Forward Simulation Model Research Articles

Three-dimensional unstructured adaptive finite element forward modeling simulation of magnetotellurics based on magnetic vector potential-electric scalar potential

Three-dimensional unstructured adaptive finite element forward modeling simulation of magnetotellurics based on magnetic vector potential-electric scalar potential

Multi-origins of pathological theta oscillation from neuron to network inferred by a combined data and model study with cubature Kalman filter

The brain rhythm is strongly associated with the brain function. Alzheimer's disease (AD) is characterized reflected by the brain rhythm switching from the alpha band (9–12 Hz) to the theta band (4–8 Hz), accompanied with the loss of brain function. However, extracting the implicating intrinsic characteristic variations of the brain network by utilizing the Electroencephalogram (EEG) information is challenging. Kaman observer, serving as an effective Bayesian technique, can provide a visualization service for probing the intrinsic characteristics underlying the pathological theta oscillations. This work first establishes an excitation-inhibitory neural network model and explores the role of the fraction of the inhibitory neurons and inhibitory synapses in the pathological theta oscillation. The results indicate that the reduced inhibitory neuronal proportion and attenuated inhibitory synaptic weight are the main neural bases of the frequency reduction of neural oscillation. Then, we further explore the intrinsic spiking characteristic by considering spike frequency adaptation (SFA) to the inhibitory neurons. The results show that the SFA reduces the firing rate of neurons, which facilitates the theta rhythm. The enhancement of SFA current by increasing the time constant of its gating variable can further decrease the theta frequency from 7 Hz to 4 Hz. Furthermore, for this high-dimensional nonlinear excitation-inhibitory neural network model, cubature Kalman filter (CKF) is employed to estimate the above potential variations from the EEG data. The observation results show that the attenuated trends of the inhibitory neuronal proportion and the decreased inhibitory SFA current result in the descending brain rhythm. Finally, the theoretical simulation is deduced by utilizing the mean field theory for simplifying high-dimensional model to verify the simulation results. The theoretical variations of inhibitory parameters and adaptation gating variable are consistent with the simulation results. In summary, we investigate the multi-origin factors related to inhibitory neuronal intrinsic characteristics from forward model simulation and inverse EEG estimation process. And we further verify the simulation and data-driven results by theoretical derivation. This work enhances the understanding of the systematic function of inhibitory intrinsic characteristics on pathological theta oscillation and provides an effective method to decode the dynamics underlying neural activities.

3-D forward modeling of shale gas fracturing dynamic monitoring using the borehole-to-ground transient electromagnetic method

The borehole-ground transient electromagnetic method enhances the detection range and resolution by placing the transmitter electrode within the borehole, allowing for close-proximity excitation of subsurface targets and compensating for the shortcomings of traditional ground-based methods. Utilizing an unstructured vector finite element method combined with a second-order backward Euler variable time-stepping difference scheme and the MUMPS solver, we have achieved three-dimensional forward modeling simulation of the transient electromagnetic field for a borehole-ground electric source. Based on the validation of its accuracy, we analyzed the effectiveness of this method for dynamic monitoring of shale gas reservoir fracturing. Through forward modeling, we examined the characteristics of the radial electric field response during shale gas reservoir fracturing monitoring and evaluated the effectiveness of the borehole-ground TEM method in dynamic monitoring of shale gas reservoirs. The research results indicate that this method can meet the requirements for reservoir fracturing dynamic monitoring and has a promising application prospect.

Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problems

We present a non-trivial integration of dimension-independent likelihood-informed (DILI) MCMC (Cui et al 2016) and the multilevel MCMC (Dodwell et al 2015) to explore the hierarchy of posterior distributions. This integration offers several advantages: First, DILI-MCMC employs an intrinsic likelihood-informed subspace (LIS) (Cui et al 2014)—which involves a number of forward and adjoint model simulations—to design accelerated operator-weighted proposals. By exploiting the multilevel structure of the discretised parameters and discretised forward models, we design a Rayleigh–Ritz procedure to significantly reduce the computational effort in building the LIS and operating with DILI proposals. Second, the resulting DILI-MCMC can drastically improve the sampling efficiency of MCMC at each level, and hence reduce the integration error of the multilevel algorithm for fixed CPU time. Numerical results confirm the improved computational efficiency of the multilevel DILI approach.

