Agriculture, Ecosystems and Environment 313 (2021) 107341

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment

journal homepage: www.elsevier.com/locate/agee

Evidence for soil pesticide contamination of an agroecological farm from a

neighboring chemical-based production system
Constanza Bernasconi a, b, c, Pablo M. Demetrio a, b, Lucas L. Alonso a, b, c,
Tomas M. Mac Loughlin a, b, c, Eduardo Cerdá c, Santiago J. Sarandón c, d, e,
Damian J. Marino a, b, c, *
a
Centro De Investigaciones Del Medio Ambiente - CIM-, Facultad De Ciencias Exactas, Universidad Nacional De La Plata, Boulevard 120 N◦ 1489, 1900, La Plata,
Buenos Aires, Argentina
b
Consejo Nacional De Investigaciones Científicas y Técnicas - CONICET, Centro Tecnológico La Plata, Buenos Aires, Argentina
c
Red Nacional De Municipios y Comunidades que Fomentan La Agroecología - RENAMA, Buenos Aires, Argentina
d
Laboratorio De Investigación y Reflexión En Agroecología - LIRA, Facultad De Ciencias Agrarias y Forestales-Universidad Nacional De La Plata, Avenida 60 y 119,
1900, La Plata, Buenos Aires, Argentina
e
Comisión De Investigaciones Científicas DeLa Provincia De Buenos Aires (CIC), Bs As., Argentina

A R T I C L E I N F O A B S T R A C T

Keywords: The global chemical-based agriculture (CBA) production system brought social and environmental consequences
Agroproductive system such as the contamination of soils, waters, bottom sediments and food, as well as negative effects on non-target
Agroecology species. As an alternative, a new paradigm emerged: agroecology-based agriculture (ABA), based on ecosystem
Glyphosate
services and the reduction of chemical inputs. More and more establishments have adopted this form of pro­
Agricultural soils
duction; however, they are located next to crops to which pesticides are applied. The objective of this work was
to study, through the spatio-temporal characterizations of pesticides in soils, how an ABA production system can
be affected by the CBA fields around it. Two sampling campaigns were conducted and soil samples were obtained
from "La Aurora", an agricultural establishment located in the Argentine pampa and recognized by the FAO for its
agroecological practices, and from neighboring fields with CBA productions. The samples were tested for 19
herbicides (including 3 metabolites) and 3 fungicides by UPLC-MS/MS, of which we detected glyphosate and its
metabolite AMPA, 2,4-D, atrazine, acetochlor, metsulfuron-methyl, desethyl-atrazine, epoxiconazole, and
tebuconazole. Three or more pesticides co-occurred in 93% and 32% of the CBA and ABA samples, respectively.
Glyphosate and AMPA, with the highest detection frequency, also accounted for 90% of the total pesticide load in
both systems. The maximum concentrations (μg kg− 1 dry weight) in the CBA/ABA fields, respectively, were
glyphosate (1268.92/98.93), AMPA (2919.17/114.01), followed by 2,4-D (38.52/31.12), and epoxiconazole
(13.35/18.41). No significant temporal differences were found in glyphosate concentration within each estab­
lishment, corroborating its pseudo-persistence in CBA establishments, and establishing it in ABA field. Moreover,
glyphosate was found in the ABA field more than 300 m from the limit with the CBA fields. Glyphosate and
AMPA concentrations are in the order of those reported to cause sublethal and lethal effects in soil organisms.
These results highlight the mobility of pesticides, as the ABA establishment is affected by its surroundings where
pesticides are used, even at sites far from the interface between them. Given their higher detection frequencies
and environmental concentrations in comparison to the other pesticides, glyphosate and AMPA are proposed as
environmental tracers of conventional agroproductive activities.

