Northern Gulf of Thailand: Differences in the Khmer and Pattani Basins
Kelly Dempster, Jeffrey R. Johnson, and Fang Lin
Chevron Energy Technology Company, Houston, Texas, USA
The choice of an analog for petroleum systems analysis usually involves assumptions regarding regional tectonics, paleoclimate, environment of deposition, style of structural deformation, stratigraphic architecture, and source rock characteristics. Experience in modeling the analog can also bring insight into appropriate techniques, including the required level of stratigraphic and faulting detail, and choice of modeling software. Block B8/32, in the offshore Thailand Pattani Basin, served as a training ground with abundant well control, production data, and 3D seismic to learn the types of geochemical analysis and level of modeling detail required to understand the sparsely-explored Khmer Basin Block A in offshore Cambodia. The understanding of the Pattani petroleum system and the modeling workflow were applied directly to the Khmer Basin and contributed to the selection of drilling locations in Block A. This work examines how results from the drilling campaign challenged our initial assumptions of source rock distribution and migration patterns, and necessitated changes in our understanding of regional heat flow, the role of faulting in lateral migration, and the causes of overpressure on hydrocarbon migration and accumulation.
In the Pattani, construction of the basin models required an increase in the number of defined intervals and introduction of multiple carrier beds to represent interconnected sand bodies. This synthetic stratigraphy was able to reproduce the discovery well’s accumulations, and calibration points in other fields. Backstripping and comparison of interval thickness were used to estimate periods of structural movement, and controlled migration along faults. Maturity trends derived from vitrinite data were consistent with other geochemical indicators from oil and gas samples. Geochemical analysis supported an Oligocene lacustrine source rock, interpreted as three distinct ‘lakes’ with differing source rock properties. Geochemical data also indicated thermal cracking in the central region. Resulting basin models identified oil- and gas- prone areas as well as migration focus, and were used to support the location of new platforms and acquisition of new exploration concessions.
In the Khmer, 2D basin models were constructed using similar techniques along profiles tying available well control. Preliminary geochemistry at well ties helped constrain the stratigraphic position of the undrilled source rocks. Potential lower source rock distribution was inferred from seismic character and upper source facies were defined as basinward of well penetrations. The newly acquired 3D seismic interpretation was used to construct a 3D basin model for thermal calibration and to characterize timing of maturation and migration. Migration analysis was further refined with high-resolution map-based flow path modeling to reflect faulting patterns. These detailed migration/accumulation maps were used to support drilling locations.
There were many unanticipated results. Oil shows were widely distributed areally, but limited to syn-rift and earliest post-rift section; only minor gas was encountered. Except at one location, vertical migration did not reach the sand-prone post-rift reservoir units. Most new directional wells demonstrated much higher than predicted bottom hole temperatures, resulting in questions regarding convective flow and resulted in changes to the basal heat flow map. The sand content was similar to the Pattani’s, but varied in the pattern of distribution across the block. Geochemical analysis provided indicators of two potential source zones, and supported the concept of high volume water flow in certain fault blocks. Maturity equivalence of the oils provided evidence for potential migration paths. Seal rock analysis indicated variable column height potential, but also a zone of early paleosol development, which helped explain overpressure formation and constraints on vertical hydrocarbon distribution. Patterns of overpressure suggested a baffled system consistent with petrographic analysis that indicated thermally-controlled mineralization in faults.
The various lines of evidence produced an integrated view of the petroleum system, and insight into key aspects of timing. The potential for hydrocarbons comes from two source rocks and with two periods of peak generation in the early- to mid-Miocene. Early generation preceded post-rift faulting responsible for most trapping geometries, and may have been lost to up-dip migration, while the second phase was partially captured in basin-flanking fault systems. Both the sand content and temperature vary from west to east, causing differences in the timing of cementation relative to generation, and controlled accumulation and vertical migration of hydrocarbons along faults into sand-prone units. Temperature, pressure, and geochemical evidence suggests the development of hydrothermal, and probably hydrocarbon, flow parallel to fault strike. Look-back experiments based on observed hydrocarbons provide constraints on the likely distribution of source rocks and extent of hydrocarbon migration. The original prediction of source distribution in Khmer held up to analysis, but the Pattani analog for vertical migration along faults was problematic. Uncertainty persists about the relative roles of hydrocarbon abundance, fault cementation, and remobilization of oil due to late stage gas generation on migration.
This study has led to a new level of understanding of aspects of mainly lateral petroleum systems, including the timing of migration relative to cementation in faults, the formation of anomalous thermal and pressure conditions, and traditional views of up-dip migration. It has also reinforced the idea that choosing a geologic analog should require modeling the analog first in order to set the appropriate level of detail and dimensionality of models.
AAPG Search and Discovery Article #120098©2013 AAPG Hedberg Conference Petroleum Systems: Modeling the Past, Planning the Future, Nice, France, October 1-5, 2012