FLOOD FREQUENCY ANALYSIS IN THE WORLD

The paper deals with the hypothesis that floods in the last few decades recorded an extreme increase in frequency. As analysis of the number of events is the first step in the flood risk modelling, an empirical study was conducted based on data from the natural disaster database of the Université Catholique de Louvain in Belgium. The Mann Kendall Trend Test is used to identify a tendency in a series (the number of floods from 1960 to 2022 in the world) by testing statistical hypotheses and a randomness check in its variations, i. e. determining whether the time series has a tendency. The test determined that the flood frequency tendency is statistically significant. Based on measures of representativeness of the tendency, standard error and mean average deviation, an exponential regression curve was chosen with a coefficient of determination of 0.8807. With a relatively high coefficient of determination value, the model gives a solid basis for predicting those events in the future, but this property could certainly be improved.
Key words: flood insurance, catastrophic weather events, floods, flood frequency tendency