GANSim-3D for conditional geomodelling: theory and field application

Geomodelling of subsurface reservoirs is important for water resources, hydrocarbon exploitation, and Carbon Capture and Storage (CCS). Traditional geostatistics-based approaches cannot abstract complex geological patterns and are thus not able to simulate very realistic earth models. We present a Generative Adversarial Networks (GANs)-based 3D reservoir simulation framework, GANSim-3D, which can capture geological patterns and relationships between various conditioning data and earth models and is thus able to directly simulate multiple 3D realistic and conditional earth models of arbitrary sizes from given conditioning data. In GANSim-3D, the generator, designed to only include 3D convolutional layers, takes various 3D conditioning data and 3D random latent cubes (composed of random numbers) as inputs and produces a 3D earth model. Two types of losses, the original GANs loss and condition-based loss, are designed to train the generator progressively from shallow to deep layers to learn the geological patterns and relationships from coarse to fine resolutions. Conditioning data can include 3D sparse well facies data, 3D low-resolution probability maps, and global features like facies proportion, channel width, etc. Once trained on a training dataset where each training sample is a 3D cube of a small fixed size, the generator can be used for geomodelling of 3D reservoirs of large arbitrary sizes by directly extending the sizes of all inputs and the output of the generator proportionally. To illustrate how GANSim-3D is used for field geomodelling and also to verify GANSim-3D, a field karst cave reservoir in Tahe area of China is used as an example. The 3D well facies data and 3D probability map of caves obtained from geophysical interpretation are used as conditioning data. First, we create a training dataset consisting of facies models of 64×64×64 cells with a process-mimicking simulation method to integrate field geological patterns. The training well facies data and the training probability map data are produced from the training facies models. Then, the 3D generator is successfully trained and evaluated in two synthetic cases with various metrics. Next, we apply the pretrained generator for conditional geomodelling of two field cave reservoirs of Tahe area. The first reservoir is 800m×800m×64m and is divided into 64×64×64 cells, while the second is 4200m×3200m×96m and is divided into 336×256×96 cells. We fix the input well facies data and cave probability maps and randomly change the input latent cubes to allow the generator to produce multiple diverse cave reservoir realizations, which prove to be consistent with the geological patterns of real Tahe cave reservoir as well as the input conditioning data. The noise in the input probability map is suppressed by the generator. Once trained, the geomodelling process is quite fast: each realization with 336×256×96 cells takes 0.988 seconds using 1 GPU (V100). This study shows that GANSim-3D is robust for fast 3D conditional geomodelling of field reservoirs of arbitrary sizes.

Taxonomies for structuring models for World–Earth systems analysis of the Anthropocene: subsystems, their interactions and social–ecological feedback loops

In the Anthropocene, the social dynamics of human societies have become critical to understanding planetary-scale Earth system dynamics. The conceptual foundations of Earth system modelling have externalised social processes in ways that now hinder progress in understanding Earth resilience and informing governance of global environmental change. New approaches to global modelling of the human World are needed to address these challenges. The current modelling landscape is highly diverse and heterogeneous, ranging from purely biophysical Earth system models, to hybrid macro-economic integrated assessments models, to a plethora of models of socio-cultural dynamics. World–Earth models capable of simulating complex and entangled human–Earth system processes of the Anthropocene are currently not available. They will need to draw on and selectively integrate elements from the diverse range of fields and approaches; thus, future World–Earth modellers require a structured approach to identify, classify, select, combine and critique model components from multiple modelling traditions. Here, we develop taxonomies for ordering the multitude of societal and biophysical subsystems and their interactions. We suggest three taxa for modelled subsystems: (i) biophysical, where dynamics is usually represented by “natural laws” of physics, chemistry or ecology (i.e. the usual components of Earth system models); (ii) socio-cultural, dominated by processes of human behaviour, decision-making and collective social dynamics (e.g. politics, institutions, social networks and even science itself); and (iii) socio-metabolic, dealing with the material interactions of social and biophysical subsystems (e.g. human bodies, natural resources and agriculture). We show how higher-order taxonomies can be derived for classifying and describing the interactions between two or more subsystems. This then allows us to highlight the kinds of social–ecological feedback loops where new modelling efforts need to be directed. As an example, we apply the taxonomy to a stylised World–Earth system model that endogenises the socially transmitted choice of discount rates in a greenhouse gas emissions game to illustrate the effects of social–ecological feedback loops that are usually not considered in current modelling efforts. The proposed taxonomy can contribute to guiding the design and operational development of more comprehensive World–Earth models for understanding Earth resilience and charting sustainability transitions within planetary boundaries and other future trajectories in the Anthropocene.

