On Tue, Jun 30, 2009 at 06:02:34PM -0700, Scott Webster wrote: > I'm wondering about the units of the "y-axis" of the PSDF tool. In > gwyddion they are length^3. I have thought about this, and seem to > think that it should be length^4. Is this just some convention that > gwyddion has chosen? Is there a rationale or does anyone know of a > reference that explains it?
http://gwyddion.net/documentation/user-guide/statistical-analysis.html#power-spectral-density-1d The integral of one-dimensional PSDF W₁(k) is σ² [z²]. Therefore, the density units must be [z²x] because k units are [1/x]. If the lateral [x] and value [z] units happen to be the same, you get [length³]. > On a similar note, the (z scale) units of the 2D-FFT are in length. > Is this also "wrong?" I would expect length^3? The 2D FFT produces the modulus, not PSDF (which is the square of the modulus). The units could be [zx] but I am afraid a different areal normalization convention is used here which means it is just [z]. The formulas are in the user guide. Yeti ------------------------------------------------------------------------------ _______________________________________________ Gwyddion-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/gwyddion-users
