Quantifying sensitivity in numerical weather prediction?modeled offshore wind speeds through an ensemble modeling approach

A decade of research has shown that numerical weather prediction (NWP)-modeled wind speeds can be highly sensitive to the inputs and setups within NWP model. For resource characterization applications, this sensitivity is often addressed by constructing a range selecting one best validates against observations. However, approach not possible in areas lack high-quality hub height observations, especially offshore areas. In such cases, techniques quantify disseminate confidence NWP-modeled absence observations are needed. We address need present study propose practices for quantifying spread speeds. implement an ensemble which we consider 24 different Weather Research Forecasting (WRF) construct considering variations WRF version, namelist, atmospheric forcing, sea surface temperature (SST) forcing. Our analysis finds standard deviation produces more consistent estimates compared interquartile tends conservative estimator variability. further find model increases closer on shorter time scales. Finally, explore methods attribute total variability components (e.g., forcing SST product) contributions also vary depending scale. anticipate results presented paper will provide reasonable foundation into ensemble-based data sets.