Four Years of Earthquake Early Warning in Southern Iberia: 2016–2019

The performance of an earthquake early warning system (EEWS) for southern Iberia during the period of 2016–2019 is analyzed. The software PRESTo (PRobabilistic and Evolutionary early warning SysTem; the University of Naples Federico II, Italy) operating at the Universidad Complutense de Madrid has detected 728 events (2 < Mw < 6.3), with 680 earthquakes occurring in southern Iberia. Differences between the EEWS origin time and epicenter and those of the Instituto Geográfico Nacional (IGN) catalog are less than 2 s and 20 km, respectively, for 70% of the detected earthquakes. The main differences correspond to the EEWS magnitude that is underestimated for earthquakes that occurred at the west of the Gibraltar Strait (Mw differences larger than 0.3 for 70%). To solve this problem, several relationships have been tested, and a modification to those that currently use PRESTo is proposed. Other improvements, such as to densify the network or to use 3D Earth models, are proposed to decrease the time needed to issue the alert and avoid the false alerts (19 events over a total of 728 events). The EEWS has estimated the depth for 680 events and compared to those from the IGN (491 events). The performance of PRESTo during the 2020–2021 Granada swarm is analyzed. The hypocentral locations for the three largest earthquakes are close to those from the IGN (differences from 1 to 7 km for the epicenter and 0 s for the time origin), although there are some differences in their magnitude estimations that varies from 0.2 to 0.5. The PRESTo first times are 17, 25, and 41 s after the origin time. This study shows that the actual PRESTo EEWS configured for the southern Iberia may generate effective warnings despite the low seismicity rate in this region. To decrease the warning time, the geometry and density of the seismic network must be improved together with the use of 3D Earth models and on-site system approaches.

Seismic Driven Geomechanical Modeling of Uplifted and Subsided Wells in Mumbai Offshore and Its Engineering Implications

Seismic data provide evidence about hydrocarbon deposition, geological and geophysical subsurface information, including geomechanical aspects. Deriving and understanding geomechanical properties is crucial for reservoir management as it can avoid drilling and production-related problems that cause environmental impacts associated with land subsidence and uplift. The Poison's ratio (PR), Young Modulus (YM), and elastic moduli for a reservoir block were estimated using 3D seismic pre-stack data and well data. 3D Mechanical Earth Models (MEM) were also developed using the well logs, seismic horizons, and drilling data. Seismic data-derived geomechanical properties were compared with the mechanical earth models for the first time for this field. Well-tie analysis was used for inversion of 3D seismic data to extract detailed waveform and amplitude information. The brittleness index of the subsurface layers was estimated, which is a critical rock property that provides information about rock hardness and fragility phenomenon. The brittleness index has a diverse range from 5-35%, with significant contrast at shallow zones. PR and YM models generated from 3D MEM and seismic data have average values of 0.2 -0.6 and 5 - 28 GPa with significant contrast from shales and carbonates. The study recommends that the drilling through these problematic zones should be avoided to avoid wellbore problems that cause challenges in maintaining wellbore integrity and reservoir management in the North-Heera field, Mumbai Offshore Basin.

A hybrid method to calculate teleseismic body waves in a regional 3D model using GEMINI and SPECFEM.