Non-invasive investigations of closed landfills: An example in a karstic area

The monitoring of existing landfills is a pending environmental issue for the years to come. This monitoring is particularly challenging in the more and more common case of closed landfills, where direct investigation is difficult or impossible, calling for non-invasive methods, which in turn are stretched to maximizing their imaging capabilities in front of difficult logistical constraints, requiring novel and well-conceived scientific approaches. In this study we present a non-invasive approach designed and calibrated to identify the state of the subsoil underlying a closed urban waste landfill. In the presented case, two main questions had to be addressed: (a) whether large karstic cavities are present under the landfill, and (b) if any leachate leakage is present. A 3D Electrical Resistivity Tomography (ERT) configuration was used to solve the problem. The survey design has made use of forward model simulations, in order to verify whether the proposed approach was capable of imaging the possible large karstic cavities. This preliminary study showed the importance of choosing a suitable measurement protocol to recover the true position of the cavities. The analysis of the real field data did not show any anomaly compatible with the presence of large cavities, and thus, in comparison with the previous simulations, led to the conclusion that such cavities are not present. However, the results showed the presence of an electrically conductive anomaly, potentially be linked to leachate release. Direct investigations (drilling and sampling) confirmed the presence of fresh water in a silty sediment environment, both contributing towards the observed larger electrical conductivity, larger than the surrounding drier limestone bedrock. The presented general approach proved to be a valuable, generalizable, tool towards for the characterization and monitoring of closed landfills.

Noise Characteristics and Denoising Methods of Long-Offset Transient Electromagnetic Method

The advantages of the long-offset transient electromagnetic method include deep detection and sensitive response to resistivity anomalies. It is widely used in underground mineral resources exploration, fluid identification in petroleum reservoirs, hydraulic fracturing, and dynamic residual oil and gas monitoring. After the primary field signal is turned off, grounded electrodes or coils are used to observe the secondary eddy field. The secondary field signal decays quickly and has a large dynamic range and a wide frequency band but is easily affected by various natural and human electromagnetic interferences. Therefore, noise reduction and distortion correction are important issues in the processing of transient electromagnetic data. This paper proposes a systematic noise interference suppression process. Multi-period and positive–negative bipolar signal stackings were used to remove random noise and suppress DC offset signals. Then, a time-domain inverse digital recursive method was applied to remove characteristic frequency signals, e.g., power frequency signals and their harmonic interference. A standard noise-free signal was constructed through forward modeling simulation and verified by adding different types of noise. Finally, high-quality transient electromagnetic secondary field attenuation signals were obtained through overlapping windowing technology. We applied this algorithm to obtain electromagnetic data from dynamic monitoring of hydraulic fracturing in Fuling shale gas and from a copper–iron metal mine in Daye City, demonstrating its effectiveness.

Energy management strategy and simulation analysis of a hybrid train based on a comprehensive efficiency optimization

Since trains usually have long travel mileage requirements, designing an optimal energy management strategy has great potential for reducing energy consumption. First, a hybrid train power component model was developed, and the train operating efficiency, as well as a map of the optimal integrated efficiency of the entire vehicle operation, was derived for each mode from simulation calculations by analyzing the power flows in different operating modes. Next, the overall optimal efficiency of the entire vehicle was used to determine the mode division rules, and proportional integral control was used to ensure that the engine and motor operated at the target speed under the target torque. Finally, with the optimization objectives of reducing the fuel consumption of the entire vehicle and maintaining a balanced battery state of charge, a forward simulation model for the overall efficiency optimization of the entire vehicle was built in the MATLAB/Simulink software. Then, the trajectories of the engine and motor operating points, the fuel consumption, and the battery SOC were obtained for certain operating conditions. The effectiveness of the energy management strategy based on integrated efficiency optimization was verified by comparing the simulation results for the hybrid train with those of a conventional internal combustion engine train.

