The Cathedral of Ica, Peru, is one of the four prototype buildings involved in the ongoing Seismic Retrofitting Project, initiative of the Getty Conservation Institute. The complex historical building, which was heavily damaged by earthquakes in 2007 and 2009, can be divided into two substructures: an external masonry envelope and an internal timber frame built by a construction method known as quincha technique. This study makes use of the information available in literature and the results obtained from experimental campaigns performed by Pontificia Universidad Católica del Perú and University of Minho. Nonlinear behaviour of masonry is simulated in the numerical models by considering specified compressive and tensile softening behaviour, while isotropic homogeneous and linear behaviour is adopted for modelling timber with appropriate assumptions on the connections. A single representative bay was initially studied by performing linear elastic analysis and verifying the compliance with the various criteria specified by the applicable normative to discuss the actual failure of Ica Cathedral. Afterwards, the structural behaviour of the two substructures composing the Cathedral is evaluated independently. Finally, the interaction of these two substructures is investigated by performing structural analysis on the entire structure of Ica Cathedral. Several structural analysis techniques, including eigenvalue, nonlinear static and dynamic analyses, are performed in order to: (1) evaluate the dominant mode shapes of the structure; (2) validate the numerical models by reproducing the structural damage observed in situ; (3) estimate the structural performance; and (4) identify the main failure mechanisms.@en

This paper describes a simplified methodology for the assessment of unreinforced masonry (URM) walls under out-of-plane two-way bending seismic action. The methodology involves a force-based check derived from the principle of virtual work. This check is proposed based on experimental observations of significant cracking resistance associated with two-way spanning URM walls, indicating methodologies considering such walls to be pre-cracked or to be non-laterally supported as overly conservative. The methodology incorporates several findings and developments from recent experimental campaigns: ranging from novel characterization tests on masonry couplets to incremental dynamic tests on full-scale buildings. Such incorporations include new formulation to calculate the torsional shear strength of a bed joint and accounting for possible changes in the boundary conditions of an OOP wall during dynamic loading. Testing standards as well as recommendations in several international guidelines for masonry structures addressing the input properties required to implement the proposed methodology are enlisted and reviewed. The methodology requires the definition of the period of vibration of the assessed URM walls, to calculate which plate theory based formulation is provided. Open research questions and potential avenues for further development of the methodology are ultimately highlighted.@en

Historical quay walls constructed in unreinforced masonry are integral elements of many cities. Originally designed as gravity retaining walls, they are nowadays often subjected to the action of traffic loads as a result of vehicles travelling on roads constructed on their backfill. This paper presents a numerical analysis procedure for carrying out the structural assessment of quay walls under traffic loads. The procedure simulates the non-linear dynamic response of the quay wall under the effect of the passage of a vehicle. Non-linear dynamic calculations are performed not only to be representative of the actual nature of loading but also to produce realistic estimations of structural safety, load redistribution capacities and displacements. Adopting a tier-based approach, the computational burden typically associated with such simulations is significantly reduced. This is obtained by adopting simplifications which allow for the modelling the 3D soil block comprising the backfill of the quay wall only in the first tier of the procedure. To demonstrate the implementation of the procedure, a detailed application to an existing quay wall in Amsterdam, the Netherlands, is presented. Different foundation damage scenarios are also considered. Though the procedure is presented in this paper for a specific typology of quay walls, it has conceptual and methodological value. With appropriate modifications it can be used for the structural assessment of other earth retaining structures as well, under the effect of vehicular traffic on their backfills.@en

Canals delimited by masonry quay walls are integral elements of many cities in the Netherlands. Historically built to enable the efficient transportation of goods, today such infrastructure also gives the cities their historical and monumental character. In recent years, many quay walls in the Netherlands have shown substantial deformation and damage, and in few cases even collapse. Historical quay walls, which are constructed in thick multi-wythe unreinforced brick masonry and are supported on a system of timber piles, nowadays sustain traffic loads larger than the one they were designed for. Instances of collapse and severe damage has given rise to a need for research assessing the safety of these structures, which are not appropriately covered by any normative or standardised guidelines. This paper presents a novel methodology to numerically assess the performance of masonry walls in historical quays under the dynamic effect of traffic loads. Application of the proposed methodology to a case study in Amsterdam, the Netherlands, is presented.@en

