Content uploaded by Christian Schürings
Author content
All content in this area was uploaded by Christian Schürings on Feb 05, 2024
Content may be subject to copyright.
Science of the Total Environment 917 (2024) 170583
Available online 1 February 2024
0048-9697/© 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Water Framework Directive micropollutant monitoring mirrors catchment
land use: Importance of agricultural and urban sources revealed
Nele Markert
a
,
b
,
1
, Christian Schürings
a
,
*
,
1
, Christian K. Feld
a
,
c
a
University Duisburg-Essen, Faculty of Biology, Aquatic Ecology, Universit¨
atsstr. 5, 45141 Essen, Germany
b
North Rhine-Westphalia Ofce of Nature, Environment and Consumer Protection (LANUV NRW), 40208 Düsseldorf, Germany
c
University Duisburg-Essen, Centre for Water and Environmental Research (ZWU), Universit¨
atsstr. 5, 45141 Essen, Germany
HIGHLIGHTS GRAPHICAL ABSTRACT
•Micropollutant concentrations were
related to urban land use individual
crop types.
•Percent urban area was strongly related
to pharmaceutical and industrial
chemicals.
•Pesticide concentrations mirror crop
type-specic pesticide applications.
•Micropollutants from diffuse and point
sources require different management.
ARTICLE INFO
Guest Editor: Daniel Hering
Keywords:
Agriculture
Crop type
Chemical monitoring
Ecological quality
Land use
Pollution source
River basin management
Urban area
ABSTRACT
River monitoring programs worldwide consistently unveil micropollutant concentrations (pesticide, pharma-
ceuticals, and industrial chemicals) exceeding regulatory quality targets with deteriorating effects on aquatic
communities. However, both the composition and individual concentrations of micropollutants are likely to vary
with the catchment land use, in particular regarding urban and agricultural area as the primary sources of
micropollutants. In this study, we used a dataset of 109 governmental monitoring sites with micropollutants
monitored across the Federal State of North Rhine-Westphalia, Germany, to investigate the relationship between
high-resolution catchment land use (distinguishing urban, forested and grassland area as well as 22 different
agricultural crop types) and 39 micropollutants using Linear Mixed Models (LMMs). Ecotoxicological risks were
indicated for mixtures of pharmaceutical and industrial chemicals for 100 % and for pesticides for 55 % of the
sites. The proportion of urban area in the catchment was positively related with concentrations of most phar-
maceuticals and industrial chemicals (R
2
up to 0.54), whereas the proportions of grassland and forested areas
generally showed negative relations. Cropland overall showed weak positive relationships with micropollutant
concentrations (R
2
up to 0.29). Individual crop types, particularly vegetables and permanent crops, showed
higher relations (R
2
up to 0.46). The ndings suggest that crop type-specic pesticide applications are mirrored
in the detected micropollutant concentrations. This highlights the need for high-resolution spatial land use to
* Corresponding author at: Department of Aquatic Ecology, Faculty of Biology, University Duisburg-Essen, Universit¨
atsstr. 5, 45141 Essen, Germany.
E-mail address: christian.schuerings@uni-due.de (C. Schürings).
1
Authors Nele Markert and Christian Schürings contributed equally to the work
Contents lists available at ScienceDirect
Science of the Total Environment
journal homepage: www.elsevier.com/locate/scitotenv
https://doi.org/10.1016/j.scitotenv.2024.170583
Received 30 November 2023; Received in revised form 28 January 2024; Accepted 29 January 2024
Science of the Total Environment 917 (2024) 170583
2
investigate the magnitude and dynamics of micropollutant exposure and relevant pollution sources, which would
remain undetected with highly aggregated land use classications. Moreover, the ndings imply the need for
tailored management measures to reduce micropollutant concentrations from different sources and their related
ecological effects. Urban point sources, could be managed by advanced wastewater treatment. The reduction of
diffuse pollution from agricultural land uses requires additional measures, to prevent pesticides from entering the
environment and exceeding regulatory quality targets.
1. Introduction
Globally, societies face three major planetary crises: biodiversity
loss, climate change and chemical pollution (UNEP, 2021). The latter is
associated with negative effects on biodiversity, ecosystem functioning
(Groh et al., 2022; Sigmund et al., 2022) as well as human health (Fuller
et al., 2022) and may impose long-term economic effects for societies
(Grandjean and Bellanger, 2017). >350,000 chemicals have so far been
registered for production and use worldwide (Wang et al., 2020) and
many compounds can be found in the environment at environmentally
relevant concentrations (Schwarzenbach et al., 2006). Hence, a reduc-
tion of chemical pollution is essential to remain within the planetary
boundaries, which describe the natural limits for human impact to
prevent unacceptable environmental change (Diamond et al., 2015;
Persson et al., 2022; Rockstr¨
om et al., 2009). Aquatic ecosystems in
particular are strongly impaired by a multitude of micropollutants
including pharmaceuticals (Fekadu et al., 2019), pesticides (Liess et al.,
2021; Sch¨
afer et al., 2011) and industrial chemicals (Koumaki et al.,
2018), which have previously been associated with ecological degra-
dation (Lemm et al., 2021; Posthuma et al., 2020; Schürings et al.,
2024a).
International policies (e.g., European Green Deal) and environ-
mental legislation (European Commission, 2019, 2020; UNEP, 2017)
have been developed to promote the sustainable use of chemical sub-
stances and achieve a toxic-free environment. Comprehensive programs
to monitor chemical pollution already exist (e.g., EU Water Framework
Directive (WFD), Directive 2000/60/EC), which, however, cannot
adequately address the numerous substances that are applied (Malaj
et al., 2014; Moschet et al., 2014; Weisner et al., 2022). The risk
assessment of micropollutants is typically based on the comparison of its
environmental concentrations with substance-specic ecotoxicological
assessment values. For several micropollutants (i.e., priority substances
and river basin-specic pollutants) environmental quality standards
(EQS) and further ecotoxicologically derived assessment values are set
by the WFD and related national legislations (e.g., the German surface
waters directive, OGewV, 2016).
The sources of micropollutants and the pathway of pollution vary
between substances, while two major pathways of pollution can be
distinguished. Point sources constitute spatially explicit points of
pollution, for example, efuents of industrial or municipal wastewater
treatment plants (WWTP) in urban areas (Finckh et al., 2022; Loos et al.,
2013). Contrastingly, diffuse sources of pollution cannot be attributed to
explicit efuents, but comprise rather broad-scale pathways such as
surface and groundwater run-off from agricultural areas into the aquatic
environment (Harrison et al., 2019; Wiering et al., 2020). Agricultural
practices and pesticide applications vary between crop types (Andert
et al., 2015). In particular, the high pesticide application rates for per-
manent crops and vegetables (Dachbrodt-Saaydeh et al., 2021) result in
enhanced and ecotoxicologically relevant concentrations for riverine
biota (Bereswill et al., 2012; Schulz, 2001; Xing et al., 2012). In contrast,
forage maize cultivations are often highly fertilized (Britz and Witzke,
2014), but associated with rather small amounts of pesticides, nearly
exclusively herbicides, whereas the use of pesticides on grassland is very
limited (Dachbrodt-Saaydeh et al., 2021; Riedo et al., 2022). Forested
areas in general show low relationships to micropollutant concentra-
tions and often relate positively to river health (Goss et al., 2020).
In this study, we investigated the relationships between catchment
land use and individual micropollutant concentrations in German rivers.
More specically, we aimed to test whether differences in concentration
patterns are observed for specic crop types, revealing crop type-specic
pesticide applications that are reported by Andert et al. (2015) and
Dachbrodt-Saaydeh et al. (2021). This differentiation between the
sources of pollution as well as the source-specic pollutants is deemed of
primary importance for water management, because the management of
diffuse and point sources would require different management strate-
gies. The following research questions were formulated:
(1) Which micropollutants do exceed the environmental quality
targets that are set by available environmental regulations and
ecotoxicological risks assessments?
(2) Do the monitored micropollutant concentrations reect the pro-
portions of urban, forested and agricultural areas in the catch-
ment of monitoring sites?
(3) Do agricultural pesticide concentrations relate to specic crop
types, thus reecting crop-specic pesticide application rates?
2. Materials and methods
2.1. Study area
In total, 109 micropollutant monitoring sites were included in this
study (Fig. 1). The sites are located in the Federal State of North Rhine-
Westphalia (NRW), Germany and cover lowland (altitude below 200 m
a.s.l) and mountainous regions (altitude 200–800 m a.s.l) as well as
small (catchment area 5–100 km
2
), mid-sized (catchment area
100–1000 km
2
) and large rivers. Catchment area ranged 5–2834 km
2
(median: 326 km
2
; see Supplementary Material Table A1 for detailed
site characterization).