A simulation and data informed approach to porosity partitioning

Information about pore shape and geometry form one of the significant inputs in determining the effective elastic properties of porous rocks. The size and structure of the pores determine the amount of fluid contents which in turn determine their bulk response. Conventionally, effective medium models have been used to study the macroscopic properties of these rocks. These models often use a fixed set of microscopic variables (like the shape parameter) to describe their bulk behavior. Barring the specification of the elastic properties of the starting host material, their original formulations often lack the provision to integrate any experimental data, later. In the context of carbonates, the rocks exhibit a large variation in pore shapes and sizes, hindering the development of rock physics models. Therefore, they eventually require calibration. This is an important issue as the size of the pores here range from nanoscale to meter scale due the presence of primary and secondary porosity systems. In this paper we develop a method that can integrate the pore shape information into differential effective medium models using simple statistical and machine learning algorithms. Avoiding the use of conceptual values for the interparticle and vug type pores, our technique facilitates a data-based inference of the shape parameters from both the image data and forward model simulations. Micro-tomography images were used to extract the pore shape information for relatively larger pores such as the vugs. The information about the interparticle pores was deduced using a simulation-based approach from well-logs. Both of these were used to specify the properties of the new inclusions, hence aiding in data informed porosity partitioning. Finally, we outline a simple regression-based scheme using a first order polynomial model that can be used to calibrate any general rock physics model.

Application of Mask R-CNN for lab-based X-ray diffraction contrast tomography

Segmentation of spots in diffraction images is critical for accurate grain mapping in 3D. When the grain mapping is done by the recently established lab-based X-ray diffraction contrast tomography (LabDCT), diffraction spots suffering from low signal-to-noise ratios impose a severe challenge in precise identification of the spots using conventional image filters, thereby hindering the detection of small grains and limiting the spatial resolution. To overcome this challenge, we have applied an automatic instance segmentation deep learning network based on Mask R-CNN (two-stage region-based convolutional neural network) for finding spots in LabDCT images. The training data for the neural network was synthesized by combining virtual noise-free images (obtained from a forward simulation model) and noise-only images (obtained by filtering out diffraction spots in experimental images). Based on the diffraction spots deducted by the forward simulation model, data labelling and annotation was thus performed in an unsupervised manner without the need for tedious human labelling. By applying the network in a PyTorch framework called Detectron2, we show that the trained model performed significantly better than the conventional method in spot segmentation, resulting in a better grain reconstruction, subsequently. The work illustrates the potential of deep learning for improving LabDCT and other grain mapping techniques in a broader sense.

An Intelligent Detection Method for Obstacles in Agricultural Soil with FDTD Modeling and MSVMs

Unknown objects in agricultural soil can be important because they may impact the health and productivity of the soil and the crops that grow in it. Challenges in collecting soil samples present opportunities to utilize Ground Penetrating Radar (GPR) image processing and artificial intelligence techniques to identify and locate unidentified objects in agricultural soil, which are important for agriculture. In this study, we used finite-difference time-domain (FDTD) simulated models to gather training data and predict actual soil conditions. Additionally, we propose a multi-class support vector machine (MSVM) that employs a semi-supervised algorithm to classify buried object materials and locate their position in soil. Then, we extract echo signals from the electromagnetic features of the FDTD simulation model, including soil type, parabolic shape, location, and energy magnitude changes. Lastly, we compare the performance of various MSVM models with different kernel functions (linear, polynomial, and radial basis function). The results indicate that the FDTD-Yee method enhances the accuracy of simulating real agricultural soils. The average recognition rate of the hyperbola position formed by the GPR echo signal is 91.13%, which can be utilized to detect the position and material of unknown and underground objects. For material identification, the directed acyclic graph support vector machine (DAG-SVM) model attains the highest classification accuracy among all soil layers when using an RBF kernel. Overall, our study demonstrates that an artificial intelligence model trained with the FDTD forward simulation model can effectively detect objects in farmland soil.

