A Characterization of 3D Karst Networks Through Fractal Concept Conference Paper uri icon

abstract

  • Abstract Carbonate reservoirs in Malaysia, particularly within the Luconia province in Sarawak, represent major hydrocarbon resources and potential CO2 storage (CCS) sites, with consideration for karst-related risks and uncertainties. The middle to upper Miocene age of these carbonate reservoirs in the JTN field is influenced by multiple diagenetic phases and varying degrees of karstification. Karst networks are characterized by their complex and heterogeneous distribution in the carbonate depositional sequence, geometry or shape, dimension, and possible karst fills, presenting significant carbonate reservoir uncertainty. This study investigates the application of fractal concepts to the rescaled 3D seismic data in the 3D grid domain for the geometry or shape of karst networks characterization, which includes feature recognition such as collapse shape within chaotic seismic reflectors or sinkholes. The fractal dimension is used to analyze the complexity and the connections between these features within the area of interest, especially related to geobody recognition for karst geometry or shape and spatial pattern. The methodology involves a detailed workflow of resampling 3D seismic cubes in 3D grid domain to capture detailed seismic characterization of the karst network and provide flexibility to conduct the karst geometry assessment in multiple grid cell sizes depending on the complexity of karst networks. The seismic data is analyzed to create a pattern recognition of the karst features within 100m x100m lateral resolution. Subsequently, fractal analysis to these lateral dimensions is applied to assess karst geometric complexity and spatial patterns. This includes the introduction of multiple region-based areas instead of one-go full-field analysis to achieve optimum estimation of the fractal dimension, which provides insights into the scaling and similarity pattern recognition of the karst geobody. The fractal dimension varies across different parts of the study area, reflecting the degree of variations in the distribution, density, and geometry of karst features. Areas with higher fractal distribution correspond to regions of greater karst activity (western and northern part of the area), suggesting more intricate karst networks. Conversely, regions with lower fractal dimensions exhibit simpler karst features, especially in the central region of the study area. The application of fractal concepts to 3D seismic data provides a more robust understanding of karst systems compared to conventional methodology. For instance, a better understanding of the fractal nature of karst networks can enhance predictions of karst networks for dynamic fluid flow analysis. This study demonstrates the benefits of integrating fractal geometry understanding with multiple cut-off 3D seismic cubes for karst network recognition within the advanced 3D grid multiple cell-size application. It highlights the potential of fractal analysis to uncover hidden geometry patterns below the base-case 100m × 100m lateral resolution and improve our ability to model and manage 3D karst networks as input for 3D static and dynamic modeling as a dual property modeling approach. The findings have significant implications for both research and practical applications in 3D reservoir modeling, providing a robust framework for future studies aimed at exploring the fractal nature of other complex geological systems in managing the uncertainty of hydrocarbon optimization and CO2 storage or containment assessment.

authors

  • Trianto, A.
  • Jalil, M. A.
  • Tewari, R. D.
  • Sedaralit, M. F.
  • Hendraningrat, L.
  • Sagar, S. F.
  • Abdul Latiff, Abdul Halim
  • Grisel, Jimenez

publication date

  • 2024

start page

  • D031S113R001