TESTING MODELS OF LATERAL FACIES HETEROGENEITY IN THE UPPER SILURIAN (ILLINOIS BASIN, NORTH-CENTRAL INDIANA): FACIES MOSAIC, STRUCTURAL SAG, OR STACKED INCISED VALLEYS?

Geologic mapping studies must employ an integrated stratigraphic approach to accurately predict the spatial distribution of map units, particularly in areas with limited surface exposure. The Silurian (Ludlow-Pridoli) in north-central Indiana contains a heterogeneous assemblage of peritidal to open marine and reefal facies packages assigned to the Mississinewa, Liston Creek, Kokomo, and Kenneth members of the Wabash Formation. Previous studies differ in their interpretations of these units from lateral age-equivalents to separate depositional packages bound by unconformities. These chronostratigraphic models were equivocal as the techniques employed did not provide the requisite temporal resolution for testing. To test these models, we employ an integrated bio-, chemo-, and sequence-stratigraphic approach.

New and archived drill cores penetrating the Silurian strata of north-central Indiana provide continuous records for sampling. In our study, elemental (pXRF) analysis preceded sampling for conodonts to target least-altered intervals. (Silurian strata in this area contain both limestones and texture-preserving hydrothermal dolostones.) Powdered samples for stable carbon isotope analysis were collected at intervals of 30 to 60 cm in bedded strata and at 60 to 120 cm in the reefs. Unfortunately, conodont samples had low yields with poor preservation. Carbon isotope results show consistent trends and absolute values between closely spaced sections in both limestone and dolostone intervals. Values range from +6 to - 4 per mil and display multiple positive excursions that are temporally constrained through integrating archived conodont collections.

It is clear from this study that the current age definition of the Wabash Formation requires substantial revision. The members of the Wabash Formation are not lateral facies equivalents of one another. The age of the Mississinewa Shale is Gorstian, the Liston Creek is lower to middle Ludfordian (the siluricus Zone missing), and the Kokomo is late Ludfordian. The exact age of the Kenneth Member remains uncertain, but is demonstrably younger than the Kokomo Member. The spatially restricted and temporally discrete distribution of parts of the Liston Creek, Kokomo, and Kenneth members best fits a model of stacked incised valley fills.