The rms slope value derived from the Hagfors model is an estimate
of the slope distribution of the surface at scales of tens to
hundreds of meters. Comparisons of lunar photogrammetric data
and radar echoes show some degree of correlation between the two
methods of measuring slopes, but the baseline over which the radar
results apply is often difficult to quantify [Moore et al., 1980].
For much of Venus, the Magellan rms slope values follow the trend
of the SAR echoes, with rougher surfaces having higher slopes
and higher 
.
Comparing these data with terrestrial analog surfaces has not
been simple, and numerous questions as to the derivation and interpretation
of such slope estimates remain [Simpson and Tyler, 1982;
McCollom and Jakosky, 1992;
Campbell and Garvin, 1993].
Published maps produced by the Stanford research group [Tyler et al., 1992],
using different techniques than those used to make the current
GsDR dataset [Ford and Pettengill, 1992],
suggest that the choice of data reduction technique may strongly
affect the estimate of rms slope in rough areas. Rms slope is
nonetheless a valuable measure of angular spreading in the near-nadir
radar echo, which can be a useful diagnostic of inhomogeneity
within the footprint or of possible porous terrain units [Campbell and Rogers, 1994].
The average value and range of this parameter, which are part
of the anc_data output, should be included in the map unit
description.