Uncertainty of upland soil carbon sink estimate for Finland

Abstract

Changes in the soil carbon stock of Finnish upland soils were quantified using forest inventory data, forest statistics, biomass models, litter turnover rates, and the Yasso07 soil model. Uncertainty in the estimated stock changes was assessed by combining model and sampling errors associated with the various data sources into variance-covariance matrices that allowed computationally efficient error propagation in the context of Yasso07 simulations. In sensitivity analysis, we found that the uncertainty increased drastically as a result of adding random year-to-year variation to the litter input. Such variation is smoothed out when using periodic inventory data with constant biomass models and turnover rates. Model errors (biomass, litter, understorey vegetation) and the systematic error of total drain had a marginal effect on the uncertainty regarding soil carbon stock change. Most of the uncertainty appears to be related to uncaptured annual variation in litter amounts. This is due to fact that variation in the slopes of litter input trends dictates the uncertainty of soil carbon stock change. If we assume that there is annual variation only in foliage and fine root litter rates and that this variation is less than 10% from year to year, then we can claim that Finnish upland forest soils have accumulated carbon during the first Kyoto period (2008)(2009)(2010)(2011)(2012).

Underestimation of boreal soil carbon stocks by mathematical soil carbon models linked to soil nutrient status

“…Some level of deviation from measurements and poorly explained spatial variation ( Fig. S7) was expected from the uncertainties of the SOC measurements, annual plant litter inputs and climate 330 variability for the model SOC stock change estimates (Ortiz et al, 2013;Lehtonen et al, 2015a).…”

Evaluating CENTURY and Yasso soil carbon models for CO2 emissions and organic carbon stocks of boreal forest soil with Bayesian multi‐model inference

“…However, prediction of the magnitude and spatial variation of SOC is far from perfect (Hashimoto et al, ; Lehtonen et al, ; Todd‐Brown et al, ; Ťupek et al, ). Model uncertainties hinder conclusions on both changes of SOC and CO 2 emissions (Lehtonen & Heikkinen, ). The imbalance between observed and modelled soil carbon stocks can be caused by incorrectly represented or missing biotic and abiotic factors of long‐term soil carbon accumulation (Schmidt et al, ; Todd‐Brown et al, ).…”

Stocks and Ecological Significance of Soil Carbon in Tanzania

“…Lal [29] argued that climatic conditions, soil types, vegetation types and agronomic practices are among the prominent factors influencing carbon stock and changes in most tropical areas. All these concepts and factors are also acknowledged by different authors [7,8,20,26,30].…”

“…On the other hand, temperature acts as catalyst to increase the rate of chemical reactions particularly the decomposition of litter which is a useful data input in SOC modeling. When other factors are constant; soil organic matter is higher in cooler climate than warmer climate [21,30].…”