Dataset for 'Investigation of turbulent flow over irregular rough surfaces using direct numerical simulations'

This dataset is associated with the PhD thesis: Thakkar, M. (2017) Investigation of turbulent flow over irregular rough surfaces using direct numerical simulations (PhD Thesis), University of Southampton, Southampton, UK Dataset description: each of the 17 folders comprises of five .csv files containing figure data from chapter 6 in the thesis. Files in each folder have the same naming convention with only the first few characters changing depending on the sample name. For example, folder s1 contains the following files. i) s1_Fig6p1.csv - consisting of data from Figure 6.1 (mean streamwise velocity profiles for the 17 rough surface samples). ii) s1_Fig6p2.csv - consisting of data from Figure 6.2 (mean streamwise velocity defect profiles for the 17 rough surface samples). iii) s1_Figs6p5-6p8-6p11-6p14.csv - consisting of data from Figures 6.5, 6.8, 6.11 and 6.14 (Reynolds streamwise, spanwise, wall-normal and shear stress profiles for the 17 rough surface samples). iv) s1_Fig6p17.csv - consisting of data from Figure 6.17 (TKE profiles for the 17 rough surface samples). v) s1_Figs6p19-6p23-6p26-6p30.csv - consisting of data from Figures 6.19, 6.23, 6.26 and 6.30 (dispersive streamwise, spanwise, wall-normal and shear stress profiles for the 17 rough surface samples). Folder s2 contains files s2_Fig6p1.csv, s2_Fig6p2.csv, s2_Figs6p5-6p8-6p11-6p14.csv, s2_Fig6p17.csv, s2_Figs6p19-6p23-6p26-6p30.csv, and so on for the remaining folders. For files containing data from more than one figure (for example, s1_Figs6p5-6p8-6p11-6p14.csv), the x-axis values of each plot are the same and are stored in the first column (titled 'z/delta') of the data matrix. Only the y-axis values differ depending on the relevant statistic and are stored in subsequent columns with appropriate titles. Some data may contain NaN or Inf values in their first and last few rows. These occur due to small errors during spatial averaging arising from the differences between fluid and solid regions of the computational domain. These values do not have a significant impact on the results and may be ignored.