.. _ifs_nemo_energy_indicators: Energy Indicators for IFS-NEMO ============================== Historical ----------- The energy indicators are derived from the IFS-NEMO coupled simulations (0.05° × 0.05°) for the historical period (1990-2014) they have been evaluated against a reference dataset. Here ERA5 (0.25° × 0.25°) is used as a reference data. The following indicators have been compared with ERA5-derived indicators: **1. Wind speed distribution at 100 m hub height** **2. Low wind events (wind speed at 100 m < cut-in speed, [3 m/s])** Given the global availability of the simulations, a preliminary assessment of hub-height (100 m) wind speed at the global scale is conducted and its distribution at four representative wind farms is analysed. The selected sites are: the Alta Wind Energy Center in California (onshore), USA (35.02° N, 118.41° W), Moray East offshore wind farm in the North Sea, UK (58.03° N, 2.80° W), Markbygden onshore wind farm in northern Sweden (65.06° N, 20.53° E), and the Dongtai–Rudong offshore wind farm in the Yellow Sea, China (32.74° N, 121.48° E). These wind farms are chosen to represent diverse climate regimes, terrain characteristics, onshore and offshore conditions, and the availability of turbine power-curve information. While the analysis focuses on these locations, the global nature of the dataset enables the same framework to be readily applied to any region of interest worldwide. The wind farms locations are depicted in **Figure 1** below. .. figure:: ../../figures/windfarm_locations.png :align: center :width: 100% :figclass: align-center **Figure 1:** Wind speed climatology (1991-1995) and four wind farm locations used in this comparison study. Wind speed distribution at 100 m hub height: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The probability density function (PDF) of wind speed at 100 m hub height has been analysed for the selected wind farm locations over one-month, one-year, and 5-year periods. In general, on monthly timescales, IFS–NEMO exhibits a wider distribution with more frequent higher wind speeds across the four wind farms. However, on annual and 5-year timescales, ERA5 and IFS–NEMO show good agreement, except at the Alta Wind Energy Center in California, where ERA5 still produces a narrower distribution (See **Figure 2**). This feature may be related to the complex background terrain associated with Alta wind farm. .. figure:: ../../figures/wind_speed_distribution_alta_wf.png :align: center :width: 95% :figclass: align-center **Figure 2a:** The wind speed distribution from IFS-NEMO model using Energy Indicator package over the Alta Wind Energy Center in California, USA (35.02° N, 118.41° W) and ERA5 distribution. Monthly distribution (left), full year (center) and 5-yr distribution (right) .. figure:: ../../figures/wind_speed_distribution_moray_wf.png :align: center :width: 95% :figclass: align-center **Figure 2b:** The wind speed distribution from IFS-NEMO model using Energy Indicator package over the Moray East offshore wind farm in the North Sea, UK (58.03° N, 2.80° W) and ERA5 distribution. Monthly distribution (left), full year (center) and 5-yr distribution (right) .. figure:: ../../figures/wind_speed_distribution_markbygden_wf.png :align: center :width: 95% :figclass: align-center **Figure 2c:** The wind speed distribution from IFS-NEMO model using Energy Indicator package over the Markbygden onshore wind farm in northern Sweden (65.06° N, 20.53° E) and ERA5 distribution. Monthly distribution (left), full year (center) and 5-yr distribution (right) .. figure:: ../../figures/wind_speed_distribution_china_wf.png :align: center :width: 95% :figclass: align-center **Figure 2d:** The wind speed distribution from IFS-NEMO model using Energy Indicator package over the Dongtai–Rudong offshore wind farm in the Yellow Sea, China (32.74° N, 121.48° E) and ERA5 distribution. Monthly distribution (left), full year (center) and 5-yr distribution (right) Low wind events: ~~~~~~~~~~~~~~~~ Additionally, the comparison of Low wind Events (LWE, wind speed below 3m/s) shows similar reasonably well spatial and temporal match across wind farms with ERA5 data. In general IFS-NEMO simulates fewer low wind event occurrence as compared to the ERA5. It can be clearly seen over Alta wind farm locations and it is less evident over offshore locations like Moray East in North Sea (See Figure 3) .. figure:: ../../figures/lwe_comparisons_at_windfarm_locations.png :align: center :width: 95% :figclass: align-center **Figure 3:** The Low Wind Event (LWE) occurrence from IFS-NEMO model and ERA5 using Energy Indicator package over the four wind farms for the year 1991-1995. Recommendations ~~~~~~~~~~~~~~~ The Energy Indicators use case provides monthly wind speed histograms as output. However, for comparison with other datasets, the probability density function (PDF) was calculated, where an equal bin width of 0.5 m/s was used for all datasets considered. In general, IFS–NEMO captures the wind speed distribution reasonably well. IFS–NEMO generally simulates fewer low-wind events compared to ERA5. Projection ----------- Currently under evaluation. Further notes ---------------- The evaluation of the full set of simulations (historical and projection) is currently under evaluation and will be added to this section in the future. Further metrics and uncertainty information will also be added when available.