Biogeochemical processes at the fringe of a landfill leachate pollution plume: potential for dissolved organic carbon, Fe(II), Mn(II), NH4, and CH4 oxidation

https://doi.org/10.1016/j.jconhyd.2004.01.001Get rights and content

Abstract

Various redox reactions may occur at the fringe of a landfill leachate plume, involving oxidation of dissolved organic carbon (DOC), CH4, Fe(II), Mn(II), and NH4 from leachate and reduction of O2, NO3 and SO4 from pristine groundwater. Knowledge on the relevance of these processes is essential for the simulation and evaluation of natural attenuation (NA) of pollution plumes. The occurrence of such biogeochemical processes was investigated at the top fringe of a landfill leachate plume (Banisveld, the Netherlands). Hydrochemical depth profiles of the top fringe were captured via installation of a series of multi-level samplers at 18, 39 and 58 m downstream from the landfill. Ten-centimeter vertical resolution was necessary to study NA within a fringe as thin as 0.5 m. Bromide appeared an equally well-conservative tracer as chloride to calculate dilution of landfill leachate, and its ratio to chloride was high compared to other possible sources of salt in groundwater. The plume fringe rose steadily from a depth of around 5 m towards the surface with a few meters in the period 1998–2003. The plume uplift may be caused by enhanced exfiltration to a brook downstream from the landfill, due to increased precipitation over this period and an artificial lowering of the water level of the brook. This rise invoked cation exchange including proton buffering, and triggered degassing of methane. The hydrochemical depth profile was simulated in a 1D vertical reactive transport model using PHREEQC-2. Optimization using the nonlinear optimization program PEST brought forward that solid organic carbon and not clay minerals controlled retardation of cations. Cation exchange resulted in spatial separation of Fe(II), Mn(II) and NH4 fronts from the fringe, and thereby prevented possible oxidation of these secondary redox species. Degradation of DOC may happen in the fringe zone. Re-dissolution of methane escaped from the plume and subsequent oxidation is an explanation for absence of previously present nitrate and anaerobic conditions in pristine groundwater above the plume. Stable carbon isotope (δ13C) values of methane confirm anaerobic methane oxidation immediately below the fringe zone, presumably coupled to reduction of sulfate, desorbed from iron oxide. Methane must be the principle reductant consuming soluble electron-acceptors in pristine groundwater, thereby limiting NA for other solutes including organic micro-pollutants at the fringe of this landfill leachate plume.

Introduction

Processes occurring at the fringe of a pollution plume are potentially of considerable importance for natural attenuation (NA) of the entire plume. Diffusive and dispersive mixing of anoxic-polluted groundwater with (sub)oxic pristine groundwater at the plume fringe results in dilution and may enhance degradation: the degradation potential for various electron-donors like aromatic hydrocarbons is absent or limited inside plumes, but excellent in the plume fringe area, due to the presence of soluble electron-acceptors like oxygen, nitrate and sulfate. Transversal mixing of oxidized and reduced groundwaters therefore enables faster biodegradation for many compounds. For example, oxidation by nitrate reduction in a 2-m-thick fringe took the main share of the total degradation in a 20-m-thick plume consisting of phenolic compounds (Mayer et al., 2001).

High-resolution groundwater analyses of plume fringes are however limitedly present and this hampers progress in understanding NA processes at the fringe of pollution plumes. The hydrochemical gradient perpendicular to a plume fringe is often badly captured in pollution plume studies, because the vertical sampling point spacing is too large with respect to the fringe thickness. Consequently, reactive transport models cannot be validated for processes operational at the plume fringe. Conventional numerical models based on rectangular grids often overpredict mixing, resulting in associated overestimation of degradation of contaminants in the plume, and consequently, underestimation of the actual plume expansion happens (Cirpka et al., 1999). However, occurrence of transient flow, which is usually not simulated, enhances dispersive mixing and hence biodegradation (Schirmer et al., 2001).