1. Introduction agricultural (bio)technology, like the development of genetically

modified organisms (GMOs), the use of pesticides, no-till (or zero tillage)
World food production is continually adopting the latest advances in and the introduction of computerized machinery. With this way of

* Corresponding author at: Centro de Investigaciones del Medio Ambiente - CIM-, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, Boulevard 120
N◦ 1489, 1900, La Plata, Buenos Aires, Argentina.
E-mail address: damianm@quimica.unlp.edu.ar (D.J. Marino).

https://doi.org/10.1016/j.agee.2021.107341
Received 28 July 2020; Received in revised form 16 January 2021; Accepted 19 January 2021
Available online 24 February 2021
0167-8809/© 2021 Elsevier B.V. All rights reserved.
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

production, the cultivated area worldwide of biotech crops increased such as prohibition of certain agrochemicals in order to meet the certi­
110 times in just 21 years of commercialization of these new techno­ fication requirements, the agroecological practices seek the stability and
logical developments (ISAAA, 2016). sustainability of the agrarian system by strengthening ecological pro­
Argentina is the third country, after the United States and Brazil, cesses or functions, thus resulting in a decrease or elimination of
with the largest area cultivated with GMOs in the world, reaching 23.6 chemical inputs (Gurr et al., 2016; Altieri, 2018). The agro-biodiversity
million ha planted in 2017 (ISAAA, 2017). This expanse represents 61% (genetic, specific, and structural), when just maintaining the key
of the area cultivated with extensive monocrops during the 2017/2018 necessary components, can provide functional ecological services such
agricultural cycle(July 1, 2017 - June 30, 2018) in Argentina (MAGyP, as pests and pathogens regulation, nutrient cycling (decomposition of
2019). Out of the total cultivated GMOs, the majority corresponds to organic matter and maintenance of soil fertility), control of erosion
soybeans (77%), followed by corn (22%) and cotton (1%) (ISAAA, (vegetation cover), pollination, among others (Altieri, 2018).
2017). As a result of the tendency towards monoculture practices and In Argentina, there are no official reports of the number of agroeco­
the increasing use of the previously mentioned biotechnological tools, logical-based agriculture (ABA) systems, since only organic certified crops
chemical control to combat weeds, pests and diseases has intensified, as are registered nationally, representing 0.1% of the total extensive pro­
well as the use of synthetic fertilizers to replenish nutrients (FAO, 1990; duction (SENASA, 2019; MAGyP, 2019). However, the growth of the
Sarandón and Flores, 2014). According to the latest official information ABA production (Sarandon and Marasas, 2017) is reflected on its addi­
reported in Argentina, the use of pesticides increased by 900% since the tion as a type of agricultural practice in the National Agricultural Census
approval of the first transgenic event in 1996, with 317 million kg or L of carried out in 2018 (INDEC, 2020 - Data is still in the analysis stage).
active ingredients being used in 2012, and showing an increasing trend The expansion of ABA systems is still in an initial stage, which means
in use (CASAFE, 2012). that these systems will be immersed in an environment of conventional
The conventional production system, also known as chemical-based systems where pesticides are used. Given their close proximity, and
agriculture (CBA), has brought socio-cultural and environmental prob­ taking into consideration the complexity of pesticide dynamics in the
lems as consequences (Leguizamón, 2014; Rauchecker, 2019). The use environment, it is relevant to assess the reach of these agrochemicals in
of pesticides, together with intensive mechanical practices, can result in these scenarios, as they can alter the ecological functions in agroeco­
the loss of natural habitats, and consequently alter the biodiversity logical plantations. Therefore, the objective of this work was to study,
associated with them (Benton et al., 2003; Sharma et al., 2018). The through the spatio-temporal characterizations of pesticides in soils, how
harmful effects of these compounds on beneficial species, for example, an ABA system can be adversely influenced by the pesticide applications
insects and arthropods that provide ecological services and functions in the CBA crop fields around it.
such as biological control of pests and pollination, or oligochaetes and
microorganisms responsible for nutrient recycling and maintaining soil 2. Materials and methods
