DECIPHER Monte Baldo Flux Tower, Crest, Processed Data (University of Trento)

1. Short Description of the Sensor The Flux Tower dataset comprises a cluster of instruments deployed at three levels of a 9-m mast to observe the atmospheric properties, airflow characteristics, turbulence (via the eddy covariance method), and local precipitation. The flux tower has three monitoring levels at 3, 6, and 9 m. The list of instruments per level is given below, together with a direct link to each sensor manufacturer's page: 3-m level: First level from the ground devoted to the characterization of the mean and turbulent properties of the airflow. It is also equipped for computing mass fluxes. Sensors deployed at this level are: N. 1 3D Sonic anemometer CSAT3A (Campbell Scientific, https://www.campbellsci.com/csat3a), measuring the three components of the wind velocity and the sonic temperature N. 1 Open-path gas analyser EC150 (Campbell Scientific, https://www.campbellsci.com/ec150), measuring the concentrations of water vapour and carbon dioxide, atmospheric pressure, and air temperature. Coupled with the sonic anemometer, fluxes of water vapour and carbon dioxide can be computed 6-m level: Second level from the ground devoted to the characterization of the mean and turbulent properties of the airflow. Sensors deployed at this level are: N. 1 3D Sonic anemometer CSAT3A (Campbell Scientific, https://www.campbellsci.com/csat3a), measuring the three components of the wind velocity and the sonic temperature N. 1 Thermohygrometer rotronic HC2S3-L (Campbell Scientific, https://www.campbellsci.com/hc2s3), measuring the air temperature and relative humidity 9-m level: Third level from the ground devoted to the characterization of the mean and turbulent properties of the airflow. It also includes a net radiometer to evaluate the radiative balance at the surface. Sensors deployed at this level are: N. 1 3D Sonic anemometer CSAT3A (Campbell Scientific, https://www.campbellsci.com/csat3a), measuring the three components of the wind velocity and the sonic temperature N. 1 Thermohygrometer rotronic HC2S3-L (Campbell Scientific, https://www.campbellsci.com/hc2s3), measuring the air temperature and relative humidity N. 1 All-in-one Weather station ATMOS 41 Gen 2 (Meter, https://metergroup.com/products/atmos-41), measuring solar radiation, precipitation, electrical conductivity, air temperature, barometric pressure, relative humidity, wind speed, direction, maximum gust, lightning, tilt. Data Storage: Data were locally stored in a CR6 (Campbell Sceintific, https://www.campbellsci.com/cr6) and ATMOS ZL6 (Meter, https://metergroup.com/it/products/zl6) dataloggers, and normal operations were monitored via remote access. Power Supply: An input power supply of 12-14 V was ensured through solar panels operating during the whole campaign. 1.2 Specification The specifications for each sensor can be found on the manufacturer's web pages previously reported. 2. Measurement Strategy during the Campaign 2.1 Description of data collection The data collection is organized in three separate folders according to the sample rate of the instrumentation: High Sampling (HF): Data from sonic anemometers and an open-path gas analyser sampled at 20 Hz, including all the levels of the flux tower. Low Sampling - atmosphere (LF - weather): Data from thermohygrometers and the net radiometer sampled at 1 second, including all the levels of the flux tower. Low Sampling - weather station (LF - weather station): Data from the weather station is sampled at 5 minutes. 2.2 Time period covered by the data 24 June 2025 - 09 October 2025 2.3 Time zone UTC+1 2.4 Physical location Latitude 45,6646646; Longitude 10,8160711; Altitude 1668 m 3. Data Processing 3.1 Description of derived parameters and processing techniques used Aggregated 30-minute averages are provided. Averaging is performed in blocks after data processing and quality checks. Second-order turbulence quantities are computed through the eddy covariance method, after the application of Reynolds decomposition or detrending, or computed directly from spectra (after dealiasing). 3.2 Description of quality assurance and control procedures All data have been checked to ensure values fall within the expected ranges for the site's geographic location. Data collected using sonic anemometers were further processed to remove outliers and then rotated to align with the streamline: the despiking procedure follows the indication by Vickers and Mahrt 1997 and Schmid 2000, by which an outlier is identified as such if its value is more than 3.5 median absolute deviations from the median of each 30 min data distribution. The alignment to the streamline is performed according to McMillen et al. 1988 (see also Kaimal and Finnigan 1994), completing a double rotation to align the wind vector to the main wind direction. For a complete application of the whole sonic anemometer cleaning procedure, consult Barbano et al. 2022. Solar radiation is corrected to set the zero in the nocturnal downward shortwave radiation: the mean bias is computed among negative downward shortwave values and removed from the time series. Net radiation is also re-computed. Warning flags are provided for non stationary 30-min intervals in the wind direction and horizontal wind speed, and if the double rotation fails to set mean(v) and mean(w) to zero. 4. Data Format 4.1 Data file structure Data are stored in a single file in .mat and .csv formats, then zipped. 4.2 File naming convention [initial date]_[final date]_DECIPHER_FluxTower_Crest_[average window].mat [initial date]_[final date]_DECIPHER_FluxTower_Crest_[average window].csv 4.3 List of relevant parameters and units Each file is provided with headers listing the quantities' names using a common nomenclature. For details, contact [e-mail]. 5. Data Remarks 5.1 Known missing data periods 5.2 Software compatibility MATLAB or any software that can read and process CSV files