Effects of fertilizer type and rate on labile soil fractions of a sandy Cambisol-long-term and short-term dynamics

The application of density fractionation is an established technique, but studies on short-term dynamics of labile soil fractions are scarce. Objectives were (1) to quantify the long-term and short-term dynamics of soil C and N in light fraction (LFOC, LFON, p <= 2.0 g cm(-3)) and microbial biomass C (C-mic) in a sandy Cambisol as affected by 28 y of different fertilization and (2) to determine the incorporation of C-4-C into these labile fractions during one growing season of amaranth. The treatments were: straw incorporation plus application of mineral fertilizer (MSI) and application of farmyard manure (FYM) each at high (MSIH, FYMH, 140-150 kg N ha(-1) y(-1)) and low (MSIL, FYML, 50-60 kg N ha(-1) y(-1)) rates at four field replicates. For all three sampling dates in 2008 (March, May, and September), stocks of LFOC, LFON and Cmic decreased in the order FYMH > FYML > MSIH, MSIL. However, statistical significance varied markedly among the sampling dates, e.g., with LFOC being significantly different (p <= 0.05) in the order given above (sampling date in March), significantly different depending on the fertilizer type (May), or nonsignificant (September). The high proportion of LFOC on the stocks of soil organic C (45% to 55%) indicated the low capacity of soil-organic-matter stabilization on mineral surfaces in the sandy Cambisol. The incorporation of C-4-C in the LFOC during one growing season of amaranth was small in all four treatments with C-4-LFOC ranging from 2.1% to 3.0% of total LFOC in March 2009, and apparent turnover times of C-3-derived LFOC ranged from 21 to 32 y for the sandy soils studied. Overall, our study indicates that stocks of LFOC and LFON in a sandy arable soil are temporarily too variable to obtain robust significant treatment effects of fertilizer type and rate at common agricultural practices within a season, despite the use of bulked six individual cores per plot, a common number of field replicates of four, and a length of treatments (28 y) in the order of the turnover time (21-32 y) of C-3-derived LFOC.