JIA-2019-11

2606 CHEN Xu et al. Journal of Integrative Agriculture 2019, 18(11): 2605–2618 promoted by aggregate structures with larger fractions that are important in protecting various forms of C and range from the light fraction composed of plant and microbial residues to decomposed and humified decomposition products (Tisdall and Oades 2012). Soil aggregates, composed of mineral particles and binding agents, are the basic units of soil structure that physically and chemically protect SOC (Bronick and Lal 2005) from microbial decomposition. Further, they are responsible for regulating the dynamics of water, trace gases and nutrients, and also provide structure to reduce erosion, in addition to other important ecosystem traits (Six et al. 2004). A better understanding of aggregate size distribution effects on microbial community structure and activity would reveal mechanisms imparting stability to SOC. SOC content and microbial abundance and community composition are the most readily manipulated factors involved in aggregation (Six et al. 2004; Tian et al. 2016). Recent studies have found that microbial communities and their biomass were primarily influenced by soil structure by providing habitat and influencing substrate availability (Garcia-Franco et al. 2015; Li et al. 2015). Microbial biomass generally increases with increasing aggregate size (Helgason et al. 2010) due to the increased protection of labile SOC substrates (Chen et al. 2015). Aggregation and the turnover of encapsulated SOC are also strong determinants of microbial community structure (Gupta and Germida 1988). Fungi play a vital role in aggregation by excreting agents such as polysaccharides, by physically enmeshing soil particles in their hyphal networks (Tisdall 1991), and by producing hydrophobic materials that decrease the wettability of the aggregates (White et al. 2000). The role of bacteria is mostly associated with the production of polysaccharides (Lynch and Bragg 1985). Understanding the effect of soil structure on the distribution of microbial functional groups is required for a better understanding of regulatory processes, including nutrient turnover and SOCmaintenance and storage. Increasing the amount of aggregate-protected SOC also has the potential to mitigate climate change, because SOC decomposition is governed by accessibility to decomposers (Dungait et al. 2012). Breaking aggregates into smaller size classes according to the hierarchical size theory provides information on the functions of the different aggregate sizes (Tisdall and Oades 2012; Tian et al. 2015). Previous studies have separated aggregates by size and incubated the sizes individually to understand their roles in influencing microbial community composition, processes and activities (Drury et al. 2004). Most studies of soil aggregates have been mainly based on the sieving of wet or air-dried soil, and these treatments likely affect both microbial community composition and activity (Dorodnikov et al. 2009; Helgason et al. 2010). For example, complete soil wetting in a wet-sieving procedure could affect the activity of soil microorganisms by creating temporary anaerobic conditions (Zhang and Zak 1998). Prolonged sieving of air-dried soil tends to increase aggregate abrasion and fragmentation due to he shearing forces during shaking (Munkholm and Kay 2002). In this study, we compared intact aggregates to reduced aggregate size distributions in the same soils with different management treatment histories that changed SOC and aggregate size distributions. We investigated the role of aggregate size distribution on the composition and activity of microbial communities by reducing large aggregates into smaller size classes. The theory of hierarchical aggregate sizes states that large aggregates are composed of smaller aggregates (Tisdall and Oades 2012), so we designed an incubation experiment to study the effects of different aggregate sizes on microbial community composition and activity after adding various substrates to a range of Mollisol soils that were managed to provide a range of SOC and aggregate size distributions. We hypothesized that: (1) reducing macroaggregates into smaller size fractions would increase microbial biomass and enzymatic activity through the release of spatially protected SOC, and (2) conversely, soils with lower SOC and smaller aggregate size classes would respond less dramatically, primarily due to less labile SOC availability. 2. Materials and methods 2.1. Site description Soil (Pachic Haploborolls originating from loamy loessal parental material, US Soil Taxonomy) was sampled from the State Key Experimental Station of Agroecology of the ChineseAcademy of Sciences, Hailun (47°26´N, 126°38´E), Heilongjiang Province. This region has a typical temperate continental monsoon climate with a mean annual air temperature of 2.2°C. Annual precipitation averages 550mm, 65% of which occurs from June through August. The sampling site was initiated in 1985 by dividing a former agricultural field into three long-term trials of a restored grassland (GL), bare fallow (BF, no vegetation) and continuing farmland (FL). The GL is 360 m 2 in area and dominated by naturally re-vegetated Leymus chinensis L. The BF is 180 m 2 in area maintained by the frequent hand hoeing of weeds. Both the GL and BF have no fertilizer inputs or tillage and were not replicated. The FL is 720 m 2 in area and is maintained in a continuous three-year crop rotation of maize ( Zea mays L.), soybean ( Glycine max (L.) Merr.) and wheat ( Triticum aestivum L.) in a randomized complete block design with three rates of nitrogen (N)