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

National Soil Surveys in Soil Carbon Monitoring

Mats Olsson

Department of Forest Soils, SLU, Uppsala, Sweden, e-mail:

Soil organic carbon may play a profound role for the greenhouse-gas balance since it is a big pool, that gobally, is 40% bigger than carbon bound in living terrestrial vegetation. The Marrakesh Accords specify that this C pool should be counted. Carbon monitoring is also justified by IPCC´s report on Good Practice Guidance. It states that it is good practice to use methods that provide the highest levels of certainty, i.e. highest tier level, and to use available resources as efficiently as possible. Soil organic carbon monitoring should therefore preferable be included in an integrated ecosystem assessment of all forest carbon pools, with explicit linkages between the soil, biomass and dead organic matter pools. Soil organic carbon is referred to as carbon in mineral soils and organic to a specified depth chosen by the country. Live fine roots can be included with soil organic matter where they cannot be distinguished from it empirically. The soil organic carbon pool does not include the O horizon (L, F and H horizons) that is classified as litter, which is not consistent with practise in soil science.

Monitoring of soil organic carbon stocks in mineral soils and their changes involves field sampling with or without stratified random sampling, extraction of soil samples for carbon analyses, laboratory analyses and evaluation of data. Changes in organic soils C stocks can not easily be measured by monitoring C stocks but have to be evaluated through measurements or estimations of emissions. Uncertainties may be due to sampling errors, systematic errors, analytical errors and errors in used models or functions. Particular errors may be due to wrong estimates of bulk densities and of the occurrence of coarse material.

Sweden, based on experiences from the National Soil Inventory, concluded that adequate depth of the soil carbon pool is 50 cm in mineral soils. Due to glacial soils with high content of coarse material core sampling is not possible. Instead soil samples are taken at set depths and by pedotransfer functions the real carbon stock is assessed. Different sample depths are used for different soils. A model for bulk density is applied, based on depth and carbon content. The content of coarse material is determined by the rod penetration technique. In total ca 23000 permanent sample plots are studied during 5-10 year cycles. Results from previous sampling, 1993 – 2002, show that the spatial variation within the country is high, eg for podsols resulting in a standard deviation of around 80 % of the mean carbon concentration in B horizons, or 70 % for the carbon stock to 50 cm . Based on knowledge of uncertainties it is possibly to estimate precision and number of samples needed to verify a certain C change. Because the inventory is carried out also for other purposes than carbon monitoring it can not be stratified in order to optimise detection of carbon stocks. ´The most important site factor for the soil carbon stock was soil moisture. Comparison of plots with podsols for two sampling periods, 1983-1988 and 1992- 2003 does not show any significant change in carbon stocks.

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