Storage of soil carbon depends on the carbon sources input, their chemical structure, and the decomposition rate or turnover rate of soil organic
matter. Some current state-of-the-art experimental approaches to determine the turnover rate of soil organic matter make use of natural stable isotope labeling experiments (Fig. 3.8) (Balesdent and Mariotti, 1996). In these experiments, the existing vegetation is replaced by structurally similar but isotopically different vegetation, i.e., C3 plants like wheat or rye (¿13C ~
-25%o) are replaced by C4 plants like maize (<513C--12%o). Initially all
SOM molecules are labeled according to the isotopic signature of the C3 crop (Fig. 3.8). Several years after the vegetation change the new vegetation differentially labels individual molecules (i.e., squares in Fig. 3.8 are already completely labeled whereas triangles are not labeled at all). The difference in <513C values of soil organic matter from a field without vegetation change and one with vegetation change can be used to calculate the fraction of remaining Q-;-derived carbon (Balesdent and Mariotti, 1996). Assuming exponential decay of carbon in soils at steady state the apparent residence time of total soil carbon or of individual compounds of soil organic matter can be determined (Gleixner et al, 1999). This apparent residence time indicates how much time is needed to label the complete pool of carbon with carbon from the new crop.
Corresponding turnover times for bulk soil organic matter, in the upper 25 cm of agricultural fields, are between 10 and 100 years (Balesdent and Mariotti, 1996; Collins et al, 2000; Paul et al, 2001). In forest systems the change from broad leaf trees to conifers or the use of labeled 13C02 m FACE experiments indicates that only a small amount of new plant-derived carbon enters the litter layer (Schlesinger and Lichter, 2001). Most carbon is immediately respired back to the atmosphere. For a vegetation change experiment involving a switch from a 120-year-old beech stand to a spruce stand at the Waldstein, Fichtelgebirge, Germany, the extremely low input of new carbon to soil organic matter is obvious (Fig. 3.9). The calculated mean residence time for soil organic matter increased from 60 years in the
Mean turnover time (a)
Figure 3.9 Turnover time of soil organic matter from different depths of a beech stand converted to spruce. Litter layers are indicated with negative depth.
litter layer down to more than 5000 years at 10-30 cm soil depth. However, these turnover rates contrast with age of the bulk soil organic matter in these profiles, which are calculated to be less than 500 years old using 14C analyses. Similar to the bulk 13C values, a mixture of old soil carbon with new highly 14C enriched 'bomb' carbon ends up with a mean 14C age that is difficult to interpret (Trumbore, 2000).
Physical fractionation of soil carbon partially overcomes this problem. Soil organic matter found in the sand fraction or in the light density fraction of soil has a shorter turnover time than carbon found in the silt/clay or the heavy density fraction (Balesdent and Mariotti, 1996). It is assumed that carbon derived from plant litter first enters the sand or light fraction, as this pool is labeled quickly by new vegetation. Later in the course of the degradation, mineral-organic complexes are formed and carbon is stabilized on these complexes, which belong to the heavy or silt/clay fraction, and which are labeled more slowly by the carbon input from new vegetation (Sohi etal, 2001; Six etal., 2002). However, so far it is not clear what source of carbon, plant or microbial carbon, enters the more stable mineral fraction. Only compound-specific isotope ratios will give insight into this process.
Was this article helpful?