<p>Modeling waveforms of teleseismic body waves requires the solution of the seismic wave equation in the entire Earth. Since fully-numerical 3D simulations on a global scale with periods of a few seconds are far too computationally expensive, we resort to a hybrid approach in which fully-numerical 3D simulations are performed only within the target region and wave propagation through the rest of the Earth is modeled using methods that are much faster but apply only to spherically symmetric Earth models.</p><p>We present a hybrid method that uses GEMINI to compute wave fields for a spherically symmetric Earth model up to the boundaries of a regional box. The wavefield is injected at the boundaries, where wave propagation is continued using SPECFEM-Cartesian. Inside the box, local heterogeneities in the velocity distribution are allowed, which can cause scattered and reflected waves. To prevent these waves from reflecting off the edges of the box absorbing boundary conditions are specifically applied to these parts of the wavefields. They are identified as the difference between the wavefield calculated with SPECFEM at the edges and the incident wavefield.</p><p>The hybrid method is applied to a target region in and around the Alps as a test case. The region covers an area of 1800 by 1350 km centered at 46.2°N and 10.87°E and includes crust and mantle to a depth of 600 km. We compare seismograms with a period of up to ten seconds calculated with the hybrid method to those calculated using GEMINI only for identical 1D earth models. The comparison of the seismograms shows only very small differences and thus validates the hybrid method. In addition, we demonstrate the potential of the method by calculating seismograms where the 1D velocity model inside the box is replaced by a velocity model generated using P-wave traveltime tomography.</p>

North American Crustal Motion and Glacial Isostatic Adjustment Model Predictions

<p>A suite of forward GIA model predictions, spanning a wide range of layered mantle viscosity and lithospheric thickness values, is compared to observed horizontal crustal motions in North America to discern optimal model parameters in order to minimize a root-mean-square (RMS) measure of the velocity residuals. To obtain the Earth model response, a combination of the full normal mode analysis and the collocation method is implemented. It provides a means to determine the surface loading response automatically and robustly to 1-dimensional (radially varying) Earth models, while retaining as much of the physics of the normal mode method as numerically feasible, given documented issues with singularities along the negative inverse-time axis in the Laplace transform domain. This method enables the exploration across a wide parameter range (for the lower mantle, transition zone, asthenosphere, and thickness of the elastic lithosphere) to find optimal combinations to explain horizontal crustal motion in North America. The analysis utilizes crustal motion rates from approximately 300 GNSS sites in central North America (Canada and United States) provided by the Nevada Geodetic Laboratory.  Preliminary results indicate that horizontal crustal motion predictions generated with a thin lithosphere, 40 – 60 km, produce horizontal motions that are strongly discrepant with the observations and have velocity residuals larger than the null model (modelled horizontal motion set to zero). As the lithospheric thickness increases, from 80 km to 240 km, the horizontal motion residuals gradually decrease with no minimum apparent for the thicknesses thus far considered. The residual velocities for the best-fitting models appear to carry a remaining signal, confirming previous inferences of limitations to spherically symmetric Earth models in modeling horizontal crustal motions in North America.</p>

Quantifying the influence of glacial isostatic adjustment on current and future sea-level change using 3-D Earth models

<p>Ongoing deformation of the Earth in response to past ice-ocean mass exchange is a significant contributor to contemporary sea-level changes and will be an important contributor to future changes. Calibrated models of this process, conventionally termed glacial isostatic adjustment (GIA), have been used to determine its influence on current and future sea-level changes. To date, the majority of these models have assumed a spherically-symmetric (1-D) representation of Earth structure. Here we apply a model that can simulate the isostatic response of a 3-D Earth in order to consider the contribution of lateral structure to model estimates of current and future sea-level change. We will present results from a global analysis based on two independent ice history reconstructions and a suite of 3-D Earth models with viscosity structure constrained using different seismic velocity models and recent estimates of lithosphere thickness variations. The accuracy of these GIA model parameter sets is assessed by comparing model output to a recently published data set of vertical land motion specifically intended to provide a robust measure of the GIA signal (Schumacher et al., Geophysical Journal International, 2018). This comparison indicates that the inclusion of lateral Earth viscosity structure results in an improved fit to the GPS-determined vertical land motion rates although significant residuals persist in some regions indicating that further efforts to improve constraints on this structure are necessary. Using the model parameter sets that best match the GPS constraints to predict the contribution of GIA to contemporary sea-level change indicates that lateral viscosity structure impacts the model estimates by order 1 mm/yr in some regions and that the model uncertainty is of a similar amplitude. Simulations of the GIA contribution to future sea-level change are also significantly affected, with differences, relative to a 1-D Earth model, reaching several decimetres on century timescales and several metres on millennial timescales. </p>