We will discuss statistical methods for predicting and simulating spatio-temporal data and the uncertainty associated with various data sources. We propose a statistical space-time model for predicting atmospheric wind speed based on deterministic numerical weather predictions (NWP) and historical measurements. The wind speed forecast is represented as stochastic predictive scenarios that are targeted for power grid applications where they account for the uncertainty associated with renewable energy generation. As forecast evaluation is a crucial step in the prediction process, we will discuss the importance of benchmark data. In particular, we model the error associated with the verification dataset and investigate the effect of this error on a decision-making process.