This article, written by Technology Editor Dennis Denney, contains highlights of paper SPE 110272, "Integrating Pressure- Transient-Test Data With Seismic-Attribute Analysis To Characterize an Offshore Fluvial Reservoir," by Akshay Sahni, SPE, Ken Kelsch, SPE, Hathaiporn Samorn, and Chalatpon Boonmeelapprasert, SPE, Chevron, prepared for the 2007 SPE Asia Pacific Oil & Gas Conference and Exhibition, Jakarta, 30 October-1 November. Interpreting pressure-transient tests in complex faulted and stratigraphic environments can be difficult. In fluvial-depositional environments, in which sand continuity is a significant uncertainty, pressure-transient-test interpretation can generate several nonunique solutions, all of which may match test data. Use of seismic-attribute analysis to constrain pressure-transient-test interpretation leads to a better understanding of reservoir heterogeneities and boundaries. Introduction The integration of geology with well-test interpretation has been discussed in industry literature, as noted in the full-length paper. It has been stated that integration of geoscience and well testing reduces uncertainty in reservoir description, especially in fluvial reservoirs. This paper builds on work done in the past, use of a data set comprising seismic-attribute analysis and a long-duration pressure-buildup test, along with core, well-log, and production data. Reservoir Reservoir sands in the area of interest primarily are Tertiary sequences, predominantly of nonmarine origin, and deposited in fluvial to lacustrine environments. The net-to-gross ratio over the entire reservoir interval (more than 3,000 ft) typically is between 0.15 and 0.20. At this low ratio, the vertical connectivity of sands generally is very poor. Areas with a greater concentration of high-amplitude seismic events likely are associated with a high probability of better net-to-gross-ratio development (more sand prone), but not directly related to better pay (hydrocarbon-bearing sands). It can be difficult to distinguish better sand quality from pay sand. Seismic modeling of sands indicates that gas sands yield higher seismic-reflection amplitudes, but only 5 to 10% gas saturation is necessary to change the acoustic impedance significantly. Reservoir-compartment size of the hydrocarbon-bearing sands identified in wells can be estimated by use of the seismic inlines, crosslines, and time slices. For the reservoir area under consideration, time-slice mapping indicates that the average reservoir compartment size is relatively small and is supported in most cases by well-test interpretation, production, and pressure data. There is a high level of faulting, subdividing the channel systems, and thus reducing the connected-sand area to approximately 30 to 50 acres. Larger connected areas (250 acres or more) are found on structural shoulders or outer fault blocks of a fault complex. The sand area/volume connected to a well remains one of the largest uncertainties in estimating reserves.