The only roughness parameter expressed in degrees is the RMS gradient of surface slope (Rdq or Sdq, in 2d or 3D characterization standards, respectively).
I presume that the friction angle between sand particles is due to high angularity of the particles shape. Is it correct? In that case, I would measure Sdq and filtrate it with Gaussian filter where the long cut-off is set to the radius of the measured sand-particle. It would give the slope (deg.) of average asperity on the sand surface. In simple words, Sdq is responsible for mechanical interlocking. If Sdq is high, then the summits on the surface are sharp and it will not slide easily.
My apologies if I misunderstand the inter-particles friction angle.
In addition to Ruslan Melentiev's answer, you may refer to "Table 1. Typical applications for various 3D parameters" by following this practical application note:
M. Zecchino, " Characterizing Surface Quality: Why Average Roughness is Not Enough ", 2010 - https://www.bruker.com/fileadmin/user_upload/8-PDF-Docs/SurfaceAnalysis/3D-OpticalMicroscopy/ApplicationNotes/AN511-Characterizing_Surface_Quality-Why_Average_Roughne.pdf
A correlation exists between roughness parameters and friction angle. Follow:
M. Angela, " Correlation between roughness and friction angle of open discontinuity surfaces ", 2011 - https://pdfs.semanticscholar.org/600a/c4ab7e371bcfc389b05f156f6e78a8fec90a.pdf?_ga=2.217552726.1932399560.1569934203-1356479370.1553781158
Sazzad et al., " Effect of Interparticle Friction Angle on the Stress-Dilatancy Responses of Granular Assembly by DEM ", 2017 - https://www2.kuet.ac.bd/JES/images/files/v42/5-JES_100.pdf