2.2. Micropollutant monitoring and ecotoxicological risk assessment
Data on micropollutant concentrations originate from WFD-related
chemical monitoring programs of the North Rhine-Westphalian Ofce
of Nature, Environment and Consumer Protection and regional water
boards. Sampling was based upon grab samples of surface water (see
OGewV (2016) and LAWA (2019) for details on sampling and analysis)
and occurred between 2016 and 2019. For each site one sampling year
was selected that temporally matched the reference timing of land use
data (2016/2017) best (Section 2.3). In total, 39 micropollutants (19
pesticides, 14 pharmaceuticals and six industrial chemicals including
personal care products and household chemicals; Table 1) were selected
for this study because of their ecotoxicological relevance, i.e. they
constitute priority substances, river basin-specic pollutants or candi-
date substances on the watch list listed by the WFD and were identied
as ecotoxicologically relevant by previous studies (e.g., Ginebreda et al.,
2014; Gustavsson et al., 2017; Markert et al., 2020).
To quantify the ecotoxicological risk of micropollutants (research
question 1), we calculated individual risk quotients (RQ) for each sub-
stance, i.e., the quotient of the measured concentration divided by the
substance-specic assessment value (Backhaus and Faust, 2012). The
estimation of chronic risks during longer exposure periods was based on
annual mean concentrations of individual micropollutants (OGewV,
2016). For all substances, the number of measured concentrations
ranged between one and 24 values with a mean of four measured values
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
3
per substance, site and year (Table A2 Supplementary Material).
Assessment values were derived from environmental quality standards
(EQS) from the WFD, national legislation (OGewV, 2016) and validated
ecotoxicological data (e.g., EQS proposals and predicted no effect con-
centrations) in accordance with the technical guidance for deriving
environmental quality standards (European Commission, 2017; Markert
et al., 2020). To account for combined risks of micropollutant mixtures,
the sum of individual RQs (SUM RQ) was calculated for each site
(Backhaus and Faust, 2012; Markert et al., 2020). RQ and SUM RQ
values above one indicate that individual and combined micropollutant
concentrations exceed the ecotoxicological effect levels and thus
constitute a potential (mixture) risk.
Since both the number of micropollutants and the composition of
substances measured at each site varied among the sites, data gaps
occurred for individual micropollutants that ranged between 1 and 32 %
of the sites (mean across all substances: 12 %). Missing values were
imputed using an iterative imputation algorithm based on random for-
ests (missForest), which has previously been shown to perform well for
data gaps extending up to 30 % (or even 50 %) of the values (Stekhoven
and Bühlmann, 2012; Tang and Ishwaran, 2017). Left-censored data, i.e.
concentrations below the limit of quantication (LOQ) were replaced by
half of the LOQ value for pharmaceuticals and industrial chemicals and
by zero for pesticides. This approach was chosen since pharmaceuticals
and industrial chemicals are ubiquitously and continuously released into
the aquatic environment (Hernando et al., 2006; McEneff et al., 2014),
where substitution with zero might lead to a critical underestimation of
concentrations. In contrast, pesticides tend to show seasonal
concentrations patterns (Vormeier et al., 2023b), where substitution
with half the LOQ might result in arbitrarily high concentration ranges.
2.3. Catchment land use
For each sampling site, we quantied the proportions of forested,
urban and agricultural terrestrial land use in the catchment area up-
stream of the site. Catchment delineation was based on a digital eleva-
tion model (©dl-zero-de/2.0, Geobasis NRW, 10 m resolution) in ESRI
ArcView 3.3 subsequently checked visually for correctness and clipped
with altogether 23 different crop types (including grassland) using ESRI
ArcGIS Pro 2.9.0 and Spyder (Phyton 3.7.0). Crop type-specic land uses
for 2017 were derived from satellite images (Sentinel-2, Landsat 8 and
Sentinel-1, 10 m resolution) (Blickensd¨
orfer et al., 2022). The propor-
tion of urban and forested areas in the catchment for 2016 were derived
from Grifths et al. (2019) and quantied alike crop type-specic land
use. To statistically account for the temporal variation of micropollutant
data (2016–2019) and land use/cover data (2016–2017), the year of
micropollutant sampling was included as a random factor in the models
(see below). However, the inuence is likely minor, as Schürings et al.
(2024b) found no major differences between year when comparing the
effect of land use on river biota using the land use data of Blickensd¨
orfer
et al. (2022) of the years 2017 and 2018.
To quantify and compare catchment land uses, the 23 different crop
types (including grassland) and urban and forested area were assigned
several categories (Table 2). Except for grassland, all crop types were
merged into a category ‘cropland’ to account for general effects of
Fig. 1. Location of micropollutant monitoring sites in the Federal State of North Rhine-Westphalia (NRW), Germany with adjacent land use derived from Blick-
ensd¨
orfer et al. (2022) and Grifths et al. (2019) using ESRI ArcGIS Pro.
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
4
intensive agricultural land use. In the category, maize and cereals were
dominant. Grassland was kept separate because it constitutes a rather
extensive form of agricultural land use. To analyze crop type-related
effects, the 22 individual crop types were categorized into maize, ce-
reals, oilseeds, permanent crops and vegetables. To further differentiate
between different vegetables that are known to be associated with high
pesticide application rates (Dachbrodt-Saaydeh et al., 2021) asparagus,
strawberries and onions were additionally kept as individual categories
(Table 2).
2.4. Statistical analyses
To investigate, whether micropollutant mixture risks are reected by
catchment land uses (research question 2), individual linear regression
models of the SUM RQ of industrial chemicals, pharmaceuticals and
pesticides (incl. The sub-groups herbicides, fungicides and insecticides)
with catchment land uses (cropland, urban area, forest) as predictors
were visualized (R package ggplot2 (Wickham, 2016) with lm smooth
function; R Core Team, 2020).
For micropollutant-specic analyses, separate linear mixed models
(LMMs) were tted for each possible combination of four land use cat-
egories (urban, forest, grassland, cropland as well as individual crop
types) and 39 micropollutants, with the micropollutant concentration as
response and the proportion of one land use type as the predictor (i.e.
the xed effect in the model). Ecoregion (lowlands, low mountains) and
the year of micropollutant sampling were included in each LMM as
random effects. No spatial autocorrelation was assumed, as all sites have
distinct sub-catchments and additionally, prior studies using the same
Table 1
Statistical parameters and calculated risk quotients of micropollutants
Application of pesticides (‘plant protection products’) refers to substance-related approvals in Germany (BVL, 2023). Three pesticide subgroups were distinguished:
herbicides (H), fungicides (F) and insecticides (I). Risk quotients (RQ) were calculated as quotients of measured concentrations and assessment values in accordance
with the technical guidance for deriving environmental quality standards (European Commission, 2017). Individual CAS numbers and assessment values used for
ecotoxicological risk assessment of micropollutants are listed in Table A2, Supplementary Material. Concentrations below the limit of quantication (LOQ) were
replaced by half of the LOQ value for pharmaceuticals and industrial chemicals and by zero for pesticides.
