Exploring the use of sequential sampling for monitoring organic matter stocks in a grazed and non-grazed Scots pine stand

The possible reduction of soil organic matter stocks under a grazing regime was studied by sequential, intensive sampling over a seven-year period. The measured soil organic matter stocks were highly variable, and overall no significant changes in soil organic matter stocks could be observed. The variation was partly due to methodological errors, but particularly to high spatial variability, which obscures small changes over time. This means that, even in this relatively homogeneous ecosystem with the same parent material and vegetation, quantifying temporal changes in stocks of soil organic matter is very difficult. In order to examine whether significant changes in organic matter stocks could be expected at all, a sensitivity analyses was done using a balance model. We used a simple balance model because we did not want to predict changes in decomposition rates or litter input, but only analyse the potential changes in soil organic matter stock under realistic and more extreme scenarios. The model calculates the development of organic matter stocks under various litter inputs (measured values reduced with 0–70%) and decomposition rates (measured k-values decreased and increased with 50%). In this way, the potential impact of grazing on soil organic matter was explored. The numerical exercise confirmed the results of the observational study and showed that meaningful differences can hardly be expected. Significant differences between control and grazed situations could only be expected under unrealistic conditions or excessively large sample sizes. Because litter input and decomposition are small compared to the stocks of soil organic matter, the latter appeared a very robust parameter, which is hardly reduced by grazing in spite of the obvious reduction in grass standing biomass. It can be concluded that in this kind of field experiments or in monitoring programmes focused on carbon sequestration, sequential sampling will not reveal changes in organic matter stocks, probably not even when measurements are continued for many years. A sensitivity analyses, based on litter input data and different estimated decomposition rates can improve the design of a sampling strategy.