Subsurface Biogeochemical Research. Click to return to home page.
Department of Energy Office of Science. Click to visit main DOE SC site.

Subsurface Biogeochemical Research Program

(SC GG 5.21.1)
Develop predictive model for contaminant transport that incorporates complex biology, hydrology, and chemistry of the subsurface. Validate model through field

Third Quarter Results

For the 1st Quarter Results (Q1 Milestone) we reported on preliminary model simulations aimed at improving our understanding of ongoing field manipulation experiments at the Subsurface Biogeochemical Research’s (ERSP’s) Field Research Center at Oak Ridge National Laboratory.  In this report (Q3 Milestone), we compare the results of these simulations (reported as a Q1 deliverable) to corresponding Oak Ridge Field Research Center field data.  These field data were collected during in situ manipulation experiments conducted in FRC Areas 1 and 2.


Model simulations were performed as part of the research project “Stability of U(VI) and Tc(VII) Reducing Microbial Communities to Environmental Perturbation: Development and Testing of a Thermodynamic Network Model”, which is being conducted by an interdisciplinary team of researchers from Oregon State University, Pacific Northwest National Laboratory, University of Tennessee, and the University of Oklahoma.  The overall goal of this project is to predict the effect of various chemical amendments on the subsurface microbial community at the FRC and other DOE sites with the goal of promoting the bioimmobilization of contaminant metals and radionuclides.  The model is unique in that it couples the thermodynamics of microbial growth and geochemical reactions resulting in quantitative and system-specific predictions of microbial community composition and function.  Model predictions are being compared with the results of small- to intermediate-scale field experiments conducted at the FRC and other DOE sites.

1.  Overall Objective and Hypothesis

The overall objective of this project is to improve our ability to use native microbial communities to speed remediation of contaminated groundwater and sediments.  The broad scientific hypothesis is that: a thermodynamic analysis of the energy-yielding reactions performed by broadly defined groups of microorganisms can be used to make quantitative and testable predictions of the change in microbial community composition and system geochemistry (including contaminant chemistry) that occur when a substrate is added to the subsurface and/or when environmental conditions change.

2.  Types of Investigations Performed

Our research consists of three main activities:

  1. Model Development.  A list of the major microbial groups present in samples of groundwater and sediment from the FRC has been compiled from published findings by SBR researchers working at the site.  Computer applications have been developed to generate the overall growth reactions and governing thermodynamic quantities for each microbial group.  Necessary model input parameters (electron transfer efficiencies) are being estimated from large numbers of published laboratory experiments.  These reactions, containing the chemical stoichiometry and standard state free energy changes that define the growth of each group (e.g. denitrifiers, iron reducers, etc.), are added to an existing geochemical thermodynamic data base (developed by Lawrence Berkeley National Laboratory).  Collectively, these reactions define the growth and energy flow in an intact microbial community with defined geochemistry.

  2. Numerical Simulations.  Representative geochemical conditions at the site are used in conjunction with the modified thermodynamic data base to predict ‘reaction paths’ that show the coupled changes in microbial community composition and system geochemistry that occur when amendments are added to the subsurface.  Simulations are being performed to investigate the effect of ethanol additions on uranium and technetium bioimmobilization for the major subsurface environments at the FRC (Areas 1 and 3 with low pH and high nitrate groundwater; and Area 2 with neutral pH and low nitrate groundwater). Model output includes predicted changes in groundwater and sediment chemistry and microbial community composition.  Model predictions have been consistent with field observations and provide important insights into the role of specific microbial groups (esp. denitrifiers) on overall system response.  Model predictions are being used to investigate alternate bioimmobilization strategies (choice of substrate, sulfate additions, etc.) and to predict the long-term stability of bioreduced uranium and technetium to changing environmental conditions.     

  3. Experimental Verification.  Ultimately, model predictions must be verified by direct measurement and we have assembled a suite of lipid and nucleic acid-based biomarkers for this purpose.  We are developing a “dictionary” that will allow predicted growth of defined microbial groups to be detected and quantified by one or more distinct biomarkers.  Model predictions for long-term experiments in small-scale microcosms and intermediate-scale physical models and for short-term small-scale field experiments are being compared with biomarker data collected on groundwater and sediment samples.  Initial comparisons are made with total biomass and groups with known functional genes.  Particular emphasis is placed on denitrifiers, sulfate reducers and iron reducers since these make up the largest portion (> 90 % in some cases) of the entire community after substrate has been added. We are actively collaborating with several SBR investigators by applying model simulations to other experimental systems.

3.   Significant Results for Q1

The most recent results of our project during the 1st Quarter were:

  • The predicted change in microbial community composition following ethanol addition was found to vary greatly from one portion of the FRC to the next (e.g., Area 1 vs Area 2).  The most important factors were initial concentrations of nitrate, sulfate, Fe(III)- and Mn(IV)-bearing minerals.
  • Model predictions showed that denitrification, sulfate-, iron-, and manganese-reduction are the major microbial processes in all environments tested.  Denitrification is the major microbial process in high-nitrate (> 10 mM) portions of the FRC including Areas 1, 3, 4, and 5.  Sulfate reduction is the major microbial process in low-nitrate (< 10 mM) portions of the FRC including Area 2.  Organisms that produce acetate and hydrogen were also important because these compounds serve as important electron donors for some microbial groups.
  • Low concentrations of uranium and technetium provide very little energy to microorganisms, and microbial uranium and technetium reduction consume very small amounts of added ethanol.
  • Bioreduced uranium can be readily reoxidized by nitrate and oxygen once electron donor additions cease; bioreduced technetium is more resistant to reoxidation.

