Deep carbonate reservoir characterization using multi seismic attributes:A comparison of unsupervised machine learning approaches

Abstract

Seismic reservoir characterization is of great interest for sweet spot identification, reservoir quality assessment, and geological model building. The sparsity of the labeled samples often limit the application of supervised machine learning for seismic reservoir characterization. Unsupervised learning methods, on the other hand, explore the internal structure of data and extract low-dimensional features of geologic interest from seismic data without the need for labels. We compare various unsupervised learning approaches, including the linear method PCA, manifold learning methods T-distributed Stochastic Neighbor Embedding (t-SNE) and Uniform Manifold Approximation and Projection (UMAP), and the Convolutional Autoencoder (CAE), on both the 3D synthetic and field seismic data of a deep carbonate reservoir, SW China. On the synthetic data, the low-dimensional features extracted by UMAP and CAE provide a better indication of porosity and gas saturation than traditional seismic attributes. In particular, UMAP better preserves the global structure of geological features, and show the potential of decoupling the gas saturation and porosity effects from seismic responses. We demonstrate that by joint use of multi type of seismic attributes instead of using single type of seismic attributes can better delineate the reservoir structures using unsupervised ML. On the field seismic data, UMAP can effectively characterize sedimentary facies distribution, which is consistent with the geological understanding. Nevertheless, the porosity and saturation can not be reliably identified from field seismic data using unsupervised ML, which is likely to be caused by the complex pore structures in carbonates complicating the mapping relationship between seismic responses and reservoir parameters.