Single column models (SCMs) are useful tools for the evaluation of parameterisations of radiative and moist processes used in general circulation models (GCMs). Most SCM studies to date have concentrated on regions where there is a sufficiently dense observational network to derive the required forcing data. This paper describes an ensemble single column modeling (ESCM) approach where the forcing data are derived from numerical weather prediction (NWP) analysis products. To highlight the benefits of the ESCM approach, four forcing data sets were derived for a two year period at the Tropical Western Pacific ARM (Atmospheric Radiation Measurement Program) sites at Manus Island and Nauru. In the first section of the study, the NWP derived forcing data are validated against a range of observations at the tropical sites. In the second section, the sensitivity of two different SCMs to uncertainties in the forcing data sets are analysed. It is shown that despite the inherent uncertainties in the NWP derived forcing data, an ESCM approach is able to identify errors in the SCM physics. This suggests the ESCM approach is useful for testing parameterisations in relatively observation sparse regions, such as the TWP.