Root effects on soil carbon and nitrogen cycling in a Pinus radiata D. Don plantation on a coastal sand.
Many factors interact to control rates of soil carbon (C) and nitrogen (N) cycling in terrestrial ecosystems (e.g. Kladivko and Keeney 1987; Paul and Clark 1996). Several studies have examined the impact of above-ground inputs on soil N cycling, particularly N mineralisation (see Hobbie 1992, and Scott and Binkley 1997, for reviews of relevant literature). Likewise, considerable attention has been given to the impact of litter quality on decomposition processes and C cycling (Cadisch and Miller 1997). However, in forest ecosystems, very few studies have examined the importance of below-ground inputs on soil C and N cycling processes. More recent work on C budgets in forests has attempted to quantify the contribution of roots to soil respiration (reviewed in Hanson et al. 2000). The impact of below-ground inputs on soil C and N pools and turnover, however, has received detailed attention in only a few studies using trenched plots to isolate blocks of soil from receiving new below-ground inputs (e.g. Fisher and Gosz 1986; Hart and Sollins 1998).
In trenched plots, C inputs to the soil via live roots are curtailed; this can influence other biochemical processes and nutrient pools (Vitousek et al. 1979; Silver and Vogt 1993; Hart and Sollins 1998). Nitrogen metabolism has been most widely investigated with measurements made of inorganic N pools (Vitousek et al. 1979; Fisher and Gosz 1986; Silver and Vogt 1993; Simard et al. 1997), and net N mineralisation under anaerobic (Hope and Li 1997; Simard et al. 1997) and aerobic, and in situ conditions (Hart and Sollins 1998). In comparison, changes in microbial biomass and interactions between C and N cycling following trenching have received little attention (Hart and Sollins 1998).
We report here on the effects of trenching on soil biochemical properties in a Pinus radiata plantation in New Zealand on a coastal sand. We measured total, extractable, and microbial C and N pools, [CO.sub.2]-C production, and net and gross N mineralisation and nitrification under standardised conditions, and also in situ mineral N and inorganic P production. We hypothesised that removal of below-ground inputs would decrease C availability and so influence rates of soil N cycling.
Materials and methods
Site and soil
Himatangi Forest is located at 40 [degrees] 22'S and 175 [degrees] 16'E on recent sand dunes about 30 km west of Palmerston North, New Zealand. Mean annual temperature is about 13 [degrees] C and mean annual rainfall about 900 mm. The soil is Hokio sand (Cowie and Smith 1958) and is classified as an Aquic Udipsamment.
The site was under marram grass (Ammophila arenaria L.) and lupins (Lupinus arboreus Sims) from 1950 to 1970, and was then planted to P. radiata. Most of the stand was clear-felled in 1993, but an area of c. 1.5 ha (c. 150 by 100 m) was left unharvested as a windbreak for subsequent second-rotation plantings (Parfitt et al. 1998). At the time of our first sampling in the unharvested area in 1996, the trees were 26 years old and at a stand density of about 300 stems/ha. The site was almost flat and had no understorey. Forest floor depth, estimated from 150 measurements, averaged 1.5 cm (s.e.m. 0.05) for L material and 3.6 cm (s.e.m. 0.15) for FH material.
Three replicate plots (c. 4 by 4 m) for trenching, and 3 similarly sized control plots, were selected between trees in the middle of the plantation to avoid, as much as possible, any edge effects. The plots were at least 1 m from the nearest tree and were about 15-30 m from each other. The layout of the plots was partly governed by the tree spacings and had features of both randomised block and systematic designs (Hurlbert 1984). Trenching was carried out on 19 June 1996 by digging a trench around each plot to about 1 m depth, which roughly corresponded with the current ground-water level. The trench was lined with a double thickness of heavy-duty polythene sheets, back-filled, and compacted with a light tractor. Care was taken not to disturb the area bounded by each trench. The few herbs present were removed; all plots, including the controls, were maintained weed-free by hand-weeding throughout the 27-month study. Above-ground litterfall, however, was not excluded, and averaged 680 g dry wt (at 65 [degrees] C)/[m.sup.2] [multiplied by] year between September 1997 and 1998, with needles contributing 81% of the total (N. A. Scott, N. J. Rodda, J. A. Townsend unpubl. data).
Samples of L and FH material and 0-10 cm and 10-20 cm of mineral soil were first taken on 20 June 1996 in 2 diagonals across each plot; possible disturbance effects were avoided by sampling at least 50 cm from the plot edges. L and FH materials were collected by hand at 9 locations and mineral soil was sampled as 2.5-cm-diameter cores at the same locations. The 9 samples were pooled to give one sample for each depth per plot. The second set of samples was taken similarly from the same plots 27 months later on 8 September 1998.
L material was cut into pieces 2-4 mm, with twigs [is greater than] 2 mm diameter being discarded. The other samples were sieved ([is less than] 5.6 mm) and stored at about 4 [degrees] C, except for a portion that was air-dried and ground in a Tema mill for total C and N analyses.
Results are expressed on the basis of oven-dry (105 [degrees] C) weight of material, unless otherwise stated. Biochemical measurements on the field-moist samples were routinely made in duplicate and commenced within 1 week of collection. Samples for a [sup.15]N experiment to determine gross N cycling rates were, however, stored for 12 weeks at about 4 [degrees] C before it could be commenced.
Soil moisture and pH (in water) were determined according to Blakemore et al. (1987), and total C and N by combustion in a Leco FP 2000 analyzer. Water-holding capacity (WHC) was measured according to Harding and Ross (1964).
