This article, written by Special Publications Editor Adam Wilson, contains highlights of paper IPTC 18234, “Constraining Reservoir Models by Integrating Data Across Disciplines—A Case Study of a Thin-Bedded Turbidite Field in the North Sea,” by Manish K. Choudhary, SPE, Amit Keshri, SPE, and Kamalraj Mohan, SPE, Shell Technology Centre Bangalore, prepared for the 2014 International Petroleum Technology Conference, Kuala Lumpur, 10–12 December. The paper has not been peer reviewed. Field X in the North Sea was discovered in the 1980s, and large subsurface uncertainties exist regarding reservoir extent, fluid contacts, and reservoir properties. An integrated study was conducted to define subsurface models that honored the available data set and captured the uncertainty range. These subsurface models then were used in a dynamic realm to understand the effect of dynamic uncertainties such as aquifer strength, gas-cap size, and relative permeability, with an overall objective of generating low-, middle-, and highcase production forecasts. Available Data Two wells have been drilled in Field X, with Well X1 being the discovery well and Well X2 being the appraisal well (Fig. 1). The two wells confirmed the presence of an oil rim, approximately 200 ft thick, overlain by a primary gas cap. The hydrocarbon-bearing Tay formation was composed of thin-bedded sands, 3–5 ft thick. Data acquired from the wells included basic lithologs, reservoir pressures, fluid samples, and core data. Both wells had been production tested, which helped in understanding the producibility of these thin bedded zones. Seismic data in this area were processed and were also available for the feasibility study. Regionally, the Eocene Tay formation is deposited in a structurally diverse basin, influenced by salt and gravity tectonics during the Late Jurassic and Early Cretaceous. The Tay formation can be subdivided into three major subunits— Upper Tay, Middle Tay, and Lower Tay. Modeling Uncertainties A detailed interpretation of the seismic data was carried out during the study, which helped in mapping the top of the Tay formation, the top of the underlying Balder formation, and the top of the Chalk formation. A number of seismic attributes were extracted at the top of Upper Tay, but only the seismic amplitudes and spectral decomposition were useful in distinguishing the channel or fan sands from the background sediments. The seismic resolution was not good enough to distinguish Upper Tay from Middle Tay because of the low netto- gross ratio (NTG) of the formation and a change in the fluid fill. In the absence of a seismic mapped surface, no attributes could be extracted separately for the Middle Tay reservoir. Depositional Model. The integration of seismic-amplitude maps along with well-log data gave some insights into the reservoir-sand geometries but did not establish a single depositional model conclusively. The amplitude map suggested that the deposits could be part of either deep marine feeder channels or turbidite fans. This uncertainty was because of the dimming of seismic amplitudes at the crest of the structure, suggesting the possibility of a lower NTG area. Multiple realizations of sand geometries were constructed with a mix of channels and lobe geometries. These models were different from one another but provided an exhaustive set of realizations.