Documents published in Scipedia

• Applicability of CPTU to characterize diatomaceous fine-grained soils: a case study in Euganean Hills (Italy)

  G. Dalla Santa*, F. Ceccato, P. Simonini

  ISC2024.

  
  

  Abstract

  Diatomaceous soils, composed of diatom fossils and clay minerals typically formed in volcanic environments, exhibit characteristics such as low unit weight, high plasticity and liquid limit, significant compressibility, and high friction angles. Despite their presence in various locations globally, knowledge about their geotechnical behavior is limited and primarily based on laboratory tests conducted on artificial samples. This paper presents data obtained from undisturbed samples of natural diatomaceous soils and discusses the interpretation of Cone Penetration Test with Pore Pressure (CPTU) data to classify these complex non-textbook soils and estimate their mechanical properties. The study area is situated in the Po Plain near the thermal anomaly region of the Euganean Hills in Northeast Italy. Three CPTUs and one borehole with the collection of four Osterberg undisturbed samples were conducted. Laboratory tests on the undisturbed samples provided values for Atterberg Limits, soil unit weight, in-situ void ratio, compressibility, and permeability, which were compared with estimations derived from CPTU data analysis. Moreover, Scanning Electron Microscope images provided insight into the distinctive microstructure of diatom microfossils embedded in a clayey matrix. Based on these comparisons, CPTU proves to be effective in estimating relevant parameters of diatomaceous soils, particularly the Soil Behavior Type (SBT) and consolidation coefficient from dissipation tests. However, the agreement in estimating the oedometric modulus is less satisfactory. Therefore, for a precise definition of the geotechnical model, it is recommended to conduct additional laboratory tests, particularly those focused on defining compressibility parameters, given the unique behavior of natural diatomaceous soils.

  Abstract
  Diatomaceous soils, composed of diatom fossils and clay minerals typically formed in volcanic environments, exhibit characteristics such as low unit weight, high plasticity [...]

  • 25
  •  read

• Soil Characterisation Using a Dynamic Penetrometer

  J. Rong*, M. Nazem

  ISC2024.

  
  

  Abstract

  In offshore engineering, a geotechnical site investigation is an important step in analysis and design to ensure the integrity and serviceability of infrastructure. The Cone Penetration Test (CPT) stands as the prevailing technology for offshore soil characterisation. However, this test method requires a substantial allocation of resources for equipment transportation and operation personnel. This proves inefficient and costly for conducting comprehensive surveys over ocean beds. Alternatively, free-falling penetrometers (FFP) have attracted attention as a CPT replacement for soil characterisation. Nevertheless, these devices can penetrate only to shallow depths within soils, limiting their applicability for offshore site investigation purposes. A new device has been created to overcome this constraint, featuring a dynamic penetrometer launched by a speargun. Unlike conventional free-falling penetrometers, this apparatus can attain a greater penetration ratio, exceeding 20 times its diameter. The process of experimental testing yielded notable enhancements, particularly in effectively addressing challenges associated with tilting when attempting low-penetration depths. By implementing rate corrections into the methodology, promising results were obtained for equivalent static penetration resistance. This approach not only represents the capacity to influence future penetrometer designs but elevates the overall efficiency of in-situ soil characterisation procedures.

  Abstract
  In offshore engineering, a geotechnical site investigation is an important step in analysis and design to ensure the integrity and serviceability of infrastructure. The Cone [...]

  • 60
  •  read

• Machine Learning-Based Modeling of Net Ecosystem Exchange Using Numerical Weather Data and Satellite Images

  N. Kim*, J. Cho, Y. Lee

  ISC2024.

