High-frequency nutrient monitoring to infer seasonal patterns in catchment source availability, mobilisation and delivery.

To explore the value of high-frequency monitoring to characterise and explain riverine nutrient concentration dynamics, total phosphorus (TP), reactive phosphorus (RP), ammonium (NH4-N) and nitrate (NO3-N) concentrations were measured hourly over a 2-year period in the Duck River, in north-western Tasmania, Australia, draining a 369-km(2) mixed land use catchment area. River discharge was observed at the same location and frequency, spanning a wide range of hydrological conditions. Nutrient concentrations changed rapidly and were higher than previously observed. Maximum nutrient concentrations were 2,577 μg L(-1) TP, 1,572 μg L(-1) RP, 972 μg L(-1) NH₄-N and 1,983 μg L(-1) NO₃-N, respectively. Different nutrient response patterns were evident at seasonal, individual event and diurnal time scales-patterns that had gone largely undetected in previous less frequent water quality sampling. Interpretation of these patterns in terms of nutrient source availability, mobilisation and delivery to the stream allowed the development of a conceptual model of catchment nutrient dynamics. Functional stages of nutrient release were identified for the Duck River catchment and were supported by a cluster analysis which confirmed the similarities and differences in nutrient responses caused by the sequence of hydrologic events: (1) a build-up of nutrients during periods with low hydrologic activity, (2) flushing of readily available nutrient sources at the onset of the high flow period, followed by (3) a switch from transport to supply limitation, (4) the accessibility of new nutrient sources with increasing catchment wetness and hydrologic connectivity and (5) high nutrient spikes occurring when new sources become available that are easily mobilised with quickly re-established hydrologic connectivity. Diurnal variations that could be influenced by riverine processes and/or localised point sources were also identified as part of stage (1) and during late recession of some of the winter high flow events. Illustrated by examples from the Duck River study, we demonstrate that the use of high-frequency monitoring to identify and characterise functional stages of catchment nutrient release is a constructive approach for informing and supporting catchment management and future nutrient monitoring strategies.