Microbial C and N, and extractable C and N
Microbial C and N were determined by fumigation-extraction methods (Ross and Sparling 1993), with samples adjusted to 60% of WHC (about -5 kPa), using 2.5 g of L material and 10.0 g of the other samples. After the 24 h fumigation with CH[Cl.sub.3], all samples were extracted with 25 mL 0.5 M [K.sub.2][SO.sub.4] on an end-over-end shaker for 30 min. Extractable C in the fumigated and non-fumigated samples was determined with a Shimadzu TOC-5000 total C analyzer, and extractable total N by persulfate oxidation (Ross 1992). Extractable C and extractable N flush were calculated as the difference between C and N concentrations in the fumigated and unfumigated samples. A [k.sub.EC]-factor of 0.34 was used for converting extractable C flush to microbial C; this was based on the mean [k.sub.EC]-factor calculated for comparable samples from another coastal sand (Ross and Sparling 1993), but adjusted for the higher extractable C values obtained with the TOC analyser than with the original dichromate oxidation method. A [k.sub.EN]-factor of 0.45 (Jenkinson 1988) was used to calculate microbial N from the extractable N flush values. Extractable organic N was calculated as the difference between extractable total N and mineral-N determined as described below in the 2 M KCl extract of a non-incubated (Day 0) sample.
Samples for [CO.sub.2]-C production by L and FH material were weighed into 125-mL polypropylene containers and adjusted with water to give 2.5 g material at 60% of WHC. Incubation was at 25 [degrees] C with the polypropylene containers enclosed in 1-L sealed glass jars fitted with a septum, and that contained 10 mL water to maintain humidity. [CO.sub.2]-C produced after 7 and 14 days was measured by sampling the jar head-space and injecting a 1-mL sample into a Carle 8700 gas chromatograph with a thermal conductivity detector. [CO.sub.2]-C production by mineral soil was determined by incubating 20.0 g of sample at 60% of WHC at 25 [degrees] C in approximately 250-mL biometer flasks (Bartha and Pramer 1965; Ross and Sparling 1993).
Additional measurements were made with mineral soil and FH samples incubated for 84 days at 25 [degrees] C in 1-L sealed jars in conjunction with the measurements of gross N mineralisation and nitrification (see below); sample weights and water contents were similar to those described above. Head-space [CO.sub.2] concentrations were measured by gas chromatography at 14, 28, 57, 70, and 84 days; the jars were vented after each measurement.
Net N mineralisation was determined at 25 [degrees] C with the above weights of samples incubated in 125-mL containers, as described by Scott et al. (1998). Mineral N (min-N = [NH.sub.4][sup.+]-N + [NO.sub.3][sup.-]-N), produced after 0, 14, and 56 days incubation, was extracted with 50 mL 2 M KCl and measured by autoanalyzer procedures (Blakemore et al. 1987).
Isotopic dilution with [sup.15][NH.sub.4][sup.+]-N and [sup.15][NO.sub.3][sup.-]-N amendments (Brooks et al. 1989; Hart et al. 1994b) was used for determining gross N mineralisation and nitrification in FH material and 0-10 cm depth mineral soil. Nitrogen cycling rates over a period of 24 h at 25 [degrees] C were measured before and after incubation of the samples for 57 days at 25 [degrees] C. The procedure used is described by Scott et al. (1998). The amount of labelled [sup.15][NH.sub.4][sup.+]-N removed from the extractable pool within 15 min of addition was also determined (Hart et al. 1994b). Nitrogen transformation rates were calculated using the isotope-dilution equations given in Kirkham and Bartholomew (1954).
In situ net N and P mineralisation was assessed in each plot with triplicate resin bags (Binkley and Matson 1983) buried at about 3-5 cm depth of mineral soil and incubated for 593 days between January 1997 and September 1998. Each bag contained 5 g of `Duolite' (MB 6113) mixed resin (BDH, Ltd). Extraction of adsorbed mineral N and inorganic P was with 100 mL 2 M KCl. [NH.sub.4][sup.+]-N and [NO.sub.3][sup.-]-N were again determined by autoanalyzer procedures and inorganic P was determined with the Murphy and Riley (1962) reagent (Blakemore et al. 1987).
Data reported are means and standard errors of the means (s.e.m.) of values from the 3 control and 3 trenched plots; each value was the average of the duplicate laboratory replicates. Carbon and N pools and cycling rates in (a) control and trenched samples taken in 1996, (b) control and trenched samples taken in 1998, and (c) control samples taken in 1996 and 1998 were compared by t-tests (SYSTAT 1996), with each depth of sample being treated separately. Before analysis, transformations of the following properties were applied to improve normality: logarithms for in situ mineral-N and mineral-P production, and [sin.sup.-1] ([square root]) for percentages of mineral N as [[NO.sub.3].sup.-]-N. Pearson correlations were used to determine relationships between [CO.sub.2]-C production, extractable C, and microbial C. Treatment differences in [CO.sub.2]-C production over the 14-84-day period were examined by repeated measures within the General Linear Models procedure (SYSTAT 1996). The degrees of significance between treatments are indicated in the Tables and Figure. In the text, differences were considered significant at P = 0.05, unless stated otherwise.
In the initial 1996 samples, only two of the differences between control and trenched plots were significant (data not shown). These occurred in L material, with total C concentration being higher (P [is less than] 0.001) in the control than in the trenched samples (542 and 518 g/kg, respectively) and net N mineralisation (0-14 days) being lower (P [is less than] 0.05) in the control than in the trenched samples (-11.1 and 5.4 mg/kg, respectively).
