We use a sub-set of European GPS network, equipped with surface meteorologi
cal sensors, to evaluate the impact of GPS precipitable water (PW) on the n
umerical weather prediction (NWP) model. GPS PW are assimilated into the PS
U/NCAR nonhydrostatic model MM5 using both the Nudging technique and the Ob
jective Analysis. The selected GPS network is made by six stations: Cagliar
i, Graz, L'Aquila, Matera, Wettzell and Zimmerwald. In this work we test th
e whole processing from GPS data reduction to GPS PW assimilation in order
to critically evaluate GPS PW impact on the weather forecast. (C) 2001 Else
vier Science Ltd. All rights reserved.