Upscaling retardation factor in hierarchical porous media with multimodal reactive mineral facies.

Aquifer heterogeneity controls spatial and temporal variability of reactive transport parameters and has significant impacts on subsurface modeling of flow, transport, and remediation. Upscaling (or homogenization) is a process to replace a heterogeneous domain with a homogeneous one such that both reproduce the same response. To make reliable and accurate predictions of reactive transport for contaminant in chemically and physically heterogeneous porous media, subsurface reactive transport modeling needs upscaled parameters such as effective retardation factor to perform field-scale simulations. This paper develops a conceptual model of multimodal reactive mineral facies for upscaling reactive transport parameters of hierarchical heterogeneous porous media. Based on the conceptual model, covariance of hydraulic conductivity, sorption coefficient, flow velocity, retardation factor, and cross-covariance between flow velocity and retardation factor are derived from geostatistical characterizations of a three-dimensional unbounded aquifer system. Subsequently, using a Lagrangian approach the scale-dependent analytical expressions are derived to describe the scaling effect of effective retardation factors in temporal and spatial domains. When time and space scales become sufficiently large, the effective retardation factors approximate their composite arithmetic mean. Correlation between the hydraulic conductivity and the sorption coefficient can significantly affect the values of the effective retardation factor in temporal and spatial domains. When the temporal and spatial scales are relatively small, scaling effect of the effective retardation factors is relatively large. This study provides a practical methodology to develop effective transport parameters for field-scale modeling at which remediation and risk assessment is actually conducted. It does not only bridge the gap between bench-scale measurements to field-scale modeling, but also provide new insights into the influence of hierarchical mineral distribution on effective retardation factor.

Mesoscale carbon sequestration site screening and CCS infrastructure analysis.

We explore carbon capture and sequestration (CCS) at the meso-scale, a level of study between regional carbon accounting and highly detailed reservoir models for individual sites. We develop an approach to CO(2) sequestration site screening for industries or energy development policies that involves identification of appropriate sequestration basin, analysis of geologic formations, definition of surface sites, design of infrastructure, and analysis of CO(2) transport and storage costs. Our case study involves carbon management for potential oil shale development in the Piceance-Uinta Basin, CO and UT. This study uses new capabilities of the CO(2)-PENS model for site screening, including reservoir capacity, injectivity, and cost calculations for simple reservoirs at multiple sites. We couple this with a model of optimized source-sink-network infrastructure (SimCCS) to design pipeline networks and minimize CCS cost for a given industry or region. The CLEAR(uff) dynamical assessment model calculates the CO(2) source term for various oil production levels. Nine sites in a 13,300 km(2) area have the capacity to store 6.5 GtCO(2), corresponding to shale-oil production of 1.3 Mbbl/day for 50 years (about 1/4 of U.S. crude oil production). Our results highlight the complex, nonlinear relationship between the spatial deployment of CCS infrastructure and the oil-shale production rate.