Vorwata is a giant gas field located in Papua Barat, Indonesia. Initial production through the two LNG trains began in June 2009. There is a large uncertainty in total gas initially in place (GIIP) in the field, primarily due to uncertainties in seismic time-to-depth conversion; the upside static model has 85% more gas than the downside model. In addition to a large uncertainty in GIIP, seismic interpretation has revealed extensive faulting with potential compartmentalization. The total GIIP and the GIIP connected to the existing 14-well development are the two primary factors in the determination of the timing and extent of the next phase of development. Reducing the GIIP uncertainty range is of critical importance during the early production period.
Analysis of a pre-production interference test and early production real-time downhole pressure data confirmed pressure communication exists within the currently developed area. However, well shut-in pressure trends began diverging within a few months of startup, indicating the potential for significant variation in reservoir connectivity, and potentially, connected GIIP.
Conventional material balance GIIP estimation methods do not provide reliable results in Vorwata at this stage of depletion due to several factors, including: small depletion, clustering of the current development in the center of the field, large variation in shut-in well pressures, potential compartmentalization, existence of baffled regions, significant interwell pressure interference, variable daily individual well rates and lack of field-wide shut-ins. A computer-assisted history matching technique was developed to assess connected GIIP uncertainty using a BP in-house uncertainty modeling toolkit ("TDRMTM, Top Down Reservoir Modelling"), Landmark Graphics’ Nexus® simulator and a simplified reservoir model. The TDRM/Nexus modeling technique matches simulated well pressures to measured well downhole shut-in pressures. Uncertainty variables used for history matching are: regional GIIP, regional permeability and fault transmissibilities for approximately 30 fault segments. The trend in the connected GIIP uncertainty range versus cumulative produced gas is updated on a regular basis. The methodology integrates static and dynamic data and models to reduce GIIP uncertainty. An additional benefit of the approach is that the history matched fault transmissibilities are used to constrain the fullfield model for depletion planning purposes. The methodology is a significant improvement compared to conventional P/Z analysis, especially for gas fields lacking field-wide static reservoir measurements. The downhole pressure sensors provide critical surveillance data for GIIP uncertainty reduction and fault characterization.
TDRM/Nexus uncertainty modeling has reduced the Vorwata field P90-P10 connected GIIP uncertainty range by over a factor of 5, from 14.9 tscf (Jan 2010) to 2.5 tscf (March 2011). This reduction in GIIP uncertainty combined with the most likely Vorwata GIIP estimate is sufficient to begin planning for a Tangguh expansion project.