  
  

  Abstract

  Recently, the increasing severity of climate change attributable to global warming has emphasized the imperative of carbon absorption to mitigate greenhouse gas emissions. The use of the carbon sink based on the carbon absorption and storage functions of forests is suggested as an effective alternative for domestic greenhouse gas reduction. Additionally, agricultural land cover comprises approximately 38% of the Earth's surface, underscoring the importance of comprehensively understanding the carbon cycle within not only forests but also agricultural landscapes. This significance arises from the fact that agricultural land locally amplifies seasonal variations in carbon dioxide by approximately 25% compared to vegetated areas. Consequently, a comprehensive understanding of both forest and agricultural land carbon cycles is imperative, necessitating quantitative analysis of carbon uptake in agricultural settings. Thus, this study aims to construct a machine learning-based model for estimating the net ecosystem exchange (NEE) of rice paddies in South Korea using ground flux data, meteorological variables, and satellite images. Through quantitative assessment, the NEE was determined, with a mean absolute error of 1.387, root mean square error of 2.203, and correlation coefficient of 0.872. Notably, observed NEE values demonstrating extremes in magnitude were associated with calculation errors, reflecting tendencies of both underestimation and overestimation. This phenomenon is likely attributed to the study's reliance on a limited dataset and the inherent challenges of training models across a broad spectrum of observations. To enhance calculation accuracy, future endeavors should focus on accumulating a more extensive repository of NEE flux observations and leveraging high-resolution satellite imagery and meteorological datasets for refining machine learningbased models.

  Abstract
  Recently, the increasing severity of climate change attributable to global warming has emphasized the imperative of carbon absorption to mitigate greenhouse gas emissions. [...]

  • 6
  •  read

• AI Modeling for Characterization of Paddy Rice Yields under Extreme Weather Conditions Using Remote Sensing and Geospatial Data

  J. Ahn*, Y. Lee

  ISC2024.

  
  

  Abstract

  Climate change, such as increase in CO2 levels and rising temperatures, can have a significant impact on paddy rice production and increase the uncertainty of yield forecasts. This study aims to employ AI modeling for forecasting paddy rice yield and present the findings of a quantitative analysis to determine its ability to generate stable forecasts under extreme weather conditions, such as heatwaves, low temperatures, and heavy rainfall. Vegetation growth indices from the Moderate Resolution Imaging Spectroradiometer (MODIS) satellite product were utilized. These indices include the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Leaf Area Index (LAI), Fraction of Absorbed Photosynthetically Active Radiation (FPAR), and Near-Infrared Reflectance of vegetation (NIRv). Meteorological variables such as downward solar radiation flux, daily temperature difference, precipitation, relative humidity, and temperature were also used. Over 23 years of experimentation (2000-2022), yields under extreme weather conditions did not exhibit a significant difference from the normal period, with a Mean Absolute Error (MAE) ranging from 0.30 to 0.33 ton/ha, representing a 4-5% error of the average yield. This study presents an AI modeling methodology that enables stable predictions of paddy rice yields, even under extreme weather conditions. Future work should focus on refining input data and optimizing the model by analyzing cases of extreme weather.

  Abstract
  Climate change, such as increase in CO2 levels and rising temperatures, can have a significant impact on paddy rice production and increase the uncertainty of yield forecasts. [...]

  • 2
  •  read

• Taillings Characterization: Exploring Laser Granulometry with Machine Learning

  I. Bernardes, G. Frainz*

  ISC2024.

  
  

  Abstract

  The conventional particle size test has been a widely used method in the characterization of soils and tailings. Such information is particularly useful in the evaluation of materials deposited in tailing stacks or compacted landfills, which must follow reference particle size ranges. However, the method has limitations, the main one being the execution time, which usually lasts around three days. On the other hand, laser testing appears as a viable alternative. This innovative method obtains the grain size curve of the soil through the light dispersion pattern and lasts a few minutes, a significant improvement over the conventional method. Furthermore, this method can cover particle size ranges of up to 0.1 micrometers, while the conventional method is limited to 1 micrometer. Despite the benefits of using this equipment, the laser grain size test does not yet have specific standardization for use in the field of soil mechanics. In this context, this work proposes the use of machine learning techniques to demonstrate the existence of compatibility between both methods. To this end, tests were carried out using both methodologies on different samples of iron ore tailings and an algorithm was developed to predict the material classification. The evaluation of the results made it possible to verify the consistency and precision of the results between the two methods, reinforcing the reliability and viability of the laser test as an efficient alternative to the traditional method

  Abstract
  The conventional particle size test has been a widely used method in the characterization of soils and tailings. Such information is particularly useful in the evaluation [...]

