although the Improved von karman question is not directly implemented in TurbSim ( even if there is a chunk of code somewhere), I wonder if you may anyway shed a light about the following question.
If I look at the ESDU “Characteristics of atmospheric turbulence near the ground Part III: variations in space and time for strong winds (neutral atmosphere)”, at pag. 21 (pag. 24 pdf version) you’ll be able to spot the assumed coherence function for the longitudinal component. In the reported link you’ll find the same definition, but given as used in Bladed (at pag. 63 of the pdf format)
Within the formula, the modified Bessel function of second order Kv(x) must be called on a variable eta which is depending on the distance among the grid points. But the Bessel function will retrieve infinity for each element laying on the diagonal (mainly all those distance entries with 0 value). So, at the end of the day, the question is: is that correct? I would say not at a real first glance, but I am not a real master in this area.
I am really interested in understanding what I am missing, since I do want to implement the improved von Karman model as part of my own turbulent wind generator in MATLAB.
I thank you in advance for considering my question and I look really forward to gathering info from you.
Best regards from Germany,