Although the understanding of mixing processes is improving, knowledge on the biogeochemistry in the plume fringe remains behind. Hunter et al. (1998) made a distinction between primary redox reactions, where organic matter is degraded, and secondary redox reactions (SRR), which involve the oxidation of reduced redox species formed by the primary redox reactions. Besides dissolved organic carbon (DOC) and aromatic hydrocarbons, especially in landfill leachate plumes several reduced redox species (Fe(II), Mn(II), NH4, and CH4) can be subject to oxidation at the plume fringe, and compete for available oxidants. Secondary redox reactions may be beneficial when the electron-donor is considered as a pollutant (ammonium), but can also be regarded as unwanted (methane) because the availability of soluble oxidants then diminishes for more harmful pollutants. The relevance of the various SRRs is under debate Griffioen, 1999, Hunter and Van Cappellen, 2000, and the occurrence and extent of SRRs and related geochemical processes at the plume fringe deserve more attention (Christensen et al., 2000). This knowledge needs to be developed in order to validate model simulations on degradation and dilution at the plume fringe.

Landfill leachate contains high ammonium concentrations, where nitrification of ammonium under aerobic conditions is a possible attenuation process. Recently, occurrence of anaerobic ammonium oxidation (anammox) by reduction of nitrite was proved (Jetten et al., 1998), and the occurrence of anammox coupled to nitrate reduction providing nitrite for ammonium oxidation, has been observed as well (Thamdrup and Dalsgaard, 2002). Anammox may be an important process in nature at oxic/anoxic interfaces including the fringes of landfill leachate plumes (Schmidt et al., 2002).

Ferrous iron is another competitor for both nitrate and oxygen. Anaerobic nitrate-dependent Fe(II) oxidation is a microbiological process and needs the presence of an organic cosubstrate such as acetate (Straub et al., 2001). Aerobic oxidation of Fe(II) and Mn(II) occur both microbiologically mediated and chemically (Stumm and Morgan, 1996). Re-oxidation of ferrous iron that is mobilized upon reductive dissolution of iron oxide in association with organic matter oxidation, can be beneficial, as for example benzene seems more often degraded with Fe(III) than with NO3 as electron-acceptor (Lovley, 2000).

Finally, methane in leachate is prone to oxidation at the plume fringe, at least with oxygen (Hanson and Hanson, 1996), but possibly with other electron-acceptors, too. Anaerobic methane oxidation (AMO) has been observed with sulfate as electron-acceptor, but the process is still poorly understood (Valentine and Reeburgh, 2000). Hoehler et al. (1994) and Schink (1997) proposed the mechanism of methanogens conducting reverse methanogenesis, in association with sulfate-reducers oxidizing and maintaining low levels of hydrogen in order to make the reaction thermodynamically feasible. In theory, hydrogen-oxidizing microorganisms could also use electron-acceptors other than sulfate (e.g., NO3, Fe(III), Mn(IV)) to keep hydrogen levels sufficiently low for reverse methanogenesis to be favorable (Hoehler et al., 1994). However, the concept of hydrogen as the electron-shuttle during AMO is questioned. Hydrogen in combination with acetic acid (Valentine and Reeburgh, 2000), formate (Sorensen et al., 2001), and transfer of an electron carrier rather than a methane-derived carbon compound (Nauhaus et al., 2002), were proposed as alternatives. Anaerobic methane oxidation coupled to nitrate-reduction has been suggested Bjerg et al., 1995, Van Breukelen et al., 2003, and observed to occur at field conditions (Smith et al., 1991) and in a reactor (Costa et al., 2000).

The present study was initiated to study the biogeochemistry of a pollution plume fringe (Banisveld landfill, the Netherlands), and in particular the relevance of several possible SRRs. For this purpose, the (isotope) hydrochemistry was determined across the fringe at three distances downstream from the landfill, via sampling multi-level samplers (MLS) where a vertical-sampling interval of only 10 cm was obtained. Reactive transport modelling was performed using PHREEQC-2 (Parkhurst and Appelo, 1999) in order to deduce the governing biogeochemical processes. A series of additional simulations were performed to evaluate the relative importance of the various possible SRRs at the fringe of a landfill leachate plume more generally.

Section snippets

Field site description: hydrogeology and biogeochemistry

The biogeochemistry and microbial ecology of the Banisveld landfill leachate plume were previously studied Röling et al., 2000, Röling et al., 2001, Van Breukelen et al., 2003, Van Breukelen et al., in press. Here, a summary will be given. The Banisveld landfill (6 ha) is situated 5 km southwest of Boxtel, the Netherlands. Disposal of primarily household refuse (400,000 m3) occurred in a former 6-m-deep sand pit between 1965 and 1977. Artificial or natural liners are absent, while most of the

