Carbon storage in a Ferrosol under subtropical rainforest, tree plantations, and pasture is linked to soil aggregation.
Managing the carbon (C) storage capacity of soils has been proposed as a method to counteract increasing atmospheric C[O.sub.2] concentration. However, better understanding is required of the mechanisms controlling soil organic C (SOC) storage and turnover in different ecosystems to ensure best practice and accurate quantification for C-trading schemes. Soils of tropical and subtropical forests account for almost 30% of total global SOC, but high rates of deforestation result in a loss of 212 Mt of SOC annually (Fearnside 2000; Achard et al. 2002; Mayaux et al. 2005). In addition, the increasing rate of plantation establishment in the tropics and subtropics (FAO 2005) highlights the need for improved understanding of the effects of land use on soil C stabilisation.
Three proposed mechanisms for SOC stabilisation are: (1) incorporation of SOC in soil aggregates that establishes a barrier between microbes, microbial enzymes, and organic matter substrates; (2) preservation of SOC through inherent biochemical recalcitrance, or selective degradation into chemically resistant materials during microbial decomposition; and (3) sorption, precipitation, or complexation of SOC with the mineral matrix via intermolecular interactions that reduce the availability of substrate through changes in conformation and binding of functional groups (Christensen 1996; Sollins et al. 1996; Jastrow and Miller 1998; Baldock and Skjemstad 2000; Six et al. 2002a; Krull et al. 2003). There is currently limited knowledge of which mechanisms are most important for C storage under different soils and land-use systems, yet such knowledge is crucial for devising systems with efficient C sequestration.
There is evidence that the ability of soils to store organic C is a function of the content of poorly crystalline minerals, and that soil C storage capacity decreases as crystalline minerals accumulate during advanced stages of weathering (Torn et al. 1997; Pal et al. 2004; Bera et al. 2005). Poorly crystalline iron (Fe) and aluminium (A1) oxides and hydroxides have a larger surface area for SOC sorption than crystalline minerals, such as goethite and hematite Fe oxides, or silicate clays (Schwertmann et al. 1986) and are responsible for preservation of SOC in the mineral matrix of various temperate and tropical forest soils (Powers and Schlesinger 2002; Kleber et al. 2005; Wiseman and Puttmann 2005; Mikutta et al. 2006). Stabilisation of SOC through sorption or complexation to soil minerals is considered to be an important mechanism for maintaining a slow-turnover, resistant pool of SOC in forest soils (Tom et al. 1997; Kiem and Kogel-Knabner 2002; Eusterhues et al. 2003). Few studies have investigated the influence of Fe and A1 oxides and hydroxides for soil C stabilisation following land-use change in subtropical and tropical soils (Lopez-Ulloa et al. 2005), which are often highly weathered, contain clays of low activity, and are rich in Fe and Al oxides and hydroxides (Isbell 1994; Feller and Beare 1997).
Soil microaggregates may be an important store of SOC that is protected from microbial attack over long time scales (Six et al. 2002a). Knowledge of aggregate protection of SOC is mostly based on studies in agricultural systems and pastures (Elliott et al. 1991; Cambardella and Elliott 1992; Beare et al. 1994a, 1994b; Jastrow 1996; Jastrow and Miller 1998; Six et al. 2000a, 2000b), but some evidence suggests that protection of SOC in stable aggregates is important for C storage in tropical and subtropical forested soils (Feller and Beare 1997; Garcia-Oliva et al. 1999; Krishnaswamy and Richter 2002). Carbon loss occurs from aggregate fractions following clearing of tropical forest and establishment of pasture or plantations (Krishnaswamy and Richter 2002; Ashagrie et al. 2005; Garcia-Oliva et al. 2006). However, little quantitative knowledge exists on aggregate stability after land-use change on highly weathered soils such as Ferrosols and Ultisols, which account for 60-70% of tropical soils (Feller and Beare 1997; Krishnaswamy and Richter 2002).
We showed that converting secondary subtropical rainforest on Ferrosols to exotic pastures resulted in similar soil C stocks after 39 years, while conversion of rainforest to native tree plantations caused a reduction of soil C stocks by 20 t/ha in the top 1.0m of soil after 50 years of plantation growth (Richards et al. 2007). The objective of this study was to examine the relationship between soil aggregation, mineralogy, and soil C in the 3 land uses to understand patterns of soil C storage. Such knowledge will contribute to understanding of the processes responsible for C sequestration in soils of forest plantations, an important consideration as tree plantations are increasingly established for C[O.sub.2] mitigation schemes.
Materials and methods
Rainforest, pasture, and plantation study sites were located in Yarraman State Forest, south-east Queensland, Australia (26[degrees]52'6"S, 151[degrees]51'1"E). The soil type is Red Ferrosol (Krasnozem) (Isbell 2002) or Rhodic Ferralsol (FAO 1998) and additional soil characteristics, including pH and texture, are described in Richards et al. (2007). Altitude ranges from 440 to 600 m a.s.l. The climate is subtropical with variable annual rainfall (range 377-1363 mm) and long-term mean rainfall of 744mm per annum (Jeffrey et al. 2001). Average maximum temperature in January is 29.7[degrees]C and average minimum temperature in July is 4.4[degrees]C (Jeffrey et al. 2001).
The rainforest, Araucarian microphyll vine forest (Webb 1959), is dominated by Flindersia australis (R.Br.), F. xanthoxyla (A.cunn. ex Hook), Brachychiton discolor (F.Muell.), Araucaria cunninghamii, A. bidwillii (Hook.), and an understorey of Excoecaria dallachyana (Baill.), Capparis arborea (F.Muell.), Casearia multinervosa (C.T.White and Sleumer ex. Sleumer), Arytera foveolata (F.Muell.), and Cleistanthus cunninghammii (Muell.Arg). The rainforest site was selectively logged in the early 20th Century but has not been logged since. Hoop pine (Araucaria cunninghamii, Araucariaceae) plantations were established on former rainforest sites. Prior to planting hoop pine, the rainforest was logged using bullock teams or small dozers. Eighteen months before planting, remaining valuable timber was salvaged and residues brushed off to a height of<0.15 m. Two months before planting, the site was burnt. The initial tree density for all plantings in 1940, 1952, 1968, and 1977 was 1517 stems/ha, which was progressively thinned to 246, 391, 532, and 735 stems/ha, respectively. All hoop pine sites were first rotation plantations and aged 63, 50, 34, and 25 years at the time of sampling. The pasture site, planted in 1965 following clearing and burning of rainforest, was adjacent to the plantation estate and consisted of Pennisetum clandestinum (kikuyu grass), an African grass species.
