Présentations Colloques

Session 8.10: Recent developments in groundwater modeling and mathematical tools in Hydrogeology
Boisson (orateur)
Using the power-law behavior of density-driven solute breakthrough curves for contaminant plume development reconstitution and prognosis in a large alluvial aquifer
The upper Rhine valley alluvial aquifer is one of the largest groundwater reservoirs of Western Europe and constitutes a major water resource for the Alsace region in France. Geological formations below alluvium were exploited for potassium hydroxide (potash) extraction through underground mines. As a consequence of this exploitation, large amounts of extracted materials made of sodium chloride and mine waste have been stored at the land surface. Fifteen heaps have been created to store residuals of 568 million tons extracted during the period 1910-2002 on an area of 220 Km2. Due to rainfall and treatment activities, large quantities of chloride infiltrated to the aquifer from the waste heaps with concentration up to 40 g L leading to a widespread contamination of the alluvial aquifer by high density plumes. Contaminant plumes migrated from the heaps downgradient and, as a result, observed chloride concentrations in treatment and observation wells are similar to the shape of breakthrough curves (BTC) resulting from classical artificial tracer point injections. These BTCs recorded long tailing of more than 15 years following a power-law behavior. Despite various studies, reasons for such tailing remained partly unknown. Based on chloride BTCs from 215 treatment observation wells over the time period 1988-2015 and interpretation of the BTCs power law behavior, we were able to reconstitute the plumes historical development and behavior. Numerical density-driven contaminant transport simulations, supported BTCs interpretations and enabled to forecast plumes future evolution. Hence we developed a methodology based on power-laws as a proxy to estimate plume behavior in a context where density-driven numerical modeling is not a realistic approach at the field site scale. Such a methodology is an efficient and cost-effective tool for contaminant transport evaluation for contaminated sites impacted by long-term contamination and where long concentration time series are available from observation pumping wells.