ArticlePDF Available

Then and now: Revisiting nutrient export in agricultural watersheds within southern Ontario's lower Great Lakes basin

Authors:
  • Ontario Ministry of the Environment Conservation and Parks
  • Ontario Ministry of the Environment Conservation and Parks

Abstract and Figures

An enhanced understanding of nonpoint source (NPS) nutrient export to the lower Great Lakes is needed to inform land use and land management decisions within southern Ontario. However, this understanding is limited by a lack of long-term, temporally-intensive monitoring. To address this knowledge gap, we revisit six agriculturally-dominated subwatersheds in southern Ontario, which were intensively studied during the mid-1970s, to assess changes in hydrology and NPS nutrient contributions. We compared 1975-1977 to 2016-2018 stream runoff, nutrient export (kg/day•km 2), and flow-weighted mean concentrations (FWMCs) of total phosphorus (TP), total dissolved phosphorus (TDP), total nitrogen (TN), nitrates (NO 3-+NO 2-) and Total Kjeldahl Nitrogen (TKN). Relative to the 1970s, runoff increased at three of six watersheds (by~20-35%) while TP and TDP export increased at five watersheds (by~50-125%). The increases in TP and TDP FWMCs were lower relative to phosphorus export changes at the three watersheds with increased runoff, suggesting that hydrology is an important driver of phosphorus export at these sites. Interestingly, export of TN and nitrates increased while TKN export decreased at most watersheds. We further note a shift in the timing of nutrient export at most sites, with~40-70% of export now occurring during the winter and fall seasons whereas~40-85% of past export occurred during spring and summer. These findings support an enhanced importance of non-growing season nutrient export from agricultural watersheds since the mid-1970s and stresses the need for targeted best management practices specific to the fall and winter seasons.
Content may be subject to copyright.
Then and now: Revisiting nutrient export in agricultural watersheds
within southern Ontario’s lower Great Lakes basin
Clare Nelligan
a,
, Ryan J. Sorichetti
b
, Meguel Yousif
c
, Janis L. Thomas
b
, Christopher C. Wellen
a
,
Christopher T. Parsons
d
, Mohamed N. Mohamed
b
a
Department of Geography and Environmental Studies, Ryerson University, 350 Victoria St, Toronto M5B 2K3, Canada
b
Ontario Ministry of the Environment, Conservation and Parks, 125 Resources Rd, Toronto M9P 3V6, Canada
c
Department of Chemical Engineering, University of Waterloo, Waterloo N2L 3G1, Canada
d
Watershed Hydrology and Ecology Research Division, Environment and Climate Change Canada, 867 Lakeshore Rd, Burlington L7S 1A1, Canada
article info
Article history:
Received 4 November 2020
Accepted 26 July 2021
Available online 25 August 2021
Communicated by Mary Anne Evans
Keywords:
Agricultural watersheds
Stream nutrient export
Runoff
Nonpoint source nutrients
Nutrient management
MWNS
abstract
An enhanced understanding of nonpoint source (NPS) nutrient export to the lower Great Lakes is needed
to inform land use and land management decisions within southern Ontario. However, this understand-
ing is limited by a lack of long-term, temporally-intensive monitoring. To address this knowledge gap, we
revisit six agriculturally-dominated subwatersheds in southern Ontario, which were intensively studied
during the mid-1970s, to assess changes in hydrology and NPS nutrient contributions. We compared
1975–1977 to 2016–2018 stream runoff, nutrient export (kg/daykm
2
), and flow-weighted mean concen-
trations (FWMCs) of total phosphorus (TP), total dissolved phosphorus (TDP), total nitrogen (TN), nitrates
(NO
3
+NO
2
) and Total Kjeldahl Nitrogen (TKN). Relative to the 1970s, runoff increased at three of six
watersheds (by ~20–35%) while TP and TDP export increased at five watersheds (by ~50–125%). The
increases in TP and TDP FWMCs were lower relative to phosphorus export changes at the three water-
sheds with increased runoff, suggesting that hydrology is an important driver of phosphorus export at
these sites. Interestingly, export of TN and nitrates increased while TKN export decreased at most water-
sheds. We further note a shift in the timing of nutrient export at most sites, with ~40–70% of export now
occurring during the winter and fall seasons whereas ~40–85% of past export occurred during spring and
summer. These findings support an enhanced importance of non-growing season nutrient export from
agricultural watersheds since the mid-1970s and stresses the need for targeted best management prac-
tices specific to the fall and winter seasons.
Ó2021 The Authors. Published by Elsevier B.V. on behalf of International Association for Great Lakes
Research. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).
Introduction
Improved water quality as a result of nutrient management was
observed in the Laurentian Great Lakes (LGL) basin following the
1972 Canada-United States Great Lakes Water Quality Agreement
(GLWQA; De Pinto et al., 1986). The GLWQA established water
quality objectives and led to the creation of programs aimed pri-
marily at reducing industrial and municipal point-sources of phos-
phorus (P). Coupled with a shift to low-phosphate detergents, the
years following the GLWQA (to approximately the late-1980s)
were marked by decreases in chlorophyll-aconcentrations (Dove
and Chapra, 2015), algal biomass (Kane et al., 2014), nuisance algae
(Painter and Kamaitis, 1987) and oxygen depletion rates (Bertram,
1993), particularly in Lake Erie. Recommendations to manage non-
point sources (NPS) of nutrients were published during the late-
1970s (Agriculture Canada et al. 1978) with subsequent manage-
ment efforts focused on reducing soil erosion and associated P
losses (Sharpley and Smith 1991). However, the efficacy of these
NPS reduction efforts was not rigorously documented and they
occurred simultaneously with decreases in point source pollution,
increases in agricultural production and regional climate warming.
Improvement in nutrient management in the LGL basin remains
critical, particularly in response to the resurgence of algal blooms
and anoxia in Lake Erie (e.g., Mohamed et al. 2019; Scavia et al.,
2014; Stumpf et al., 2012; Watson et al., 2016), and nuisance ben-
thic algae in the Great Lakes nearshore (e.g., Auer et al., 2010;
Howell, 2018).
Agricultural land use represents a significant source of NPS
nutrients to the lower Great Lakes (Robertson and Saad, 2011)
https://doi.org/10.1016/j.jglr.2021.08.010
0380-1330/Ó2021 The Authors. Published by Elsevier B.V. on behalf of International Association for Great Lakes Research.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Corresponding author.
E-mail address: clarenelligan@gmail.com (C. Nelligan).
Journal of Great Lakes Research 47 (2021) 1689–1701
Contents lists available at ScienceDirect
Journal of Great Lakes Research
journal homepage: www.elsevier.com/locate/ijglr
and is the largest contributor of NPS P to Lake Erie (Maccoux et al.
2016). Nutrients applied to the landscape during agricultural activ-
ity can be lost to surrounding surface waters through a variety of
pathways including subsurface drainage (King et al. 2015, Macrae
et al. 2007). Specifically, tile drainage has been linked to increased
contributions of soluble reactive P to surface waters where farming
practices require minimal or no tillage (e.g., Kelly et al., 2019;
Williams et al., 2016), where the water table is high (i.e., reducing
conditions can facilitate P mobility from soils; reviewed in King
et al., 2015), and where there is increased connectivity between
soils and tiles (either through preferential flow paths or macrop-
ores; e.g., Geohring et al., 2001). Agricultural subsurface drainage
can be a dominant pathway for runoff resulting in the edge-of-
field acting as a contributor of nutrients into aquatic ecosystems
(Plach et al. 2019). While best management practices (BMPs)
including crop and fertilizer management, conservation tillage,
cover crops and filter strips aim to control NPS loadings (Richards
et al. 2002, Sharpley et al. 2012), the success of BMPs is variable
(Bosch et al. 2014), and it has been suggested that some BMPs
may even enhance the movement of soluble nutrient fractions to
streams (Jarvie et al. 2017). Further, most studies evaluating the
efficacy of BMPs empirically analyse export at the edge-of-field
and do not consider the effects of BMPs on in-stream physical
and biogeochemical processes (Sharpley and Smith 1990). There-
fore, empirical evidence for water quality improvements and
decreases to nutrient export as a result of BMPs at the watershed
scale is elusive (Rittenburg et al., 2015; Tomer and Locke, 2011)
and often based on modeling rather than field observations
(Lintern et al., 2020). A better understanding of agricultural NPS
nutrient loads, with respect to nutrient mobility and speciation
at the watershed scale, is therefore needed to aid in informing
effective nutrient management actions.
The relationship of land use and land management on NPS
nutrient export is further complicated by regional climate warm-
ing, particularly with respect to the influence of climate on stream
discharge. In Ontario, increases in winter air temperatures, rainfall
and annual freeze–thaw days and decreases in the number of days
with snowfall have occurred from 1900 to 2016 (Vincent et al.,
2018). The consequence is a change in the amount (Norton et al.,
2019; Stow et al., 2015) and seasonality (Champagne et al., 2019)
of stream discharge, and as a result nutrient transport (Kelly
et al., 2019; Stow et al., 2015). Specifically, a study of four southern
Ontario watersheds identified a >40% increase in winter stream-
flow with corresponding decreases in April streamflow (due to
more high-flow events during the month of January; Champagne
et al., 2019). More frequent and extreme precipitation events have
also been projected to increase across North America (Kirchmeier-
Young and Zhang, 2020), which can disproportionately influence
NPS nutrient loads as a few high-discharge events can contribute
a large portion of the annual nutrient load in some systems
(Long et al., 2015; Macrae et al., 2007). Thus, it is important to con-
sider the influence of climate change on NPS nutrient export when
developing management strategies.
Data from long-term monitoring programs are needed to detect
environmental change and evaluate management efforts. However,
these datasets are rare and when they are available, sampling fre-
quency and design often limits their utility for calculating nutrient
export (e.g., Wellen et al. 2020). Further, methodological changes
over time can make comparisons between past and present condi-
tions challenging. In the mid-1970s, the Pollution from Land Use
Activities Reference Group (PLUARG, 1974–1978) study was estab-
lished by the International Joint Commission to assess NPS pollu-
tion to the Great Lakes and recommend actions to limit these
inputs (Sonzogni et al., 1980). The PLUARG study was considered
one of the most intensive investigations of NPS pollution during
its time, with the aim of developing relationships among land
use, land features, land management and nutrient loads
(Sonzogni et al. 1980). Forty years later, the Ontario Ministry of
the Environment, Conservation and Parks (MECP) re-visited many
of the goals and objectives of the PLUARG study by developing
the Multi-Watershed Nutrient Study (MWNS, 2015–2020). The
MWNS is listed in the Canada-Ontario Lake Erie Action Plan as a
priority research action to better understand nutrient dynamics
in the Lake Erie basin (ECCC and OMOECC, 2018).
In this study, we present a then-and-now comparison of agri-
cultural NPS nutrient export in southern Ontario relative to
40 years ago. Using two robust datasets, the goals of this investiga-
tion were to: (1) Quantify changes in the direction, magnitude and
seasonality of P and nitrogen (N) export; and (2) assess the poten-
tial influence of changes in hydrology and climate on nutrient
export. We also provide an evaluation of the PLUARG and MWNS
methods, and discuss important data caveats, limitations and cor-
rective measures taken to ensure that nutrient export represented
in both studies are comparable. Understanding long-term (deca-
dal) changes, such as those presented in this study, are essential
to developing effective management strategies across the diverse
landscape of southern Ontario.
