The objective of this work was to compare the hygrothermal responses and the moisture performance of four wood-frame walls as predicted by four hygrothermal (HAM) simulation tools, namely: DELPHIN, WUFI, hygIRC and COMSOL. The four wall systems differ only in their cladding type; these were fibreboard, vinyl, stucco and brick. Three Canadian cities having different climates were selected for simulations: Ottawa, Ontario; Vancouver, British Columbia and Calgary, Alberta. In each city, simulations were run for 2 years. Temperature and relative humidity of the outer layer of OSB sheathing were compared amongst the four simulation tools. The mould growth index on the outer layer of the OSB sheathing was used to compare the moisture performance predicted by the respective hygrothermal simulation tools. Temperature profiles of the outer layer of the OSB sheathing were all in good agreement for the four HAM tools in the three locations. For relative humidity, the highest discrepancies amongst the four tools were found with stucco cladding where differences as high as 20% could be found from time to time. Mould growth indices predicted by the four HAM tools were similar in some cases but different in other cases. The discrepancies amongst the different HAM tools were likely related to: the material property processing, how the quantity of wind-driven rain absorbed at the cladding surface is computed and some implementation details. Despite these discrepancies, The tools generally yielded consistent results and could be used for comparing the impacts of different designs on the risk of premature deterioration, as well as for evaluating the relative effects of climate change on a given wall assembly design.

The objective of this study was to explore the potential of a machine learning algorithm, the Support Vector Machine Regression (SVR), to forecast long-term hygrothermal responses and the moisture performance of light wood frame and massive timber walls. Hygrothermal simulations were performed using a 31-year long series of climate data in three cities across Canada. Then, the first 5 years of the series were used in each case to train the model, which was then used to forecast the hygrothermal responses (temperature and relative humidity) and moisture performance indicator (mold growth index) for the remaining years of the series. The location of interest was the exterior layer of the OSB and cross-laminated timber in the case of the wood frame wall and massive timber wall, respectively. A sliding window approach was used to incorporate the dependence of the hygrothermal response on the past climatic conditions, which allowed SVR to capture time, implicitly. The variable selection was performed using the Least Absolute Shrinkage and Selection Operator, which revealed wind-driven rain, relative humidity, temperature, and direct radiation as the most contributing climate variables. The results show that SVR can be effectively used to forecast hygrothermal responses and moisture performance on a long climate data series for most of the cases studied. In some cases, discrepancies were observed due to the lack of capturing the full range of variability of climate variables during the first 5 years.

The objective of this study was to assess the potential effects of climate change on the moisture performance and durability of massive timber walls on the basis of results derived from hygrothermal simulations. One-dimensional simulations were run using DELPHIN 5.9.4 for 31 consecutive years of the 15 realizations of the modeled historical (1986–2016) and future (2062–2092) climates of five cities located across Canada. For all cities, water penetration in the wall assembly was assumed to be 1% wind-driven rain, and the air changes per hour in the drainage cavity was assumed to be 10. The mold growth index on the outer layer of the cross-laminated timber panel was used to compare the moisture performance for the historical and future periods. The simulation results showed that the risk of mold growth would increase in all the cities considered. However, the relative change varied from city to city. In the cities of Ottawa, Calgary and Winnipeg, the relative change in the mold growth index was higher than in the cities of Vancouver and St. John’s. For Vancouver and St. John’s, and under the assumptions used for these simulations, the risk was already higher under the historical period. This means that the mass timber walls in these two cities could not withstand a water penetration rate of 1% wind-driven rain, as used in the simulations, with a drainage cavity of 19 mm and an air changes per hour value of 10. Additional wall designs will be explored in respect to the moisture performance, and the results of these studies will be reported in a future publication.