EGU24-6255, updated on 08 Mar 2024
EGU General Assembly 2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

A Synthesis of Sphagnum Litterbag Experiments: The Role of Initial Leaching Losses and a Test of the Holocene Peatland Model

Henning Teickner1, Edzer Pebesma2, and Klaus-Holger Knorr1
Henning Teickner et al.
  • 1University of Münster, Institute of Landscape Ecology, Ecohydrology and Biogeochemistry Group, Münster, Germany
  • 2University of Münster, Institute for Geoinformatics, Spatio-temporal Modelling Lab, Münster, Germany

Decomposition is one of the major controls of long-term sequestration of carbon in northern peatlands. Our knowledge of the magnitude and controls of decomposition rates is derived to a large extent from litterbag experiments and estimated decomposition rates and environmental controls inform decomposition modules in dynamic peatland models.

Here, we combine synthesized Sphagnum litterbag data from 15 studies with simulation and modeling to address the following questions:

1. How large are initial leaching losses in Sphagnum litterbag experiments?

2. How does considering or ignoring initial leaching losses affect decomposition rate estimates?

3. Can the Holocene Peatland Model (HPM) (Frolking et al., 2010) predict decomposition rates from litterbag experiments?

We provide a systematic overview on Sphagnum decomposition rates and initial leaching losses. Data from litterbag experiments suggest that the assumption that leaching losses from Sphagnum litterbag experiments generally account for only few percent of the initial mass is wrong. Average initial leaching loss estimates range between 2 to as much as 22 percent of the initial mass. Ignoring initial leaching losses when estimating one-pool decomposition rates can bias predicted remaining masses when extrapolated to several decades because decomposition rates are overestimated.

With standard parameters, the HPM had an average root-mean square error (RMSE) of 0.06 yr-1 for decomposition rates estimated separately from litterbag data (reference decomposition rate estimates). The HPM and reference decomposition rate estimates could be made compatible with each other (training RMSE = 0.02 yr-1) by constraining the reference decomposition rate estimates and by adjusting HPM parameters with information from the litterbag experiments.

In terms of HPM parameters, the analysis suggests that oxic decomposition rates may be fastest at larger water contents and that anoxic decomposition rates may be less limited with depth below the water table (= larger under anoxic conditions) than assumed by the HPM, indicating either misspecification of the HPM or the influence of varying water table levels on the litterbag data. Since a previous sensitivity analysis of the HPM has shown that limitation of anoxic decomposition rates is important for peat accumulation (Quillet et al., 2013), the HPM may currently overestimate peat accumulation rates.


Frolking, S., N. T. Roulet, E. Tuittila, J. L. Bubier, A. Quillet, J. Talbot, and P. J. H. Richard. 2010. “A New Model of Holocene Peatland Net Primary Production, Decomposition, Water Balance, and Peat Accumulation.” Earth System Dynamics 1 (1): 1–21.

Quillet, Anne, Michelle Garneau, and Steve Frolking. 2013. “Sobol’ Sensitivity Analysis of the Holocene Peat Model: What Drives Carbon Accumulation in Peatlands?” Journal of Geophysical Research: Biogeosciences 118 (1): 203–14.

How to cite: Teickner, H., Pebesma, E., and Knorr, K.-H.: A Synthesis of Sphagnum Litterbag Experiments: The Role of Initial Leaching Losses and a Test of the Holocene Peatland Model, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6255,, 2024.

Comments on the supplementary material

AC: Author Comment | CC: Community Comment | Report abuse

supplementary materials version 2 – uploaded on 11 Apr 2024
  • CC1: Comment on EGU24-6255, Xiaoying Shi, 16 Apr 2024

    This research study using the observed litter bag experiemental data to constrain the peat model and improve the model predictive capacity of decomposition rate. Like this data-model integrated approach.

supplementary materials version 1 – uploaded on 09 Apr 2024, no comments