Abstract
Phosphorus (P) is a nutrient of concern for its potential role in promoting nuisance growth in rivers. It can enter streams either through runoff or sub-surface flow pathways in dissolved or particulate form. The equilibrium phosphorus concentration (EPC0) is the point at which phosphorus in a stream is in equilibrium with the phosphorus in the bed sediment. It is thought that bed sediment acts as a control on P exchange, and is important for understanding effective P attenuation in waterways. By determining the EPC0 values for various sites within a stream, calculations can be made to assess whether the bed sediment is acting as a source or a sink of P in the waterway. At a larger scale, determining the sources of sediment to a stream can help to control how much P gets into a waterway in the first place. In Otago, previous studies have shown that phosphorus and sediment losses in overland flow are large. This study looks firstly at using the EPC0 to determine the controls on P in a small stream in Otago under baseflow conditions over a 12-month sampling period, and secondly at trying to identify sources of sediment to the stream. Sediment samples from an unnamed tributary of the Dow Stream on the AgResearch Invermay farm near Mosgiel, Otago, were analysed to calculate EPC0 and water samples were collected and measured to determine their P concentrations. Both wet and dry sediments were used for analysis, however there was too much variation in the wet sediment results to be used. Dried sediment results were used to infer patterns of seasonality but instream processes were not able to be determined. The dry sediment EPC0 values showed significant (P < 0.05) changes over the sampling period that were attributed to three stages, before and after a large flow event, and after grazing of a beet crop. Repeated small rain events combined with grazing were responsible for changes to bed sediment EPC0, not just large-scale rain events that cause sediment mobilisation as is established in literature. Source samples were also taken from transects surrounding the bed sediment sampling sites and the elemental profile of the bed sediments, as well as that of potential source sediments, were determined. A discriminant function analysis was used to determine predicted group membership of the bed sediments which also changed before and after grazing. The five source groups that were created did not capture all five of the bed sediment sites accurately, therefore although changes were observed the sources could not be accurately identified due to oversimplifying sampling using the transect method. Understanding both the equilibrium processes in stream and source areas externally are important for determining the lag effect of sediment bound P, entering a stream during storm events, and then contributing to the dissolved P load in the stream under base-flow conditions. Further research is still needed to link these in the tributary of Dow Stream.