4.3. Analysis on factors affecting soil CO2 flux at different layers
Climate-driven losses of soil carbon are currently occurring across many ecosystems, with a detectable and sustained trend emerging at the global scale (Bond-Lamberty, Bailey, Chen, Gough, & Vargas, 2018). As the major driving factors of soil CO2 production, temperature and moisture vary vertically within the soil profile (Davidson E A, 2006), soil temperature rises slowly from the surface to the bottom in spring and begins to cool down from the surface (Subke et al., 2003). Soil organic carbon (SOC) is one of the key indexes for assessing soil quality. The content of soil organic carbon decreased with soil depth increasing and is significantly different at vertical profiles(P <0.01), and changes in SOC may have a huge potential impact on global climate. (Wang, Fu et al. 2010). Furthermore, related studies have shown that soil temperature and moisture explain the temporal and spatial variation of soil CO2emissions. Similarly, changes in soil organic carbon significantly increase soil CO2 flux, and there was a strong positive correlation between CO2 flux and SOC (Mande et al., 2015).
Some scholars believe that the average Q10 of forests in the world is 1.5(Luo, Wan, Hui, & Wallace, 2001), and the Q10 value of the forest in the north temperate zone was between 0.9 and 2.2 (Gulledge & Schimel, 2000). In our study,the range of Q10 in Robinia pseudoacacia of different stand ages was 1.245±0.077 to 2.121±0.404, and Q10 value at different layers was significantly different (P<0.01). With the increase of soil depth, the Q10 at different layers increased, and the Q10 (2.121±0.404) at deep layer (200 cm) was significantly higher than that(1.245±0.077) at surface layer. A similar phenomenon observed in Georgia, USA, was that the Q10increased with increasing the depth of the soil layer. and the researchers believed that this phenomenon may be caused by a decrease in soil temperature as the depth of the soil increases(Pingintha, Leclerc, Beasley, Zhang, & Senthong, 2010), which may reflect root growth and root input at oligorganic layer has a relatively greater important effecct than that at O-horizon(Davidson, Savage, et al., 2006). And higher Q10 at deep layers may also be related to factors such as humidity limitation, nutrient availability and so on(Davidson, Janssens, & Luo, 2006; Graf et al., 2008). Some other scholars have found that, in a deep warming experiment in mineral soil, all depths responded to warming with similar temperature sensitivities, driven by decomposition of decadal-aged carbon. Whole-soil warming revealed a larger soil respiration response than many in situ experiments and models (Pries, Castanha, Porras, & Torn, 2017). Our study further confirmed that: under the external environment disturbance, the same temperature change occurs in deep layers as shallow layers, deep layers revealed a larger soil CO2 flux response than shallow layers. Therefore, the disturbance occurred at deep layers should be close concerned in future human activities and natural environment improvement.
The response mechanism of moisture to soil respiration is complicated, and the change of moisture in soil surface was generally larger than that in deep layers (Davidson E A, 2006). Therefore, soil CO2 flux at different layers will change due to fluctuations in moisture. In Kog.Ma experimental site of Thailand, Hashimoto et al. (Hashimoto et al., 2007) found that the CO2 flux at different layers varied with the change of moisture, and the CO2 flux of surface soil was larger in rainy seasons than that in dry seasons, while there presented the opposite trend at deep layers. Davidson et al. (Davidson E A, 2006)also found that the release rate of CO2 in the surface layer and O-layer of mineral soil was significantly correlated with moisture(P<0.001), but the correlation between the upper and middle layers of A-layer was vary low. This indicates that the effect of alternate drying-wetting on soil CO2 release only occurs at the O-layer. In our study, the changes of soil moisture inRobinia pseudoacacia of stand ages or at different layers were significant. There was certain quadratic function relation between soil CO2 flux and moisture(Fig6.), due to temperature and matrix and other factors at profile of (20cm- 80cm) lying between surface layer and deep layer, the response of moisture to soil CO2 flux was relatively slow compared to the surface and deep layers. The reasons of above phenomenon may be that: destruction of soil aggregates accelerated by near-surface water replenishment leads to increase of matrix availability, and then which leads to enhance of microbial activity, all this stimulates CO2 emissions. At deep layes, soil CO2 flux, under the stimulation of water replenishment, the dominant mechanism affecting CO2 flux may shift from ”substrate supply” to ”microbial stress”, and there presents a distinct feedback (J. Wang, Liu, Chen, Liu, & Sainju, 2015).
In this study, soil CO2 flux increased with the increase of soil organic carbon at the lower layers. while it decreased at near-surface layers. Related studies suggested that soil organic carbon concentration decreased with increasing of soil depth in arid and semi-arid loess hilly area, and there was a significant difference in vertical profiles (P <0.01) (Y. F. Wang, Fu, Lu, Song, & Luan, 2010). Based on the theory of Michaelis equation (Moorhead & Weintraub, 2018), the concentration of organic carbon substrate used for mineralization reaction at the lower layer was not saturated, and it still had a positive feedback on mineralization rate. At the near-surface layers, in addition to the normal organic carbon sequestration process, the forest floor litter had a great influence on the physic-chemical environment, and the activities of microorganisms and roots were greatly changed. The feedback mechanism of organic carbon substrate on the CO2 flux became a far more complex, and thus soil organic carbon substrate showeds a certain negative feedback effect.