Spatial evaluation of high-resolution modeled offshore winds using estimated winds derived from a network of HF radars

The temporal and spatial variability in the New Jersey offshore wind resource has large implications on the energy production for proposed offshore wind parks. The Rutgers University Weather Research and Forecasting (RUWRF) mesoscale atmospheric modeling system can begin to diagnose as well as predict key sources of variability both in space and time, including sea and land breezes and frontal passages. While vertically validating model performance in coastal and offshore regions is readily achieved through the use of meteorological towers and approved remote sensing systems, horizontal evaluation of winds-especially at sufficiently high resolutions-can be difficult with pre-existing systems. We apply the high-resolution surface current mapping capabilities of a high frequency (HF) radar network to infer wind fields over the offshore domain of RU-WRF. Surface wind fields derived from the HF radar network are compared to 10m wind fields modeled by the RU-WRF; correlations are generally between 0.5 and 0.8 in the study domain. Finally, to demonstrate the feature-tracking ability of the HF radar and RU-WRF model simulated offshore winds, we focus on a passing front and its associated thunderstorms. An area of divergence in RU-WRF modeled near-surface winds is evident as the thunderstorm line passes, likely caused by either a strong outflow boundary ahead of the thunderstorm front or directly from the cold downdraft in the core of the cold rain, which would reach the ocean's surface and diverge outwards. The forcing was so strong that the response was evident in both the HF radar currents and HF radar-inferred surface winds. The case warrants future analysis in surface ocean response to thunderstorm outflow boundaries and downdrafts, especially via the use of the HF radar-derived surface winds.