CLIMATHUNDERR: A database of LS-PIV velocity and temperature data from experimental downbursts. A combination of the impinging jet and gravity current techniques (Q10018)

Thunderstorm downbursts are intense, negatively buoyant winds that descend from cumulonimbus clouds, spreading horizontally near the ground and potentially impacting natural and built environments. They form due to instability between cold, dense air in the cloud and the warmer, lighter surrounding atmosphere. This instability generates vortical structures that descend at the radial edge of the downdraft and, upon impingement, constrain the outflow boundary layer near the ground. The most prominent of these structures, the primary vortex (PV), is often associated with high-intensity winds close to the surface, i.e., where human activity predominantly occurs. Despite extensive research, full-scale measurements of downbursts remain scarce due to their short-lived, localized nature, which limits comprehensive data collection. Consequently, controlled experimental simulations in specialized wind laboratories are preferred. Two primary techniques are used: Impinging Jet (IJ): Simulates downbursts mechanically using air fans, though it lacks the thermodynamic contributions of natural phenomena. Gravity Current (GC): Relies on density differences between fluids to produce a buoyant jet but often operates on smaller scales, making IJ preferable for reproducing realistic Reynolds numbers relevant to engineering applications. The CLIMATHUNDERR Project The EU-funded CLIMATHUNDERR (CLIMAtic investigation of THUNDERstorm winds) project for the first time combined GC and IJ techniques at large fluid dynamics scales at the Jules Verne Climatic Wind Tunnel (JVCWT) at the Centre Scientifique et Technique du Btiment (CSTB) in Nantes, France. Experiments investigated the buoyant contributions driving downburst winds by introducing temperature differences (ΔT) between the testing chamber air (~25C) and a cold air jet. Experimental Setup An impinging-jet plenum (IJP) measuring 2 2 2 m3 was positioned H = 3 m above ground level. Cold air filled the plenum via an air conditioning (A/C) unit in closed-loop mode, while temperature differences (ΔT) ranged from 0C to 25C. Temperature uniformity within the plenum and chamber was challenging due to the large geometries involved and consequent onset of temperature stratifications. Target ΔT values are recorded in experiment names, while the actual ΔT achieved are reported in the file test_plan.txt. A piston (2 2 0.6 m3) at the plenum top was released at two velocities (0.15 m s-1 or 0.25 m s-1) to regulate the initial jet speed at the nozzle (D = 1 m) exit (w_IJ= 0.8 m s-1 or 1.3 m s-1). This, combined with the buoyant contribution (w_GC), formed the final jet velocity (w). The nozzles louvers opening, controlled electronically, synchronized all measurements. Testing Campaign Fifty-seven experiments were conducted, with the first 39 focusing on geometric and dynamic properties of pure downbursts and the final 18 analyzing downburst interactions with a 1:2000-scaled model of Genoas Polcevera Valley, fabricated from expanded polystyrene (EPS). The model spanned 3.5 2 m, representing 7 4 km of terrain extending from the Genoa port area to the inland hills. A three-dimensional (3D) STL drawing of the model is attached, model.stl. Velocity measurements were performed using Large-Scale Particle Image Velocimetry (LS-PIV), offering high spatial resolution and insensitivity to temperature fluctuations. Helium-filled soap bubbles (HFSBs) served as tracers, illuminated by a dual-pulsed laser (532 nm). Two cameras captured images: initially, an Imager Pro X 4M with a CCD sensor (4 MP resolution) until it malfunctioned at experiment #15, then an Imager SX6M with a CMOS sensor (6 MP resolution). PIV pairs of images were recorded at 7 Hz and 12 Hz, respectively. Data processing employed DAVIS 10 software using Fast Fourier Transform correlation and interrogation windows of 32 32 pixels with 50% overlap. The cameras field of view (FOV) covered a vertical area of 2.5 2.5 m and 2.5 3.2 m (vertical horizontal), respectively. The FOV origin was (r0, z0) = (0, 0) corresponding to the jet vertical centerline and floor level, respectively. The orography model and the PIV camera were placed at two different locations to analyze the effect of different downdraft touchdown scenarios on the reproduced orography. The southern