Methods
Study site selection and description
The 11 agricultural watersheds studied during PLUARG (1975–
1977) were small (20–70 km
2
) and selected to be representative of
the various agricultural landscapes of southern Ontario
(Agriculture Canada et al., 1978; Sonzogni et al. 1980). The MWNS
also explored 11 agricultural watersheds in southern Ontario;
however, sites were selected to span a nutrient source and trans-
port gradient (Rosamond et al. 2018). Five PLUARG watersheds
were excluded from the MWNS due to new nutrient point sources
(e.g., greenhouses, sewage treatment plants, fish farms) or urban
land-uses that now exist within the watershed; therefore, six
watersheds were common to both studies (Big Creek, Little Ausable
River, Nissouri Creek, North Creek, North Maitland River and Veni-
son Creek; Fig. 1). Because these sites were selected to be inclusive
of the diverse agricultural landscapes of southern Ontario, sites
varied with respect to physical watershed characteristics and the
type and extent of agricultural activity (Table 1; Electronic Supple-
mentary Material (ESM) Table S1). Most MWNS sampling sites
were at or close to the known PLUARG stations; however, this
was not always possible as each station required safe site access,
the ability to install utilities infrastructure (e.g., electricity) for
automated sampling equipment and telemetry, and an appropriate
section of stream channel for flow gauging. All MWNS stations
were <1.5 km from the PLUARG station, with the exception of Veni-
son Creek which was moved ~6 km upstream to avoid the influence
of a fish farm that discharges into the stream. Variation in the
watershed delineations between PLUARG and the MWNS also
occurred due to methodology (i.e., hand-mapping of watershed
boundaries during PLUARG vs. delineation using GIS tools during
MWNS) and physical changes to the watersheds between the study
periods (e.g., changes to drainage networks and topography). As
such, the watershed areas between the two studies did exhibit
some differences (Fig. 1;Table 1).
Data used to characterize the land use and land management
during the 1970’s were taken from a report on watershed agricul-
tural activities generated by the PLUARG project (Frank and Ripley.
1977). These data include detailed surveys of crops grown, animal
populations, and tile drainage prevalence and densities (summa-
rized in Table 1). Land use data were collected for the current per-
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1690
iod through interviews with farmers in a project led by one of the
authors (CW) and is available at https://www.agri-model.com/.
Regional physiography and climate
All study watersheds are located within the Mixedwood Plains
ecozone of southern Ontario. The fertile soils found in this region
are underlain by limestone, sandstone and shale bedrock and
diverse lacustrine and glaciofluvial landforms (Crins et al., 2009).
Regional climate is highly influenced by the Great Lakes with mean
July annual temperatures and annual precipitation ranging from 18
to 22 °C and 720 to 1000 mm, respectively (Crins et al, 2009).
Meteorological data from Hamilton, Ontario (near the North Creek
watershed) were used to characterize regional climate conditions
during both study periods. The Hamilton A meteorological station
(ECCC, WMO ID: 71263) was selected because it is centrally located
among the study watersheds and contains the longest continuous
data record for rain and snowfall measurements within the study
region. Regional air temperature, precipitation, snowfall and freez-
ing degree days (number of days the mean daily air temperature
was < 0 °C) are presented to compare meteorological conditions
during the PLUARG study and MWNS (Fig. 2;Table 2). Climate data
were accessed from the Government of Canada Historical Climate
Data online archives (GOC, 2020).
Stream discharge data collection and preparation
Although continuous discharge (calculated from level record-
ings and rating curves) was recorded during the PLUARG study,
only daily discharge data were available from the printed PLUARG
reports (OMOE, 1979). A discrete discharge measurement was also
recorded at the same time as each water chemistry sample and is
included in our analysis. MWNS water level and discharge are
available in five-minute intervals from 2014 to 2020 (collected
and made publicly accessible by WSC; station and rating curve
details are presented in ESM Table S2). When application of the
WSC rating curve resulted in a negative discharge estimate (i.e.,
during extreme low-flow conditions), the preceding positive dis-
charge value was substituted.
To make comparisons between the PLUARG study and the
MWNS stream discharge data, total stream discharge was calcu-
lated at each site over the entire two-year period for each study.
The MWNS five-minute discharge data were averaged to produce
mean daily discharge values to match the daily discharge data
available in the PLUARG reports (OMOE, 1979). If there were any
dates where discharge data were missing (i.e., due to power
outages or sampling gaps), the seasonal mean was substituted so
that every day over the two-year period was assigned a stream dis-
charge value. Total stream discharge and seasonal stream dis-
Fig. 1. Locations of the six Pollution from Land Use Activities Reference Group (PLUARG) study and Multi-Watershed Nutrient Study (MWNS) watersheds and monitoring
stations in southern Ontario. Insets depict differences in the watershed area for PLUARG study (hatched lines) and the MWNS (orange fill). (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1691
charge values were generated by summing the mean daily dis-
charge for all dates and seasons, respectively. To account for differ-
ences in watershed area, total and daily stream discharge were
converted to total and daily runoff (mm).
Water chemistry sample collection and data preparation
During the PLUARG study, water samples were collected
weekly, with twice-weekly samples collected during high-flow
periods. Sampling was conducted both manually and through the
use of automated samplers. Depth-integrated grab samples were
collected either by hand or through lowering a weighted bottle
on a rope. The automatic samplers (CAE Aircraft) used in the
PLUARG study were installed at the Big Creek, Little Ausable River
and Nissouri Creek sites. Samples were analyzed for a variety of
water quality parameters including TP, TDP, filtered nitrates
(NO
3
+NO
2
; hereafter referred to as nitrates) and TKN. TDP and
nitrates were filtered prior to analysis using a glass fiber filter
(1.5 mm pore size). All analyses were performed at the Ontario Min-
istry of the Environment laboratories in Toronto and London,
Ontario (Agriculture Canada et al. 1975).
A temperature-controlled automated sampler (Teledyne ISCO
Model 6712FR) was used to capture a range of flow events at each
MWNS site. The ISCO automated sampler contains 24 bottles that
were progressively filled over storm events or other high-flow
periods, including snowmelt. For each flow event, the sample col-
lection frequency was dependent on the duration of the event, typ-
ically collecting a sample every four hours (ESM Fig. S1). After each
event, and upon return of stream water level to antecedent condi-
tions, a subset of samples were selected for analysis so that sam-
ples spanned the hydrograph for the whole event. Grab samples
were also collected over the study period, primarily during antece-
dent conditions, and were typically pumped directly from the ISCO
automated sampler or obtained using a sampling pole during sum-
mer low-flow conditions. An effort was made to collect one grab
sample during antecedent conditions prior to each sampling event
(representing on average ~27% of data collected). Whole-water
samples were analyzed for TP and total nitrogen (TN). A portion
of the ISCO sample was then filtered through a 0.45 mm nylon filter
to be analyzed for total dissolved phosphorus (TDP) and nitrates.
Samples were shipped on ice to the Ontario Ministry of the Envi-
ronment, Conservation and Parks (OMECP) Laboratory Services
Branch (LaSB; Toronto, ON) and analyzed using standard methods
outlined in Chow et al. (2010). Analyses for TP and TDP were con-
ducted at the OMECP Dorset Environmental Sciences Centre (DESC;
Dorset, ON) using a low-level P colourimetry method described in
OMOE (2011).
Method detection limits (MDLs), instrumentation and proce-
dure for MWNS and PLUARG TP, TDP, TN, nitrates and TKN analy-
ses are presented in ESM Table S3. Nitrogen (N) MDLs were 36 and
25 mg/L for PLUARG and MWNS, respectively, and P MDLs were
6.6 mg/L and 0.02 mg/L for PLUARG and MWNS, respectively.
Increased sensitivity for P levels during the MWNS was coincident
with the implementation of the acid-heat digestion and low-P level
spectrophotometric method and instrumentation. Samples that
typically fall around P MDLs are coincident with low stream flow
samples in both studies, while event and high flow samples had
concentrations 10–100 MDLs. Despite a thorough evaluation of
the equivalency of PLUARG and MWNS methods (to minimize
inherent bias), differences in instrumentation and sample handling
may add uncertainty to the presented findings.
Nitrates data were unavailable from June 2016 to January 2017
(with only sparse filtered nitrate samples collected between Jan-
uary to June 2017). As a result, nitrates were estimated using a lin-
ear regression of the filtered and unfiltered nitrates data (which
was available over the entire five-year MWNS dataset). For each
Table 1
Physical and land use characteristics for each study watershed. Columns denoted with an ‘‘M” represent the Multi-Watershed Nutrient Study (MWNS) data and column denoted with a ‘‘P” represent thePollution from Land Use Activities
Reference Group (PLUARG) study data. Soil type and PLUARG tile drainage and spacing are documented in Frank and Ripley (1977).
Site Soil Type Average Slope (m/
km)
Watershed
Area (km
2
)
Tile
Drainage
Density (%)
Horizontal
Spacing (m)
Dominant Crop Types Livestock Type Livestock
Density
(animal
units/ha)
PMP MP MP M P M PM
Big Clay 0.34 43.65 55.51 80 96 8 10 11 Cash crop, soybean, wheat,
corn
Cash crop, soybean Few Swine 0.08 0.26
Little Ausable Clay Loam 1.25 58.10 63.86 50 98 15 14 Cash corn, beans, grains,
pasture
Corn, soybean, wheat Beef, dairy, swine Beef, dairy 0.48 1.14
Nissouri Loam/Silt
Loam
1.98 30.60 31.30 75–
100
99 24 14 Pasture, mixed grains, corn Corn, soybean, wheat,
hay
Beef, dairy,
poultry
Beef, dairy,
swine
0.77 3.08
North Clay 0.78 31.76 36.55 0 31 N/A 13.5 Corn, pasture Soybean, wheat, pasture Beef, dairy Beef, poultry 0.61 0.26
North
Maitland
Loam/Silt
Loam
2.7 55.85 58.78 25 78 Random 13.5 Mixed grains, corn, pasture Corn, soybean, wheat Beef, dairy, swine Beef, dairy 0.51 0.53
Venison Sand 0.96 80.98 44.17 0–25 80 N/A 18 Tobacco, cash crops Corn, soybean Few Few 0.04 0
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1692
site, the R
2
of the filtered to unfiltered fit was > 0.90 (ESM
Table S4). For the MWNS dataset, Total Kjeldahl Nitrogen (TKN)
was calculated by subtracting nitrates from TN and for the PLUARG
dataset, TN was estimated by calculating the sum of nitrates and
TKN.
Data analysis and nutrient export estimates
We compared total stream runoff, nutrient export and flow-
weighted mean concentrations (FWMC) at the six study water-
sheds. Data from two years of continuous seasonal sampling were
included in this study; specifically, June 1975 to May 1977 (inclu-
sive) for PLUARG and June 2016 to May 2018 (inclusive) for the
MWNS. This subset of data for both studies included two of each
meteorological season: winter (December to February), spring
(March to May), summer (June to August) and fall (September to
November).
The PLUARG study utilized the Beale Ratio Estimator (Beale,
1962) to generate nutrient loads from stream discharge and water
chemistry data. The Beale technique is best applied in situations
when there are limited concentration data yet high-frequency flow
data (Quilbé et al., 2006); and when applied to flow-stratified data,
Beale estimates are considered robust and unbiased (Lee et al.,
2016). Relative to PLUARG, the MWNS water chemistry and dis-
charge data collection frequency were higher due to the automated
remote sampling and telemetry opportunities. Daily mean MWNS
stream discharge was computed to match the sample frequency
during PLUARG. The log–log relationship between discharge and
concentration for each nutrient parameter was calculated for each
site with 95% confidence intervals (CI) used to evaluate significant
changes in each regression slope and intercept (ESM Table S6).
Using the Beale Ratio Estimator, we calculated seasonal load
estimates for the six study watersheds. Estimates were boot-
strapped (B = 2000) to generate CIs. Loads were considered signif-
icantly different if the 95th percentiles of the bootstrapped
distribution of load estimates did not overlap (Wood, 2004). To cal-
culate the average two-year load, one value from each seasonal
bootstrapped load distribution (seasonal n = 2000) was randomly
selected and averaged. This process was repeated for 2000 itera-
tions, and the mean of the corresponding distribution was used
to represent the average load. CIs were again applied to evaluate
significant changes between PLUARG and MWNS loads. To account
for differences in watershed size between studies, loads (and the
respective CIs) were divided by the watershed area to produce an
estimate of nutrient loss, or unit-area export, kg/day x km
2
here-
after referred to as export. Each load estimate was also converted
Table 2
Winter season meteorological conditions at the North Creek watershed during the
PLUARG study and MWNS.
