Goal Oriented Parametrization of Transport Models – Recent Advances in Field Characterization and Data Analysis Methods

Monday, October 5, 2015: 11:45 AM
Alraune Zech1, Sabine Attinger1, Alberto Bellin2, Vladimir Cvetkovic3, Gedeon Dagan4, Peter Dietrich1, Aldo Fiori5, Yoram Rubin6 and Georg Teutsch1, (1)Helmholtz Centre for Environmental Research UFZ Leipzig, Leipzig, Germany, (2)University of Trento, Department of Civil, Environmental and Mechanical Engineering, Trento, Italy, (3)Royal Institute of Technology, Stockholm, Sweden, (4)Tel Aviv University, Tel Aviv, Israel, (5)Universita' di Roma Tre, Dipartimento di Ingegneria, Rome, Italy, (6)University of California Berkeley, Berkeley, CA, United States
Abstract:
Despite the achievements of three decades of research on transport processes in heterogeneous formations, ADE based transport models assuming homogeneity are still widely used in practice for the prediction of solute movement in aquifers. Due to the lack of adequate field data, these models are commonly of deterministic nature and not accounting explicitly for prediction uncertainty due to heterogeneity. Furthermore, they are limited in the representation of mixing and dispersion, which are essential for capturing important properties of the plume, such as probability of exceeding a given concentration threshold, early time arrivals and tailing. Field scale dispersion coefficients are commonly estimated by simple rules of thumb.

So far, the time and effort needed to acquire the statistical parameters for an adequate representation of the subsurface heterogeneity were generally prohibitive for field sites. Applications were limited to a few selected research sites. However, very significant conceptual and technological advancements have been achieved during the past 10-15 years which led to an entirely new perspective on data acquisition quality and cost effectiveness [Bohling et al., 2012 (WRR), Brauchler et al., 2013 (WRR)].

Our study uses data gathered during a detailed literature review of well-defined field experiments [Zech et al., 2015 (under revision)] and investigates the applicability of some of the new characterization technologies in a few extensively investigated field sites, such as MADE. In analyzing transport at these sites we discuss the characterization requirements depending on the specific goal of modeling: (a) a risk analysis targeted study which focuses on first arrivals of a contaminant plume, (b) a general plume propagation description which focuses on average mass movement and (c) a clean-up targeted study where the behavior of the tail of the contaminant plume is relevant. Similarly, we discuss the appropriate goal oriented data analysis and modeling methods.

Ultimately, we want to show that targeted site characterization investigations can provide efficiently the data for an adequate representation of subsurface heterogeneity toward modeling transport in geological formations.