Elsevier

Environmental Pollution

Volume 159, Issue 10, October 2011, Pages 2831-2839
Environmental Pollution

Using advanced surface complexation models for modelling soil chemistry under forests: Solling forest, Germany

https://doi.org/10.1016/j.envpol.2011.05.002Get rights and content

Abstract

Various dynamic soil chemistry models have been developed to gain insight into impacts of atmospheric deposition of sulphur, nitrogen and other elements on soil and soil solution chemistry. Sorption parameters for anions and cations are generally calibrated for each site, which hampers extrapolation in space and time. On the other hand, recently developed surface complexation models (SCMs) have been successful in predicting ion sorption for static systems using generic parameter sets. This study reports the inclusion of an assemblage of these SCMs in the dynamic soil chemistry model SMARTml and applies this model to a spruce forest site in Solling Germany. Parameters for SCMs were taken from generic datasets and not calibrated. Nevertheless, modelling results for major elements matched observations well. Further, trace metals were included in the model, also using the existing framework of SCMs. The model predicted sorption for most trace elements well.

Highlights

► Surface complexation models can be well applied in field studies. ► Soil chemistry under a forest site is adequately modelled using generic parameters. ► The model is easily extended with extra elements within the existing framework. ► Surface complexation models can show the linkages between major soil chemistry and trace element behaviour.

Introduction

Soil and soil solution chemistry under (semi-)natural ecosystems, such as forests, is determined by atmospheric inputs of water and elements, element cycling by vegetation, biochemical and geochemical processes and chemical equilibria. To gain insight into the impact of acid atmospheric deposition on soil and soil solution chemistry, various field studies were set up in the 1970s and 1980s (Abrahamsen et al., 1976, Driscoll et al., 1985, Van Breemen et al., 1986), whereas national (e.g. De Vries et al., 1995) and European wide (e.g. De Vries et al., 2003) forest monitoring programmes started in the 1990s. One monitoring site, at Solling Germany, started even as early as 1966 and monitoring has continued up to now, leading to a unique monitoring dataset of more than 40 years (Meesenburg et al., 1995). At the same time, various soil acidification models have been developed since the mid 1980s to understand, describe and predict effects of sulphur and nitrogen deposition on the soil solution chemistry under nature areas (e.g. Cosby et al., 1985, De Vries et al., 1989, Groenenberg et al., 1995, Warfvinge et al., 1993). In most of these models sorption equilibria are modelled using Gapon or Gaines–Thomas ion exchange equations for cations and Langmuir or Freundlich isotherms for sulphate and phosphate (Tiktak and Van Grinsven, 1995). Equilibrium parameters, like exchange constants and sorption constants, are then generally calibrated for specific sites. This approach is nicely illustrated in a special issue of Ecological Modelling (Van Grinsven, 1995) in which a total of 16 forest–soil–atmosphere models were applied to a dataset of the Solling Norway spruce site (Tiktak et al., 1995).

Exchange equations like Gaines–Thomas use a single exchange constant for each element to describe sorption to all sorption surfaces, like organic matter, clay minerals, iron and aluminium (hydr)oxides. This exchange constant will be some average of the exchange constants of the individual sorption surfaces. This means that exchange constants, and also other sorption constants, differ for different soil types, because the relative abundances of the reactive surfaces differs. Even more, this average exchange constant depends on the saturation of the total exchange complex and thus on the concentrations of elements in the soil solution. This is because elements tend to bind first to the surface for which they have the highest affinity. Once this surface becomes saturated they will bind to a different surface with a different exchange constant. Altogether, this hampers the extrapolation and regional scale application of such models. An alternative is so-called surface complexation models (SCMs), which are process-based models with a detailed description of complexation by a specific sorption surface. These SCMs have been developed for most import sorption surfaces in soils (e.g. Dzombak and Morel, 1990, Kinniburgh et al., 1999). Also generic parameter values have been determined for these SCMs (e.g. Milne et al., 2003). A combination of several SCMs, each for a specific surface, into a so-called assemblage model makes it possible to model the chemical equilibria of a complete soil system. So far, these SCMs have mainly been tested on batch extraction experiments and controlled laboratory column experiments. Lofts and Tipping (1998) and Weng et al. (2001) were among the first to apply such an assemblage model to calculate trace element concentrations in soil extracts. Bonten et al. (2008) and Dijkstra et al., 2004, Dijkstra et al., 2009 showed that these models can be used in batch experiments for a wide range of soil types and soil pH values with a single generic parameter set. Fest et al. (2005) and Almas et al. (2006) used these types of models to predict changes of both pH, concentrations of major cations and concentrations of trace metals in laboratory experiments with acid and/or with base titrated soils. These results indicate that SCMs could be used in field scale dynamic modelling without any calibration of chemical equilibrium parameters. So far, the only field scale applications of SCMs for soil system dynamics have been on a few UK moorland catchments (Tipping et al., 2006, Tipping et al., 2010). However, in these studies important model inputs such as historical deposition rates were estimated and weathering rates and S and N dynamics were calibrated to match observations and simulations for major solutes.

