CONICET | Buscador de Institutos y Recursos Humanos

Since the 1980s, pollution became a subject of increasing interest and the scientific community was aware of pollution consequences. There was a need for monitoring techniques that was approached by several fields of research. Although the man-made contribution of heavy metal can be studied by careful geochemical methods (time-consuming, laborious and costly), magnetic monitoring constitutes an alternative tool for pollution studies. In particular, in our previous studies (e.g. Chaparro et al., Environ. Geol. 54: 365371, 2008), several statistical analyses were performed, univarite and multivariate techniques, revealing a link between magnetic and chemical variables.In this contribution, continuing the above-mentioned pollution studies, we report on a fuzzy rule-based system (FRBS) in order to model pollution in soils. Modelling based on fuzzy set theory and inference techniques seems to be an adequate mathematical tool for this problematic. The used dataset (n= 46) comprises magnetic and chemical variables; and the soil samples were collected in Marambio station from Antarctica (64º14S; 56º37W). The dataset comprises unpolluted, moderately and polluted samples. For this work, we selected, on a set of nine magnetic variables and seven chemical variables, the following relevant variables, magnetic concentration variables: mass-specific magnetic susceptibility (X) and anhysteretic remanent magnetization (ARM); magnetic features-dependant variables: KARM/K-ratio and S-ratio; and a chemical variable: the Tomlinson pollution load index PLI (a composite index using Cr, Ni, Cu, Zn and Pb). The model was test using a fitting goodness coefficient (R2=0.91), indicating a very good fitting between data and model

enviar mensaje