PARCELL, WILLIAM C. , Department of Geological Sciences, University of Alabama, Tuscaloosa, AL
A new computer model predicts depositional facies distribution and productivity rates on a 3D carbonate platform using continuous set theory (fuzzy logic), which accepts complex, linguistic data with large uncertainties. Earlier deterministic models have become valuable tools for stratigraphic interpretation, but have a difficult time with linguistic concepts whose parameters are difficult to define. For example, deterministic models estimate "depositional facies" based solely on water depth. The term "depositional facies" involves many more complex, nonlinear processes including water energy, circulation, climate, and distance from siliciclastic input. The current model integrates water depth with such complex variables. The user approximates parameters for each of the variables using linguistic terminology. The model produces depositional facies including sabkha, tidal flat, lagoon, reef, shoal, and subtidal.
Carbonate productivity is controlled by many complex processes such as water depth, temperature, latitude, turbidity, climate, and water energy. Deterministic models use either simple water-depth functions or apply a constant rate to model productivity within the shallow water zone. Both deterministic methods apply the same productivity rates per given water depth across a model's extent often resulting in erroneously high sediment volumes. The current model integrates water depth with interrelated variables allowing variation in productivity depending on processes occurring at specific locations.
Development of this model makes it possible to incorporate linguistic and imprecise geologic data into stratigraphic simulations and thus allows visualization of the response of carbonate systems to interactions of complex processes. This also allows greater potential for accurately predicting carbonate reservoir distribution.
AAPG Search and Discovery Article #90908©2000 GCAGS, Houston, Texas