This repository contains the data used in the paper <Local wind regime induced by giant linear dunes: comparison of ERA5 re-analysis with surface measurements>,where wind data measured at 4 different places in and across the Namib Sand Sea are compared to the data from the ERA5/ERA5Land climate reanalyses. The use this data, one should first look at the following GitHub repository: https://github.com/Cgadal/GiantDunes. The description sometimes refers to scripts used in https://github.com/Cgadal/GiantDunes/tree/master/Processing. The two folders 'raw_data' and 'processed_data' contain the input raw_data, and the output data after procesing used to make the paper figures, respectively. In each of them, '.npy' files contain python dictionaries with different variables in it. They can be loaded using the python library numpy as 'data = np.load('file.npy', allow_pickle=True).item()' and the different keys (variables) can be printed with 'data.keys()' or 'data[station].keys()' if 'data.keys()' return the different stations. Note that, unless specified otherwise below, all variables are given in International System of Units (SI), and wind direction are given anticlockwise, with the 0 being a wind blowing from the West to the East. raw_data: - DEM: contain the Digital Elevation Models of the two stations from the SRTM30, downloaded from here: https://dwtkns.com/srtm30m/ - ERA5: hourly data from the ER5 climate reanalysis, on surface (_BLH) and pressure levels (_levels). Downloaded from https://cds.climate.copernicus.eu/ - ERA5Land: hourly data from the ER5Land climate reanalysis Downloaded from https://cds.climate.copernicus.eu/ - KML_points: kml points of the measurement station. Can be opened directly in GoogleEarth. - measured_wind_data: contain the measured in situ data. The windspeed is measured using Vector Instruments A100-LK cup anemometers, the wind direction using Vector Instruments W200-P wind vane and the time using Campbell Instruments CR10X and CR1000X dataloggers. processed_data: - 'Data_preprocessed.npy': preprocessed_data, output of 1_data_preprocessing_plot.py - 'Data_DEM.npy': properties of the processed DEM, output of 2_DEM_analysis_plot.py - 'Data_calib_roughness.npy': data from the calibration of the hydrodynamic roughnesses, output of 3_roughness_calibration_plot.py - 'Data_final.npy': file containing all computed quantities - 'time_series_hydro_coeffs.npy': file containing the time series of the calculated hydrodynamic coefficients by '5_norun_hydro_coeff_time_series.npy'. Depending on the loaded data file, main dictionnary keys can be: - 'lat': latitude, in degree - 'lon': longitude, in degree - 'time': time vector, in datetime objects (https://docs.python.org/3/library/datetime.html) - 'DEM': elevation data array in [m], with dimensions matching 'lat' and 'lon' vectors - 'z_mes', 'z_insitu', 'z_ERA5LAND': height of the corresponding velocity - 'direction': measured wind direction, in [degrees] - 'velocity': measured wind velocity, in [m/s] - 'orientaion': dune pattern orientation, [deg] - 'wavelength': dune pattern wavelength, [km] - 'z0_insitu': chosen hydrodynamic roughness for the considered station. - 'U_insitu', 'Orientation_insitu': hourly averaged measured wind velocities and direction - 'U_era', 'Orientation_era': hourly 10m wind data from the ERA5Land data set - 'Boundary layer height', 'blh': boundary layer height from the hourly ERA5 dataset - 'Pressure levels', 'levels': Pressure levels from the pressure levels ERA5 dataset - 'Temperature', 't': Temperature from the pressure levels ERA5 dataset - 'Specific humidity', 'q': Specific humidity from the pressure levels ERA5 dataset - 'Geopotential', 'z': Geopotential from the pressure levels ERA5 dataset - 'Virtual_potential_temperature': Virtual potential temperature calculated from the pressure levels ERA5 dataset - 'Potential_temperature': Potential temperature calculated from the pressure levels ERA5 dataset - 'Density': Density calculated from the pressure levels ERA5 dataset - 'height': Vertical coordinates calculated from the pressure levels ERA5 dataset - 'theta_ground': Averaged virtual potential temperature within the ABL. - 'delta_theta': Virtual potential temperature at the ABL. - 'gradient_free_atm': Virtual potential temperature gradient in the FA. - 'Froude': time series of the Froude number U/((delta_theta/theta_ground)*g*BLH) - 'kH': time series of the number 'kH' - 'kLB': time series of the internal Froude number kU/N Other keys are not relevant, and stored for verification purposes. For more details, please contact Cyril Gadal (see authors), and look at the following GitHub repository: https://github.com/Cgadal/GiantDunes, where all the codes are present.

Note that this repository contains modified Copernicus Climate Change Service Information (2021). Neither the European Commission nor ECMWF is responsible for any use that may be made of the Copernicus Information or Data it contains. Multiple grants have supported the collection of wind data through visits to the four sites between 2013 and 2020 (John Fell Oxford University Press (OUP) Research Fund (121/474); National Geographic (CP-029R-17); Natural Environment Research Council UK (NE/R010196/1 and NE/H021841/1 NSFGEO-NERC); , Southampton Marine and Maritime Institute SMMI EPSRC-GCRF UK), along with research permits (1978/2014, 2140/2016, 2304/2017, 2308/2017, RPIV00022018, RPIV0052018, RPIV00230218). The authors are very grateful for support from Etosha National Park (especially Shyane Kötting, Boas Erckie, Pierre du Preez, Claudine Cloete, Immanuel Kapofi, Wilferd Versfeld, and Werner Kilian), Gobabeb Namib Research Institute (Gillian Maggs-Kölling and Eugene Marais), The Skeleton Coast National Park (Joshua Kazeurua). Various researchers and desert enthusiasts have assisted with instruments and the logistics of expeditions, especially Mary Seely for expert guidance at the North Sand Sea site. Finally, we acknowledge financial support from the Laboratoire d'Excellence UnivEarthS Grant ANR-10-LABX-0023, the Initiative d'Excellence Universite de Paris Grant ANR-18-IDEX-0001, the French National Research Agency Grants ANR-17-CE01-0014/SONO and the National Science Center of Poland Grant 2016/23/B/ST10/01700.