1 © France Energies Marines Characterizing Atmospheric Boundary Layer Turbulence in the Gulf of Lion: Insights from the Meso-NH atmospheric model and a wind lidar profiler Tanguy LUNEL1,2, Didier RICARD1, Maxime THIEBAUT2
[email protected] 1 Centre National de Recherches Météorologiques, Météo-France - CNRS 2 France Energies Marines WESC Nantes - 2025-06-25 Présentation France Energies Marine
Présentation France Energies Marines2 ●Difficulty in obtaining reliable turbulence information from models for the wind industry: –Example with AROME model: underestimation of TKE in comparison to an offshore fixed wind lidar profiler Context –At 140m: ●TKEAROME = 0.5 m2s-2 – TIAROME = 10 % ●TKELidar = 1.5 m2s-2 –TILidar = 17 % m2/s2 m/s . Mean profiles on 2023-01 Wind speed Wind direction TKE → How to improve TKE/TI estimates from mesoscale models?
Présentation France Energies Marines3 ●Methodology: –Based on atmospheric simulations - Mesoscale and Large Eddy Simulations (LES) - and wind lidar profiler data –Real case studies at: ●Le Planier island, Marseille, Mediterranean Sea: Materials and methods 15 km Taken From Thiébaut et al. 2024, Characterization of the offshore wind dynamics for wind energy production in the Gulf of Lion, Western Mediterranean Sea
Présentation France Energies Marines4 Materials and methods ●Observational data: –Wind lidar profiler ●Sampling rate: 1Hz ●Measurement heights from 60 to 220m a.s.l. ●Meso-NH: –Research atmospheric model with prognostic TKE, 1.5 order, 2nd-order moments are diagnosed –Dynamical downscaling approach, coupled to the surface model SURFEX 1 km 200 m 40 m AROME analysis (1.3 km)
Présentation France Energies Marines5 TKE concepts ●Concepts of TKE: –In observations: ●Temporal variance measured inside a small volume –In models: ●Subgrid vs Resolved TKE: –Resolved → temporal variance of mean wind in the grid point volume ● > 80% of total TKE in LES ●~ 50 % in the grey zone ● Low at the mesoscale –Subgrid → spatial variance of wind inside the volume of the grid point ● Needed for vertical mixing of variables Evolution of instantaneous wind speed at 140 m above Le Planier ws_obs or ws_LES ws_model_1km
Présentation France Energies Marines6 TKE concepts ●Concepts of TKE: –In observations: ●Temporal variance measured inside a small volume –In models: ●Subgrid vs Resolved TKE: –Resolved → temporal variance of mean wind in the grid point volume ● > 80% of total TKE in LES ●~ 50 % in the grey zone ● Low at the mesoscale –Subgrid → spatial variance of wind inside the volume of the grid point ● Needed for vertical mixing of variables Evolution of instantaneous wind speed at 140 m above Le Planier ws_obs or ws_LES ws_model_1km ●Comparison of model results to observations should consider: –Subgrid and Resolved TKE ●Well accepted in the community ●≈ Subgrid TKE for mesoscale ●≈ Resolved TKE for LES
Présentation France Energies Marines7 TKE concepts ●Concepts of TKE –TKE mixing via K-gradient theory: –TKE (e) is used for the Eddy-Diffusivity related mixing: with K = C Lm e1/2 Θ, r, u, v, w, e for a grid cell (= resolved)
Présentation France Energies Marines8 TKE concepts ●Concepts of TKE –TKE mixing via K-gradient theory: –Lack of mixing in counter-gradient zone → Eddy-Diffusity / Mass-Flux (EDMF) schemes –TKE (e) is used for the Eddy-Diffusivity related mixing: with K = C Lm e1/2 –Mass-Flux related mixing does not use directly TKE Mass-Flux (MF) Eddy-Diff (ED) wu Θu, ru, uu, vu, ee Eddy-Diff (ED) Θe, re, ue, ve, we, ee Θ, r, u, v, w, e for a grid cell (= resolved) Detrainment Entrainment au With au : fraction of updraft / wu, we : vertical velocity in updraft/environment
