Research Article

Comparison of Three Supervised Learning Methods for Digital Soil Mapping: Application to a Complex Terrain in the Ecuadorian Andes

Table 1

Terrain parameter calculation with SAGA GIS software.

ParameterModuleReferenceSaga author/year

1Elevation (El)Fill sinks [9]Wichmann 2003
2Slope (α)Slope, aspect, curvature [10]Conrad 2001
3Aspect
4Valley depthRelative heights and slope positionsBöhner and Conrad 2008
5WindWind effectBoehner and Ringeler 2008, Conrad 2011
6CIConvergence index [11]Conrad 2003
7Specific catchment areaFlow tracing/kinematic routing algorithm (KRA)[12]Conrad 2001
8OFDStrahler order ≥5[13]Olaya 2004
Overland flow distance to channel network[14]Conrad 2001 & 2011
9SWISaga wetness index[15]Böhner and Conrad 2001
10LS factorLength of slope factor[16]Conrad 2003