Elsevier

Applied Geochemistry

Volume 22, Issue 12, December 2007, Pages 2606-2617
Applied Geochemistry

Gradients controlling natural attenuation of ammonium

https://doi.org/10.1016/j.apgeochem.2007.06.009Get rights and content

Abstract

Oxidation of reduced pollutants such as NH4+ in groundwater often takes place at steep redox gradients where oxygenated water is being mixed into polluted water such as landfill leachate. In order to identify controlling parameters and quantify the influence of environmental factors for NH4+ degradation, sensitivity analysis was performed by means of scenario specific numerical modelling. Geometrical factors such as aquifer thickness have been shown to be very influential on the capability of natural attenuation of pollutants in groundwater. The scenarios investigated here include biodegradation at redox gradients in groundwater, so called fringe processes, for (i) a partly contaminated aquifer with two reaction fronts, (ii) and a spatially variable aquifer thickness. In addition, (iii) the influence of groundwater recharge and (iv) restricted supply of O2 to contaminated water by slow dispersion and diffusion across the capillary fringe are investigated. Contaminated aquifer thickness, zones of enhanced mixing due to flow focussing and diffusion/dispersion coefficients in the capillary fringe are identified qualitatively as controlling factors for natural attenuation under complex conditions, whereas predictive functions will require further research.

Introduction

Ammonium is a common pollutant at landfills and former gas work sites. This is due to the generally large amounts of N present at these sites (e.g., proteins) that are transformed to NH4+ under reducing conditions and thus represent a strong potential of forming large scale groundwater pollution (contaminant plumes). Ammonium is fish toxic and may cause problems in drinking water wells. Several processes lead to attenuation of NH4+, such as sorption, dilution and biodegradation. Dilution, however may cause contaminant concentrations to drop below the legal limit, but does not lead to a decrease of contaminant fluxes in environmental compartments.

Sorption of NH4+ is mainly controlled by ion exchange and leads to retardation but not to the destruction of contaminants. Retardation may increase the available time to perform remedial measures at a receptor of concern, and should be taken into account to decide whether an NH4+ plume has reached its maximum extent. Biodegradation thus remains as the only process that is able to efficiently transfer NH4+ to less hazardous (such as NO3-) or harmless compounds (e.g., N2). Whether biodegradation takes place, is highly dependent on redox conditions.

Studies on NH4+ attenuation processes at the field scale so far have mainly focussed on characterizing reaction processes and corresponding redox conditions (Buss et al., 2003, Bjerg et al., 1995, Erskine, 2000, Christensen et al., 2001). Rügner et al. (2004) determined NH4+ degradation rates in the field based on measurements of NH4+ mass flow rates at two consecutive control planes and conceptual (scenario specific) numerical modelling. Liedl et al., 2005, Maier and Grathwohl, 2006 have described analytical and numerical approaches to actually predict the steady state length for NH4+ plumes or other aerobically degradable contaminants for simple scenarios. However, the assessment of NH4+ degradation and the length of NH4+ plumes under more complex boundary conditions is still challenging.

Biotransformation may take place as core controlled processes within the interior of the plume or as fringe controlled processes driven by external electron acceptors. Oxygen as a latter one is usually not present in landfill leachate. Core processes which allow for degradation without external mixing of electron acceptors are strongly dependent on the available electron acceptors, degradation kinetics and stoichiometry. Anaerobic NH4+ degradation under reducing conditions using NO3-, Mn or Fe oxides as intrinsic electron acceptors (anammox) may contribute to NH4+ degradation (Christensen et al., 2001, Jetten, 2001, Schink, 2002, Buss et al., 2003). However, field data indicate that these mechanisms, which are not considered in this study, are rather slow and may thus only contribute significantly to natural attenuation of NH4+ in very long plumes i.e. of more than several hundred meters to 1 km length, which was complied as tolerable at the Osterhofen case study (Rügner et al., 2004). Thus, core processes are unlikely to contribute to natural attenuation of NH4+.

The most important degradation process is aerobic nitrification of NH4+ to NO3- in the presence of O2 (Stumm and Morgan, 1996). As two different bacterial genuses are involved, the reaction proceeds in two steps. The intermediate product NO2- is instable, so that nitrification is commonly combined to the overall reaction:NH4++1.5O2NO2-+H2O+2H+NO2-+0.5O2NO3-NH4++2O2NO3-+H2O+2H+The kinetics of biodegradation (i.e. chemical reaction of two components A and B) are commonly expressed according to Michaelis and Menten (1931):R=-kmax·CACA+K1/2A·CBCB+K1/2B·CBCB+KthrB2·CACA+KthrA2In this case A represents NH4+ and B O2. R [M L−3 T−1] denotes the reaction rate at a certain location, kmax [M L−3 T−1] is the Monod maximum utilization rate of the reaction, K1/2 [M L−3] is the Monod half utilization concentration of the reaction for each compound and Kthr [M L−3] is the threshold concentration of the specific compound that is just sufficient to maintain the reaction.