structural properties, has been reported (Benamú et al., 2010; Pelosi
et al., 2014; Evans et al., 2018; Sharma et al., 2018; Wołejko et al., 2.1. Study area
2020).
Glyphosate (GLP) is the most extensively used pesticide worldwide The study area is located in Benito Juárez, in the south of the
and in Argentina, accounts for 62% of the pesticides sold in the country Province of Buenos Aires, Argentina. The region is a landscape of soft to
(CASAFE, 2012). GLP is a broad-spectrum, systemic and post-emergent moderately undulating plains, with slight depressions of the "Pampa
herbicide, applied to GM crops resistant to it for weed control in Deprimida Occidental Sector" sub-region, in a well-drained hillside po­
no-tillage systems (Okada et al., 2016; Primost et al., 2017). Due to its sition, developed in loessic sediments on a calcareous crust of regional
physicochemical properties, it binds strongly to soil components, with extension, non-saline, non-alkaline, on slopes of 1% (INTA, 2002). The
reported half-life of up to 197 days, conditioned by soil type, climate climate is mesothermal humid (temperate), with average annual pre­
conditions and microbial activity. The main degradation product is the cipitations between 800 and 900 mm. The annual average temperature
aminomethylphosphonic acid (AMPA), which in turn degrades more is 13.8 ◦ C, with an average maximum and minimum temperature of 21.4
slowly than glyphosate in soils, with a half-life range from 60 to 240 ◦
C and 6.9 ◦ C, respectively (INTA, 2014). The soil of the region is a
days (Giesy et al., 2000). Moreover, Primost et al. (2017) has classified typical Argiudol. The main use of the land is livestock production and
both molecules as pseudo-persistent in Argentinian soils. The increase in extensive agriculture, with the less suitable soils used for raising cattle,
pesticide use is due to, not only the larger expanse of cultivated area, but and the intermediate to good ones dedicated to agriculture (Cerda et al.,
also to the development and spread of pesticide resistance by some pests 2014). The selected study system is an extensive ABA and livestock
and pathogens, which consequently leads to the use of higher doses production farm adjacent to conventional agricultural fields with a CBA
and/or the release of new active ingredients and formulations (Sarandón production system.
and Flores, 2014).
Once released into the environment, the fate of every compound is 2.1.1. ABA: Agroecological-based agriculture
variable, as pesticide dynamics fundamentally depend on their physi­ The agroecological farm “La Aurora” has a total surface of 650 ha, of
cochemical properties and weather conditions, as well as soil properties which 15 ha are not agriculturally exploited,186 ha correspond to low
(Azcarate et al., 2015; Okada et al., 2016). Concentrations of some of grounds, 152 ha to hills, and 297 ha to agricultural soils, in turn divided
these compounds have been reported in surface waters, soils and sedi­ into 14 plots (Fig. 1). In 1997, an agroecological transition process
ments (Ronco et al., 2016; Mac Loughlin et al., 2017; Etchegoyen et al., began, which involved strengthening rotation, maximizing carbon fix­
2017; Van Bruggen et al., 2018; Silva et al., 2019), in food (Mac ation, increasing the surface area of crop associations with legumes
Loughlin et al., 2018), and in rainwater and air particulate matter (biological nitrogen fixation), enriching soil organic matter, improving
(Chang et al., 2011; Alonso et al., 2018). Concurrently, there is evidence the animals’ diet and the use of their manure to balance nutrients in the
to the adverse effects they cause on non-target species, and, ultimately, soil, among others. This ABA-management succeeded in reducing the
pose a risk for biodiversity (Van Bruggen et al., 2018; Iturburu et al., use of agrochemicals, with the last herbicide application in 2011
2019; Trudeau et al., 2020). (Table 1) (Iermanó, 2015; Cerdá et al., 2014). In 2016, “La Aurora” was
Agroecology emerged in the 1970s as a new approach and paradigm recognized by the FAO as one of 52 worldwide agroecological farm, for
in Agricultural Sciences in Latin America, and has been gaining strength having demonstrated that productive agriculture without agrochemicals
on account of the environmental, social and productive problems caused is not only possible, but also profitable (FAO, 2016).
by the conventional system (Altieri, 2017; Sarandon and Marasas, The plots that have agricultural aptitude are planted with annual
2017). Different from organic agriculture, that adopts specific measures crops, winter greens and pastures. The main winter crops are wheat

