Investigating Historic Wood Moisture Data

Wood and water have a delicate relationship. For over a century, the Forest Products Laboratory (FPL) has provided data on how wood products and moisture interact under various conditions. This includes equilibrium moisture content (EMC) values of wood for given temperature and relative humidity conditions.

Much of this data has been used for practical purposes related to drying lumber and controlling moisture content. FPL experts Samuel Glass and Sam Zelinka, a research physical scientist and a materials research engineer, respectively, have reevaluated the FPL EMC data with an eye towards reliability. Their report, called Investigation of Historic Equilibrium Moisture Content Data from the Forest Products Laboratory, is available as FPL General Technical Report (GTR) 229.

Rain-map

Excessive moisture can impact the durable life span of wood-in-use. The map above show areas of low, medium, and high average rainfall in the United States.

FPL data on wood and moisture is commonly cited in articles for commercial and homeowner use, such as the Wood Handbook, as well as in scientific works, covering topics such as the thermodynamics of water vapor sorption in wood and evaluation of physical models. Using the data for such scientific purposes presupposes that the methods by which the data were acquired are well documented and accepted.

Glass and Zelinka questioned previous assumptions about the historic EMC data and worked to uncover and evaluate the original data sources. In doing so they also addressed related topics, including how the presentation of data has evolved in the literature and whether the data are practically applicable to all wood species. The researchers affirm that the data are indeed useful for practical applications, such as lumber drying, conditioning of wood specimens prior to testing physical or mechanical properties, and modeling of moisture content of wood members in buildings.

On the other hand, the data are unreliable for scientific purposes, such as thermodynamic analysis and testing of physical models. There are three reasons: lack of proper documentation of methodology; the unsolvable problem of knowing which values are determined from direct observations and which are interpolated; and the absence of definitive measurement error analysis.