Morphology-driven interfacial shear behavior of cement-stabilized macadam asphalt composite pavements based on in-situ surface reconstruction

Interfacial shear failure is a critical distress mode in cement-stabilized macadam (CSM) base–asphalt surface layer systems, but laboratory evaluation remains challenging because construction-induced base-course morphology is difficult to reproduce. This study proposes a quantitative field-to-laboratory framework for repeatable evaluation of morphology-driven interfacial shear behavior. Field CSM base surfaces were digitized using high-resolution three-dimensional scanning, and ISO 25178 areal texture parameters were extracted to characterize interface morphology. Principal component analysis was used to construct a Roughness Composite Index for selecting representative smooth, medium-roughness, and rough morphology states. Negative molds were then fabricated, and the fidelity of the reconstructed interfaces was verified before direct shear testing. Tests conducted at −10°C, 20°C, and 50°C showed pronounced temperature dependence: peak shear strength decreased by approximately 93.7% for smooth interfaces and 93.5% for rough interfaces from −10°C to 50°C. Within the tested three-template system, specimens reconstructed from the selected rough boundary template exhibited mean peak shear strengths 33.8–42.0% higher than those reconstructed from the selected smooth boundary template, whereas the medium-roughness template produced intermediate mean responses. The observed ordering was consistent with increased asperity engagement and mechanical interlocking within the tested morphology range. Peak material volume (Vmp), core roughness depth (Sk), and ten-point height (S10z) were identified as shear-relevant morphology descriptors. These descriptors were used to develop specimen-level, locally calibrated temperature-specific empirical strength-estimation relationships, with adjusted R 2 values of 0.792–0.864 within the tested three-template material system and temperature range. The proposed procedure provides a practical basis for linking field-derived interface morphology with laboratory-measured monotonic interlayer shear resistance within the tested morphology and experimental ranges.