RESIDUAL EFFECTS OF TILLAGE AND CROP-ROTATION ON SOIL PROPERTIES, SOIL INVERTEBRATE NUMBERS AND NUTRIENT-UPTAKE IN AN IRRIGATED VERTISOL SOWN TO COTTON

The residual effects of tillage and cropping sequence on soil physical and chemical properties, surface-active and soil invertebrate composi tion and abundance, nutrient uptake, growth and yield of cotton were e valuated from 1994 to 1996 in a compacted Typic Haplustert (Vertisol) of north-western New South Wales, Australia. The experimental treatmen ts from 1985 to 1992 were intensive tillage (disc-ploughing to 200 mm, chisel ploughing to 300 mm followed by ridging every year) sown with continuous cotton (Gossypium hirsutum L.); minimum tillage (planting o n ridges retained intact from previous years with soil disturbance bei ng limited to deepening of the furrows with disc-hillers and shallow c ultivation on ridge surfaces) sown with either continuous cotton or a cotton-winter wheat (Triticum aestivum L.)-fallow rotation where wheat was sown with no-tillage. The tillage treatments were repeated in May 1993, and the plots were either fallowed or cropped by sowing either cowpea (Vigna unguiculata Walp.) or cotton. Cotton was sown with minim um tillage in 1994 and 1995 in all plots. Soil was sampled from the 0- 150 mm, 150-300 mm, 300-450 mm and 450-600 mm depths, and analyzed for organic carbon, dispersion index, soil resilience (a measure of the s elf-mulching ability of the soil), plastic limit, soil strength, pH, e xchangeable Ca, Mg, K and Na, and nitrate-N. Profile water content, nu trient uptake, numbers of soil invertebrates, cotton growth and lint y ield, and fibre quality were also quantified. Soil strength was lowest where intensively tilled continuous cotton had been sown, whereas in plots where minimum tillage and cotton-wheat-fallow rotation were comb ined soil fertility was best (indicated by lowest values of pH, exchan geable Na, exchangeable sodium percentage and dispersion, and highest values of organic C) and water extraction by cotton greatest during pe riods of reduced water availability. The latter was attributed to cott on utilizing stable pores with a high degree of pore-continuity create d by the root systems of preceding crops or associated macrofauna as ' by-pass channels' to avoid the restrictions of the soil matrix, thereb y facilitating rapid access to sub-soil water. Cotton growth reflected these differences such that vegetative and reproductive growth, nutri ent uptake and lint yield were greater and fibre quality superior wher ever minimum tillage had been imposed, and best in plots under minimum tilled cotton-wheat-fallow rotation. Composition and abundance of sur face-active and soil invertebrates were determined primarily by soil m icroclimate and pesticide application regime rather than by tillage an d cropping system. Ant numbers were lowest in intensively tilled plots whereas Collembola activity was limited to periods when the soil was moist. (C) 1997 Elsevier Science B.V.