MaNGIA: 10 000 mock galaxies for stellar population analysis

Context. Modern astronomical observations give unprecedented access to the physical properties of nearby galaxies, including spatially resolved stellar populations. However, observations can only give a present-day view of the Universe, whereas cosmological simulations give access to the past record of the processes that galaxies have experienced in their evolution. To connect the events that happened in the past with galactic properties as seen today, simulations must be taken to a common ground before being compared to observations. Therefore, a dedicated effort is needed to forward-model simulations into the observational plane. Aims. We emulate data from the Mapping Nearby Galaxies at Apache Point Observatory (MaNGA) survey, which is the largest integral field spectroscopic galaxy survey to date with its 10 000 nearby galaxies of all types. For this, we use the latest hydro-cosmological simulations IllustrisTNG to generate MaNGIA (Mapping Nearby Galaxies with IllustrisTNG Astrophysics), a mock MaNGA sample of similar size that emulates observations of galaxies for stellar population analysis. Methods. We chose TNG galaxies to match the MaNGA sample selection in terms of mass, size, and redshift in order to limit the impact of selection effects. We produced MaNGA-like datacubes from all simulated galaxies, and processed them with the stellar population analysis code pyPipe3D. This allowed us to extract spatially resolved maps of star formation history, age, metallicity, mass, and kinematics, following the same procedures used as part of the official MaNGA data release. Results. This first paper presents the approach used to generate the mock sample and provides an initial exploration of its properties. We show that the stellar populations and kinematics of the simulated MaNGIA galaxies are overall in good agreement with observations. Specific discrepancies, especially in the age and metallicity gradients in low- to intermediate-mass regimes and in the kinematics of massive galaxies, require further investigation, but are likely to uncover new physical understanding. We compare our results to other attempts to mock similar observations, all of smaller datasets. Conclusions. Our final dataset is released with this publication, consisting of ≳10 000 post-processed datacubes analysed with pyPipe3D, along with the codes developed to create it. Future work will employ modern machine learning and other analysis techniques to connect observations of nearby galaxies to their cosmological evolutionary past.

Design of a Virtual Multi-Interaction Operation System for Hand–Eye Coordination of Grape Harvesting Robots

In harvesting operations, simulation verification of hand–eye coordination in a virtual canopy is critical for harvesting robot research. More realistic scenarios, vision-based driving motion, and cross-platform interaction information are needed to achieve such simulations, which are very challenging. Current simulations are more focused on path planning operations for consistency scenarios, which are far from satisfying the requirements. To this end, a new approach of visual servo multi-interaction simulation in real scenarios is proposed. In this study, a dual-arm grape harvesting robot in the laboratory is used as an example. To overcome these challenges, a multi-software federation is first proposed to establish their communication and cross-software sending of image information, coordinate information, and control commands. Then, the fruit recognition and positioning algorithm, forward and inverse kinematic model and simulation model are embedded in OpenCV and MATLAB, respectively, to drive the simulation run of the robot in V-REP, thus realizing the multi-interaction simulation of hand–eye coordination in virtual trellis vineyard. Finally, the simulation is verified, and the results show that the average running time of a string-picking simulation system is 6.5 s, and the success rate of accurate picking point grasping reached 83.3%. A complex closed loop of “scene-image recognition-grasping” is formed by data processing and transmission of various information. It can effectively realize the continuous hand–eye coordination multi-interaction simulation of the harvesting robot under the virtual environment.

Cross-mode stabilized stochastic shallow water systems using stochastic finite element methods

The development of surrogate models to study uncertainties in hydrologic systems requires significant effort in the development of sampling strategies and forward model simulations. Furthermore, in applications where prediction time is critical, such as prediction of hurricane storm surge, the predictions of system response and uncertainties can be required within short time frames. Here, we develop an efficient stochastic shallow water model to address these issues. To discretize the physical and probability spaces we use a Stochastic Galerkin method and an Incremental Pressure Correction Scheme to advance the solution in time. To overcome discrete stability issues, we propose cross-mode stabilization methods which employs existing stabilization methods in the probability space by adding stabilization terms to every stochastic mode in a modes-coupled way. We extensively verify the developed method for both idealized shallow water test cases and hindcasting of past hurricanes. We subsequently use the developed and verified method to perform a comprehensive statistical analysis of the established shallow water surrogate models. Finally, we propose a predictor for hurricane storm surge under uncertain wind drag coefficients and demonstrate its effectivity for Hurricanes Ike and Harvey.

