Maintenance and improvement of soil organic matter levels is an important concern in dryland farming systems of temperate regions. The Century soil organic matter model was used to simulate changes in soil organic C and total N under long-term wheat (Triticum aestivum L.) and pasture rotations at five sites in southern Australia. Average declines in soil organic C and total N of 14 and 10%, respectively, in continuous and wheat-fallow systems over a 10 to 20 year period were closely simulated by the model at each site. Additions of N fertilizer (80 kg N ha-1), which prevented soil organic matter decline in continuous wheat systems, was also well represented by the model. Trends in soil organic matter under long-term legume pasture were not adequately simulated by the model, probably due to the 'annual' nature of subterranean clover (Trifolium subterranean L.) in dry seasons and subsequent changes in the ratio of live to dead plant biomass and shoot to root ratios. Overall, the study emphasizes the importance of adequate total plant C production to prevent a decline in soil organic C.

No abstract

Soil hydraulic properties in the sandy clay loam surface horizon of duplex soils were assessed under cultivated, stubble burnt (CCB); direct drilling, stubble burnt (DDB); and direct drilling, stubble retained (DDR) systems for three wheat-lupin rotations, established from 7 to 10 years, in north-eastern Victoria. Disc permeameters were used to determine sorptivity and steady state infiltration in each rotation. A rainfall simulator was used on the 10 year rotation to characterize saturated infiltration and surface soil stability. Cultivation caused an increase in soil bulk density and decreased organic C at the soil surface. Application of a -40 mm water supply potential removed macropore flow from the infiltration process and mainly characterized water flow in the soil matrix. Differences in sorptivity among tillage treatments at this potential mainly reflected initial soil moisture, rather than soil structure. Sorptivity measured on soil cores was related (T = 0.74, P = 0.01)= to sorptivity measured in the field. Under Rainfall simulation , DDR increased sorptivity, wetting depth and time to runoff, and decreased runoff rate and sediment loss, compared with DDB and CCB. Overall, steady state infiltration rate was controlled over time by the permeability of lower soil horizons. The combination of disc permeameter and rainfall simulation measurements provided a useful description of unsaturated and saturated infiltration under field conditions. The DDR system improved the potential for saturated infiltration, maximized rainfall storage in the surface horizon, and increased the stability of macroporous infiltration.

A field experiment was conducted at Rutherglen, in north-eastern Victoria, to determine the effects of liming and deep ripping on soil water extraction by wheat, sorptivity of water into the soil profile and soil resistance to a penetrometer. The site was typical of many cropping paddocks in the region. In the unmodified state the top 20 cm of the soil profile was acid (pH 4.80) and there was a dense hardpan between 7.5 and 17.5 cm depth. Deep ripping increased water extraction by wheat by an average of 8 mm during a drought season (1982), but had no effect on water use in a wet season (1983). The major effect of ripping was to increase the water use in winter from below the ripped zone (40 cm) compared with the unripped treatment. Lime, either with or without ripping, had no significant effect on crop water extraction. Sorptivity, a measure of infiltration, was increased by ripping alone and by ripping plus lime. Soil resistance to a penetrometer was reduced by deep ripping; an effect which had persisted at least 30 months after the last ripping operation. Economic wheat yield responses were obtained by using deep ripping and liming to improve soil physical properties at this site.

The environment, duplex soil types and trends in crop production in South Australia, southern New South Wales, north-eastern and north-central Victoria, the southern Wimmera and the Victorian Western District are reviewed. In the latter 2 regions, pastoral industries dominate and crop production is curtailed by regular and severe soil waterlogging, except for limited areas of lower rainfall. Subsurface drainage can eliminate waterlogging, but is feasible only for the Western District where subsoils are sufficiently stable. The other regions all have a long history of soil degradation due to cropping practices, but these effects can now be minimised with the use of direct drilling and stubble retention cropping methods. A vigorous pasture ley phase is still considered necessary to maintain nitrogen levels and to restore soil structure to adequate levels for sustainable farming. Future productivity improvements will require increased root growth in the subsoils. Deep ripping, 'slotting' of gypsum, and crop species capable of opening up subsoils are techniques which may hold promise in this regard. The inclusion of lucerne, a perennial species, in annual pastures and intercropping at intervals is a technique being pioneered in north-central and western Victoria and may provide the best opportunity to crop duplex soils successfully without associated land degradation.