1984; Westrop, 1990), but no clear consensus (or even majority opinion) has yet emerged. Data from carbonate platform and off-platform strata have been used to propose process-response models that invoke sea level change as a forcing mechanism for extinctions and/or radiations within the Cambrian and Early Ordovician. Some workers (Palmer, 1984; Saltzman et al., 1995) have suggested that oceanic overturn or destratification, perhaps triggered by sea level rise and fueling a strong shift in marine _13C values, caused the Late Cambrian extinctions. Others, noting evidence of onlap of slope lithofacies onto the outer margin of the carbonate platform near horizons of faunal turnover, attribute the extinctions to biogeographic shifts (Westrop and Ludvigsen, 1987) or flooding of the platform with cool and anoxic waters (Palmer and Taylor, 1984; Palmer, 1984). However, studies in coeval, more proximal platform strata (Stitt, 1977; Taylor et al., 1992; Loch and Taylor, 1995) reveal no evidence of transgression, but instead document facies changes suggestive of regression associated with at least some of these extinctions. Some regressive features observed near horizons of faunal change within the Cambrian-Ordovician boundary interval on various continents have been used to propose a series of "eustatic events" (Nicholl et al., 1992). These include the "Lange Ranch Eustatic Event" and "Black Mountain Eustatic Event " of Miller (1984, 1992) and the Acerocare Regressive Event and Peltocare Regressive Event of Erdtmann (1986). There is much debate about the nature of these proposed events (Ludvigsen et al., 1986; Taylor et al., 1992; Landing, 1993) based, at least in part, on the ambiguous nature of the sedimentological data and insufficient precision of correlation.

Ultimately, a rigorous test of the proposed linkage between extinctions and paleoceanographic events within the Cambrian-Ordovician boundary interval will require varied, high-resolution stratigraphic data from a complete onshore-offshore profile, including inner shelf, platform, and shelfbreak settings. Detailed data from inner shelf environments are particularly critical because facies and fauna in these settings are highly responsive to relative sea level changes and other environmental perturbations. However, the mixed carbonate and siliciclastic facies that dominate inner shelf successions in the Cambrian-Ordovician deposits of Laurentia are generally less fossiliferous than coeval carbonate platform facies, and in places contain numerous stratigraphic gaps. As a result, they have received less attention than outer shelf facies. Although much has been learned over the past two decades about environmental controls on the content of Upper Cambrian and Lower Ordovician trilobite faunas (Taylor, 1977; Ludvigsen and Westrop, 1983; Westrop, 1986, 1989; Hohensee and Stitt, 1989; Loch and Taylor, 1995; Taylor et al., in press), data on trilobite biofacies is sparse for faunas that occupied proximal shelf and cratonic environments. In addition, sedimentologic information in most previous biostratigraphic studies of inner shelf assemblages is either lacking or of insufficient precision to determine whether horizons and intervals of faunal change coincide with sequence boundaries or significant facies transitions. Chemostratigraphic techniques such as carbon isotope stratigraphy, when used in conjunction with high-resolution biostratigraphic information and detailed sedimentologic and sequence stratigraphic data, can significantly enhance temporal correlations between these deposits and the more heavily-studied distal platform facies. Additionally, high quality chemostratigraphic data are a proxy for secular changes in ocean chemistry, thus providing vital information concerning paleoceanic changes that may be tied to geological and biological events.

The proposed research will create an integrated stratigraphic framework in which sea level behavior (as recorded in the stratigraphic succession of lithofacies and sequence boundaries), paleoceanographic events (reflected in the isotope stratigraphy), and bioevents (extinction horizons and intervals of adaptive radiation) are monitored with data collected from inner shelf successions. This will allow testing of null hypotheses that horizons of extinction are marked by isotopic anomalies and/or lithofacies shifts, suggesting that paleoceanographic events forced the faunal change. Unprecedented precision of correlation will be possible through utilization of this varied and precise (sub-meter scale) data for Graphic Correlation (Shaw, 1964; Mann and Lane, 1995), in which all measured sections will be compared with one another, as well as with the extensively studied Cambrian-Ordovician "standards" in the Ibex area of Utah (Ross et al., 1997; Ripperdan and Miller, 1995) and the Arbuckle Mountains of southern Oklahoma (Stitt, 1977, 1983).

PROPOSED STUDY

General - Results of the previous grant for work in Colorado established that proximal marine deposits of the central and southern Rocky Mountain region are amenable to integrated stratigraphic study, and that high-precision data can be recovered from these inner shelf facies through modern stratigraphic techniques. They also confirm the precise location of the Transcontinental Arch, allowing confident assignment of sections in westernmost Colorado (White River Plateau and northwestern Sawatch Range) and Wyoming/Montana to the western shelf. Exposures to the east and south of the Homestake Shear Zone, including those in New Mexico and southeast Arizona, formed on the opposite side of Laurentia, facing the Iapetus Ocean. The previous study also revealed that sections in Colorado contain numerous unconformities that omit critical intervals for evaluation of hypotheses linking extinctions to paleoceanographic events. For this reason, the proposed research focuses on sections deposited slightly farther out on the two shelves where subsidence rates were somewhat higher, although not as great as those that produced thicker, carbonate-dominated successions in outer platform and failed rift settings like the Great Basin and the Southern Oklahoma Aulacogen, respectively.