North Ck. Winter (Dec-Jan-
Feb)
PLUARG (1975–
1977)
MWNS (2016–
2018)
Mean Temp (Deg C) 5.5 3.0
Median Temp (Deg C) 5.3 1.8
Min Temp (Deg C) 22.0 20.2
Max Temp (Deg C) 10.6 12.3
Temp Range 32.6 32.5
Mean Rain (mm) 1.0 1.6
Mean Rain > 0 mm (mm) 6.5 5.2
Median Rain > 0 mm (mm) 4.3 3.0
Mean Snow (cm) 1.5 0.8
Mean Snow > 0 cm (cm) 3.3 2.5
Median Snow > 0 cm (cm) 1.6 1.2
Mean Consecutive FDD (days) 11.1 3.8
Median Consecutive FDD
(days)
5.0 2.0
Max Consecutive FDD (days) 62 22
Consecutive FDD Cycles 21 32
Number of Days Below 0 Deg C 214 177
Fig. 2. Mean daily air temperature, snowfall, rainfall and number of consecutive freezing degree days (FDD, days with mean daily air temperature > 0 °C) at the Government
of Canada Hamilton A meteorological station, near the North Creek Watershed, during the PLUARG study (1975–1977) and the MWNS (2016–2018) years.
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1693
into a FWMC by dividing the total load by the total stream dis-
charge for each site with values expressed as
l
g/L.
The percent change in total runoff, FWMC and export was calcu-
lated for each nutrient parameter. Changes in FWMC and load esti-
mates were assessed using Cis, and a Welch’s two-sample t-test
was performed to assess the difference among PLUARG and MWNS
daily mean runoff. Significance was evaluated on critical
a
= 0.05.
The percent seasonal contribution for nutrient export and runoff
were calculated for both PLUARG and MWNS datasets, and the dif-
ference in the percent seasonal contribution between studies was
also computed. All analyses were performed in the R software
environment (R Core Team, 2020).
An evaluation of sampling coverage over the entire range of the
probability of exceedance curve was conducted and indicated there
was poor sampling coverage over the summer at Venison Creek
during the MWNS (ESM Fig. S2) and all seasons during PLUARG
(with >5% of high-flow conditions missing for all seasons; ESM
Figs. S3-S6). Peak flow conditions were not captured for the
PLUARG study during the summer at Big Creek, Nissouri Creek
and North Creek (missing the top 5%, 9% and 25% of high-flow con-
ditions, respectively), during the fall at Nissouri Creek (missing the
top 10% of high-flow conditions), and during the winter at North
Creek and North Maitland River (missing the top 10% and 4% of
high-flow conditions, respectively). As high flows can potentially
deliver substantial loads, we conducted a sensitivity analysis for
the Beale Ratio Estimator method to explore whether the reduced
high-flow sampling frequency during PLUARG relative to the
MWNS had an influence on nutrient load estimates. The range in
MWNS streamflow was reduced to match that of the PLUARG study
probability of exceedance curves (ESM Fig. S7) and export was re-
calculated. Despite decreases in the reported TP and TDP export
and increases in nitrates and TN export at most sites, the direction-
ality of change was typically preserved when comparing PLUARG
to MWNS although it is important to note that the percent changes
in export at Nissouri Creek (for TP), North Maitland River (for TP,
TDP, and TKN), Little Ausable River (TDP and TKN) and at Big Creek
(TN, nitrates and TKN) were no longer significant when run on the
reduced dataset (ESM Table S5, Fig. S8). All export and FWMC esti-
mates presented here were generated on the full MWNS dataset,
with the results from the ‘‘reduced” data scenario presented in
Fig. S8.
Results
Stream runoff
An increase in total runoff between 1975–1977 and 2016–2018
was observed at five of six study watersheds, with a significant
increase in daily mean runoff observed at three sites (Fig. 3A).
The largest percent increases in runoff occurred at North Creek
Fig. 3. Percent change between PLUARG study (1975–1977) and MWNS (2016–2018) for: (A) stream runoff, (B) unit-area export and (C) flow-weighted mean concentrations
(FWMC) at Big Creek (BIG), Little Ausable River (LA), Nissouri Creek (NIS), North Creek (NOR), North Maitland River (NMA) and Venison Creek (VEN). Export and FWMCs are
presented for total phosphorus (TP), total dissolved phosphorus (TDP), total nitrogen (TN), nitrates (NO
3
+NO
2
) and Total Kjeldahl Nitrogen (TKN). Sites with significant
changes in mean daily runoff (
a
= 0.05), nutrient export and FWMCs (i.e., the 95% confidence intervals do not overlap between the MWNS and PLUARG studies) are shaded in
black. All changes considered to be not significant are shaded in grey.
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1694
(~52%), Big Creek (~39%), and Little Ausable River (~29%) alongside
significant increases in mean daily runoff. The percent change in
total runoff at Nissouri Creek and North Maitland River was rela-
tively smaller when compared to the other watersheds (increasing
by 12%), and was 0.05% at Venison Creek.
Phosphorus and nitrogen export
Significant increases in TP and TDP export were observed at five
of the six watersheds, with all five watersheds undergoing
increases greater than ~50% (Fig. 3B, Table 3). The percent change
in TP export at Venison Creek was not significantly different
between the PLUARG and MWNS study periods; however, there
was a significant decrease in the estimated TDP export (Fig. 3B).
Increases in the TN and nitrates export relative to the PLUARG
were observed at Little Ausable River, Nissouri Creek, North Mait-
land River and Venison Creek, whereas North Creek underwent an
increase in only nitrate (Fig. 3B, Table 4). Big Creek was the only
site where there was a significant decrease in the export of TN
and nitrates. The increase in nitrates export was larger than the
increase in TN export at North Maitland River and Venison Creek,
increasing by a factor of ~1.4 and ~2.0, respectively. Interestingly,
TKN export decreased at Big Creek, North Creek, North Maitland
River and Venison Creek relative to export observed during
PLUARG, with the only positive percent change in TKN export
observed at Little Ausable River (Fig. 3B).
Phosphorus and nitrogen flow-weighted mean concentrations
During the MWNS, the highest FWMCs for TP and TDP were
observed at North Creek (Table 3), and the highest FWMCs for all
N fractions (nitrates, TKN and TN) were observed at Nissouri Creek
(Table 4). Conversely during the PLUARG study period, the highest
TP, nitrates, and TN FWMCs were observed at Big Creek and the
highest TDP and TKN FWMCS were observed at North Creek
(Table 3–4). Significant increases in FWMCs of TP were observed
at Little Ausable River, Nissouri Creek and North Maitland River
and significant increases in FWMCs of TDP were observed at Big
Creek, Nissouri Creek and the North Maitland River (Fig. 3C). Veni-
son Creek was the only site with a negative percent change in TDP
FWMCs (Fig. 3C), indicating a decrease from PLUARG
concentrations.
FWMCs underwent similar directional changes as export esti-
mates among N fractions. Big Creek and North Maitland River
had negative percent changes in the FWMCs for TN, with a negative
percent change also observed in nitrate FWMCs at Big Creek. Sig-
nificant increases in nitrates were observed at Little Ausable River,
Nissouri Creek, North Maitland River and Venison Creek. The per-
cent change in nitrate FWMC at Venison Creek was ~2the
increase observed at North Maitland River and Nissouri Creek
and ~6the increase observed at Little Ausable River (Fig. 3C).
Seasonality of nutrient export and stream discharge
During the PLUARG study, the greatest contribution of the total
nutrient export generally occurred during the growing season (i.e.,
spring and summer seasons), often representing >50% of the total
export for N and P (Fig. 4). The only instances where this did not
occur were TN and nitrate export at Little Ausable River, nitrate
export at Nissouri Creek and TDP export at North Creek.
In comparison, the majority of total nutrient export generally
occurred in the non-growing season (i.e., winter and fall seasons)
during the MWNS for all nutrients. The sites where winter and fall
export was not greater than 50% of the total export include: North
Creek (for TP), Big Creek (for TN and TKN), and Venison Creek (for
TN and nitrate). At Big Creek, although the TN, nitrates and TKN
Table 3
Sample size, unit-area export, flow-weighted mean concentrations (FWMC) and total runoff during the MWNS and PLUARG study. Nutrient parameters include: total phosphorus (TP) and total dissolved phosphorus (TDP). Load
estimates were generated using the Beal Ratio estimator and were bootstrapped to generate 95% confidence intervals (presented beside each export and FWMC value).
Site Analyte Sample Size Unit-Area Export (kg/day km
2
) FMWC (mg/L) Runoff (mm/year)
PLUARG MWNS PLUARG MWNS PLUARG MWNS PLUARG MWNS
Estimate Lower CI Upper CI Estimate Lower CI Upper CI Estimate Lower CI Upper CI Estimate Lower CI Upper CI
Big TP 375 350 0.40 0.34 0.47 0.63 0.57 0.70 0.48 0.40 0.56 0.53 0.48 0.59 309.71 431.89
TDP 362 350 0.072 0.060 0.088 0.16 0.15 0.18 0.085 0.071 0.10 0.14 0.13 0.15
Little Ausable TP 487 278 0.21 0.18 0.23 0.37 0.33 0.42 0.17 0.15 0.18 0.23 0.20 0.26 457.44 587.88
TDP 493 280 0.11 0.10 0.11 0.16 0.14 0.17 0.085 0.078 0.091 0.10 0.088 0.11
Nissouri TP 539 306 0.27 0.22 0.33 0.43 0.37 0.50 0.24 0.20 0.29 0.37 0.32 0.43 414.97 425.70
TDP 534 310 0.10 0.091 0.11 0.20 0.17 0.22 0.089 0.080 0.10 0.17 0.15 0.19
North Creek TP 494 307 0.40 0.36 0.45 0.72 0.64 0.81 0.46 0.41 0.51 0.54 0.48 0.61 319.20 484.01
TDP 487 303 0.16 0.15 0.17 0.24 0.23 0.26 0.18 0.17 0.20 0.18 0.17 0.20
North Maitland TP 154 253 0.045 0.034 0.059 0.084 0.074 0.095 0.032 0.025 0.043 0.054 0.048 0.061 504.02 565.85
TDP 150 253 0.019 0.016 0.022 0.034 0.030 0.038 0.014 0.012 0.016 0.022 0.019 0.025
Venison TP 84 239 0.085 0.066 0.10 0.12 0.10 0.14 0.066 0.051 0.080 0.095 0.078 0.11 470.21 452.84
TDP 82 237 0.022 0.017 0.028 0.014 0.012 0.015 0.017 0.013 0.022 0.011 0.010 0.012
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1695
fall/winter export did not represent > 50% of the total export dur-
ing the MWNS, winter/fall export still increased since PLUARG,
with the export of all N parameters now occurring ~50% in the
spring/summer and ~50% in the fall/winter (Fig. 4). A similar
change was observed at Venison Creek where fall/winter export
of TN and nitrates increased relative to PLUARG so that now
~55% of TN and nitrate export occurs during the spring/summer
and ~45% in the fall/winter (previously 65% spring/summer and
36% fall/winter for TN and 61% spring/summer and 39% fall/winter
for nitrates; Fig. 4). It is important to note that typically, most
growing season export occurred during the spring and most non-
growing export typically occurred during the winter at most sites
(with only a few exceptions; Fig. 4).
Winter export tended to increase over time, but it is evident
that the magnitude of these changes varied among sites and ana-
lytes (Table 5). For example, the percent difference between
PLUARG and MWNS winter export for TP and TDP are greater than
the winter export changes observed among N parameters (nitrates,
TKN and TN) at Nissouri Creek, North Maitland River and Little
Ausable River (Table 5). Further, the contribution of winter TP
and TDP export increased between the PLUARG and MWNS studies
at Little Ausable, whereas the relative contribution of winter
export for all N parameters decreased.
