Are Static 2-D Measures of Connectivity a Good Proxy for 3-D Dynamic Connectivity in Deepwater Slope Channels?
Accurately capturing flow path connectivity in reservoir models is critical to predicting fluid flow performance. In deep-water slope channel reservoirs, flow path connectivity is controlled by multi-scale stratigraphy including a combination of internal channel architecture and the stacking patterns of channels into channel complexes. Although flow connectivity is inherently three-dimensional, measurements in the subsurface are sparse (wireline logs and core) or low resolution (seismic reflectivity) providing limited information for reliable prediction. Outcrop analog studies are often used to fill this gap by attempting to quantify static connectivity (e.g. percentage of sand on sand contacts) or as the basis for two-dimensional flow simulations. However, outcrop exposures are inherently two-dimensional and the link between two-dimensional static connectivity measures and three-dimensional flow connectivity is poorly understood. In this study we use a three-dimensional outcrop model of deep-water slope channels from the Cretaceous Tres Pasos Formation in the Magallanes Basin of Chile to provide insights into the link between available subsurface data from wells and our ability to accurately predict reservoir performance. We analyze two-dimensional static connectivity along the outcrop, varying channel width, internal architecture and drape coverage. These static connectivity metrics are compared to three-dimensional dynamic connectivity using flow simulation. We evaluate sweep efficiency to elucidate whether static two-dimensional connectivity measures from outcrop are good proxies for three-dimensional dynamic connectivity. Finally, we tie the three-dimensional connectivity results back to statistics from measured sections to test the ability of one-dimensional statistics to help predict the three-dimensional connectivity and answer the question of how many wells would be required to generate an accurate prediction. Our results show that, as expected, the static connectivity measures are weakly correlated with sweep efficiency and are strong affected by fine scale flow barriers, such as percentage of drape coverage. Furthermore, an increase number of wells improve the probability of an accurate static connectivity prediction.
AAPG Datapages/Search and Discovery Article #90189 © 2014 AAPG Annual Convention and Exhibition, Houston, Texas, USA, April 6–9, 2014