Hoop pine plantations, rainforest and pasture were sampled for soil and plant litter within 32 by 32 m quadrats near the centre of each site, with a minimum distance of 10 m from the edge. Five transects, parallel to the slope and 8 m apart, were sampled at 5-m intervals through the quadrat. Soil from hoop pine and pasture sites was sampled to 0.3 m depth using a truck-mounted, hydraulically operated sampler with a 45-mm-diameter steel tube. The rainforest site was sampled using a 75-mm-diameter hand auger, since it was inaccessible by vehicle. The soil core was sectioned into 4 depth intervals: 0-0.05, 0.05-0.1, 0.1-0.2, and 0.2-0.3m. Five composite soil samples were collected for each depth interval at each site. These composite samples were formed by mixing 5 individual soil samples. The rainforest and hoop pine sites aged 25, 34, and 50 years were sampled in April 2003 after summer rainfall. The 63-year-old hoop pine and pasture were sampled in November 2003. The samples were air-dried (40[degrees]C), ground to pass a 2-mm sieve, and stored in sealed plastic containers. In the rainforest site, bulk density was measured from 3 hand-excavated pits. Metal rings 69 mm in diameter were used to collect soil from the wall of each pit down to 0.3m. Collected soil from rainforest ring samples or subsamples of hoop pine and pasture cores were oven-dried at 105[degrees]C for at least 24h before weighing for bulk density determination. All bulk density samples were corrected for gravel and woody roots.
Three composite soil samples from the rainforest, pasture, and hoop pine sites aged 25, 50, and 63 years required for aggregate analysis were collected separately, using similar field methods to 0.1 m depth only, in November 2003. Fresh soil samples were stored in sealed plastic containers and transferred to a refrigerator within 24 h after sampling. Samples were sieved, while fresh, to 9.5 mm before air drying (40[degrees]C).
Soil was separated into water-stable aggregate fractions according to the procedure described by Six et al. (2002b). Briefly, 100g of air-dried (<9.5mm) soil was submerged for 5 min in distilled water on a 2000-[micro]m sieve to allow slaking of unstable aggregates. Following this, the sample was separated through a nest of sieves (2000 [micro]m, 250 p-m, and 53 [micro]m) using wet sieving for 15 min via a reciprocal shaker (38 cycles/min). Litter floating on the water in the 2000-[micro]m sieve was removed with a strainer, as it is by definition not considered part of soil organic matter (SSSA 1997). The material from each fraction was dried on the sieve for 3.5-4h (50[degrees]C) before removal and further drying in separate containers (~10-12h, 60[degrees]C). The smallest fraction (<53 [micro]m) was recovered after evaporation of the water in aluminium containers. The <53-[micro]m fraction was termed the silt and clay associated organic C (s+cOC). After drying, any gravel pieces from the samples >2000 [micro]m were removed by hand and samples were weighed.
Following initial separation, the free light fraction (LF) material from a 5-g subsample of the aggregate fractions >2000p-m and 250-2000 [micro]m, and a 3-g subsample from the 53-250 [micro]m fraction, was removed using density flotation in 1.6 g/[cm.sup.3] of sodium polytungstate (Richards et al. 2007). After isolation of the LF, the remaining aggregate sample was washed in distilled water to remove residual sodium polytungstate and then dispersed in 0.5% sodium hexametaphosphate by shaking for 18 h. The dispersed fraction was then further separated by passing through 2000-, 250-, and 53-[micro]m sieves (using the same method described previously). The dispersed fractions were combined to form 2 fractions: intra-aggregate particulate organic matter >53 [micro]m (iPOM) and intra-aggregate mineral associated C <53 [micro]m (imSOC). These terms follow the naming convention of Six et al. (2002b) where any organic material >53 [micro]m following soil dispersion is termed particulate organic matter. All aggregate samples (both before and after dispersion) were ground and analysed for C and N. Sand correction of all aggregate size fractions was performed by subtracting weights of the 5 intra-aggregate particulate organic matter and sand (iPOM+sand) fractions, following density separation, from the initial water-stable aggregate fractions (Six et al. 1998, 2002b). There was a mean recovery of 97% of whole soil mass in soil aggregate fractions separated by wet-sieving, and 88% recovery of whole soil mass following dispersion. Recovery of C in aggregate fractions was 96% and 76% of whole soil C for fractions separated by wet-sieving and dispersion, respectively. Some C may have been dissolved in the chemical solutions of sodium polytungstate and sodium hexametaphosphate, used to separate aggregates by density and dispersion, respectively, causing the lower recovery of whole soil C in the dispersed fractions.
Analysis of total carbon and nitrogen
Soil total C and N were determined on finely ground (<0.25 mm) composite soil samples by dry combustion and infrared detection in a LECO analyser (CNS-2000, LECO Corporation, MI, USA). Subsets of the same samples were also analysed for C and N using a mass spectrometer where only small sample sizes were required for C and N analysis. C content of the soils was similar (paired t-test, P> 0.05) with both analysis methods. Therefore, the remaining samples were analysed via continuous flow isotope ratio mass spectrometry (CF-IRMS, Tracer Mass, Europa Scientific), for total C and N.
Adjustment of carbon stocks to equivalent soil mass due to differences in bulk density
In order to compare SOC stocks between sites, total soil C values were scaled to the same equivalent soil mass measured under the 63-year hoop pine site. Soil C stocks for each site were recalculated from polynomial relationships ([r.sup.2] > 0.99) between cumulative soil C stock and soil mass calculated for that site (Dalai et al. 2005). This method represents the relationship between soil mass and C as a continuous variable rather than segregating soil into discrete depth intervals, and thus reduces error in the recalculated C stock values.
Soil iron and aluminium oxide and hydroxide analysis
Sodium citrate dithionite extractable ALl and Fe ([Fe.sub.d], Aid) were determined on whole soil samples according to the procedure of Mehra and Jackson (1960), while ammonium oxalate extractable (poorly crystalline) Fe and A1 ([Fe.sub.o], [Al.sub.o]) were determined following the procedure of McKeague and Day (1966). Fe and A1 concentrations in the extracts were measured by inductively coupled plasma-atomic emission spectrometry (ICP-AES) (Rayment and Higginson 1992). Crystalline Fe content was calculated as [Fe.sub.d]--Fen ([Fe.sub.d-o]) (Parfitt and Childs 1988).
