|dc.description.abstract||Hydrocarbon recovery from unconventional reservoirs is driven by the presence of natural and induced fractures in the reservoir. While estimation of these fractures using single component seismic data is well practiced, the process may be ambiguous. Nine-component (9-C), 3-D seismic acquisition was developed with an interpretive emphasis on improving fracture network mapping in these unconventional plays. This research explores the application of pre-stack multicomponent interpretation for fracture identification. Prior to interpretation, multicomponent acquisition requires rotation of the horizontal components such that the wave modes are separated into meaningful data sets. I discuss the implications of processing 9-C data in different coordinate systems and demonstrate the effectiveness of the radial-transverse system regardless of anisotropic conditions. A technique of 9-C fracture interpretation takes advantage of the idea that in the presence of horizontal traverse isotropy (HTI), waves show variation in velocity depending on the source-receiver azimuth (VVAz). HTI is assumed for modeling presented in this research as it best describes a reservoir with a single, dominant vertical fracture set. Conventional post-stack techniques attempt to map these velocity variations using shear wave splitting (SWS) calculations, however, the most insightful observations are in the details of pre-stack multicomponent data. I show the advantages of common-offset, common-azimuth (COCA) gathers and limited azimuth stacks (LAS) for shear wave splitting analysis on pre-stack 9-C data. All nine components uniquely expose HTI conditions in the reservoir. VVAz is observed at reflectors below the HTI interval since waves must traverse the fractures in order to polarize, and accumulate a travel time delay. While the VVAz signature is unique to each component, the observations are complementary due to the orthogonality of particle motion of all wave modes. Assuming an HTI medium, along fracture strike, the P-wave and SV-wave velocities are fast and the SH-wave velocity is slow. The cross-term components show no energy parallel or perpendicular to fracture strike. This research demonstrates the added value of shear wave components based on the increased sensitivity of their VVAz response at all offsets and for thinner HTI layers. For this anisotropic feasibility study, the Niobrara stratigraphy of the Wattenberg field is modeled with varying extents of the fractured reservoir. The largest VVAz exposed on the noise-free model with the entire reservoir fractured shows 2.5s of splitting on the pure shear wave components and 1.8s on the converted wave components. Where the fractured interval thins to less than 25m, the P-wave velocity anisotropy is visually indistinguishable on COCA gathers. Based on modeling, field data is expected to show little HTI-related VVAz response on all components. The Reservoir Characterization Project (RCP), Colorado School of Mines has explored the added interpretive value of 9-C data on unconventional reservoir development. Often, the nine data sets are interpreted separately even though all components are responding to the same subsurface conditions. This research evaluates the converted wave and shear wave inversion techniques used to solve for fracture azimuth and shear wave splitting and introduces a joint converted-shear wave inversion for improved fracture azimuth detection. In recent years, the RCP has focused on the unconventional Niobrara-Codell reservoir within a 1-square mile section of the Wattenberg Field (Wishbone Section). The Niobrara-Codell formations are a typical unconventional petroleum system in which the efficiency of natural and induced fracture networks critically influence production in the low permeability, low porosity reservoir. The hypotheses and conclusions of this research are evidenced by both synthetic data and field data. Synthetic examples are based on simple HTI and Niobrara models and the simultaneous interpretation of the RCP multicomponent field survey over the Wishbone Section emphasizes the implications of noise, acquisition and processing on multicomponent data.