Differences between the 1996 and 1998 sets of samples from the control plots were also non-significant (P [is greater than] 0.10; data not shown). The only exception was with mineral N in non-incubated mineral soil. Mineral-N values at 0-10 cm depth were 0.43 and 1.73 mg/kg in the 1996 and 1998 samples (P [is less than] 0.10), respectively, and at 10-20 cm depth were 0.07 and 2.04 mg/kg (P [is less than] 0.01), respectively.
Moisture, pH, and C and N pools
As might be expected in soil without actively growing roots, moisture concentrations were higher in the trenched than in the control samples (Table 1); the only exception was L material, in which roots did not occur. Trenching resulted in a slight decrease in pH values, with the changes being weakly significant (P [is less than] 0.10) in 0-10 cm depth soil only (Table 1).
Table 1. Moisture concentrations, pH, and total and extractable C and N concentrations of soil from the control and trenched (for 27 months) plots Values are means with s.e.m. in parentheses Plot Moisture pH Total C Treatment (g/kg) (g/kg) L material Control 836 (11) 5.8 (0.2) 537 (1) Trenched 750 (41) 5.6 (0.04) 538 (1) FH material Control 2160 (99) 5.1 (0.1) 413 (28) Trenched 2520 (125)([dagger]) 4.9 (0.1) 370 (17) 0-10 cm Control 76 (6) 6.3 (0.1)([dagger]) 8.6 (1.0) Trenched 111 (1)(*) 6.1 (0.1) 7.4 (0.2) 10-20 cm Control 51 (0.3) 6.8 (0.1) 3.6 (0.3) Trenched 85 (2)(**) 6.7 (0.04) 3.4 (0.1) Plot Total N C/N Extractrable C Extractable Treatment (g/kg) (mg/kg) organic N (mg/kg) L material Control 9.8 (0.2) 54 (1) 1170 (127) 81 (4) Trenched 9.2 (0.2) 58 (2) 1110 (59) 94 (22) FH material Control 12.3 (0.6) 33 (1) 477 (28)(*) 38 (15) Trenched 11.0 (0.3) 33 (1) 332 (8) 49 (3) 0-10 cm Control 0.49 (0.09) 18 (l) 42 (9) 2.3 (0.8) Trenched 0.38 (0.01) 19 (1) 21 (2) 1.6 (0.7) 10-20 cm Control 0.20 (0.02) 18 (1) 25 (5) 2.0 (0.1)(*) Trenched 0.18 (0.01) 19 (1) 15 (1) 1.4 (0.1) ([dagger]) P < 0.10, (*) P < 0.05, (**) P < 0.01, for the difference between control and trenched plots.
Total organic C and N concentrations and C/N ratios did not differ between treatments (Table 1). Extractable C and microbial C and N concentrations tended to be lower in the trenched than in the control samples, with some of the differences being significant (Tables 1, 2). Treatment differences in extractable organic N were less consistent (Table 1). Ratios of microbial C/N, microbial C/total C, and microbial N/total N also tended to be lower in the trenched than control samples (Table 2).
Table 2. Microbial C and N pools and ratios of soil from the control and trenched (for 27 months) plots Values are means with s.e.m. in parentheses Plot Microbial C Microbial N Microbial C/ Treatment (mg/kg) (mg/kg) microbial N L material Control 18700 (756) 1630 (79) 11.5 (0.3) Trenched 17900 (460) 1590 (12) 11.2 (0.3) FH material Control 5090 (700) 613 (75) 8.3 (0.4) Trenched 3260 (141) 427 (20) 7.5 (0.1) 0-10 cm Control 139 (9)(**) 30 (4)([dagger]) 4.8 (0.4) Trenched 78 (5) 19 (2) 4.1 (0.5) 10-20 cm Control 77 (6)(*) 13 (3) 7.0 (1.6) Trenched 48 (7) l0 (1) 5.7 (1.3) Plot Microbial C (x 100)/ Microbial C (x 100)/ Treatment total C total N L material Control 3.5 (0.1) 19.0 (0.4) Trenched 3.3 (0.1) 19.4 (0.8) FH material Control 1.3 (0.2) 5.0 (0.6) Trenched 0.9 (0.1) 4.0 (0.2) 0-10 cm Control 1.6 (0.2)(*) 6.4 (0.9) Trenched 1.0 (0.1) 5.1 (0.5) 10-20 cm Control 2.2 (0.3)([dagger]) 6.3 (1.0) Trenched 1.4 (0.2) 5.7 (0.6) ([dagger]) P < 0.10, (*) P < 0.05, (**) P < 0.01, for the difference between control and trenched plots.
Soil [CO.sub.2]-C production similarly tended to be lower in samples from the trenched than from the control plots, with the difference in 0-10 cm mineral soil being significant (Table 3). In the measurements made in conjunction with the [sup.15]N experiment, [CO.sub.2]-C production values were likewise lower in the trenched than control treatment (Fig. 1 a, b). Although t-tests indicated that treatment differences were only weakly significant (P [is less than] 0.10) or non-significant at each of the incubation periods, the repeated measures analysis of the 14-84 day data showed that the treatment differences were significant (P [is less than] 0.05) for both FH material and 0-10 cm depth mineral soil (Fig. 1 a, b).