There are no comprehensive studies of conodont and trilobite biostratigraphy and carbon isotopic signatures of proximal deposits in the Rocky Mountain belt from Montana to Arizona. However, previous trilobite studies of these strata, generally regional work from the 1950's and 1960's (e.g., Grant 1965), indicate great potential. Although proximal facies are locally problematic, the Colorado work has clearly indicated that useful, relatively high-resolution data are recoverable through intensive and systematic sampling. This is corroborated by trilobite and conodont collections made by Kurtz (1976) and by a chemostratigraphic pilot study by McGlaughlin (1998), a Wooster College student supervised by Myrow. Thus, the potential for comparison of stratigraphic patterns at various locations along the strike of the Rocky Mountains, and across strike into the Great Basin, has been established.

Our goal is to determine the spatial and temporal variability in the sedimentological and geochemical signatures of Cambrian-Ordovician boundary events in order to gain insight into the mechanisms responsible for extinction and radiation. Precise correlations made possible through use of multiple fossil groups and non-paleontologic chronostratigraphic correlation methods will be enhanced by graphic correlation in which physically and faunally defined event horizons are utilized. This will establish the distribution and magnitude of unconformities within the inner shelf successions along strike on both sides of the Transcontinental Arch for comparison with the more outboard settings of the Great Basin and the Appalachians. In combination with sedimentological data, these correlations will allow comparison of relative sea level curves from opposites sides of the arch and establish the presence or absence of evidence for sea level change in association with extinction horizons and _13C variations.

Chemostratigraphic curves are a key aspect to this work as they provide for more detailed correlation of sections and hold important information about secular changes in ocean chemistry that may have accompanied the bioevents. Additionally, evidence from inner shelf facies collected in the proposed study may prove critical for assessing claims of Cambro-Ordovician eustatic signals. Strongly contrasting sea level histories recorded from opposite sides of the Transcontinental Arch would effectively falsify eustatic hypotheses. Similarly, evidence for uplift and faulting, which may be preserved in the stratigraphic architecture of proximal facies, would cast doubt on a eustatic influence.

Data recovered in the proposed research will be used to try and falsify the following null hypotheses:

1. Extinction horizons are associated with physical evidence of sea level rise or fall (sequence boundaries or facies change), consistent with hypotheses that link faunal change to excursions in sea level.

2. Horizons of faunal turnover are marked by predictable _13C variation, reflecting a linkage to paleoceanographic or paleoclimatological events such as oceanic overturn, oceanic destratification, or glaciation.

3. Chronostratigraphically constrained relative sea level curves from opposite sides of the Transcontinental Arch are similar in structure and validate the existence of "eustatic events" proposed in previous studies for Upper Cambrian and Lower Ordovician strata.

Although the aforementioned hypotheses are the focus of the study, the integrated data set that will be compiled will be useful in assessing other aspects of Laurentian Cambrian-Ordovician faunas and environments that are important for continental and global scale reconstructions of the Early Paleozoic. For example, it will reveal the degree to which certain lithofacies (e.g., extremely glauconitic sandstone, shale with interbedded flat-pebble conglomerate, etc.) and/or biofacies are constrained to one side of the arch or the other at specific times. These data will elucidate the physical and/or chemical controls on depositional systems and the spatial and temporal distribution of specific facies. Such a database is necessary for development of models that reconstruct such paleoceanographic-paleoclimatic phenomena as the position of shorelines relative to wind belts, nature of oceanic circulation, and degree of oceanic stratification.

Similarly, these data will establish the degree to which endemism and biofacies differentiation developed in trilobite faunas on either side of the Transcontinental Arch. Precise chronocorrelation between sections on opposite sides of the arch (using widespread species, physical event stratigraphy, and chemostratigraphic events) will allow for comparison of faunas from coeval intervals and assessment of degrees of endemism displayed by faunas on opposite shelves at various times in the Late Cambrian and Early Ordovician. To this end, the planned research will include a comparison of trilobite faunas from several zones of moderate to high diversity within both the Upper Cambrian (Prosaukia, Illaenurus, and Saukia Zones) and Lower Ordovician (Bellefontia-Xenostegium and Leiostegium-Kainella Zones). There will not be sufficient time in this 3-year grant to completely characterize and compare faunas from all these intervals; selection of the faunas to be described fully will be based on the results of our initial sampling (see Research Plan). The proposed integrated study should be a model for the type of work emphasized in such programs as the Ordovician Global Biodiversity Work Program — IGCP Project 410, an initiative under the Ordovician Subcommission of the IUGS, concerned with Late Cambrian to Early Ordovician history.

Previous chemostratigraphic studies of the Upper Cambrian and Lower Ordovician have failed to resolve the precise interrelationships between variation in _13C values, trilobite evolution, and sea level. A dramatic shift in _13C values is known to occur globally directly below the base of the Ibexian within the Ptychaspid biomere (Ripperdan et al., 1992; Ripperdan and Miller, 1995; Ripperdan 1997), but the precise relationship of the _13C excursion with trilobite faunal changes has not yet been established. A complete understanding of is crucial in order to assess the universality of models developed from the better-studied, underlying Pterocephaliid biomere (Saltzman et al. 1998).

High-resolution chronostratigraphic frameworks, assembled through utilization of multiple correlation tools, provide an essential foundation for subsequent investigations that address an even wider range of geologic problems than those outlined above. For example, recent hypotheses on the mechanism(s) responsible for rapid change in continental plate orientation in the Late Neoproterozoic and Cambrian (Kirschvink et al., 1997) were formulated with paleomagnetic data tightly constrained within a chemostratigraphic and biostratigraphic framework. In this regard, the interval proposed for study (Upper Cambrian and Lower Ordovician) is of particular interest as a "quiescent interval” between two suggested periods of rapid true polar wander, the Neoproterozoic-Early Cambrian IITPW event (Kirschvink et al. 1997) and the Late Ordovician to Early Silurian (Van der Voo, 1994). The proposed project will set the stage for future research in this interval by providing critical temporal constraints for studies that require a precise framework for chronocorrelation.