Time-series modeling of moisture exposure on timber structures

This paper presents a methodology with the purpose of modeling and simulating outdoor temperature and relative humidity (vapor concentration) related to moisture exposure on timber structures. The methodology is based on time-series analysis with recorded meteorological data used as input. In situations where recorded data is missing or consists of limited sequences, this methodology offers the possibility to use synthetic data instead. It was found that temperature and vapor concentration could be modeled by a combination of deterministic functions and stochastic processes, which consider the dynamics of the climate, including slow seasonal variations and fast daily changes. Specifically, a model for temperature and vapor concentration in Stockholm, Sweden, is presented together with simulation results. It is shown that the model produces simulated sequences with statistical properties close to the real recorded ones. (c) 2006 Elsevier Ltd. All rights reserved.