  • 8
  •  read

• Integrated geophysical methods in identifying preferential flow paths in an earth dam

  L. Rezende, C. Aguiar*, L. Soares, C. Lemos, L. Dias

  ISC2024.

  
  

  Abstract

  The evaluation of safety conditions in dams is of utmost importance to ensure stability and often involves subsurface investigation methods. Geophysical methods have emerged as a modern and relevant alternative, often more practical than traditional direct methods. This study aims to integrate the application and interpretation of resistivity and selfpotential methods to identify preferential flow paths in a small earth dam. The investigation was conducted at a dam located on the Viçosa Campus of the Federal University of Viçosa (UFV), with three main soil layers: embankment, silty clay, and alluvium. Analysis of the results revealed potential conductive zones and negative spontaneous potential anomalies, suggesting the occurrence of piping and the presence of buried structures in the spillway area. Moreover, the geophysical investigation methodology proved effective in evaluating geotechnical characteristics and flow conditions of the dam, contributing to the foundation for future safety and stability analyses of the structure.

  Abstract
  The evaluation of safety conditions in dams is of utmost importance to ensure stability and often involves subsurface investigation methods. Geophysical methods have emerged [...]

  • 4
  •  read

• Data-driven Multi-stage Sampling Strategy for Machine Learning of Underground Digital Twins Consdiering Stratigraphic Uncertainty

  C. Shi*, Y. Wang, V. Kamchoom

  ISC2024.

  
  

  Abstract

  A sound understanding of subsurface geological conditions is crucial for the digitalisation of underground infrastructure. The building and updating of underground digital twins heavily rely on sparse geotechnical measurements (e.g., boreholes) retrieved from the ground, and an efficient sampling strategy can facilitate the interpretation of subsurface heterogeneities. Geotechnical sampling design can be viewed as a constrained optimization process that aims to obtain as much geological information as possible from a limited number of sampling locations within a given site boundary. In this study, a data-driven intelligent sampling strategy is proposed to optimize borehole locations for a multi-stage site investigation of a three-dimensional (3D) geological domain. The initial sampling plan is determined using weighted centroidal Voronoi tessellation, which assigns uniform sampling densities to zones of different importance. Measurements obtained from the initial stage are combined with prior geological knowledge to build underground digital twins using an image-based stochastic modelling method. Multiple realizations of the geological domain can be developed under the framework of Monte Carlo simulation, and stratigraphic uncertainties associated with multiple random realizations can be quantified using information entropy. The location with the maximum entropy is adaptively selected as the next optimal sampling location. The proposed method is the first sampling strategy that can explicitly consider 3D subsurface stratigraphic variations. The performance of the proposed multi-stage sampling strategy is demonstrated using a simulation example. Results indicate that the proposed method can efficiently identify the optimal sampling locations while accounting for irregular site geometries and 3D subsurface stratigraphic uncertainties.

  Abstract
  A sound understanding of subsurface geological conditions is crucial for the digitalisation of underground infrastructure. The building and updating of underground digital [...]

  • 7
  •  read

• Predicting Soil Behavior Types Along the Danube: An AI-Driven Approach Using CPT Data in the Szigetköz Floodplain Area

  E. Koch*, S. Alsamia

  ISC2024.

