4.2. Contribution of CO2 flux in deep layers to soil-atmosphere interface
The contribution of soil CO2 flux at different layers to the total CO2 flux at the soil-atmosphere interface exist some differences. The contribution is always evaluated with average proportion of soil CO2 flux at certain layer to soil-atmosphere interface. Nevertheless, with temporal and spatial change, the factors affecting the CO2 flux change, and there are dynamic changes in CO2 flux of each layers. In this study, variance contribution of proportions was introduced for dynamic evaluation,and this method can further reveal the dynamic process in the contribution of CO2 flux at profiles.
Davidson et al. (Davidson & Trumbore, 1995) took study in forests and cultivated land of Amazon and found that soil CO2emissions at depth of 1 m accounted for 70%-80% of total soil emissions at soil-atmosphere interface; Gaudinski’ studies showed that Soil CO2 fluxes within shallow layers of 15 cm accounted for 63% of total soil respiration in temperate forests(Gaudinski, Trumbore, Davidson, & Zheng, 2000); Davidson et al. (Davidson E A, 2006) found that the contribution of CO2 flux at in the O-horizon soils to total soil respiration was about 40% -48% in the broadleaved mixed forest of Massachusetts. Compared with the previous research results, our study focuses on the contribution of soil CO2 flux at deep layers (< 80 cm) to the total CO2 flux. The static evaluation results show that The contribution rate of CO2 flux in Robinia pseudoacacia of stand ages (CH10a, CH20a, CH30a, CH40a) at deep layers to soil-atmosphere interface is generally larger in growing seasons, and the difference among different stand ages is not obvious. The soil CO2 flux of at deep layers accounts for 17.76%-21.86% of the total CO2 flux at soil-atmosphere interface. To sum up, that phenomenon may be due to the rich soil organic carbon content at deep layers, combined with the results of soil microbes and roots. Therefore, in the study of carbon emissions at soil-atmosphere interface,we cannot follow the traditional understanding and ignore the impact of CO2 flux deep layers just owning to its low content. According to the results of dynamic evaluation,the variation of contribution in growing season was lager than that in non-growing season, and the contribution rate in Robinia pseudoacacia of CH10a was 27.46% in the growing season, which was larger than that in other stand ages, showing a certain heterogeneity. Similar results have been reported that, Pumpanen (Pumpanen, Ilvesniemi, & Hari, 2003) believeed that the contribution of CO2 flux at bottom soils in winter to the total flux is greater than that in summer, while Risk (Risk, Kellman, & Beltrami, 2008) proved that the contribution rate at deep soil was larger in the later growing season, all of which corroborated the results of our study to a certain extent, but there is also a certain heterogeneity. The reasons for this phenomenon may be that: thick soil layer was formed due to its unique soil development process in the loess hilly region, the hysteresis of the seasonal temperature change, the dry soil layer and temporary dry layer formed by water deficiency and provisional water deficiency have a great influence on the soil CO2 flux at the deep soil layer. The heterogeneity of the contribution in Robinia pseudoacacia of CH10a in the just confirms the fact that intensification of organic carbon mineralization induced by the rhizosphere stimuli in agricultural land and abandoned arable land (Shahzad et al., 2018).