Uncertainties In Soil Properties Research Articles

Assessing Soil Prediction Distributions for Forest Management Using Digital Soil Mapping

Texture, soil organic matter (SOM), and soil depth (SoD) are crucial properties in forest management because they can supply spatial information on forest site productivity and guide fertilizer applications. However, soil properties possess an inherent uncertainty that must be mapped to enhance decision making in management applications. Most digital soil mapping predictions primarily concentrate on the mean of the distribution, often neglecting the estimation of local uncertainty in soil properties. Additionally, there is a noticeable scarcity of practical soil examples to demonstrate the prediction uncertainty for the benefit of forest managers. In this study, following a digital soil mapping (DSM) approach, a Quantile Regression Forest (QRF) model was developed to generate high-resolution maps and their uncertainty regarding the texture, SoD, and SOM, which were expressed as standard deviation (Sd) values. The results showed that the SOM (R2 = 0.61, RMSE = 2.03% and with an average Sd = 50%), SoD (R2 = 0.74 and RMSE = 19.4 cm), clay (R2 = 0.63, RMSE = 10.5% and average Sd = 29%), silt (R2 = 0.59, RMSE = 6.26% and average Sd = 33%), and sand content (R2 = 0.55, RMSE = 9.49% and average Sd = 35%) were accurately estimated for forest plantations in central south Chile. A practical demonstration of precision fertilizer application, utilizing the predictive distribution of SOM, effectively showcased how uncertainty in soil attributes can be leveraged to benefit forest managers. This approach holds potential for optimizing resource allocation and maximizing economic benefits.

Dimension-Reduction Spectral Representation of Soil Spatial Variability and Its Application in the Efficient Reliability Analysis of Seismic Response in Tunnels

Reliability analysis of tunnels considering soil spatial variability was traditionally carried out using Monte Carlo simulations, which are computationally expensive. This study proposes a dimension-reduction spectral representation method (SRM) for simulating parameter stochastic fields to capture uncertainties in soil properties, which can be coupled with the probability density evolution method (PDEM). Additionally, the PDEM is employed to conduct seismic reliability assessments of tunnels in a more efficient manner. By integrating a non-intrusive dynamic random finite element method (FEM), this study explores the horizontal drift angle and vertical relative settlement of tunnel dynamic response under different coefficient of variations (COVs) and horizontal autocorrelation lengths of elastic modulus (E). Results show the changes in COVs have more pronounced effects on the mean and standard deviation of horizontal drift angle and vertical relative settlement of tunnel than alterations in horizontal autocorrelation lengths. A limit state pertaining to the horizontal drift angle and vertical relative settlement is established through the use of a probability density function (PDF) and cumulative distribution function (CDF). The results suggest that the combination of soil spatial variability and PDEM can serve as an effective approach for practical tunnel design.

Reliability analysis using Limit equilibrium and Smoothed Particle Hydrodynamics-based method for homogeneous soil slopes.

This paper develops a combined method to predict the volume of sliding mass for homogeneous slopes in an efficient manner. Firstly, the failure surface with minimum factor of safety (FS) in Limit Equilibrium Method is equated to that one determined by Smoothed Particle Hydrodynamics algorithm to obtain the threshold displacement value for unstable and stable particles. Secondly, the threshold displacement value is used to identify the volume of sliding mass using SPH. Finally, a regression model is developed based on a finite number of SPH simulations for homogeneous soil slopes. The proposed LEM-SPH based method is illustrated through a cohesive soil slope. It is concluded that the use of failure surface with minimum FS in LEM tends to underestimate the volume of sliding mass and to give an unconservative risk value. The Coefficient of Variation (Cov) of volume of sliding mass are 0.14, 0.28, 0.4, 0.48, 0.53 for Cov of soil properties = 0.2, 0.3, 0.4, 0.5, and 0.6, respectively. The uncertainty of soil properties has a significant effect on the mean value of volume of sliding mass and therefore the landslide risk value. The proposed method is necessitated for cases where large uncertainties in soil properties exist.

