A soil’s physical properties, mineral types, and pore structure significantly influence the shape and properties of the soil-water characteristic curve (SWCC). This study investigated the effects of the soil’s physical properties and mineral types on the SWCC and pore-size distribution (PSD). Eight different soils from an alluvial deposit in Istanbul and Adapazarı/Türkiye were used in the study. The test samples were prepared by compaction at optimum water content (OWC) and wet side of optimum water content (wet of OWC). The samples were prepared by consolidation from the slurry. The PSDs of the samples were calculated using the SWCCs and evaluated with scanning electron microscope (SEM) analysis. In addition, the mineral types of all soils were determined by X-ray diffraction analysis. The soil which contains illite-type minerals has higher matric suction than containing kaolin-type. The effect of the clay percentage is more pronounced in silty soils than in plasticity and activity. Soil suction increased with decreasing compaction water content in clayey soils. The air entry water contents rose as the void ratio, liquid limit, clay content, and plasticity increased. The compaction conditions affected the macropore structure more than the micropore structure. In addition, the ratio of macro-micro pore sizes increased with the rise of the compaction water content.

The focus of this study is to examine the soil properties and liquefaction potential of the Erenler center district in a critical tectonic zone that can produce large earthquakes, such as the North Anatolian Fault Zone (NAFZ). In this sense, 40 surface wave measurements and 52 geotechnical drillings were employed. Accordingly, the liquefaction potential index (LPI), liquefaction severity index (LSI), Ishihara boundary (IB) curve, and Ishihara-inspired index (LPI_ISH) liquefaction approaches from geotechnical and geophysical methods were used as integrated. All liquefaction analyses were examined for two scenarios, Mw: 7.4 1999 Izmit (a_max: 0.41 g) and Mw: 7.0 1967 (a_max: 0.28 g) Mudurnu. According to the analyses, almost all of the study area showed liquefaction risk in the Izmit scenario. In the Mudurnu scenario, liquefaction risk distribution decreased parallel to acceleration. The LPI, LPI_ISH, and IB liquefaction risk results for both scenarios support each other. On the other hand, it was determined that the probability of liquefaction was lower in the LSI evaluation. The spatial distribution of the liquefaction potential of the Vs-based and SPT-based LPI assessments had a similar pattern. These results show that the LPI approach, originally SPT-based, can also be calculated based on Vs. In addition, according to Vs₃₀-based (average shear wave velocity at 30 m depth) soil classification criteria, low-velocity character E and D soil groups dominate the Quaternary alluvial basin. This indicates that deformation-induced failures may occur even in areas with a low probability of liquefaction in similar strong ground motions. In addition, the evaluation of liquefaction with many approaches in two different bases within the scope of the study constitutes a novelty for the study area and liquefaction analyses. While performing diversified liquefaction analyses and approaches will contribute to obtaining more reliable soil liquefaction results, more case studies are needed to elucidate these comparisons.

The accurate prediction of ultimate settlement is of critical importance for the longevity of structures, as it helps prevent structural damage and functional deterioration. Settlement, a function of the soil’s compression index ( C _c ), is typically determined through oedometer tests. This research seeks to enhance the prediction accuracy of C _c within a sustainability framework by employing advanced machine learning techniques. A series of soil samples with varying clay contents were synthesized by incorporating bentonite into silt, resulting in nine distinct soil classes ranging from low plasticity silt (ML) to high plasticity clay (CH). Consolidation tests were performed on reconstituted samples subjected to 100 kPa pre-consolidation pressure, revealing significant correlations between the compression index and various physical properties. Leveraging these correlations, a comprehensive dataset was compiled, integrating current experimental data with relevant studies from the literature. The predictive capabilities of several machine learning models, including Support Vector Regression (SVR), Artificial Neural Networks (ANN), Random Forest (RF), and Multiple Regression Analysis (MRA), were evaluated. Among these, the RF model exhibited superior performance, achieving an R ² of 0.83. The study introduces a novel prediction framework that extends beyond conventional settlement estimation by incorporating a comparative ML-based modeling approach supported by both experimental and literature-based data. Furthermore, the methodology demonstrates its practical value by offering a faster and less resource-dependent alternative to traditional consolidation testing, contributing to sustainability oriented and cost-efficient geotechnical design.