2
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

Fig. 1. Geographic location of the “La

Aurora” (ABA field), in a darker shade,
surrounded by CBA fields, in a lighter
shade, in Benito Juarez, Province of
Buenos Aires, Argentina. The 14 parcels
in which the ABA field is subdivided are
shown. Sampling sites are shown with
yellow dots. Graphical representation of
the sampling methodology for border
sampling sites, shown with S7, and
center sites (>300 m), shown with S3.
The transect is only shown in for the 40
m distance in the ABA field, but the
same procedure was applied for the
other distances (For interpretation of
the references to colour in this figure
legend, the reader is referred to the web
version of this article.).

Table 1
Land use in the ABA (Agroecological Based Agriculture) field for each campaign, and registry of pesticide applications until 2011, when the last application occurred.
Land Use

Site Last application (year)* C1 (July-winter) C2 (November-Spring)

S1-ABA n.p.a. Pasture (cattle) Pasture (cattle)

S2-ABA 2,4D - Dicamba (2011) Sorghum Sorghum (cattle)
S3-ABA 2,4D- Dicamba- Metuslfuron Methyl (2011) Natural Field Natural Field
S4-ABA n.p.a. Pasture Pasture
S5-ABA Metsulfuron methyl (2010) Tillage wheat + clover
S6-ABA n.p.a. Pasture + oat + Vicia sp. (cattle) Pasture + oat + Vicia sp. (cattle)
S7-ABA 2,4 D Amina-Dicamba (2011) Oat + Vicia sp. Oat + Vicia sp.
S8-ABA 2,4D- Dicamba- Metuslfuron Methyl (2011) Clover (cattle) Clover
S9-ABA Dicamba (2010) Corn Corn harvested
S10-ABA n.p.a. Tillage wheat + clover

n.p.a. : no pesticide applications since 2006.

*
Personal records on pesticide applications and management practices provided by the farmer and owner of the “La Aurora”, Juan Khier.

(Triticum aestivum L.), barley (Hordeum vulgare L.), and oats (Avena sativa exclusively to livestock farming, but no plot is exclusively arable.
L.), all in consortium with legumes such as red clover (Trifolium pratense
L.) or Vicia sp. During the summer, the crops are forage sorghum (Sor­ 2.1.2. CBA: chemical-based agriculture
ghum bicolor (L.) Moench), graniferous sorghum (Sorghum sudanense The fields neighboring “La Aurora” currently rotate their crops be­
(Piper) Stapf), and corn (Zea mays). In general, saved crop seeds are used tween soy, sunflower or corn during the summer, and wheat, rapeseed,
for new plantings. During each crop rotation, cattle are introduced into or barley during the winter. Unlike the ABA field, these CBA establish­
these plots in order to feed on the stubble and fertilize the soil. As pre­ ments do not have any livestock activity. As the latter’s production is
viously mentioned, due to soil restrictions, there are plots dedicated carried out in a conventional way, agrochemicals are used during the

Table 2
Land use in the CBA (Chemical-Based Agriculture) fields for each campaign, and pesticides used for each crop.
Pesticide Applieda (month)*
C1 (July- C2 (November-
Site Fallow Growing
winter) spring)
Herbicides Herbicides Fungicides

S4- Wheat Wheat Glyphosate, metsulfuron-methyl (April- Glyphosate, metsulfuron-methyl, 2,4-D Epoxiconazol, tebuconazole
CBA June) (September-October) (October-December)
S5-
CBA
S6- Chemical Glyphosate, metsulfuron-methyl, 2,4-D Glyphosate, acetochlor (December-
soybean Epoxiconazole (January-April)
CBA fallow (May-October) January)
S10-
CBA
S7- Chemical Metsulfuron-methyl, dicamba (July- Epoxiconazol, tebuconazole (July-
Barley Glyphosate (April-June)
CBA fallow September) November)
S8- Chemical Glyphosate, metsulfuron-methyl (April- Glyphosate, metsulfuron-methyl, 2,4-D Epoxiconazol, tebuconazole
Wheat
CBA fallow June) (September-October) (October-December)
S9- Chemical Glyphosate, metsulfuron-methyl, 2,4-D
Chemical fallow
CBA fallow (May-October)
*
Based on surveys carried out with producers in the province of Buenos Aires, regarding the crops and pesticides applied (DP, 2015).