Out-of-plane (OOP) mechanisms are one of the major causes of structural collapse in unreinforced masonry (URM) buildings as observed in recent as well as past seismic events. Among such failures, two-way bending mechanisms involving at least one restrained vertical edge constitute a very distinct majority. Nevertheless, very little research pertaining to such mechanisms can be found currently in literature. The paper takes a step forward in this topic through the execution and interpretation of dynamic tests on four full-scale single leaf and one full-scale cavity URM walls. Each tested full-scale specimen consisting of a OOP panel and two return walls, varying in terms of commonly encountered boundary conditions, applied overburden or the presence/absence of an opening. The specimens were subjected to sequences of incremental input motion till collapse and these results are presented in terms of deformed shapes, failure mechanisms and force-displacement hysteretic curves. State of the art analytical techniques based on the method of virtual work are applied to evaluate their reliability as simplified tools for assessing the behaviour of such wall subjected to OOP two-way bending excitation. The testing campaign resulted in all specimens exhibiting a rather brittle response, despite sustaining accelerations of 1 g without any damage.@en

This paper provides information related to the sensor measurements obtained from five different unreinforced masonry (URM) walls subjected to incremental dynamic shake-table tests at EUCENTRE, Pavia, Italy. This information has been made available to assist in the development and calibration of analytical and numerical models intended to simulate the out-of-plane (OOP) two-way bending response of URM walls. For further interpretation of the sensor recordings, and for a detailed discussion on the observed seismic performance of the specimens, the reader is referred to the article entitled “Experimental Response of URM Single Leaf and Cavity Walls in Out-Of-Plane Two-Way Bending Generated by Seismic Excitation” [1]. Videos documenting the failure of each specimen are also available on YouTube [2].@en

An applied element method-based modelling strategy to model rubble stone masonry is presented for the first time. The strategy represents rubble stone masonry as rigid blocks connected by deformable interface springs where non-linear material properties are lumped. Cyclic in-plane shear-compression experiments on rubble stone masonry piers representative of Portuguese historical constructions are reproduced with good accuracy. The robustness of the strategy is then investigated further by performing a parametric study and comparing results with existing literature on the subject. The numerical procedure marks a significant improvement in terms of computational burden with respect to conventionally used finite element-based strategies, making it attractive for performing non-linear dynamic seismic safety assessment of large structures.@en

This paper presents the results of incremental dynamic tests on full-scale unreinforced masonry (URM) walls constructed with an intentionally weakened mortar under two-way bending excitation in the out-of-plane (OOP) direction. A specimen was also subjected to simultaneous vertical and horizontal excitation. Walls were also tested in boundary conditions that have not been previously reported in the literature. Emphasis was placed on the development and interpretation of a characterisation test to estimate the torsional shear strength of bed-joints, a parameter of prime importance in the OOP two-way bending behaviour of URM walls. Analytical formulation to estimate the initial stiffness of walls based on the theory of plates is then proposed and validated. Being the first experimental campaign to address the dynamic OOP two-way bending response of full-scale URM wall panels provided the opportunity to propose simple yet mechanically rational methodologies with which both the peak force resistance as well as the displacement at which such resistance is attained can be easily estimated.@en

An experimental programme was conducted on six different batches of unreinforced masonry (URM) considering different unit-mortar combinations to study their behaviour under torsional shear. Experiments performed included a specific characterisation test to measure the torsional shear strength of URM bed joints. Dilatancy measurements during both direct and torsional shear testing of masonry were recorded. Refined finite element modelling was then performed to evaluate whether parameters evaluated from standardised direct shear tests on masonry triplets can be used to estimate the torsional shear strength of URM bed joints. A possible mechanism by which dilatancy can increase the torsional shear resistance of bed joints without affecting the external level of applied normal force is also presented. Based on both experimental and numerical findings, a rational mechanics based formula to evaluate the torsional shear strength of URM bed joints is proposed and validated. This formula represents an improvement on currently used empirical formulation correlating the torsional shear strength of a URM bed joint to their flexural tensile strength.@en

