Christopher J. Murray

Quantifying community assembly processes and identifying features that impose them

Spatial turnover in the composition of biological communities is governed by (ecological) Drift, Selection and Dispersal. Commonly applied statistical tools cannot quantitatively estimate these processes, nor identify abiotic features that impose these processes. For interrogation of subsurface microbial communities distributed across two geologically distinct formations of the unconfined aquifer underlying the Hanford Site in southeastern Washington State, we developed an analytical framework that advances ecological understanding in two primary ways. First, we quantitatively estimate influences of Drift, Selection and Dispersal. Second, ecological patterns are used to characterize measured and unmeasured abiotic variables that impose Selection or that result in low levels of Dispersal. We find that (i) Drift alone consistently governs ∼25% of spatial turnover in community composition; (ii) in deeper, finer-grained sediments, Selection is strong (governing ∼60% of turnover), being imposed by an unmeasured but spatially structured environmental variable; (iii) in shallower, coarser-grained sediments, Selection is weaker (governing ∼30% of turnover), being imposed by vertically and horizontally structured hydrological factors;(iv) low levels of Dispersal can govern nearly 30% of turnover and be caused primarily by spatial isolation resulting from limited exchange between finer and coarser-grain sediments; and (v) highly permeable sediments are associated with high levels of Dispersal that homogenize community composition and govern over 20% of turnover. We further show that our framework provides inferences that cannot be achieved using preexisting approaches, and suggest that their broad application will facilitate a unified understanding of microbial communities.

Physicochemical Characteristics of the Hyporheic Zone Affect Redd Site Selection by Chum Salmon and Fall Chinook Salmon in the Columbia River

Abstract Chum salmon Oncorhynchus keta and fall chinook salmon O. tshawytscha spawned at separate locations in a side channel near Ives Island, Washington, in the Columbia River downstream of Bonneville Dam. We hypothesized that measurements of water depth, substrate size, and water velocity would not sufficiently explain the separation in spawning areas and began a 2-year investigation of physicochemical characteristics of the hyporheic zone. We found that chum salmon spawned in upwelling water that was significantly warmer than the surrounding river water. In contrast, fall chinook salmon constructed redds at downwelling sites, where there was no difference in temperature between the river and its bed. An understanding of the specific factors affecting chum salmon and fall chinook salmon redd site selection at Ives Island will be useful to resource managers attempting to maximize available salmonid spawning habitat within the constraints imposed by other water resource needs.

Technetium reduction in sediments of a shallow aquifer exhibiting dissimilatory iron reduction potential

Pertechnetate ion [Tc(VII)O(4) (-)] reduction rate was determined in core samples from a shallow sandy aquifer located on the US Atlantic Coastal Plain. The aquifer is generally low in dissolved O(2) (<1 mg L(-1)) and composed of weakly indurated late Pleistocene sediments differing markedly in physicochemical properties. Thermodynamic calculations, X-ray absorption spectroscopy and statistical analyses were used to establish the dominant reduction mechanisms, constraints on Tc solubility, and the oxidation state, and speciation of sediment reduction products. The extent of Tc(VII) reduction differed markedly between sediments (ranging from 0% to 100% after 10 days of equilibration), with low solubility Tc(IV) hydrous oxide the major solid phase reduction product. The dominant electron donor in the sediments proved to be (0.5 M HCl extractable) Fe(II). Sediment Fe(II)/Tc(VII) concentrations >4.3 were generally sufficient for complete reduction of Tc(VII) added [1-2.5 micromol (dry wt. sediment) g(-1)]. At these Fe(II) concentrations, the Tc (VII) reduction rate exceeded that observed previously for Fe(II)-mediated reduction on isolated solids of geologic or biogenic origin, suggesting that sediment Fe(II) was either more reactive and/or that electron shuttles played a role in sediment Tc(VII) reduction processes. In buried peats, Fe(II) in excess did not result in complete removal of Tc from solution, perhaps because organic complexation of Tc(IV) limited formation of the Tc(IV) hydrous oxide. In some sands exhibiting Fe(II)/Tc(VII) concentrations <1.1, there was presumptive evidence for direct enzymatic reduction of Tc(VII). Addition of organic electron donors (acetate, lactate) resulted in microbial reduction of (up to 35%) Fe(III) and corresponding increases in extractable Fe(II) in sands that exhibited lowest initial Tc(VII) reduction and highest hydraulic conductivities, suggesting that accelerated microbial reduction of Fe(III) could offer a viable means of attenuating mobile Tc(VII) in this type of sediment system.

