Aquatic Cycles at the Earth’s Surface
Calcite dissolution in two deep eutrophic lakes

https://doi.org/10.1016/S0016-7037(99)00256-2Get rights and content

Abstract

The calcium cycle, in particular carbonate dissolution, was analyzed in two deep eutrophic lakes, Lago di Lugano (288 m maximum depth) and Sempachersee (87 m) located in Switzerland. A box model approach was used to calculate calcite dissolution in the water column and at the sediment-water interface based on various lake monitoring data such as sediment traps, sediment cores, water and pore-water analysis. A model for stationary conditions allowing the calculation of calcite dissolution in the water column for a given particle size distribution was developed. The relative values of the simulated flux were consistent with sediment trap observations. The best fit of the dissolution rate constant of sinking calcite in Lago di Lugano was on the same order of magnitude (3 · 10−10 kg1/3 s−1) as published laboratory values for this surface controlled process.

Both lakes show a similar specific calcite precipitation rate of 170 g Ca m−2 a−1. The diffusive flux across the sediment-water interface amounts to about 15 and 10% of total calcite precipitation in Sempachersee and Lago di Lugano, respectively. However, 61% of the precipitated calcite is dissolved in the water column of Lago di Lugano compared to only 13% in Sempachersee. These results point towards the importance of grain size distributions and settling times in stratified deep waters as the two most important factors determining calcite retention in sediments of hard water lakes.

Introduction

Werner Stumm’s thinking as a coordination chemist had a large impact on the research agenda in low temperature geochemistry. The concept of surface complex formation provides an integrative view of geochemical processes at the mineral-water interface Stumm 1992, Stumm and Morgan 1996. This coordination chemistry approach has been first developed for oxide minerals and has been successfully extended to the surface chemistry of carbonates Kunz and Stumm 1984, Charlet et al 1990, Van Cappellen et al 1993, Schosseler et al 1999. At the same time, the laboratory chemist, Werner Stumm, had a strong interest in geochemical cycles (Stumm, 1977). In this context he advocated the use of lakes as “test tubes” in geochemical research (Stumm, 1985).

In this paper we address the topic of calcite dissolution in productive hard water lakes. In contrast to the marine environment, calcite is precipitated in lakes mainly during spring and summer by chemical mechanisms as a consequence of high supersaturation in the productive zone Stabel 1986, McKenzie 1985, Kelts and Talbot 1990, Hodell et al 1998. The question, “which fraction of the precipitated calcite finally accumulates in the sediment?” is relevant for two reasons. First, the size of calcite crystals has been proposed as a paleo-productivity indicator. High concentrations of phosphate inhibit calcite nucleation. As a consequence larger calcites are found in recent sediments deposited under eutrophic conditions compared to older sediment strata corresponding to mesotrophic or oligotrophic lakes Niessen and Sturm 1987, Lotter et al 1997. In order to interpret the size distribution of calcite in lake sediments, a quantitative estimate of calcite dissolution after precipitation is necessary. Second, the process of calcite dissolution is a central factor influencing water column stability in deep lakes. In the absence of intense wind-induced turbulence, the accumulation of dissolved Ca2+ and HCO3 in the water column can produce a density gradient which stabilizes the water column even against the geothermal heat flux Wuest et al 1992, Imboden and Wuest 1995. Thus, in deep hardwater lakes intense calcite dissolution may induce meromixis with important consequences for the redox chemistry in the deep water.

The Northern Basin of Lago di Lugano provides an ideal setting for a case study on calcite dissolution in a meromictic lake. The vertical thermal convection during winter does not reach the deepest point of the basin (288 m) and the deep water has not been in contact with the atmosphere since 1963 Barbieri and Mosello 1992, Wuest et al 1992. The mineralization of biomass leads to an accumulation of dissolved solids in the deep waters, a higher partial pressure of CO2, a decrease of pH and an accumulation of HCO3 which is balanced mainly by Ca2+Wuest et al 1992, Karagounis et al 1993. Thus the calcite dissolution plays a key role for the stabilization of the water column.

In contrast, Sempachersee is shallower (maximum depth 87 m) and is artificially mixed by compressed air during winter. Both lakes are ideal systems for case studies because large data sets from long term monitoring programs are available. In addition, sediment trap observations were performed in both lakes at depths of 20 m and between 85 and 90 m, which facilitates the comparison of these two depth intervals. In order to evaluate these data we will use two types of models: (1) We adopt a box model approach to determine the rates of calcite dissolution in the water column and fluxes at the sediment-water interface; and (2) Because calcite dissolution depends on particle size (Kunz and Stumm, 1984) we develop a one-dimensional model to quantify calcite dissolution in the water column as a function of the particle size distribution. We will show that differences in the size of precipitated calcite crystals and their residence time in the hypolimnion are the most relevant factors determining the burial efficiency of calcite in eutrophic hardwater lakes.