Reliability analysis of geosynthetic-reinforced slopes under rainfall infiltration

The rainfall-induced instability of geosynthetic-reinforced is a time-dependent phenomenon owing to the infiltration process, and is influenced by rainfall patterns. Catering to the inherent uncertainty in soil properties, this study conducted a reliability analysis of three-dimensional (3D) vertical geosynthetic-reinforced slopes, in order to explore how the probabilistic stability of slope evolves over time under different rainfall patterns. A 3D horn-like mechanism incorporating the Conte-Troncone (CT) model is adopted as a framework for deterministic analysis. Through a Fourier transform-based theoretical reasoning, the CT model assesses the time-variable pore-water pressure of soils in response to any continuously varying rainfall intensity over time. Subsequently, the pore water pressure-driven changes in soil unsaturated strength and the corresponding extend power are integrated into the three-dimensional mechanism, enabling a rapid determination of the instantaneous safety factor at discrete time instants. To avoid the tedious computation generated by Monte Carlo simulation, a simplified Hasofer-Lind-Rackwitz-Fiessler (HLRF) algorithm is used to calculate the time-varying reliability indices. Using the implantation of the proposed method, the effects of rainfall pattern, slope width, and reinforcement tensile strength are investigated by parametric analysis.

Sustainable and resilient seismic design of reinforced concrete frames with rocking isolation on spread footings

The rapidly evolving climate change together with the urgent need of modern societies for resilience against catastrophic threats set the sustainable and resilient seismic design of reinforced concrete (RC) structures as a priority. Rocking isolation of RC frames resting on spread footings has been proven numerically and experimentally to offer superior seismic performance with reduced seismic demands in the superstructure. At the same time, rocking footings do not require special construction solutions and can be readily implemented in the current state of practice. To exploit these benefits, the present study, for first time, utilizes rocking footings in the optimum design of RC frames for high seismic resilience and reduced environmental impact. This is achieved by incorporating a resilient seismic design methodology into a numerical optimization procedure that is aiming to minimize embodied carbon. Applications of the proposed approach show that the carbon footprint of RC frames can be reduced by 40% due to rocking isolation on spread footings. The benefits become more important as the level of seismic hazard increases. It is also found that the environmental benefits of rocking footings for RC frames are rather insensitive to the characteristics of ground motions and the uncertainties of soil properties.

Surrogate-assisted uncertainty modeling of embankment settlement

The structural optimization of basal reinforced piled embankments is usually conducted by examining design alternatives while ignoring the inherent variability of soil properties and studying only a limited number of structural variables. As an alternative, this paper proposes a hybrid modeling framework to introduce soil property uncertainty into embankment settlement calculations. This is important because settlement is critical in the serviceability assessments considered during structural optimization. The proposed framework consists of uncertainty modeling, finite element method, surrogate modeling, and probabilistic analysis. More specifically, a neural network with Monte Carlo dropout that accounts for uncertainty is employed to correlate the soil properties which affect the long-term performance of embankments over soft clay. Next, a coupled finite element analysis is performed using two constitutive soil parameters generated by the neural network to predict post-construction settlements. Combining the finite element (input source) with a surrogate model (data-driven approximation) yields substantial settlement outcomes for structure evaluations. A case study is then used to validate the effectiveness and applicability of this framework. Finally, an exhaustive search approach is used to design a cost-effective improved ground within ultimate and serviceability limit state constraints. Pareto front is computed using a logistic function at different settlement reliability levels.

Reliability analysis of pile-founded T-walls using the random variable method and random field method

The pile-founded walls (e.g., T-walls) are important civil infrastructure for flood protection. The stability analysis of floodwalls in the face of flooding hazards is a critical problem for the design and maintenance of floodwalls. However, most current studies for the stability assessment of floodwalls are based on deterministic analyses. The uncertainties of soil properties, which cannot be avoidable due to complex geological processes and depositional environment, are not sufficiently studied. To address the uncertainties of soil properties, a two-dimensional pile-founded T-wall system with several clay layers on the top and a sand layer on the bottom is taken as an example for the reliability analysis. Both the random variable method and random field method are adopted for the probabilistic analysis of floodwalls. The effect of the coefficient of variation (COV) and scales of fluctuation of soil properties are investigated by parametric studies. The results show that the probability of failure increases with the COV of soil properties for both methods. For the random field method, the effect of the vertical scale of fluctuation of soil properties is more profound than the horizontal scale of fluctuation. The probability of failure derived from the random variable method is generally larger than that from the random field method at high flood water elevations when the same COV of soil properties is applied. This study can provide useful references for risk-informed decision-making in the stability assessment of floodwalls in the face of flooding hazards.

