This study presents a new experimental procedure for evaluating the durability of stabilized soils subjected to multiple wetting and drying (W-D) cycles. An integrated experimental program combining dynamic shear rheometer (DSR) testing with W-D cycles was designed and implemented to assess moisture-induced performance degradation in natural sand stabilized with two types of rapid-setting cementitious stabilizers. Small cylindrical specimens (10.5 mm in diameter and 35.0 mm in height) of stabilized sand mixes were compacted, cured, and subjected to up to seven W-D cycles. Each W-D cycle was meticulously controlled to gauge its impact on the material's durability. The mechanical properties of the stabilized samples were evaluated at different stages of the W-D cycles using the strain-sweep DSR testing based on a methodology developed from preliminary work. The proposed test method focuses on the shear properties of the material, measuring its mechanical response under the torsional loading of a cylindrical sample and providing dynamic mechanical properties and fatigue-resistance characteristics of the stabilized soils under cyclic loading. Test results demonstrate water-induced deterioration of stiffness and reduced resistance to cyclic loading with good testing repeatability, efficiency, and material-specific sensitivity. By combining dynamic mechanical characterization with durability assessment, the new testing method provides a high potential as a simple, scientific, and efficient method for assessing, engineering, and developing stabilized soils, which will enable more resilient transportation infrastructure systems.

Almost all of the existing testing methods to determine elastic modulus of the soil or aggregate for pavement design involve the application of repetitive loads applied at a single point. This approach falls short of representing the conditions that are observed when the wheel of a vehicle rolls over the surface. This study presents a new methodology, in which light weight deflectometer (LWD) is used to apply three adjacent sequential loads repetitively to replicate a multipoint loading of the surface. The elastic modulus values obtained from these multipoint LWD tests were compared against the repetitive single point LWD test results. The multipoint LWD test elastic modulus values were consistently lower than the values obtained from the single point LWD tests. The single point LWD tests showed an increase in elastic modulus with increased load repetition. The multipoint LWD results did not show an increase in the elastic modulus as a function of repetitive loading. This study showed that damping ratio values provide guidance to explain differences in the elastic modulus with an increased number of load repetitions. In repetitive single point tests, the applied load caused initial compaction, and in multipoint LWD tests, it caused disturbance in the ground. With increased load cycles, the ground reached a stabilized condition in both tests. The methodology presented in this study appeared to minimize the unintended compaction of the ground during the single point LWD tests to determine the elastic modulus.

This study demonstrates the feasibility of utilizing machine learning (ML) for routine identification of sand particles. Identifying different types of sand is necessary for various geotechnical exploration projects because understanding the specific sand type plays an important role in estimating the physical and mechanical properties of the soil. To accomplish this, dynamic image analysis was employed to generate a substantial volume of sand particle images. Individual size and shape descriptors were automatically extracted from each particle image. The analysis involved use of 40,000 binary particle images representing 20 different sand types, and a corresponding six size and four shape descriptors for each particle (400,000 parameters). Six ML models were trained and tested. The work demonstrates that using size and shape features the models efficiently identified up to 49% of individual sand particles. However, when clusters of particles were considered in conjunction with a voting algorithm, classification accuracy significantly improved to 90%. Among the ML models studied, neural networks performed the best, while decision tree exhibited the lowest accuracy. Finally, the use of size consistently outperformed shape as a classification parameter but combining size and shape parameters yielded superior results across all sands and classifiers. These findings suggest that ML holds much promise for automating sand classification using ordinary images.

