Adam Pidlisecky

SimPEG: An open source framework for simulation and gradient based parameter estimation in geophysical applications

Inverse modeling is a powerful tool for extracting information about the subsurface from geophysical data. Geophysical inverse problems are inherently multidisciplinary, requiring elements from the relevant physics, numerical simulation, and optimization, as well as knowledge of the geologic setting, and a comprehension of the interplay between all of these elements. The development and advancement of inversion methodologies can be enabled by a framework that supports experimentation, is flexible and extensible, and allows the knowledge generated to be captured and shared. The goal of this paper is to propose a framework that supports many different types of geophysical forward simulations and deterministic inverse problems. Additionally, we provide an open source implementation of this framework in Python called SimPEG (Simulation and Parameter Estimation in Geophysics, http://simpeg.xyz). Included in SimPEG are staggered grid, mimetic finite volume discretizations on a number of structured and semi-structured meshes, convex optimization programs, inversion routines, model parameterizations, useful utility codes, and interfaces to standard numerical solver packages. The framework and implementation are modular, allowing the user to explore, experiment with, and iterate over a variety of approaches to the inverse problem. SimPEG provides an extensible, documented, and well-tested framework for inverting many types of geophysical data and thereby helping to answer questions in geoscience applications. Throughout the paper we use a generic direct current resistivity problem to illustrate the framework and functionality of SimPEG.

Compensating for temperature variations in time-lapse electrical resistivity difference imaging

Variations in temperature during time-lapse electrical resistivity imaging (ERI) surveys introduce changes in electrical conductivity (EC). When the goal of the time-lapse ERI survey is to image changes in EC due to changes in saturation or pore water salinity, compensation must be made for the effect of temperature variations. A temperature-compensation method can approximate time-lapse ERI data with the effect of temperature variations removed. First uncompensated ERI data are inverted. The inversion model then is adjusted to a standard temperature image. Forward simulations are performed using the uncompensated inversion and the standard temperature equivalent model. The temperature-compensated simulated resistance data are subtracted from the uncompensated simulated resistance data, forming data correction terms. The data correction terms then are subtracted from the measured data to yield temperature-compensated data. Using the temperature-compensated data, inversions have been carried out on two syn...

Informed experimental design for electrical resistivity imaging

Electrical resistivity imaging has been successfully used to monitor near-surface hydrologic processes but use of standard measurement arrays may not provide the greatest data sensitivity to the imaged region. We present a method of experimental design based on the concept of informed imaging for creating an electrical resistivity imaging experiment to monitor flow beneath a recharge pond. Informed imaging is the integration of all available data about a site into the acquisition, inversion and interpretation of electrical resistivity data. Informed experimental design uses all available information to develop an a priori model of the subsurface conductivity structure that guides the selection of measurement arrays for an electrical resistivity imaging experiment given spatial and temporal constraints on the acquisition. Selection of arrays focuses on maximizing the amount of unique information acquired with each source pair. We apply the method to the selection of arrays for imaging the top 5 m of the subsurface beneath a recharge pond in Northern California, which is part of an aquifer storage and recovery project. Decreasing infiltration rates over time reduce the effectiveness of the recharge pond. We seek to monitor infiltration processes at the contact between a fines-rich sand layer and coarser sand layer in an effort to understand the hydrologic controls on infiltration. The performance of the arrays selected using informed experimental design relative to two standard arrays (Wenner and dipole-dipole) is validated on two synthetic subsurface conductivity models, which are representative of conductivity structures that may arise during an infiltration event. Performance is evaluated in terms of a singular value decomposition of the sensitivity matrix produced by the three types of arrays, as well as a measure of the region of investigation. Results demonstrate that arrays selected using informed experimental design provide independent information about the imaged region and are robust in the presence of noise, improving the ability to image changes in a conductivity structure that result from infiltration processes.

Inversion of time-lapse electrical resistivity imaging data for monitoring infiltration

Monitoring hydrologic processes in the vadose zone is of great importance in developing management strategies for groundwater resources. Flow and transport behavior in variably saturated media control the rate at which, and path along which, fluids or contaminants reach groundwater aquifers. Electrical resistivity imaging (ERI) is sensitive to changes in the subsurface conductivity structure, which is highly dependent on the water saturation distribution in the vadose zone. Time-lapse ERI, therefore, has been widely used for environmental and hydrologic monitoring experiments.

Hydrogeologic structure underlying a recharge pond delineated with shear-wave seismic reflection and cone penetrometer data

With the goal of improving the understanding of the subsurface structure beneath the Harkins Slough recharge pond in Pajaro Valley, California, USA, we have undertaken a multimodal approach to develop a robust velocity model to yield an accurate seismic reflection section. Our shear-wave reflection section helps us identify and map an important and previously unknown flow barrier at depth; it also helps us map other relevant structure within the surficial aquifer. Development of an accurate velocity model is essential for depth conversion and interpretation of the reflection section. We incorporate information provided by shear-wave seismic methods along with cone penetrometer testing and seismic cone penetrometer testing measurements. One velocity model is based on reflected and refracted arrivals and provides reliable velocity estimates for the full depth range of interest when anchored on interface depths determined from cone data and borehole drillers’ logs. A second velocity model is based on seismic cone penetrometer testing data that provide higher-resolution 1D velocity columns with error estimates within the depth range of the cone penetrometer testing. Comparison of the reflection/refraction model with the seismic cone penetrometer testing model also suggests that the mass of the cone truck can influence velocity with the equivalent effect of approximately one metre of extra overburden stress. Together, these velocity models and the depth-converted reflection section result in a better constrained hydrologic model of the subsurface and illustrate the pivotal role that cone data can provide in the reflection processing workflow.

