Read Me For Rough wall PIV dataset supporting the publication " Effect of pressure gradient histories on turbulence characteristics of turbulent boundary layers over smooth and rough walls"’
Dataset DOI: 

Date that the file was created: 01/26
-------------------
GENERAL INFORMATION
-------------------
ReadMe Author: Thomas Preskett, University of Southampton
Date of data collection: 
06/22-01/24
--------------------------
SHARING/ACCESS INFORMATION
-------------------------- 
Licenses/restrictions placed on the data, or limitations of reuse: CC-BY
Recommended citation for the data: Please use the above DOI
This dataset supports the publication:
AUTHORS: T. Preskett
TITLE: Effect of pressure gradient histories on turbulence characteristics of turbulent boundary layers over smooth and rough walls 
Type: Thesis
Links to other publicly accessible locations of the data:
Links/relationships to ancillary or related data sets: 
DATA & FILE OVERVIEW
This file contains rough-wall PIV data for five angles of attack at 20 m/s. It includes the instantaneous fields at each position, as well as the mean flow fields and boundary-layer properties. 
 Each angle of attack has its own folder and in each folder is the following -
There are four folders corresponding to each position in which PIV data was taken, see chapter 3 of “Effect of pressure gradient histories on turbulence characteristics of turbulent boundary layers over smooth and rough walls” for more detail. 
In the position folder is a Coords file which contains X and Y grids corresponding to the position of the velocity vectors. The coords file also contains the mean viscosity and density grids, which can be used for calculations. Also UInfinity (U0_mean) is given from the reference pitot tube when combined with other positions. 
Also in each position folder are approximately 2000 vector fields containing U and V corresponding to the X and Y coordinates, with a file name in the format XXXX.mat
Finally there is a Mean_Flow_Fields.mat which contains the U_mean, V_mean, X, Y, nu_mean, rho_mean, uu_mean, vv_mean and uv_mean along with U0_mean which are averaged across the 2000 images. 
In each AOA folder there is also a Mean_Flow_Fields_AOA_X.mat where X is the angle of attack. This contains U_mean, V_mean, X, Y, nu_mean, rho_mean, uu_mean, vv_mean and uv_mean which have been stitched together from the four positions. The data is normalised by the corresponding U0 value from a given position then converted back using the mean U0 value from all four positions given by U0_mean in Mean_Flow_Felds_AOA_X.mat.

In the main PIV folder is Summary_PIV_All_AOA_RW.pkl which contains a dictionary, the first layer is the string of the angle of attack, for each angle of attack the following keys are used. 
'X', 'Y', 'U_mean', 'V_mean', 'rho_mean', 'nu_mean', 'uu_mean', 'vv_mean', 'uv_mean', 'Uinf', 'delta_array', 'deltastar_array', 'theta_array', 'H_array', 'U99'

Note:
Pickle (.pkl) files can be read into python using the following command
with open(dir_pkl, 'rb') as f:
      data = pickle.load(f)
matlab data files can be read in python using: 
data = mat73.loadmat('file.mat')