Table 3. [CO.sub.2]-C production and metabolic quotients (q[CO.sub.2]) of soil from the control and trenched (for 27 months) plots Values are means with s.e.m. in parentheses Plot treatment [CO.sub.2]-C (mg/kg [multiplied by] h) 0-7 days 7-14 days L material Control 107 (4) 68 (4) Trenched 99 (6) 60 (4) FH material Control 20 (4) 18 (3) Trenched 15 (1) 12 (1) 0-10 cm Control 0.49 (0.04)(*) 0.29 (0.03)([dagger]) Trenched 0.28 (0.01) 0.21 (0.01) 10-20 cm Control 0.22 (0.03) 0.093 (0.017) Trenched 0.14 (0.02) 0.088 (0.017) Plot treatment q[CO.sub.2] (mg [CO.sub.2]-C/ g microbial C [multiplied by] h)(A) L material Control 5.7 (0.1) Trenched 5.5 (0.2) FH material Control 3.9 (0.2) Trenched 4.5 (0.1)([dagger]) 0-10 cm Control 3.5 (0.2) Trenched 3.7 (0.1) 10-20 cm Control 2.8 (0.2) Trenched 3.0 (0.5) ([dagger]) P < 0.10, (*) P < 0.05, for the difference between control and trenched plots. (A) [CO.sub.2]-C production calculated over 0-7 days.
Treatment differences in [CO.sub.2]-C production paralleled those in microbial C concentrations and so resulted in similar q[CO.sub.2] values in the trenched and control samples, except in FH material (Table 3). Rates of [CO.sub.2]-C production correlated strongly with microbial C in all samples, and less strongly with extractable C (Table 4). Microbial C and extractable C were also correlated significantly in the L, FH, and 0-10 cm mineral soil layers (Table 4).
Table 4. Pearson correlation coefficients (r) of soil C pools and [CO.sub.2]-C production Sample Extractable C Microbial C Microbial C L material 0.909(*) FH material 0.887(*) 0-10 cm 0.866(*) 10-20 cm 0.493 [C0.sub.2]-C (0-7 days) L material 0.696 0.907(*) FH material 0.699 0.947(**) 0-10 cm 0.816(*) 0.953(**) 10-20 cm 0.599 0.821(*) [C0.sub.2]-C (7-14 days) L material 0.579 0.819(*) FH material 0.736([dagger]) 0.965(**) 0-10 cm 0.819(*) 0.912(*) 10-20 cm 0.251 0.368 ([dagger]) P < 0.10, (*) P < 0.05, (**) P < 0.01 (d.f. = 4).
Net N mineralisation and nitrification
In FH material and mineral soil, the pool of mineral N in non-incubated (0 day) samples tended to be higher in the trenched than control samples (Table 5). Net N mineralisation over 0-14 days was significantly higher in trenched than control FH material (Table 5). In samples that had been incubated for 56 days, however, total mineral N was similar in the trenched and control treatments at all depths.
Table 5. Mineral N (min-N) concentrations, net mineral N production, and percentage of mineral N as nitrate-N ([NO.sub.3][sup.-]-N) of soil from the control and trenched (for 27 months) plots Values are means with s.e.m. in parentheses Net min-N production Plot Min-N (mg/kg) Min-N treatment (mg/kg) (mg/kg) 0 days 0-14 days 14-56 days 56 days L material Control 23 (0.2) -3.7 (2.7) 10 (3)([dagger]) 29 (5) Trenched 23 (1) -1.1 (2.4) -1.9 (2) 20 (3) FH material Control 25 (3) 55 (3) 371 (57) 452 (59) Trenched 54 (4)(**) 130 (16)(*) 289 (44) 471 (34) 0-10 cm Control 1.7 (0.4) 5.2 (0.6) 12 (3) 19 (4) Trenched 2.4 (0.1) 4.7 (0.2) 9 (1) 16 (2) 10-20 cm Control 0.7 (0.3) 2.9 (0.4) 5.6 (0.3)(*) 9.1 (0.5) Trenched 1.6 (0.2)(*) 2.9 (0.3) 4.2 (0.3) 8.6 (0.8) Plot [NO.sub.3][sup.-]-N (% of min-N) treatment 0 days 14 days 56 days L material Control 2 (1) 8 (1) 16 (1) Trenched 7 (0.3)(**) 6 (1) 18 (6) FH material Control 4 (1) 3 (1) 37 (6) Trenched 39 (4)(***) 89 (5)(**) 98 (0.2)(**) 0-10 cm Control 9 (1) 4 (1) 69 (7) Trenched 53 (3)(***) 38 (4)(**) 99 (0.2)(*) 10-20 cm Control 21 (2) 12 (1) 99 (0.3) Trenched 55 (4)(**) 36 (2)(***) 99 (1) ([dagger]) P < 0.10, (*) P < 0.05, (**) P < 0.01, (***) P < 0.001, for the difference between control and trenched plots.
Trenching stimulated net nitrification throughout the entire 56-day incubation in the FH and 0-10 cm layers (Table 5). After an incubation period of 14 days, [NO.sub.3][sup.-]-N concentrations (as a percentage of mineral-N) were about 30-fold higher in FH material and 10-fold higher in 0-10 cm mineral soil (P [is less than] 0.01) in the trenched than control samples; corresponding values after 56 days were only 2.6- and 1.4-fold higher, respectively (Table 5), mainly because of the large increase in [NO.sub.3][sup.-]-N production in the control samples between 14 and 56 days.
The effects of trenching on soil N dynamics were even greater during the in situ incubation. Total min-N production, as assessed using resin bags, was about 40-fold higher (P [is less than] 0.05), and [NO.sub.3][sup.-]-N production over 200-fold higher (P [is less than] 0.01), in the trenched than in the control plots (Table 6). In situ inorganic P production did not differ between the trenched and control plots (Table 6).