Uncertainty propagation of soil property in dynamic site response under different site conditions

AbstractIt has been recognized that the soil profile and the associated parameters are usually not known with certainty in probabilistic seismic hazard analysis, however, little is known about the influence of site conditions on the propagation of uncertainty. This work aims to analyze and quantify the propagation of soil property uncertainties in site response analysis under different site conditions using a recently proposed DCZ constitutive model. Sensitivity analysis is performed to determine the most important parameter, which is regarded as a random variable in the subsequent uncertainty analysis. A Monte Carlo simulation procedure is applied to quantify the uncertainty propagation caused by the random variable. Detailed case studies involving multiple site conditions are presented to quantify the impacts of soil property variability under different site conditions in site responses. This study has found that variation in site response strongly depends on the shear wave velocity compared to other parameters. For softer soil, the PGA values at the ground surface show more dispersion (i.e., uncertainties) compared to stiffer soil. This indicates that for improving the accuracy of site response analysis and earthquake disaster prediction, it is necessary to operate probabilistic dynamic site response analysis rather than deterministic analysis, especially for soft soil.

Parametric geological model update and probabilistic analysis of shield tunnel excavation: A borehole-based conditional random fields approach

Shield tunnel construction has faced considerable challenges due to the uncertain geological conditions in real scenarios. The borehole data obtained from geological investigations are often too limited to effectively evaluate uncertainties at nonsampled points. Moreover, the existing probabilistic safety assessment (PSA) frameworks that consider the uncertainties of soil properties have encountered limitations in engineering practice due to the complexity of conditional random field (CRF) theory and the operability of numerical modeling. Based on the digitization of the underground infrastructure, this study implements the object-oriented parametric modeling (PM) to address these limitations. A PM-CRF engine is first developed to automate the process of parametric geological modeling using CRFs. Then, a PM-CRF-PSA interoperability framework is proposed to automatically generate a PSA model for comprehensive safety assessment considering the uncertainties of soil properties. Finally, a case study is conducted with the proposed framework to assess the whole process of parametric modeling from CRF characterization to PSA. The results of this study highlight the clear advantages of this framework compared to traditional deterministic safety assessment (DSA) methods in terms of characterization of the uncertainty of soil properties and the efficiency of PSA modeling. Thus, the proposed framework enables the efficient and comprehensive exploration of shield tunnel design schemes by considering the uncertainties of soil properties.

A multivariate approach for mapping a soil quality index and its uncertainty in southern France

AbstractPedometricians have spent a lot of effort on mapping soil types and basic soil properties. However, end‐users typically need a more elaborate soil quality index for land management. Soil quality indices are typically derived from multiple individual soil properties by evaluating whether specific criteria are met. If this is based on individually mapped soil properties, then an important consequence is that cross‐correlations between soil properties are ignored. This makes it impossible to quantify the uncertainties associated with the mapped indices. The objective of this study was to map a soil potential multifunctionality index for agriculture (Agri‐SPMI) over a 12,125 km2 study region located along the French Mediterranean coast to help urban planners preserve soils of the highest quality. The index considered the ability of soils to fulfil four functions under five land use scenarios. A binary map represented each soil function fulfilment for a given scenario. The final soil quality index map was the sum of the 20 binary maps. A regression cokriging model was developed to map the basic soil properties first individually from legacy soil data and spatial soil covariates using a random forest algorithm, and next, interpolate the residuals using cokriging and the linear model of coregionalisation. The mapping uncertainties of soil properties were propagated by calculating the soil quality index over 300 stochastic simulations of soil properties derived from the linear models of coregionalisation. Results showed a poor prediction accuracy of the quality index, mainly because some soil properties were poorly predicted (notably available water capacity and coarse fragments) and used in combination with extreme thresholds that defined land suitability. Overall, the uncertainty was correctly quantified because the stochastic simulations reproduced the width of the observed distribution well, but the shapes of the distributions differed considerably from those of the observations. We envisage some ways for improvement, such as creating probability maps instead of the mean from simulations, and changing the prediction support from point to area.Highlights The study mapped a soil quality index (Agri‐SPMI) to help preserve soils of highest quality along the French Mediterranean coast. The Agri‐SPMI considered the ability of soils to fulfill four functions under five land use scenarios, derived from multiple individual soil properties. A regression cokriging model was developed to map the basic soil properties, followed by interpolating the residuals using cokriging and the linear model of coregionalisation. The study accurately quantified mapping uncertainties using stochastic simulations, but the soil quality index prediction accuracy was poor, with suggestions for improvement.

