n through 
rn region, 
creating a 
aluate the 
ind use 
  
Moraes, Jener Fernando 
  
2.1 Soil Map 
The methodology iniciates by a soil survey resource inventory comprising aerial photograph interpretation, field survey 
and laboratory work. In total, 35 soil samples were collected in 0-20cm and 80-100 cm depth. Chemical and physical 
analyses were performed. Soil survey inventory showed that Ultisol and Alfisol (Table 1) are the predominant soil 
types, comprising 90% of the total area and the predominant parent material is the sandstone. These soils are very 
susceptible to erosion due to the great difference in clay content between A and B horizons. 
  
  
  
Soil Type Occurrence K-factor T-factor 
(ha) M.J.mm.ha" k'! (T.ha! yr!) 
Latossolos Vermelho Vermelho Escuro — LE (Oxisol) 202.26 0.0178 15.0 
Latossolos Vermelho Amarelo — LV — (Oxisol) 103.07 0.0172 14.2 
Podzólico Vermelho Escuro — PE — (Alfisol) 80.49 0.0357 77 
Podzólico Vermelho Amarelo distrófico — PVd — (Ultisol) 176.85 0.0419 7.7 
Podzólico Vermelho Amarelo distrófico/mesotrófico — PVe — (Ultisol) 614.19 0.0419 77 
Solos Hidromórficos — (Aquents) 88.1 --- --- 
TOTAL 1263.00 
  
Table 1. Soil types occurrence, K-Erodibily factor and T- maximum tolerance of soil loss accepted 
2.2 Topographic Map 
The area topography is characterized by undulate hills with altitudes ranging between 410 and 520 m. A detailed 
topographic map (with vertical distance of level curves between 10 meters) was digitized in Ilwis. A contour 
interpolation was performed to generate the Digital Elevation Model (DEM). Slope and slope length map were 
obtained from DEM. Slope length was obtained according to the equation: 
SL = 740.05*S 12312 (1) 
where: SL = Slope length and S = Slope (%) 
2.3 Land Use Map 
One quadrant of digital Landsat thematic map image (Path:222, Row 75) was acquired in 1998. Image processing was 
performed using ILWIS. The image enhancement techniques consisted of linear contrast modification, image filtering 
operations and band ratio. A principal component analysis was performed on the TM bands 3, 4, 5 and 7. Considering 
the characteristics of the vegetation cover in the study area that presents land use classes in different development stages 
we have calculated the normalized vegetation index, using the following band ratio: 
IVN = 128*[1+(TM4-Tm3)/(TM4+TM3)] Q) 
Using a color composition (TM bands 4, 3 and 7), training areas were sampled for the image classification. Ground 
information collected with GPS showed that land cover in the area include: Forest, Reforestation, Sugarcane, Pasture, 
Corn, Coffee and some fruit-bearer. A MAXVER supervised classification was applied. Some confusion about land use 
classes were reduce with visual analysis interpretation performed on the color composition image and on the aerial 
photographs. 
2.4  USLE-factors 
The Universal Soil Loss Equation (Wischmeier & Smith, 1978) was used to estimate the average extent of soil loss and 
to elaborate soil erosion risk map and land use planning map. In this case the USLE parameters were obtained as 
follow: 
(1) Rainfall intensity (R-factor) — It was considered as homogeneous for the watershed with a value of 6757MJ.mm/ha.h 
(2) Soil erodibility (K-factor) was calculated according to Bertoni & Lombardi Neto (1992) 
(3) Topography (LS-factor) was determined by the equation: LS — 0.00984*L?? *S!18 (Bertoni and Lombardi Neto, 1992) 
(4) C (Land-use factor) was obtained from the land-use map according to Bertoni & Lombardi Neto (1992) 
(5) P (P-factor) was obtained by the methodology proposed by Lombardi Neto (in preparation), and varied between 0 to 1. 
  
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000. 897