Variably saturated flow and multicomponent biogeochemical reactive transport modeling of a uranium bioremediation field experiment.

Three-dimensional, coupled variably saturated flow and biogeochemical reactive transport modeling of a 2008 in situ uranium bioremediation field experiment is used to better understand the interplay of transport and biogeochemical reactions controlling uranium behavior under pulsed acetate amendment, seasonal water table variation, spatially variable physical (hydraulic conductivity, porosity) and geochemical (reactive surface area) material properties. While the simulation of the 2008 Big Rusty acetate biostimulation field experiment in Rifle, Colorado was generally consistent with behaviors identified in previous field experiments at the Rifle IFRC site, the additional process and property detail provided several new insights. A principal conclusion from this work is that uranium bioreduction is most effective when acetate, in excess of the sulfate-reducing bacteria demand, is available to the metal-reducing bacteria. The inclusion of an initially small population of slow growing sulfate-reducing bacteria identified in proteomic analyses led to an additional source of Fe(II) from the dissolution of Fe(III) minerals promoted by biogenic sulfide. The falling water table during the experiment significantly reduced the saturated thickness of the aquifer and resulted in reactants and products, as well as unmitigated uranium, in the newly unsaturated vadose zone. High permeability sandy gravel structures resulted in locally high flow rates in the vicinity of injection wells that increased acetate dilution. In downgradient locations, these structures created preferential flow paths for acetate delivery that enhanced local zones of TEAP reactivity and subsidiary reactions. Conversely, smaller transport rates associated with the lower permeability lithofacies (e.g., fine) and vadose zone were shown to limit acetate access and reaction. Once accessed by acetate, however, these same zones limited subsequent acetate dilution and provided longer residence times that resulted in higher concentrations of TEAP reaction products when terminal electron donors and acceptors were not limiting. Finally, facies-based porosity and reactive surface area variations were shown to affect aqueous uranium concentration distributions with localized effects of the fine lithofacies having the largest impact on U(VI) surface complexation. The ability to model the comprehensive biogeochemical reaction network, and spatially and temporally variable processes, properties, and conditions controlling uranium behavior during engineered bioremediation in the naturally complex Rifle IFRC subsurface system required a subsurface simulator that could use the large memory and computational performance of a massively parallel computer. In this case, the eSTOMP simulator, operating on 128 processor cores for 12h, was used to simulate the 110-day field experiment and 50 days of post-biostimulation behavior.

Transport and biogeochemical reaction of metals in a physically and chemically heterogeneous aquifer

Biologically mediated reductive dissolution and precipitation of metals and radionuclides play key roles in their subsurface transport. Physical and chemical properties of natural aquifer systems, such as reactive iron-oxide surface area and hydraulic conductivity, are often highly heterogeneous in complex ways that can exert significant control on transport, natural attenuation, and active remediation processes. Typically, however, few data on the detailed distribution of these properties are available for incorporation into predictive models. In this study, we integrate field-scale geophysical, hydrologic, and geochemical data from a well-characterized site with the results of laboratory batch-reaction studies to formulate two-dimensional numerical models of reactive transport in a heterogeneous granular aquifer. The models incorporate several levels of coupling, including effects of ferrous iron sorption onto (and associated reduction of reactive surface area of) ferric iron surfaces, microbial growth and transport dynamics, and cross-correlation between hydraulic conductivity and initial ferric iron surface area. These models are then used to evaluate the impacts of physical and chemical heterogeneity on transport of trace levels of uranium under natural conditions, as well as the effectiveness of uranium reduction and immobilization upon introduction of a soluble electron donor (a potential biostimulation remedial strategy).

