The concentration of an element in soil is the end result of its input and output. The input includes parent material, atmospheric deposition, fertilizers, agrochemicals, organic wastes and inorganic pollutants. The output includes removal in harvested crops, leaching and volatilisation. Thus the natural variation of potentially toxic elements (PTE) concentration is a preliminary step in assessing PTE contamination in soils. Data are scarce on qualitative and quantitative trace elements content of Argentinian soils. The soils of Catamarca province (NW of Argentina) are thought to be generally non-contaminated. Nevertheless, an intensive urbanization of agricultural soils around Catamarca provincial capital has led to a situation where agricultural soils are entrapped within suburban areas. Therefore the influence of urban and industrialised environments on the accumulation of metals in agricultural soils has been investigated. In general, a distinction can be made between empirical and statistical methods. The empirical background is mainly determined as a fixed value (mean or median). This determination requires expert knowledge about the geochemical behaviour of the investigated elements under the prevailing environmental conditions. No information is needed about the distribution function. Due to its lack of objectivity, the statistical approach was preferred. With respect to the statistical approach, same as most statistical tests it is helpful in eliminating outliers and thus reducing the original data set to a “clean” data set. From this reduced data set, which is being addressed as anthropogenically undisturbed, the essential descriptive statistical parameters can be calculated. A representative set of thirty soil profiles of Catamarca were sampled. The samples were acid digested using aqua regia extraction technique. Cadmium, chromium, copper, lead, nickel and zinc were determined with inductively coupled argon plasma emission spectrometry. PTE content was calculated with the 4σ-outlier test, Iterative 2σ-outlier test and calculated distribution function described by Mataschullat et al. (2000) (table 1). The normal range of a sample is defined by upper values assessed by the mean +2σ, which means that 97% of the samples lie below this value. Only this type of data includes sufficient information about the natural scatter of the background. The upper limits obtained applying Iterative 2σ–technique and the calculated distribution function differ from those obtained from the original data set and the application of the 4σ–outlier test. Iterative 2σ–technique is more robust reducing background upper limit compared with the original data set, 4σ–outlier test and calculated distribution. Measured data above background are geochemical anomalies. Based in previous knowledge about the natural scatter of investigated elements it was verified that no anthropogenic contamination occurred. The obtained values are realistic and plausible with all applied statistics. All three methods seem appropriate for a fast and reliable evaluation of likely upper limits of background values.

Table. Upper limit of studied Catamarca soil PTE background

Methods	Potentially Toxic Elements (mg·kg-1,*mg·kg-1)
	*Cd	Cr	Cu	Ni	Pb	Zn
Original data set	212.4	43.8	35.8	28.4	16.5	106.4
Iterative 2σ-technique	173.6	38.8	33.9	26.2	15.0	96.3
4σ-outlier test	212.4	43.8	35.8	28.4	16.5	106.4
Calculated distribution	208.2	46.8	36.5	30,1	17,5	115,6