International workshop on

Koli (c) Metla/Erkki Oksanen

Development of Models and Forest Soil Surveys for Monitoring of Soil Carbon

April 5-8, 2006 at Koli, Finland

Effect of the model choice for simulating soil carbon stocks and stock changes: comparison of four soil models.

Patrick Faubert1, Esther Thürig1, Marcus Lindner1, Mikko Peltoniemi2, Taru Palosuo1, Oleg Chertov3, Alexander Komarov4, Alexey Mikhailov4, Felicitas Suckow5, Petra Lasch5, Martin Wattenbach6, Pete Smith6, Pia Gottschalk6.

1 European Forest Institute, Joensuu, Finland
2 Finnish Forest Research Institute (Metla), Vantaa, Finland
3 St. Petersburg State University, St. Petersburg-Peterhof, Russia
4 Russian Academy of Sciences, Institute of Physico-Chemical and Biological Problems in Soil Science, Moscow, Russia
5 Potsdam Institute for Climate Impact Research, Potsdam, Germany
6 University of Aberdeen, UK
* Contact information of the first author: European Forest Institute, Torikatu 34, 80100 Joensuu, Finland,
e-mail:, Phone: +358 13 252 0242, Fax: +358 13 124 393

Models are essential tools to analyze the effects of climate change and forest management on carbon stocks in forest soils. In this study, we evaluated how four different soil models affected the assessment of carbon stocks and stock changes in forest soils The models were 4C, ROMUL (the soil sub-module of EFIMOD), RothC and Yasso. All models were run for four forest sites in Finland and two sites in Germany. The sites differed in their climatic and soil conditions and stand composition (dominated by Scots pine or Norway spruce). The models were initialized with measured soil data (4C, EFIMOD), equilibrium runs with stand litter data (Yasso) or a combination of both (RothC). The simulation was run over a 20-year time period. All models were driven by litter input data generated with 4C and EFIMOD. Soil carbon stocks were compared for the carbon in the organic layer (4C, ROMUL), the mineral soil (4C, ROMUL, RothC), and the total soil carbon (4C, ROMUL, Yasso). Preliminary results showed that the simulated soil carbon stocks and stock changes varied considerably between the models. The initial carbon stocks of the models differed in a range of ± 45% from the average. The inter-annual total stock changes simulated by the four models ranged from -60% to +110% from the average. The 4C model simulated the highest stock increase followed by ROMUL, RothC (for the mineral soil) and Yasso. The differences can be partly explained by the different initialization procedures (measured values versus equilibrium assumption) and by the internal model assumptions (e.g. fine root allocation). This model comparison showed that the choice of the model considerably influences the estimation of carbon stocks. However, in order to assess the models results in terms of reality, they should be compared to measured soil carbon stock changes.

   Päivitetty:   02.03.2006 / EKel Metla : Events   Palaute Metlan etusivulle
. .