Gross nitrogen mineralization-, immobilization-, and nitrification rates as a function of soil C/N ratio and microbial activity

Abstract

A laboratory experiment was designed to challenge the idea that the C/N ratio of forest soils may control gross N immobilization, mineralization, and nitrification rates. Soils were collected from three deciduous forests sites varying in C/N ratio between 15 and 27. They were air-dried and rewetted to induce a burst of microbial activity. The N transformation rates were calculated from an isotope dilution and enrichment procedure, in which ¹⁵NH₄Cl or Na¹⁵NO₃ was repeatedly added to the soils during 7 days of incubation. The experiments suggested that differences in gross nitrogen immobilization and mineralization rates between the soils were more related to the respiration rate and ATP content than to the C/N ratio. Peaks of respiration and ATP content were followed by high rates of mineralization and immobilization, with 1–2 days of delay. The gross immobilization of NH₄⁺ was dependent on the gross mineralization and one to two orders of magnitude larger than the gross NO₃⁻ immobilization. The gross nitrification rates were negatively related to the ATP content and the C/N ratio and greatly exceeding the net nitrification rates. Taken together, the observations suggest that leaching of nitrate from forest soils may be largely dependent on the density and activity of the microbial community.

Introduction

The low concentration of NO₃⁻ found in forest soils has often been attributed to low rates of nitrification (Vitousek et al., 1982, Gosz and White, 1986). This interpretation is supported by observations of low net nitrification rates during incubation assays of soil samples. However, occasional measurements of gross nitrification suggest a rapid turnover of a small NO₃⁻ pool in forest soils, and that the dominant fate of NO₃⁻, as well as NH₄⁺, is immobilization in the soil organic matter pool (Davidson et al., 1992, Groffman et al., 1993, Stark and Hart, 1997). In this way, immobilization may prevent nitrogen leakage to ground and surface waters.

The extent of immobilization varies between 35 and 95% from one soil to another (Mead and Pritchett, 1975, Heilman et al., 1982, Melin et al., 1983, Schimel and Firestone, 1989, Hart and Firestone, 1991, Hart et al., 1993). Nitrogen is immobilized either by abiotic or biotic processes. Fixation of NH₄⁺ to clay minerals is fast (<30 min) but usually less than 10% of the added NH₄⁺ is fixed (Drury and Beauchamp, 1991, Trehan, 1996). Most of the immobilization to the soil organic matter is biotic. Early work estimated soil microorganisms to be responsible for 10–50% of NH₄⁺ immobilization (Brookes et al., 1985) and more recent analyses by NMR of the organic matter of ¹⁵NH₄⁺ or ¹⁵NO₃⁻ treated soils found the organic ¹⁵N in microbially derived peptides and proteins (>80%), nucleic acids, and aliphatic amine groups after several months of in situ incubation (Clinton et al., 1995). The average turnover time of N in microbial biomass is estimated to 1–2 months (Davidson et al., 1992), but some of the immobilized N becomes very stable (Kelley and Stevenson, 1985), as a result of fixation by 2:1 clay minerals (van Veen et al., 1985, Breitenbeck and Paramasivam, 1995), repeated cycles of immobilization and mobilization (Jansson and Persson, 1982, He et al., 1988), and accumulation of persistent residues of microbial cells (Paul and Juma, 1981).

Despite the obvious importance of soil microorganisms in the turnover and retention of nitrogen in forest soils, the causes of variation of immobilization of N from one soil to another and between abiotic and biotic processes are not well known. Some observations suggest that the immobilization is controlled by the concentration of available C (Woodmansee and Duncan, 1980), and others that it is inversely dependent on the amount of available inorganic N (Priha and Smolander, 1995, Bengtsson and Bergwall, 2000).

Mineralization of soil N depends on a wide range of factors, including the C/N-ratio (Frankenberger and Abdelmagid, 1985), the N content (Iritani and Arnold, 1960), lignin content (Frankenberger and Abdelmagin, 1985, De Neve et al., 1994), water soluble N and cellulose content (Iritani and Arnold, 1960, Bending et al., 1998) of the litter, the light fraction organic matter of the soil (Janzen et al., 1992, Sierra, 1996), microbial respiration and ATP content (Alef et al., 1988), microbial biomass (Dalal and Meyer, 1987), and microbial N content (Fisk and Schmidt, 1995). The diversity of factors correlated with N mineralization reflects the variation in substrates and microbial communities being used and temporal changes in substrate quality.

Models predicting nitrogen turnover and retention in soils often include the C/N ratio as an important factor in determining the rate of mineralization, immobilization and nitrification (van Veen et al., 1984, Aber, 1992, Bradbury et al., 1993, Janssen, 1996). The idea to use the soil C/N ratio to predict variations in N mineralization and immobilization rates among soils is based on the fact that heterotrophic soil bacteria usually have a lower C/N ratio than the soil they inhabit. If it is assumed that the cells have a C/N ratio of 10 and respire about 50% of their C uptake, they may be N limited above a soil C/N ratio of 20 and C limited below (Tate, 1995). Soils with a high C/N ratio may then be characterized by rapid immobilization of N and soils with a low C/N ratio by slower N immobilization and a surplus of available NH₄⁺, derived from deamination of organic carbon sources.

The evidence for a correlation between the soil C/N ratio and nitrogen immobilization and mineralization is equivocal, at least partly because early work was confined to measurements of net rates of N transformation. Within-site variations of the C/N ratios of different pools of organic matter and between-site variations of the C and N assimilation efficiency of the microbial biomass may also confound the usefulness of the soil C/N ratio to predict gross nitrogen transformation rates. Therefore, we designed a ¹⁵N pool dilution experiment to simultaneously measure gross immobilization, mineralization, and nitrification, in an effort to challenge the hypotheses of a positive relationship between the C/N ratio and N immobilization and a negative relationship between the C/N ratio and N mineralization and nitrification in three forest soils. Even if the soil C/N ratio may prove useful to predict site-to-site variations of the N transformations, temporal variation of soil temperature and moisture will trigger dynamics of microbial biomass and activity that influence N transformations more than the spatial variation of the C/N ratio. Therefore, we developed the sub-hypothesis that the variation in N transformation rates is controlled by the variations in microbial biomass and activity—a high biomass and activity would correspond (1) to high rates of N immobilization and mineralization, and, as nitrifiers tend to be competitively inferior to heterotrophs (Verhagen and Laanbroek, 1991, Verhagen et al., 1995), (2) to low nitrification rates.

It is well established from laboratory experiments that soil drying and rewetting may bring about a flush of C and N mineralization (Birch, 1958, Agarwal et al., 1971, van Gestel et al., 1991) and an increase in microbial numbers and activity (West et al., 1986). The origin of the mineralized N may be both biomass lyzed by the drying, and non-biomass organic matter that becomes more accessible to microbial attack after drying and rewetting (Jager and Bruins, 1975, Marumoto et al., 1982). We used this evidence to test the sub-hypothesis and induced a microbial growth phase by drying and rewetting the three soils, and then evaluated the subsequent gross N transformations rates. Although short-time laboratory ¹⁵N pool dilution experiments do not necessarily reflect N transformation rates in the field, they facilitate experimental manipulations and help explaining mechanisms causing variation in N transformation rates from one soil to another.