PARENT SESSION
Oral Session #60: Remote Sensing and GIS. Presiding: A. Gallant.
Thursday, August 9, 2001. 8:00 AM to 12:00 PM. Hall of Ideas E.
GIS, expert knowledge, and fuzzy logic-based approach to soil mapping over areas with moderate relief.
Zhu, A-Xing1, Simonson, Duane2, Qi, Feng1, Shi, Xun1, English, Edward1, 1 2
ABSTRACT- A GIS, expert knowledge, and fuzzy logic-based approach to soil mapping was developed. The degree of success for this approach (referred to as SoLIM) largely depends on the ability to characterize the environmental conditions using GIS techniques. It has been shown that the approach was successful when applied to area with a strong environmental gradient due to the fact that soil formative environmental conditions can be easily characterized using GIS with 30-meter USGS digital elevation models (DEM). However, characterizing soil formative environmental conditions using the 30-meter DEMs presents challenges for areas with moderate to gentle environmental gradients. This paper examines the applicability of this approach to areas with a moderate environmental gradient with a regular USGS 30-meter DEM and a 10-meter DEM. A recently produced order 2 level soil map was used to compare with the results from the SoLIM approach. Field validation shows that with a 10-meter DEM the SoLIM approach was able to achieve 83% accuracy while the order 2 soil map was only at 66% accuracy. For areas mapped as complexes on the order 2 soil map, the SoLIM approach achieved 89% accuracy while the conventional soil map was only at 73% accuracy. However, with a 30-meter DEM, the SoLIM approach only achieved a moderate improvement over the order 2 soil map.
KEY WORDS: GIS, Expert Systems, Fuzzy Logic, Soil Mapping