Conceptual model of leachate migration in a granular aquifer derived from the integration of multi-source characterization data (St-Lambert, Canada)Modèle conceptuel de la migration d’un lixiviat dans un aquifère poreux, intégrant des données de caractérisation provenant de sources multiples (St-Lambert, Canada)Modelo conceptual de migración de lixiviados en un acuífero granular a partir de la integración de datos de caracterización de múltiples fuentes (St-Lambert, Canada)Modelo concetual de migração de lixiviados num aquífero granular, resultante da integração de dados de caraterização com diferentes origens (St-Lambert, Canadá)

Numerical models encompassing source zones and receptors, based on representative conceptual models and accounting for aquifer heterogeneity, are needed to understand contaminant migration and fate; however, aquifer characterization seldom provides the necessary data. This study aimed to develop a workflow for field characterization and data integration, which could: (1) be adapted to the definition of subwatershed-scale aquifer heterogeneity (over 10 km2) and (2) adequately support mass transport model development. The study involved the field investigation of a shallow granular aquifer in a 12-km2 subwatershed in Saint-Lambert-de-Lauzon, Canada, in which a decommissioned landfill is emitting a leachate plume managed by natural attenuation. Using proven field methods, the characterization sequence was designed to optimize each method in terms of location, scale of acquisition, density and quality. The emphasis was on the acquisition of detailed indirect geophysical data that were integrated with direct hydraulic and geochemical data. This report focuses on the first qualitative and geostatistical data integration steps of the workflow leading to the development of a hydrogeological conceptual model. This is a prerequisite for further integration steps: prediction of hydrofacies and hydraulic conductivity (K), geostatistical simulations of K, studies of geochemical processes and numerical modeling.