WORK PLANS - Laserfiche WebLink

Home Browse Search

ARCADIS GERAGHTY&MILLER requires an investigator to make a calculation which partitions the total amount reported in the sample analysis into the amount sorbed and the amount in the sample which exists in the pore liquids as a dissolved fraction. Next a calculation must be made which relates the constituent amount sorbed to the amount that is desorbed and the rate at which the desorption (into the dissolved phase) occurs. Next, the rate of advection in the moving groundwater must be identified. Next, the rate of migration of a constituent must be related to the rate of movement of site groundwater by a retardation factor. This rate of migration must be further modified to account for the amount of attenuation of the compound due to dispersion and biodegradation. This series of calculations must be performed for each constituent of interest which is identified and quantified in the soil analysis. At each step of the calculation process, there are inherent and significant sources of error which ultimately make this quantification process highly uncertain. For example, the above- described partitioning calculations required the use of a partitioning coefficient. The value of the partitioning coefficient can only be obtained experimentally and varies significantly with soil type. In instances where experimentally determined partitioning coefficients are not available, they can be estimated using regression equations. The partition coefficient has been recognized as a key parameter in predicting the environmental fate of organic compounds. The uncertainty (error) associated with estimation of the partitioning coefficient, when one is not able to relate the site soil to one identified under carefully controlled laboratory conditions can be orders of magnitude (ie. greater that a factor of 10). A second example of the uncertainty in predicting the potential impact on groundwater based on a soil sample is that the hydraulic conductivity of the medium must be know in order to predict the advective rate of transport. Since the value of hydraulic conductivy varies significantly with soil type and even spatially within a soil type, the error associated with a calculation dependent upon assuming a hydraulic conductivity value or spatially extrapolating a measured hydraulic conductivity value can also be orders of magnitude. Theses errors ultimately are multiplicative and can result a highly misleading assessment of the potential for future impact to downgradient groundwater. Because the task of hand-calculating predictions of future downgradient groundwater impact is so onerous, some investigators have utilized numerical computer models to ease the calculation work load. However, it is important to emphasize that while the burden of the onerous work load has been removed, the uncertainty attendant to the process described above has not been removed. To make matters worse. the uncertainty is hidden by the