Site effects were assessed using 1-way analysis of variance (ANOVA) with a Tukey HSD post-hoe comparison of means (SPSS Inc. ver. 12.0). Bivariate correlations and sequential multiple regression procedures were used to determine the relationship between soil properties (texture, depth, mineralogy, aggregation, pH) and different soil C fractions. Variables were combined to reduce multi-collinearity (variance inflation factor <2.5) and transformed to fit assumptions of normality and homoscedastic distribution of residuals (Tabachnick and Fidell 2007).
Carbon and N stocks in whole soil and density fractions
Soil C stocks in the top 0.3 m of soil under 25-, 34-, and 63-year-old hoop pine plantations were lower than under pasture, rainforest, and the 50-year hoop pine plantation, and this difference was also observed when comparing C stocks to 0.1 m depth (Table 1). Soil N stocks at 0.1 m depth were highest under rainforest and the 50-year hoop pine site, followed by pasture, while the lowest soil N stocks were measured in 63-year hoop pine plantations (Table 1).
Soil C associated with aggregates and primary particles
Soil C and N concentrations under hoop pine plantations were almost uniformly distributed across all 4 aggregate fractions (Fig. 1). Only the 63-year hoop pine plantation showed significant (P<0.05) differences in C content between fractions, with a lower C concentration in the aggregate fractions <53 [micro]m and 53-250 [micro]m than the large macroaggregates (>2000 [micro]m) (data not shown). However, there were significantly (P<0.05) higher C and N concentrations under rainforest and pasture in both the macroaggregate (250-2000 [micro]m) and microaggregate (53-250 [micro]m) fractions than at the other sites (Fig. 1). Rainforest soil had higher C and N concentrations in macroaggregates >2000 [micro]m than all other sites (Fig. 1). Expressed on a whole soil basis, 80-91% of total C and N in the top 0.1 m soil depth was found in aggregate fractions >250 [micro]m diameter at all sites.
[FIGURE 1 OMITTED]
Following the dispersion of water-stable aggregates, iPOM in the top 0.1 m of soil was significantly (P<0.05) higher under pasture and rainforest than the 25- and 63-year hoop pine plantations, and rainforest iPOM content was higher than the 50-year hoop pine plantation (Table 1). Similarly, intra-aggregate mineral-associated C (imSOC) was also highest under rainforest compared with all other sites (Table 1). C/N ratios of iPOM fractions were significantly (P<0.05) greater than imSOC fractions (data not shown), and ratios of both fractions were higher under 25- and 50-year plantations than rainforest and pasture (Table 1). As a percentage of total soil C, however, rainforest and pasture had significantly (P < 0.05) more iPOM and less mineral-associated C (imSOC + s+cOC) than plantations (Fig. 2).
Soil Fe and AI oxides and hydroxides
There was a significant effect of soil depth on Fe and A1 oxide and hydroxide concentrations, with an increasing concentration of [Fe.sub.d-o] (2-way ANOVA, F = 17.9, P < 0.001) and Aid (F= 3.6, P=0.02), and a decreasing concentration of [Fe.sub.o] (F=9.9, P<0.001) and [A1.sub.o] (F=6.8, P<0.001) with depth. There was a 10-fold higher concentration of Fed than [Al.sub.d] at each site, while concentrations of Feo and Alp were similar, and only [Fe.sub.o] was different between sites at each depth interval (Table 2). Soil under the 63-year hoop pine plantation had significantly (P<0.05) higher concentrations of [Fe.sub.d-o] at all depths than the other sites (Table 2). There was a similar pattern for [Al.sub.d] concentrations under the 63-year hoop pine site, although the values were not significantly higher than the 25- and 34-year hoop pine plantations at all depths (Table 2). The [Fe.sub.o]/[Fe.sub.d] ratio was generally highest under the 34-year hoop pine site and pasture, and lowest under rainforest and the 63-year hoop pine site, but this was not the case for the [Al.sub.o]/[Al.sub.d] ratio (Table 2).
[FIGURE 2 OMITTED]
Relationships between soil [Fe.sub.o], [Fe.sub.d], [Al.sub.o] and [Al.sub.d], aggregation and organic C
A bivariate correlation was performed to uncover the relationship between soil Fe and AI oxides and hydroxides or aggregation, and C fractions or other soil properties (Table 3). The concentrations of different forms of soil Fe and AI were significantly correlated with soil depth, and this was particularly pronounced for [Al.sub.o] (Table 3). However, the relationship was weak for [Fe.sub.d-o] and Aid, which had higher negative correlations with soil pH (r=-0.60 and -0.62, respectively), than did [Fe.sub.o] and [Al.sub.o] (Table3).Crystalline [Fe.sub.d-o] and Aid concentrations were also negatively correlated with concentrations of total C, N, and C associated with imSOC and iPOM (Table 3). [Fe.sub.d-o] and Aid concentrations were significantly correlated with the C concentration of water-stable aggregates in the size fractions 250-20001 [micro]m ([Fe.sub.d-o] r=-0.54, [A1.sub.d] r= -0.46), 53-250 [micro]m ([Fe.sub.d-o] r= -0.50, [A1.sub.d] r = -0.41), and <53 [micro]m ([Fe.sub.d-o] r = -0.52, [Al.sub.d] r=-0.48). Total concentrations of [Fe.sub.o] and [Al.sub.o] were only weakly (r < 0.3) correlated with different C fractions, including a small positive correlation between [Al.sub.o] and whole soil C and N concentrations (Table 3).
Multivariate analyses were performed to account for the effects of depth on soil C concentrations and find further relationships between soil mineralogy, aggregation, and soil C fractions (Table 4). Only data from the top 0-0.1 m soil depth are presented, but results were similar for the 0.1-0.3 m depth interval. [Fe.sub.d-o] and [Fe.sub.o] concentrations were significantly and negatively correlated with total soil C, and explained 30% of variation in soil C content (Table 4). Fe and A1 oxide and hydroxide concentrations in the soil were not correlated with the amount of iPOM across all sites.
When individual sites were compared, [Fe.sub.d-o] and the [Fe.sub.o]/[Fe.sub.d] ratio showed the strongest relationship with total soil C compared with other mineral indices (Fig. 3). The negative relationship between total C concentration and Fed_o was highly significant under rainforest ([r.sup.2]=0.96, P<0.01), but was weakest under pasture ([r.sup.2]=0.41, P<0.05) (Fig. 3). In contrast, the [Fe.sub.o]/[Fe.sub.d] ratio was strongly and positively correlated with total soil C at all sites except the 25-year hoop pine plantation (Fig. 3).
