Seismic facies analyses as aid in regional gas hydrate assessments. Part-II: Prediction of reservoir properties, gas hydrate petroleum system analysis, and Monte Carlo simulation

Abstract

A new strategy to assess occurrences of marine gas hydrate has been implemented for the study area of the Ulleung Basin, East Sea, where two comprehensive gas hydrate drilling programs completed logging and coring at 18 sites. In this study we introduce multi-attribute analyses using five drill sites along a regional 2D seismic line to predict physical properties of P-wave velocity, density, and porosity required for gas hydrate saturation calculations. The five well-sites allow a reasonable cross-validation of the final multi-attribute-based prediction and to derive statistical parameters used in a Monte Carlo simulation of total gas hydrate volume in the study area. A similar analysis was completed for a 3D seismic volume around Site UBGH1-4. The crucial boundaries of the top of gas hydrate occurrence zone (TGHOZ) and base of the gas hydrate stability zone (BGHSZ) that are part of the gas hydrate petroleum system were calculated using constraints on geothermal gradient, pore-fluid salinity, type of hydrocarbon gases present, sedimentation rates, and total organic carbon content based on results from the two drilling expeditions carried out in the Ulleung Basin as well as other regional studies. Gas hydrate saturations were determined following the effective medium theory and using the sedimentological data obtained at the drill sites to define required mineral compositions. Seismic facies classification allowed the definition of occurrence of certain sediment types along the seismic data used in this study, and physical properties from the multi-attribute analyses and mineral compositions were assigned to each facies class. Statistical parameters of mean and standard deviation were estimated for each input parameter (P-wave velocity, bulk density, mineral compositions, grain density, density-porosity, TGHOZ, BGHSZ) in the gas hydrate saturation calculations based on individual uncertainties in the measurements. In the final step, a Monte Carlo simulation is used to calculate the total amount of gas hydrate (in m3) present in the study region for the two example data set of a 2D seismic line and one 3D seismic volume. The Monte Carlo simulation draws on the various statistical distributions of the input parameters and we have implemented 150,000 simulation runs in this study. The final histogram distributions of total volume of gas hydrate (in m3) allow determination of the mean, median, and mode values, as well as the 95% and 5% probability thresholds for gas hydrate volumes being present in the two study regions, though they are not meant as estimates of possible commercial amounts of gas associated with gas hydrates in the Ulleung Basin.