The Factors Controlling Microbial Distribution and Activity in the Shallow Subsurface

The relationships among sedimentological, geochemical, and microbiological parameters in the vadose zone of a barrier sediment from the eastern shore of Virginia were examined. Pairs of samples were taken 10 cm apart in the vertical direction and 2 cm apart in the horizontal direction along three transects with one sample from each depth being processed aerobically and the other being processed anaerobically. Little variation was observed in the sedimentological and microbiological parameters tested. The sediment of all samples was fine-to-coarse sand, and the grain sizes ranged from 0.19-1.16 mm. Sediment moisture was low for all samples, but increased near the top and bottom of each sampling transect. These were regions where bioavailable Fe(III) concentrations were high. Rates of H 2 uptake ranged from below detection limit to 0.064 w mol H 2 · day m 1 · g m 1 of sediment with a median rate of 0.01 w mol H 2 · day m 1 · g m 1 of sediment. The variation in bacterial numbers was slightly more than an order of magnitude range over the entire sampling face. Phospholipid fatty acid analysis showed a diverse but fairly uniform microbial community from sample to sample. We found that the quartile of aerobically processed samples with the highest H 2 uptake rates had statistically higher moisture content and bioavailable iron content than did the rest of the samples. The quartile of aerobically processed samples with the lowest H 2 uptake rates had significantly more gravel, less moisture, and less bioavailable Fe(III) than did the rest of the samples. Similar trends were observed for anaerobically processed samples, but the differences were not significant. Our data indicate that the spatial variation in microbial parameters is low within strata with uniform grain sizes.

Vadose Zone Transport Field Study: Summary Report

From FY 2000 through FY 2003, a series of vadose zone transport field experiments were conducted as part of the U.S. Department of Energy’s Groundwater/Vadose Zone Integration Project Science and Technology Project, now known as the Remediation and Closure Science Project, and managed by the Pacific Northwest National Laboratory (PNNL). The series of experiments included two major field campaigns, one at a 299-E24-11 injection test site near PUREX and a second at a clastic dike site off Army Loop Road. The goals of these experiments were to improve our understanding of vadose zone transport processes; to develop data sets to validate and calibrate vadose zone flow and transport models; and to identify advanced monitoring techniques useful for evaluating flow-and-transport mechanisms and delineating contaminant plumes in the vadose zone at the Hanford Site. This report summarizes the key findings from the field studies and demonstrates how data collected from these studies are being used to improve conceptual models and develop numerical models of flow and transport in Hanford’s vadose zone. Results of these tests have led to a better understanding of the vadose zone. Fine-scale geologic heterogeneities, including grain fabric and lamination, were observed to have a strong effect on the large-scale behavior of contaminant plumes, primarily through increased lateral spreading resulting from anisotropy. Conceptual models have been updated to include lateral spreading and numerical models of unsaturated flow and transport have revised accordingly. A new robust model based on the concept of a connectivity tensor was developed to describe saturation-dependent anisotropy in strongly heterogeneous soils and has been incorporated into PNNL’s Subsurface Transport Over Multiple Phases (STOMP) simulator. Application to field-scale transport problems have led to a better understanding plume behavior at a number of sites where lateral spreading may have dominated waste migration (e.g. BC Cribs and Trenches). The improved models have been also coupled with inverse models and newly-developed parameter scaling techniques to allow estimation of field-scale and effective transport parameters for the vadose zone. The development and utility of pedotransfer functions for describing fine-scale hydrogeochemical heterogeneity and for incorporating this heterogeneity into reactive transport models was explored. An approach based on grain-size statistics appears feasible and has been used to describe heterogeneity in hydraulic properties and sorption properties, such as the cation exchange capacity and the specific surface area of Hanford sediments. This work has also led to the development of inverse modeling capabilities for time-dependent, subsurface, reactive transport with transient flow fields using an automated optimization algorithm. In addition, a number of geophysical techniques investigated for their potential to provide detailed information on the subtle changes in lithology and bedding surfaces; plume delineation, leak detection. High-resolution resistivity is now being used for detecting saline plumes at several waste sites at Hanford, including tank farms. Results from the field studies and associated analysis have appeared in more than 46 publications generated over the past 4 years. These publications include test plans and status reports, in addition to numerous technical notes and peer reviewed papers.