An increase in the relative contribution of winter runoff was
observed at Little Ausable River (~19%), Nissouri Creek (~19%)
and North Maitland River (~15%), and an increase in the relative
contribution of fall runoff was observed at Big Creek (~12%) and
Little Ausable River (~5%; Fig. 4). Relatively modest changes were
observed with respect to the seasonal contribution to the total run-
off at North Creek (all changes <5%; Table 5). Relative increases in
fall/winter runoff were generally accompanied by corresponding
decreases in spring/summer runoff (Table 5). The only sites where
spring/summer runoff did not decrease were North Maitland River,
where total summer runoff increased by ~9%, North Creek where
spring runoff increased by ~5%, and Venison Creek where spring
runoff increased by ~2%.
Discussion
This study compares changes in the magnitude and seasonality
of P and N export in agricultural headwater watersheds in the LGL
basin between two studies conducted 40 years apart. Despite
widespread implementation of various agricultural conservation
measures in southern Ontario (Filson et al, 2009; Lamba et al.,
2009), export of TP, TDP, TN and nitrates increased relative to the
mid-1970s at most study watersheds (Fig. 3A). At most sites, grow-
ing season (particularly spring) export often represented the
majority (i.e., >50%) of total P and N export during PLUARG
whereas the non-growing season (particularly winter) exhibited
the highest estimated export during the MWNS. The variability
observed in the changes to the magnitude and seasonality of
export between PLUARG and MWNS among the studied water-
sheds highlights the complex influence of multiple drivers on the
diverse agricultural landscape of southern Ontario.
Total and dissolved phosphorus export changes
Increased runoff in present day relative to the 1970s may be a
driver of the P export increases observed at the Big Creek, Little
Ausable River and North Creek watersheds. At these sites, positive
and significant increases in TP and TDP export were observed, but
increases in TP and TDP FWMCs were lower or non-significant
(Fig. 3). Many studies have associated enhanced stream discharge
with P export to aquatic systems through runoff (Kelly et al.,
2019; Stow et al., 2015). In one Hamilton Harbour tributary, two
Table 4
Sample size, unit-area export, flow-weighted mean concentrations (FWMC) and total runoff during the MWNS and PLUARG study. Nutrient parameters include: total nitrogen (TN), nitrates (NO3 + NO2) and Total Kjeldahl Nitrogen
(TKN). Load estimates were generated using the Beal Ratio estimator and were bootstrapped to generate 95% confidence intervals (presented beside each export and FWMC value).
Site Analyte Sample Size Unit-Area Export (kg/day km
2
) FMWC (mg/L) Runoff (mm/year)
PLUARG MWNS PLUARG MWNS PLUARG MWNS PLUARG MWNS
Estimate Lower CI Upper CI Estimate Lower CI Upper CI Estimate Lower CI Upper CI Estimate Lower CI Upper CI
Big TN 330 280 5.63 5.00 6.37 3.93 3.43 4.48 6.65 5.90 7.52 3.31 2.90 3.78 309.71 431.89
NO3 + NO2 356 267 4.27 3.55 5.15 3.08 2.73 3.43 5.04 4.19 6.08 2.60 2.31 2.90
TKN 374 261 1.77 1.57 1.99 0.76 0.59 0.97 2.09 1.85 2.35 0.64 0.50 0.82
Little Ausable TN 433 216 6.74 6.23 7.26 11.53 10.14 13.12 5.37 4.97 5.79 7.16 6.30 8.14 457.44 587.88
NO3 + NO2 482 211 6.03 5.57 6.51 9.94 8.75 11.23 4.81 4.44 5.19 6.17 5.43 6.97
TKN 500 169 0.88 0.81 0.94 1.25 1.03 1.52 0.70 0.65 0.75 0.78 0.64 0.94
Nissouri TN 453 254 6.62 6.30 7.03 10.68 9.59 11.89 5.82 5.54 6.18 9.18 8.24 10.21 414.97 425.70
NO3 + NO2 524 253 5.37 5.08 5.72 9.38 8.37 10.56 4.72 4.47 5.03 8.06 7.19 9.07
TKN 542 216 1.34 1.18 1.54 1.43 1.25 1.62 1.18 1.04 1.35 1.23 1.07 1.39
North Creek TN 441 294 4.46 4.09 4.80 4.14 3.69 4.67 5.09 4.67 5.49 3.12 2.78 3.52 319.20 484.01
NO3 + NO2 486 285 2.42 2.15 2.66 3.16 2.71 3.74 2.77 2.46 3.04 2.38 2.05 2.82
TKN 499 295 2.06 1.95 2.16 1.13 0.96 1.32 2.35 2.22 2.47 0.85 0.73 0.99
North Maitland TN 142 233 2.98 2.54 3.55 5.06 4.59 5.51 2.16 1.84 2.57 3.27 2.96 3.56 504.02 565.85
NO3 + NO2 147 236 2.30 1.95 2.76 4.65 4.19 5.13 1.67 1.41 2.00 3.00 2.70 3.31
TKN 156 201 0.70 0.61 0.80 0.49 0.44 0.55 0.51 0.44 0.58 0.32 0.29 0.36
Venison TN 74 221 2.05 1.68 2.51 3.65 3.53 3.79 1.59 1.30 1.95 2.95 2.84 3.06 470.21 452.84
NO3 + NO2 80 216 1.35 1.06 1.77 3.41 3.29 3.54 1.05 0.82 1.37 2.75 2.65 2.86
TKN 84 198 0.68 0.52 0.88 0.32 0.29 0.35 0.53 0.40 0.68 0.26 0.23 0.28
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1696
storm events in 2009 were responsible for contributing 88% of the
total summer TP load (Long et al., 2015). It is recognized that a
small number of high-discharge events have the potential to
account for a large proportion of the TP annual load (Macrae
et al., 2007). This may be further supported when considering
the Beale Ratio Estimator sensitivity test conducted by reducing
the MWNS range of streamflow data to match conditions captured
within the PLUARG study (ESM Fig. S2). When excluding the high-
est flows captured within the MWNS, the average percent change
in TP and TDP export among all sites was reduced by 20–30%
(ESM Table S5).
Seasonal changes in the magnitude of stream discharge may
have contributed to P export increases. At the North Maitland River
and Nissouri Creek watersheds, increases in TP and TDP export
were recorded despite no significant increase in stream runoff. At
these two sites, the relative contribution of stream runoff, TP and
TDP export all increased during winter months (Fig. 4). Apart from
North Creek, all other sites also underwent increases in the relative
proportion of TP and TDP export occurring during the fall and/or
winter seasons (Fig. 4). Our findings are consistent with non-
growing season nutrient load increases observed in other southern
Ontario watersheds (Long et al., 2015; Macrae et al., 2007; Plach
et al., 2019). Interestingly, while some Ontario-based studies have
not documented significant long-term trends in streamflow (Raney
and Eimers, 2014) or discharge (Stammler et al., 2017), these stud-
ies excluded data from the winter season. The runoff results pre-
sented here are consistent with modeled discharge increases in
other southern Ontario watersheds (Rahman et al., 2012;
Shouquan Cheng et al., 2012) and suggest that examination of
the trends in streamflow and chemistry year round are essential
to detect long-term nutrient export and stream runoff increases.
Legacy P stores in agricultural soils may have hampered P
export reductions despite efforts to mitigate nutrient losses
through BMPs. Globally, the imbalance between P applied to the
landscape and its uptake by plants has led to an accumulation of
P in soils at a rate of ~7 Tg P each year (Yuan et al. 2018). Even if
Fig. 4. Percent seasonal contribution for stream runoff and export of total phosphorus (TP), total dissolved phosphorus (TDP), total nitrogen (TN), nitrates (NO
3
+NO
2
), and
Total Kjeldahl Nitrogen (TKN) for Big Creek (BIG), Little Ausable River (LA), Nissouri Creek (NIS), North Creek (NOR), North Maitland River (NMA) and Venison Creek (VEN). At
each site, the seasonal breakdown is presented for the PLUARG study and the MWNS (denoted with a ‘‘P” or ‘‘M” above each bar, respectively).
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1697
P application stopped, it would take years to decades for P stores to
be depleted (Sharpley et al. 2013), with the rate of recovery highly
variable across the landscape due to differences in soil type and
agricultural production systems (McDowell et al., 2020). Within
the Great Lakes basin region, specifically the Maumee River basin,
P is mobilized at a faster rate than it accumulates in soils resulting
in net P export from the watershed since the late-1990s (Powers
et al., 2016). Despite enhanced mobilization, meeting target P loads
in the Maumee River Watershed is projected to take decades when
assuming no additional P is applied to the landscape (with longer
projections associated with a wetter climate; Muenich et al.,
2016). However, differences in soil type and fertilizer application
practices between the Maumee River watershed and southern
Ontario limit our ability to extend legacy P trends to our study
region. To address this knowledge gap, work is currently underway
to assess soil test P concentrations at all 11 MWNS watersheds to
assess the influence of legacy P on water quality in southern
Ontario.
FWMCs have been proposed as an alternative metric to load for
tracking watershed progress because they are less sensitive to
inter-annual weather fluctuations (e.g., Mohamed et al., 2019).
Our data provide us an opportunity to explore the interplay of
hydrological and non-hydrological factors in influencing TP and
TDP losses. Comparing the changes in FWMC, runoff, and export
in Fig. 3, we can categorize watersheds into: i) those which had
export changes clearly driven by hydrology, where runoff and TP
export increased but FWMC did not increase (Big Creek and North
Creek); ii) those with a complex interplay of hydrology and other
factors, where runoff and export did increase but FWMC also
increased (Little Ausable), and iii) those where the FWMC changed
without a change in runoff, indicating a change in non-hydrological
aspects of TP export (Nissouri Creek, North Maitland, Venison).
While the information provided by FWMCs can identify cases
where hydrology was primarily responsible for water quality dete-
rioration and those where it was not, there are still cases where
FWMC was not able to resolve the persistent ambiguity in ascrib-
ing changes in water quality to hydrological, biogeochemical, and
management factors. In such cases, detailed analysis of
concentration-discharge relationships would be necessary to
resolve this ambiguity (e.g., Makarewicz et al., 2009). When we
compare changes in runoff to those in TDP export and TDP FWMC,
some watersheds did not follow the same patterns as they did for
TP. For instance, changes to the TP export of Big Creek were due to
hydrology, whereas hydrology alone cannot account for changes in
TDP at Big Creek, as there were increases in the TDP FWMC. At Lit-
tle Ausable, increases in runoff were sufficient to explain the
increases in TDP, whereas they were not for TP. While comparing
FWMCs, runoff, and export can be illuminating, there are still
ambiguities present, and comparisons must be done separately
for each analyte. Additional research with a much larger set of sites
at field and headwater scales would help develop more nuanced
interpretations of the interplay of runoff, export, and FWMC.
Nitrates, total and Kjeldahl nitrogen export changes
Increases in nitrate export occurred at five of six watersheds
and increases in TN at four of six watersheds (Fig. 3). Nitrate trans-
port from agricultural landscapes to surface waters is known to
occur during precipitation events (Rixon et al., 2020) and periods
of high stream discharge (Nangia et al., 2010). A decrease in
nitrates and TN export were observed at Big Creek, where the lar-
gest percent increase in runoff occurred (Fig. 3). It is therefore pos-
sible that the drivers of TN and nitrate export may differ at Big
Creek when compared to the other sites and other factors, in addi-
tion to increased runoff, may have contributed to the increased
nitrates and TN export at the other studied watersheds. These
other factors may include changes in subsurface drainage (e.g.,
Arenas Amado et al., 2017; Kokulan, 2019; Williams et al., 2015),
or potential changes in the composition, amount and timing of
N-based fertilizer application.
Increasing nitrate concentrations have been observed in other
streams within the LGL basin. For example, nitrate concentrations
doubled in 13 Lake Ontario tributaries with mixed urban and agri-
cultural land use over ~30 years (Eimers and Watmough, 2016).
Offshore concentrations of nitrates have also increased in most of
the Great Lakes since the 1970s, even in non-agricultural areas
such as Lake Superior (Dove and Chapra, 2015) with increasing
trends in nitrates across such a large geographic area suggestive
of a regional driver. Atmospheric N deposition is a potential driver
of such regional limnological changes; however, wet nitrate depo-
sition within a 100-km area around each Great Lake has declined
since ~1980 (Reavie et al., 2017). This study supports the conclu-
sions of others (e.g., Eimers and Watmough, 2016), that while
the exact mechanism may be unknown, increasing concentrations
of nitrate within the LGL basin may be at least partially linked to
increasing tributary export.
