engleski

Application of fractals in spatial variability analysis of soil

In theory, fractal is a rough or fragmented geometric shape that can be subdivided in parts, each of which is a smaller copy of the whole. For this reason, fractals are often discontinuous and so have fractal dimensions D which is quantitative indicator of spatial variation phenomena. Fractal theory has been used to investigate and quantitatively characterize spatial variability of different soil properties, including EC, Cl-, pH, CaO, Humus, P2O5, K2O and soil texture. In total, n = 221 soil samples collected at a depth of 0 - 25 cm from a 391-ha irrigated agricultural field located in the delta of river Neretva in Adriatic sea, Croatia, were used in the study. The semivariogram methods were used to estimate parameter D. The objective was, to evaluate the appropriation of interpolation procedure for soil data mapping on the base of the calculated D values. This research showed that the soil properties, included in study, have high fractal dimensions, which vary over a wide range (D = 1, 6 – 2, 0). The large D value indicated that short-range sources of the variation dominate with high proportion of the irregular and largely uncorrelated soil variations. These results suggest that application of interpolation techniques, such as kriging, for making spatial distribution maps of soil properties with high D like pH, P2O5 and K2O are unsuitable.

fractal dimension; semivariogram; interpolation mapping; soil

nije evidentirano