Concentration (
μ
g/L) Risk quotient (RQ)
Micropollutant
group
Substance Application Min. Max. Mean SD Min. Max. %Sites with RQ
>1
Industrial chemicals Benzo(a)pyrene Polycyclic aromatic hydrocarbon- 0.0003 0.010 0.002 0.001 1.47 56.44 100 %
Industrial chemicals Benzotriazole Diverse, i.a. corrosion inhibitor 0.025 5.267 1.156 0.991 0.001 0.28 0 %
Industrial chemicals Bisphenol A Diverse, mainly plastic production 0.005 0.135 0.040 0.029 0.01 0.40 0 %
Industrial chemicals Fluoranthene Polycyclic aromatic hydrocarbon 0.001 0.028 0.007 0.004 0.11 4.48 44 %
Industrial chemicals Galaxolide
(HHCB)
Fragrance 0.032 0.243 0.104 0.051 0.007 0.06 0 %
Industrial chemicals Triclosan
1
Disinfecting and preserving agent 0.001 0.014 0.006 0.002 0.06 0.70 0 %
Pharmaceuticals Azithromycin Antibiotic agent 0.005 0.079 0.027 0.015 0.28 4.15 60 %
Pharmaceuticals Bezabrate Antiepileptic agent 0.005 0.610 0.040 0.091 0.002 0.27 0 %
Pharmaceuticals Carbamazepine Antiepileptic agent 0.005 0.710 0.119 0.124 0.01 1.42 2 %
Pharmaceuticals Ciprooxacin Antibiotic agent 0.005 0.018 0.013 0.002 0.06 0.20 0 %
Pharmaceuticals Clarithromycin Antibiotic agent 0.006 0.310 0.039 0.052 0.06 3.10 7 %
Pharmaceuticals Clindamycin Antibiotic agent 0.006 0.054 0.021 0.010 0.14 1.23 3 %
Pharmaceuticals Clobric acid Antibiotic agent 0.001 0.062 0.012 0.006 0.0001 0.01 0 %
Pharmaceuticals Diclofenac Pain medication 0.005 3.900 0.352 0.586 0.1 78.00 74 %
Pharmaceuticals Erythromycin Antibiotic agent 0.003 0.200 0.020 0.023 0.02 1.00 0 %
Pharmaceuticals Ibuprofen Pain medication 0.004 0.250 0.028 0.039 0.40 25.00 98 %
Pharmaceuticals Naproxen Pain medication 0.002 0.840 0.050 0.108 0.001 0.49 0 %
Pharmaceuticals Paracetamol Pain medication 0.005 0.107 0.025 0.021 0.0001 0.002 0 %
Pharmaceuticals Sulfamethoxazol Antibiotic agent 0.005 0.600 0.071 0.086 0.008 1.00 0 %
Pharmaceuticals Venlafaxine Antidepressant agent 0.005 1.000 0.085 0.126 0.006 1.14 1 %
Pesticides (H) Aclonifen Field crop, Vegetable 0.000 0.006 0.001 0.001 0.00 0.05 0 %
Pesticides (F) Azoxystrobin Field crop, Vegetable, Fruit, Wine, Biocide, Hop,
Ornamental plant
0.000 0.781 0.016 0.094 0.00 3.91 3 %
Pesticides (H) Chlortoluron Field crop 0.000 0.042 0.003 0.007 0.00 0.11 0 %
Pesticides (I) Clothianidin Field crop, Vegetable, Ornamental plant, Biocide 0.000 0.016 0.000 0.002 0.00 0.20 0 %
Pesticides (H) 2,4-D Field crop, Fruit, Ornamental plant 0.000 0.031 0.001 0.003 0.00 0.15 0 %
Pesticides (H) Dimethenamid Field crop, Vegetable, Fruit, Ornamental plant 0.000 0.617 0.010 0.063 0.00 2.37 1 %
Pesticides (H) Diuron
2
Field crop, Fruit, Wine, Biocide 0.000 0.173 0.006 0.023 0.00 0.87 0 %
Pesticides (H) Ethofumesat Field crop, Vegetable 0.000 0.035 0.001 0.004 0.00 0.01 0 %
Pesticides (H) Flufenacet Crop, Vegetable, Fruit, Ornamental plant 0.000 0.097 0.008 0.014 0.00 2.42 3 %
Pesticides (I) Imidacloprid
2
Field crop, Vegetable, Fruit, Wine, Biocide,
Ornamental plant, Hop
0.000 0.159 0.006 0.017 0.00 79.50 43 %
Pesticides (H) Isoproturon
2
Field crop, Ornamental plant, Biocide 0.000 0.053 0.003 0.009 0.00 0.18 0 %
Pesticides (H) MCPA Field crop, Hop, Fruit, Ornamental plant 0.000 0.163 0.010 0.022 0.00 0.25 0 %
Pesticides (H) Metazachlor Field crop, Vegetable, Ornamental plant 0.000 0.195 0.004 0.024 0.00 0.49 0 %
Pesticides (H) Metolachlor Field crop, Vegetable 0.000 0.117 0.004 0.015 0.00 0.58 0 %
Pesticides (H) Nicosulfuron Field crop 0.000 0.020 0.002 0.003 0.00 2.22 4 %
Pesticides (F) Tebuconazole Field crop, Biocide 0.000 0.161 0.004 0.017 0.00 0.28 0 %
Pesticides (H) Terbuthylazine Field crop, Vegetable 0.000 0.243 0.012 0.034 0.00 0.49 0 %
Pesticides (H) Terbutryn
2
Biocide 0.000 0.103 0.008 0.016 0.00 1.59 3 %
Pesticides (I) Thiacloprid
2
Field crop, Vegetable, Fruit, Ornamental plant 0.000 0.014 0.001 0.002 0.00 1.43 2 %
1
Triclosan was used as a biocidal active substance for human hygiene, disinfection and preservation but the approval was withdrawn in the EU in 2016 (ECHA,
2023); yet it is used in cosmetic and personal care products (European Commission, 2014).
2
The substances, diuron, imidacloprid, isoproturon, thiacloprid and terbutryn have been banned for (outdoor) use as ‘plant protection product’ in the EU since 2002
(terbutryn), 2007 (diuron), 2016 (isoproturon), 2018 (imidacloprid) and 2020 (thiacloprid) (BVL, 2023b), but are still approved as biocidal active substances for
preservatives, for example, in facade paint or construction material (diuron, terbutryn, isoproturon; ECHA, 2023), for insecticide (imidacloprid, ECHA, 2023) or as
veterinary medicinal products (imidacloprid, EMA, 2021).
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
5
data basis did not nd strong autocorrelation (e.g. Schürings et al.,
2024b). A gaussian distribution was selected for LMMs, as preliminary
analyses of the data using Generalized Additive Models (GAMs) sug-
gested that a linear relationship of xed effects can be assumed. LMMs
were run in R with the ‘gamlss’ package (v5.2–0, Rigby and Stasino-
poulos, 2005).
In each of the models, 70 % of the data were bootstrapped 1000 times
to calculate a mean-peudo-R
2
(from here on referred to as R
2
) for the
xed effect including condence intervals. The R
2
of the xed effect was
calculated as the squared correlation between the tted response and the
predicted response, solely based on the xed effect. Alongside the in-
dividual R
2
(and condence intervals) for individual micropollutant
concentrations, an overall R
2
was calculated for each group of micro-
pollutants (concentrations of pesticides, pharmaceuticals, and industrial
chemicals), using a random effect model with the metafor package
(Viechtbauer, 2010). Effect size (Fig. 2) was based on the individual and
grouped R
2
, however, a sign was added to the plot axis to distinguish
positive and negative regression coefcients, i.e. positive or negative
effects of land uses on micropollutant concentrations. To analyze
whether pesticide concentrations reect crop-specic pesticide appli-
cation rates (research question 3), additional models were calculated for
the individual crop types, i.e., maize, cereals, oilseeds, permanent crops,
vegetables as well as asparagus, strawberries and onions, following the
same procedure.
3. Results
3.1. Ecotoxicological risk assessment
Six out of the 39 micropollutants frequently (i.e. at >10 % of the
sampling sites) exceeded regulatory assessment values (RQ >1) and
hence imposed individual ecotoxicological risks: ibuprofen, diclofenac,
azithromycin (pharmaceuticals), benzo(a)pyrene, uoranthene (indus-
trial chemicals), and imidacloprid (pesticide; Table 1). Furthermore, the
pharmaceuticals clarithromycin, clindamycin, carbamazepine and ven-
lafaxine as well as the pesticides thiacloprid, azoxystrobin, nicosulfuron,
ufenacet, dimethenamid and terbutryn were found in concentrations
exceeding the assessment values, although at <10 % of the sites. In
contrast to individual RQs, when considering chemical mixtures, eco-
toxicological risks of pharmaceuticals and industrial chemicals were
indicated at 100 % of the sites (SUM RQ >1, Table A3 Supplementary
Material). Mixture Risks of pesticide mixtures were evident at 55 % of
Table 2
Statistical parameters of the proportion of land uses in the catchments upstream
of the sampling sites and categorization of crop types into sub-groups.
Land use Min.
%
Max.
%
Mean
%
SD %
Forest 0.04 74.94 31.78 20.07
Urban area 4.05 59.09 20.04 11.51
Grassland 2.89 34.86 12.57 6.54
Cropland 0.11 76.77 29.87 21.66
Individual crop type
Maize
(silage maize, grain maize)
0.00 76.91 29.13 17.61
Cereals
(wheat, rye, barley, oat, other cereals)
5.93 87.39 46.75 15.80
Oilseeds
(rapeseed, sunowers)
0.00 25.38 5.56 5.68
Permanent crops
(vineyards, hops, orchards)
0.00 27.52 2.66 4.83
Vegetables
(potatoes, sugar beets, legumes,
strawberries, asparagus, onions, carrots,
other vegetables)
1.51 48.96 13.72 10.46
Asparagus 0.00 7.20 1.72 1.65
Strawberries 0.00 24.06 2.72 4.23
Onions 0.00 4.53 0.52 0.95
Fig. 2. Relationship between the proportion of cropland, urban and forested areas in the catchment and mixture risk quotients (log SUM RQ) of industrial chemicals,
pharmaceuticals, herbicides, fungicides and insecticides. The solid line marks the t of a linear regression model with 95 % condence interval indicated in gray;
dashed red lines mark the threshold of SUM RQ =1, which translates to 0 along the log-transformed y-axis.
