| As an important part of terrestrial ecosystems, cropland plays a large role in the terrestrial carboncycle. On the one hand, cropland has a large potential to sequestrate carbon through photosynthesis inthe cropping system. On the other hand, cropland may release a large amount of carbon in associationwith cultivation for crop production. Hence, how to maintain or increase the cropland carbon pool hasattracted much attention.This study is based on four long-term experimental sites in China: three in the temperate zone andone in the sub-tropical zone. Both statistical analysis and CENTURY model are chosen to complete theresearch. We use statistical analysis to examine:(i) responses of soil C/N ratio to various fertilizationsacross the sites;(ii) the relationship between soil carbon change and carbon inputs. We also integratelong-term experimental data with CENTURY model simulation to find out soil carbon sequestrationpotential for different area in China. The main findings are as following:1. Soil organic carbon (SOC) and total nitrogen (TN) had different responses to the treatments.There was an increasing trend in SOC, even without fertilizer. However, applying inorganic fertilizersonly (NPK) did not maintain TN contents at some sites. The NPKM treatment resulted in a largeincrease in both SOC (35-147%) and TN (33-10%) contents, relative to the initial values.2. The soil C:N ratio shows a significant increase over time at the sub-tropical site but little changeat the three temperate sites. Our analysis shows similar C:N ratios (37-38) in gross input of organicmaterials under the NPK treatments. However, the estimated C:N ratio during decomposition was muchsmaller at the sub-tropical site (23.7) than at the three temperate sites (44.0-48.2) under the NPKtreatments, which may explain the increased soil C:N ratio at the sub-tropical site. Thus, we concludethat variations in soil C:N ratio are not caused by organic matter inputs but by decomposition in thewheat-corn double cropping systems.3. Different amounts of balanced fertilization show little impact on the C inputs derived by plants,reaching to3.5Mg C ha-1yr-1. The SOC change rate is much higher under the manure application thantreatments with chemical fertilizers only. Statistical analysis shows that the linear and non-linearequations perform equally well (p<0.01) within the experimental data interval. But the non-linearequation is more suitable for specific purpose. Using the non-linear equation, we can predict that minimum C input to maintain the current SOC level would be0.03-1.32Mg C ha-1yr-1at the four sites.The chemical nitrogen and phosphate fertilization yield sufficient carbon biomass inputs to maintain thecurrent SOC levels. However, to increase SOC at1Mg C ha-1yr-1, soils need over10Mg C ha-1yr-1atmost sites. Our results suggest that the increment of SOC stocks would be mainly related to theadditional carbon inputs for the long-term perspectives.4. The CENTURY model (version4.5) can simulate fertilization effects on SOC change in differentclimate conditions and soil properties (n-RMSE<15%). After running the CENTURY over the period of1990-2100, the SOC levels are supposed to increase to31.6-54.7Mg ha-1across the sites. With thecomparison of SOC stocks in1990and2100, we estimate that the carbon sequestration potential wouldbe9.2-38.2Mg ha-1under the current high manure application (hNPKM). Analysis of organic carbon ineach carbon pool indicates that long-term fertilization enhances the slow pool proportion but decreasethe passive pool proportion. Our results suggest that the change in slow carbon pool determines theSOC dynamics under long-term fertilization.In summary, soil C:N ratio change is mainly influenced by the decomposition of soil organic matter.Long-term fertilizations would affect the proportion of slow carbon pool, and thus the soil oraganiccarbon pool. Applying additional carbon inputs is the most effective way for enhancing soil organiccarbon level.
|
PDF Full Text Request