Cosmogenic 3He paleothermometry on post-LGM glacial bedrock within the central European Alps

Abstract. Diffusion properties of cosmogenic 3He in quartz at Earth surface temperatures offer the potential to directly reconstruct the evolution of past in situ temperatures from formerly glaciated areas, which is important information for improving our understanding of glacier–climate interactions. In this study, we apply cosmogenic 3He paleothermometry to rock surfaces gradually exposed from the Last Glacial Maximum (LGM) to the Holocene period along two deglaciation profiles in the European Alps (Mont Blanc and Aar massifs). Laboratory experiments conducted on one representative sample per site indicate significant differences in 3He diffusion kinetics between the two sites, with quasi-linear Arrhenius behavior observed in quartz from the Mont Blanc site and complex Arrhenius behavior observed in quartz from the Aar site, which we interpret to indicate the presence of multiple diffusion domains (MDD). Assuming the same diffusion kinetics apply to all quartz samples along each profile, forward model simulations indicate that the cosmogenic 3He abundance in all the investigated samples should be at equilibrium with present-day temperature conditions. However, measured cosmogenic 3He concentrations in samples exposed since before the Holocene indicate an apparent 3He thermal signal significantly colder than today. This observed 3He thermal signal cannot be explained with a realistic post-LGM mean annual temperature evolution in the European Alps at the study sites. One hypothesis is that the diffusion kinetics and MDD model applied may not provide sufficiently accurate, quantitative paleo-temperature estimates in these samples; thus, while a pre-Holocene 3He thermal signal is indeed preserved in the quartz, the helium diffusivity would be lower at Alpine surface temperatures than our diffusion models predict. Alternatively, if the modeled helium diffusion kinetics is accurate, the observed 3He abundances may reflect a complex geomorphic and/or paleoclimatic evolution, with much more recent ground temperature changes associated with the degradation of alpine permafrost.

Sensory attenuation from action observation

A central claim of many embodied approaches to cognition is that understanding others' actions is achieved by covertly simulating the observed actions and their consequences in one's own motor system. If such a simulation occurs, it may be accomplished through forward models, a component of the motor system already known to perform simulations of actions and their consequences in order to support sensory-monitoring of one's own actions. Forward-model simulations cause an attenuation of sensory intensity, so if the simulations hypothesized by embodied cognition are indeed provided by forward models, then action observation should trigger this sensory attenuation. To test this hypothesis, the experiments reported here measured the perceived intensity of a touch sensation on the finger when participants observed an active touch (a finger reaching to touch a ball) vs. a passive touch (a ball rolling to touch an unmoving finger). The touch sensation was perceived as less intense during observation of active touch in comparison with observation of passive touch, providing evidence that forward models are indeed engaged during action observation. The strength of this sensory attenuation is compared and contrasted with a well-established sensory-amplification effect caused by visual attention. This sensory-amplification effect has not generally been considered in studies related to sensory attenuation in action observation, which may explain conflicting results reported in the field.

CO2-Containing Reservoir Evaluation Based on the Joint Inversion of Nuclear Logs and Resistivity Logs: A Case Study of Ultrahigh-Temperature and High-Pressure Gas Reservoirs in the Yinggehai Basin, China.