The aim of this study is to investigate whether SCMs are able to reproduce the field behaviour of major ions and trace elements without any calibration of chemical equilibrium parameters using a site with measured information on historical rainfall and deposition rates. To do so, we present an assemblage of SCMs calculating chemical equilibria within the soil acidification model SMARTml, and evaluate it on the intensively monitored Solling spruce site, using data for the period 1968–2004. The extensive and continuous monitoring at the Solling site makes such a model comparison possible, without estimating historic model inputs like atmospheric deposition and meteorology, and without calibrating processes like weathering, uptake and organic matter dynamics. The model SMARTml was developed as a multi-layer version of the model soil acidification model SMART (De Vries et al., 1989) and extended with trace elements, including the linkage between soil acidification and trace element behaviour. The latter extension makes the model quite unique and allowed us to validate both the solution chemistry of major ions and of trace elements on measurements at the Solling spruce site.

Section snippets

Site description and measurements

The Solling site is located at the centre of the Solling plateau at 508 m above sea level about 33 km northwest of Göttingen, Germany. Its geographical position is 9°30′E, 51°40′N. We used data of the Solling F1 monitoring site, a Norway spruce (Picea abies Karst.) plantation forest which has been planted in 1888 with 4-year-old spruce saplings. The Solling massive consists of Triassic sandstone, locally known as ‘‘Solling-Folge’’, which is covered with 60–80 cm thick solifluction layers of

Results and discussion

The goal of this study is to demonstrate the applicability of SCMs for field applications, more especially in soil acidification modelling. Therefore, the result section focuses on parameters that are affected by the use of SCMs, i.e. pH and concentrations and contents of aluminium, base cations, sulphate and heavy metals. Results for other parameters, like Cl and nitrogen, are only discussed shortly. These will be shown because they can affect results for other parameters. We show the results

Concluding remarks

In this study we evaluated the use of surface complexation models (SCMs) in predicting dynamics in soil chemistry without any site-specific calibration. For this an assemblage of SCMs were incorporated into the dynamic soil system model SMARTml. Overall, model results matched observations well. Simulations deviated from observations only for soil layers or parameters for which insufficient information was available. These positive results demonstrate the potential to further apply SCMs in

Acknowledgements

This research was funded by the Dutch Ministry of Housing, Spatial Planning and the Environment and by the Dutch National Institute for Public Health and the Environment.

References (46)

  • G.J. Reinds et al.

    Modelling recovery from soil acidification in European forests under climate change

    Science of the Total Environment

    (2009)
  • A. Tiktak et al.

    Review of sixteen forest–soil–atmosphere models

    Ecological Modelling

    (1995)
  • A. Tiktak et al.

    The Solling dataset. Site characteristics, monitoring data and deposition scenarios

    Ecological Modelling

    (1995)
  • E. Tipping et al.

    Simulating the long-term chemistry of an upland UK catchment: major solutes and acidification

    Environmental Pollution

    (2006)
  • E. Tipping et al.

    Dynamic modelling of atmospherically-deposited Ni, Cu, Zn, Cd and Pb in Pennine catchments (northern England)

    Environmental Pollution

    (2010)
  • P. Warfvinge et al.

    Modelling long-term cation supply in acidified forest stands

    Environmental Pollution

    (1993)
  • J.D. Allison et al.

    MINTEQA2/PRODEFA2, a Geochemical Assessment Model for Environmental Systems: Version 3.0 User’s Manual. EPA/600/3-91/021

    (1991)
  • L.T.C. Bonten et al.

    Dynamic modelling of effects of deposition on carbon sequestration and nitrogen availability: VSD plus C and N dynamics (VSD+)

  • B.J. Cosby et al.

    Modelling the effects of acid deposition: assesment of a lumped parameter model of soil water and streamwater chemistry

    Water Resources Research

    (1985)
  • W. De Vries et al.

    Simulation of the long-term soil response to acid deposition in various buffer ranges

    Water, Air, and Soil Pollution

    (1989)
  • J.J. Dijkstra et al.

    Leaching of heavy metals from contaminated soils: an experimental and modeling study

    Environmental Science and Technology

    (2004)
  • J.J. Dijkstra et al.

    Evaluation of a generic multisurface sorption model for inorganic soil contaminants

    Environmental Science and Technology

    (2009)
  • Cited by (0)

    View full text