Section snippets

Study sites

Lago di Lugano is located south of the Alps at the Swiss-Italian border (Fig. 1). It is subdivided in two main basins. The Northern and the Southern Basin are separated by a frontal moraine, on which an artificial dam was built in 1844 (Barbieri and Mosello, 1992). Our study concentrated on the eutrophic Northern Basin with its deep anoxic hypolimnion (Table 1). Eutrophic, pre-alpine Sempachersee is situated in Central Switzerland (Fig. 1, Table 1). A rapid eutrophication due to excessive

Box model

Calcium discharge was calculated from the balance equation for Lago di Lugano, while it was based on measurements for Sempachersee (Table 3). The resulting discharge into Lago di Lugano was higher (460 g m−2 a−1) than in Sempachersee (280 g m−2 a−1) and the measured net sedimentation was lower for Lago di Lugano 44 g Ca m−2 a−1 than for Sempachersee 143 g Ca m−2 a−1Fig. 5, Fig. 6. Net sedimentation amounts to 10% of calcium discharge in Lago di Lugano and 50% in Sempachersee. The accumulation

Discussion

The box model approach used in this study combines observations in order to obtain average calcite dissolution rates at various depths in the water column and in the sediment. The calculation of the fluxes of turbulent diffusion included a typical error of ∼20% resulting from the eddy diffusion coefficient and calcium gradients (Aeschbach-Hertig, 1994). Net sedimentation rates have a standard deviation of ∼20% in Lago di Lugano and ∼10% in Sempachersee resulting from heterogeneity of the

Conclusions

Mass balance calculations based on sediment trap data and dissolved transport by turbulent mixing in the meromictic Lago di Lugano indicate that 61% of the precipitated calcite is redissolved in the water column. Model calculations starting with a measured particle size distribution indicate that in this system only calcite particles with a diameter larger than 40 μm reach the lake bottom at 288 m water depth. The dissolution process in a deep, meromictic hypolimnion can therefore dramatically

Acknowledgements

We thank Anette Hofmann and Januz Dominik from the Institute Forel and Alberto Barbieri and the Laboratorio di Studi Ambientali for access to their data on Lago di Lugano. René Gächter and Michael Sturm supplied data on Sempachersee and provided helpful insights. We are grateful to Erwin Grieder, Antonin Mares, André Steffen, Wisy Zwyssig, Michael Schurter and Christian Dinkel for their lab- and fieldwork. We wish to thank Jim Bischoff and two anonymous reviewers for their critical comments,

References (37)

  • R. Gächter et al.

    Gewässerschutz und Seeforschung

    Mitteilungen der Naturforschenden Gesellschaft Luzern

    (1993)
  • R. Gächter et al.

    Ten years of artificial mixing and oxygenationNo effect on the internal P loading of two eutrophic lakes

    Environ. Sci. Technol.

    (1998)
  • D.A. Hodell et al.

    Biologically induced calcite and its isotopic composition in Lake Ontario

    Limnol. Oceanogr.

    (1998)
  • Hofmann A. (1996) Caractéristiques géochimiques et processus de transport de la matière particulaire dans le bassin...
  • D.M. Imboden et al.

    Mixing mechanisms in lakes

  • I. Karagounis et al.

    A coupled physical-biochemical lake model for forecasting water quality. Application to the Northern Basin of Lake Lugano

    Aquatic Sciences

    (1993)
  • K. Kelts et al.

    Lacustrine carbonates as geochemical archives of environmental change and biotic/abiotic interactions

  • B. Kunz et al.

    Kinetik der Bildung und des Wachstums von Calciumcarbonat

    Vom Wasser

    (1984)
  • Cited by (40)

    • Calcite as a candidate for non-invasive phosphorus removal from lakes

      2021, Ecohydrology and Hydrobiology
      Citation Excerpt :

      Therefore the vast majority of the applied material will be probably preserved in hypolimnion and thus removable. Loss of Cc resulting from dissolution will depend on in-lake conditions (i.a. saturation with respect to calcite in hypolimnion) and grain size of the applied Cc-material, as smaller grains are more prone to dissolution (Ramisch et al., 1999). Thus, due to problem of Cc dissolution, application of very fine Cc powdered materials to hypolimnion should be avoided (median particle size of synthetic Cc materials can be even smaller than 0.1 µm (Heberling et al., 2016; Bańkowska-Sobczak et al., 2020)).

    • Quantitative evaluation of human and climate forcing on erosion in the alpine Critical Zone over the last 2000 years

      2021, Quaternary Science Reviews
      Citation Excerpt :

      For this reason, our analysis focuses on siliciclastic material, and we removed the carbonate component of the lake and river sediments before geochemical analyses. Although the authigenic carbonate production is poorly preserved in the deep basin of peri-Alpine lakes (Ramisch et al., 1999), this procedure also ensures that authigenic lake carbonates do not affect our interpretation. To that effect, after crushing in an agate mortar and drying at 60 °C for 72 h in a laboratory oven, river and lake sediment samples were decarbonated by two successive 5 mL HCl 0.5 N leaching steps to remove authigenic and detrital carbonate (supplementary S4).

    • Predicting the response of Hamilton Harbour to the nutrient loading reductions: A modeling analysis of the "ecological unknowns"

      2011, Journal of Great Lakes Research
      Citation Excerpt :

      While the latter point casts doubt on the anticipated efficacy of the on-going restoration efforts, we believe that it will be more prudent to first consider the relative contribution of other potentially important nutrient sources (i.e., internal loading, episodic events) that may intermittently fuel epilimnetic algal growth and therefore the role of the microbial loop as a nutrient supplier could have been overstated in our earlier work. Sediments act as an important source for a wide variety of chemicals in aquatic ecosystems, where a number of microbiological, geochemical, and physical processes determine the fraction of organic matter, nutrients, and pollutants released into the overlying water (Ramisch et al., 1999). Detailed knowledge of the processes occurring in the top few centimeters of the sediment can be essential for the assessment of water quality and the management of surface waters.

    View all citing articles on Scopus
    View full text