Surface Deformation Mapping and Automatic Feature Detection Over the Permian Basin Using InSAR

Download or read book Surface Deformation Mapping and Automatic Feature Detection Over the Permian Basin Using InSAR written by Scott Staniewicz and published by . This book was released on 2022 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: The Permian Basin has become the United States' largest producer of oil and gas over the past decade. During the same time, it has experienced a sharp rise in the number of induced earthquakes. In order to better understand the damage potential from induced earthquakes, new data and monitoring approaches are critically needed. Interferometric Synthetic Aperture Radar (InSAR) is a remote sensing technique that measures surface deformation over broad areas with 10s-100s meter spatial resolution and up to millimeter-to-centimeter accuracy. These measurements can be used to derive information about Earth’s subsurface and assess induced seismic risks. However, it is difficult to perform basin-scale surface deformation mapping and automatic feature detection using InSAR because the signal-to-noise ratio (SNR) of the deformation signals compared to tropospheric noise is extremely low. It is common to assume that the Permian Basin is rigid enough that the subtle deformation associated with oil and gas production and wastewater injection are not detectable by InSAR. In this dissertation, we develop methods for characterizing tropospheric noise and its power spectral density directly from InSAR observations. We show that the tropospheric noise distribution is non-Gaussian, and a small portion of SAR scenes are corrupted by up to ±15 cm noise outliers associated with storms and heat waves. This finding is significant because most of the InSAR time series solutions are optimal only when noise follows a Gaussian distribution. We design robust and scalable time series algorithms to reconstruct the temporal evolution of surface deformation in this challenging scenario, and we achieved basin-wide millimeter-level accuracy based on independent GPS validation. We observe numerous subsidence and uplift features near active production and disposal wells, as well as linear deformation patterns associated with fault activities near clusters of induced earthquakes. Furthermore, we designed a new computer vision algorithm for detecting the size and location of unknown deformation features in large volumes of InSAR data. We are able to determine whether a detected feature is associated with tropospheric artifacts or real deformation signals based on a realistic tropospheric noise model derived from InSAR data