Procedure for Calculating the Surface Configuration Index

This is the "secret-decoder" message for calculating the surface configuration index contained

in the soon-to-be-online Excel workbook describing all three map surface comparison approaches…

1) Calculate the "normalized" difference in slope:

Most grid-based GIS packages calculate % slope. Percent slope can be converted to

degrees slope by DEGREES(ARCTAN(%slope/100). The difference between the two slopes

is obtained by ABS(M1_DegSlope - M2_DegSlope). The difference in slope angles can be

normalized between 0 to 100 by (((Diff_DegSlope - min) * 100) / (max - min)), where

min = 0 and max = 90 for degree_slope possible range.

2) Calculate the "normalized" difference in azimuth:

Most grid-based GIS packages calculate precise aspect in degrees azimuth. Degrees azimuth

must be converted to radians by RADIANS (Deg_Azimuth). The difference between two

azimuths can be calculated in degrees by

DEGREES( ACOS( SIN(Map1_RadAzimuth) * SIN(Map2_RadAzimuth) +

COS(Map1_RadAzimuth) * COS(Map2_RadAzimuth) ) )

Normalized between 0 to 100 by (((Diff_DegAzimuth - min) * 100) / (max - min)), where

min = 0 and max = 180 for degree_azimuth possible range.

3) Calculate the average "normalized" differences in slope & azimuth:

Simple arithmetic average of the normalized differences in slope and aspect.

(Optional) Most interesting "problem"…

…Note: the azimuth "361" represents perfectly

flat, not 1 degree NE. In the column to the left,

TRUE identifies such locations. How should

one handle these "no azimuth" locations--

Disregard azimuth differences in calculating the

Surface Configuration Index?

On another note, what about a weighted average

of slope and azimuth differences? Like a log function?

What do you think?