N and P concentration-discharge relationships across a range of Waikato catchments

Abstract

Waikato Regional Council operates a river water quality monitoring programme where samples are taken monthly at 114 sites and analysed for concentrations (C) of a range of water quality parameters. Water flow (or discharge, D) is measured at or nearby 26 of these sites, which allows nutrient concentration–river discharge relationships (C-D relationships) to be established. The patterns of the C-D relationships were surprisingly similar across the region in spite of substantial differences in natural conditions, land use and the potential effect of point source discharges. Statistically highly significant (p ≤ 0.005) C-D relationships were found at nearly all sites (n=24–26) for total nitrogen (TN), nitrate nitrogen (NO₃-N) and total phosphorus (TP); all but two of which were positive. Ammonium nitrogen (NH₄-N) and dissolved reactive phosphorus (DRP) were less frequently correlated to discharge. While all significant correlations (n=18) were positive in the case of NH₄-N, 9 of the 14 significant correlations for DRP were negative. In spite of many C-D relationships being statistically highly significant, there is typically a wide spread in the data, resulting in substantial uncertainty if the equations are used for predictions. To evaluate to what extent these relationships provide information on the transfer pathways from the land surface to river monitoring sites, we stratified all concentration data into those from sampling dates when baseflow (BF) dominated discharge versus those when quickflow (QF) was dominant. As the positive C-D relationships imply, average baseflow concentrations of TN, NO₃-N and NH₄-N were lower than quickflow concentrations at the same site, whereas negative C-D relationships for TP or DRP were associated with higher concentrations of these solutes in baseflow compared with quickflow. As BF is largely due to discharge of (older) groundwater, while quickflow is predominantly due to (younger) near-surface flows, trends in these data reflect land management changes at different time scales. Deteriorating NO₃-N concentrations were predominantly the combined effect of BF and QF deteriorations, while TP trends, both positive and negative, were largely due to trends in BF. The prominent role of baseflow in determining these river concentration trends highlights the importance of understanding a nutrient’s mean residence time in the groundwater system of a particular catchment when trying to link a concentration change to a land management change.