The Fitzroy and Burdekin basins are Queensland’s largest coastal catchments, and both drain directly to the Great Barrier Reef. Greater than 80% of the catchment area in each basin is impacted by grazing, on both native and improved pastures. Despite substantial historical plantings of legume pasture species and ongoing programs of pasture improvement incorporating legumes, no information exists on the potential for increased exports of nitrogen in runoff from these systems to the Great Barrier Reef. This study sought to determine if broad-scale plantings of pasture legumes, particularly leucaena and butterfly pea, pose a risk to Great Barrier Reef water quality by increasing loads of nitrogen in runoff waters compared to grass only pastures or the virgin brigalow scrub landscape.
Comparison of pasture type effects on water quality at the paddock scale was undertaken in the Fitzroy Basin using a paired, calibrated catchment study approach combined with simple regression based modelling developed from the long-term Brigalow Catchment Study. This work was complimented by plot scale rainfall simulation experiments in the Burdekin and Burnett-Mary Basins.
At the paddock scale (12 to 24 ha) in the hydrological years 2010 and 2011, loads of total suspended solids and nitrogen in runoff from grass only and leucaena pastures were typically lower than or equal to loads from virgin brigalow scrub. Phosphorus loads from the pastures showed the opposite trend, being typically equal to or higher than loads from virgin brigalow scrub. In the same period, all parameters from butterfly pea ley pasture were equal to or higher than loads from virgin brigalow scrub. High event mean concentrations did not necessarily equate to high loads. During 2012, no runoff occurred from the virgin brigalow scrub, so loads from all catchments were an absolute increase compared with their pre-European condition.
Plot scale rainfall simulation conducted in the Burdekin and Burnett-Mary Basins showed that runoff in the late dry season typically had higher loads of total nutrients than runoff in the late-wet season. No significant interaction between pasture type and season for total suspended solids was observed. Rainfall simulations at the plot scale in the Burdekin and Burnett-Mary Basins also indicated that results from paddock scale catchment studies at the long-term Brigalow Catchment Study in the Fitzroy Basin are applicable to other grazing areas in the Brigalow Belt Bioregion.
Modelling flow and water quality from cropping and grazed buffel grass pasture between 1984 and 2012 showed similar trends to the rainfall simulation studies. Using virgin brigalow scrub as a reference, cropping exported more dissolved inorganic nitrogen and phosphorus, total phosphorus and total suspended solids; whilst grazing exported less total nitrogen and dissolved inorganic nitrogen, but more total and dissolved inorganic phosphorus and more total suspended solids. Cropping exports were always greater than grazing exports.
Within the grazing landscape, soil and pasture nutrient concentrations exhibited high variability and limited temporal response within the study. With no discernible period of potential high nutrient availability during the year, soil and pasture management should focus on minimising runoff, rather than manipulation of the natural nutrient cycle to reduce risks to water quality.
Newly planted legume based ley pastures pose a risk to water quality as they contribute higher nutrient loads than grass only pasture systems, established leucaena pastures, and the virgin brigalow scrub landscape representative of the environment in its pre-European condition. However, they do reduce total suspended sediment loads compared with the cropping system that they replaced. Dissolved inorganic nitrogen loads from well-established leucaena exceeded those from grass, indicating a potential risk to water quality from the legume component of permanent pasture. This may have implications for parts of northern Australia, such as the Burdekin Basin, with large areas of naturalised Stylosanthes spp. pasture. These findings have been synthesised into a series of values and trends suitable for use in model development and validation to further refine estimations of the impact of changed land use, management and the adoption of leguminous pastures on water quality.