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
6
the sites, with insecticides (44 % of the sites) dominating the risk
assessment over herbicides (27 %) and fungicides (3 %).
The relationship between risks of chemical mixtures and catchment
land use varied among micropollutant groups (Fig. 2) but showed strong
positive relationships of cropland with herbicides (R
2
=0.31) and fun-
gicides (R
2
=0.30) and of urban area with pharmaceuticals (R
2
=0.38).
The proportion of forested area was negatively related with risks of
chemical mixtures of all micropollutant groups, which was most pro-
nounced for herbicides (R
2
=0.45) and fungicides (R
2
=0.33).
3.2. Link between micropollutant concentrations and land use
The proportion of urban and forested areas, cropland and grassland
revealed clear differences in their relationship to individual micro-
pollutants and micropollutant groups (Figs. 3 and 4). Urban land use
(Fig. 3a) was positively related to numerous pollutants, particularly to
pharmaceuticals (R
2
=0.31) and industrial chemicals (R
2
=0.39), while
its relationship with pesticides (R
2
=0.02) was almost negligible.
Among the pharmaceuticals, antibiotics (azithromycin: R
2
=0.54,
clindamycin. R
2
=0.45 and clarithromycin: R
2
=0.44) revealed the
strongest relationship to proportion of urban areas. The effect sizes for
industrial chemicals were in a similar range and showed particular
strong relationships to galaxolide (R
2
=0.51) and triclosan (R
2
=0.48).
The strongest individual relationship of a pesticide to urban area was
found for terbutryn (R
2
=0.27).
Cropland showed a weak, but positive relationship to pesticides
(pooled R
2
=0.08), while its relationship to pharmaceuticals and in-
dustrial chemicals was negligible (both R
2
=0.02). The strongest indi-
vidual relationship between proportion of cropland and pesticides were
found for ufenacet (R
2
=0.29) and nicosulfuron (R
2
=0.21), indi-
vidual relationships to pharmaceuticals and industrial chemicals were
negligible (R
2
up to 0.04), except for a weak negative relation with
ciprooxacin (R =0.13).
Grassland (Fig. 4a) showed weak and negative relationships to all
micropollutant groups, with pooled effect sizes of R
2
=0.04, R
2
=0.07
and R
2
=0.06 for pesticides, pharmaceuticals and industrial chemicals,
respectively. Individual effects of the proportion of grassland were most
pronounced and negative for the pharmaceutical ciprooxacin (anti-
biotic, R
2
=0.17) and for the pesticide ufenacet (herbicide, R
2
=0.13).
Eventually, forest (Fig. 4b) showed weak and negative relationships to
all micropollutant groups with pooled effect sizes of R
2
=0.07, R
2
=
0.08 and R
2
=0.08 for pesticides, pharmaceuticals, and industrial
Fig. 3. Relationship (effect size) of the proportion of urban areas (a) and cropland (b) in the catchment with micropollutant concentrations. Effect sizes represent
model ts (pseudo-R
2
) derived from bootstrapped (n =1000) univariate linear mixed models (LMM) with 95 % condence intervals indicated in brackets. Negative
signs were added to account for negative relationships (i.e., negative regression coefcients) – although R
2
values are positive by denition.
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
7
chemicals, respectively. Again, the strongest individual relationships to
the proportion of forested areas were found for ciprooxacin (R
2
=0.33)
and ufenacet (R
2
=0.22).
3.3. Link between micropollutant concentrations and individual crop
types
In contrast to the overall weak effects of the proportion of cropland
in the catchment on the majority of micropollutants as described in the
previous section, much more pronounced relationships on pesticides
were evident for individual crop types. Permanent crops (vineyards,
hops and orchards) were strongly related to two insecticides: thiacloprid
(R
2
=0.46) and imidacloprid (R
2
=0.26.; Fig. 5a). Vegetables also
showed strong relationships to both insecticides (imidacloprid: R
2
=
0.28, thiacloprid: R
2
=0.23) and in addition to the herbicides aclonifen
(R
2
=0.30) and dimethenamid (R
2
=0.25; Fig. 5b). Imidacloprid and
thiacloprid were (at the time of the data) approved for, among others,
applications to fruits and hops (both) and viticulture (imidacloprid),
while aclonifen and dimethenamid were approved for various eld
crops and vegetables (Table 1; BVL, 2023). Notably, both insecticides
imidacloprid and thiacloprid have been banned for (outdoor) use as
plant protection product in the EU in 2018 and 2020. Cereals and maize
constitute the dominating crop types in this dataset of the Federal State
of North Rhine-Westphalia and showed the strongest relationships to
ufenacet (R
2
=0.29 and R
2
=0.27, respectively) and nicosulfuron (R
2
=0.16 for both crop types; Fig. A4.1, Supplementary Material). These
herbicides are approved for eld crops including maize (both) and ce-
reals such as winter barley, winter rye, winter soft wheat (only
Fig. 4. Relationship (effect size) of the proportions of grassland (a) and forested areas (b) in the catchment with micropollutant concentrations. Effect sizes represent
model ts (pseudo-R
2
) derived from bootstrapped (n =1000) univariate linear mixed models (LMM) with 95 % condence intervals indicated in brackets. Negative
signs were added to account for negative relationships (i.e., negative regression coefcients) – although R
2
values are positive by denition.
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
8
ufenacet; Table 1, BVL, 2023).
Strong relationships were also found between the proportion of
strawberry elds and the herbicides dimethenamid (R
2
=0.40) and
diuron (R
2
=0.33), and between the proportion of asparagus elds, and
the herbicides dimethenamid and MCPA (both R
2
=0.33; Fig. A4.2,
Supplementary Material). Interestingly, MCPA has been approved only
for pome and stone fruits (e.g., apple or peach), but not for other fruits or
vegetables (BVL, 2023). Furthermore, the proportion of onion elds was
related to dimethenamid (herbicide, R
2
=0.38), imidacloprid (insecti-
cide, R
2
=0.36) and aclonifen (herbicide, R
2
=0.33; Fig. A4.3, Sup-
plementary Material), all of which are approved for – and applied to
cultivations of onions (BVL, 2023). The proportion of oilseeds (e.g.,
rapeseed, sunowers) showed comparatively weak relationships with
pesticides (max. R
2
=0.11 for 2,4-D; Fig. A4.3, Supplementary
Material).
4. Discussion
4.1. Micropollutant concentrations exceed regulatory assessment values
Several micropollutants were found to exceed existing regulatory
assessment values at multiple sites. Especially, concentrations of phar-
maceuticals, the non-steroidal anti-inammatory drugs diclofenac and
ibuprofen, the antibiotic azithromycin as well as concentrations of
polycyclic aromatic hydrocarbons, benzo(a)pyrene and uoranthene,
exceeded assessment values, thus indicating a widespread and enhanced
ecotoxicological risk for riverine biota (e.g, Beckers et al., 2018; Beek
et al., 2016; Markert et al., 2020). For pesticides, ecotoxicological risks
Fig. 5. Relationship (effect size) of the proportion of permanent crops (a) and vegetables (b) in the catchment with micropollutant concentrations. Effect sizes
represent model ts (pseudo-R
2
) derived from bootstrapped (n =1000) univariate linear mixed models (LMM) with 95 % condence intervals indicated in brackets.
Negative signs were added to account for negative relationships (i.e., negative regression coefcients) – although R
2
values are positive by denition.
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
9
were evident for <10 % of sites and found only for the insecticide imi-
dacloprid, while other pesticides (thiacloprid, azoxystrobin, nic-
osulfuron, ufenacet, dimethenamid) exceeded assessment values at <5
% of sites. Pesticide risk assessment, however, substantially changed,
when risks of chemical mixtures were evaluated, which exceeded the
threshold of one (SUM RQ >1) at 55 % of the sites. Thus, while for
pharmaceuticals and industrial chemicals ecotoxicological risk were
already driven by single substances, pesticide risks originate primarily
from joint mixture risks.
Based on the (mixture) toxicity risk quotients calculated in this
study, adverse effects of micropollutant exposure on river biota are very
likely. The calculated risks, however, might underestimate actual
toxicity risks, because micropollutant monitoring was based on grab
sampling. In contrast to high-frequent and event-based monitoring, grab
sampling is likely to miss the peak concentrations of micropollutants,
pesticides in particular, as they often occur directly after stormwater
rain events and with strong seasonal differences (Halbach et al., 2021;
Munz et al., 2017; Rumschlag et al., 2019; Weisner et al., 2021; Weisner
et al., 2022). When using event-based sampling and increasing sampling
efforts (Liess et al., 2021; Rumschlag et al., 2019; Weisner et al., 2022),
measured concentrations can exceed concentrations found by grab
samples by more than an order of magnitude. Moreover, the detection of
(mixture) toxicity risks may also be limited by the selection and number
of regularly measured micropollutants and their individual detection
limits (Malaj et al., 2014; Moschet et al., 2014; Weisner et al., 2022).