TKN export declined at four of six watersheds (Fig. 3). While
decreases in TKN loads and concentrations have been observed
Table 5
Percent difference in seasonal contribution for stream discharge, total phosphorus (TP), total dissolved phosphorus (TDP), total nitrogen (TN), nitrates (NO
3
+NO
2
) and Total
Kjeldahl Nitrogen (TKN) export between the Pollution from Land Use Activities Reference Group (PLUARG) study and the Multi-Watershed Nutrient Study.
Big Little Ausable
TP TDP TN NO3 + NO2 TKN Runoff TP TDP TN NO3 + NO2 TKN Runoff
Winter 10.79 18.82 18.04 23.12 13.44 0.64 24.28 22.38 8.94 6.45 7.93 18.82
Spring 5.38 –22.56 17.80 –22.11 7.63 3.82 13.31 19.32 7.63 12.86 18.53 17.77
Summer –22.47 10.18 12.51 13.68 14.56 8.92 16.89 8.72 7.09 12.03 5.65 6.01
Fall 17.06 13.92 12.26 12.67 8.74 12.10 5.92 5.66 8.40 5.62 32.11 4.96
Nissouri North Creek
TP TDP TN NO3 + NO2 TKN Runoff TP TDP TN NO3 + NO2 TKN Runoff
Winter 46.65 45.12 3.55 3.17 27.82 18.75 4.17 10.56 6.92 14.86 1.70 0.87
Spring 37.68 42.48 11.27 3.59 40.11 13.45 6.44 5.93 10.07 15.62 2.54 4.84
Summer 12.27 5.03 4.33 5.79 6.02 4.64 1.53 0.66 4.42 8.95 2.71 2.86
Fall 3.30 2.39 3.39 0.97 18.31 0.66 0.74 3.97 7.57 9.71 6.94 1.11
North Maitland Venison
TP TDP TN NO3 + NO2 TKN Runoff TP TDP TN NO3 + NO2 TKN Runoff
Winter 31.80 35.08 9.77 8.16 23.58 14.76 24.20 17.69 7.42 4.79 16.31 1.25
Spring 38.97 30.31 4.26 4.04 12.27 –23.91 18.94 0.57 11.54 13.23 11.30 0.75
Summer 1.39 7.06 10.09 9.30 13.80 9.13 7.53 17.73 1.58 7.99 9.84 1.65
Fall 5.78 2.28 4.58 5.18 2.49 0.02 2.27 0.61 2.54 0.45 4.84 2.15
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1698
elsewhere (Burniston et al., 2018; Stow et al., 2001), regional
trends are variable, with other studies reporting TKN increases
(Choquette et al., 2019; Lebo et al., 2012). A significant decrease
in the slope and intercept of the log–log regression between dis-
charge and TKN concentration at Big Creek, North Creek and Veni-
son Creek (ESM Table S6) suggest decreases in export at these sites
are not the result of changes in hydrology but rather a change in
nutrient source. TKN is comprised of organic N (present in plant
material and animal manure fertilizer) and ammonia/ammonium
(present in commercial fertilizers). A reduction in manure use in
favour of synthetic fertilizers can reduce the amount of organic
N, and thus TKN, in receiving surface waters; however, this was
likely not the case in our study as livestock densities have
increased in most study watersheds since the mid-1970s (Table 1).
A better understanding of land use and land management change
(including fertilizer composition and application rates) is needed
to confidently identify potential drivers of reduced TKN export in
present day relative to the 1970s.
Influence of agricultural land use and management
The changes we observe in nutrient export (Fig. 3) are likely the
result of a number of interacting factors, including changes in land
use, land management, and/or hydrological variability. While we
are not able to complete detailed attribution within the scope of
this study, we can pose or rule out some testable hypotheses. None
of the watersheds experienced significant changes in land use; all
were predominantly agricultural land in the 1970s and remain so
to this day. Further, most watersheds were cash cropped in the
1970s with some mix of corn, soy, and wheat, and the watersheds
that had significant hay or pasture in the 1970s still have those
types of land use in roughly the same amounts (ESM Table S1).
Notable changes in land use include an increase in soy relative to
the mid-1970s in all watersheds (ESM Table S1) and an increase
in livestock density at all sites except for North Creek and Venison
Creek (Table 1). It is also possible that the study watersheds are
cropped more intensively now than they were in the 1970s.
Indeed, regional studies have shown this to be the case throughout
the lower Great Lakes (Bruulsema et al., 2011; Reid and Schneider,
2019). A nutrient balance of the watersheds (encompassing legacy
P) is needed to further investigate this hypothesis, as it is possible
that agricultural intensification has led to a greater quantity of
source material (i.e., fertilizer/manure) in these watersheds. How-
ever, it is also necessary to account for changes in the placement
and timing of nutrient inputs to the land, as significant campaigns
have taken place to encourage farmers to apply subsurface nutri-
ents (Flis, 2017).
A widespread increase in the amount and density of tile drai-
nage is also observed (Table 1). While the effects of tile drainage
on increasing water yield are clear and widely established, their
effect on nutrient loss is more subtle and variable (King et al.,
2015). However, the expansion of tile drainage at these catchments
represents potential for increased transport of nutrients. More
research is needed to disentangle the roles of increased drainage
and agricultural intensification. Specifically, an investigation of
the specific agricultural land use and land management practices
within each MWNS watershed is underway and will be compared
to farmer survey results that were conducted during PLUARG in the
1970s. These data will provide the landscape context to better
understand stream water quality changes and how these may have
changed between PLUARG and MWNS.
Maumee River comparison to MWNS
The province of Ontario and the state of Ohio have reported
nearly identical trends in agricultural fertilizer, manure and crop
removal P balance between 1955 and 2010 (Bruulsema et al.,
2011). With these similar trends, it may be expected that changes
in P export between Ontario and the Maumee River watersheds
would also be comparable. Of the six watersheds studied, Big Creek
was the most similar in landscape features when compared to the
Maumee River watershed. Both watersheds have poorly-drained
clay soils, a low-slope relief, similar crop types and predominately
underlain by tile drainage (Muenich et al., 2016). Changes observed
in Big Creek nutrient export relative to PLUARG are generally con-
sistent with temporal trends in nutrient concentrations of the
Maumee River. Specifically, concentrations of TP and TDP in the
Maumee River have increased since the 1990s with corresponding
decreases in nitrate and TN concentrations (Stow et al., 2015).
While increases in stream discharge were also observed in the
Maumee, seasonal increases in stream discharge were only
observed during March (Stow et al., 2015) which contrasts obser-
vations from Big Creek where an increase in fall runoff was
observed (+13% relative to PLUARG; Fig. 4).
Importantly, changes in nutrient export and runoff at Big Creek
differed from the other watersheds (i.e., decreases in TN and
nitrates export and increases in fall runoff), suggesting that
insights gained from the Maumee River watershed with respect
to nutrient management should be applied with caution in other
areas of southern Ontario. Since the agricultural watersheds within
the LGL basin, including southern Ontario, encompass a range of
soil types, land uses, and land management (OMAFRA, 2020),
applying management strategies similar to those used in the Mau-
mee may not result in comparable outcomes.
Conclusions and implications for management
Despite differences among watershed characteristics, some
common themes emerge from this investigation. Increased runoff,
particularly during the non-growing seasons, is likely an important
contributor to the increased TP and TDP export at three of six
watersheds. TN and nitrate export also increased at most sites,
except for Big Creek, and may be associated with enhanced tile
drainage or changes in fertilizer management. Although the extent
of tile drainage in each watershed has increased since the 1970s
(Table 1), the spatial extent and density of tile drainage is not well
characterized in this region and thus limits our ability to assess
relationships.
Findings from the farmer survey component of the MWNS will
aid in developing linkages between stream nutrient export and
land use/land management. The unique changes observed at Big
Creek (with respect to the decreases in TN and nitrate export)
and Venison Creek (with respect to a decrease in TDP export)
emphasizes the variability of agricultural watersheds to climate
and landscape changes over time. This study highlights the need
for management plans specific to southern Ontario watersheds
that account for differences in land use, land drainage, future sus-
ceptibility to runoff increases and potentially the legacy P accumu-
lation in soils.
The observed seasonal shift in the timing of P and N export from
the growing season (during PLUARG) to the non-growing season
(during MWNS) highlights the need to revisit pre- existing
assumptions that downplay the role of winter in lake and stream
nutrient balances and nutrient bioavailability. This finding stresses
the importance of year-round monitoring of surface waters to
ensure all critical seasons and time periods are captured to gener-
ate accurate estimates of nutrient export. Where watershed BMPs
exist with a primary focus on spring and summer agricultural prac-
tices, a further need for BMPs specific to fall and winter activities is
warranted if these are now the important seasons for P and N
export in Ontario headwater agricultural watersheds. A combina-
tion of continued monitoring and experimental lab and field-
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1699
based studies will be needed to track progress towards NPS reduc-
tions and the effectiveness of BMPs.
The increases we observed in runoff, export, and FWMC were
striking. If such patterns are pervasive across southern Ontario,
they would have important implications for our understanding of
nutrient export from agricultural watersheds, potential manage-
ment options, and implications for nutrient loading to the LGL.
However, a key limitation of the present study is the limited num-
ber of watersheds where such direct, detailed comparisons were
possible. As mentioned previously, work currently underway, such
as the MWNS, will provide more systematic comparisons across
agricultural regions of southern Ontario.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgements
This work was funded by the Canada-Ontario Agreement (Pro-
ject ID #1504), through the Ontario Ministry of the Environment,
Conservation and Parks (OMECP) with Ryerson University. Grace
Arabian and Craig Onafrychuk contributed data management and
mapping support, Laura Benakoun developed standardized proto-
cols for field sampling, partner training and data QA/QC, and Derek
Smith and Dave Supper coordinated field sampling efforts. Addi-
tional field and sampling support was provided from OMECP South
West Region and West Central Region, Ganaraska Region Conserva-
tion Authority and Maitland Valley Conservation Authority.
PLUARG report access and valuable discussion was provided by
Dr. Trevor Dickinson. The Ontario Ministry of Agriculture, Food
and Rural Affairs contributed to project development and steering.
All codes and scripts used in data preparation and analysis can be
obtained from the corresponding author upon request.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jglr.2021.08.010.
References
Agriculture Canada, Ontario Ministry of Agriculture and Food, Ontario Ministry of
the Environment, 1975. Agricultural watershed studies Great Lakes drainage
basin Canada detailed study plan 1975–1976 Accessed 09/29/2020 https://
atrium.lib.uoguelph.ca/xmlui/bitstream/handle/10214/14512/
PLUARG_summary.pdf?sequence=1&isAllowed=y,.
Agriculture Canada, Ontario Ministry of Agriculture and Food, Ontario Ministry of
Environment, 1978. Agricultural watershed studies Great Lakes drainage basin
Canada final summary report Accessed 09/29/2020 https://atrium.lib.uoguelph.
ca/xmlui/bitstream/handle/10214/14512/PLUARG_summary.pdf?sequence=1&
isAllowed=y,.
Arenas Amado, A., Schilling, K.E., Jones, C.S., Thomas, N., Weber, L.J., 2017.
Estimation of tile drainage contribution to streamflow and nutrient loads at
the watershed scale based on continuously monitored data. Environ. Monit.
Assess. 189, 426.
Auer, M.T., Tomlinson, L.M., Higgins, S.N., Malkin, S.Y., Howell, E.T., Bootsma, H.A.,
2010. Great Lakes Cladophora in the 21st century: Same algae-different
ecosystem. J. Great Lakes Res. 36, 248–255.
Beale, E.M.L., 1962. Some uses of computers in operational research. Ind. Organ. 31,
51–52.
Bertram, P.E., 1993. Total phosphorus and dissolved oxygen trends in the central
basin of Lake Erie, 1970–1991. J. Great Lakes Res. 19, 224–236.
