Flike is an extreme value analysis package that calculates the probability of flood events based on historical records. It has been created for flood frequency analysis, but can be applied to any extreme value analysis. Developed by Professor George Kuczera from the School of Civil Engineering at the University of Newcastle Australia, Flike is compliant with the recent major revision of industry guidelines for flood estimation, documented in the draft update of Australian Rainfall and Runoff (ARR).

Key Technical Features:

  • A range of probability models including Log Normal, LPIII, Gumbel, GEV and Generalised Pareto distributions
  • Higher order L-moments for parameter estimation
  • Bayesian inference methodology for estimation of distribution parameters which allows:
    • The use of historic data outside of the gauged record
    • Ability to safely censor data
    • Incorporation of the multiple Grubbs-Beck test for low outliers
    • Ability to incorporate regional information as prior knowledge