3.   Main Results for Q3

  • Detailed simulations were performed for two main flow regimes: “batch” and “flush”.  Batch simulations were used to describe the results of field “push-pull” tests conducted at the FRC by the Istok et al. field project.   Flush simulations were used to describe results of natural- and forced-gradient (i.e. well-to-well) field tests conducted by the Schiebe et al. and Criddle et al. field projects.  Both types of simulations are also being used in an ongoing effort to interpret laboratory microcosm (batch) and column experiments (flush) using FRC sediments and groundwater.  These experiments are being conducted by our project and in collaboration with other FRC investigators.
  • Batch model predictions are in qualitative agreement with most observed features of field push-pull tests in Area 2.  In these low nitrate environments, more energetically favorable electron acceptors (esp. O2 and nitrate) are readily consumed resulting in conditions that favor the precipitation of reduced uranium as uraninite, and technetium as Tc2S7, under both sulfate- and Fe(III)-reducing conditions. 
  • Flush model predictions were able to match measured increases in microbial biomass in two intermediate-scale physical models operated for 18 months at the FRC.  Additional microbial community analyses are in progress to compare predicted and observed growth of denitrifiers, sulfate reducers, and Fe(III) reducers.  Flush model simulations accurately predicted observed precipitation of sulfide minerals and uraninite.  Predicted amounts of precipitated Tc2S7 could not be verified because measurement techniques were unable to detect this mineral at such low concentrations.
  • Batch model predictions were in agreement with observations made during push-pull tests in Area 1, which has very high nitrate concentrations (> 100 mM).  In these environments, high nitrate levels prevented uraninite precipitation but not Tc2S7 precipitation.  The predicted growth of denitrifying organisms as the major microbial group after ethanol addition was verified by Q-PCR measurements.
  • To support continued use of the model to predict and interpret field experiments, a real-time PCR protocol to quantify adenosine-5’-phosphosulfate (APS) reductase was developed to supplement enumeration of dissimilatory sulfite reductase (DSR) as an indicator of sulfate reducing bacteria.  In addition, real-time PCR primers were designed and tested to quantify formyl-tetrahydrofolate synthase as an indicator of acetogens.
  • In collaboration with other NABIR researchers, a prototype 16S rRNA gel element microarray was developed.  The current version of the microarray includes probes for more than 20 genera of known nitrate-, iron-, and sulfate-reducing bacteria and is readily expandable to include additional sequences of interest.

4.  Planned Activities:

The activities planned for the coming year include:

  • We will continue to refine the numerical model by including additional groups of microorganisms and additional microbial processes in order to better align simulations with observations from FRC field experiments.  In particular we wish to include groups that are involved in the nitrate-dependent oxidation of Fe(II) and U(IV).
  • Last year we designed a series of laboratory microcosm experiments to explicitly test model predictions for a range of defined growth conditions.  Microcosm experiments targeted nitrate, iron, sulfate, manganese and methanogenic processes.  In these experiments, we monitored changes in geochemistry and collected sediment samples for microbial community analyses.  The initial experiments have been completed and microbial community analysis is underway.  Real-time PCR quantification will be performed to enumerate denitrifying, iron reducing, sulfate reducing, and methanogenic members of the microbial community in each microcosm described above.  DGGE with subsequent sequencing of excised bands will be performed to determine dominant members of the bacterial community.  PLFA/Quinone profiles will also be performed to quantify viable biomass and bacterial functional groups.
  • We will continue to collaborate with other SBR investigators to design probes targeting terminal electron accepting processes, nutrient limitation, and stress responses for use in an mRNA based microarray. 
  • We have initiated a series of collaborations with other SBR investigators to test the model’s utility in interpreting laboratory and field data, and for transferring results from one experimental system to another.  To this end we are attempting to simulate the results of the Schiebe et al. field experiment at the FRC and the results of the Long et al. field experiments at the Old Rifle site.
  • We will conduct a series of sensitivity analyses to examine the effects on community composition of various environmental perturbations including initial nitrate concentration, available iron, lower pH, and sulfate concentration.  Preliminary results showed initial growth of nitrate reducers, followed by growth of Fe reducers as donor concentration increased.  When the Fe source is exhausted, methanogenesis begins to dominate the community, and uranium begins to be geochemically reduced to uraninite.  It was interesting to note that the microbial community rapidly adjusted to a lower initial pH after ~ 1 mM of ethanol was reacted.

5.   Project Significance

This project is showing that the ability to predict the effects of donor addition on changes in microbial community composition is essential for creating conditions that favor the long-term stability of bioreduced uranium and technetium.  Moreover, the ability of a microbial community to maintain functional stability (i.e. maintain high rates of uranium and technetium reduction) when subjected to various environmental perturbations is of critical importance for the ultimate use of bioimmobilization at DOE legacy waste sites.  The longer-term significance of this project will be to provide a comprehensive theoretical framework for designing and interpreting complex field experiments, and to aid in “bridging-the-gap” between basic research and field applications.

Submit Research Highlights

Now Featuring

Subsurface Biogeochemical Research Overview

July 2017 [PDF]


Subsurface Biogeochemical-Related BER Research Highlights