Nitrogen transformations in fallen tree boles and mineral soil of an old-growth forest

Ecology, June, 1999 by Stephen C. Hart

Statistical analyses

Because some of the variables exhibited nonnormality, I used the Wilcoxon signed-rank test (a nonparametric equivalent of the paired t test) to determine significant differences in soil properties and processes for a given sampling date or incubation period between bole and mineral soil samples. All statistical analyses were performed using SigmaStat V. 2 software (Jandel Scientific, San Rafael, California, USA) at the P = 0.05 significance level.

RESULTS

Well-decayed boles had higher total C and N concentrations than adjacent surface mineral soil (Table 1). Even for this relatively C-rich mineral soil, well-decayed boles had a C concentration over six times that of the mineral soil (Table 1). Nitrogen concentrations were also higher in boles than in mineral soil, although these differences were much smaller than was the case for total C. The resulting C:N ratios of the materials were 117 and 26 for bole and mineral soil, respectively. Because of the higher C (i.e., organic matter content) of well-decayed boles compared to mineral soil, boles had a much lower bulk density and a corresponding higher gravimetric water content at the onset of both field incubation periods (Table 1). On a mass basis, well-decayed boles contained 4 to 5 times as much water as mineral soil.

Expressed on a per unit area of material basis (or volume to a depth of 0.15 m), net rates of N mineralization over either incubation period (winter or summer) or over the entire year (sum of the two periods) were similar for well-decayed boles and mineral soil [ILLUSTRATION FOR FIGURE 1 OMITTED]. For both materials, net N mineralization rates were much higher over the summer compared to the winter [ILLUSTRATION FOR FIGURE 1 OMITTED]. Net nitrification rates in both materials were low, but annual net nitrification rates were significantly higher in mineral soils than in boles (19 and 6 mg N [multiplied by] [m.sup.-2] [multiplied by] [yr.sup.-1], respectively; data not shown).

However, a different pattern emerges from these results when rates are expressed as per mass of total N to assess the quality of the total N as a substrate for net mineralization (see Powers 1990). On a per-unit total N basis, net N mineralization rates were significantly higher in well-decayed boles compared to mineral [TABULAR DATA FOR TABLE 2 OMITTED] soil during the summer incubation period [ILLUSTRATION FOR FIGURE 2 OMITTED] and over the entire 1-y period (data not shown); winter rates were still similar [ILLUSTRATION FOR FIGURE 2 OMITTED]. Over the 1-y period, 1.0% of the total N in the well-decayed bole was mineralized on a net basis, compared to only [approximately]0.4% of the mineral soil total N.

Laboratory estimates of available C (i.e., microbial biomass C, microbial respiration, specific respiration rate) and N (i.e., microbial biomass N, anaerobically mineralizable N, aerobic net N transformations) were significantly higher for well-decayed boles compared to adjacent mineral soil when values were expressed on a mass basis (Table 2). Exceptions to this pattern occurred for net N transformation rates during aerobic laboratory incubation in samples collected in June 1991, where there was no significant difference in rates between well-decayed boles and mineral soils (Table 2). Additionally, microbial biomass N was statistically similar in well-decayed boles and mineral soil materials collected in October 1990 (Table 2).