Maximising the use of soil water by herbaceous species in the high rainfall zone of southern Australia: a review

The planting of deep-rooted pasture species, herbaceous shrubs, and trees has been widely recommended to reduce deep drainage and recharge to the groundwater in the high rainfall zone (HRZ). However, in more recent years, the value of perennial pastures to reduce recharge has been questioned in areas with >600 mm annual rainfall. Currently, pastures dominated by annual species with relatively low productivity occur across much of the HRZ where deep drainage is most likely contributing to recharge. This review outlines our current understanding of water use by various herbaceous species, and indicates ways in which their water use may be increased in the HRZ of southern Australia. To reduce deep drainage in the HRZ, the soil water deficit must be increased prior to the opening autumn rains. This will allow a greater storage of water before any potential deep drainage occurs. There are two ways that this can be achieved with the use of herbaceous species. Firstly, change to or encourage species that use more water annually. Although plants with deeper root systems including lucerne have the ability to dry the soil to depth, a combination of winter- and summer-active species, rotational grazing, and pasture spelling would extend the active growing season and soil water use of annual and perennial species. A second option is to increase the productivity of the pasture, as there is a direct link between growth and water use. For example, improving pasture productivity by 50%, say from 8 to 12 t dry matter/ha, could use (transpire) approximately 160 mm more water annually by a C 3 species, irrespective of evaporation from the soil surface or evaporative demand factors. This is supported by strong correlations between plant dry mass and water use among a wide range of C 3 and C 4 plants of diverse growth form and habitat. This relationship appears to have been overlooked in recent studies of various components of the soil water balance model, possibly due to limited and unreliable estimates of evapotranspiration (ET). An improved relationship between 'estimated' ET and measured dry matter production should improve the capability of the soil water balance model to predict deep drainage, which is primarily dependent on the ET. Ways to increase pasture productivity and soil water use include regular applications of fertiliser and lime, and better management of waterlogged and acidic soils in the HRZ. Summer-active native species may also be useful on soils where the