[FIGURE 3 OMITTED]
Previous studies of soil C dynamics following tree plantation establishment have observed a net loss of soil C, and this is thought to result from a reduction in C inputs to the soil and from organic matter decomposition caused by disturbance during site preparation (Guo and Gifford 2002; Paul et al. 2002; Turner et al. 2005). However, few studies have examined C storage in the context of soil properties such as aggregation and Fe and A1 oxide content (Six et al. 2002b; Pai et al. 2004). In our study we cannot rule out that differences in rates of C input and changes to soil properties during site preparation and thinning have contributed to the reduced C storage in the top 0.3 m under most hoop pine plantations, compared with rainforest and pasture. However, our results indicate that differences in soil aggregation between sites are associated with the observed patterns of SOC storage after land-use change.
All studied sites had a comparatively uniform spread of C concentrations in each water-stable aggregate size fraction. This finding, combined with the observation that soil C and N at all sites were mostly associated with macroaggregate-sized material, suggests that soil aggregates are resistant to rapid wetting treatments and may not display an aggregate hierarchy, as suggested by Tisdall and Oades (1982). In this sense, an aggregate hierarchy refers to the breakdown of aggregates in a stepwise fashion from larger macroaggregates to smaller microaggregates, to which Ferrosols do not conform (Oades and Waters 1991).
To further investigate soil C stabilisation through physical protection in the studied soils, we determined concentrations of intra-aggregate particulate organic matter (iPOM) and mineral associated C (imSOC + s+cOC). The greater concentration of iPOM under pasture and rainforest indicates that SOC is protected to a greater extent within aggregates than in the hoop pine soils, which are dominated by C associated with silt- and clay-sized particles (<53 [micro]m). Initial loss of soil C from the iPOM fraction could be caused by burning before plantation establishment. Ashagrie et al. (2005) observed a loss of soil C from the iPOM fraction following conversion of native forest to 21-year-old Eucalyptus monoculture plantations in Ethiopia, and suggested that burning during site clearing caused loss of stabilised C in aggregates. The pasture site was also burned before establishment; however, we observed a much greater recovery of the pasture iPOM pool compared with plantations, indicating that hoop pine plantations had a direct effect on soil aggregation.
We showed that C in imSOC fractions was possibly more highly processed and persistent than iPOM due to its lower C/N ratio, and longer mean C residence time (22-77 years; unpubl, data) calculated from soil [[delta].sup.13]C isotope signatures of pasture fractions (Richards et al. 2007). If this is also the case for plantations, then it appears that hoop pine soils, with greater proportions of total C in imSOC fractions, are dominated by a pool of old stabilised C protected by silt- and clay-sized particles, whereas physical protection of SOC in larger aggregates may not be an important mechanism for SOC stabilisation. This supports our previous work on hoop pine where we observed that a large, decomposition-resistant pool of rainforest-derived C remained in plantation soils (Richards et al. 2007). Further support for the observation that plantation establishment is responsible for a reduction in aggregate-protected soil C was the lower amount of iPOM in the 50-year hoop pine plantation. The 50-year hoop pine plantation had levels of total C stock to 0.1 m soil depth similar to those under rainforest, and at 0.3 m similar to SOC under rainforest and pasture. This indicates that change in land use, rather than differences in initial site conditions, may have resulted in a decrease in aggregate-protected C in soil under hoop pine.
The greater storage of C within aggregates in pasture and rainforest soil, compared with hoop pine plantations, may be due to different root characteristics. Roots are important binding agents for aggregates through their exudates and decomposition products, which support microbial populations in the rhizosphere, as well as their ability to enmesh soil particles (Tisdall and Oades 1982). Larger input of fine roots by pasture grasses and rainforest vegetation could explain the higher concentration of C in iPOM, and the significantly greater amount of C in the macroaggregate soil fraction >250 [micro]m, compared with hoop pine plantations. Preliminary observations suggest greater fine root C input under pasture, and possibly under rainforest due to a high diversity of plant species (Berish and Ewel 1988) compared with plantation sites (A. Richards, unpubl. data).
At all sites, [Fe.sub.d-o] and [Al.sub.d] (assumed to be crystalline Fe and Al oxides and hydroxides) (Schwertmann et al. 1986; Parfitt and Childs 1988) increased but C decreased with soil depth, suggesting that SOC may be involved in retarding crystallisation of Fe and AI oxides (Schwertmann et al. 1986; Hsu 1989; Schwertmann and Taylor 1989). On the other hand, since age of weathering increases with soil depth, this would allow a longer period for crystallisation of Fe and Al oxides and hydroxides as soil depth increases (Blume and Schwertmann 1969). Another possibility for the higher concentration of crystalline Fe and AI oxides and hydroxides at increasing depth is preferential leaching of these minerals through the soil profile (Park and Burt 1999).
The degree of crystallinity of Fe and Al oxides and hydroxides can influence organic matter storage (Schwertmann et al. 1986; Torn et al. 1997; Jones and Edwards 1998; Miltner and Zech 1998; van Hees et al. 2003; Kalbitz et al. 2005). Crystalline minerals have smaller specific surface areas and hydroxyl site densities than poorly crystalline minerals and, therefore, have a lower sorption capacity for organic material than poorly crystalline minerals (Guggenberger and Haider 2002; Kaiser and Guggenberger 2003). Several studies of forest soils have shown preferential stabilisation of C in the soil matrix through interaction with poorly crystalline Fe and Al oxides and hydroxides (Powers and Schlesinger 2002; Kleber et al. 2005; Lopez-Ulloa et al. 2005; Wiseman and Puttmann 2005). We observed significant, site-specific relationships between [Fe.sub.d-o] and [Fe.sub.o]/[Fe.sub.d] ratio and SOC, indicating a greater concentration of soil C when poorly crystalline Fe oxides dominate the soil profile. However, when all sites were combined, and the effect of soil depth was controlled in multiple regression analyses, we observed no positive correlation between poorly crystalline [Fe.sub.o] content and SOC.
The lack of significant correlations between soil C and mineral content across all sites may be because our study focused on the topsoil layers where organic matter is relatively abundant. In these layers soil Fe and AI oxides and hydroxides may have a limited capacity to sorb soil C because micro- and small mesopores on ferrihydrite or other poorly crystalline minerals would already be saturated by sorption to organic material (Kaiser et al. 2002; Guggenberger and Kaiser 2003; Kaiser and Guggenberger 2003; Kleber et al. 2005; Wiseman and Puttmann 2005; Lutzow et al. 2006). For example, ratios of organic C to total Fe oxide content at all sites and depths were mostly greater than the maximum sorption capacity of ferrihydrite (0.22 g C/g Fe) reported from laboratory studies (Wagai and Mayer 2007). Values from the current study ranged from 1.45 g C/g Fe under pasture at 0-0.05 m depth to 0.17 g C/g Fe under the 63-year hoop pine plantation at 0.2-0.3 m depth. Overall, the lack of significant relationships between SOC and Fe and Al oxide and hydroxide concentrations in these Ferrosols is consistent with other studies which showed limited capacity for sorptive preservation of organic matter in a range of soils (Guggenberger and Kaiser 2003; Wagai and Mayer 2007).
