Sourcerocker: A Heuristic Computer Program that Predicts the Occurrence of Source Rocks Using Information from Palogeography and Paleoclimate Models

Christopher Scotese¹, Thomas Moore², Harold Illich³, and John Zumberge^3
1 PALEOMAP Project, Arlington, TX
² PaleoTerra, Bollingbrook, IL
³ GeoMark Research, Inc, Houston, TX

Source rocks do not occur randomly in either space or time. Rather, source rocks seem to occur more abundantly at some periods in Earth History, than at other times . Similarly, source rocks do not occur equally frequently in all depositional environments or under all climatic and oceanographic conditions.

To solve the problem of source rock production in space and time, we have assembled 3D paleogeographic maps (Paleo DEMs) that model surface topography and bathymetry for eight time intervals. These topographic and bathymetric models provide the framework for paleoclimatic simulations, (FOAM) that give important information concerning the past state of the ocean and atmosphere. The results from these paleoclimatic simulations, in turn, allow us to model the key surficial processes, such as sediment transport, the depth of the mixed layer, and biological productivity, and other criteria that to a large extent control the spatial and temporal distribution of source rocks . The goal of the GANDOLPH Project is to use the artificial intelligence tools developed in SourceRocker to predict and better evaluate the potential of source rocks in basins that are under-explored or in deeply buried strata that have yet to be drilled.