Toxicity risk assessment is often biased by missing or left-censored data
(i.e. unknown concentrations between zero and the technical limit of
quantication; von der Ohe et al., 2011), which constitutes a main
obstacle for multivariate comparisons of effects among sites and in
relation to potential sources and biological responses. Despite these
sources of uncertainty in the chemical risk assessment, however, our
ndings conrm those of previous studies (e.g., Finckh et al., 2022;
Halbach et al., 2021; Markert et al., 2020): regulatory assessment values
for micropollutants are frequently exceeded in the aquatic environment
so that freshwater biota are exposed to critical levels of both individual
micropollutants and mixtures thereof.
4.2. Micropollutant concentrations relate to catchment land uses
Our results point at clear relationships between particular land use
types and individual micropollutants as well as micropollutants groups.
Cropland was related to pesticide concentrations while relations to
pharmaceuticals and industrial chemicals were negligible. This is partly
in line with recent studies describing agriculture as a main determinant
for pesticide exposure (Sz¨
ocs et al., 2017). Previous studies also sug-
gested urban point sources to substantially contribute to pesticide
pollution due to the use of pesticides in urban gardens or as biocidal
products, for example in façade paints (Münze et al., 2017; Tauchnitz
et al., 2020). We, however, found the major part of the monitored pes-
ticides to relate to the proportion of agricultural areas in the catchment,
except for terbutryn, which in fact is no longer approved for agricultural
use but for biocidal façade paint; thus, this herbicide showed a stronger
relationship to proportion of urban areas. Urban areas were found to be
strongly associated with individual and mixture risks of pharmaceuticals
and industrial chemicals (Bradley et al., 2020; Ebele et al., 2017).
Notably, detailed characteristics of urban areas, such as the population
density or the proportion of industrial areas, were not specied in this
study but may inuence the association with micropollutant concen-
trations (Mandaric et al., 2018). Nonetheless, the proportion of cropland
and urban areas in the catchment can apparently explain – and differ-
entiate between – distinct patterns of micropollutant exposure. In
contrast, the proportion of forested and grassland areas primarily
showed a negative relationship to micropollutants. Despite strong
negative correlation between the proportion of forests and cropland
(pearson r = − 0.82), this indicates that both forms of extensive land use
relate to lower pollution (Dachbrodt-Saaydeh et al., 2021; Goss et al.,
2020; Riedo et al., 2022).
4.3. Individual pesticide concentrations relate to crop-specic pesticide
application
Our ndings conrm that individual pesticide concentrations can be
linked to individual crop types in the catchment of rivers (Andert et al.,
2015; Dachbrodt-Saaydeh et al., 2021; Schürings et al., 2024b). Pesti-
cide concentrations, particularly of fungicides and insecticides, were
strongly related to permanent crops and vegetables, in particular to
onion elds. These crop types are associated with intensive pesticide
application, in particular with insecticides (Dachbrodt-Saaydeh et al.,
2021). Further studies reported a deterioration of riverine biota in
agricultural catchments with a high areal coverage of permanent crops,
vegetables, vineyards or orchards (Bereswill et al., 2012; Schulz, 2001;
Schürings et al., 2024b; Xing et al., 2012). Cereals and maize showed
weaker relationships to pesticide concentrations (except for the herbi-
cides ufenacet and nicosulfuron), which suggests a less intensive
pesticide application connected to these crop types, except for herbi-
cides (Andert et al., 2015; Roßberg, 2016). Although the uncertainties in
the detection of pesticides in our data (see above) prevent us from
drawing nal conclusions as to the relationship between pesticides, in-
secticides in particular, and agricultural land uses (Weisner et al., 2022),
our ndings support the clear demand to distinguish between crop
types. The use of rather general categories like ‘cropland’ in our study
showed that relationships between individual herbicides and in-
secticides, and individual crop types would have been largely
overlooked.
4.4. Implications for micropollutant risk assessment and management
This study shows that both the proportion of urban and agricultural
areas in the catchment of rivers are notably related to the micropollutant
exposure in the rivers. Agricultural effects on micropollutant concen-
trations and joint mixture risks are not uniform and strongly vary be-
tween individual crop types. The mere differentiation between cropland
and grassland does not adequately represent agricultural stress. Notably,
the individual pesticides that were found to be strongly associated with
individual crop types largely reected their approved area of application
in Germany (BVL, 2023). Thus, in the absence of site-specic data on
pesticide concentrations, proportion of individual crop types cultivated
in the catchment (or at ner scales) may provide a good proxy to inform
the assessment of potential toxicity risks (Schürings et al., 2024a). The
same areal data could also support the identication of specic pollution
sources and the assessment of (mixture) risks of micropollutants in the
environment. In order to improve the assessment of (mixture) risks of
micropollutants, chemical monitoring programs need to further imple-
ment high frequent and event-based monitoring or composite sampling
(Bundschuh et al., 2014; Carvalho et al., 2019).
Industrial chemicals and pharmaceuticals were mainly related to
proportion of urban areas in the catchment, thus indicating a high
relevance of urban point sources, especially wastewater treatment plants
(Beek et al., 2016). Therefore, advanced wastewater treatment using
ozonation or activated carbon (or a mixture of both) require imple-
mentation to reduce the concentrations of micropollutants and hence
the ecotoxicological risks originating from them (Bundschuh et al.,
2011; Finckh et al., 2022; Kienle et al., 2022; Spilsbury et al., 2024;
Triebskorn et al., 2019). However, advanced wastewater treatment
cannot remove all micropollutants and neither can it remove the sec-
ondary (transformation) products that result, for example, from the
ozonation of primary pollutants (Bundschuh et al., 2011).
Intensive agriculture constitutes another major source of micro-
pollutants that imposes strong negative effects on riverine biota (Hughes
and Vadas, 2021; Schürings et al., 2022). In contrast to waste water
treatment plants, the diffuse pollution (and related ecological risks)
from agricultural areas cannot be reduced by selective local measures
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
10
(Rothe et al., 2021). Instead, agricultural approaches minimizing or
eliminating pesticide application, such as integrated pest management,
organic farming, agroecology or precision agriculture (Barzman et al.,
2015; Gebbers and Adamchuk, 2010; Gonz´
alez-Chang et al., 2020;
Reganold and Wachter, 2016) are required. Additionally, constructed
wetlands, vegetated buffer strips and riparian vegetation have been
shown to reduce pesticide exposure in surface waters (Lerch et al., 2017;
Stehle et al., 2011; Turunen et al., 2019; Vormeier et al., 2023a).
However, these approaches rely on substantial changes in agricultural
management and successful implementation of ambitious regulations
(Pe'er et al., 2022).
CRediT authorship contribution statement
Nele Markert: Writing – review & editing, Writing – original draft,
Visualization, Methodology, Investigation, Formal analysis, Data cura-
tion, Conceptualization. Christian Schürings: Writing – review &
editing, Writing – original draft, Visualization, Methodology, Investi-
gation, Formal analysis, Data curation, Conceptualization. Christian K.
Feld: Writing – review & editing, Supervision, Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data availability
The authors do not have permission to share data on micropollutant
concentrations, but data are available on request from the correspond-
ing agency North Rhine-Westphalia Ofce of Nature, Environment and
Consumer Protection. Land use data were derived from Grifths et al.
(2019) and Blickensd¨
orfer et al. (2022).
Acknowledgements
The individual projects of this work are funded by the Ministry of the
Environment, Nature and Transport of the State of North Rhine-
Westphalia (MUNV) to Nele Markert and by a scholarship funding
from the German Federal Environmental Foundation (DBU) to Christian
Schürings. Christian K. Feld was supported by the Collaborative
Research Centre 1439 RESIST (Multilevel Response to Stressor Increase
and Decrease in Stream Ecosystems; www.sfb-resist.de) funded by the
Deutsche Forschungsgemeinschaft (DFG, German Research Foundation;
CRC 1439/1, project number: 426547801). We thank Lukas Blick-
ensd¨
orfer et al. for the detailed data on land uses and crop types.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.scitotenv.2024.170583.
References
Andert, S., Bürger, J., Gerowitt, B., 2015. On-farm pesticide use in four northern German
regions as inuenced by farm and production conditions. Crop Prot. 75, 1–10.
https://doi.org/10.1016/j.cropro.2015.05.002.
