PAWN is a new moment-independent GSA method that can be used in place of or as a complement of variance-based (Sobol’) GSA. Moment-independent methods differ from Sobol’ in that they consider the entire distribution of the model output, rather than its variance only. As such, they can be preferable when variance is not an adequate proxy of uncertainty, for example when the output distribution is highly-skewed or multi-modal. The difference between PAWN and other moment-independent (or “density-based”) methods is that PAWN uses the Cumulative Distribution Function of the output, rather than the Probability Density Function, which simplifies numerical implementation. Other advantages of PAWN are that it can be easily tailored to focus on output sub-ranges, for instance extreme values, and that intermediate results generated in the application of PAWN can be visualized to gather insights about the model behaviour (“factor mapping”).

PAWN was introduced in the paper:

Pianosi, F., Wagener, T. (2015), A simple and efficient method for global sensitivity analysis based on cumulative distribution functions, Environmental Modelling & Software, 67, 1-11. (Open Access)

and it is implemented in SAFE.

A comparison between PAWN and Sobol’ for parameter screening and ranking of a relatively complex environmental model (SWAT) is presented in the paper:

Farkhondeh KZ, Nossent J, Sarrazin F, Pianosi, F, van Griensven A, Wagener T, Bauwens, W (2017), Comparison of variance-based and moment-independent global sensitivity analysis approaches by application to the SWAT model, Environmental Modelling & Software, 91, 210–222.