Environmental risk assessment (ERA) is a useful methodology to estimate the possible adverse effects to human health due to contaminants exposure. In the case of agricultural scenarios, this method requires knowing the concentrations of contaminants in soil solution and vegetation, among other parameters. This study aimed to develop multicorrelation models to estimate metal extractable from soil as a function of total metal concentration in soil and soil properties in a cattle manure application scenario. It also aimed to estimate metal concentrations in plant by soil-plant uptake factors (UF). All the multicorrelation models obtained were significant, ranging R(2) values from 0.44 for Cd to 0.92 for Cu. Soil-plant UF were an adequate method for the estimation of metal concentration in plant, since the relationship between the soil-plant UF and the extractable metal concentration from soil was significantly described by a power model, for all the heavy metals.