Efficient modeling of random fields by using Gaussian process inducing-point approximations

Gaussian process (GP) regression provides an elegant way to estimate the trend and associated uncertainty of soil properties from datasets by using flexible and expressive kernels. However, the exact computation is intensive due to the dense operation of the matrix that scales cubically with the size of the datapoints. This paper proposes using inducing-point approximations as an efficient method to address the problem. The approximated posterior mean and covariance are reformulated for efficient sampling. Numerical results indicate that in terms of efficiency, sparse approximations are a significant improvement over exact computations. The locations of arbitrarily selected inducing points are compared in terms of approximation accuracy, and recommendations are provided based on empirical evidence. In addition, the relations between the relevance vector machine (RVM), Bayesian compressive sensing (BCS), and GP for random field modeling are clarified. The results from different models are compared. The sparse GP provides a more reasonable estimation of the uncertainty at locations far from the observations while being competitive with the RVM and BCS in terms of computational efficiency.

Stochastic dynamic response and seismic fragility analysis for high concrete face rockfill dams considering earthquake and parameter uncertainties

The multiple uncertainties of soil properties and ground motions significantly affect the seismic response of high concrete face rockfill dams (CFRDs). Three types of randomness (the material parameter randomness, stochastic seismic excitation, and their coupling) are generated by stochastic ground motion model and generalized F-discrepancy (GF-discrepancy) method. The generalized probability density evolution method (GPDEM) is used to analyze the influence of randomness on the stochastic dynamic response and system reliability of high CFRDs. The limit state of CFRD is defined according to the crest subsidence and the stress of face-slab, and a performance-based reliability evaluation framework is established. The results show that the randomness of ground motions dominates stochastic dynamic response of the CFRD and the dynamic reliability at different safety thresholds can be obtained by the above methods. Moreover, the proposed seismic safety evaluation framework provides a significant guidance for designing and analyzing practical engineering projects.

Seismic fragility analysis of free-spanning submarine pipelines incorporating soil spatial variability in soil-pipe interaction and offshore motion propagation

Geotechnical engineering problems are always faced with various sources of uncertainty, especially the uncertainties of soil properties. This paper presents the seismic fragility analysis of free-spanning submarine pipelines (FSSPs) with consideration of soil spatial variability. Firstly, the two-dimensional random field theory is employed to account for the spatial distribution characteristics of soil properties, and the levels of spatial variability are shifted by changing the autocorrelation lengths. Then, based on the generated random field data, the finite element (FE) models of the soil-pipe system are established and the offshore spatial motions applied to pipelines are stochastically synthesized. Meanwhile, the impacts of soil spatial variability on the soil-pipe interaction and offshore motion propagation are investigated. Next, the probabilistic seismic demand models of FSSPs are derived via dynamic analyses and statistical regression, and the seismic fragility curves are generated. The influence of soil spatial variability on the seismic performance of FSSPs is examined by comparing the fragility curves between the random-field and random-variable cases. Finally, a fast assessment method for the ultimate bearing capacity is proposed to simply incorporate soil spatial variability in the seismic performance estimation of FSSPs. Numerical results highlight the necessity of incorporating soil spatial variability in seismic fragility analysis of FSSPs. The proposed assessment method can provide a great convenience for designers and engineers to obtain the accurate ultimate bearing capacity of FSSPs or other underground structures in the preliminary stage of seismic design.

