Fuzzy Predictive Earth Analysis Constrained by Heuristics Applied to Stratigraphic Modeling

A three-dimensional computer simulation was developed to examine the sensitivity of stratigraphic evolution of a continental margin to variable sediment influx, subsidence and sea level. The preliminary results of this model, created using MATLAB and referred to as fuzzyPEACH ( Predictive Earth Analysis Constrained by Heuristics), are presented in this chapter. Fuzzy logic provides a method to quantify subjective variables with a collection of linguistic-based rules and fuzzy sets. To date, few geological investigations have utilized fuzzy logic to simulate complex, geologically reasonable stratigraphic responses to variable depositional systems. Fuzzy logic was chosen as a modeling platform because it is easy to use, adaptable, and a computationally efficient alternative to traditional, mathematically complex models. FuzzyPEACH uses a series of five fuzzy inference systems that rely on 15 rules and 47 fuzzy sets to control grain size, sediment distribution, fluvial avulsions, compaction of sediment, and isostatic (load-driven) subsidence. User-defined variables include margin physiography (width, depth of shelf-slope break), sedimentation rates, sea level, tectonic subsidence, duration of model and number of time steps. Graphical tools allow output visualization via strike-and dip-oriented stratigraphic cross sections, isopach (sediment thickness) maps, and structure maps. The locations and time-step intervals for these tools are user-defined. Preliminary comparisons of model output to an extensive seismic dataset from the East China Sea continental margin help validate the ability of fuzzy logic to recreate complex depositional conditions and advance further the idea that fuzzy logic is well suited not only for stratigraphic analysis but geological applications in general.