We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
A classification scheme for concentration-discharge relationships based on long-term low-frequency water quality data.
- Authors
Pohle, Ina; Glendell, Miriam; Baggaley, Nikki; Stutter, Marc
- Abstract
Riverine nutrient concentrations represent the integrated response to hydrological andbiogeochemical processes in the terrestrial phase of the catchment and in stream. Effectivemitigation measures against nutrient pollution require in-depth knowledge of thespatio-temporal controls on water quality. The analysis of concentration-discharge (c-Q)relationships allows investigating solute sources and pathways. The hysteresis behaviour inc-Q relationships is often analysed based on high-frequency data (event-based or continuoustime series in experimental catchments). We developed a classification scheme to describe hysteresis patterns in c-Qdynamics also for low frequency observations of concentrations (given sufficiently longrecords). Nine classes of c-Q relationships are distinguished as a combination of exportbehaviour (dilution, neutral, enrichment) and rotational pattern of the hysteresis(clockwise, no rotation, anti-clockwise). To that end, power-law functions are used todescribe the relationships between concentration and discharge. These relationshipsuse distinct parameter values for rising and falling hydrograph limb and for lowflows and high flows. The export behaviour is assessed based on the theoretical c-Qrelationships by testing whether concentrations decrease, do not change or increase withdischarge (Mann-Kendall test). The rotational pattern is determined by comparingconcentrations at the rising and the falling limb of the hydrograph (Kruskal-Wallistest). The classification scheme has been applied to daily stream-flow and monthly nutrientconcentration data of 45 Scottish catchments (time period: 1987-2016, catchment areabetween 55 and 4587 km2, covering in total around 50 % of mainland Scotland). For thesecatchments, similarity in terms of stream-flow variability and nutrient exports hasfurthermore been investigated by hierarchical cluster analysis of mean values and variabilityindices of runoff depth and nutrient concentrations. The classification of c-Q relationships often relates to catchment characteristics.Dilution behaviour of total organic nitrogen, ammonia and nitrate often occurs incatchments with high annual precipitation and low proportions of arable land. Areas withhigh proportions of (sub-)urban land often show dilution behaviour of nitrate, totaland soluble reactive phosphorous. Total and dissolved organic carbon mostly arecharacterized by enrichment behaviour especially in catchments with relatively lowannual runoff. Total and soluble reactive phosphorous mostly show clock-wiserotation, which is most pronounced in catchments with relatively high (sub-)urbanland and low wetland proportions. The clusters of stream-flow variability can beascribed to climatic conditions and topography, whereas land cover is the main controlfor concentration variability clusters. Furthermore, we will explore the extent towhich soil hydrological characteristics and soil organic horizons explain the c-Qrelationships. The classification scheme for c-Q relationships presented is transferable to other catchmentsand allows increasing the information content of low frequency data of concentrationsas typical for regulatory monitoring data. Understanding the catchment-specificcontrols on water quality can inform efficient pollution mitigation measures against,monitoring strategies, as well as water quality modelling and regionalization approaches.
- Subjects
SCOTLAND; WATER quality; HIERARCHICAL clustering (Cluster analysis); WATER quality management; DATA quality; WATERSHEDS; ARABLE land
- Publication
Geophysical Research Abstracts, 2019, Vol 21, p1
- ISSN
1029-7006
- Publication type
Article