Despite clear evidence, changes in mechanical properties (i.e., stiffness or modulus) of compacted subgrades in response to subgrade moisture regime changes after construction have rarely been investigated in the geotechnical profession. In particular, when in-service assessment of pavement subgrade is made, the modulus-moisture variation should be addressed on the basis of unsaturated soil mechanics. This study presents the unsaturated small-strain modulus behavior of five predominately fine-grained compacted subgrade soils. The small-strain shear modulus ( Go ) of saturated compacted specimens subjected to a desorption soil-water characteristic curve (SWCC) was evaluated using bender elements. A test apparatus was designed to apply two stress state variables, the net confining pressure and matric suction, during the Go measurements. The relationship between Go and the SWCC under a constant mean net stress was developed. Additionally, the effect of compaction moisture content, compaction energy, and soil...

Hydraulic properties of soils used for water balance covers measured at the time of construction and one to four years after construction are compared to assess how the hydraulic properties of cover soils change over time as a result of exposure to field conditions. Data are evaluated from ten field sites in the United States that represent a broad range of environmental conditions. The comparison shows that the saturated hydraulic conductivity ( Ks ) can increase by a factor of 10,000, saturated volumetric water content ( θs ) by a factor of 2.0, van Genuchten’s α parameter by a factor of 100, and van Genuchten’s n parameter can decrease by a factor of 1.4. Larger changes occur for denser or more plastic fine-textured soils that have lower as-built Ks , α , and θs and higher as-built n , resulting in a reduction in the variation in hydraulic properties that can be attributed to compaction. After two to four years, many water balance cover soils can be assumed to have Ks between 10−5 and 10−3 cm∕s , θs be...

The effect of median particle size and breadth of particle sizes in clean sands on the SWCC and the van Genuchten parameters α and n is described. For uniform sands, the air entry suction increases (α decreases) and the steepness diminishes slightly (n decreases) for the drying SWCC as the particles become finer. Increasing the breadth of particle sizes also increases the air entry suction (lower α) and reduces the steepness (lower n) of the SWCC. Breadth of the particle size distribution has a larger effect on steepness and n of the SWCC, and median particle size has a larger effect on the air entry suction and α. Similar effects occur for the wetting SWCC, although the impact on the water entry suction is less than the impact on the air entry suction. The systematic effects of median particle size and breadth of particle sizes on the SWCC are captured in a pedotransfer function (PTF) to estimate α and n for clean sand based on d₆₀ and C_u. Comparison with other SWCC data shows that α estimated with the PTF typically is within a factor of 2 of the actual α and n estimated with the PTF is within ± 2 of the actual n. These findings are applicable to α for sands with nonplastic fines provided the fines content is less than 20%, but apply to n only for sands with negligible fines content. A procedure to convert between van Genuchten and Brooks-Corey parameters for sands is also presented.

Road maintenance is essential for supporting road safety and user comfort. Developing predictive models for road surface conditions enables highway agencies to optimize maintenance planning and strategies. The international roughness index (IRI) is widely used as a standard for evaluating road surface quality. This study compares the performance of deep neural networks (DNNs) and graph convolutional networks (GCNs) in predicting IRI values. A unique aspect of this research is the inclusion of additional predictor features, such as the type and timing of recent roadwork, hypothesized to affect IRI values. Findings indicate that, overall, the DNN model performs similarly to the GCN model across the entire highway network. Given the predominantly linear structure of national highways and their limited connectivity, the dataset exhibits a low beta index, ranging from 0.5 to 0.75. Additionally, gaps in IRI data collection and discontinuities in certain highway segments present challenges for modeling spatial dependencies. The performance of DNN and GCN models was assessed across the network, with results indicating that DNN outperforms GCN when highway networks are sparsely connected. This research underscores the suitability of DNN for low-connectivity networks like highways, while also highlighting the potential of GCNs in more densely connected settings.