Due to the unique characteristics of reservoirs in the Yinggehai Basin in the South China Sea, such as high temperature and high pressure (HPHT), low porosity, low permeability, complex pore structure, and high lime content, the log responses of these reservoirs have very complex characteristics, which makes it difficult to evaluate reservoir parameters accurately. In addition, most reservoirs in Ledong Block of the Yinggehai Basin in the South China Sea contain CO2, posing great difficulties for subsequent exploration and development. Accurate evaluation of CO2 layers is of paramount importance for the development of oil and gas fields. In this study, we used a method for the joint inversion of multiple well logs to evaluate the reservoirs and determine CO2 saturation level and other formation parameters. We optimized the joint inversion model based on the characteristics of the reservoirs in the Yinggehai Basin and adjusted the forward simulation model to consider the effects of high temperature and high pressure on gas density. In view of high lime content in the formations in this area, we adjusted the resistivity forward simulation model to consider the effect of lime content. The inversion results show that the values of porosity, permeability, and water saturation level obtained through inversion are largely consistent with the core data. The CO2 saturation level determined through joint inversion is 22%, which represents a deviation of less than 10% from the drilling system testing (DST) result, indicating that the joint inversion method is accurate. The error in the water saturation level determined through the joint inversion method is smaller than that in the calculated results from conventional multimineral inversion models. We performed forward simulation of the results calculated with the joint inversion method and compared the results of forward simulation with actual log curves. For the sandstone interval, the results of forward simulation are largely consistent with the actual log curves, indicating that the joint inversion method is accurate. In summary, the method presented in this paper can accurately determine reservoir parameters and provide strong support for the exploration and development of oil and gas fields in the Yinggehai Basin in the South China Sea.

VLIDORT-QS: A quasi-spherical vector radiative transfer model

We present a new VLIDORT-based quasi-spherical vector radiative transfer (RT) model (VLIDORT-QS) that was developed for forward-model simulations of long-path measurements in limb- or nadir-viewing spherical-atmosphere configurations. This paper provides a theoretical description of the RT model, in which the single-scatter radiation field is treated accurately in spherical geometry, while the diffuse-scatter source terms at points along the line-of-sight are approximated locally through pseudo-spherical simulations based on the VLIDORT RT code; these multiple scatter source terms are then folded into the source function integration to yield the radiation field at any line-of-sight observation point in the atmosphere. In addition, the VLIDORT-QS model has a complete linearization facility (ability to generate analytic Jacobians of the radiation field with respect to any atmospheric or surface property). Validation of the model has been carried out against a fully-spherical Monte Carlo code, and also against the standard VLIDORT model for nadir-view scenarios with relatively low sphericity effects. A sample application of VLIDORT-QS for a biomass-burning plume scenario is demonstrated. A second paper will present a detailed analysis of the retrieval of ice-cloud properties from high-altitude aircraft measurements taken as part of the NASA Airborne Tropical Tropopause Experiment (ATTREX) campaign, with VLIDORT-QS providing forward-model simulations of the radiation fields and associated Jacobian functions.

Deep learning to estimate permeability using geophysical data

Time-lapse electrical resistivity tomography (ERT) is a popular geophysical method to estimate three-dimensional (3D) permeability fields from electrical potential difference measurements. Traditional inversion and data assimilation methods are used to ingest this ERT data into hydrogeophysical models to estimate permeability. Due to ill-posedness and the curse of dimensionality, existing inversion strategies provide poor estimates and low resolution of the 3D permeability field. Recent advances in deep learning provide us with powerful algorithms to overcome this challenge. This paper presents a deep learning (DL) framework to estimate the 3D subsurface permeability from time-lapse ERT data. To test the feasibility of the proposed framework, we train DL-enabled inverse models on simulation data. Each measurement in both synthetic and field data is standardized by removing the mean and scaling the time-series to unit variance. This pre-processing step is necessary to bring simulation data closer to field observations. Subsurface process models based on hydrogeophysics are used to generate this synthetic data. Training performed on limited simulation data resulted in the DL model over-fitting. An advanced data augmentation based on mixup is implemented to generate additional training samples to overcome this issue. This mixup technique creates weakly labeled (low-fidelity) samples from strongly labeled (high-fidelity) data. The weakly labeled training data is then used to develop DL-enabled inverse models and reduce over-fitting. As both time-lapse ERT (1133048 features/realization) and 3D permeability (585453 features/realization) data samples are from a high-dimensional space, principal component analysis (PCA) is employed to reduce dimensionality. Encoded ERT and encoded permeability are generated using the trained PCA estimators. A deep neural network is then trained to map the encoded ERT to encoded permeability. This mixup training and unsupervised learning allowed us to build a fast and reasonably accurate DL-based inverse model under limited simulation data. Results show that proposed weak supervised learning can capture salient spatial features in the 3D permeability field. Quantitatively, the average mean squared error (in terms of the natural log) on the strongly labeled training, validation, and test datasets is less than 0.5. The R2-score (global metric) is greater than 0.75, and the percent error in each cell (local metric) is less than 10%. Finally, an added benefit in terms of computational cost is that the proposed DL-based inverse model is at least O(104) times faster than running a forward model once it is trained. Data generation, DL model training, and hyperparameter tuning to identify optimal neural network architectures utilized high-performance computing resources while the DL inference is performed on a standard laptop. Approximately, O(105) processor hours are used for generating data and DL tuning and training. We acknowledge that the data generation and DL model development are expensive. But once a DL model is trained, it can be re-used for inversion rapidly for the given system, with set physics and domain. Note that traditional inversion may require multiple forward model simulations (e.g., in the order of 10 to 1000), which are very expensive. This computational savings ≈O(105)−O(107) makes the proposed DL-based inverse model attractive for subsurface imaging and real-time ERT monitoring applications due to fast and yet reasonably accurate estimations of permeability field.

