Validation of Offshore Winds in the ERA5 Reanalysis and HRRR Model Using Two Floating LIDARS South of Long Island, NY

Wind forecasting over the coastal waters is challenging given the lack of observations and the limitations of operational numerical weather prediction (NWP) models in predicting the marine atmospheric boundary layer. Limited observations have required the use of empirically based boundary layer wind profile estimates (i.e., extrapolation using log-law and power-law relationships), mesoscale models, and model reanalyses for wind power resource assessments. These methods can, however, exhibit considerable wind prediction errors, especially when low-level jets (LLJs) are present, a particularly important warm season coastal wind phenomenon south of Long Island, NY. Using approximately three years (4 September 2019 - 30 November 2022) of hourly vertical wind profiler data from two New York State Energy Research and Development Authority (NYSERDA) floating lidars in the New York Bight, ERA5 Reanalysis, and HRRR model 1-18h forecasts, this study aims to understand how well short-term model forecast and reanalysis wind speed profiles compare against observations in the lowest 200 m ASL and for LLJ events.

We identified warm season (May-September) LLJ events using the observed floating lidar wind profiles and a combination of three different types of LLJ detection algorithms whose criterion have been slightly modified from their original usage in prior coastal LLJ papers (i.e., Andreas et al., 2000 and Debnath et al., 2021). An event was included in the warm season LLJ dataset, if at least two out of the three different algorithms identified a jet.

We found that the ERA5 reanalysis (31 km grid spacing and 8 vertical levels in the lowest 200 m) tends to underpredict wind speeds by 1 – 9% through the 200 m profile during spring and summer months. April through June have the largest negative ERA5-lidar wind speed biases of 9.5-17.5% (1.0 – 1.75 m s^-1), with a clear diurnal trend in this bias. During these LLJ events, the ERA5 had 29.9% of hours that completely missed the LLJ profile. On average, the lidar observed LLJ profile had a jet nose maximum of 11.4 ± 0.3 m s^-1 at 100 m ASL, while the corresponding ERA5 average had an elevated nose of 9.5 ± 0.3 m s^-1 at 140-160 m. The ERA5 tends to overpredict the mean height of wind speed maximum by 40-60 m, with only 9.33% of hours correctly depicting the observed jet nose height. Overall, this work highlights the challenges in using analyses with the limited offshore observations available for data assimilation in offshore wind prediction and wind resource assessment. This presentation will also highlight some of the HRRR 1-18h forecast results and some of the synoptic and mesoscale patterns associated with the largest wind speed biases.

References

Andreas, E. L., K. J. Claffy, and A. P. Makshtas, 2000: Low-level atmospheric jets and inversions over the western Weddell Sea. Boundary-Layer Meteorology, 97, 459–486, doi:10.1023/a:1002793831076.

Debnath, M., P. Doubrawa, M. Optis, P. Hawbecker, and N. Bodini, 2021: Extreme wind shear events in US offshore wind energy areas and the role of induced stratification. Wind Energy Science, 6, 1043–1059, doi:10.5194/wes-6-1043-2021.