Understanding the impact of biological activities on the soil phosphorus (P) distribution under long-term fertilizer application can facilitate better soil P fertility management.  Therefore, the primary objectives of this study were to investigate the effect of long-term (since 1981) fertilizer application on the soil P fractions and microbial community and to evaluate correlations between the microbial community structure and P distribution.  The following treatments were implemented in a long-term field trial: no fertilization (CK), inorganic N and K (NK), inorganic P and K (PK), inorganic N, P and K (NPK) and manure+NPK (MNPK) fertilization.  The study showed that the soil pH, soil organic carbon and total and available N and P concentrations were considerably higher in the MNPK treatment than in the CK treatment.  The soil microbial biomass C, N and P concentrations were also significantly higher in the MNPK treatment than in the CK treatment.  Among fertilization treatments, the β-1,4-glucosidase, α-1,4-glucosidase, urease, acid phosphatase and phosphodiesterase activities were the highest in the MNPK treatment.  Compared to inorganic fertilization, the MNPK treatment increased the labile soil P fractions and decreased the residual soil P concentration.  Continuous fertilization significantly affected the soil microbial composition.  The total phospholipid fatty acid (PLFA) concentrations in the NK, PK, NPK and MNPK treatments were 23.3, 43.1, 48.7 and 87.7% higher, respectively, than in the CK treatment.  A significant correlation was observed between the microbial community and soil P fractions.  Moreover, the aggregated boosted tree (ABT) model showed that among the various soil biochemical properties, the total PLFA concentration was the factor that most influenced the active P pool, accounting for 35.4% of the relative influence of all soil biochemical properties examined.  These findings reveal that combined manure and inorganic fertilizer application is a better approach than applying inorganic fertilizer alone for sustaining long-term P fertility by mediating soil biological activity.

There is limited information on carbon sequestration efficiency (CSE) of soil aggregates in upland and paddy soils under long-term fertilization regimes.  In a red soil region of southern China, an upland soil experiment started in 1986 and a paddy soil experiment commenced in 1981.  These experiments were conducted using different fertilization treatments.  After 30 years, soil organic carbon (SOC) content and stock of different aggregate components were analyzed.  The results showed that the SOC contents and stocks in upland soil were lower than in paddy soil.  In both upland and paddy soils, the SOC contents and stocks of all aggregate components in NPKM (combined treatment with chemical nitrogen (N), phosphorus (P), potassium (K) fertilizers and manure) were the highest among all treatments.  Compared with CK (no fertilizer), SOC content of all aggregate components in NPKM was increased by 13.21–63.11% and 19.13–73.33% in upland and paddy soils, respectively.  Meanwhile, the change rates in SOC stock of all aggregate components in upland soil were lower than in paddy soil, although the change rate of SOC stock of all aggregate components in NPKM was higher than in other treatments.  Furthermore, a linear equation could fit the relationships between carbon (C) input and change rate of SOC stock (P<0.05).  Results indicated that the sum of CSE from all aggregate components in upland soil (16.02%) was higher than that of paddy soil (15.12%) in the same climatic condition and from the same parent material.  However, the CSEs from all aggregates were higher than that of bulk soil, although the result from bulk soil also showed that the CSE of upland soil was higher than that of paddy soil.