Forecasting diurnal variations in meteorological parameters for predicting fire behaviour

Abstract

Empirical models for predicting diurnal variations in air temperature, wind speed, and relative humidity, given daily maximum and minimum values, latitude, and Julian date, have been developed. Solar heating during the daytime is quantitatively depicted using a truncated sine function, and nighttime cooling is characterized by an exponential decay function. Daily wind-speed trends echo diurnal temperature patterns, and relative humidity is modelled as a function of temperature by assuming that absolute humidity remains constant throughout the day. The time of occurrence of daily maxima and minima in temperature and wind speed varies seasonally according to changes in day length. Maximum temperature and wind speed typically occur later in the day than the maximum solar insolation. The magnitude of this lag depends on both time of year and latitude. The models were parameterized for two locations in Australia and observed, and modelled variations in temperature, relative humidity, and wind speed were used to forecast diurnal trends in fire behaviour. Predicted fire behaviour compared favourably in the two cases considering the very simple modelling approach employed.