  
  

  Abstract

  The Szigetköz (Hungary) is a hotbed of sand boil formation, owing to the combination of a 100-250 m thick gravel layer beneath a relatively thin covering of poor soil with varying thickness. Soil behavior is critical for flood protection in this region. This work proposes a novel way to predict Soil Behaviour Types (SBT) based on detailed CPT data collected from 29 sites in the Szigetköz area using an artificial intelligence (AI) model. The study follows a methodically planned approach that includes data collecting, preprocessing, SBT categorization based on the SBT chart developed by Robertson et al. (1986), and AI model building. The CPT dataset contains critical metrics like cone resistance and friction ratio, which are essential in characterising soil behavior. The AI model, built with powerful machine learning algorithms, is intended to learn complicated associations within data to forecast SBT classifications. Extensive feature selection, hyperparameter tuning, and cross-validation are all necessary steps in model construction to ensure accuracy and generalizability. The results show that the model can accurately forecast SBT classifications for the Szigetköz area, shedding information on the soil's behavior near the Danube River. Spatial distribution visualizations emphasize the region's many SBT categories, giving valuable information for engineering projects, land use planning, and environmental conservation activities. The AI model's interpretability elucidates the major CPT parameters driving SBT forecasts, providing stakeholders with actionable information for decision-making. Furthermore, validation of the model with new, previously unseen CPT data confirms its applicability and robustness in real-world circumstances.

  Abstract
  The Szigetköz (Hungary) is a hotbed of sand boil formation, owing to the combination of a 100-250 m thick gravel layer beneath a relatively thin covering of poor soil with [...]

  • 24
  •  read

• Artificial Intelligence in Dynamic Compaction for Geotechnical Site Characterization

  J. Park*, C. Lee

  ISC2024.

  
  

  Abstract

  We present a novel method using four artificial intelligence (AI) algorithms to anticipate the cumulative degree of soil compaction (CDSC) after dynamic compaction (DC). Four AI algorithms adopted in this study include support vector regression SVR, artificial neural network (ANN), random forest (RF), and gradient boosting machine (GBM). Input variables for AI algorithms involve the average SPT N-value before dynamic compaction, cumulative applied energy normalized with a cross-sectional area of tamper, and the number of the tamper drops. Apart from cross-validation with a testing set, additional in situ test data compiled from a different section within the studied site are used to estimate the generalized capacity of the AI models. In addition, we conduct out-of-distribution analyses for the four AI algorithms in view of parametric studies. The CDSC prediction performance for the four AI models results in high prediction metrics of accuracy with the r2 higher than 0.9 for the testing scenario while the r2 of the other AI models is more than 0.9 when out-of-sample data are considered except for the GBM. The ANN seems to be the best model as the parametric study considers out-of-distribution data and suggests a strong relationship between input variables and CDSC that is more coherent with engineering principles for DC. Finally, the ANN model can be utilized to develop a mathematical model for CDSC prediction.

  Abstract
  We present a novel method using four artificial intelligence (AI) algorithms to anticipate the cumulative degree of soil compaction (CDSC) after dynamic compaction (DC). Four [...]

  • 23
  •  read

• Integrating laboratory and geophysical data considering measurement errors

  J. Xie, J. Huang*, S. Huang

  ISC2024.

  
  

  Abstract

  Laboratory and geophysical tests are commonly used in site characterization. Combining these data sets based on empirical relationships can essentially enhance data interpretation. While in traditional approaches, the uncertainties in the relationship between these data sets are ignored. The Bayesian updating method is used to consider these uncertainties. Besides, the uncertainties due to measurement errors in the laboratory tests, particularly for preconsolidation pressure, are considered based on the kriging fitting method. The outcomes of kriging fitting are utilized to establish the prior distribution, and these outcomes are then compared against the baseline established by the trend fitting method. The Markov chain Monte Carlo (MCMC) algorithm is applied to incorporate the shear wave velocity measurements from a seismic dilatometer test to derive the posterior distribution. Bayesian updating of parameters considering measurement errors is able to get a more convincing design profile.

  Abstract
  Laboratory and geophysical tests are commonly used in site characterization. Combining these data sets based on empirical relationships can essentially enhance data interpretation. [...]

  • 16
  •  read