Table 6. Production of mineral N and mineral P, as assessed by ion-exchange resin adsorption (mg/kg resin), in mineral soil (c. 3-8 cm depth) of the control and trenched plots during in situ incubation for 593 days (January 1997-September 1998) Values are means with s.e.m. in parentheses Plot treatment [NH.sub.4][sup.+]-N [NO.sub.3][sup.-]-N Control 46 (27) 9 (4) Trenched 474 (151)(*) 1940 (152)(**) Plot treatment Min-N Inorganic P Control 54 (31) 64 (21) Trenched 2420 (57)(*) 205 (94) (*) P < 0.05; (**) P < 0.01, for the difference between control and trenched plots.
Gross N mineralisation and nitrification
Gross N mineralisation in both FH material and mineral soil was significantly lower in the trenched than control samples before, but not after, incubation for 57 days (Fig. 1 c, d). Gross N immobilisation also tended to be lower in the trenched than control samples before incubation. In incubated samples, gross N immobilisation remained lower in the trenched than control samples of FH material, but in mineral soil was similar in both treatments (Fig. 1 c, d). Gross nitrification in mineral soil and FH material was, in contrast, significantly higher in the trenched than control samples before incubation, but did not differ between treatments after incubation (Fig. 1 e, f).
Contribution of root-associated factors to the labile C pools and C mineralisation
An assessment of the contribution of root-associated materials, such as exudates, decomposing tissues, and associated organisms, to soil biochemical properties was obtained by comparing values on an area basis to 20 cm depth of mineral soil. The bulk densities (Mg/[m.sup.3]) used were 0.08 for L and FH material, 1.42 for 0-10 cm depth mineral soil, and 1.51 for 10-20 cm depth mineral soil (F. J. Cook, unpubl, data). Property values (to 20 cm depth of mineral soil) were calculated as the sum of the weighted values for each depth of sample. The difference between control and trenched-plot values, expressed as a percentage of the control value, was taken to be the root-associated input. On this area basis, inputs from roots in the control soil would have accounted for 40% of the extractable C pool, 28% of microbial C, 26% of microbial N, and 23 or 21% of [CO.sub.2]-C production (calculated over 0-7 or 7-14 days, respectively).
Soil property values were, with very few exceptions, similar in the control and trenched plots at the commencement of the trial, and in the control samples at the beginning and end of the trial. The changes that occurred after trenching can therefore be attributed with confidence to the severance of roots and the associated reduction in below-ground C inputs and root metabolism in the trenched plots. Visual examination of the plots at the completion of the trial showed that trenching, with one minor exception, had been effective in preventing root growth. Actively growing roots were seen in only one of the plots where a few had invaded in one corner over the top of the polythene sheet; this particular area was avoided when samples were taken.
The exceptions noted above occurred in (a) L material at the beginning of the trial, and (b) mineral soil in the control 1996 and 1998 samples. (a) The initially higher net N mineralisation (0-14 days) in L material from the trenched than control plots may have been associated with a sampling effect; the total C concentration was lower in the trenched than control samples, suggesting that some decomposed needles could have been included in the trenched material. That this difference was not a constant feature of L material in these plots is shown by the indistinguishable values of net mineral-N production (0-14 days) at the completion of the trial in the trenched and control samples (Table 5); these samples had then essentially the same total C concentration (Table 1). (b) The somewhat higher mineral-N concentration in the spring (1998) samples than in the winter (1996) samples of the control mineral soil may have resulted from seasonal effects.
In the closed canopy plantation at Himatangi, moisture contents were appreciably higher in samples from the trenched than from the control plots, as in most (Horn 1985; Fisher and Gosz 1986; Hope and Li 1997), but not all (Silver and Vogt 1993; Simard et al. 1997), trenching studies. The slight increase in soil acidity after trenching is probably dependent to some extent on the increase in [NO.sub.3][.sup.-]-N formation (Binkley and Richter 1987). A decline in soil pH after trenching was also found by Silver and Vogt (1993), but was non-detectable in other systems (Simard et al. 1997; Hart and Sollins 1998).
Carbon and N pools and C mineralisation
One of the effects of trenching in a P radiata plantation is a loss of ectomycorrhizal fungi (Gadgil and Gadgil 1975). In the non-trenched system, these fungi could have reduced the activity of the decomposer population by competing more effectively for N and other nutrients (Smith and Read 1997). A reduction in ectomycorrhizal activity after trenching is likely to have occurred and contributed to the differences found in C and N pools and metabolism between our control and trenched samples of FH material.
Whereas trenching had a non-significant effect on total C concentrations, it resulted in some significant (P [is less than] 0.05) reduction in FH material and mineral soil of the pools of extractable C and microbial C, and to a lesser extent (P [is less than] 0.10) extractable organic N and microbial N. This reduction in both labile substrates and microbial biomass after trenching would probably have been responsible for the reduced respiratory activity in samples from the trenched plots. Other studies suggest that the effects of trenching on [CO.sub.2]-C production under laboratory conditions may be site-dependent. Respiratory activity was lower in litter and mineral soil from trenched than from control plots in Douglas fir stands (Hope and Li 1997), but not in moisture-adjusted samples from a mixed-conifer forest 2 years after trenching (Fisher and Gosz 1986).
