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

Ecology, June, 1999 by Stephen C. Hart

The soil is an unclassified loamy-skeletal, mixed, frigid, Typic Dystrochrept. The stand has a moderately thick (averaging [approximately]10 cm), mor-type forest floor (i.e., O horizon) and a moderately acidic surface (0-15 cm) mineral soil (p[H.sub.saturated paste] = 5.4). Selected properties of well-decayed boles and mineral soil are shown in Table 1.

Field methods

Eight plots ([approximately equal to]9 [m.sup.2]) were chosen randomly from a group of 20 spread over about a 4-ha area, where a well-decayed bole (class 4 or 5; Triska and Cromack 1980, Sollins 1982) was present alongside mineral soil. Decay-class 4 and 5 boles are those where the bark is detached or absent, branch stubs (if present) pull out easily, and little if any wood structural integrity remains leading to an elliptical cross-section. The well-decayed boles were present as ill-defined, elongate mounds ([greater than]0.50 m in width and 5 m in length) extending from the forest floor into the mineral soil, but all were recognizable from the ground surface (Sollins et al. 1987). This selection process resulted in two plots with class 4 boles and six plots with class 5 boles, which approximates roughly the relative areal abundance of each bole decay-class in this stand (M. E. Harmon, unpublished data). All identified boles within the selected plots were Douglas-fir, but three of the six selected class 5 boles could not be identified. The precise sampling location within a plot on each of two sampling dates (19 October 1990 and 4 June 1991) was determined randomly. On each sampling date, the forest floor was carefully peeled back to expose the outer surface of a well-decayed bole or the mineral soil. Two adjacent, intact bole cores and two adjacent, intact soil cores (0-0.15 m) were then removed using thin-walled polyvinyl chloride (PVC) pipe (0.05 m inner diameter x 0.20 m long) that had been sharpened at one end. Only brown-rot wood (Harmon et al. 1986) was found in the boles at this sampling depth. Bole and mineral soil sampling locations were always within 0.5 m of each other within a given plot. For each material (bole or soil), one of the paired cores (the initial core) was placed in a polyethylene bag, kept cool ([approximately]4 [degrees] C), and returned to the laboratory for analysis (within 72 h of sampling). These bole and soil samples were used for laboratory incubations and determinations of initial inorganic N pool sizes, gravimetric water content, and microbial biomass (see Laboratory' incubations and analyses). The other two cores (one bole and one soil) were used to assess net N mineralization and net nitrification rates under field conditions, using the resin-core method (Di Stefano and Gholz 1986, Binkley and Hart 1989, Hart and Firestone 1989).

Resin cores consisted of intact bole or soil cores capped at both ends by ion exchange resin (IER) bags. The purpose of the top resin bag was to capture incoming ions originating from above the bole or soil core (i.e., forest floor), while allowing water to enter freely. The bottom resin bag collected ions leached from the confined bole or soil core. The IER bags were constructed by placing 30 mL (7.8 g oven-dry equivalent) of cation anion exchange resin beads (JT Baker [Phillipsburg, New Jersey] #M-614 16-50 mesh mixed-bed IER that had been pre-extracted with 2 mol/L KCl; Hart and Binkley 1984) in nylon stockings that contained a 0.05-m diameter latex rubber tubing ring, and then were tied shut. The resulting bag fit tightly within the PVC pipe. However, to insure against solution losses through boundary flow along the inner PVC wall, silicon glue was used to seal the outside ring to the PVC tube. The resin cores were then returned to their original holes within the plot. After making sure that a solid contact was made between the bottom resin bag and the underlying bole or soil material, the overlying forest floor horizon was carefully replaced over the resin core. Bole and soil cores sampled in October 1990 were incubated until the June 1991 sampling (hereafter called the "winter incubation"). Cores sampled in June 1991 were incubated until 19 October 1991 (hereafter called the "summer incubation"). This experimental design resulted in eight replicates per material type (bole or soil) and incubation period (winter and summer).