edge of the model, corresponding to the port dam, was placed respectively at 0.3 m with respect to the geometric center of the IJ touchdown. Temperature measurements were collected using 20 fast-response thermocouples (Tf, reaction time: ~0.5 s) and four standard thermocouples (Ts, reaction time: ~1 s). Fast-response thermocouples were strategically positioned in the outflow field and along the jet centerline, while the standard thermocouples were installed inside the IJP and at the connection between A/C system and IJP. Temperature differences between the plenum and chamber were calculated using thermocouple averages inside the IJP and along the jet centerline prior to the jet release. The thermocouple coordinates are reported in two separate files for experiments without and with model installed in the flow. The file thermocouples_noModel.txt reports the absolute horizontal (radial, r) and vertical (z)coordinates relative to the jet touchdown geometric center, (r0, z0) = (0, 0). For the tests with the model in place, the file thermocouples_wModel.txt reports the thermocouple coordinates (x, y, z), where xis the models longitudinal coordinate, y indicates the models transversal coordinate, while z is the vertical coordinate relative to the models surface (z = 0 denotes thermocouple installed at the models surface). (x, y) = (0, 0) corresponds to the southwesternmost point of the model. Data Organization A comprehensive list of the experimental runs is provided in the file test_plan.txt along with the temperature and velocity parameters characterizing each experiment. Two datasets accompany the experiments: Velocity Data (PIV.zip): Contains 57 folders, each corresponding to an experiment. Folders are named i0AB_dTCD_VpEF_repG, where AB represents the experiment number, CD the target ΔT (C), EF the jet velocity (w_IJ, m/s), and G the repetition number. For experiments involving the orography model (#4057), folder names also include the model position (posH, H = 1 or 2 for jet touchdown onshore or offshore). Each folder contains tab-delimited .txt files representing time frames of PIV measurements. Inside each folder, the tab-delimited .txt files correspond to specific time frames of the PIV measurements. Folders for experiments #1 to #15 contain 210 text files (ranging from V0001.dat to V0210.dat), reflecting 30 s of data recorded at a 7 Hz acquisition frequency. Folders for experiments #16 to #57 contain 360 text files (from V0001.dat to V0360.dat), representing 30 s of data collected at a 12 Hz acquisition frequency. The first file corresponds to measurements recorded at the time t = 0 of nozzle opening. The piston descent began 1000 ms after the nozzle opening. Each .txt file comprises various variables extracted from the raw PIV measurements, organized into 8 columns: the longitudinal coordinate r (r) [mm]; the vertical coordinate z (z) [mm]; the horizontal wind speed u(u) [m s-1], measured as positive in the outgoing direction of downburst propagation; the vertical wind speed w(w) [m s-1], measured as positive upward; the correlation value between point(s) in consecutive frames that determine the resulting velocity vector (Correlation value); uncertainty quantification of u (Uncertainty u); uncertainty quantification of w (Uncertainty w); a data validity flag (isValid). Temperature Data (T.zip): Contains 57 text files, each corresponding to an experimental run and named according to the PIV folder structure defined earlier. Files for experiments #1 to #19 include 421 rows (representing sampling times of 30 s at a 14 Hz frequency plus nozzle opening time t = 0), while files for experiments #20 to #57 contain 361 rows (30 s at a 12 Hz frequency plus nozzle opening time t = 0). Analogously to PIV records, the first measurement of each file corresponds to the opening time of the nozzle, t= 0. Each file consists of 28 columns containing the following variables: sample time (Time) [s]; temperature T,wt [C], relative humidity RH,wt [%], and atmospheric pressure Patm,wt [hPa] from a sensor located upstream of the test section inside the main JVCWT horizontal nozzle; temperatures from the fast thermocouples labeled Tf01 to Tf20 [C]; and readings from the monitoring standard thermocouples labeled Ts01 to Ts04 [C]. A comprehensive description of the database of measurements and of its potential uses, along with a few output examples, will be soon published in the data paper by Canepa et al. (2025).