Influence of River Level on Temperature and Hydraulic Gradients in Chum and Fall Chinook Salmon Spawning Areas Downstream of Bonneville Dam, Columbia River

Abstract Chum salmon Oncorhynchus keta and fall Chinook salmon O. tshawytscha segregate spatially during spawning in the Ives Island side channel of the lower Columbia River downstream from Bonneville Dam. Previous research during one spawning season (2000) suggested that these species selected spawning habitats based on differences in hyporheic temperature and vertical hydraulic gradient (VHG). In this study we confirmed the spatial segregation of spawning based on hyporheic characteristics over 4 years (2001–2004) and examined the effects of load-following operations (power generation to meet short-term electrical demand) at Bonneville Dam on hyporheic function and characteristics. We found that during the study period hyporheic temperature and VHG in chum salmon spawning areas were highly variable during periods of load-following operation, when river levels fluctuated. In contrast, hyporheic water temperature and VHG within chum salmon spawning areas fluctuated less when river levels were not changing...

Application of ensemble‐based data assimilation techniques for aquifer characterization using tracer data at Hanford 300 area

Subsurface aquifer characterization often involves high parameter dimensionality and requires tremendous computational resources if employing a full Bayesian approach. Ensemble-based data assimilation techniques, including filtering and smoothing, are computationally efficient alternatives. Despite the increasing number of applications of ensemble-based methods in assimilating flow and transport related data for subsurface aquifer charaterization, most are limited to either synthetic studies or two-dimensional problems. In this study, we applied ensemble-based techniques for assimilating field tracer experimental data obtained from the Integrated Field Research Challenge (IFRC) site at the Hanford 300 Area. The forward problem was simulated using the massively-parallel three-dimensional flow and transport code PFLOTRAN to effectively deal with the highly transient flow boundary conditions at the site and to meet the computational demands of ensemble-based methods. This study demonstrates the effectiveness of ensemble-based methods for characterizing a heterogeneous aquifer by sequentially assimilating multiple types of data. The necessity of employing high performance computing is shown to enable increasingly mechanistic non-linear forward simulations to be performed within the data assimilation framework for a complex system with reasonable turnaround time.

Establishing a geochemical heterogeneity model for a contaminated vadose zone — Aquifer system

A large set of sediment samples from a 1600 m² experimental plot within a 2.2 km² vadose zone and groundwater uranium (VI) plume was subject to physical, chemical, and mineralogic characterization. The plot is being used for field experimentation on U(VI) recharge and transport processes within a persistent groundwater plume that exists in the groundwater-river interaction zone of the Columbia River at the U.S. DOE Hanford site. The samples were obtained during the installation of 35 tightly spaced (10 m separation) groundwater monitoring wells. The characterization measurements for each sample included total contaminant concentrations (U and Cu primarily), bicarbonate extractable U(VI), sequential ²³⁸U(VI) contaminant desorption Kd, ²³³U(VI) adsorption K(d), grain size distribution, surface area, extractable poorly crystalline Fe(III) oxides, and mineralogy. The characterization objective was to inform a conceptual model of coupled processes controlling the anomalous longevity of the plume, and to quantify the spatial heterogeneity of the contaminant inventory and the primary properties effecting reactive transport. Correlations were drawn between chemical, physical, and reaction properties, and Gaussian simulation was used to compute multiple 3-D realizations of extractable U(VI), the ²³³U(VI) adsorption K(d), and the distribution of the reactive <2 mm fraction. Adsorbed contaminant U(VI) was highest in the vadose zone and the zone of seasonal water table fluctuation lying at its base. Adsorbed U(VI) was measureable, but low, in the groundwater plume region where very high hydraulic conductivities existed. The distribution of adsorbed U(VI) displayed no apparent correlation with sediment physical or chemical properties. Desorption [²³⁸U(IV)] and adsorption [²³³U(VI)] K(d) values showed appreciable differences due to mass transfer controlled surface complexation and the effects of long subsurface residence times. The ²³³U(VI) adsorption K(d), a combined measure of surface complexation strength and site concentration, was relatively uniform throughout the domain, displaying correlation with fines distribution and surface area. The characterization results revealed U(VI) supplied to the groundwater plume through spatially heterogeneous recharge from residual contamination in the zone of seasonal water table fluctuation, and transport of U(VI) controlled by weak, kinetically-controlled surface complexation in the coarse-textured saturated zone. Geostatistical relationships for the adsorbed contaminant U distribution in the characterization domain allow an extrapolation to inventory at the plume scale, a critical unknown for remedial action.