Fuel Moisture Content in Croatian wildfire spread simulator AdriaFirePropagator

Abstract

Fuel moisture content (FMC) is the mass of water contained within vegetation in relation to the dry mass. It is one of the most important variables in all wildfire prediction and spread simulation models. FMC has great influence on wildfire ignition and combustion. For accurate wildfire spread simulations and wildfire risk index estimations, fuel moisture is a very important input variable. Since 2016, Croatian firefighters in everyday practice use Web based System for wildfire behaviour modelling and wildfire spread simulation named AdriaFirePropagator. The simulator is based on semi-empirical Rothermel’s surface fire spread model for wildfire behaviour modelling and cellular automata for wildfire spread simulation. Fuel moisture, both live and dead is a very sensitive parameter in wildfire behaviour modelling. Live fuel moisture defines the moisture content of live fuels and dead fuel moisture is defined as moisture of dead fuels with time-lag of 1 hour, 10 hour and 100 hours. In Croatia there is no organised service for daily measurement of fuel moisture content, so values of these variables has to be estimated from meteorological parameters. This paper compares three approaches to fine dead fuel estimation, all implemented in AdriaFirePropagator. The first one, used in most wildfire simulation software, was based on standard mathematical models that relate air temperature, air humidity, 24-hours rainfall and wind speed with fine dead fuel moisture (FFMC). The second one was based on standard Fire Behavior Analysis Tables (FBA Tables) and the third one was based on intensive experimental research of dead fine fuel moisture content of Aleppo pine (Pinus halepenses Mill.). After intensive experimental research new Croatian fine dead fuel models PhFFMC was developed, tested in selected pine species stand and applied in AdriaFirePropagator for fuel regions where this pine species dominate. Croatian model is much better correlated with experimental data, therefore for more accurate wildfire simulations, similar models have to be developed also for other typical vegetation fuel types.