Soil respiration (Rs) is the largest terrestrial carbon flux into the atmosphere, accurate estimation of Rs is not easy at finer spatial scales (even within the one ecosystem) due to the temporal-spatial variability of Rs, which is generally caused by different biotic and abiotic factors. Soil temperature (ST) and water content (SWC) are recognized as the main factors in controlling the temporal variability of Rs. However, the relative importance of these abiotic and biotic factors in the spatial heterogeneity of Rs has yet to be directly assessed, and the complicated interactions of these factors weaken our understanding of the spatial heterogeneity in Rs.
We conducted a field experiment in a subtropical forest to examine spatial heterogeneity in Rs for 2 years in relation to soil, root, and forest structural parameters. We intended to explain spatial heterogeneity in Rs on small-scale differences among point measurements within a stand at inter-annual time scale. We predicted the spatial distributions of Rs, ST and SWC based on Krige interpolation. Linear regressions were performed to examine the relationships between these predictors and Rs. We fitted a piecewise structural equation model (SEM) to infer relative importance of microclimate, topography, root, litterfall, soil characteristics and forest structure on Rs.
Results/Conclusions
ST and SWC exerted considerable influence on the temporal variability in Rs. Considering the spatial covariability of ST and SWC, a significant exponential relationship [Rs = α × exp (β × ST) × SWC] between ST, SWC and Rs was observed in this study (R2 = 0.76, p < 0.001). The maximum correlation (Rmax) was found at a radius distance of 8 m between forest structural parameters and Rs (R2max = 0.08, 0.08, and 0.10). The SEM explained 34.2 % of the spatial heterogeneity in Rs (χ2 = 28.75, df = 26, p = 0.32). The model showed that the elevation had the strongest total effect on Rs through its direct and indirect effects on SWC (both p < 0.05), and the forest structural parameters (TBA (8), diversity (8) and productivity (8), respectively) had relatively strong effects on Rs (both p < 0.05). The findings highlight the importance of considering the influence of topography and forest structure on spatial heterogeneity in Rs. Although current earth system models have considered both plant and environmental factors in explaining Rs variability, we highly recommend the models to better represent the topography and forest structure especially in the subtropical forest.