Wall Material and Collapse-Related Failure in Rural Masonry Buildings after the 2023 Kahramanmaraş Earthquakes: Near-Fault Observational Comparison in Hatay, Türkiye

Abstract

This study examines whether wall material, hollow concrete block or solid clay brick, was associated with collapse-related failure in 516 nonengineered legacy rural masonry buildings in Antakya, Hatay, after the 2023 Kahramanmaraş earthquake sequence. Damage states were grouped into binary Failure and Nonfailure classes within a spatially constrained observational comparison. Because nonfailure outcomes were sparse, Firth penalized logistic regression was applied to reduce small-sample bias. The overall failure rate was 89.9% (95% confidence interval (CI) = 87.0%–92.2%). Solid brick buildings showed a higher observed failure rate (96.4%) than hollow block buildings (89.2%), but the association was not statistically significant [odds ratio (OR) = 2.56; 95% profile-penalized-likelihood interval = 0.83–12.66; p = 0.109 ]; building height was likewise nonsignificant (OR = 1.16; p = 0.746 ). Adjustment for construction period and neighborhood gave a similar material estimate (OR = 2.47; p = 0.134 ), whereas neighborhood showed the strongest recorded association with failure ( p < 0.001 ), with failure rates of 78.9% to 96.0% across three adjacent neighborhoods. A design-based simulation indicated limited detection probability; the data therefore neither demonstrates nor excludes a material effect. System-level features were not recorded as building-level variables. Wall material alone is insufficient as a standalone screening criterion.