Performance-based assessment of the monopile foundation of offshore wind turbines

Due to uncertainties in metocean conditions and soil properties at the offshore wind farm, the performance of the monopile foundation of offshore wind turbines (OWTs) is also uncertain. In this study, a performance-based assessment method is suggested to calculate the annual probability of exceedance of a performance measure of monopile foundations of OWTs. In the suggested method, a finite element model is constructed to analyze the responses of the monopile foundation when the OWT system is subjected to aerodynamic and hydrodynamic loadings. A procedure is suggested to develop the fragility function of the monopile foundation considering both uncertainties in metocean conditions and soil properties. The annual probability of exceedance of the performance measure is then calculated by integrating the fragility function with respect to the occurrence rate of the metocean conditions. An example is provided to demonstrate the application of the suggested method. It is found that the annual probability of exceedance of the allowable lateral displacement of the monopile foundation at mudline is affected by both uncertainties in metocean conditions and soil properties. Specifically, it is unconservative to ignore the uncertainties in soil properties.

RELIABILITY BASED DESIGN OF GEOTECHNICAL SYSTEMS BASED ON A DECOUPLED RELIABILITY ANALYSIS AND OPTIMIZATION APPROACH

Reliability-based design intends to design an engineering system subject to probabilistic design constraints. This paper presents a reliability design optimization approach of geotechnical engineering problems based on a decoupled reliability analysis and optimization concept. Geotechnical and foundation systems involve considerable uncertainties in soil properties, loading conditions, and inadequate site characterization. A traditional reliability design optimization method, such as the reliability index approach, requires a double-loop optimization structure, since reliability analysis is implemented for constraint function evaluation using a numerical optimization algorithm. The decoupled approach separates the reliability analyses of the constraint functions from the design optimization loop. An adaptive metamodeling method is proposed to create approximate functions of the reliability indices, so that the reliability design optimization is performed using the approximate functions of the reliability constraints. Radial basis functions are implemented to build the metamodels. To progressively improve the accuracy of predicting the reliability indices, an adaptive sampling technique is employed, so that new samples are added using the optimal points found in previous design iterations. Numerical examples are solved and presented, and the proposed design approach works well. It is shown to be a useful alternative for solving reliability-based design of geotechnical and foundation systems.

The generation of uniform hazard Floor Response Spectra

In the generation of Floor Response Spectra (FRS), there are significant uncertainties in the earthquake excitations and site conditions, which lead to uncertainty in the resultant FRS. A fully probabilistic method is developed to account for the uncertainties and to generate FRS with a desired return period (RP). The uncertainties in seismic inputs and geotechnical models from nuclear industry are considered in this research. A set of seismic inputs at the bedrock with different loading levels are used to represent the uncertainty in earthquake excitations. A large number of site profiles are developed by Monte Carlo simulation to account for the uncertainty in soil properties. Following site response analysis and soil-structure interaction (SSI) analysis, amplification factors are determined for each site under the excitations of different ground motions to address the effects of seismic wave propagation and SSI. The uncertainties are propagated from site response analysis to SSI analysis consistently. Based on these analyses and seismic hazard curves at the bedrock, uniform hazard FRS are obtained with a complete probabilistic description. In the numerical examples based on the uncertainty model in nuclear industry, the uniform hazard FRS generated by the proposed method can reduce the seismic demands significantly (up to 46% as compared to results obtained by the current practice when the RP is set to 10,000 years) to provide an economical solution for seismic design. Furthermore, it overcomes the underestimation of FRS in some frequency ranges by the current practice. Sensitivity study is performed to study the influence of the logarithmic standard deviation and the correlation coefficient between adjacent soil layers of shear-wave velocity.

