Masonry arch bridges are characterised by three-dimensional (3D) behaviour when subjected to external eccentric loading (e.g., vehicle loads). The arch ring, abutments, backfill and spandrel walls may interact with each other in a complex manner, leading to a 3D mode of response that can have a significant impact on the initiation and propagation of damage. However, there is a dearth of experimental data from tests designed to investigate the 3D behaviour of masonry arch bridges, particularly under loading levels below those required to cause failure. This paper presents results from tests on a large-scale brickwork masonry arch bridge subjected to low- and mid-level static loads under laboratory conditions. Point loads of increasing magnitude were applied at different locations on the top of the backfill in order to investigate 3D response and damage accumulation. Details of the experimental setup, material characterisation, and the results obtained from static and repeated load tests at low- and mid-level load magnitudes are presented herein. Results demonstrate that the bridge exhibited a 3D mode of response under eccentric point loads. Loading at the mid-span resulted in greater deformation of the arch barrel compared to loading at the quarter- and three-quarter-span points, due to the shallower backfill depth over the crown. Under the mid-level loading, stiffness degradation was observed during the testing regime, suggesting an accumulation of damage in the bridge. Moreover, when loading was applied close to a spandrel wall, measurable out-of-plane deformation of the spandrel wall was observed, with this deformation increasing significantly as the load was increased from 150 kN to 250 kN. This results from a combination of increased lateral soil pressure and decreased shear resistance at the arch-spandrel wall interface.

In-service masonry arch road bridges, mainly realised before the first half of the last century, represent a wide portion of the entire worldwide infrastructural asset. Given their age, during their service life these structures could have experienced damage due to anthropic (i.e. traffic) and natural (i.e. earthquakes, soil settlements, degradation, etc.) actions which may have inevitably affected their load-bearing capacity. The present study addresses the problem of the residual capacity estimation of damaged bridges by investigating the impact of previous loading on the actual strength of the structure. In particular, reference to a past experimental activity retrieved from the literature on reduced-scale bridges subjected to concentrated vertical loads has been made to calibrate a reliable detailed finite element model in Abaqus software. Then, damage of different extent has been introduced by simulating the transit of vehicles of various weights on the structure and the residual capacities of the bridge have been assessed and compared against the undamaged configuration. The results confirm that preexisting damage due to traffic loading may significantly influence the capacity of such structures, with peak load reductions up to 60% estimated through the proposed methodology.