In many cases, processes at plume fringes and along gradients control natural attenuation. As these processes require the transport of external electron acceptors to the contaminant plume, the geometry of the aquifer and the contamination as well as the properties of the porous medium controlling mixing processes are very influential on the total amount of mass transformation (Liedl et al., 2005, Maier and Grathwohl, 2006). Werth et al., 2006, Cirpka and Kitanidis, 2000 identified zones where groundwater flow converges as major contribution to dispersive mass fluxes. In addition, Koussis et al., 2003, Chu et al., 2005, Maier and Grathwohl, 2006 justified that reaction kinetics are not limiting aerobic biodegradation in most cases under field conditions. When mixing of contaminant and O2 is the limiting process, steady state plume length is independent of groundwater flow velocity (Liedl et al., 2005, Maier and Grathwohl, 2006), whereas geometry and mixing characteristics were shown to be the crucial parameters as they play a decisive role in delivering the required electron acceptors to sustain biodegradation to the fringe of contaminant plumes. For a scenario with a homogeneous aquifer uniformly contaminated at the influx boundary and O2 supply from the water table, the steady state plume length L was predicted by Liedl et al., 2005, Maier and Grathwohl, 2006:L=M2αtf(γ)where M is the aquifer thickness, αt the transverse vertical dispersivity and f(γ) a function of the stoichiometric ratio of contaminant to O2 demand, which generally appears as a weak non-linear relationship to plume length (logarithm or power  0.3). Furthermore, slow aqueous phase molecular diffusion or transverse dispersion may limit mass transfer across the capillary fringe (McCarthy and Johnson, 1993, Jellali et al., 2003), which is important for O2 supply to aquifers.

The aim of this study is to identify the main factors that control natural attenuation of NH4+ for more complex boundary conditions which are frequently observed at the field scale. In particular, these are (i) partly contaminated aquifers with two reaction fronts and (ii) a spatially variable aquifer thickness, (iii) the influence of groundwater recharge and (iv) the influence of restricted supply of O2 to contaminated water by slow dispersion and diffusion processes across the capillary fringe. For this purpose, scenario specific numerical modelling was performed to assess the potential of natural attenuation at the aquifer scale by variation of environmental factors (Grathwohl et al., 2003). In contrast to narrow contaminant plumes caused by DNAPL infiltration it is frequently observed at landfills that the width of an NH4+ plume usually corresponds to a major proportion to the width of the landfill itself (e.g. 110 m of 180 m at the “Osterhofen” field site; Rügner et al., 2004). Thus, an emphasis is given on gradients that control the vertical transport of reactants. Thus, numerical simulations were performed in 2-D vertical cross sections. This also reduces computation time.

Section snippets

Methodology – scenario specific modeling

The numerical code MIN3P (Mayer et al., 2002) was used, that allows for the simulation of a variable number of geochemical compounds and reactions. Biogeochemical reactions and transport processes are coupled by a global implicit solution method and solved using a finite volume algorithm. Physical–chemical properties of the geochemical species are defined in the external database of the geochemical equilibrium model MINTEQ (Allison et al., 1991). Advective-dispersive transport in the aqueous

Partly contaminated aquifer

Landfills may be entirely placed within the unsaturated zone, emitting contaminated seepage water into the aquifer, or comprise the unsaturated zone and the upper part of an aquifer. If contaminated water is underlain by oxygenated groundwater, a second reaction front will develop contributing to natural attenuation. Simulated concentration contours for a steady state plume of that scenario are given in Fig. 1 for NH4+ and O2 and the reaction product NO3-. The NH4 plume is confined by two

Conclusions

The spatial zones and their relative importance for NH4+ attenuation were evaluated with a focus on geometric factors affecting flow and transport in and into aquifers, depending on plume and aquifer geometry, groundwater recharge, and limited O2 supply.

A strong influence on the steady state plume length affected by biodegradation has to be expected from parameters such as the aquifer or plume thickness and geometry. Aquifer thickness M was reported by Liedl et al., 2005, Maier and Grathwohl,

Acknowledgement

The study was funded by the LfU (Environmental Agency) Baden-Württemberg which is gratefully acknowledged.

References (29)

  • M. Chu et al.

    Modeling microbial reactions at the plume fringe subject to transverse mixing in porous media: when can the rates of microbial reaction be assumed to be instantaneous?

    Water Resour. Res.

    (2005)
  • O.A. Cirpka et al.

    Characterization of mixing and dilution in heterogeneous aquifers by means of local temporal moments

    Water Resour. Res.

    (2000)
  • M. Dentz et al.

    Temporal behavior of a solute cloud in a heterogeneous porous medium 1. Point-like injection

    Water Resour. Res.

    (2000)
  • A.D. Erskine

    Transport of ammonium in aquifers: retardation and degradation

    Quart. J. Eng. Geol. Hydrogeol

    (2000)
  • Cited by (17)

    • A high-resolution non-invasive approach to quantify oxygen transport across the capillary fringe and within the underlying groundwater

      2011, Journal of Contaminant Hydrology
      Citation Excerpt :

      In the unsaturated zone, the effective diffusion coefficient additionally takes the volumetric water and gas content into account. With increasing water saturation from the unsaturated zone toward the groundwater, the effective gaseous diffusion coefficient decreases (Affek et al., 1998; Maier et al., 2007). Simultaneously, the horizontal flow velocity increases and hydrodynamic dispersion in the aqueous phase becomes important.

    View all citing articles on Scopus
    View full text