Changes in soil water content under annual- and perennial-based pasture systems in the wheatbelt of southern New South Wales

Abstract

Nine pasture treatments differing in species composition were monitored for changes in soil water content at a depth of 0.10–1.70 m, at 2 sites (Kamarah and Junee), in the wheatbelt of eastern Australia. Treatments containing perennial species, viz. lucerne (Medicago sativa L.), phalaris (Phalaris aquatica L.), cocksfoot (Dactylis glomerata L.), mixture (lucerne + phalaris + cocksfoot), wallaby grass (Austrodanthonia richardsonii Cashmore.), and lovegrass (Eragrostis curvula (Schrader) Nees.), were sown with subterranean clover (Trifolium subterraneum L.). In addition, 3 treatments based solely on annual species were examined: subterranean clover (sown by itself and kept weed-free with herbicides), annual (sown to subterranean clover but weed invasion not controlled), and serradella (Ornithopus compressus L.). The experiment was conducted from 1994–97 at the Junee site (annual average rainfall 550 mm/year) and from 1995–97 at the Kamarah site (annual average rainfall 450 mm per year). At the higher rainfall site (Junee), there were few differences among pasture types in soil water content to 0.70 m. Below 0.70 m the soil profile was drier under all the perennial swards than under the annual pasture treatments by the end of the 4-year pasture phase. At the drier Kamarah site, where the pasture phase was shorter due to an initial sowing failure, all the perennials, except cocksfoot, dried the profile below 1.05 m. At both sites, lucerne dried the 1.05–1.70 m section of the soil profile more rapidly than the other perennials, which apparently took longer to reach this depth. At the Junee site, the soil water deficit in May (SWD(MAY), defined as field capacity (mm) – stored soil water (mm) at the beginning of May) was largest in the phalaris, mixture, lucerne, and cocksfoot treatments (155–162 mm), whereas as under a pasture of subterranean clover alone, SWD(MAY) was only 89 mm. At the drier Kamarah site, the largest SWD(MAY) was created by the lovegrass (114 mm) and lucerne (107 mm) treatments. The cocksfoot and subterranean clover treatments created the smallest SWD(MAY) at this site, at 79 and 72 mm, respectively. The study showed that currently available C3 and C4 perennial grasses can be as effective as lucerne in drying the soil profile to 1.70 m in the 450–600 mm rainfall areas of the southern NSW wheatbelt, creating a dry soil buffer to reduce the risk of deep drainage during subsequent cropping phases. As the rate at which grasses dried the profile was slower than lucerne, pastures based on perennial grasses may have to be retained longer to achieve the same level of dewatering.