In summary, our study provides evidence that greater SOC storage under native rainforest and pasture, compared to hoop pine plantations, may be largely due to stabilisation in soil aggregates rather than via sorptive preservation through organo-mineral interactions. Since hoop pine soils had low amounts of C stored in soil aggregates, and since most Fe and Al oxide and hydroxide surfaces are already saturated with organic matter, there are probably limited options for stabilisation of new hoop pine derived C in plantation soil. The occurrence of lower aggregate-protected C in hoop pine soils is also consistent with previous research using the Century model. Model simulations predicted that plantations do not store sufficient C in the Century slow turnover C pool to compensate for loss of rainforest-derived C from this pool (Richards et al. 2007). It can be surmised that hoop pine biomass enters the LF soil C pool initially (Richards et al. 2007), but that the majority of these C inputs are not further stabilised within the soil matrix, possibly due to lower amounts of fine root C input. These findings have implications for accounting of carbon stocks in subtropical regions, especially where establishment of native tree plantations is an emerging practice.
AER received an Australian Postgraduate Award from the University of Queensland and the Queensland Department of Natural Resources and Water, and support from the Cooperative Research Centre (CRC) for Greenhouse Accounting and the Rainforest CRC. The Queensland Department of Primary Industries kindly provided access to the study sites. We thank Ben Harms for help with field sampling, Gordon Moss for carbon analysis, Steven Reeves for whole soil light fraction C separation, and several anonymous reviewers for helpful comments on previous drafts of the manuscript.
Manuscript received 15 July 2008, accepted 6 March 2009
Achard F, Eva HD, Stibig H, Mayaux P, Gallego J, Richards T, Malingreau J (2002) Determination of deforestation rates of the world's humid tropical forests. Science 297, 999-1002. doi: 10.1126/science. 1070656
Ashagrie Y, Zech W, Guggenberger G (2005) Transformation of a Podocarpus falcatus dominated natural forest into a monoculture Eucalyptus globulus plantation at Munesa, Ethiopia: soil organic C, N and S dynamics in primary particle and aggregate-size fractions. Agriculture, Ecosystems & Environment 106, 89-98. doi: 10.1016/ j.agee.2004.07.015
Baldock JA, Skjemstad JO (2000) Role of the soil matrix and minerals in protecting natural organic materials against biological attack. Organic Geochemistry 31, 697-710. doi: 10.1016/S0146-6380(00) 00049-8
Beare MH, Cabrera ML, Hendrix PF, Coleman DC (1994a) Aggregate-protected and unprotected organic matter pools in conventional-and no-tillage soils. Soil Science Society of America Journal 58, 787-795.
Beare MH, Hendrix PF, Coleman DC (1994b) Water-stable aggregates and organic matter fractions in conventional- and no-tillage soils. Soil Science Society of America Journal 58, 777-786.
Bera R, Seal A, Banerjee M, Dolui AK (2005) Nature and profile distribution of iron and aluminium in relation to pedogenic processes in some soils developed under tropical environments in India. Environmental Geology 47, 241-245. doi: 10.1007/s00254-004-1149-2
Berish CW, Ewel JJ (1988) Root development in simple and complex tropical successional ecosystems. Plant and Soil 106, 73-84. doi: 10.1007/BF02371197
Blume HP, Schwertmann U (1969) Genetic evaluation of profile distribution of aluminium, iron, and manganese oxides. Soil Science Society of America Proceedings 33, 438-444.
Cambardella CA, Elliott ET (1992) Particulate soil organic-matter changes across a grassland cultivation sequence. Soil Science Society of America Journal 56, 777-783.
Christensen BT (1996) Matching measurable soil organic matter fractions with conceptual pools in simulation models of carbon turnover: Revision of model structure. In 'Evaluation of soil organic matter models'. (Eds DS Powlson, P Smith, JU Smith) pp. 143-159. (Springer-Verlag: Berlin)
Dalal RC, Harms BP, Krull ES, Wang WJ (2005) Total soil organic matter and its labile pools following mulga (Acacia aneura) clearing for pasture development and cropping 1. Total and labile carbon. Australian Journal of Soil Research 43, 13-20. doi: 10.1071/SR04044
Elliott ET, Palm CA, Reuss DE, Monz CA (1991) Organic matter contained in soil aggregates from a tropical chronosequence: correction for sand and light fraction. Agriculture, Ecosystems & Environment 34, 443-451. doi: 10.1016/0167-8809(91)90127-J
Eusterhues K, Rumpel C, Kleber M, Kogel-Knabner I (2003) Stabilisation of soil organic matter by interactions with minerals as revealed by mineral dissolution and oxidative degradation. Organic Geochemistry 34, 1591-1600. doi: 10.1016/j.orggeochem.2003.08.007
FAO (1998) 'World reference base for soil resources. No. 84.' (FAO: Rome)
FAO (2005) 'Global forest resources assessment 2005: Progress towards sustainable forest management.' (FAO: Rome)
Fearnside PM (2000) Global warming and tropical land-use change: Greenhouse gas emissions from biomass burning, decomposition and soils in forest conversion, shifting cultivation and secondary vegetation. Climatic Change 46, 115-158. doi: 10.1023/A:1005569915357
Feller C, Beare MH (1997) Physical control of soil organic matter dynamics in the tropics. Geoderma 79, 69-116. doi: 10.1016/S0016-7061(97) 00039-6
Garcia-Oliva F, Lancho JFG, Montano NM, Islas P (2006) Soil carbon and nitrogen dynamics followed by a forest-to-pasture conversion in western Mexico. Agroforestry Systems 66, 93-100. doi: 10.1007/s10457-0052917-z
Garcia-Oliva F, Sanford RL Jr, Kelly E (1999) Effects of slash-and-burn management on soil aggregate organic C and N in a tropical-deciduous forest. Geoderma 88, 1-12. doi: 10.1016/S0016-7061(98) 00063-9
Guggenberger G, Haider KM (2002) Effect of mineral colloids on biogeochemical cycling of C, N, P, and S in soil. In 'Interactions between soil particles and microorganisms: Impact on the terrestrial ecosystem'. (Eds PM Huang, J-M Bollag, N Senesi) pp. 267-322. (John Wiley & Sons Ltd: Chichester, UK)
Guggenberger G, Kaiser K (2003) Dissolved organic matter in soil: challenging the paradigm of sorptive preservation. Geoderma 113, 293-310. doi: 10.1016/S0016-7061(02)00366-X
Guo LB, Gifford RM (2002) Soil carbon stocks and land use change: a meta analysis. Global Change Biology 8, 345-360. doi: 10.1046/ j.1354-1013.2002.00486.x
Hsu PH (1989) Aluminium hydroxides and oxyhydroxides. In 'Minerals in soil environments'. (Eds JB Dixon, SB Weed) pp. 331-378. (Soil Science Society of America: Madison, WI)
Isbell RF (1994) Krasnozems--A profile. Australian Journal of Soil Research 32, 915-929. doi: 10.1071/SR9940915
Isbell RF (2002) 'The Australian Soil Classification.' (CSIRO Publishing: Collingwood, Vic.)