This study describes the full-scale traffic evaluation of a prototype submersible matting system (SUBMAT) at a test site on the U.S. Army Engineer Research and Development Center's campus in Vicksburg, MS. The SUBMAT prototype was designed to bridge the gap between high- and low-tide at a beach interface to enable 24-h vehicle offloading operations at an expeditionary watercraft landing site. This unique system is made from common geotextile materials, is filled with indigenous sand using simple commercially available pumps, and creates a robust driving surface. The results of the study showed that the SUBMAT system was able to sustain an accumulation of 1,000 Medium Tactical Vehicle Replacement, 350 Heavy Expanded Mobility Tactical Truck, and over 150 M1A1 main battle tank passes without experiencing any significant damage. The ease of deployment, relatively low cost, and trafficability results could make the SUBMAT a suitable candidate for expedient low-volume roads in austere environments such as stream beds, low-water crossings, recently flooded or flood prone areas, and areas with weak soil.

In cold regions, the soil temperature gradient and depth of frost penetration can significantly affect roadway performance because of frost heave and thaw settlement of the subgrade soils. The severity of the damage depends on the soil index properties, temperature, and availability of water. While nominal expansion occurs with the phase change from pore water to ice, heaving is derived primarily from a continuous flow of water from the vadose zone to growing ice lenses. The temperature gradient within the soil influences water migration toward the freezing front, where ice nucleates, coalesces into lenses, and grows. This study evaluates the frost heave potential of frost-susceptible soils from Iowa (IA-PC) and North Carolina (NC-BO) under different temperature gradients. One-dimensional frost heave tests were conducted with a free water supply under three different temperature gradients of 0.26 degrees C/cm, 0.52 degrees C/cm, and 0.78 degrees C/cm. Time-dependent measurements of frost penetration, water intake, and frost heave were carried out. Results of the study suggested that frost heave and water intake are functions of the temperature gradient within the soil. A lower temperature gradient of 0.26 degrees C/cm leads to the maximum total heave of 18.28 mm (IA-PC) and 38.27 mm (NC-BO) for extended periods of freezing. The maximum frost penetration rate of 16.47 mm/hour was observed for a higher temperature gradient of 0.78 degrees C/cm and soil with higher thermal diffusivity of 0.684 mm(2)/s. The results of this study can be used to validate numerical models and develop engineered solutions that prevent frost damage.

Triaxial tests has been routinely used to measure the stress-strain relationship for geomaterials. During triaxial testing, many sources of errors that cannot be completely avoided but are often ignored or approximated using empirical equations. This paper presents a systematic investigation of soil volume change, volume strain nonuniformity, and cross-sectional calculation along the specimen during triaxial testing using a photogrammetry-based method. Consolidated drained triaxial tests were performed in which two parallel measurements were taken: (1) relative volume using the conventional triaxial testing and (2) absolute volume using the photogrammetry-based method. The difference in the observed volume and void ratio measurements between the two methods highlighted the importance of absolute soil volume over the relative volume method. The results revealed the effect of soil volume change in the interpretation of triaxial testing findings. Various preparation steps and procedures during triaxial testing have influenced the initial measured volume and thus caused deviation of the subsequent associated measurements. This method would present a quality control approach for different aspects of triaxial testing to improve the test simulations to accurately measure the soil behavior. The proposed method is important for more refined simulations and identification of setup-induced errors. Additional applications of the current research would allow correct determination of the stress path followed during triaxial testing, the critical-state soil mechanics parameters, and the stress-strain relationship, including deformation and strength parameters.

This paper focuses on evaluating the increase in axial pile resistance subjected to both consolidation and aging setups. Consolidation and aging setup models were first developed to estimate the setup parameters based on databases collected from literature, which include 10 instrumented piles for consolidation setup and 26 test piles for long-term aging. The eight top-performing pile cone penetration test (CPT) methods that were evaluated in a previous study were used to estimate the side resistance of soil layers at 14 days after pile driving. The developed consolidation and aging setup models were then used to extrapolate the results to evaluate the side resistance of each soil layer at the end of consolidation and for long-term aging. The estimated side and total resistances were compared with the measurements from pile load tests considering both consolidation and aging setups. The resistances estimated before and after completion of excess pore water pressure dissipation indicates that significant aging takes place after consolidation setup. The value of consolidation setup parameter (Ac) was 0.53, and, for aging, the setup parameter (Ag) was 0.23 in clay and 0.16 in sand. The results show that all pile CPT methods with/without using a consolidation setup model tend to underestimate the unit side resistance of clay soil layers. The use of pile CPT methods in combination with an aging model improved the accuracy of pile CPT methods, and this was verified using load test results for five piles subjected to aging. The Philipponnat and University of Florida (UF) methods showed the best performance on estimating the total resistance of piles subjected to aging.

