Resubmitted to Biogeochemistry, date 21/09/2010

Supplemental Information

Drainage-induced forest growth alters belowground carbon biogeochemistry in the Mer Bleue bog, Canada

Christian Blodau*1,2,3 and Melanie Siems1,2

1Limnological Research Station and Department of Hydrology, University of Bayreuth, Bayreuth, Germany

2Department of Geography, McGillUniversity, Montreal, Canada

3present address: School of Environmental Sciences, University of Guelph, Canada

*author for correspondence; School of Environmental Sciences, University of Guelph, phone: +1 /519-824-4120 ext. 56203; fax: +1 /519-824-5730; e.mail:

1. Assessment of vertical solute transport mechanism

Method

To assess whether a diffusion based inverse pore water modelling was adequate to estimate production rates of dissolved inorganic carbon and methane we first assessed vertical transport mechanism using a simple solute box model implemented in STELLA as described in (Beer & Blodau 2007).We attempted to reproduce measured chloride concentration profiles with and without vertical advection of water. Concentrations at the water table and in the lowest segment were used as constant boundary conditions. The diffusion coefficient of chloride, 9.9×10-5 m2 d-1 (5°C) (Li & Gregory 1974), was corrected for average soil temperature (8°C) (Lerman 1979) and porosity, which was calculated for bulk and peat density (1.61gcm3) taken from Blodau & Moore (2002). Several advection rates of water were manually tested to obtain a best fit of measured and modelled chloride concentration profiles. Based on the advection rate producing best fit, the dominating vertical transport processes was estimated from calculation of the dimensionless Peclet number (Pe=vzdD-1; d: average particle diameter) using the same parameters as described in Beer & Blodau (2007). The average particle diameter was assumed to be 1mm, based on data obtained for similar peat soil types (Heiskanen 1995).

Results

Measured chloride concentrations ranged from 50µmolL-1 to 2,789µmolL-1 with highest values at 200m site and lowest values at 30m site (Figure 2s). As expected, concentrations strongly increased with depth at 60m, 200m and forested site, whereas at 30m site concentrations were surprisingly more or less constant, very low and hardly changed with depth (see Figure 2s). At forested site chloride concentrations increased almost linearly from 133µmolL-1 at water table to a maximum of 1,745µmolL-1 at 180cm depth (1648µmolL-1 at 205 cm), which suggests a diffusion-dominated process.

This was confirmed by the application of the diffusive-advective box-modelsince the diffusion-only scenario (q=0) correlated best with the measured data (R²=0.95). According toFetter (1993) a Peclet number of Pe <0.4 (non-dimensional) represents diffusion-dominated transport.Calculated values alongthe Mer Bleue transect did not exceed 0.02 at any plot. The comparison of measured chloride concentrations from MLP-data to those modelled by STELLA and the calculated Peclet numbers for all depth profiles lead to the conclusion that diffusion is the more effective vertical transport process at all sites, despite some advective vertical transport.

Even though the q-values are too small to indicate an advection dominated transport mechanism, they nevertheless indicate a vertical transport direction. Negative q-values represent a downward transport, positive q-values an upward movement of water. Thus, solute transport at forested site was directed upwards. By contrast at 30m site transport was directed downwards, which is in agreement with the more detailed investigation of groundwater movement by Kopp(Kopp et al. 2010).

Figure 1s: Measured (symbols) and modeled (lines) chloride concentrations. Advection rates q in units of m d-1.

2. Dissolved organic matter concentrations in pore water peepers

Figure 2s: Dissolved organic carbon concentrations (DOC) in pore water peepers (PP) (± s.d. when n = 3). The data show the enrichment in DOC at the forest site relative to the other sites in the surface layers of the peat, in agreement with MLP data.

References

Beer J & Blodau C (2007) Transport and thermodynamics constrain belowground carbon turnover in a northern peatland. Geochimica et Cosmochimica Acta 71: 2989-3002

Blodau C & Moore TR (2002) Macroporosity affects water movement and pore water sampling in peat soils. Soil Science 167: 98-109

Fetter C (1993) Contaminant Hydrogeology. Macmillan Publishing Company, New York

Heiskanen J (1995) Physical properties of two-component growth media based on Sphagnum peat and their implications for plant-available water and aeration. Plant and Soil 172: 45-54

Kopp B, J., Fleckenstein JH, Roulet NTR, Humphreys E & Blodau C (2010) Impact of long-term drainage on groundwater flow patterns in the Mer Bleue peatland, Ontario, Canada. Hydrologic Processes, submitted

Lerman A (1979) Geochemical processes: Water and Sediment Environments. John Wiley & Sons, New York

Li Y & Gregory S (1974) Diffusion of ions in sea water and in deep-sea sediments. Geochimica Cosmochimica Acta 38: 703-714