Title: Draugen - 123 4D - an integrated perspective from acquisition processing and reservoir geophysics Authors: M. Zwaan, K.L. Wong, F. Smit, J. Brain (Shell U.K. Limited), G-B. Strøm , P. Søiland, B. Nystrand, K. Guderian (A/S Norske Shell) In 2004 the third 4D monitor survey over the Draugen field has been acquired and processed. This third monitor survey is a high- resolution survey which will be used to address two main objectives: - the 4D compatible version will be used for fluid flow monitoring and reservoir management optimization; - the high-resolution version will be used to firm up high side potentials through improved lithology differentiation (sand versus shale) and fluid type determination. The main Draugen reservoir is the Rogn, which is an upper Jurassic shore-face sandstone. In the West of the field the Rogn on-laps onto a second upper Jurassic reservoir, the Garn, which is partly in communication with the Rogn through sand to sand juxtaposition across faults. The 4D was designed to firm-up and de-risk infill drilling projects for both the Rogn and Garn formation by assessing drained and un- swept zones. In the remainder of this paper the discussion is restricted to the results of the 4D processing, as the interpretation of the high-resolution processing and inversion are currently not finalised. The 4D processing started with a fast-track processing to identify the behaviour of the waterfront and to identify any potential crucial reservoir management issues already very early in the data-cycle. A comparison with synthetics (seismic forward modeling based on the parameters of the dynamic reservoir model) showed that the waterfront-flow on the synthetic and recorded data is very similar, hence confirming the current water-injection and production strategy. The previous 4D survey in 2001, however, showed significant differences as compared to the synthetic data from the reservoir model, clearly indicating an unwanted imbalanced waterfront. Both times the value of the fast-track processing was apparent: an early indication of to which degree the reservoir model reflects the true dynamic nature of the field and to enable timely reservoir management adjustments if necessary. The 4D processing has continued to produce both a dedicated 4D baseline and a set of monitor surveys. For each vintage, four angle sub-stacks have been generated. The results of the synthetic modeling from log-data, were used to determine the optimum angle ranges of these four different sub-stacks. Elastic impedance inversions on