Comparison of interface moisture conductivity averaging methods for moisture-induced deterioration prediction of outdoor cultural properties

In this study, we examined the impact of different approximation methods for physical property values on degradation prediction simulations of outdoor cultural properties due to moisture. We compared integral mean, weighted harmonic mean, weighted arithmetic mean, and three types of interface value evaluation methods, simulating both the water absorption process and the dry-wet cycle under outdoor environment fluctuation. The results showed that, in situations such as the water absorption process where there is rapid water penetration from dry conditions, the harmonic mean and the interface value evaluation methods, using the average chemical potential of water, significantly underestimated water transport. This presents a significant challenge when modeling rainfall-induced moisture infiltration. In the dry–wet cycle under outdoor environment, the interface value evaluation method using the average moisture content and the logarithmic mean of the chemical potential of water yielded favorable results. Given that the integral mean is the most accurate but also the highest computationally cost, it is essential to select an appropriate approximation method based on the analysis objective and the required level of precision. In addition to providing important insights into balancing accuracy and efficiency in numerical simulations for cultural property preservation, this study helps to understand trends in numerical computation errors in cases where the approximation methods cannot be freely chosen, such as with general-purpose software.