Variability: a non-parametric measure

calculate Variability from an array, without assuming distribution; robust & no rare event issues

This technique simply calculates variation based on a comparison of all abundances in a time series, and is free of assumptions of an underlying 'normal' distribution. It is more robust that the coefficient of variation CV and/or the standard deviation of log transformed abundances SDL, it is not biased by rare events, zero abundances can be included if desired, it more accurately measures known long term variation from short time series and unlike CV / SDL it doesn't artificially suggest spectral reddening. See Heath, J.P. 2006. Quantifying temporal variability in population abundances. Oikos, 115:573-581.