Backhaus, T., Faust, M., 2012. Predictive environmental risk assessment of chemical
mixtures: a conceptual framework. Environ. Sci. Technol. 46 (5), 2564–2573.
https://doi.org/10.1021/es2034125.
Barzman, M., B`
arberi, P., Birch, A.N.E., Boonekamp, P., Dachbrodt-Saaydeh, S., Graf, B.,
et al., 2015. Eight principles of integrated pest management. Agron. Sustain. Dev. 35
(4), 1199–1215. https://doi.org/10.1007/s13593-015-0327-9.
Beckers, L.-M., Busch, W., Krauss, M., Schulze, T., Brack, W., 2018. Characterization and
risk assessment of seasonal and weather dynamics in organic pollutant mixtures from
discharge of a separate sewer system. Water Res. 135, 122–133. https://www.scienc
edirect.com/science/article/pii/S0043135418301015.
Beek, T. aus der, Weber, F.-A., Bergmann, A., Hickmann, S., Ebert, I., Hein, A., et al.,
2016. Pharmaceuticals in the environment—global occurrences and perspectives.
Environ. Toxicol. Chem. 35 (4), 823–835. https://doi.org/10.1002/etc.3339.
Bereswill, R., Golla, B., Streloke, M., Schulz, R., 2012. Entry and toxicity of organic
pesticides and copper in vineyard streams: Erosion rills jeopardise the efciency of
riparian buffer strips. Agric. Ecosyst. Environ. 146 (1), 81–92. https://doi.org/
10.1016/j.agee.2011.10.010.
Blickensd¨
orfer, L., Schwieder, M., Pugmacher, D., Nendel, C., Erasmi, S., Hostert, P.,
2022. Mapping of crop types and crop sequences with combined time series of
Sentinel-1, Sentinel-2 and Landsat 8 data for Germany. Remote Sens. Environ. 269,
112831 https://doi.org/10.1016/j.rse.2021.112831.
Bradley, P.M., Journey, C.A., Button, D.T., Carlisle, D.M., Huffman, B.J., Qi, S.L., et al.,
2020. Multi-region assessment of pharmaceutical exposures and predicted effects in
USA wadeable urban-gradient streams. PloS One 15 (1), e0228214. https://doi.org/
10.1371/journal.pone.0228214.
Britz, W., Witzke, P., 2014. CAPRI model documentation 2014. https://www.capri-mode
l.org/lib/exe/fetch.php?media=docs:capri_documentation.pdf. (Accessed 2 March
2023).
Bundschuh, M., Zubrod, J.P., Seitz, F., Stang, C., Schulz, R., 2011. Ecotoxicological
evaluation of three tertiary wastewater treatment techniques via meta-analysis and
feeding bioassays using Gammarus fossarum. J. Hazard. Mater. 192 (2), 772–778.
https://doi.org/10.1016/j.jhazmat.2011.05.079.
Bundschuh, M., Goedkoop, W., Kreuger, J., 2014. Evaluation of pesticide monitoring
strategies in agricultural streams based on the toxic-unit concept — experiences from
long-term measurements. Sci. Total Environ. 484, 84–91. https://doi.org/10.1016/j.
scitotenv.2014.03.015.
BVL, 2023a. Authorised plant protection products with crops and pests (July 2023).
https://www.bvl.bund.de/SharedDocs/Downloads/04_Panzenschutzmittel/ppp_
crops_pests.xlsx;jsessionid=51820C91EBCC37F895B8D0BE02DBC11D.internet012?
__blob=publicationFile&v=23. (Accessed 18 September 2023).
BVL, 2023b. Beendete Zulassungen von Panzenschutzmitteln (Stand: Januar 2023).
Carvalho, L., Mackay, E., B., Cardoso, A.C., Baattrup-Pedersen, A., Birk, S.,
Blackstock, K., et al., 2019. Protecting and restoring Europe’s waters: an analysis of
the future development needs of the Water Framework Directive. Sci. Total Environ.
658 https://doi.org/10.1016/j.scitotenv.2018.12.255.
Dachbrodt-Saaydeh, S., Sellmann, J., Strassemeyer, J., Schwarz, J., Klocke, B.,
Krengel, S., 2021. Netz Vergleichsbetriebe Panzenschutz Jahresbericht 2017
Analyse der Ergebnisse der Jahre 2007-2017: Julius Kühn-Institut. Report No. 210.
Diamond, M.L., de Wit, C.A., Molander, S., Scheringer, M., Backhaus, T., Lohmann, R.,
et al., 2015. Exploring the planetary boundary for chemical pollution. Environ. Int.
78, 8–15. https://www.sciencedirect.com/science/article/pii/S01604120150002
88.
Ebele, A.J., Abou-Elwafa Abdallah, M., Harrad, S., 2017. Pharmaceuticals and personal
care products (PPCPs) in the freshwater aquatic environment. Emerging
Contaminants 3 (1), 1–16. https://www.sciencedirect.com/science/article/pii/
S2405665016300488.
ECHA, 2023. Information on biocides. https://echa.europa.eu/de/information-on-ch
emicals/biocidal-active-substances. (Accessed 17 September 2023).
EMA, 2021. Imoxat (imidacloprid / moxidectin). EMA/575985/2021.
European Commission, 2014. Commission Regulation (EU) No 358/2014 of 9 April 2014
amending Annexes II and V to Regulation (EC) No 1223/2009 of the European
Parliament and of the Council on cosmetic products Text with EEA relevance.
Brussels. Off. J. Eur. Union.
European Commission, 2017. Technical Guidance for Deriving Environmental Quality
Standards. Directorate-General for Health and Food Safety, European Commission,
Brussels.
European Commission, 2019. Communication from the Commission to the European
Parliament, the European Council, the Council, the European Economic and Social
Committee and the Committee of the Regions. The European Green Deal. COM
(2019) 640 nal. Brussels.
European Commission, 2020. Communication from the Commission to the European
Parliament, the Council, the European Economic and Social Committee and the
Committee of the Regions. Chemicals Strategy for Sustainability. Towards a toxic-
free environment. COM(2020) 667 nal. Brussels.
Fekadu, S., Alemayehu, E., Dewil, R., van der Bruggen, B., 2019. Pharmaceuticals in
freshwater aquatic environments: a comparison of the African and European
challenge. Sci. Total Environ. 654, 324–337. https://doi.org/10.1016/j.
scitotenv.2018.11.072.
Finckh, S., Beckers, L.-M., Busch, W., Carmona, E., Dulio, V., Kramer, L., et al., 2022.
A risk based assessment approach for chemical mixtures from wastewater treatment
plant efuents. Environ. Int. 107234. https://www.sciencedirect.com/science/artic
le/pii/S016041202200160X.
Fuller, R., Landrigan, P.J., Balakrishnan, K., Bathan, G., Bose-O'Reilly, S., Brauer, M.,
et al., 2022. Pollution and health: a progress update. The Lancet Planetary Health 6
(6), e535–e547. https://www.sciencedirect.com/science/article/pii/S2542519
622000900.
Gebbers, R., Adamchuk, V.I., 2010. Precision agriculture and food security. Science (New
York, N.Y.) 327 (5967), 828–831. https://doi.org/10.1126/science.1183899.
Ginebreda, A., Kuzmanovi´
c, M., Guasch, H., de Alda, M.L., L´
opez-Doval, J.C., Mu˜
noz, I.,
et al., 2014. Assessment of multi-chemical pollution in aquatic ecosystems using
toxic units: compound prioritization, mixture characterization and relationships
with biological descriptors. Sci. Total Environ. 468-469, 715–723. https://doi.org/
10.1016/j.scitotenv.2013.08.086.
Gonz´
alez-Chang, M., Wratten, S.D., Shields, M.W., Costanza, R., Dainese, M., Gurr, G.M.,
et al., 2020. Understanding the pathways from biodiversity to agro-ecological
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
11
outcomes: A new, interactive approach. Agric. Ecosyst. Environ. 301, 107053
https://doi.org/10.1016/j.agee.2020.107053.
Goss, C.W., Sullivan, S.M.P., Goebel, P.C., 2020. Effects of land-cover transitions on
emerging aquatic insects and environmental characteristics of headwater streams in
an agricultural catchment. River Res. Appl. 36 (7), 1097–1108. https://doi.org/
10.1002/rra.3639.
Grandjean, P., Bellanger, M., 2017. Calculation of the disease burden associated with
environmental chemical exposures: application of toxicological information in
health economic estimation. Environmental health a global access science source 16
(1), 123. https://doi.org/10.1186/s12940-017-0340-3.