The expression of our results on an area basis has also shown the marked effect of root-associated factors on labile C pools and [CO.sub.2]-C production. Based on these results, heterotrophic [CO.sub.2]-C production from the decomposition of root-associated inputs would have comprised [is greater than] 20% of the total in situ respiratory activity. At this site, the respiratory activity of roots themselves, viz. the activity of living roots and associated mycorrhizae and rhizosphere microoganisms, was estimated to be about 55-60% of total in situ respiration (K. R. Tate, D. J. Ross, N. A. Scott, N. J. Rodda, J. A. Townsend unpubl. data). Similar contributions from roots to soil respiratory activity in situ have been reported for other pine-dominated forests (Andrews et al. 1999; Hanson et al. 2000; Raich and Tufekcioglu 2000).
Although [CO.sub.2]-C production was lower in FH material from the trenched than from the control plots (Table 3; Fig. 1 a, b), it was higher in the trenched-plot samples when expressed per unit of microbial C (viz. the q[CO.sub.2] value). This suggests that microbial populations in FH material were under some stress (Wardle and Ghani 1995) after trenching. In mineral soil, however, microbial metabolic efficiency appears to have been similar in the trenched and control treatments, based on their indistinguishable q[CO.sub.2] values.
Nitrogen mineralisation and nitrification
In the [sup.15]N experiments, the absolute values for gross N transformations could have been influenced to some extent by the preliminary storage of the samples for 12 weeks. Mineral-N concentrations had then increased appreciably, although treatment differences remained proportionately similar to those in the unstored samples (data not shown). These stored, non-incubated (Day 0 samples) did, however, differ between the 2 treatments in many of their rates of gross N transformations (Fig. 1 c-e), with the differences in nitrification, in particular, being consistent with those found in net N transformations. Our use of the stored samples is, therefore, considered valid for indicating gross N cycling patterns in the control and trenched treatments.
Our finding that mineral-N concentrations in non-incubated samples were higher in the trenched than control treatment is consistent with the results of other studies that have generally shown increased mineral-N concentrations (Vitousek et al. 1979; Fisher and Gosz 1986; Silver and Vogt 1993), or increased net N mineralisation (Hart and Sollins 1998) after trenching. Such changes, however, may depend on ecosystem properties, and possibly time since trenching, as no significant effect on [NH.sub.4][sup.+]-N concentrations was found in a system with Douglas fir seedlings (Simard et al. 1997). Trenching, likewise, had no consistent effect on mineralisable N in 2 stands of Douglas fir, although the trenches there were only 35 cm deep and may not have prevented the lateral growth inwards of roots from below this depth (Hope and Li 1997).
One of the most striking aspects of our N mineralisation results is the relatively high net production of [NO.sub.3][sup.-]-N in the trenched treatment. An increase in net [NO.sub.3][sup.-]-N production has been commonly observed in other trenching studies (Vitousek et al. 1979; Silver and Vogt 1993) and also after the harvesting of P radiata (Dyck et al. 1983; Smith et al. 1994). Several mechanisms might be responsible for these increases in [NO.sub.3.][sup.-]-N. They include: reduced concentrations of inhibitory monoterpenes from roots (Janson 1993; Paavolainen et al. 1998), increased populations of nitrifying organisms as a result of increased availability of [NH.sub.4][sup.+]-N (Vitousek and Matson 1985; Van Miegroet et al. 1990), and lower microbial demand for [NO.sub.3][sup.-]-N by soil heterotrophs as a result of reduced C availability (Hart et al. 1994a; Scott et al. 1998). In both FH material and mineral soil, the marked increases in gross nitrification during incubation, and the similarity of gross nitrification and nitrate consumption in control and trenched samples after 57 days, appeared to coincide with reduced C availability, based on [CO.sub.2]-C production rates (Fig. 1). Our results therefore support the hypothesis of Hart et al. (1994a), but do not exclude the possibility of other mechanisms also influencing rates of [NO.sub.3][sup.-]-N production.
Carbon availability could also have influenced other N cycling pathways. The decline in rates of gross N mineralisation and immobilisation in samples of the control mineral soil, and the similarity of values in the control and trenched samples, after incubation for 57 days (Fig. 1d) suggest that pools of labile N and C in the control soil had been reduced during this incubation period. In the net N-mineralisation experiment, net mineral-N pools after incubation for 56 days were indistinguishable in control and trenched samples of FH material and both depths of mineral soil, suggesting that the pool of readily mineralisable N was similar in both treatments.
Strong support for our laboratory mineralisation results was provided by the in situ experiment, although the uptake of mineral N by roots would probably have contributed to the comparatively low mineral-N adsorption by the resin in the control soil. However, the ratio of [NO.sub.3][sup.-]-N / [NH.sub.4][sup.+]-N adsorbed was markedly higher in the trenched than in the control plots, even though the adsorption of [NH.sub.4][sup.+]-N is likely to have been relatively greater in the trenched plots because of their higher moisture content (Binkley 1984). Hart and Sollins (1998), in contrast, did not find a close correspondence between laboratory and in situ determinations (using a resin-core method; Hart and Perry 1999) of soil N mineralisation in untrenched and trenched plots in an old-growth conifer forest. There, net N mineralisation in mineral soil (0-15 cm depth) was significantly greater in the trenched than untrenched samples in the laboratory, but tended to be higher in the untrenched than trenched samples incubated in situ for c. 7.5 months in `winter' and 4.5 months in `summer' (Hart and Sollins 1998). Their experimental technique was, however, different from ours with excised roots in their field cores (5 cm diam.) possibly resulting in increased substrate availability for heterotrophic microorganisms and N cycling rates; any such effect should have been greater in the untrenched plots because of their root content being higher than in the trenched plots (Hart and Sollins 1998).