Bosch, N.S., Evans, M.A., Scavia, D., Allan, J.D., 2014. Interacting effects of climate
change and agricultural BMPs on nutrient runoff entering Lake Erie. J. Great
Lakes Res. 40, 581–589.
Bruulsema, T.W., Mullen, R.W., Halloran, I.P.O., Warncke, D.D., 2011. Agricultural
phosphorus balance trends in Ontario , Michigan and Ohio. Can. J. Soil Sci. 91,
437–422.
Burniston, D., Dove, A., Backus, S., Thompson, A., 2018. Nutrient concentrations and
loadings in the St. Clair River-Detroit River Great Lakes interconnecting channel.
J. Great Lakes Res. 44, 398–411.
Champagne, O., Arain, M.A., Coulibaly, P., 2019. Atmospheric circulation amplifies
shift of winter streamflow in southern Ontario. J. Hydrol. 578, 124051.
Choquette, A.F., Hirsch, R.M., Murphy, J.C., Johnson, L.T., Confesor, R.B., 2019.
Tracking changes in nutrient delivery to western Lake Erie: Approaches to
compensate for variability and trends in streamflow. J. Great Lakes Res. 45, 21–
39.
Chow, J., Abbey, A.I., Khan, Z., Dermicheva, S., Jennings, W., Wilson, P., 2010. 2009
Performance report: General chemistry and microbiology analyses section.
Ontario Ministry of the Environment Report. Queen’s Printer for Ontario,
Ontario, Canada.
Crins, W.J., Gray, P.A., Uhlig, P.W.C., Wester, M.C., 2009. Ecosystems of Ontario. Part
1, Ecozones and ecoregions. Ontario, Ministry of Natural Resources, Inventory,
Monitoring and Assessement Section. Accessed 09/29/2020 https://files.ontario.
ca/mnrf-ecosystemspart1-accessible-july2018-en-2020-01-16.pdf.
De Pinto, J.V., Young, T.C., Mchroy, L.M., 1986. Great Lakes water quality
improvement. Environ. Sci. Technol. 20, 752–759.
Dove, A., Chapra, S.C., 2015. Long-term trends of nutrients and trophic response
variables for the Great Lakes. Limnol. Oceanogr. 60, 696–721.
Eimers, M.C., Watmough, S.A., 2016. Increasing nitrate concentrations in streams
draining into Lake Ontario. J. Great Lakes Res. 42, 356–363.
[ECCC, OMOECC] Environment and Climate Change Canada, Ontario Ministry of the
Environment and Climate Change, 2018. Canada-Ontario Lake Erie Action Plan.
Accessed 09/29/2020:https://www.canada.ca/content/dam/eccc/documents/
pdf/great-lakes-protection/dap/action_plan.pdf.
Filson, G.C., Sethuratnam, S., Adekunle, B., Lamba, P., 2009. Beneficial management
practice adoption in five Southern Ontario watersheds. J. Sustain. Agric. 33,
229–252.
Flis, S., 2017. Phosphorus management research and the 4Rs. Crops and Soils 3, 28–
67.
Frank, R., Ripley, B.D., 1977. Land use activities in eleven agricultural watersheds in
Southern Ontario, Canada, 1975-76. Ontario Ministry of Agriculture and Food.
Geohring, L.D., McHugh, O.V., Walter, M.T., Steenhuis, T.S., Akhtar, M.S., Walter, M.F.,
2001. Phosphorus transport into subsurface drains by macropores after manure
applications: Implications for best manure management practices. Soil Sci. 166,
896–909.
[GOC] Government of Canada. 2020. Historical Climate Data. Accessed 09/29/2020:
https://climate.weather.gc.ca/historical_data/search_historic_data_e.html.
Howell, E.T., 2018. Cladophora (green algae) and dreissenid mussels over a nutrient
loading gradient on the north shore of Lake Ontario. J. Great Lakes Res. 44, 86–
104.
Jarvie, H.P., Johnson, L.T., Sharpley, A.N., Smith, D.R., Baker, D.B., Bruulsema, T.W.,
Confesor, R., 2017. Increased soluble phosphorus Loads to Lake Erie:
Unintended consequences of conservation practices? J. Environ. Qual. 46,
123–132.
Kane, D.D., Conroy, J.D., Peter Richards, R., Baker, D.B., Culver, D.A., 2014. Re-
eutrophication of Lake Erie: Correlations between tributary nutrient loads and
phytoplankton biomass. J. Great Lakes Res. 40, 496–501.
Kelly, P.T., Renwick, W.H., Knoll, L., Vanni, M.J., 2019. Stream nitrogen and
phosphorus loads are differentially affected by storm events and the
difference may be exacerbated by conservation tillage. Environ. Sci. Technol.
53, 5613–5621.
King, K.W., Williams, M.R., Macrae, M.L., Fausey, N.R., Frankenberger, J., Smith, D.R.,
Kleinman, P.J.A., Brown, L.C., 2015. Phosphorus transport in agricultural
subsurface drainage: A review. J. Environ. Qual. 44, 467–485.
Kirchmeier-Young, M.C., Zhang, X., 2020. Human influence has intensified extreme
precipitation in North America. Proc. Natl. Acad. Sci. U.S.A. 117, 13308–13313.
Kokulan, V., 2019. Environmental and economic consequences of tile drainage
systems in Canada. Ottawa. Accessed 09/29/2020 https://capi-icpa.ca/wp-
content/uploads/2019/06/2019-06-14-CAPI-Vivekananthan-Kokulan-Paper-
WEB.pdf.
Lamba, P., Filson, G., Adekunle, B., 2009. Factors affecting the adoption of best
management practices in southern Ontario. Environmentalist 29, 64–77.
Lebo, M.E., Paerl, H.W., Peierls, B.L., 2012. Evaluation of progress in achieving TMDL
mandated nitrogen reductions in the Neuse river basin, North Carolina. Environ.
Manage. 49, 253–266.
Lee, C.J., Hirsch, R.M., Schwarz, G.E., Holtschlag, D.J., Preston, S.D., Crawford, C.G.,
Vecchia, A.V., 2016. An evaluation of methods for estimating decadal stream
loads. J. Hydrol. 542, 185–203.
Lintern, A., McPhillips, L., Winfrey, B., Duncan, J., Grady, C., 2020. Best management
practices for diffuse nutrient pollution: wicked problems across urban and
agricultural watersheds. Environ. Sci. Technol. 54, 9159–9174.
Long, T., Wellen, C., Arhonditsis, G., Boyd, D., Mohamed, M., O’Connor, K., 2015.
Estimation of tributary total phosphorus loads to Hamilton Harbour, Ontario,
Canada, using a series of regression equations. J. Great Lakes Res. 41, 780–793.
Maccoux, M.J., Dove, A., Backus, S.M., Dolan, D.M., 2016. Total and soluble reactive
phosphorus loadings to Lake Erie: A detailed accounting by year, basin, country,
and tributary. J. Great Lakes Res. 42, 1151–1165.
Macrae, M.L., English, M.C., Schiff, S.L., Stone, M., 2007. Capturing temporal
variability for estimates of annual hydrochemical export from a first-order
agricultural catchment in southern Ontario, Canada. Hydrol. Process. 21, 1651–
1663.
Makarewicz, J.C., Lewis, T.W., Bosch, I., Noll, M.R., Herendeen, N., Simon, R.D.,
Zollweg, J., Vodacek, A., 2009. The impact of agricultural best management
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1700
practices on downstream systems: Soil loss and nutrient chemistry and flux to
Conesus Lake, New York, USA. J. Great Lakes Res. 25, 23–36.
McDowell, R., Dodd, R., Pletnyakov, P., Noble, A., 2020. The ability to reduce soil
legacy phosphorus at a country scale. Front. Enviro. Sci. doi: 10.3389/
fenvs.2020.00006.
Mohamed, M.N., Wellen, C., Parsons, C.T., Taylor, W.D., Arhonditsis, G., Chomicki, K.
M., Boyd, D., Weidman, P., Mundle, S.O.C., Van Cappellen, P., Sharpley, A.N.,
Haffner, D.G., 2019. Understanding and managing the re-eutrophication of lake
erie: Knowledge gaps and research priorities. Freshw. Sci. 38, 675–691.
Muenich, R.L., Kalcic, M., Scavia, D., 2016. Evaluating the impact of legacy P and
agricultural conservation practices on nutrient loads from the Maumee River
Watershed. Environ. Sci. Technol. 50, 8146–8154.
Nangia, V., Mulla, D.J., Gowda, P.H., 2010. Precipitation changes impact stream
discharge, nitrate-nitrogen load more than agricultural management changes. J.
Environ. Qual. 39, 2063–2071.
Norton, P.A., Driscoll, D.G., Carter, J.M., 2019. Climate, streamflow, and lake-level
trends in the Great Lakes Basin of the United States and Canada, water years
1960–2015.
[OMAFRA] Ontario Ministry of Agriculture, Food and Rural Affairs, 2020. Cassifying
prime and marginal agricultural soils and landscapes: guidelines for application
of the Canada Land Inventory in Ontario Accessed on 09/29/2020 http://www.
omafra.gov.on.ca/english/landuse/classify.htm,.
[OMOE] Ontario Ministry of the Environment, 1979. Streamflow data for Pollution
from Land Use Activities Reference Group program 1975-1977. Toronto.
[OMOE] Ontario Ministry of the Environment, 2011. Method E3036: The
determination of total phosphorus in water by colourimetry, version 3.0,
April 1, 2011. Dorset, Ontario.
Painter, D.S., Kamaitis, C., 1987. Reduction of Cladophora biomass and tissue
phosphorus in Lake Ontario, 1972–83. Can. J. Fish. Aquat. Sci. 44, 2212–2215.
Plach, J., Pluer, W., Macrae, M., Kompanizare, M., McKague, K., Carlow, R., Brunke, R.,
2019. Agricultural edge-of-field phosphorus losses in Ontario, Canada:
Importance of the nongrowing season in cold regions. J. Environ. Qual. 48,
813–821.
Powers, S.M., Bruulsema, T.W., Burt, T.P., Chan, N.L., Elser, J.J., Haygarth, P.M.,
Howden, N.J.K., Jarvie, H.P., Yang, L., Peterson, H.M., Sharpley, A.N., Shen, J.,
Worrall, F., Zhang, F., 2016. Long-term accumulation and transport of
anthropogenic phosphorus in three river basins. Nature Geosci. 9, 353–356.
Quilbé, R., Rousseau, A.N., Duchemin, M., Poulin, A., Gangbazo, G., Villeneuve, J.P.,
2006. Selecting a calculation method to estimate sediment and nutrient loads in
streams: Application to the Beaurivage River (Québec, Canada). J. Hydrol. 326,
295–310.
R Core Team, 2020. R: A Language and Environment for Statistical Computing. R
Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.
org/.
Rahman, M., Bolisetti, T., Balachandar, R., 2012. Hydrologic modelling to assess the
climate change impacts in a Southern Ontario watershed. Can. J. Civ. Eng. 39,
91–103.
Raney, S.M., Eimers, M.C., 2014. Unexpected declines in stream phosphorus
concentrations across southern Ontario. Can. J. Fish. Aquat. Sci. 71, 337–342.
Reavie, E.D., Sgro, G.V., Estepp, L.R., Bramburger, A.J., Shaw Chraïbi, V.L., Pillsbury, R.
W., Cai, M., Stow, C.A., Dove, A., 2017. Climate warming and changes in
Cyclotella sensu lato in the Laurentian Great Lakes. Limnol. Oceanogr. 62, 768–
783.
Reid, K., Schneider, K.D., 2019. Phosphorus accumulation in Canadian agricultural
soils over 30 yr. Can. J. Soil Sci. 99, 520–532.
Rittenburg, R.A., Squires, A.L., Boll, J., Brooks, E.S., Easton, Z.M., Steenhuis, T.S., 2015.
Agricultural BMP effectiveness and dominant hydrological flow paths: Concepts
and a review. J. Am. Water Resour. Assoc. 51, 305–329.
