Organic matter characteristics under native forest, long-term pasture, and recent conversion to Eucalyptus plantations in Western Australia: microbial biomass, soil respiration, and permanganate oxidation.

Introduction

Extensive areas of native vegetation in south-western Australia have been cleared within the last century for agricultural production, with the majority used for pasture-based agriculture. Conversion from native vegetation to pasture has resulted in increased soil fertility, largely due to regular application of chemical fertilisers, inputs of nitrogen from leguminous pasture species, and subsequent addition of pasture- and livestock-derived residues. These factors are likely to have changed the quality of the soil organic matter. Another land use that is becoming more prevalent in the south-west is tree farming, with Eucalyptus globulus currently being established in south-western Australia at a rate exceeding 25 000 ha per year (National Forest Inventory 2000). First-rotation E. globulus plantations established on farmland are less intensively managed than pasture, and the amounts and nature of organic residues would be different to those in pasture-based agriculture. These changes have been shown to result in decreased N and P availability under E. globulus plantations (Grove et al. 2000).

Organic matter plays a pivotal role in soil fertility (Tiessen et al. 1994), as it is the main store of plant nutrients such as N (Russell 1973). Residues from different vegetation types can have different qualities for decomposition (Swift et al. 1979), and leguminous residues have lower C :N and lignin and tannin contents than eucalypt residues. These changes in residue input quality should be reflected in changes to the composition of the soil organic pool, possibly affecting microbial activity, turnover times, and nutrient release. Changes in soil organic matter composition may also be a sensitive measure of changes to the total organic pool in the longer term.

Although soil organic matter quality is an often-used term, it has proven difficult to quantify. Permanganate-oxidisable C is one approach that has been useful for measuring land-use-induced changes to the labile organic pool in the cropping systems of New South Wales (Blair et al. 1995; Conteh et al. 1998) and Queensland (Blair et al. 1995; Moody et al. 1997), but it has not been tested for pasture or plantation soils in Western Australia. Loginow et al. (1987) used 3 concentrations of permanganate (33, 167, and 333 mm), but generally 333 mm permanganate has been used in Australia (Blair et al. 1995). The proportion of total C oxidised with 333 mm is relatively high (10-25%), and labile soil C may be a smaller proportion than this. Hence lower concentrations of permanganate may be more sensitive at detecting small changes in labile C. Microbial biomass and carbon mineralisation (respiration) have been used to measure microbial C availability in many studies (Bauhus et al. 1993; Hassink 1994a; Franzluebbers 1999), and under constant conditions it also may be a sensitive indicator of changes in organic matter composition/ quality. The objective of this study was to examine the influence of land-use management on indicators of soil organic matter quality, including total C concentration, permanganate-oxidisable C, microbial biomass and respiration under controlled conditions. We tested the hypothesis that land-use change has resulted in a change in soil organic matter quality in surface soils (0-10 cm) at 10 sites in south-western Australia. Land use types at each site included native vegetation, pasture (20-71 years old), and plantation (7-10 years old).

Materials and methods

Site selection and soil sampling

Ten sites representing the climatic range in which E. globulus is grown in south-western Australia were selected (Table 1). The sites represented a range in soil texture, with clay contents between 2.4 and 15.2%. Adjacent pasture (20-71 years old) and E. globulus plantation (7-10 years old) plots, and a plot in nearby native vegetation on the same landform element, were selected at each site for sampling. Native vegetation was generally consisted of xerophytic communities with a jarrah (E. marginata) overstorey, except Anning, which was karri (E. diversicolor) forest. Eucalyptus globulus plantations in south-western Australia generally reach canopy closure by 2-3 years, after which time, growth in the understorey is minimal, so it contributes little to cycling of organic matter and nutrients. The plantations in this study were planted at a density of 1000-1200 stems/ha, and ranged in productivity from 8.5 to 44.1 [m.sup.3]/ha.year (Table 1). Soil profiles were evaluated at each of the three land-use types to ensure they were well matched at each site.