Although our results have shown clearly that trenching influenced several aspects of soil C and N cycling, the quantitative effects of living roots on these biochemical properties cannot be assessed with a high degree of accuracy. Factors responsible include uncertainties about the viability of severed roots (Ferrier and Alexander 1985; Santantonio and Santantonio 1987; Publicover and Vogt 1993) and the contribution of decomposing roots to the various pools and fluxes (Silver and Vogt 1993). Neither of these aspects was examined here. Root viability may not, however, have been a major factor, based on the finding of Arneth et al. (1998) that roots of 22-year-old P. radiata appeared to be dead 4 months after tree harvesting. Their site, moreover, had lower annual rainfall and air temperature, and therefore probably lower rates of decomposition than those at Himatangi. Overall, we consider that our observed trenching effects of decreased labile C pools and C mineralisation, and increased N-cycling rates (including nitrification), are likely to be conservative estimates of how living roots influence these properties because of the unquantified input of C and other nutrients from root decomposition. We do, however, recognise that our results are from only one site and that further work is needed to determine their applicability to P. radiata at other locations.
We thank P. McCarthy and B. Guild of Earnslaw One Ltd for site information and permission to work in Himatangi forest, and the Foundation for Research Science and Technology for financial support.
Andrews JA, Harrison KG, Matamala R, Schlesinger WH (1999) Separation of root respiration from total soil respiration using carbon-13 labeling during free-air carbon dioxide enrichment (FACE). Soil Science Society of America Journal 63, 1429-1435.
Arneth A, Kelliher FM, Gower ST, Scott NA, Byers JN, McSeveny TM (1998) Environmental variables regulating carbon dioxide efflux following clear-cutting of a Pinus radiata D. Don plantation. Journal of Geophysical Research 103, 5695-5705.
Bartha R, Pramer D (1965) Features of a flask and method for measuring the persistence and biological effects of pesticides in soil. Soil Science 100, 68-70.
Binkley D (1984) Ion exchange resin bags: Factors affecting estimates of nitrogen availability. Soil Science Society of America Journal 48, 1181-1184.
Binkley D, Matson P (1983) Ion exchange resin bag method for assessing forest soil nitrogen availability. Soil Science Society of America Journal 47, 1050-1052.
Binkley D, Richter D (1987) Nutrient cycles and [H.sup.+] budgets of forest ecosystems. Advances in Ecological Research 16, 1-15.
Blakemore LC, Searle PL, Daly BK (1987) Methods for chemical analysis of soils. New Zealand Soil Bureau Scientific Report No. 80.
Brooks PD, Stark JM, McInteer BB, Preston T (1989) A diffusion method to prepare KCl extracts for [sup.15]N analysis. Soil Science Society of America Journal 53, 1707-1711.
Cadisch G, Miller KE (1997) `Driven by Nature: plant litter quality and decomposition.' (CAB International: Wallingford, UK)
Cowie JD, Smith BAJ (1958) Soils and agriculture of Oroua Downs, Taikorea, and Oroua Districts, Manawatu County. Department of Scientific and Industrial Research, New Zealand Soil Bureau Bulletin 16.
Dyck WJ, Goss JR, Hodgkiss PD (1983) Nitrate losses from disturbed ecosystems in New Zealand--a comparative analysis. New Zealand Journal of Forestry Science 13, 14-24.
Ferrier RC, Alexander IJ (1985) Persistence under field conditions of excised fine roots and mycorrhizas of spruce. In `Ecologial interactions in soil. Plants, microbes and animals'. (Eds AH Fitter, D Atkinson, D Read, MB Usher) pp. 175-179. Special Publication number 4 of the British Ecological Society (Blackwell: Oxford)
Fisher FM, Gosz JR (1986) Effects of trenching on soil processes and properties in a New Mexico mixed-conifer forest. Biology and Fertility of Soils 2, 35-42.
Gadgil RL, Gadgil PD (1975) Suppression of litter decomposition by mycorrhizal roots of Pinus radiata. New Zealand Journal of Forestry Science 5, 33-41.
Hanson PJ, Edwards NT, Garten CT, Andrews JA (2000) Separating root and soil microbial contributions to soil respiration: A review of methods and observations. Biogeochemistry 48, 115-146.
Harding DE, Ross DJ (1964) Some factors in low-temperature storage influencing the mineralisable-nitrogen of soils. Journal of the Science of Food and Agriculture 15, 829-834.
Hart SC, Perry DA (1999) Transferring soils from high- to low-elevation forests increases nitrogen cycling rates: climate change implications. Global Change Biology 5, 23-32.
Hart SC, Sollins P (1998) Soil carbon and nitrogen pools and processes in an old-growth conifer forest 13 years after trenching. Canadian Journal of Forest Research 28, 1261-1265.
Hart SC, Nason GE, Myrold DD, Perry DA (1994a) Dynamics of gross nitrogen transformations in an old-growth forest: the carbon connection. Ecology 75, 880-891
Hart SC, Stark JM, Davidson EA, Firestone MK (1994b) Nitrogen mineralization, immobilization, and nitrification. In `Methods of soil analysis: Part 2. Microbiological and biochemical properties'. 3rd edn. (Eds RW Weaver, JS Angle, PS Bottomley) pp. 985-1018. (Soil Science Society of America: Madison, WI)
Hobbie SE (1992) Effects of plant species on nutrient cycling. Trends in Ecology and Evolution 7, 336-339.