3
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

pre-sowing period (chemical fallow with herbicides), and at growth supernatants, 1 mL was filtered through a 0.22-μm pore size nylon filter
stages (pesticides and fertilizers). and transferred into a chromatographic vial for instrumental analysis.
For the extraction and analysis of glyphosate and its environmental
2.2. Sampling metabolite AMPA, the procedure proposed by Aparicio et al. (2013) was
followed. All the samples were spiked with isotopically
Two sampling campaigns were carried out, C1 during the winter labeled-glyphosate (glyphosate-2-13C,15N, 99 atom% 13C, 98 atom%
15
(July 2016) and C2 in the spring (November 2016), in order to evaluate N, purchased from Sigma Aldrich) as internal standard, in order to
temporal variations in environmental pesticide concentrations as a achieve a 100 μg L− 1 nominal concentration at instrumental analysis. In
consequence of the different application frequency and rates (DP, 2015), short, 5 g of soil was weighed into a 50 mL polypropylene tube and
in accordance with production cycles (Table 2). Ten sampling sites were extracted with 25 mL of a 100 mM K2HPO4 solution at pH = 9. Three
selected according to the spatial extent and surrounding CBA fields. Of sonication cycles of 15 min each were applied and, then, samples were
these, 3 locations were in the middle of the ABA field (S1 to S3), more centrifuged at 3000 rpm for 10 min. Of the supernatants, 1 mL was
than 300 m from the CBA-ABA borders, and 7 locations were in the derivatized with a 1 mg mL− 1 solution of 9-fluorenylmethyl chlor­
border area between CBA and ABA fields (S4 to S10). At each site, oformate (≥99.0%, for HPLC derivatization, Sigma Aldrich) in aceto­
subsurface soil samples from the first 10 cm were collected (Bento et al., nitrile, leaving it to react overnight in the dark. A clean-up was
2016). The location of the sampling sites and a scheme of the sampling performed by adding dichloromethane. Finally, the aqueous phase was
methodology is summarized in Fig. 1. filtered through a 0.22-μm nylon filter prior to instrumental analysis.
For sites S1-S3, which were in the ABA farm more than 300 m from
the CBA borders, samples were taken every 5 m on a 10 m by 10 m grid, 2.3.2. Instrumental analysis
as shown for S3 in Fig. 1, and combined to create a composite sample. At Instrumental analysis was performed with a Waters Acquity Ultra
the border sites (S4 to S10), samples were taken at different distances Performance Liquid Chromatography (UPLC) system coupled to a
from the fence that divides the fields: on the CBA side, at 40 m from the Quattro Premier XE tandem quadrupole mass spectrometer (MS/MS),
fence, and on the ABA side at 2, 10 and 40 m from the fence, as exem­ with an electrospray ionization (ESI) source. High purity nitrogen was
plified with the S7 in Fig. 1. At each distance, moving perpendicularly 50 used as the nebulizer and drying gas, and argon was used as the collision
m to each side, subsamples were collected every 5 m; then combined to gas. For the analysis of glyphosate and AMPA, the UPLC was equipped
obtain one sample (approximately 1 kg) for each distance from the with a C18 Acquity UPLC BEH column (1.7 μm, 50 × 2.1 mm), operating
fence. Therefore, in each campaign, 3 samples were taken from the at a flow of 0.50 mL min− 1, with a methanol-nanopure water gradient,
middle of the ABA farm (S1-S3), 7 from the CBA fields (S4-S10), and 21 both solvents 5 mM NH4Ac. For the chromatographic separation of the
samples from the agroecological side at different distances from the rest of the herbicides and fungicides, a C18 column (1.7 μm, 100 × 2.1
border with the conventional fields (7 sites × 3 distances). mm) was used, at a flow of 0.3 mL min− 1 with acetonitrile/methanol-
The distances from the border were chosen based on the recom­ nanopure water gradient (previously conditioned with formic acid).
