4.2 Sequestration of soil organic carbon by vegetation
Sequestration of soil C implies the transfer of atmospheric CO2 into the soil of a land unit through plants grown on that land (Lal et al. 2015). Although the effects of afforestation on soil C storage in arid and semi-arid areas have been examined in the past, the results continue to be uncertain: some found that afforestation of grasslands had no effect on SOC from 5 to 30 years after planting (Davis and Condron 2002,Perez-Quezada et al. 2010, Cunningham et al. 2012) whereas Paul et al. (2002) found that following such afforestation, soil C decreased during the first 5 years but recovered later so that after about 30 years, the difference was either minimal or that soil C was slightly more and continued to increase gradually. Hu et al. (2008) suggested that the loss of C in the mineral component of soil was partly compensated for by the increased stocks of C in roots. Jackson et al. (2002) found a clear negative relationship between precipitation and changes in soil organic matter in grasslands that had been invaded by woody vegetation, with drier sites gaining SOC and wetter sites losing it. In contrast with these studies, our results suggest that planting trees in grasslands could increase SOC significantly, a finding supported by Li et al. (2013), who maintain that the accumulation of C in soil may be due to annual inputs through net primary production that exceeds the amount lost by decomposition. However, these effects were found to vary with the region, because they are influenced by soil texture, tree species, age of the stand, and management practices. Therefore, more empirical data are needed to identify the point in time after which soil C begins to be sequestered as a result of afforesting a desertified grassland, such as the one examined in the present study.
Storage of soil C decreased with depth, an observation consistent with that made by Grandy and Robertson (2007), the storage being maximum in the 10–20 cm layer. The concentration of soil C across all depths (except the 90–100 cm layer) differed significantly between the three treatments and the control. Among the three treatments, the concentration in the 0–40 cm layer was significantly higher in FG than in UG and MG. Total C in the uppermost layer (0–5 cm) was found to increase substantially after tree planting, but that in the 0–30 cm layer remained the same (Cunningham et al. 2012). Noble et al. (1999) also found higher SOC in the 0–5 cm layer of a plantation than that of a pasture, but not at greater depths (5–50 cm). Similarly, at most of the 28 sites studied by Davis and Condron (2002), afforestation showed no pronounced influence on soil C at depths greater than 10 cm.
The total estimated pool of SOC up to a depth of 100 cm was 23.63–79.98 t/ha, which amounted to 80.12% of the total pool in the ecosystem, showing that terrestrial C is sequestered mainly in soil. This finding is consistent with the findings of Fang et al. (2001) and Li et al. (2007). Dixon et al. (1994) reported that more than a third of the C pool in terrestrial ecosystems is contributed by soil organic matter. Detwiler (1986) reported that 35%–80% of the C stored up to a depth of 100 cm is concentrated in the 0–40 cm layer in tropical and subtropical soils; we obtained similar results, with 41.81%–60.13% stored in the 0–30 cm layer and 59.42%–80.80% in the 0–50 cm layer. The accumulation of SOC is controlled by the rates of biomass formation and decomposition (Lal et al. 1995). The turnover of fine roots is another major component of the dynamics of SOC; the size of this turnover matches that of leaf litter (Rasse et al. 2005). In the present study, the lowest amount of soil C was in the control, probably because of much lower input of litter, greater abundance of woody and herbaceous vegetation, different patterns of distribution of roots at different depths, and the differences in the availability of soil moisture and soil temperature, all of which accelerated the decomposition of organic matter (Covington 1981, Jackson et al. 2000). The storage of soil C was significantly greater in MG than in FG and UG, an observation consistent with earlier observations on the enrichment of soil organic matter as a result of greater vegetation cover ensured by appropriate intervention (Fearnehough et al. 1998, Wezel et al. 2000).