Présentation France Energies Marines9 TKE concepts ●Concepts of TKE –TKE mixing via K-gradient theory: –Lack of mixing in counter-gradient zone → Eddy-Diffusity / Mass-Flux (EDMF) schemes –TKE (e) is used for the Eddy-Diffusivity related mixing: with K = C Lm e1/2 –Mass-Flux related mixing does not use directly TKE Mass-Flux (MF) Eddy-Diff (ED) wu Θu, ru, uu, vu, ee Eddy-Diff (ED) Θe, re, ue, ve, we, ee Θ, r, u, v, w, e for a grid cell (= resolved) Detrainment Entrainment au ●TKE (e) is in this case a variable used for mixing ●But TKE does not account for the vertical velocity variance coming from the presence or not of updraft ●Ex: MYNN scheme in WRF / EDKF scheme in Meso-NH and AROME With au : fraction of updraft / wu, we : vertical velocity in updraft/environment
Présentation France Energies Marines16 Preliminary results ●Vertical profiles above Le Planier Island –Improvement with LES –But insufficient TKE in the LES? Altitude a.s.l. [m] m/s m2/s2 Mean vertical profiles above Le Planier on 2023-01-22 a.m.
Présentation France Energies Marines17 Preliminary results ●Vertical profiles above Le Planier Island –Improvement with LES –But insufficient TKE in the LES? ~100m –Or questionable lidar data? ●2 main uncertainties sources: –Inter-beam and intra-beam effects ●2 TKE retrieval methods: –standard method (inter + intra) –variance method (intra) Altitude a.s.l. [m] m/s m2/s2 Mean vertical profiles above Le Planier on 2023-01-22 a.m.
Présentation France Energies Marines18 Preliminary results ●Critical view on the TKE data from wind lidar profiler –Variance method is impacted only by intra-beam effect → systematic underestimation of wind variance –But variance method should be used only if wind direction is aligned with a pair of opposite beams → In the end, LES and lidar data provide consistent results Altitude a.s.l. [m] m/s m2/s2 Mean vertical profiles above Le Planier on 2023-01-22 a.m.
Présentation France Energies Marines19 Preliminary results ●Vertical profiles above Le Planier Island : –With consideration of Mass-Flux related wind variance: TKE_EQ → Mesoscale model able to give reasonably good TKE values in the middle of ABL → More difficulties for the surface boundary layer Altitude a.s.l. [m] m/s m2/s2 Mean vertical profiles above Le Planier on 2023-01-22 a.m.
Présentation France Energies Marines20 Preliminary results ●Cross-sections of TKE –On the morning (unstable) – 07:00 UTC ●TKE rather well captured offshore, i.e. at Le Planier island ●Still major differences near the coast –to investigate 1 km 40 m TKE and Ri at 09:00 UTC Total TKE [m2 s-2] Total TKE [m2 s-2]
Présentation France Energies Marines21 Discussion / conclusion ●Preliminary result: –A proper accounting for EDMF related wind variance allows for better match between mesoscale models and observations for the TKE –Variance method for lidar measurements is useful for interpretation of observed TKE ●Next steps: –Other case studies –Work on the TKE dissipation rate
Présentation France Energies Marines22 © France Energies Marines Thank you for your attention
Présentation France Energies Marines24 Preliminary results ●Components of mean profiles
Présentation France Energies Marines25 Preliminary results ●Variance vs Standard method for TKE retrieval from Lidar Taken From Thiébaut et al. 2024, Characterization of the offshore wind dynamics for wind energy production in the Gulf of Lion, Western Mediterranean Sea
Présentation France Energies Marines26 Preliminary results ●Critical view on the TKE data from wind profiling lidar –Reliability seems to depend on height and dimensions Vertical profiles above Le Planier on 2023-01-05 at 10:00 UTC
