Variability Of Soil Organic Carbon Transit Time In Forest Systems And Its Driver At Global And Transect Scales

One known bias in current Earth System Models?ESMs?is the underestimation of global mean soil organic carbon transit time(?_soil),which quantifies the age of the carbon atoms at the time they leave the soil.However,it remains unclear where such underestimations are located globally.Research over the past several decades has clarified the mechanisms and timescales involved in stabilizing organic matter in soils,but we still lack processbased understanding for predicting how long will the soil organic carbon stay in the ecosystem,given environmental changes across a range of landscapes.Part of the problem is the emphasis on short-term studies and processes that dominate carbon balance at the point or soil profile scale,whereas other processes that dominate over longer timescales and larger spatial scales may actually be more important for determining the carbon balance of soils in a region.Thus,we constructed a global database of measured ?_soil in order to evaluate the uncertainties in soil organic carbon transit times.We further quantified the ?_soil with radiocarbon dating approach at the region scales.The main findings were as follows:?1?In order to accurately define the ?_soil among approaches,we construsted a dataset of ?_soil.Based on our dataset,the ?_soil estimated by incubation,pool-over-flux and stable isotope are generally smaller by an order of magnitude than ?_soil estimated by radiocarbon dating.?2?The underestimations of modeled ?_soil are mainly located in cold and dry biomes,especially tundra and desert.Furthermore,we showed that one ESM?i.e.,CESM?has improved its ?_soil estimate by incorporation of the soil vertical profile.These findings indicate that the spatial variation of ?_soil is a useful benchmark for ESMs,and we recommend more observations and modeling efforts on soil carbon dynamics in regions limited by temperature and moisture.?3?Comparisons of ?_soil estimated by the observational data previously described a wide range of ?_soil.The variations within each method are attributable to difference in vegetation and climate.?4?By grouping the ?_soil into different climatic categories,we found that the observed ?_soil significantly covaried with MAT?y =-5.28 x + 156.04,r² = 0.48,P < 0.01?and MAP?y =-68.19 x + 1222.6,r² = 0.60,P < 0.01?.?5?To quantify the ?_soil and its drives at the regional scales,we collected the soil sample from five forest ecosystems in eatern China.The longest ?_soil were 3849 ± 770 years at the Changbai Mountain site,followed by Tiantong site was 1348 ± 270 years.The ?_soil were of Dongling Mountain,Badagong Mountain and Heishiding Site is relatively similar: 361 ± 72 years,436 ± 87 years and 383 ± 77 years.?6?In order to quantify the interaction of aboveground litter stock and ?_soil in a mature subtropical forest,the aboveground litter and soil were sampled in a subtropical mature broad-leave forest,eastern China.The mean aboveground litter stock was 367.5 g m-2 in the Tiantong dynamics forest plot across all the 187 grids.The relatively influence of abiotic factors?environmental and topographical factors?to litter stock was 71.4%,which is larger than biotic factors?28.6%?.For the surface layers,the ?_soil was corralted with soil organic carbon content.For the deep layers,the ?_soil was affected by soil organic carbon content and topography.Our research summarized and clarified the differences among different approaches to estimated ?_soil.Based on our dataset,we found that the ESMs significantly underestimated the ?_soil,especially in the dry and cold regions.This is mainly due to the fact that the ?_soil simulated by model was limited by temperature and moisture.Overall,this study shows the great spatial variations of ?_soil in the natural ecosystems,and we recommend more research efforts to improve its representation by ESMs in the future.