Non-additive Effects of Nitrogen and Phosphorus Fertilization on Microbial Biomass and Residue Distribution in a Subtropical Plantation

High nitrogen (N) and phosphorus (P) availability has significant influence on microbial-driven soil carbon (C) sequestration. Microbial residues are a significant contributor of soil stable C pool, their distribution among aggregate fractions determines long-term soil C stability. However, very little is known about the interactive effects of N and P fertilization on soil microbes, especially their residues, at aggregate scale in plantation ecosystems. Since 2012, a field-manipulated experiment with N (200 kg N ha(-1) year(-1)) and/or P fertilization (50 kg P ha(-1) year(-1)) has been conducted to examine their interactive effects on microbial community and residues in bulk soil and three soil aggregate fractions: large macroaggregates (> 2 mm, LMA), small macroaggregates (0.25-2 mm, SMA), and microaggregates (< 0.25 mm, MA) in a subtropical Chinese fir (Cunninghamia lanceolata) plantation. Results showed that N and P fertilization, either individually or in combination, decreased microbial biomass of bulk soils to a similar extent (by up to 37.0%). This reduction was due to the decreased bacterial biomass in SMA and MA and fungi in LMA. By contrast, adding N and P fertilizer together (NP) significantly stimulated fungal residues in SMA and further redistributed microbial residues from LMA to SMA, although single fertilization had no effects on microbial residues or their distribution. Changes in root biomass moderated the direct effects of fertilization on aggregate-associated microbial groups and the indirect effects of NP fertilization on microbial residue distribution. Together, our results provide new insights into the microbial mechanisms through which multiple fertilization control soil C persistence in subtropical plantation. These findings highlight that separating bulk soil into distinct aggregate fractions and considering the interactive effect of N and P fertilization are needed to predict the soil C dynamics under fertilization.