High‐Dimensional Groundwater Flow Inverse Modeling by Upscaled Effective Model on Principal Components

AbstractThe main computational costs of gradient‐based inverse methods for high‐resolution groundwater flow inverse problems include costly forward model simulations and the large number of such simulations required to determine the Jacobian matrix. We develop an upscaling‐based inverse approach, named upscaled principal component inverse approach (UPCIA), which achieves dimensionality reduction and reduces computational cost of forward model simulations by evaluating the Jacobian through upscaled effective models on a coarse‐resolution grid constructed from upscaled principal components. UPCIA integrates downscaling into the inverse problem by estimating principal component coefficients based on the coarse‐resolution forward model, which are then used to generate high‐resolution parameter fields. Various numerical experiments demonstrate the effectiveness and efficiency of UPCIA, including 2‐D/3‐D high‐dimensional steady‐state and transient hydraulic tomography with known storativity to estimate multi‐Gaussian transmissivity or hydraulic conductivity fields. Results show that the hydraulic head is insensitive to small‐scale variability of conductivity, and UPCIA provides high‐quality inversion results similar to inverse methods with high‐resolution forward model simulations and significantly reduces computation time by orders of magnitude. In addition to supporting the characterization of heterogeneity in sufficient detail, UPCIA can also be used to examine whether finer resolution is necessary and possibly to determine an optimal inverse resolution.

PSP/WISPR Observations of Dust Density Depletion near the Sun. II. New Insights from within the Depletion Zone

Visible light observations from the Wide-field Imager for Solar PRobe (WISPR) aboard the Parker Solar Probe (PSP) mission offer a unique opportunity to study the dust environment near the Sun. The existence of a dust-free zone (DFZ) around stars was postulated almost a century ago. Despite numerous attempts to detect it from as close as 0.3 au, observational evidence of a circumsolar DFZ has remained elusive. Analysis of WISPR images obtained from heliocentric distances between 13.3–53.7 R ⊙ over multiple PSP orbits shows a gradually decreasing brightness gradient along the symmetry axis of the F-corona for coronal heights between 19 and 9 R ⊙. Below 9 R ⊙, the gradient reverses its trend, approaching the radial dependence exhibited at heights above 19 R ⊙. After taking into account the effects of both the electron corona background and the nonresolved starlight, the WISPR observations down to 4 R ⊙ are consistent with forward-modeling simulations of the F-corona brightness within [−6, 5]% if a circumsolar region of depleted dust density between 19 and 5 R ⊙ enclosing a DFZ is considered. In addition, we show, for the first time, that the F-corona brightness inward of about 15 R ⊙ depends on the observer’s location for observing distances below 35 R ⊙.