Streambed scour in cohesive sediment is complex because erosion processes depend on the physical, geochemical, and biological properties of the sediment. The scouring processes can also be characterized as a slow fatigue phenomenon. Therefore, repetitive hydraulic loadings from multiple stormflow events are likely necessary for equilibrium scour depths to develop in cohesive sediment compared with non-cohesive sediment. Cumulative effective stream power, which is a surrogate measure of effective stream power duration, showed a significant relation with scour development and propagation in cohesive sediments around bridge piers, where results from this study identified a statistically significant correlation between cumulative effective stream power and the observed scour depths around different bridge piers (R-2 = 0.56, p < 0.001). However, some localized and site-specific variations were observed. It was also observed that scour depth development in cohesive soil appeared to be dependent on effective shear duration, rather than the number of flow events above erosion threshold values. In addition, the relationship between an erodibility index (K) and critical stream power showed a significant statistical correlation (R-2 = 0.61, p = 0.017). Results from this study deviated from the Annandale empirical relationship for sediments when K < 0.1. This finding supports that site-specific critical stream power should be measured using an empirical relationship for cohesive bed sediments to predict scour depths.

This paper presents a case study on instrumenting, monitoring, and finite element modeling (FEM) of geosynthetic-reinforced pile-supported (GRPS) mechanically stabilized earth (MSE) walls. The GRPS-MSE wall was monitored using various instruments such as piezometers, earth pressure cells, shape-acceleration arrays (SAAs), and strain gauges. The performance criteria included efficacy, stress concentration ratio (SCR), differential settlement, and reinforcement tension. Collected data, such as excess pore-water pressures, contact pressures on pile and soft soil, differential settlements, and lateral displacement of MSE wall, were analyzed thoroughly. A 3D FEM was also developed to simulate the GRPS MSE wall, and the results are in good agreement with field data. The results demonstrated significant load transfer from soil to piles as a result of soil arching, yielding 30-32 SCR. The field efficacy was measured at 37.69 %, while the FEM efficacy was estimated as 42.4. Strains in geogrids within the geosynthetic-reinforced load transfer platform (GLTP) system were under 1%, less than the 5% maximum recommended by FHWA. The maximum differential settlement measured between pile cap and soft soil from SAAs is 7.1 mm, while it is estimated to be 8.3 mm from FEM. The MSE wall exhibited low lateral displacement (<25 mm), indicating enhanced stability because of GLTP. The comparison between five analytical GLTP design methods showed that the CUR226 methods gave the closest results to field measurements and FEM results. This study offers crucial insights into leveraging GLTP and MSE walls in highway construction.

The plastic mold compaction device (PM Device) was developed in Mississippi to compact cementitiously stabilized soil inside plastic molds to improve soil-cement quality by adding value during design and construction activities. The PM Device has been incorporated as an AASHTO provisional standard (AASHTO PP92-19) and, to date, prevailing activities have been Mississippi Department of Transportation projects and have included controlled laboratory evaluations and field projects. This paper goes beyond previous efforts, to document a field study in which the PM Device was successfully used on an Alabama Department of Transportation project to evaluate its effectiveness within another state's construction specifications. The PM Device was capable of capturing quantifiable variation in mechanical properties over the duration of the construction project, as well as producing similar mechanical properties to cores taken from the compacted pavement surface. Additionally, molds described in AASHTO PP92-19 were compared with one another to evaluate their potential within the standard practice. AASHTO PP92-19 protocols were sufficient to produce viable, repeatable test specimens within another state department of transportation construction environment for quality control and quality assurance.