mendations of the farmer and the agroecological field advisor, who The ESI ionization source operated in positive mode for all compounds,
observed less growth of their crops at a distance of up to 40 m from the except for 2,4-D and dicamba, for which the source was used in negative
CBA field. mode. The software MassLynx v4.1 and the TargetLynx package were
In each sampling campaign, the crop present in the ABA (Table 1) used for data analysis.
and CBA (Table 2) production lots was recorded, so as to be able to
associate pesticide concentrations results in the soil samples with the 2.3.3. Quality control and quality assurance
crop being grown. The performance of each analytical method was carried out by
quantifying isotopically labeled glyphosate and atrazine in each sample,
2.3. Pesticides studied using their recovery as a quality criterion. At the same time, blank re­
agents and random duplicate samples were performed. Pesticide quan­
The pesticides analyzed consisted of 16 herbicides: 2,4-D (2,4- tification was performed by means of an external standard calibration
dichlorophenoxy acetic acid), acetochlor, ametryn, atrazine, chlor­ curve, in a range of 0− 200 μg L− 1.
imuron, dicamba, diclosulam, glyphosate (GLP), flurochloridone, ima­ At least two transitions were used for each analyte, with the transi­
zapic, imazapyr, imazaquin, imazethapyr, metolachlor, metribuzin, tion of higher abundance used for quantification (Q) and the second
metsulfuron-methyl; 3 herbicides metabolites: amino­ used for confirmation (q). Subsequently, the Q/q area ratio in the pos­
methylphosphonic acid (AMPA), desethyl-atrazine, desisopropyl-atra­ itive samples was used in comparison to the standard as a criterion to
zine; and 3 fungicides: tebuconazole, epoxiconazole, metconazole. identity the pesticide (Furlong et al., 2001), accepting a deviation no
greater than 20% from the Q/q ratio of the standard (SANTE, 11945/,
2.3.1. Sample analysis 2015). The limits of detection (LOD) and quantification (LOQ) were
The soil samples were homogenized, ground and sieved through 2 calculated from the signal-to-noise ratio (S/N), applying times 3 and 5 as
mm mesh. A sub-sample (1 g) was taken and dried until constant weight factors, respectively. Recovery, linearity, precision, matrix effect, LOD
at 105 ◦ C to determine moisture content and express pesticide concen­ and LOQ were evaluated in accordance with the criteria established by
tration as μg kg− 1 dry weight (dw). SANTE, 11945/, 2015.
Herbicides and fungicides were extracted by employing the QuECh­ Percentage recoveries for isotopically labeled standards ranged from
ERS (Quick, Easy, Cheap, Effective, Rugged and Safe) procedure with 60% to 110% for atrazine-D5, and 80%–100% for glyphosate-2-13C,15N.
modifications proposed by Masiá et al. (2015) and Mac Loughlin et al. These factors were considered to correct the concentration in each in­
(2017). The procedure stated in brief: 5 g of wet soil were weighted in a dividual soil sample. The LOD and LOQ obtained for the QuEChERS
50 mL polypropylene tube, spiked with isotopically labeled atrazine procedure ranged from 0.01–7.8, and 0.03 to 30.4 μg kg− 1, respectively.
(atrazine-D5, purchased from Sigma Aldrich) as internal standard, so as With regard to glyphosate and AMPA, LOD and LOQ were 2 and 4 μg
to have a nominal concentration of 100 μg L− 1 at instrumental analysis, kg− 1.
and then 10 mL of nanopure water were added, the tubes were shaken,
and left to stand for 5 min. Then, 15 mL of acetonitrile were added, and 2.4. Data analysis
two 15-minutes sonication cycles were performed. The extraction salt
mixture (2 g NaCl and 6 g anhydrous MgSO4) was added and vigorously Descriptive statistics (median, minimum and maximum ranges,
shaken manually for 1 min, then centrifuged 10 min at 3000 rpm. Of the detection frequency) were performed with concentrations above the

