High quality wind data is needed to conduct wind resource assessment. Numerical Weather Prediction (NWP) models were recently used to provide the desired data. The current uncertainty sources in NWP models introduce uncertainties in wind resource assessment. More accurate NWP wind data will lead to more accurate resource assessment, wind turbine micrositing and energy short-term forecasting. Using one single NWP model may lead to undesired results compared to the huge investment in wind energy projects. New capabilities and strategies to quantify and reduce the uncertainty of wind data from NWP models are required. This paper proposed the use of nested ensemble NWP model for the wind resource assessment. This approach will provide information about the uncertainty of the NWP models and provide probabilistic information for researchers compared to the current use of single NWP model. A case study of the proposed approach is presented with four ensemble members using 7 km resolution over Oman. Verification results shows that the NWP ensemble mean performs better in average than individual members. (C) 2011 Elsevier Ltd. All rights reserved.