Design of multi-level sampler

A permanent multi-level sampler (MLS) was designed for the present study in order to obtain cost-effectively small groundwater samples of flexible volume at small vertical spacing, with a minimum of flushing, and minimal disturbance of the biogeochemistry. Fig. 2 shows the MLS constructed for the present study. Teflon sampling tubing [inner diameter (ID) is 2 mm, outer diameter (OD) is 4 mm] and porous borosilicate glass filters (pore size is 100–160 μm, screen length is 2 cm, OD is 8 mm,

Mixing between leachate and pristine groundwater

Three positions in the hydrochemical depth profiles (see Fig. 5, Fig. 6) are of importance for later discussion and defined here: the pristine-end of the fringe (F1, Br or Cl concentration is 100% pristine groundwater), the leachate-end of the fringe (F2, Br or Cl concentration is 100% leachate), and the divalent-cation front (F3, concentrations of divalent cations are 100% leachate). Note that Ca, Mg, Fe(II) and Mn(II) show a concentration decrease in upward direction from F3. The height of

Model set-up and calibration

A reactive transport model was constructed in PHREEQC-2 (Parkhurst and Appelo, 1999) to simulate the hydrochemistry of the plume fringe at location M1, in order to verify quantitatively if proton buffering and degassing triggered by a rising plume determine the hydrochemical patterns across the fringe. Redox processes (oxidation of DOC and methane with nitrate or sulfate) and redissolution of escaped methane were not simulated. The model consisted of 100 cells each having a length of 0.05 m.

Discussion and conclusion

Summarizing, the plume rose gradually to the surface in the period 1998–2003. The plume uplift might be caused by enhanced exfiltration to the “Heiloop” stream due to increased precipitation over this period and an artificial lowering of the water level in the stream. Rising of the plume triggered proton buffering including cation exchange, and induced degassing of methane. Cation exchange resulted in spatial separation of NH4 and in particular Fe(II) and Mn(II) from potential

Acknowledgements

Niek van Harlingen and Michel Groen are acknowledged for their contribution to the design of the multi-level sampler. The two anonymous reviewers are thanked for their critical review, which significantly improved the contents of this article.

References (49)

  • I.D. Klenk et al.

    Transverse vertical dispersion in groundwater at the capillary fringe

    J. Contam. Hydrol.

    (2002)
  • K.U. Mayer et al.

    Reactive transport modeling of processes controlling the distribution and natural attenuation of phenolic compounds in a deep sandstone aquifer

    J. Contam. Hydrol.

    (2001)
  • K. Revesz et al.

    Methane production and consumption monitored by stable H and C isotope ratios at a crude oil spill site, Bemidji, Minnesota

    Appl. Geochem.

    (1995)
  • I. Schmidt et al.

    Aerobic and anaerobic ammonia oxidizing bacteria competitors or natural partners?

    FEMS Microbiol. Ecol.

    (2002)
  • K.L. Straub et al.

    Iron metabolism in anoxic environments at near neutral pH

    FEMS Microbiol. Ecol.

    (2001)
  • B.M. Van Breukelen et al.

    Biogeochemistry and isotope geochemistry of a landfill leachate plume

    J. Contam. Hydrol.

    (2003)
  • P. Bergamaschi

    Seasonal variations of stable hydrogen and carbon isotope ratios in methane from a Chinese rice paddy

    J. Geophys. Res., [Atmos.]

    (1997)
  • P.L. Bjerg et al.

    Distribution of redox-sensitive groundwater quality parameters downgradient of a landfill (Grindsted Denmark)

    Environ. Sci. Technol.

    (1995)
  • M.G.M. Bruggenwert et al.

    Survey of experimental information on cation exchange in soil systems

  • J.B. Chung et al.

    Ammonium–potassium and ammonium–calcium exchange equilibria in bulk and rhizosphere soil

    Soil Sci. Soc. Am. J.

    (1994)
  • C. Costa et al.

    Denitrification with methane as electron donor in oxygen-limited bioreactors

    Appl. Microbiol. Biotechnol.

    (2000)
  • I.M. Cozzarelli et al.

    Geochemical and microbiological methods for evaluating anaerobic processes in an aquifer contaminated by landfill leachate

    Environ. Sci. Technol.

    (2000)
  • S.N. Davis et al.

    Uses of chloride/bromide ratios in studies of potable water

    Groundwater

    (1998)
  • D.A. Dzombak et al.

    Surface Complexation Modeling: Hydrous Ferric Oxide

    (1990)
  • Cited by (0)

    View full text