Agriculture is a major source of nitrogen (N) and phosphorus (P) in freshwater ecosystems, and different management strategies exist to reduce farmland nutrient losses and thus mitigate freshwater eutrophication. The importance of agricultural sources of N and P as drivers of water quality is known to vary spatially, but quantification of the relative importance of the nutrient sources shaping this variability remains challenging, especially with reference to inputs from waste water treatment works. Addressing this knowledge gap is key for targeting management strategies to where they are likely to have the greatest effect. To advance our understanding in this area, this study assesses the impact of population density as a driver of the relative importance of agricultural land use for predicting mean Total Oxidised Nitrogen (TON) and Reactive Phosphorus (RP) concentrations in rivers in England, using two different data-driven, statistical approaches: a generalised linear model and random forest. Our results show that agricultural N and P sources dominate in catchments with low population density, where stream water concentrations are lower and waste water treatment works are numerous, but smaller in terms of the population equivalent served. Agricultural N and P sources are not important predictors of N and P in catchments with high population density, where contributions from waste water treatment works dominate. These results require cautious interpretation, as model validation outcomes show that high TON and RP concentrations are consistently underpredicted. Altogether, our results suggest that the relative contribution of agricultural sources may be overestimated in densely populated catchments, relative to point sources from waste water treatment works, and that management strategies to reduce the contribution of agriculture to N and P in rivers may be better targeted towards catchments with lower population density, as this is where agricultural land use is the primary source of N and P.
Keywords: England; Spatially targeted management; Statistical techniques; Waste water; Water quality.
Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.