Rixon, S., Levison, J., Binns, A., Persaud, E., 2020. Spatiotemporal variations of
nitrogen and phosphorus in a clay plain hydrological system in the Great Lakes
Basin. Sci. Total Environ. 714. doi: 10.1016/j.scitotenv.2019.136328.
Robertson, D.M., Saad, D.A., 2011. Nutrient inputs to the Laurentian Great Lakes by
source and watershed estimated using SPARROW watershed models. J. Am.
Water Resour. Assoc. 47, 1011–1033.
Rosamond, M.S., Wellen, C., Yousif, M.A., Kaltenecker, G., Thomas, J.L., Joosse, P.J.,
Feisthauer, N.C., Taylor, W.D., Mohamed, M.N., 2018. Representing a large
region with few sites: The Quality Index approach for field studies. Sci. Total
Environ. 633, 600–607.
Scavia, D., David Allan, J., Arend, K.K., Bartell, S., Beletsky, D., Bosch, N.S., Brandt, S.B.,
Briland, R.D., Daloǧlu, I., DePinto, J.V., Dolan, D.M., Evans, M.A., Farmer, T.M.,
Goto, D., Han, H., Höök, T.O., Knight, R., Ludsin, S.A., Mason, D., Michalak, A.M.,
Peter Richards, R., Roberts, J.J., Rucinski, D.K., Rutherford, E., Schwab, D.J.,
Sesterhenn, T.M., Zhang, H., Zhou, Y., 2014. Assessing and addressing the re-
eutrophication of Lake Erie: Central basin hypoxia. J. Great Lakes Res. 40, 226–
246.
Sharpley, A.N., Smith, S.J., 1991. Effects of cover crops on surface water quality.
Surface Water Impacts 9, 41–49.
Sharpley, A., Richards, P., Herron, S., Baker, D., 2012. Case study comparison
between litigated and voluntary nutrient management strategies. J. Soil Water
Conserv. 67, 442–450.
Sharpley, A., Jarvie, H.P., Buda, A., May, L., Spears, B., Kleinman, P., 2013. Phosphorus
legacy: Overcoming the effect of past management practices to mitigate future
water quality impairment. Environ. Sci. Technol. 42, 1308–1326.
Shouquan Cheng, C., Li, Q., Li, G., Auld, H., 2012. Possible impacts of climate change
on daily streamflow and extremes at local scale in Ontario, Canada. Part II:
Future projection. Atmos. Clim. Sci. 02, 427–440.
Sonzogni, W.C., Chesters, G., Coote, D.R., Jeffs, D.N., Konrad, J.C., Ostry, R.C.,
Robinson, J.B., 1980. Pollution from land runoff. Great Lakes Basin
Commission Ann Arbor, Mich. 48106. Environ. Sci. Technol. 14, 148–153.
Stammler, K.L., Taylor, W.D., Mohamed, M.N., 2017. Long-term decline in stream
total phosphorus concentrations: a pervasive pattern in all watershed types in
Ontario. J. Great Lakes Res. 43, 930–937.
Stow, C.A., Borsuk, M.E., Stanley, D.W., 2001. Long-term changes in watershed
nutrient inputs and riverine exports in Neuse River, North Carolina. Water Res.
35, 1489–1499.
Stow, C.A., Cha, Y., Johnson, L.T., Confesor, R., Richards, R.P., 2015. Long-term and
seasonal trend decomposition of maumee river nutrient inputs to western lake
erie. Environ. Sci. Technol. 49, 3392–3400.
Stumpf, R.P., Wynne, T.T., Baker, D.B., Fahnenstiel, G.L., 2012. Interannual variability
of cyanobacterial blooms in Lake Erie. PLoS One 7, e42444.
Tomer, M.D., Locke, M.A., 2011. The challenge of documenting water quality
benefits of conservation practices: A review of USDA-ARS’s conservation effects
assessment project watershed studies. Water Sci. Technol. 64, 300–310.
Vincent, L.A., Zhang, X., Mekis, Wan, H., Bush, E.J., 2018. Changes in Canada’s
climate: Trends in indices based on daily temperature and precipitation data.
Atmos. - Ocean 56, 332–349.
Watson, S.B., Miller, C., Arhonditsis, G., Boyer, G.L., Carmichael, W., Charlton, M.N.,
Confesor, R., Depew, D.C., Höök, T.O., Ludsin, S.A., Matisoff, G., McElmurry, S.P.,
Murray, M.W., Peter Richards, R., Rao, Y.R., Steffen, M.M., Wilhelm, S.W., 2016.
The re-eutrophication of Lake Erie: Harmful algal blooms and hypoxia. Harmful
Algae 56, 44–66.
Wellen, C., Cappellen, P.V., Gospodyn, L., Thomas, J.L., Mohamed, M.N., 2020. An
analysis of the sample size requirements for acceptable statistical powerin
water quality monitoring for improvement detection. Ecol. Indic. 188, 106684.
Williams, M.R., King, K.W., Fausey, N.R., 2015. Contribution of tile drains to basin
discharge and nitrogen export in a headwater agricultural watershed. Agric.
Water Manag. 158, 42–50.
Williams, M.R., King, K.W., Ford, W., Buda, A.R., Kennedy, C.D., 2016. Effect of tillage
on macropore flow and phosphorus transport to tile drains. Water Resour. Res.
52, 2868–2882.
Wood, M., 2004. Statistical inference using bootstrap confidence intervals.
Significance 1 (4), 180–182.
Yuan, Z., Jiang, S., Sheng, H., Liu, X., Hua, H., Liu, X., Zhang, Y., 2018. Human
perturbation of global phosphorus cycle: changes and consequences. Environ.
Sci. Technol. 52, 2438–2450.
C. Nelligan, R.J. Sorichetti, M. Yousif et al. Journal of Great Lakes Research 47 (2021) 1689–1701
1701
... These patterns may shift over time as the lower Great Lakes basin is expected to become warmer and wetter from climate change (Meehl et al., 2007), potentially leading to changes in basin hydrology and solute transport. Consequences of climate change, such as more intense summer precipitation events and warmer winters with more freeze-thaw cycles, rain-on-snow events, and increased runoff could intensify nutrient loading in the lower Great Lakes basin (Long et al., 2015(Long et al., , 2014Mohamed et al., 2019;Nelligan et al., 2021). Additionally, physical landscape features, land use, and land management can also influence nutrient loads and transport mechanisms (Arbuckle and Downing, 2001;International Joint Commission, 2014;Mohamed et al., 2019). ...
... Steps towards these goals have been made via the Multi-Watershed Nutrient Study (MWNS), which aims to better understand nutrient dynamics in the lower Great Lakes basin (ECCC and OMOECC, 2018;Mohamed et al., 2019). To date, nutrient load estimates for the MWNS have been made using the Beale Ratio Estimator (Beale, 1962;Nelligan et al., 2021), and a critical step towards achieving the aforementioned goals is to develop more flexible load models that leverage higher frequency water quantity and quality data. ...
... Multiple linear regression is commonly used to predict nutrient loads from tributaries using surrogate data of discharge, turbidity, and seasonality (Leigh et al., 2019;Long et al., 2015;Nelligan et al., 2021;Robertson et al., 2018;Runkel et al., 2004). These predictions are often being used as calibration data for more complex coupled hydrology and water quality models (Dagnew et al., 2019). ...
Article
Full-text available
Eutrophication has re-emerged in the lower Great Lakes basin resulting in critical water quality issues. Models that accurately predict nutrient loading from streams are needed to inform appropriate nutrient management decisions. Generalized additive models (GAMs) that use surrogate data from sensors to predict nutrient loads offer an alternative to commonly applied linear regression and may better handle relationship non-linearities and skewed water quality data. Five years (2015–2020) of water quantity and quality data from 11 agricultural watersheds in southern Ontario were used to develop GAMs to predict total phosphorus (TP) and nitrate (NO3⁻) loads. This study aimed to 1) use GAMs to predict nutrient loads using both common and novel predictors and 2) quantify and examine the variability in seasonal and annual nutrient loads. Along with routine surrogate model predictors (i.e., flow, turbidity, and seasonality), the addition of the baseflow proportion and the hydrograph position of flow observations improved model performance. Conversely, including the antecedent precipitation index minimally affected model performance, regardless of constituent. Seasonal and annual patterns in TP and NO3⁻ load predictions mirrored that of the hydrologic regime. This study showed that parsimonious GAMs featuring novel model predictors can be used to predict nutrient loads while accounting for the partitioning of surface and subsurface flow paths and hysteresis between streamflow and water quality parameters that are frequently observed in a wide range of environments.
... The proportion of agricultural land planted to row crops was a significant predictor of stream NO 3 -N concentrations and export at these predominantly agricultural sub-watersheds. This suggests that 'agricultural intensification', herein defined as row crop expansion at the expense of mixed agricultural land likely contributed to observed increases in tributary NO 3 -N reported at numerous agricultural watersheds in southern Ontario (Eimers and Watmough, 2016;DeBues et al., 2019;Nelligan et al., 2021). Total dissolved P concentrations were also significantly correlated with row crop area, whereas TP was not associated with any form of LULC. ...
... Significant differences (Welch's T-Test) between flow conditions are indicated by the number of asterisks: * (p < 0.05); ** (p < 0.01); *** (p < 0.001). shifted earlier in seasonally snow-covered systems, associated with warmer winters and more rain vs. snow (Casson et al., 2012;Nelligan et al., 2021). These observations, as well as relatively high nutrient concentrations in the NGS compared with GS (see Fig. 3) suggest that multi-season monitoring is necessary to adequately quantify nutrient losses. ...
Article
Full-text available
Eutrophication remains the most widespread water quality impairment globally and is commonly associated with excess nitrogen (N) and phosphorus (P) inputs to surface waters from agricultural runoff. In southern Ontario, Canada, increases in nitrate (NO3-N) concentrations as well as declines in total phosphorus (TP) concentration have been observed over the past four decades at predominantly agricultural watersheds, where major expansions in row crop production at the expense of pasture and forage have occurred. This study used a space-for-time approach to test whether ‘agricultural intensification’, herein defined as increases in row crop area (primarily corn-soybean-winter wheat rotation) at the expense of mixed livestock and forage/pasture, could explain increases in NO3-N and declines in TP over time. We found a clear, positive relationship between the extent of row crop area within watersheds and NO3-N losses, such that tributary NO3-N concentrations and export were predicted to increase by ~0.4 mg/L and ~ 130 kg/km² respectively, for every 10% expansion in row crop area. There was also a significant positive relationship between row crop area and total dissolved phosphorus (TDP) concentration, but not export, and TP was not correlated with any form of landcover. Instead, TP was strongly associated with storm events, and was more sensitive to hydrologic condition than to landcover. These results suggest that pervasive shifts toward tile-drained corn and soybean production could explain increases in tributary NO3-N levels in this region. The relationship between changes in agriculture and P is less clear, but the significant association between dissolved P and row crop area suggests that increased adoption of reduced tillage practices and tile drainage may enhance subsurface losses of P.
... In agricultural watersheds in general, a very high proportion of total annual P loads may be exported in just a few large events, typically outside of the growing season (e.g., Long et al., 2015;Macrae et al., 2007;Nelligan et al., 2021;Plach et al., 2019). Hence, our study reinforces the critical importance of sampling across all seasons, as well as targeting high flow events, in order to derive accurate load and retention estimates (Lee et al., 2016). ...
Article
Full-text available
Extensive efforts are underway to reduce phosphorus (P) export from the Lake Erie watershed. On the Canadian side, the Thames River is the largest tributary source of P to Lake Erie’s western basin. However, the role of dams in retaining and modifying riverine P loading to the lake has not been comprehensively evaluated. We assessed whether Fanshawe Reservoir, the largest dam reservoir on the Thames River, acts as a source or sink of P, using year-round discharge and water chemistry data collected in 2018 and 2019. We also determined how in-reservoir processes alter P speciation by comparing the dissolved reactive P to total P ratio (DRP:TP) in upstream and downstream loads. Annually, Fanshawe Reservoir was a net sink for P, retaining 25% (36 tonnes) and 47% (91 tonnes) of TP in 2018 and 2019, respectively. Seasonally, the reservoir oscillated between a source and sink of P. Net P release occurred during the spring of 2018 and the summers of 2018 and 2019, driven by internal P loading and hypolimnetic discharge from the dam. The reservoir did not exert a strong influence on DRP:TP annually, but ratio increases occurred during both summers, concurrent with water column stratification. Our analysis demonstrates that Fanshawe Reservoir is not only an important P sink on the Thames River, but also modulates the timing and speciation of P loads. We therefore propose that the potential of using existing dam reservoirs to attenuate downstream P loads should be more thoroughly explored alongside source based P mitigation strategies.
