Evaluating diffuse and point source phosphorus inputs to streams in a cold climate region using a load apportionment model
Introduction
Tributaries play a key role in the biogeochemical connectivity between terrestrial and aquatic ecosystems by controlling the routing and delivery of nutrients such as phosphorus (P) to downstream waterbodies. Knowledge of the timing of P release (seasonal or episodic versus year-round), mode of P delivery (point versus non-point) and within-river transformations of P (through interactions with sediments and river biota) is critical for predicting the magnitude of nutrient export and, in turn, the water quality (eutrophication, algal blooms, species richness) of downstream lakes (Sharpley et al., 1994, Edwards and Withers, 2007). However, identifying the timing of P release and its mode of delivery can be challenging, especially for rural landscapes, and can lead to management inaction when the sources of P cannot be identified or distinguished.
Targeting nutrient sources that will produce the greatest and most cost-effective improvement to water quality requires an understanding of the relative contributions of P inputs to streams. One approach used for source attribution of riverine nutrient loads is the export coefficient method. The method uses coefficients (derived from previous studies) to estimate nutrient loads associated various land cover types, livestock excretion and wastewater treatment (WWT) plant inputs (Johnes and Heathwaite, 1997, Bowes et al., 2005a, Palviainen et al., 2016). Whilst the method is relatively simple to apply and allows comparison of the relative contributions from diffuse and point sources, major drawbacks are the need for previously-derived coefficients relevant to the study catchment as well as current detailed data on land cover, livestock types and numbers, and population served by WWT plants. Although this approach has been used to assess source contributions to annual nutrient loads (Bowes et al., 2008), it cannot adequately infer seasonal sources of nutrient delivery (Burt and Johnes, 1997). This is a serious disadvantage as temporal resolution is required to determine nutrient inputs during summer when eutrophication is most likely to occur.
To overcome these drawbacks, P concentration-flow plots have been used as an integrated indicator of nutrient sources and delivery pathways (Godsey et al., 2009, Basu et al., 2011, Ali et al., 2017). Interpretation of concentration-flow plots is based on the observation that rivers receiving diffuse inputs have a tendency to show an increase in P concentration and load with increasing river flow (Bowes et al., 2009, Jarvie et al., 2010). In contrast, loadings from point sources, such as WWT plants, tend to be relatively constant throughout the year and are generally independent of river flow. Thus, in rivers that are point source dominated, the constant rate of P input results in concentrations that are highest at low flow and that decrease with increasing river flow due to dilution. The benefit of concentration-flow analyses is that they are entirely based on water quality and flow monitoring data that are often relatively easy to obtain. However, the interpretation of nutrient-flow plots to quantify diffuse versus point P inputs, especially seasonally, can be difficult unless these plots are fitted using statistical models (Bowes et al., 2009, Ali et al., 2017).
A load apportionment model (LAM) was developed by Bowes et al. (2008) and uses differences in the mode of P delivery to provide a simple and rapid method for the relative contribution of diffuse versus point-source inputs using only paired nutrient concentration - flow datasets (Jarvie et al., 2010, Bowes et al., 2011, Greene et al., 2011). The model has been successfully applied in temperate regions (Bowes et al., 2009, Jarvie et al., 2010, Halliday et al., 2015) but has never been applied to a cold climate region such as the Northern Great Plains (central Canada and northern U.S.A). Here, excessive nutrient loading as a result of human activity has led to eutrophication of downstream waterbodies, including proliferation of harmful algal blooms in the world’s 10th largest lake, Lake Winnipeg, Manitoba, Canada (McCullough et al., 2012, Bunting et al., 2016). Application of the LAM in cold climate regions, such as the Northern Great Plains, ought to provide an understanding of potential drivers of the region’s eutrophication issues. Such regions differ from mild temperate and tropical regions in that a sizable proportion of total annual precipitation falls as snow during winter, resulting in a hydrological regime that is dominated by snowmelt runoff (Intergovernmental Panel on Climate Change Assessment Reports, 2018). In addition, unlike mild temperate regions where human population density is often high and wastewater is discharged continuously from treatment facilities, settlements in cold regions tend to be less populated and wastewater inputs are often discrete, with effluent flow occurring over several weeks during summer.