Jastrow JD (1996) Soil aggregate formation and the accrual of particulate and mineral-associated organic matter. Soil Biology & Biochemistry 28, 665-676. doi: 10.1016/0038-0717(95)00159-X
Jastrow JD, Miller RM (1998) Soil aggregate stabilisation and carbon sequestration: Feedbacks through organomineral associations. In 'Soil processes and the carbon cycle'. (Eds R Lal, JM Kimble, RF Follett, BA Stewart) pp. 207-223. (CRC Press LLC: Boca Raton, FL)
Jeffrey SJ, Carter JO, Moodie KB, Beswick AR (2001) Using spatial interpolation to construct a comprehensive archive of Australian climate data. Environmental Modelling & Software 16, 309-330. doi: 10.1016/S1364-8152(01)00008-1
Jones DL, Edwards AC (1998) Influence of sorption on the biological utilization of two simple carbon substrates. Soil Biology & Biochemistry 30, 1895-1902. doi: 10.1016/S0038-0717(98)00060-1
Kaiser K, Eusterhues K, Rumpel C, Guggenberger G, Krgel-Knabner I (2002) Stabilisation of organic matter by soil minerals--investigations of density and particle-size fractions from two acid forest soils. Journal of Plant Nutrition and Soil Science 165, 451-459. doi: 10.1002/1522-2624(200208) 165:4<451::AID-JPLN451>3.0.CO;2-B
Kaiser K, Guggenberger G (2003) Mineral surfaces and soil organic matter. European Journal of Soil Science 54, 219-236. doi: 10.1046/j.13652389.2003.00544.x
Kalbitz K, Schwesig D, Rethemayer J, Matzner E (2005) Stabilisation of dissolved organic matter by sorption to the mineral soil. Soil Biology & Biochemistry 37, 1319-1331. doi: 10.1016/j.soilbio.2004.11.028
Kiem R, Kogel-Knabner I (2002) Refactory organic carbon in particle-size fractions of arable soils II: organic carbon in relation to mineral surface area and iron oxides in fractions <6 [micro]m. Organic Geochemistry 33, 1699-1713. doi: 10.1016/S0146-6380(02)00112-2
Kleber M, Mikutta R, Tom MS, Jahn R (2005) Poorly crystalline mineral phases protect organic matter in acid subsoil horizons. European Journal of Soil Science 56, 717-725.
Krishnaswamy J, Richter D (2002) Properties of advanced weathering-stage soils in tropical forests and pastures. Soil Science Society of America Journal 66, 244-253.
Krull ES, Baldock JA, Skjemstad JO (2003) Importance of mechanisms and processes of the stabilisation of soil organic matter for modeling carbon turnover. Functional Plant Biology 30, 207-222. doi: 10.1071/ FP02085
Lopez-Ulloa M, Veldcamp E, de Koning GHJ (2005) Soil carbon stabilisation in converted tropical pastures and forests depends on soil type. Soil Science Society of America Journal 69, 1110-1117.
Lutzow Mv, Kogel-Knabner I, Ekschmitt K, Matzner E, Guggenberger G, Marschner B, Flessa H (2006) Stabilisation of organic matter in temperate soils: mechanisms and their relevance under different soil conditions a review. European Journal of Soil Science 57, 426-445. doi: 10.111l/j.1365-2389.2006.00809.x
Mayaux P, Holmgren P, Achard F, Eva H, Stibig H, Branthomme A (2005) Tropical forest cover change in the 1990s and options for future monitoring. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 360, 373-384. doi: 10.1098/ rstb.2004.1590
McKeague JA, Day JH (1966) Dithionite- and oxalate-extractable Fe and Al as aids in differentiating various classes of soils. Canadian Journal of Soil Science 46, 3-22.
Mehra OP, Jackson ML (1960) Iron oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate. Clays and Clay Minerals 7, 317-327. doi: 10.1346/CCMN.1958.0070122
Mikutta R, Kleber M, Torn MS, Jahn R (2006) Stabilisation of soil organic matter: association with minerals or chemical recalcitrance? Biogeochemistry 77, 25-56. doi: 10.1007/s10533-005-0712-6
Miltner A, Zech W (1998) Beech leaf litter lignin degradation and transformation as influenced by mineral phases. Organic Geochemistry 28, 457-463. doi: 10.1016/S0146-6380(98)00019-9
Oades JM, Waters AG (1991) Aggregate hierarchy in soils. Australian Journal of Soil Research 29, 815-828. doi: 10.1071/SR9910815
Pai C-W, Wang M-K, Zhuang S-Y, King H-B (2004) Free and noncrystalline Fe-oxides to total iron concentration ratios correlated with [sup.14]C ages of three forest soils in central Taiwan. Soil Science 169, 582-589. doi: 10.1097/01.ss.0000138419.22546.00
Parfitt RL, Childs CW (1988) Estimation of forms of Fe and Al: a review, and analysis of contrasting soils by dissolution and Moessbauer methods. Australian Journal of Soil Research 26, 121-144. doi: 10.1071/SR9880121
Park SJ, Burt TP (1999) Identification of throughflow using the distribution of secondary iron oxides in soils. Geoderma 93, 61-84. doi: 10.1016/S0016-7061 (99)00042-7
Paul KI, Polglase P J, Nyakuengama JG, Khanna PK (2002) Change in soil carbon following afforestation. Forest Ecology and Management 168, 241-257. doi: 10.1016/S0378-1127(01)00740-X
Powers JS, Schlesinger WH (2002) Relationships among soil carbon distributions and biophysical factors at nested spatial scales in rain forests of northeastern Costa Rica. Geoderma 109, 165-190. doi: 10.1016/S0016-7061(02)00147-7
Rayment GE, Higginson FR (1992) 'Australian laboratory handbook of soil and water chemical methods.' (Inkata Press: Melbourne, Vic.)