Article
It is of critical importance to understand the relationships between crop yield, soil properties and topographic characteristics for agricultural management. This study’s objective was to compare techniques to quantify the relationship between soil and topographic characteristics for predicting crop yield using high-resolution data and analytical techniques. The study was conducted on a multiple field dataset located in Southwestern Ontario, Canada, where few studies have assessed the impact of applications for precision agriculture and machine learning (ML) to the soil property-yield relationship in this region. The dataset included 145,500 observations of corn and soybean yield, topographic and soil nutrient characteristics. The attributes considered for this study included pH, soil organic matter (OM) content, cation exchange capacity (CEC), soil test phosphorus, zinc (Zn), potassium (K), elevation and topographic wetness index. Multiple linear regression (MLR), artificial neural networks, decision trees and random forests were compared to identify methods able to relate soil properties and crop yields on a subfield scale (2 m). Random forests were the most successful at predicting yield with an R2 value of 0.85 for corn and 0.94 for soybeans. MLR was the least successful with an R2 of 0.40 for corn and 0.45 for soybeans. Cross-validation experiments showed that random forest models in most cases could predict low- and high-yield areas from fields excluded from training datasets, but this was not possible in all cases. Techniques tested the models and identified significant soil and topographic attributes when predicting yield, though the identification was subject to some uncertainty. These results suggest that ML techniques might be used to predict high yield areas of fields without existing yield maps, if those fields have similar relationships of soil properties to yield.
Article
Full-text available
Precipitation extremes have implications for many facets of both the human and natural systems, predominantly through flooding events. Observations have demonstrated increasing trends in extreme precipitation in North America, and models and theory consistently suggest continued increases with future warming. Here, we address the question of whether observed changes in annual maximum 1- and 5-d precipitation can be attributed to human influence on the climate. Although attribution has been demonstrated for global and hemispheric scales, there are few results for continental and subcontinental scales. We utilize three large ensembles, including simulations from both a fully coupled Earth system model and a regional climate model. We use two different attribution approaches and find many qualitatively consistent results across different methods, different models, and different regional scales. We conclude that external forcing, dominated by human influence, has contributed to the increase in frequency and intensity of regional precipitation extremes in North America. If human emissions continue to increase, North America will see further increases in these extremes.
Article
Full-text available
The build-up of soil phosphorus (P) beyond plant requirements can lead to a long-term legacy of P losses that could impair surface water quality. Using a database of ∼4,50,000 samples collected from 2001–2015 we report the level of soil P enrichment by soil type, land use and region and the time it would take for Olsen P to decline to agronomic targets (20–0 mg L–1) if P fertilizer was stopped. We also modeled the time it would take for water extractable P (WEP), an indicator of P losses in surface runoff, to decline to an environmental target (0.02 mg L–1). Some 63% of the samples were enriched beyond agronomic targets. The area-weighted median time to reach the agronomic target was predicted to occur within a year for 75% of samples but varied up to 11.8 years in some land uses. However, the area-weighted time to reach an environmental target was 26–55 years for the 50th and 75th percentile of areas. This indicates that while an agronomic target can be easily met, additional strategies other than stopping P fertilizer inputs are required to meet an environmental target.
Article
Full-text available
(This work has been prepared for Canadian Agri-food Policy Institute and the original article can be found at https://capi-icpa.ca/wp-content/uploads/2019/06/2019-06-14-CAPI-Vivekananthan-Kokulan-Paper-WEB.pdf) A substantial proportion of agricultural lands, which, in most cases are vulnerable to waterlogged conditions, rely on artificial drainage systems. subsurface drainage is enhanced through the installation of tile drains, which are perforated plastic or clay tubes installed in the vadose zone (unsaturated soil profile). Even though agronomically effective, tiles can also be the cause of several environmental problems. In Canada, extensive research on tile drainage has been done in Ontario and Quebec whereas little literature is available from other provinces (Christianson et al., 2015, 2016). Currently there is no literature that exclusively looks at the impacts of tile drainage from a pan-Canadian perspective. This paper has three objectives: (1) The first part of this study reviews the impact of tile drainage on edge of field runoff and agrochemical pollution in Canada; (2) The second part details the best management practices that can reduce tile nutrient losses without compromising the productivity; and (3) The last part of this study identifies and outlines research gaps and their practical importance in a changing climate from a policy perspective. Outcomes of this study will be useful for Canadian farmers, researchers and policy makers in identifying and adopting tools to increase the efficiency of subsurface drainage while minimizing its negative impacts.
Article
Many water quality managers seek to demonstrate reductions in pollutants after a remedial program or policy change of some sort is implemented, but there is little information in the literature to help guide the extent of water quality sampling that is required to be confident that a change has occurred. Statistical power refers to the likelihood of avoiding a Type II error in hypothesis testing. It is critical to examine statistical power levels to ensure results are not unduly influenced by insufficient quantity of data. This study presents the first published record, to the best of our knowledge, on sample size requirements to achieve acceptable levels of statistical power in hypothesis testing of annual water quality (nutrients) in streams. We examined 13 temperate agricultural watersheds spanning a gradient of size from 11 to 16,000 km 2 using data synthesized from long-term flow and water quality records. We found that achieving commonly accepted levels of statistical power (0.8) after reductions of 20% in load or flow-weighted mean concentration (FWMC) required an inordinate quantity of data (50-250 years for load, 10-120 years for FWMC), while achieving statistical power of 0.8 after reductions of 80% of load or FWMC required very little data (2-4 years for FWMC, 2-7 years for load). Load reductions of 40% required a range of 8-50 years of data depending on analyte, while FWMC reductions of 40% required 3-10 years of total phosphorus (TP) data, 5-25 years for soluble reactive phosphorus (SRP), and 2-6 years for nitrate (NO 3). We examined relationships among times to achieve statistical power and a number of common landscape descriptors (discharge, baseflow index, basin size, concentration-discharge slope) and found no dis-cernable relationships for either TP or SRP, whereas catchments with higher baseflow indices were found to have lower data requirements for achieving statistical power of 0.8 for NO 3. We also show through subsampling experiments that higher frequency sampling tended to reduce data requirements to achieve acceptable statistical power, though these gains diminish as the sample frequency increases. The information presented will help those tasked with watershed monitoring to design appropriate sampling regimes to ensure adequate data are obtained to detect change.
Article
Extensive time and financial resources have been dedicated to address nonpoint sources of nitrogen and phosphorus in watersheds. Despite these efforts, many watersheds have not seen substantial improvement in water quality. The objective of this study is to review the literature and investigate key factors affecting the lack of improvement in nutrient levels in waterways in urban and agricultural regions. From 94 studies identified in the academic literature, we found that although 60% of studies found improvements in water quality after implementation of Best Management Practices (BMPs) within the watershed, these studies were mostly modelling studies rather than field monitoring studies. For studies that were unable to find improvements in water quality after the implementation of BMPs, the lack of improvement was attributed to: lack of knowledge about BMP functioning, lag times, non-optimal placement and distribution of BMPs in the watershed, post-implementation BMP failure, and socio-political and economic challenges. We refer to these limiting factors as known unknowns. We also acknowledge the existence of unknown unknowns that hinder further improvement in BMP effectiveness and suggest that machine learning, approaches from the field of business and operations management, and long-term convergent studies could be used to resolve these unknown unknowns.
Article
Nutrient imbalance in groundwater and surface water resources can have severe implications on human and aquatic life, including contamination of drinking water sources and the degradation of ecosystems. A field-based watershed-scale study was completed to investigate nutrient dynamics and hydrologic processes in an agriculturally-dominant clay plain system within the Great Lakes Basin. Spatial and temporal variations of nitrogen and phosphorus were examined by sampling groundwater and surface water regularly over a period of one year (June 2017 to July 2018) for nutrients including nitrate, soluble reactive phosphorus, total dissolved phosphorus and total reactive phosphorus. Nitrate transport from surrounding agricultural land to surface water was intensified with an increase in precipitation events in spring and early winter and phosphorus transport to surface water was increased during freeze-thaw cycles in the winter. The results are pertinent to the improvement of current nutrient and water management policies in clay plain systems where nutrient imbalances in surface water are a concern.
Article
Phosphorus (P) loss to freshwater is a key driver of eutrophication, and understanding the scale and spatial distribution of potential P sources is a key pre-requisite for implementing policies for P management to minimize environmental impacts. Soil test P (STP) is a useful indicator of the accumulation of P in soils, but these data are not readily available for most agricultural land in Canada, so the cumulative P balance (P inputs as manure or fertilizer minus removal of P in crops) is calculated as a proxy for this value. Cumulative P balance is an important calculation within the indicator of risk of water contamination by P, so allocations of manure and fertilizer P to cropland were updated within the calculation of P balance, and for Ontario, data from 1961 to 1980 were added to account for P applications during that period. The STP concentrations were calculated from the resulting cumulative P balances. When compared with reported STP concentrations, the predicted concentrations showed a statistically significant regression at the national (R2 = 78%) and provincial scale (Ontario, R2 = 36%; Prince Edward Island, R2 = 36%; Manitoba, R2 = 72%; British Columbia, R2 = 40%). There was significant variation in the cumulative P balance across Canada, with the highest values corresponding with areas of high livestock density, whereas large zones of P deficit were detected across the Prairies.
Article
Eutrophication of freshwaters is already a problem in many regions globally and will probably worsen as human populations grow and consume more resources. The ability of researchers and governments to anticipate, mitigate, and restore eutrophic freshwaters in a cohesive, integrated manner suffers from key uncertainties in our understanding of the watershed-to-lake continuum. Here, we use Lake Erie and its watershed as an example of a system in which there is a pressing need to resolve these uncertainties. In recent history, Lake Erie both suffered and recovered from serious eutrophication and related issues. More recently, however, there has been a resurgence of eutrophication and associated harmful algal blooms in Lake Erie, with symptoms reminiscent of prior eutrophication. This resurgence has led the USA and Canadian governments to commit to substantially reducing P inputs into Lake Erie in an effort to control eutrophication. We illustrate how key uncertainties about Lake Erie and its watershed contribute to challenges we face in restoring this ecosystem and propose avenues for their resolution. To this end, we contend that an ecosystem approach will be required for managing the eutrophication of freshwaters.
Article
Agricultural P losses are a global economic and water quality concern. Much of the current understanding of P dynamics in agricultural systems has been obtained from rainfall-driven runoff, and less is known about cold-season processes. An improved understanding of the magnitude, form, and transport flow paths of P losses from agricultural croplands year round, and the climatic drivers of these processes, is needed to prioritize and evaluate appropriate best management practices (BMPs) to protect soil-water quality in cold regions. This study examines multiyear, year-round, high-frequency edge-of-field P losses (soluble reactive P and total P [TP]) in overland flow and tile drainage from three croplands in southern Ontario, Canada. Annual and seasonal budgets for water, P, and estimates of field P budgets (including fertilizer inputs, crop uptake, and runoff) were calculated for each site. Annual edge-of-field TP loads ranged from 0.18 to 1.93 kg ha yr (mean = 0.59 kg ha yr) across the region, including years with fertilizer application. Tile drainage dominated runoff across sites, whereas the contribution of tiles and overland flow to P loss differed regionally, likely related to site-specific topography, soil type, and microclimate. The nongrowing season was the dominant period for runoff and P loss across sites, where TP loss during this period was often associated with overland flow during snowmelt. These results indicate that emphasis should be placed on BMPs that are effective during both the growing and nongrowing season in cold regions, but that the suitability of various BMPs may vary for different sites.