The purpose of this research was to determine the contribution of ‘continuous’ (point) versus ‘flow-dependent’ (diffuse) sources of P (total, dissolved, and particulate fractions) to eight sub-watersheds in southern Manitoba, Canada. Small watersheds (<700 km2) were selected because they have a clear hydrochemical signal resulting from land use and background geology; whereas, water quality patterns of larger rivers are an integration of multiple upstream land uses and effluent discharges, which tend to obscure the influences of individual human activities (Jarvie et al., 2010). The present study used the LAM devised by Bowes et al. (2008) to apportion contributions between point and diffuse sources and determine the influence of seasonality on P inputs to prairie watersheds. Information resulting from application of the model to southern Manitoba streams will assist in identification of P sources and the most effective mitigation strategies to reduce P concentrations and eutrophication risk.
Section snippets
Site description
This study was conducted in the Red River Valley (RRV) of southern Manitoba, Canada (Fig. 1a). The RRV comprises the historical bed of glacial Lake Agassiz and is characterized by a wide flat plain of fine glacio-lacustrine soils (Yates et al., 2014). The region experiences a cold continental climate with cold winters and warm summers (mean temperatures of −9.3 °C from November–March and 19 °C from June–August; 1980–2010 records for Morden, MB; Government of Canada, 2018a, Government of Canada,
Human activities
In the RRV, land use during our study was dominated by crop (corn, flaxseed, grain, soybean) cultivation and livestock production. Crop cover as a portion of watershed area varied from 59 to 92% across all eight sub-watersheds, with the quantity of synthetic P applied to cropland varying by 5-fold (Table 1, Fig. 1b). Based on estimates of crop yield and average crop P content, the mass of P removed during crop harvest ranged from 155 to 731 t among sub-watersheds. Livestock density expressed as
Discussion
Modelling of phosphorus loads using concentration and flow data showed that the Load Apportionment Model (LAM; Bowes et al., 2008) produced, on average, realistic estimates of diffuse and point-source inputs for watersheds in a sparsely-populated cold-climate region of southern Manitoba, Canada. The LAM predicted P loads that agreed, on average, with observed loads; expected: observed ratios for eight study sites averaged 0.99–1.10 for four P fractions and three time periods. Moreover, the LAM
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
We thank Ross MacKay, Zoey Duggan, Jon Challis, Julie Anderson, and Alistair Brown for their contributions to field data collection. This project was funded by Environment and Climate Change Canada through the Lake Winnipeg Basin Initiative and a National Science Engineering Research Council (NSERC) Visiting Fellowship Award to K.J.R.
References (41)
- et al.
The relative contribution of sewage and diffuse phosphorus sources in the River Avon catchment, southern England: implications for nutrient management
Sci. Tot. Environ.
(2005) - et al.
Modelling of phosphorus inputs to rivers from diffuse and point sources
Sci. Total Environ.
(2008) - et al.
Changes in point and diffuse source phosphorus inputs to the River Frome (Dorset, UK) from 1966 to 2006
Sci. Total Environ.
(2009) - et al.
Changes in water quality of the River Frome (UK) from 1965 to 2009: is phosphorus mitigation finally working?
Sci. Total Environ.
(2011) - et al.
Identifying priorities for nutrient mitigation using river concentration–flow relationships: the Thames basin, UK
J. Hydrol.
(2014) - et al.
Presence and hazards of nutrients and emerging organic micropollutants from sewage lagoon discharges into Dead Horse Creek, Manitoba, Canada
Sci. Total Environ.
(2013) - et al.