Grifths, P., Nendel, C., Hostert, P., 2019. Intra-annual reectance composites from
Sentinel-2 and Landsat for national-scale crop and land cover mapping. Remote Sens.
Environ. 220, 135–151. https://www.sciencedirect.com/science/article/pii/
S0034425718304851.
Groh, K., vom Berg, C., Schirmer, K., Tlili, A., 2022. Anthropogenic chemicals as
underestimated drivers of biodiversity loss: scientic and societal implications.
Environ. Sci. Technol. 56 (2), 707–710. https://doi.org/10.1021/acs.est.1c08399.
Gustavsson, M., Kreuger, J., Bundschuh, M., Backhaus, T., 2017. Pesticide mixtures in
the Swedish streams: environmental risks, contributions of individual compounds
and consequences of single-substance oriented risk mitigation. Sci. Total Environ.
598, 973–983. https://doi.org/10.1016/j.scitotenv.2017.04.122.
Halbach, K., M¨
oder, M., Schrader, S., Liebmann, L., Sch¨
afer, R.B., Schneeweiss, A., et al.,
2021. Small streams-large concentrations? Pesticide monitoring in small agricultural
streams in Germany during dry weather and rainfall. Water Res. 203, 117535.
https://www.sciencedirect.com/science/article/pii/S0043135421007314.
Harrison, S., McAree, C., Mulville, W., Sullivan, T., 2019. The problem of agricultural
‘diffuse’ pollution: getting to the point. Sci. Total Environ. 677, 700–717. https://
www.sciencedirect.com/science/article/pii/S0048969719316985.
Hernando, M.D., Mezcua, M., Fern´
andez-Alba, A.R., Barcel´
o, D., 2006. Environmental
risk assessment of pharmaceutical residues in wastewater efuents, surface waters
and sediments. Talanta 69 (2), 334–342. https://www.sciencedirect.com/science/
article/pii/S0039914005006594.
Hughes, R.M., Vadas Jr., R.L., 2021. Agricultural effects on streams and rivers: a western
USA focus. Water 13 (14), 1901. https://doi.org/10.3390/w13141901.
Kienle, C., Werner, I., Fischer, S., Lüthi, C., Schifferli, A., Besselink, H., et al., 2022.
Evaluation of a full-scale wastewater treatment plant with ozonation and different
post-treatments using a broad range of in vitro and in vivo bioassays. Water Res.
212, 118084 https://doi.org/10.1016/j.watres.2022.118084.
Koumaki, E., Mamais, D., Noutsopoulos, C., 2018. Assessment of the environmental fate
of endocrine disrupting chemicals in rivers. Sci. Total Environ. 628-629, 947–958.
https://doi.org/10.1016/j.scitotenv.2018.02.110.
LAWA, 2019. Rahmenkonzeption Monitoring Teil B: Bewertungsgrundlagen und
Methodenbeschreibungen: Arbeitspapier IV.1: Untersuchungsverfahren für
chemische und physikalisch-chemische Qualit¨
atskomponenten.
Lemm, J.U., Venohr, M., Globevnik, L., Stefanidis, K., Panagopoulos, Y., van Gils, J.,
et al., 2021. Multiple stressors determine river ecological status at the European
scale: towards an integrated understanding of river status deterioration. Glob.
Change Biol. 27, 1962–1975. https://doi.org/10.1111/gcb.15504.
Lerch, R.N., Lin, C.H., Goyne, K.W., Kremer, R.J., Anderson, S.H., 2017. Vegetative
buffer strips for reducing herbicide transport in runoff: effects of buffer width,
vegetation, and season. JAWRA J. Am. Water Resour. Assoc. 53 (3), 667–683.
https://doi.org/10.1111/1752-1688.12526.
Liess, M., Liebmann, L., Vormeier, P., Weisner, O., Altenburger, R., Borchardt, D., et al.,
2021. Pesticides are the dominant stressors for vulnerable insects in lowland
streams. Water Res. 201, 117262 https://doi.org/10.1016/j.watres.2021.117262.
Loos, R., Carvalho, R., Ant´
onio, D.C., Comero, S., Locoro, G., Tavazzi, S., et al., 2013. EU-
wide monitoring survey on emerging polar organic contaminants in wastewater
treatment plant efuents. Water Res. 47 (17), 6475–6487. https://www.sciencedirec
t.com/science/article/pii/S0043135413006647.
Malaj, E., von der Ohe, P.C., Grote, M., Kühne, R., Mondy, C.P., Usseglio-Polatera, P.,
et al., 2014. Organic chemicals jeopardize the health of freshwater ecosystems on the
continental scale. Proc. Natl. Acad. Sci. U. S. A. 111 (26), 9549–9554. https://doi.
org/10.1073/pnas.1321082111.
Mandaric, L., Mor, J.-R., Sabater, S., Petrovic, M., 2018. Impact of urban chemical
pollution on water quality in small, rural and efuent-dominated Mediterranean
streams and rivers. Sci. Total Environ. 613-614, 763–772. https://www.sciencedirec
t.com/science/article/pii/S0048969717324701.
Markert, N., Rhiem, S., Trimborn, M., Guhl, B., 2020. Mixture toxicity in the Erft River:
assessment of ecological risks and toxicity drivers. Environ. Sci. Eur. 32 (1) https://
doi.org/10.1186/s12302-020-00326-5.
McEneff, G., Barron, L., Kelleher, B., Paull, B., Quinn, B., 2014. A year-long study of the
spatial occurrence and relative distribution of pharmaceutical residues in sewage
efuent, receiving marine waters and marine bivalves. Sci. Total Environ. 476-477,
317–326. https://www.sciencedirect.com/science/article/pii/S0048969
714000023.
Moschet, C., Wittmer, I., Simovic, J., Junghans, M., Piazzoli, A., Singer, H., et al., 2014.
How a complete pesticide screening changes the assessment of surface water quality.
Environ. Sci. Technol. 48 (10), 5423–5432. https://doi.org/10.1021/es500371t.
Munz, N.A., Burdon, F.J., De Zwart, D., Junghans, M., Melo, L., Reyes, M., et al., 2017.
Pesticides drive risk of micropollutants in wastewater-impacted streams during low
ow conditions. Water Res. 110, 366–377. http://www.sciencedirect.com/science/
article/pii/S0043135416308466.
Münze, R., Hannemann, C., Orlinskiy, P., Gunold, R., Paschke, A., Foit, K., et al., 2017.
Pesticides from wastewater treatment plant efuents affect invertebrate
communities. Sci. Total Environ. 599-600, 387–399. http://www.sciencedirect.com/
science/article/pii/S0048969717305132.
OGewV, 2016. Ober¨
achengew¨
asserverordnung vom 20. Juni 2016, BGBl. I S. 1373.
von der Ohe, P.C., Dulio, V., Slobodnik, J., de Deckere, E., Kühne, R., Ebert, R.-U., et al.,
2011. A new risk assessment approach for the prioritization of 500 classical and
emerging organic microcontaminants as potential river basin specic pollutants
under the European water framework directive. Sci. Total Environ. 409 (11),
2064–2077. https://www.sciencedirect.com/science/article/pii/S0048969
711001136.
Pe’er, G., Finn, J.A., Díaz, M., Birkenstock, M., Lakner, S., R¨
oder, N., et al., 2022. How
can the European common agricultural policy help halt biodiversity loss?
Recommendations by over 300 experts. Conserv. Lett. 15 (6) https://doi.org/
10.1111/conl.12901.
Persson, L., Carney Almroth, B.M., Collins, C.D., Cornell, S., de Wit, C.A., Diamond, M.L.,
et al., 2022. Outside the safe operating space of the planetary boundary for novel
entities. Environ. Sci. Technol. 56 (3), 1510–1521. https://doi.org/10.1021/acs.
est.1c04158.
Posthuma, L., Zijp, M.C., De Zwart, D., van de Meent, D., Globevnik, L., Koprivsek, M.,
et al., 2020. Chemical pollution imposes limitations to the ecological status of
European surface waters. Sci. Rep. 10 (1), 14825. https://doi.org/10.1038/s41598-
020-71537-2.
R Core Team, 2020. R: A Language and Environment for Statistical Computing. R
Foundation for Statistical Computing.
Reganold, J.P., Wachter, J.M., 2016. Organic agriculture in the twenty-rst century. Nat.
Plants 2 (2), 1–8. https://doi.org/10.1038/nplants.2015.221.
Riedo, J., Herzog, C., Banerjee, S., Fenner, K., Walder, F., van der Heijden, M.G.A., et al.,
2022. Concerted evaluation of pesticides in soils of extensive grassland sites and
organic and conventional vegetable elds facilitates the identication of major input
processes. Environ. Sci. Technol. 56 (19), 13686–13695. https://doi.org/10.1021/
acs.est.2c02413.
