In this article an additional extreme value distribution (AEVD) is introduced by Box-Cox transformation. Both
generalized Pareto distribution under power normalization (GPDP) and generalized extreme value distribution under
power normalization (GEVP) can be obtained from AEVD as special cases. The AEVD can be applied in a wide variety
of application elds especially for modelling extreme data. A system of nonlinear equations is solved numerically to get
maximum likelihood estimation (MLE) for the distribution parameters. The proposed distribution is applied for modelling
daily maximum air pollution from Barking Dagenham station. Various criteria reveal that AEVD improves the tting of
this data compared with other distributions.