« Back

Abstract: Variability of soil resources in space and time can be a useful indicator in evaluating desertification risk in arid and semi-arid regions. We have compared the coefficient of variation (CV) and the fractal dimension of spatial variation (D) for some selected properties of soils with a slope of <3°, designated as mild-to-moderate sloping (MMS) and soils with the slope of >3°, designated as moderate-to-steep sloping (MSS) soils, in a 1,041.2 ha catchment located in North-Central Anatolia, Turkey. The study area was sampled based on a random sampling scheme, taking a total of 142 geo-referenced samples from 0-0.30 m soil depth. All soil samples were analyzed for soil properties of electrical conductivity (EC), pH, soil organic matter (SOM), sand, silt, clay, and coarse material, cation exchange capacity (CEC), crusting index (CI), penetration resistance (PR), and soil erodibility factor (K). The spatial variation of soil variables was characterized in MHS and SMS soils by CV, semivariogram analysis, and D. In general, higher CV values occurred in SMS soils, and the nugget effect values calculated for these variables was lower in SMS soils, revealing a greater spatial dependency of these variables in these soils, and a greater potential for desertification in these areas. Variables with a stronger spatial structure had a higher CV and a lower D. In general, higher CV values and lower D values occurred in SMS soils, indicating a greater desertification potential of these soils. Strong spatial distribution of soil properties in both slope classes suggested that D should be preferred over CV in evaluating the desertification risk in these areas.

Keyword: Desertification risk, probabilistic fractals, coefficient of variation, self-similarity, slope steepness