An evaluation of catchment-scale phosphorus mitigation using load apportionment modelling
Sci. Total Environ.
(2011) - et al.
Streamwater phosphorus and nitrogen across a gradient in rural–agricultural land use intensity
Agric. Ecosyst. Environ.
(2010) - et al.
Hydrological forcing of a recent trophic surge in Lake Winnipeg
J. Great Lakes Res.
(2012) - et al.
Quantifying seasonal variation in total phosphorus and nitrogen from prairie streams in the Red River Basin, Manitoba Canada
Sci. Total Environ.
(2017)
Hydrological variability affects nitrogen and phosphorus export from streams of the Northern Great Plains
J. Hydrol. Reg. Stud.
A hydrodynamic approach to modeling phosphorus distribution in Lake Erie
J. Great Lakes Res.
Nitrogen speciation and phosphorus fractionation dynamics in a lowland Chalk catchment
Sci. Total Environ.
Estimating nutrient production from human activities in sub-catchments of the Red River, Manitoba
J. Great Lakes Res.
Phosphorus export dynamics and hydrobiogeochemical controls across gradients of scale, topography and human impact
Hydrol. Process.
Manure Production and Characteristics
Standard Methods for the Examination of Water and Wastewater
Hydrologic and biogeochemical functioning of intensively managed catchments: a synthesis of top-down analyses
Water Resour. Res.
Increased variability and sudden ecosystem state change in Lake Winnipeg, Canada, caused by 20th century agriculture
Limnol. Oceanogr.
Managing water quality in agricultural catchments
Trans. Inst. Br. Geogr.
Cited by (11)
Techno-economic and life cycle analysis of circular phosphorus systems in agriculture
2023, Science of the Total EnvironmentDevelopment of remote sensing algorithm for total phosphorus concentration in eutrophic lakes: Conventional or machine learning?
2022, Water ResearchCitation Excerpt :The net reception of TP in Lake Taihu exceeds its discharge (Wang et al., 2019b), and the areas surrounding the lake have dense population and well-developed industries. P in domestic sewage and industrial wastewater mainly exists in the dissolved form (Carlson et al., 2013; Rattan et al., 2021). The discharge of the domestic sewage and industrial wastewater causes TP enrichment in the upstream lake area, and the spatial distribution of TP concentration is heterogeneous.
Response of phosphorus in estuaries to soil erosion – A new insight into sediment phosphorus fractions and sources
2021, CatenaCitation Excerpt :Ni et al. (2020) believe that with the implementation of the ‘‘Zero O’ clock Action” in 1999, strict wastewater discharge standards were promulgated, that effectively prevented an increase in P loads of industrial wastewater and domestic sewage. P emitted from point sources mainly exists in dissolved form (Carlson et al., 2013; Rattan et al., 2021), and the adsorption capacity of sediments for DP is mainly affected by water quality. In Lake Chaohu, dissolved oxygen (DO) and pH are the main factors that influence the Fe-P and Al-P content in sediments (Table 2).
A combined GIS-MCDA approach to prioritize stream water quality interventions, based on the contamination risk and intervention complexity
2021, Science of the Total EnvironmentCitation Excerpt :In these stations, wastewaters are treated to release the effluents into rivers under legislated thresholds of contaminant concentrations, which should be environmentally sustainable (Pérez Andrés et al., 2019). Still, the released effluents decrease the water quality of rivers which defines WWTP discharge sites as point source pressures (Rattan et al., 2020). Also, concerning hydric resources, other types of pollution sources are threatening, particularly diffuse pressures (Mockler et al., 2017).
Review: The application of source analysis methods in tracing urban non-point source pollution: categorization, hotspots, and future prospects
2024, Environmental Science and Pollution ResearchTemporal and spatial dynamics of nitrogen and phosphorus flux in the Dongjiang River mainstream(2014—2019)
2024, Huanjing Kexue Xuebao/Acta Scientiae Circumstantiae