Probably black vertical dotted lines for each velocity profile correspond to u=0, meaning that they are x/H values.

Anton Noev Thank you for your reply. Yes, I think you're right but there is no more scale for the rest points. Any idea?

Following. Interested question.

@Roh_Dargahi there is scale on the horizontal axis if I get you right

Anton Noev Yes but we cannot use it, can we? for example let's assume we need to export the data of the third profile. It apears that u=0 of the third profile is laid on 6. Thus we have 6=x/H+10u/U. Ok, as I said u=0 then we have 6=x/H. If assume that H=2 then x=12. So what? doesn't make any sense to me. Maybe I'm wrong and I need to assume that the abscissa (horizontal axis) is the x in the equation. e.g. for 7 we have 7/2+10u/U=0 => 10u/U=-7/2 =>u/U=-7/20 It's negative. What do you think? What's your idea? How can we obtain u/U for each axis?

Hi Roh

are the different profiles plotted at successive vertical sections ie u=f(x=cst,y) or u/Ub=g(x=cst/H,y/H)? if so then it is strait forward. for each profile, x is constant and is deduced from the position of the vertical grey dashed line for which it corresponds. Then u/Ub is the difference between the horizontal position of any point of the chosen profile and the position of the corresponding vertical grey dashed line.

Deset Roh,

the best and fastest way is to ask the authors od the paper. Maybe they are willing to send you numerical data. Regads, Zdravko

You can perhaps use getdata graph digitizer

You can use the software of xy extract