Evaluation and comparison of simple empirical models for dead fuel moisture content.

Background: The moisture content of litter and woody debris is a key determinant of fire potential and fire behaviour. Obtaining reliable estimates of the moisture content of dead fine fuels (i.e. 1-h and 10-h fuels) is therefore a critical requirement for effective fire management. Aims: We evaluated and compared the performance of five simple models for fuel moisture content. The models belong to two separate classes: (1) exponential functions of the vapour pressure deficit; and (2) affine functions of the (weighted) difference between air temperature and relative humidity. Methods: Model performance is assessed using error and correlation statistics, calculated using cross validation, over four empirical datasets. Key results: Overall, the best performing models were the relaxed and generalised models based on the weighted difference between temperature and relative humidity. Conclusions: Simple functions of the difference between air temperature and relative humidity can perform as well as, if not better than exponential functions of vapour pressure deficit. However, it is important to note the limitations of all these models when applied to fuels with moisture contents <10%. Implications: The moisture content of fine dead fuels and woody debris can be reliably estimated using simple models that are amenable to easy application. Five models for the moisture content of fine dead fuels were evaluated using empirical datasets. Simple functions of the difference between air temperature and relative humidity were shown to perform as well as, if not better than exponential models based on vapour pressure deficit. [ABSTRACT FROM AUTHOR]