Richards AE, Dalal RC, Schmidt S (2007) Soil carbon turnover and sequestration in native subtropical tree plantations. Soil Biology & Biochemistry 39, 2078-2090. doi: 10.1016/j.soilbio.2007.03.012
Schwertmann U, Kodama H, Fischer WR (1986) Mutual interactions between organics and iron oxides. In 'Interactions of soil minerals with natural organics and microbes'. (Eds PM Huang, M Schnitzer) pp. 223-250. (Soil Science Society of America: Madison, WI)
Schwertmann U, Taylor RM (1989) Iron oxides. In 'Minerals in soil environments'. (Eds JB Dixon, SB Weed) pp. 379-438. (Soil Science Society of America: Madison, WI)
Six J, Callewaert P, Lenders S, De Gryze S, Morris S J, Gregorich EG, Paul EA, Paustian K (2002b) Measuring and understanding carbon storage in afforested soils by physical fractionation. Soil Science Society of America Journal 66, 1981-1987.
Six J, Conant RT, Paul EA, Paustian K (2002a) Stabilisation mechanisms of soil organic matter: Implications for C-saturation of soils. Plant and Soil 241, 155-176. doi: 10.1023/A:1016125726789
Six J, Elliott ET, Paustian K (2000a) Soil macroaggregate turnover and microaggregate formation: a mechanism for C sequestration under no-tillage agriculture. Soil Biology & Biochemistry 32, 2099-2103. doi: 10.1016/S0038-0717(00)00179-6
Six J, Elliott ET, Paustian K, Doran JW (1998) Aggregation and soil organic matter accumulation in cultivated and natural grassland soils. Soil Science Society of America Journal 62, 1367-1377.
Six J, Paustian K, Elliott ET, Combrink C (2000b) Soil structure and organic matter: I. Distribution of aggregate-size classes and aggregate-associated carbon. Soil Science Society of America Journal 64, 681-689.
Sollins P, Homann P, Caldwell BA (1996) Stabilisation and destabilisation of soil organic matter: mechanisms and controls. Geoderma 74, 65-105. doi: 10.1016/S0016-7061 (96)00036-5
SSSA (1997) 'Glossary of soil science terms.' (Soil Science Society of America, Inc.: Madison, WI)
Tabachnick BG, Fidell LS (2007) 'Using multivariate statistics.' (Pearson Education Inc.: Boston, MA)
Tisdall CA, Oades JM (1982) Organic matter and water-stable aggregates in soils. Journal of Soil Science 33, 141-163. doi: 10.1111/j.13652389.1982.tb01755.x
Torn MS, Trumbore SE, Chadwick OA, Vitousek PM, Hendricks DM (1997) Mineral control of soil organic carbon storage and turnover. Nature 389, 170-173. doi: 10.1038/38260
Turner J, Lambert M, Johnson DW (2005) Experience with patterns of change in soil carbon resulting from forest plantation establishment in eastern Australia. Forest Ecology and Management 220, 259-269. doi: 10.1016/j.foreco.2005.08.025
van Hees PAW, Vinogradoff SI, Edwards AC, Godbold DL, Jones DL (2003) Low molecular weight organic acid adsorption in forest soils: effects on soil solution concentrations and biodegradation rates. Soil Biology & Biochemistry 35, 1015-1026. doi: 10.1016/S0038-0717(03) 00144-5
Wagai R, Mayer LM (2007) Sorptive stabilisation of organic matter in soils by hydrous iron oxides. Geochimica et Cosmochimica Acta 71, 25-35. doi: 10.1016/j.gca.2006.08.047
Webb L (1959) A physiognomic classification of Australian rain forests. Journal of Ecology 47, 551-570. doi: 10.2307/2257290
Wiseman CLS, Puttmann W (2005) Soil organic carbon and its sorptive preservation in central Germany. European Journal of Soil Science 56, 65-76. doi: 10.1111/j.1351-0754.2004.00655.x
Anna E. Richards (A,B,C,D,E), Ram C. Dalal (B,C), and Susanne Schmidt (A)
(A) School of Biological Sciences, The University of Queensland, St Lucia, Qld 4072, Australia. (B) Cooperative Research Centre for Greenhouse Accounting. (C) Queensland Department of Natural Resources and Water, 80 Meiers Rd, Indooroopilly, Qld 4068, Australia. (D) Current address: CSIRO Sustainable Ecosystems, Tropical Ecosystems Research Centre, PMB 44, Winnellie, NT 0822, Australia. (E) Corresponding author. Email: Anna.Richards@csiro.au
