A Regional Study on Champlain Sea Clay: Empirical Prediction of Consolidation Parameters and Statistical Characterization of Undrained Shear Strength Based on Historical Geotechnical Data

This thesis explores the geotechnical challenges associated with the performance of geo-infrastructures on Champlain Sea Clay (CSC) deposits, with a particular focus on the Ottawa Valley region of Eastern Ontario. CSCs are highly compressible sensitive marine clays, which pose serious risks both with respect to deformation as well as short-term stability behavior. Geotechnical infrastructures placed on CSC are prone to excessive settlement, differential movement, and cracking, often exceeding acceptable serviceability limits. CSCs are also sensitive in nature due to high-water content, low coefficient of permeability and pore-water salinity and typically exhibit strain-softening behavior. In many scenarios, significant post-peak shear strength loss can trigger progressive failures. There are conventional methods and recently developed advanced techniques that can be used for reliably determining and interpreting the behavior of CSCs. For example, settlement behavior can be estimated from consolidation tests performed in the laboratory testing using high-quality undisturbed soil samples. Undrained shear strength, S_u, of clay is one of the key parameters required in the ultimate limit state (ULS) design of geotechnical structures for short-term stability analyses, which can be determined from vane shear tests (VST). In addition, VST also provides valuable information about the sensitivity of CSCs. In recent years, there have also been significant efforts to develop correlations using Cone Penetration Testing (CPT) and CPT with pore pressure measurements (CPTu) for estimating and interpreting the behavior of CSCs. However, these methods are time-consuming and expensive for extending them in low budget geotechnical projects. During the last two decades there has been rapid urban expansion, particularly in the CSCs of Ottawa Valley region. Many new subdivisions have been developed by converting forest and agricultural lands into residential areas. As per the provincial housing mandate, Ottawa's municipal housing target, presently, is approximately 15,000 homes per year. In low-budget projects, key consolidation parameters are typically estimated using empirical correlations developed from fundamental soil index properties reported in the literature. However, these empirical relationships often lack the regional specificity associated with key factors such as stress history, pore water salinity, and soil fabric needed to reliably reflect the complex behavior of CSCs. For this reason, the key objective of this thesis is directed towards refining existing correlations and methodologies to better suit site-specific conditions of CSCs in Ottawa Valley region. For achieving this objective, over 500 geotechnical reports, including data from the Ontario Ministry of Transportation, were compiled to develop two databases: one for consolidation parameters and another for undrained shear strength, S_u. The datasets related to consolidation and other relevant tests were statistically cleaned using interquartile range, skewness, kurtosis, and normalization techniques to remove outliers and enhance reliability. Using the cleaned data, empirical models were developed for key consolidation parameters; compression index (C_c), recompression index (C_r), overconsolidation ratio (OCR), and preconsolidation pressure (σ_P^') that are tailored for eastern and western subregions of the Ottawa Valley. The field data from VSTs were compiled into a separate database to analyze the spatial and depth-related variability of S_u across the Ottawa Valley. The collected dataset did not provide a strong correlation between S_u and other soil properties, despite such correlations being used for other fine-grained soils in the literature. For this reason, the focus of the analysis on the S_u database was put on quantifying the Coefficient of Variation (COV) for the geotechnical variability of S_u across the Ottawa Valley region. The COV presented in this thesis provides valuable insights that can be used in probabilistic modeling and reliability-based design. In addition, sensitivity (S_t) and its relationship to other soil index properties were investigated. Strong correlations were not obtained as the investigations revealed pronounced heterogeneity due to depositional and post-depositional processes. Such trends may partly be attributed to variations in field sampling practices across the region over several decades of data collection. The primary challenge of the research undertaken in this thesis lies in the fact that the data set was compiled by numerous operators extending different practices over a period exceeding fifty years. This introduces potential vulnerabilities in the consolidation sampling and testing process and measurement of S_u resulting in the development of unreliable correlations for future use. Nonetheless, assembling a dataset of comparable size from a single source of high quality is both expensive and a practical challenge. The approach extended in this study represents a reasonable option for addressing the challenges. In summary, the research summarized in this thesis contributes to a statistically robust and regionally tailored framework for estimating several key geotechnical properties of CSCs, providing a practical alternative to conventional testing methods. The novelty of this study lies in compiling and analyzing a dataset specific to the Ottawa Valley region, treating it as a micro-zone within the broader CSC formation and proposing better empirical correlations for use in low budget geotechnical engineering projects.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.