This data article provides experimental data obtained from the incremental dynamic shake-table testing of four single leaf unreinforced masonry (URM) walls constructed in Calcium Silicate (CS) bricks reported in “Two-way bending experimental response of URM walls subjected to combined horizontal and vertical seismic excitation” [1]. These walls were tested in the second phase of a larger experimental campaign addressing the out-of-plane (OOP) response of full-scale URM panels in two-way bending configurations at EUCENTRE, Pavia, Italy. Data corresponding to the first phase of testing for four single leaf and one cavity wall has already been made available through Tomassetti et al. [2] and the information necessary to interpret these results can be found in Graziotti et al. [3]. The walls of the tests reported in this data article were constructed in an intentionally weakened mortar but using the same Calcium Silicate (CS) bricks as Graziotti et al. [3]. The walls were also tested in boundary conditions that were not addressed in [3] and a specimen was also subjected to simultaneous horizontal OOP and vertical seismic excitation in one of the tests. Videos documenting the failure of specimens tested in both phases of the experimental campaign can be viewed online on YouTube [4].@en

The multi-storied masonry “Gopuram” or the ceremonial entrance gateway, an ubiquitous structure in a South Indian temple, was a feature introduced circa 14 th c. AD in order to confer architectural status to structurally insignificant ancient shrines. The gopurams of the Meenakshi Temple in Madurai (1600 AD), the Ranganathaswamy Temple in Srirangam (17 th c. AD) and the Ekambareswar Temple in Kancheepuram (16 th c. AD), 52 m, 72 m and 59 m tall, respectively, are representative of the highly evolved Dravidian temple architecture. The current research is an attempt to understand the structural behaviour of the gopuram with a focus on a centuries-old incomplete structure in the town of Thiruvellarai in Tamil Nadu. The paper presents outcomes of detailed field and laboratory investigations on the sub-structure, the superstructure and structural materials of the gopuram, that provide insight into the structural configuration of the gopuram. It then examines the structural response to gravity and lateral forces, through non-linear finite element models of the structure. One of the significant aspects studied is the role of the core masonry in the structural response of the multi-leaf masonry structure of the gopuram. In addition the role of floor diaphragms in the structural stability of the gopuram is investigated. The outcome of the study is expected to provide important insights to the reasons for structural distress and collapse of such structures, particularly the Vijayagopuram at Srikalahasti in Andhra Pradesh in South India that collapsed in May 2010.@en

Damage observations from recent seismic events have confirmed that the activation of out-of-plane (OOP) local mechanisms is one of the major causes of structural collapse in unreinforced masonry (URM) buildings. Such failures are mostly due to the attainment of displacement levels incompatible with equilibrium configurations for the kinematic chain of the considered mechanism rather than the exceedance of stress capacity in structural elements. However, very little research can be found currently in literature regarding the two-way bending OOP failure mechanism of walls involving at least one restrained vertical edge despite it being one of the most commonly reported and surveyed cause of structural damage. The paper takes a step towards addressing this lack of knowledge in the form of dynamic testing of full-scale URM masonry walls. In particular, results of three of these tests are reported in this paper: a test on a specimen representing the calcium silicate inner/loadbearing wall, a test representing the clay outer wall, and a cavity wall composed of both leaves together.@en

A non-linear discrete homogenized strategy is used to reproduce out-of-plane two-way bending shake table tests on full-scale unreinforced masonry walls. The numerical strategy represents unreinforced masonry walls as an assemblage of rigid quadrilateral plates connected through non-linear interfaces represented. The material response laws for these non-linear interfaces are derived from a homogenization procedure performed at the meso-scale by a finite-element plate model. This discrete model can simulate masonry orthotropy, full softening behaviour and failure. Bending and torsional deformation modes are represented. The performance of the numerical models in describing the response of the experimentally tested full-scale specimens is presented. The numerical strategy is then used to study the behaviour of unreinforced masonry walls in configurations hitherto untested. Such numerical results are then compared to state-of-the-art analytical approaches. Potential avenues of improvement for the implemented analytical methods are also highlighted.@en