Chunmiao Zheng. . 05-37668 Steven M. Gorelick
After many years of research, it is evident that the prevalent, advection-dispersion model (ADM) does not adequately describe solute behavior in media with even modest heterogeneity. The ADM is based on the premise that groundwater velocity variations are due to heterogeneity in hydraulic conductivity, K, which is assumed to be correlated but otherwise random. From a geologic perspective, this premise is often unjustified because connectedness rather than randomness is expected in heterogeneous aquifers. Although a growing body of field evidence now points to the critical importance of mass transfer controlled by connected conduits in low-K matrix, there has been no systematic evaluation of non-ADM alternative models. The goal of this project is to investigate alternative transport model formulations that are based on the mass-transfer concept, and determine the best means to obtain the parameter values controlling rate-limited migration of solutes between relative aquifer conduits and the surrounding matrix. A comprehensive set of laboratory experiments and field tests at the Macro-Dispersion Experiment (MADE) site in Mississippi will be conducted to assess and contrast alternative theoretical models of solute transport in aquifers containing connected high-K networks. The research project will address four important questions: (1) What is the nature, geometry, and scale of connected high-K preferential flow channel networks in a heterogeneous fluvial aquifer? (2) How are such flow channel networks (and flow barriers) related to the texture, structure and grain-size distribution of fluvial sediments? (3) Without specific knowledge of conduit-network geometry or model calibration, can a previous theoretical relation determining the value of the mass-transfer coefficient in the embedded network systems be verified in the laboratory and field? (4) What is the most appropriate model to represent solute transport in aquifers containing small-scale connected high-K networks, and how can the parameters for that model be obtained from readily available field and laboratory data? The proposed project will allow us to establish a comprehensive, sound conceptual and modeling framework that is useful in practice to account for the controlling effects of connected high-K networks. In hydrogeology, this is important to the accurate prediction of contaminant transport, management of groundwater quality, and understanding of the migration and distribution of solutes in natural environments. This work is relevant to other areas of hydrology and to other areas of science. The topic of transport and mass transfer involving preferential pathways has important cross-over value to other disciplines, such as botany and animal physiology where systems of highly conductive networks embedded in less conductive matrix are commonplace. In hydrology, the mechanism of slow mass transfer occurs when dendritic networks of surface water interact with groundwater in deltas and estuaries. In plant and animal tissue, delivery of nutrients and drugs involves analogous mass-transfer processes, albeit at different scales.
