Automated Hydrocarbon Prospect Delineation Through Integrated Contour-Fault Analysis: Case Study from the Gulf of Thailand

ABSTRACT Traditional prospect delineation workflows require 3-5 days per prospect and introduce interpreter-dependent variability constraining scenario analysis. This research presents the first comprehensive trap-specific automated system for Gulf of Thailand structural traps, addressing three dominant configurations—ramp, closure, and nose traps—through specialized algorithms tailored to each trap type's geometric characteristics. The system employs half-angle bisector projection for ramp traps with automatic closing/opening classification, iterative hill-climbing with adaptive validation for closure traps, and curvature-minimum spanning tree analysis for nose traps. The algorithmic approach encodes geological knowledge explicitly, providing transparency essential for petroleum industry adoption without requiring extensive training data. Validation against expert interpretations demonstrates strong performance: three-way closures (1-7% area differences), closing ramps (4-6%), half grabens (6-8%), and nose traps (10-12%) meet industry accuracy requirements. Opening ramps show 34-62% overestimation due to geometric sensitivity when projections occur nearly parallel to contours, requiring continued refinement. The system achieves 4-times processing acceleration, enabling comprehensive probabilistic evaluation and rapid scenario analysis. Regional specialization demonstrates that basin-specific workflows achieve superior performance versus generic commercial software, with potential applicability to other extensional basins exhibiting similar structural configurations.