Hope SM, Li C-Y (1997) Respiration, nitrogen fixation, and mineralizable nitrogen spatial and temporal patterns within two Oregon Douglas-fir stands. Canadian Journal of Forest Research 27, 501-509.
Horn JC (1985) Responses of understory tree seedlings to trenching. The American Midland Naturalist 114, 252-258.
Hurlbert SH (1984) Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54, 187-211.
Janson RW (1993) Monoterpene emissions from Scots pine and Norwegian spruce. Journal of Geophysical Research 98, 2839-2850.
Jenkinson DS (1988) Determination of microbial biomass carbon and nitrogen in soil. In `Advances in nitrogen cycling in agricultural ecosystems'. (Ed. JR Wilson) pp. 368-386. (CAB International: Wallingford, UK)
Kirkham D, Bartholomew WV (1954) Equations for following nutrient transformations in soil, utilizing tracer data. Soil Science Society of America Proceedings 18, 33-34.
Kladivko EJ, Keeney DR (1987) Soil nitrogen mineralization as affected by water and temperature interactions. Biology and Fertility of Soils 5, 248-252.
Murphy J, Riley JP (1962) A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta 27, 31-36.
Paavolainen L, Kitunen V, Smolander A (1998) Inhibition of nitrification in forest soil by monoterpenes. Plant and Soil 205, 147-154.
Parfitt RL, Ross DJ, Salt GJ (1998) Nitrogen and phosphorus mineralization in Pinus radiata harvest residue samples from a coastal sand. New Zealand Journal of Forestry Science 28, 347-360.
Paul EA, Clark FH (1996) `Soil microbiology and biochemistry.' 340 pp. (Academic Press Inc.: San Diego)
Publicover DA, Vogt KA (1993) A comparison of methods for estimating fine root production with respect to sources of error. Canadian Journal of Forest Research 23, 1179-1186.
Raich JW, Tufekcioglu A (2000) Vegetation and soil respiration: Correlations and controls. Biogeochemistry 48, 71-90.
Ross DJ (1992) Influence of sieve mesh size on estimates of microbial carbon and nitrogen by fumigation-extraction procedures in soils under pasture. Soil Biology and Biochemistry 24, 343-350.
Ross DJ, Sparling GP (1993) Comparison of methods to estimate microbial C and N in litter and soil under Pinus radiata on a coastal sand. Soil Biology and Biochemistry 25, 1591-1599.
Santantonio D, Santantonio E (1987) Effect of thinning on production and mortality of fine roots in a Pinus radiata plantation on a fertile site in New Zealand. Canadian Journal of Forest Research 17, 919-928.
Scott NA, Binkley D (1997) Foliage litter quality and annual net N mineralization: comparison across North American forest sites. Oecologia 111, 151-159.
Scott NA, Parfitt RL, Ross DJ, Salt GJ (1998) Carbon and nitrogen transformations in New Zealand plantation forest soils from sites with different N status. Canadian Journal of Forest Research 28, 967-976.
Silver WL, Vogt KA (1993) Fine root dynamics following single and multiple disturbances in a subtropical wet forest ecosystem. Journal of Ecology 81,729-738.
Simard SW, Perry DA, Smith JE, Molina R (1997) Effects of soil trenching on occurrence of ectomycorrhizas on Pseudotsuga menziesii seedlings grown in mature forests of Betula papyrifera and Pseudotsuga menziesii. New Phytologist 136, 327-340.
Smith SE, Read DJ (1997) `Mycorrhizal symbiosis.' (Academic Press: London)
Smith CT, Dyck WJ, Beets PN, Hodgkiss PD, Lowe AT (1994) Nutrition and productivity of Pinus radiata following harvest disturbance and fertilization of coastal sand dunes. Forest Ecology and Management 66, 5-38.
SYSTAT (1996) `SYSTAT 6.0 for Windows: Statistics.' (SPSS Inc.: Chicago, IL)
Van Miegroet H, Johnson DW, Cole DW (1990) Soil nitrification as affected by N fertility and changes in forest floor C/N ratio in four forest soils. Canadian Journal of Forest Research 20, 1012-1019.
Vitousek PM, Matson PA (1985) Causes of delayed nitrate production in two Indiana forests. Forest Science 31, 122-131.
Vitousek PM, Gosz JR, Grier CC, Mellilo JM, Reiners WA, Todd RL (1979) Nitrate losses from disturbed ecosystems. Science 204, 469-474.
Wardle DA, Ghani A (1995) A critique of the microbial metabolic quotient (q[CO.sub.2]) as a bioindicator of disturbance and ecosystem development. Soil Biology and Biochemistry 27, 1601-1610.
Manuscript received 24 July 2000, accepted 23 February 2001
Des J. Ross(A), Neal A. Scott, Kevin R. Tate, Natasha J. Rodda, and Jackie A. Townsend
Landcare Research, Private Bag 11052, Palmerston North, New Zealand.
(A) Corresponding author; email: email@example.com
|Printer friendly Cite/link Email Feedback|
|Author:||Ross, Des J.; Scott, Neal A.; Tate, Kevin R.; Rodda, Natasha, J.; Townsend, Jackie A.|
|Publication:||Australian Journal of Soil Research|
|Date:||Sep 1, 2001|
|Previous Article:||Validation of the GLEAMS simulation model for estimating net nitrogen mineralisation and nitrate leaching under cropping in Canterbury, New Zealand.|
|Next Article:||Application of VARLEACH and LEACHM models to to experimental data on leaching of a non-reactive tracer and three sulfonylurea herbicides.|