Numerical modelling of the aerial drop of firefighting agents by fixed-wing aircraft. Part II: model validation

Abstract

The validation of the Aerial Drop Model consisted of the comparison of computed ground patterns with experimental data from a set of real-scale drop tests using water and a wide range of fire retardant viscosities. Results were analysed in terms of pattern length and area. A total of 78% of the computed line lengths per coverage level were within a 10% error, with an average normalised mean squared error of 0.01 and a Pearson correlation coefficient above 0.9. In all cases, nearly 90% of the results were within a factor of 2 of observations. The accuracy of the simulations showed no strong relation with the viscosity, although better results were obtained in the range from 700 to 1100 cP. In general, the model produced a good representation of the spatial distribution of the agent for various coverage levels and its accuracy was, in fact, within the statistical uncertainty of the cup-and-grid sampling method. The good performance obtained demonstrates that this tool, for the tested range of drop conditions, fulfils the requirements for use in the optimisation of firefighting operations, as a complementary method to real-scale drop testing, and in firefighter training activities.