Runoff and leaching of dissolved phosphorus in streams from a rainfed mixed cropping and grazing catchment under a Mediterranean climate in Australia

Abstract

Managing phosphorus (P) is a global priority for environmental water quality due to P lost from agricultural land through leaching, runoff and subsurface flow. In Western Australia (WA), following decades of P fertiliser application to crops and pastures in low rainfall regions, questions have been raised about this region's contribution to environmental P loss. This study was conducted on the Fitzgerald River catchment in the south Western Australia (WA) with mixed cropping and grazing land uses and a Mediterranean climate with low mean rainfall (~350 mm yr−1). Phosphorus forms were monitored continuously over a three-year period in five separate streams, each draining a defined sub-catchment. The P concentrations in streams consistently exceeded Australian and New Zealand Environment Conservation Council (ANZECC) trigger values throughout the monitoring period. Of the measured total P concentration, ~75% was dissolved P (DRP; <0.45 μm) and 80% of that fraction was in the filterable reactive form (FRP). These water quality measurements and other independent soil investigations at this site, suggest that transport of dissolved P rather than erosion of sediment-bound P was dominant in this environment. Based on extractable soil P (Colwell P) and the P buffering index (PBI), predicted concentrations of dissolved reactive P (DRP) in soil solution in topsoils (0–10 cm) across this catchment, generally exceeded ANZECC's values of 0.07 mg PL−1. The level of exceedance was spatially variable. Streams draining areas with the lowest predicted DRP concentrations also had the lowest measured FRP concentrations. Elsewhere stream water FRP concentrations depended on both DRP concentration and the PBI of the land being drained. Our findings suggest that deployment of practices that physically filter runoff, for example riparian vegetation, would be ineffective in restricting P transport into stream in this environment. This conclusion is consistent with previous findings of the ineffectiveness of riparian buffers on coarse textured sandy soils in higher rainfall areas of southwest WA. A reduction in DRP losses without yield loss could be achieved by following evidence-based fertiliser advice from soil testing to limit losses of legacy P”.