Characterizing Heterogeneity in Infiltration Rates During Managed Aquifer Recharge.

Infiltration rate is the key parameter that describes how water moves from the surface into a groundwater aquifer during managed aquifer recharge (MAR). Characterization of infiltration rate heterogeneity in space and time is valuable information for MAR system operation. In this study, we utilized fiber optic distributed temperature sensing (FO-DTS) observations and the phase shift of the diurnal temperature signal between two vertically co-located fiber optic cables to characterize infiltration rate spatially and temporally in a MAR basin. The FO-DTS measurements revealed spatial heterogeneity of infiltration rate: approximately 78% of the recharge water infiltrated through 50% of the pond bottom on average. We also introduced a metric for quantifying how the infiltration rate in a recharge pond changes over time, which enables FO-DTS to be used as a method for monitoring MAR and informing maintenance decisions. By monitoring this metric, we found high-spatial variability in how rapidly infiltration rate changed during the test period. We attributed this variability to biological pore clogging and found a relationship between high initial infiltration rate and the most rapid pore clogging. We found a strong relationship (R2  = 0.8) between observed maximum infiltration rates and electrical resistivity measurements from electrical resistivity tomography data taken in the same basin when dry. This result shows that the combined acquisition of DTS and ERT data can improve the design and operation of a MAR pond significantly by providing the critical information needed about spatial variability in parameters controlling infiltration rates.

Structurally constrained impedance inversion

AbstractIn complex geology, the presence of highly dipping structures can complicate impedance inversion. We have developed a structurally constrained inversion in which a computationally well-behaved objective function is minimized subject to structural constraints. This approach allows the objective function to incorporate structural orientation in the form of dips into our inversion algorithm. Our method involves a multitrace impedance inversion and a rotation of an orthogonal system of derivative operators. Local dips used to constrain the derivative operators were estimated from migrated seismic data. In addition to imposing structural constraints on the inversion model, this algorithm allows for the inclusion of a priori knowledge from boreholes. We investigated this algorithm on a complex synthetic 2D model as well as a seismic field data set. We compared the result obtained with this approach with the results from single trace-based inversion and laterally constrained inversion. The inversion carr...

Borehole vibration response to hydraulic fracture pressure

Summary Passive-seismic recordings in cased wellbores, including data acquired to monitor hydraulic fracture treatments, can be used to detect resonant vibrations of the casing. The resonant vibrations are linked to the geometry of geophone arrays, which are clamped to the borehole casing with sufficient force to affect the fundamental casing vibration frequency. Using field examples from western Canada, we document temporal variations of vibration frequency that occur over the course of a multistage hydraulic fracture treatment. Finite-element simulations suggest that vibrational frequency is sensitive to local stress conditions. We propose that this effect can be used for in situ monitoring of stress variations.

Chapter 4: Visible Geology: Creative Online Tools for Teaching, Learning, and Communicating Geologic Concepts

Abstract Computer-based learning tools are becoming more prevalent in classrooms from elementary school to higher education. The potential value of interactive learning tools is particularly high in geoscience education. Students can benefit from interactive tools that allow them to explore different processes in one-dimensional (1-D), two-dimensional (2-D), and three-dimensional (3-D) space. Traditionally, geoscience education has relied on laboratory exercises to provide students with the opportunity to explore dimensionality. In this chapter, we introduce Visible Geology, an innovative web-based application designed for geoscience education. Visible Geology enables visualization of geologic structures and processes through the use of interactive 3-D models. As Visible Geology has been designed from a student-centric perspective, it has resulted in a simple and intuitive interface, allowing students to creatively explore concepts. We present a case study of a large first year class at the University of British Columbia, and show the utility of Visible Geology in teaching geoscience concepts of relative dating and cross-cutting relationships. The ease of use of the software for this assignment, including automatic grading, made this tool practical for deployment in classrooms of any size. The outcome of this type of large-scale deployment is that students, who would normally not experience a lab exercise, gain exposure to 3-D thinking. The level of ownership and interactivity inherent in Visible Geology encourages engagement, leading learners to practice visualization and interpretation skills and discover geologic relationships.

Monitoring managed aquifer recharge with electrical resistivity probes

The use of managed aquifer recharge (MAR) to supplement groundwater resources can mitigate the risks to an aquifer in overdraft. However, limited information on subsurface properties and processes that control groundwater flow may lead to low levels of recapture of infiltrated water, reducing the efficacy of MAR operations. We used long 1D electrical resistivity probes to monitor the subsurface response over one diversion season at five locations beneath an operating recharge pond in northern California. The experiment demonstrated the benefits of integrating geophysical and standard hydrologic measurements. The water table response interpreted from time-lapse electrical resistivity images was in good agreement with traditional pore-pressure transducer measurements at coincident locations. Moreover, the electrical resistivity measurements were able to identify vertical variations in water saturation that would not have appeared in pore-pressure data alone. Changes in saturation estimated from electrical resistivity models indicated large hydraulic gradients at early time and suggested the presence of highly permeable conduits and baffles between the surface and the screened interval of recovery wells. The interpreted structure of these conduits and baffles would contribute to the movement of a large amount of infiltrated water beyond the capture zone of recovery wells before pumping begins, accounting in part for the low recovery rates.