Hygrothermal modelling of historic rubble masonry walls: Accounting for geometric and compositional variability

An established issue when assessing the hygrothermal performance of heritage buildings is the correlation between measured data and model outputs. This can be attributed to many factors including climate data and hygrothermal material property data. Two overlooked factors are the variation in geometry and level of decay in the structure. This can be true for several historic wall assemblies with irregular geometries and elevated levels of decay. There is a lack of guidance on how and when to incorporate this into a hygrothermal model. The goal of this paper is to examine how this variability in geometry and composition may affect model outcomes using a simulation approach with stochastic methods for a sample rubble core masonry wall. Geometrically stochastic models were generated from a script accounting for varying levels of air voids in the wall and irregular stone patterns. The presence of voids alters heat, capillary and vapour transport. This was combined with hygrothermally dissimilar mortar and stone units. Simulations were performed under four boundary conditions: heat, drying, wetting and actual weather data. The results showed that geometric stochasticity can lead to a large variation of results. The presence of voids in the wall also led to a reduction in heat and moisture transport in the assemblies. The results of the stochastic simulations were then compared to a set of simplified one and two-dimensional models to judge their appropriateness. It was found that 2-D parallel like models that account for the mortar joint in some fashion are most appropriate.