Hydrological Controls on the Carbon Dynamics of Boreal Peatlands

Pristine Boreal River Valley Mire Ust-Pojeg in NW-Russia

Project Background

Within an Integrated Research Activity of CliSAP-1, the CRG studied the “Hydrological controls on the carbon dynamics of boreal peatlands”. This project was conducted in close cooperation with the Komi Science Centre, Syktyvkar, the Ernst Moritz Arndt University Greifswald and the CRG Chemistry of Natural Aqueous Solutions. Further collaborations have developed within CliSAP through the MPI Research Group "Climate-Biogeosphere Interactions" to develop upscaling modeling strategies for dealing with variations in peatland topography on a microtopographic scale.

The Ust-Pojeg study site (62° N, 50 'E, 119 m a.s.l.) is situated in one of the large boreal river valley mire complexes that are widespread in northwest-Russia and West Siberia. The studied pristine mire is in a transitional state from fen to bog following paludification, and contains ombrogenous, minerogenous and transitional forest swamp zones. The vegetation ranges from forest to sedge and moss cover within the wetlands, and contains an abundance of microtopographic features such as hummocks, hollows, and lawns.


Previous investigations have focused on the vertical fluxes of CO2 and CH4 between the mire and the atmosphere. This gas exchange is measured by the closed chamber method on the micro site scale and by the micrometeorological eddy covariance technique on the ecosystem scale. Preliminary results demonstrate the great effect of hydrologic dynamics on the vertical fluxes of CO2 and CH4 in the temporal as well as in the spatial domain. The lateral input, throughput, and output fluxes of water and carbon require better characterization to complete the local carbon and water budgets. In the summer of 2010 we began detailed investigations of the hydrological processes and the mire water biogeochemistry (DOC, DIC) of the Ust-Pojeg mire complex. This project has provided data and process knowledge on the coupled water and carbon cycles of mire wetlands necessary for the validation and advancement of regional hydrological models and currently developed wetland modules of dynamic earth system models.


The overall goals of our integrated research activities are to enhance understanding of how the hydrology of boreal taiga landscapes controls the carbon dynamics of the extensive boreal peatlands, and to evaluate the role of the peatlands in the regional water and carbon budgets of the boreal taiga landscapes. These goals shall be achieved by combining hydrological field measurements in the Ust-Pojeg mire complex, biogeochemical laboratory analyses and process-based modeling studies.

In detail, the objectives of the project are:

  • to analyse the spatial and temporal dynamics of the mire hydrological processes and to characterise all components of the water balance of the investigated mire, including overland flow, lateral flow, infiltration, precipitation, and evaporation, and to investigate the role of the peatlands in flood attenuation and/or amplification;
  • to quantify the import and export of DOC and DIC to and from the mire, and to model the average residence time of carbon in each compartment;
  • to characterise the biogeochemical composition of DOC and DIC in inflow, storage and outflow waters and thus gain new insights into the processes that control the generation, transport and transformation of DOC and DIC in boreal mires;
  • to quantify the complete carbon balance of the investigated peatland by combining the results of lateral carbon fluxes with the results for the vertical mire-atmosphere carbon fluxes investigated by the Komi Science Center and the University of Greifswald.


Monthly mean diurnal pattern of site’s energy fluxes – net radiation (Rn, black), latent energy (LE, blue), sensible heat (H, red), and ground heat flux (G, green) – during the study period (April 2008 – February 2009).

The JRG is working towards a comprehensive understanding of the energy balance of this typical boreal mire site, with a particular focus on the evapotranspiration (ET) rate. Since ET is a predominant factor in controlling the relative water table height, these energy and water balance dynamics play a key role in determining the proportion of aerobic and anaerobic sites within the peatland. These factors together control the CO2 and CH4 emissions from the site as well as govern the amount of water discharged to the fluvial system. We have performed a careful look at the energy balance on daily, monthly, and annual time scales (Runkle et al, 2014, J. of Hydrology). This work has revealed the predominance of net radiation in driving ET, as opposed to vapour pressure deficit, a more challenging term to interpolate from scarce weather stations.

This finding encourages relatively simplistic modelling approaches for this landscape, which are being taken up in a statistical approach in collaboration with V. Brovkin’s research group at the Max Planck Institute for Meteorology (Cresto-Aleina et al, in prep). We are also working to integrate measurements of local changes in water table, spatial variations in surface water stable isotope ratios, and a reanalysis of transparent chamber data to generate a spatially varying picture of ET rates across the peatland’s heterogeneous landscape cover (Runkle et al., in prep). These measurements will then be compared to the modelling work performed by Cresto-Aleina et al.

Our water-chemical work showed substantial spatial variation in dissolved carbon concentrations across the mire complex, with higher concentrations in the marginal forest swamp of the mire complex relative to the fen and bog portions of the site (Avagyan et al., 2014, submitted). The importance of this interface region between peatland and forests was discovered also through spring fieldwork showing that a high proportion of spring snowmelt flowed through this zone, carrying outflow waters with a chemical signature more similar to this source region than that of the central mire complex (Avagyan et al., in prep). The description of site biochemistry has also benefited from improvements on the field-usage and calibration techniques of a portable UV-Visible spectrophotometer (Avagyan et al, 2014, J. of Hydrology). When choosing the absorbance of light at selected wavelengths as a proxy for dissolved organic carbon concentration – the UV wavelength 254 nm is a classic choice – our work reveals that it is essential to use multiple wavelengths in different light ranges in order to achieve the optimal explanatory power and accuracy of the proxy-based dissolved organic carbon quantification. Furthermore, our work demonstrates the variation in spectral absorption characteristics that occur in waters from different regions of the peatland site (e.g., bog, fen, or marginal forest swamp).

Dissolved organic carbon concentrations (cDOC) at three ecohydrological landscape units (swamp forest, fen, bog) within the studied boreal river valley mire complex at different soil depths through the summer-autumn period, 2010.

The hydrological work focused on the spring snowmelt period in mire-forest landscapes, which is special in that its carbon balance and its hydrology are dominated by lateral fluxes composed of snowmelt carrying dissolved organic carbon (Runkle et al., in prep.). Field work from the spring of 2011 demonstrated how the snowmelt dynamics differ between the mire complex, its surrounding upland forest soils, and a regional river draining a forest-mire catchment. These partially-nested spatial delineations reveal the importance of size and scale in understanding lateral dissolved organic carbon fluxes and provide appropriate case studies for up-scaling to regional hydrological and carbon stock modelling. In particular, smaller peat-filled watersheds show reduced dissolved organic carbon concentrations due to the diluting effect of snowmelt (M.Sc. thesis H. Haupt). In contrast, at the same time, the regional rivers show increased dissolved organic carbon concentrations due to the increasing influence of organic surface layers on flow while the underlying mineral layers have remained frozen.