How Do Adjacent Masonry Buildings Interact During Earthquakes?

Title, Authors, and Publication

The paper, titled "Seismic Testing of Adjacent Interacting Masonry Structures – Shake Table Test and Blind Prediction Competition," was authored by Tomi? Igor, Andrea Penna, Matthew DeJong, Christoph Butenweg, António Araújo Correia, Paulo Xavier Candeias, Ilaria Senaldi, Gabriele Guerrini, Daniele Malomo, Bastian Wilding, Didier Pettinga, Mark Spanenburg, Nikos Galanakis, Stuart Oliver, Francesco Parisse, Rui Marques, Serena Cattari, Paulo B. Louren?o, Francisco Galvez, Dmytro Dizhur, Jason Ingham, Giancarlo Ramaglia, Gian Piero Lignola, Andrea Prota, Omar AlShawa, Domenico Liberatore, Luigi Sorrentino, Raffaele Gagliardo, Michele Godio, Francesco Portioli, Raffaele Landolfo, Fabio Solarino, Nicoletta Bianchini, Maria Pia Ciocci, Antonio Romanazzi, Abide A??ko?lu, Jennifer D’Anna, Rafael Ramirez, Federico Romis, Marko Marinkovi?, and Filip ?or?evi?. It was presented at the 3rd European Conference on Earthquake Engineering & Seismology (3ECEES) in Bucharest, Romania, 2022.

Objective and Background

Masonry structures in historical European city centers are often constructed as building aggregates, where adjacent buildings share walls. These interacting structures exhibit complex seismic behavior, leading to increased vulnerability during earthquakes. The study investigates how adjacent masonry buildings interact dynamically under seismic loads, with a focus on:

• Understanding the influence of weak mortar joints on seismic response.
• Providing experimental data to improve numerical models of masonry aggregates.
• Evaluating the variability in predictions of different seismic analysis methods.

A shake table test was performed on a half-scale masonry building aggregate, consisting of two stone masonry units connected by mortar joints. The results were compared with blind predictions from researchers worldwide, highlighting the challenges of modeling masonry aggregates.

Introduction

Masonry buildings in historical centers lack uniformity in material properties, construction methods, and interconnections, making their seismic response difficult to predict. Recent earthquake observations in Italy and other European countries have shown that connections between adjacent buildings often fail first, leading to complex damage mechanisms such as:

• Out-of-plane collapse of fa?ades.
• Separation of buildings at mortar joints.
• In-plane shear and flexural failures.

Due to the lack of large-scale experimental data, previous numerical models of masonry aggregates have relied on simplified assumptions. This study, as part of the SERA AIMS (Seismic Testing of Adjacent Interacting Masonry Structures) project, aims to bridge this gap through controlled experimental testing and blind prediction comparisons.

Methodology

1. Shake Table Test on Masonry Aggregate A half-scale prototype was built, consisting of: Unit 1: One-story masonry structure. Unit 2: Two-story masonry structure. The two units were connected by a weak mortar joint, replicating historical masonry construction. Shake table tests were performed at LNEC laboratory in Lisbon, Portugal.
2. Seismic Testing Sequence The structure was subjected to incrementally increasing earthquake excitations, with peak ground accelerations (PGAs) from 0.11g to 0.6g. Tests were conducted in three directions: X-direction (transversal shaking). Y-direction (longitudinal shaking). Bidirectional shaking (X+Y combined).
3. Blind Prediction Competition Twelve research teams submitted 13 numerical models, including: Discrete Element Models (DEM). Finite Element Models (FEM). Equivalent Frame Models (EFM). Hand calculations and limit analysis. Researchers were asked to predict: Damage mechanisms. Displacements and base shear values. Failure sequence and collapse mode.

Key Findings

1. Masonry Aggregates Exhibit Highly Nonlinear Behavior Significant out-of-plane displacement was observed in fa?ades and wall connections. The mortar joint between units was the weakest link, leading to early separation and rocking motion.
2. Blind Predictions Showed Large Variability Limit analysis models predicted correct failure modes but underestimated PGA at failure. FEM and DEM models provided reasonable base shear values but struggled with damage localization. Predicted displacements varied widely, highlighting modeling uncertainties.
3. Strengthening Influences Seismic Response After Run 2.1 (PGA = 0.593g), significant structural damage occurred. Strengthening measures improved out-of-plane stability, but residual displacement remained high.
4. Numerical Models Require Further Refinement The study revealed the need for improved nonlinear models that account for: Unit-to-unit interaction. Floor diaphragm flexibility. Material degradation over multiple seismic cycles.

Conclusion

This study confirms that adjacent masonry structures exhibit highly unpredictable seismic behavior due to nonlinear interactions and weak mortar joints. The blind prediction competition highlighted significant variability in modeling approaches, emphasizing the need for more experimental data.

• Out-of-plane fa?ade displacement remains a primary concern.
• Numerical models struggle to capture the full complexity of masonry aggregates.
• Future research must refine modeling techniques to improve seismic risk assessments.

The findings provide valuable insights for structural engineers and conservationists working on heritage masonry preservation and seismic strengthening strategies.

Future Work and Applications

1. Refining Numerical Models Developing nonlinear FEM and DEM models that account for dynamic separation of adjacent units.
2. Expanding Experimental Testing Conducting shake table tests on different masonry configurations, including interlocking stones and steel-tied joints.
3. Investigating Strengthening Techniques Evaluating fiber-reinforced polymers (FRP), post-tensioning, and anchored steel ties for seismic reinforcement.
4. Application to Earthquake Risk Assessment Integrating masonry aggregate response into seismic hazard mapping for historical city centers.
5. Developing Guidelines for Masonry Aggregate Seismic Design Establishing engineering best practices for assessing and retrofitting masonry buildings in urban settings.

?