Probabilistic modeling for reliability analysis of buried pipelines subjected to spatiotemporal earthquakes

Reliability analysis of buried pipelines subjected to spatiotemporal seismic involves modeling multiple uncertainties related to loads, soil, and mechanical mechanisms. The present study proposes a probabilistic modeling method that can couple random spatiotemporal seismic vibrations in the axial and transverse directions of the pipeline, enabling probabilistic incorporation of stochastic uncertainty (random variability) into the seismic response analysis of structures. Uncertainties of soil properties are probabilistically modeled to simulate the random pipe–soil interaction effect. Spatiotemporal dynamic response of the pipelines is discretized using extreme value distribution theory, thus making the seismic reliability analysis of the pipeline within a time-invariant framework. The applicability and accuracy of the proposed method are illustrated by numerical studies. The application of the probabilistic modeling method supports reliability- and risk-based inspection and maintenance planning for buried pipelines subjected to spatiotemporal earthquakes.

RELIABILITY BASED DESIGN OF GEOTECHNICAL SYSTEMS BASED ON A DECOUPLED RELIABILITY ANALYSIS AND OPTIMIZATION APPROACH

Reliability-based design intends to design an engineering system subject to probabilistic design constraints. This paper presents a reliability design optimization approach of geotechnical engineering problems based on a decoupled reliability analysis and optimization concept. Geotechnical and foundation systems involve considerable uncertainties in soil properties, loading conditions, and inadequate site characterization. A traditional reliability design optimization method, such as the reliability index approach, requires a double-loop optimization structure, since reliability analysis is implemented for constraint function evaluation using a numerical optimization algorithm. The decoupled approach separates the reliability analyses of the constraint functions from the design optimization loop. An adaptive metamodeling method is proposed to create approximate functions of the reliability indices, so that the reliability design optimization is performed using the approximate functions of the reliability constraints. Radial basis functions are implemented to build the metamodels. To progressively improve the accuracy of predicting the reliability indices, an adaptive sampling technique is employed, so that new samples are added using the optimal points found in previous design iterations. Numerical examples are solved and presented, and the proposed design approach works well. It is shown to be a useful alternative for solving reliability-based design of geotechnical and foundation systems.

Assessment of Slope Stability under Bounded Uncertainties Using Differential Evolution and Particle Swarm Optimization

Abstract In the assessment of slope stability, the uncertainty of soil properties affects directly the factor of safety, which is the objective function to be minimized. Therefore, we propose an approach that treats the uncertainties as additional variables besides three common variables, i.e. the radius of the critical slip surface R and its centre coordinates X, Y. This approach is also compared with two other approaches: i) minimizing the objective function (with variables R, X and Y) at all vertices of the feasible region defined by lower and upper bounds of uncertainties, and then choosing the best solution from them; ii) using the two-level optimization in which the optimal solution (R, X and Y) satisfies the least favourable condition of uncertainties at these vertices. The proposed approach with extra variables shows the efficiency in finding good solutions with a significant decrease of computation time. The study also shows that both Differential Evolution and Particle Swarm Optimization can find good solutions, although the former is marginally better.

Probabilistic analysis of subway station excavation based on BIM-RF integrated technology

In the past decade, BIM technology has developed considerably, especially in the building industry. Although BIM technology has excellent potential in promoting information and digitization technology in the construction industry, its advantages still cannot be fully utilized during subway station excavations, and engineering computations based on complex theory cannot be well performed in the BIM technology. Particularly, with the rapid development of subway engineering and considering the great uncertainties in soil properties, it is necessary to improve the engineering management and risk assessment in construction process. In this paper, a BIM-RF integrated technology for performing probabilistic analysis of BIM-based subway station excavation based on random field theory (RF) is proposed to meet practical needs. Specifically, a three-dimensional parametric modeling method in Revit software is introduced. Then, a process of probabilistic analysis combined with RF theory for subway station excavation is presented. In the BIM-RF integrated framework, the necessary information for engineering computations is decoded from the BIM model; combined with the generated RF, a probabilistic analysis based on Monte Carlo simulations is performed to provide an engineering risk assessment. The BIM-RF integrated technology enables comprehensive engineering computations to be implemented in analyses of BIM-based subway station excavation and can improve the expandability of BIM technology.