Rigby, R.A., Stasinopoulos, D.M., 2005. Generalized additive models for location, scale
and shape. J. R. Stat. Soc. Ser. C. Appl. Stat. 54 (3), 507–554. https://doi.org/
10.1111/j.1467-9876.2005.00510.x.
Rockstr¨
om, J., Steffen, W., Noone, K., Persson, Å., Chapin, F.S., Lambin, E.F., et al., 2009.
A safe operating space for humanity. Nature 461 (7263), 472–475. https://doi.org/
10.1038/461472a.
Roßberg, D., 2016. Erhebungen zur Anwendung von Panzenschutzmitteln im Ackerbau.
J. Kult. 68 (2), 25–37. https://doi.org/10.5073/JfK.2016.02.01.
Rothe, L.E., Botha, T.L., Feld, C.K., Weyand, M., Zimmermann, S., Smit, N.J., et al., 2021.
Effects of conventionally-treated and ozonated wastewater on mortality, physiology,
body length, and behavior of embryonic and larval zebrash (Danio rerio). Environ.
Pollut. (Barking, Essex 1987) 286, 117241. https://www.sciencedirect.com/science/
article/pii/S026974912100823X.
Rumschlag, S.L., Bessler, S.M., Rohr, J.R., 2019. Evaluating improvements to exposure
estimates from fate and transport models by incorporating environmental sampling
effort and contaminant use. Water Res. 156, 372–382. https://www.sciencedirect.co
m/science/article/pii/S004313541930257X.
Sch¨
afer, R.B., van den Brink, P.J., Liess, M., 2011. Impacts of pesticides on freshwater
ecosystems. Ecological impacts of toxic chemicals 2011, 111–137.
Schulz, R., 2001. Rainfall-induced sediment and pesticide input from orchards into the
Lourens River, Western cape, South Africa: importance of a single event. Water Res.
35 (8), 1869–1876. https://doi.org/10.1016/S0043-1354(00)00458-9.
Schürings, C., Feld, C.K., Kail, J., Hering, D., 2022. Effects of agricultural land use on
river biota: a meta-analysis. Environ. Sci. Eur. 34 (1) https://doi.org/10.1186/
s12302-022-00706-z.
Schürings, C., Hering, D., Kaijser, W., Kail, J., 2024a. Assessment of cultivation intensity
can improve the correlative strength between agriculture and the ecological status in
rivers across Germany. Agric. Ecosyst. Environ. 361, 108818.
Schürings, C., Kail, J., Kaijser, W., Hering, D., 2024b. Effects of agriculture on river biota
differ between crop types and organism groups. Sci. Total Environ. 912, 168825
https://doi.org/10.2139/ssrn.4499828.
Schwarzenbach, R.P., Escher, B.I., Fenner, K., Hofstetter, T.B., Johnson, C.A., von
Gunten, U., et al., 2006. The challenge of micropollutants in aquatic systems. Science
(New York, N.Y.) 313 (5790), 1072–1077. https://doi.org/10.1126/
science.1127291.
Sigmund, G., Ågerstrand, M., Brodin, T., Diamond, M.L., Erdelen, W.R., Evers, D.C.,
et al., 2022. Broaden chemicals scope in biodiversity targets. Science (New York, N.
Y.) 376 (6599), 1280. https://doi.org/10.1126/science.add3070.
Spilsbury, F., Kisielius, V., Bester, K., Backhaus, T., 2024. Ecotoxicological mixture risk
assessment of 35 pharmaceuticals in wastewater efuents following post-treatment
with ozone and/or granulated activated carbon. Sci. Total Environ. 906, 167440.
https://www.sciencedirect.com/science/article/pii/S0048969723060679.
Stehle, S., Elsaesser, D., Gregoire, C., Imfeld, G., Niehaus, E., Passeport, E., et al., 2011.
Pesticide risk mitigation by vegetated treatment systems: a meta-analysis. J. Environ.
Qual. 40 (4), 1068–1080. https://doi.org/10.2134/jeq2010.0510.
Stekhoven, D.J., Bühlmann, P., 2012. MissForest—non-parametric missing value
imputation for mixed-type data. Bioinformatics (Oxford, England) 28 (1), 112–118.
https://doi.org/10.1093/bioinformatics/btr597.
Sz¨
ocs, E., Brinke, M., Karaoglan, B., Sch¨
afer, R.B., 2017. Large scale risks from
agricultural pesticides in small streams. Environ. Sci. Technol. 51 (13), 7378–7385.
https://doi.org/10.1021/acs.est.7b00933.
Tang, F., Ishwaran, H., 2017. Random forest missing data algorithms. Statistical analysis
and data mining 10 (6), 363–377. https://doi.org/10.1002/sam.11348.
Tauchnitz, N., Kurzius, F., Rupp, H., Schmidt, G., Hauser, B., Schr¨
odter, M., et al., 2020.
Assessment of pesticide inputs into surface waters by agricultural and urban sources
- A case study in the Querne/Weida catchment, Central Germany. Environ. Pollut.
(Oxford, U. K.) 267, 115186. https://www.sciencedirect.com/science/article/pii/
S0269749120335028.
N. Markert et al.
Science of the Total Environment 917 (2024) 170583
12
Triebskorn, R., Blaha, L., Gallert, C., Giebner, S., Hetzenauer, H., K¨
ohler, H.-R., et al.,
2019. Freshwater ecosystems prot from activated carbon-based wastewater
treatment across various levels of biological organisation in a short timeframe.
Environ. Sci. Eur. 31 (1), 1–16. https://doi.org/10.1186/s12302-019-0267-0.
Turunen, J., Markkula, J., Rajakallio, M., Aroviita, J., 2019. Riparian forests mitigate
harmful ecological effects of agricultural diffuse pollution in medium-sized streams.
Sci. Total Environ. 649, 495–503. https://www.sciencedirect.com/science/article/
pii/S0048969718334028.
UNEP, 2017. Towards a Pollution-Free Planet Background Report. United Nations
Environment Programme, Nairobi, Kenya.
UNEP, 2021. Making Peace with Nature: A Scientic Blueprint to Tackle the Climate,
Biodiversity and Pollution Emergencies. Nairobi, Kenya.
Viechtbauer, W., 2010. Conducting meta-analyses in R with the metafor package. J. Stat.
Softw. 36 (3), 1–48. https://doi.org/10.1863/jss.v036.i03.
Vormeier, P., Liebmann, L., Weisner, O., Liess, M., 2023a. Width of vegetated buffer
strips to protect aquatic life from pesticide effects. Water Res. 231, 119627. https://
www.sciencedirect.com/science/article/pii/S0043135423000623.
Vormeier, P., Schreiner, V.C., Liebmann, L., Link, M., Sch¨
afer, R.B., Schneeweiss, A.,
et al., 2023b. Temporal scales of pesticide exposure and risks in German small
streams. Sci. Total Environ. 871, 162105. https://www.sciencedirect.com/science/
article/pii/S0048969723007210.
Wang, Z., Walker, G.W., Muir, D.C.G., Nagatani-Yoshida, K., 2020. Toward a global
understanding of chemical pollution: A rst comprehensive analysis of national and
regional chemical inventories. Environ. Sci. Technol. 54 (5), 2575–2584. https://
doi.org/10.1021/acs.est.9b06379.
Weisner, O., Frische, T., Liebmann, L., Reemtsma, T., Roß-Nickoll, M., Sch¨
afer, R.B.,
et al., 2021. Risk from pesticide mixtures – the gap between risk assessment and
reality. Sci. Total Environ. 796, 149017. https://www.sciencedirect.com/science/
article/pii/S0048969721040894.
Weisner, O., Arle, J., Liebmann, L., Link, M., Sch¨
afer, R.B., Schneeweiss, A., et al., 2022.
Three reasons why the water framework directive (WFD) fails to identify pesticide
risks. Water Res. 208, 117848 https://doi.org/10.1016/j.watres.2021.117848.
Wickham, H., 2016. ggplot2: Elegant Graphics for Data Analysis. Springer International
Publishing; Imprint: Springer, Cham.
Wiering, M., Boezeman, D., Crabb´
e, A., 2020. The water framework directive and
agricultural diffuse pollution: ghting a running battle? Water 12 (5), 1447. https://
doi.org/10.3390/w12051447.
Xing, Z., Chow, L., Cook, A., Benoy, G., Rees, H., Ernst, B., et al., 2012. Pesticide
application and detection in variable agricultural intensity watersheds and their
river systems in the maritime region of Canada. Arch. Environ. Contam. Toxicol. 63,
471–483. https://doi.org/10.1007/s00244-012-9789-9.
N. Markert et al.