Nightjar wake flow fields
Description
This data constitutes of flow fields capturing the wake behind European nightjars (Caprimulgus europaeus) flying in a wind tunnel. Most of them are for one wing and the body only. The data is captured using stereo particle image velocimetry in a plane perpendicular to the free stream flow of the wind tunnel. The planes are then stacked and the files stored here represent volumes. We defined a righthanded coordinate system with x in the downstream direction and z vertically upwards. The data consists of six columns, three giving the coordinates (m) of the vectors and three giving the flow (m/s) along the different axes.
Files
Steps to reproduce
Animals Data from three individuals of European nightjar (Caprimulgus europaeus) flew in our wind tunnel (Pennycuick et al., 1997). The morphometrics of these individuals are listed in the table below (Tab. 1). Procedures were approved by the Malmö-Lund animal ethics committee (M 33-13). Table 1. Morphometric information for the three European nightjars included in the analysis. Individual Mass (kg) Wing span (m) Wing area (m2) Wing loading (N/m2) Aspect ratio Left 0.0579 0.566 0.04225 13.4 7.6 Right2 0.0789 0.562 0.04420 17.5 7.1 Alfa 0.0775 0.558 0.04224 18.9 7.7 Experimental setup The animals were flown at speeds (U∞) ranging from 2.6 m/s to 11 m/s in approximately 2 m/s intervals. To capture the flow induced by the birds we used a standard stereo particle image velocimetry (sPIV) setup with a plane perpendicular to the wind tunnel flow, as described in (Johansson et al., 2018a). The field of view was approximately 47 x 55 cm (w x h) with a vector resolution of 2.5 vectors per cm and a sampling frequency (fs) of 640 Hz. PIV settings The PIV analysis was done in Davis 8 (LaVision, Göttingen, Germany). Before the analysis we preprocessed the images using subtract sliding background (size 8). For the velocity vector determination we used a decreasing box size, starting with 64x64, at 50% overlap and finishing with 3 times 32x32 boxes with 50% overlap. We used the GPU option, starting with a maximum shift of 5 pixels decreasing to 1 pixel in the final steps and used an initial shift of 3 pixels in the streamwise direction in the first round. Between each round we applied a two-times universal outlier detection, removing vectors deviating more than 2 standard deviations (SD) from their neighbors and reinserting the second-choice vector if deviating less than 3 SD from its neighbors. After filling up empty spaces we performed an optimal smoothing. After the analysis was done, the vector fields were post processed using a “Strongly remove and iteratively replace” filter with vectors removed if differing more than 2 SD from their neighbors and inserted if less than 3 SD. Empty spaces were filled by interpolation and a single “Optimal” smoothing was performed.
Institutions
- Lunds Universitet Biologiska institutionenLund
Categories
Funders
- Lund UniversitySkåne, Sweden