Taken two agricultural lands as the study areas (loess, Haplic Luvisols; loamy to sandy soil, Eutric Cambisols), which are representative of the soils of northern Germany, this dissertation studies the upscaling of the adsorptive binding of heavy metals in soils and their variability by calculating the scale factors. The adsorptive binding of heavy metals in soils is mostly quantified by sorption isotherms with large variability at the field scale. The aim of this work is to search the correlation of sorption isotherms by means of scale factors between different heavy metals and further with physico-chemical soil properties, so that only a few measurements are necessary to make sufficient statements on heavy metal’s binding and mobility at field-scale. At both study sites, upscaling can capture the linear parts of sorption’s variability well. Scenario study discussed satisfied simulations of heavy metals transport process, where the scale factors are treated as the measure of variability. However, in the statistical and geostatistical studies, no significant correlations were found between the scale factors of different heavy metals and with physicochemical soil properties. Depending on the location and soil horizon, the correlation of scale factors between different heavy metals varied so different and not transferrable. In addition, the reference isotherm calculated directly from measurements did not match the sorption isotherm from a composite sample, which indicates that scaling is favorable to homogenous sites. Thus, the important finding in this dissertation can be summarized that the application of scale factors for heavy metal sorption isotherms, such as statistical or geostatistical evaluation, is limited only to specific case studies or a scenario modeling.