Table 1. Whole soil and particle size separated C and N stocks, and C/N ratios at each site Particle size separates include intra-aggregate particulate organic matter (iPOM >53[micro]m, mg C-g sand-free soil), or infra-aggregate mineral associated C (imSOC <53[micro]m, mg C-g sand-free soil). Standard errors are in parentheses (n = 3-5). Within columns, means followed by the same letter are not significantly different at P = 0.05. n.d., Not determined Site 0-0.1 m depth N C iPOM C imSOC (t/ha) (mg/g) Rainforest 7.1 (0.2) a 72.0 (1.9) a 77.1 (18.7) a 53.4 (1.4) a 25-year hoop 4.5 (0.1) c 46.4 (1.5) c 20.3 (4.0) c 33.7 (1.1) c 34-year hoop 4.0 (0.1) c 46.7 (1.4) c n.d. n.d. 50-year hoop 7.1 (0.1) a 77.1 (1.6) a 23.2 (2.1) bc 35.6 (1.9) c 63-year hoop 3.4 (0.1) d 44.6 (1.5) c 18.7 (6.5) c 32.7 (3.3) c Pasture 5.7 (0.1) b 62.8 (2.1) b 52.6 (15.0) ab 44.8 (3.0) b Site 0-0.3 m depth C (t/ha) (A) iPOM imSOC C/N Rainforest 10.2 (0.3) b 8.6 (0.1) b 137.9 (1.7) a 25-year hoop 15.5 (0.2) a 11.2 (0.2) a 99.7 (1.4) b 34-year hoop n.d. n.d. 95.4 (1.2) b 50-year hoop 13.8 (0.4) a 10.4 (0.2) a 131.9 (1.4) a 63-year hoop 10.4 (0.9) ab 7.1 (0.5) c 99.3 (1.5) b Pasture 10.1 (0.4) b 8.4 (0.2) bc 136.1 (1.9) a (A) Adapted from Richards et al. (2007). Table 2. Mean sodium citrate dithionite extractable Fe and A1 ([Fe.sub.d], [Al.sub.d]), ammonium oxalate extractable Fe and A1 (poorly crystalline) ([Fe.sub.o], [Al.sub.o]), and crystalline Fe ([Fe.sub.d-0]) from rainforest, hoop pine, and pasture sites at diffrent soil depths Within columns and depths, means followed by the same letter are not significantly different at P = 0.05 (n = 3) Site [Fe.sub.o] [Fe.sub.d-o] [Fe.sub.o]/ [Fe.sub.d] (g/kg) 0-0.05 m Rainforest 2.73 d 71.3 c 0.04 c 25-year hoop 5.23 ab 84.4 b 0.06 b 34-year hoop 6.31 a 16.3 c 0.09 a 50-year hoop 3.03 cd 58.3 c 0.05 bc 63-year hoop 4.10 bc 112.7 a 0.04 c Pasture 5.45 a 58.5 c 0.09 a 0.05-0.1 m Rainforest 2.66 c 79.8 bc 0.03 d 25-year hoop 5.55 a 89.7 b 0.06 bc 34-year hoop 6.21 a 67.9 cd 0.08 a 50-year hoop 3.12 c 65.2 cd 0.05 cd 63-year hoop 4.07 bc 119.1 a 0.03 d Pasture 5.25 ab 64.8 d 0.08 ab 0.1-0.2 m Rainforest 2.59 c 85.0 bc 0.03 c 25-year hoop 5.35 ab 89.5 b 0.06 ab 34-year hoop 5.76 a 67.6 d 0.08 a 50-year hoop 3.17 cd 70.2 cd 0.04 bc 63-year hoop 3.86 bc 124.7 a 0.03 c Pasture 4.44 abd 69.0 d 0.06 ab 0.2-0.3 m Rainforest 2.39 c 88.4 b 0.03 c 25-year hoop 5.06 ab 97.2 b 0.05 b 34-year hoop 5.41 a 70.3 c 0.07 a 50-year hoop 2.82 c 71.4 c 0.04 bc 63-year hoop 3.18 c 122.1 a 0.03 c Pasture 3.63 bc 69.7 c 0.05 b Site [Al.sub.o] [Al.sub.d] [Al.sub.o]/ [Al.sub.d] (g/kg) Rainforest 3.77 7.80 bcd 0.48 25-year hoop 4.46 9.30 abc 0.48 34-year hoop 4.14 9.71 ab 0.42 50-year hoop 3.54 5.93 d 0.60 63-year hoop 4.51 11.54 a 0.39 Pasture 3.48 6.58 cd 0.53 Rainforest 3.65 8.53 bc 0.43 abc 25-year hoop 3.52 9.72 ab 0.36 bc 34-year hoop 3.44 10.03 ab 0.34 c 50-year hoop 3.54 6.58 c 0.54 a 63-year hoop 3.74 11.55 a 0.32 c Pasture 3.86 7.69 bc 0.51 ab Rainforest 3.59 9.32 abc 0.39 bc 25-year hoop 3.14 9.61 abc 0.3 cd 34-year hoop 3.1 9.98 ab 0.31 cd 50-year hoop 3.51 6.86 c 0.51 a 63-year hoop 3.43 11.42 a 0.30 d Pasture 3.56 8.20 bc 0.43 ab Rainforest 3.45 9.56 b 0.36 ab 25-year hoop 3.07 10.08 ab 0.30 bc 34-year hoop 2.63 9.84 ab 0.27 c 50-year hoop 2.89 6.68 c 0.43 a 63-year hoop 3.34 11.95 a 0.28 bc Pasture 3.51 8.39 bc 0.42 a Table 3. Pearson's correlation matrix for poorly crystalline Fe and Al ([Fe.sub.o], [Al.sub.o]), crystalline Fe ([Fe.sub.d-o]), and Al ([Al.sub.d]), and soil properties from all sites Each soil property was measured to 0.3 m soil depth (n = 72). Sodium citrate dithionite extractable Fe and Al ([Fe.sub.d], [Al.sub.d]). * P < 0.05; ** P < 0.01; *** P < 0.001 Y/X Soil depth pH % Clay % C (m) [Fe.sub.o] (g/kg) -0.219 -0.367 ** -0.222 -0.026 [Fe.sub.o]/[Fe.sub.d] -0.290 * 0.015 -0.397 ** 0.222 [Fe.sub.d-o] (g/kg) 0.207 -0.603 *** 0.334 ** -0.401 *** [Al.sub.o] (g/kg) -0.460 *** 0.124 -0.248 * 0.273 * [Al.sub.o]/[Al.sub.d] -0.494 *** 0.643 *** -0.380 ** 0.610 *** [Al.sub.d] (g/kg) 0.172 -0.616 *** 0.194 -0.442 *** Y/X % N iPOM imSOC Aggregate C (mg C/g) (A) [Fe.sub.o] (g/kg) -0.081 -0.202 -0.219 [Fe.sub.o]/[Fe.sub.d] 0.193 0.128 0.138 [Fe.sub.d-o] (g/kg) -0.461 ** -0.416 * -0.467 ** [Al.sub.o] (g/kg) 0.264 * -0.118 -0.067 [Al.sub.o]/[Al.sub.d] 0.641 *** 0.168 0.297 [Al.sub.d] (g/kg) -0.497 *** -0.306 -0.386 * (A) Not measured at 34-year hoop pine site, n = 30, only 0-0.l m soil depth measured. Table 4. Multiple regression analysis for whole soil C (%) at 0-0.1 m soil depth for all sites * P < 0.05; ** P < 0.01; *** P < 0.001 Model Coefficients Semi-partial d.f. [R.sup.2] terms (s.e.) correlations Log(depth) -4.70 (0.75) *** 0.45 *** 32 0.75 *** l/[Fe.sub.d-o] 5.29 (1.05) *** 0.18 *** [Fe.sub.o] -4.93 (1.30) ** 0.12 **
|Printer friendly Cite/link Email Feedback|
|Author:||Richards, Anna E.; Dalal, Ram C.; Schmidt, Susanne|
|Publication:||Australian Journal of Soil Research|
|Date:||Jul 1, 2009|
|Previous Article:||Distribution and causes of intricate saline--sodic soil patterns in an upland South Australian hillslope.|
|Next Article:||Burning management and carbon sequestration of upland heather moorland in the UK.|