4
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

LOD, and concentrations below LOQ (detectable, non-quantifiable) were farmalso detected in the CBA fields, but not the other way around. For
replaced by the mean value between corresponding LOD and LOQ. example, metsulfuron-methyl and acetochlor were only found at the
Statistical analyses were performed only on glyphosate and AMPA CBA. The herbicide GLP and its environmental metabolite AMPA
concentrations, with concentrations below LOD were replaced by LOD/ exhibited the highest detection frequencies in both production systems,
2. (Etchegoyen et al., 2017; Antweiler, 2015). The percentage AMPA (% with a 100% occurrence of the pair in the CBA samples, and 91%% and
AMPA) was calculated for each soil sample as the ratio of AMPA con­ 54% for GLP and AMPA in the ABA samples, respectively. These results
centration to the sum of glyphosate and AMPA concentration, AMPA= of the CBA fields are similar to those reported for soils in the European
[AMPA/(Glyphosate + AMPA)]*100 (Battaglin et al., 2014; Silva et al., Union (Silva et al., 2019), where GLP and AMPA also proved to be the
2018). Since the concentration data did not follow a normal distribution most detected compounds. In Argentina, the occurrence of GLP and
(Shapiro-Wilk test), nonparametric tests were used for its analysis. AMPA detected for the CBA are comparable to those reported by other
Wilcoxon–Mann–Whitney test was used to assess significant differences authors for conventional fields (Aparicio et al., 2013; Primost et al.,
in concentrations between production systems, and for temporal varia­ 2017). Distinctly, the detection frequencies of GLP and AMPA in the
tions within each system. For spatial analysis, the Kruskal-Wallis test ABA soils were 25% and 75%, respectively, higher than those reported
was employed; if differences were statistically significant, multiple a by Aparicio et al. (2013) in fields without target applications: 25% GLP
posteriori comparisons were performed according to the guidelines and 75% AMPA.
proposed by Conover (1999). The relationship between compounds, Additionally, by analyzing the profiles of the quantified pesticides, it
where concentrations were above the LOD, each production system was is observed that GLP and AMPA represented 90% of the total mass load
analyzed using Spearman correlations. For all tests, the level of signifi­ on average in both production systems (Table 3), in accordance to trends
cance was set at p < 0.05. Statistical analysis was performed using published for soils in Europe by Silva et al. (2019), and in other envi­
InfoStat (version 2020I) and STATISTICA (Stat Soft, Inc. 2001; version ronmental matrices, such as sediments, in Argentina (Mac Loughlin
7). et al., 2017). This result shows the ubiquity of GLP and AMPA in agro­
productive soils from the region, both where the herbicide is sprayed
3. Results and discussion (Primost et al., 2017; Soracco et al., 2018), as well as in agroecological
soils. The results of both detection frequency and mass load are
3.1. Detection, mass load and relative proportion of occurrence of the consistent with the market data available, with glyphosate being the
studied pesticides most widely used pesticide in Argentina, representing 62% of the
commercialized pesticides, and an average use of 200 million liters ac­
Recovery values, LOD and LOQ were consistent to those published by cording to the last available report, published in 2012 (Woodburn, 2000;
other authors (Aparicio et al., 2013; Primost et al., 2017; Masiá et al., CASAFE, 2012).
2015) and in accordance with the ranges accepted by the SANTE regu­ In the case of fungicides, the frequency of detection of tebuconazole
lation, 11945/, 2015 for the analysis of pesticide residues. in CBA fields was higher than reported by Silva et al. (2019) in the
Only 9 of the 22 analyzed pesticides were found above LOD: 5 her­ European Union (60%versus 12%), while epoxiconazole was detected at
bicides: 2,4-D, atrazine, acetochlor, glyphosate (GLP), metsulfuron- similar levels (22 versus 24%). Regional comparisons are limited since
methyl; 2 metabolites: desethyl-atrazine and AMPA; and 2 fungicides: there are few works in Argentina about multiresidue analysis of pesti­
epoxiconazole and tebuconazole. The detected pesticides were in cides in soils (with the exception of GLP and AMPA). Nevertheless,
agreement to those reported for registered applications in CBA accord­ atrazine, tebuconazole, metsulfuron-methyl, acetochlor, GLP, and
ing to regional official surveys (DP, 2015) for crops observed at the time AMPA were reported in surface waters of the same present study region
of sampling (cf.Table 2). Regarding the co-occurrence of these com­ (De Gerónimo et al., 2014; Okada et al., 2018), reflecting the regional
pounds, mixtures of 3 or more pesticides were found in 93% and 32% of relevance of the detected pesticides. The reach and ubiquity of the
the CBA (maximum 6), and ABA (maximum 5) soil samples, respec­ pesticides used in chemical-based systems in the Pampas region, imply a
tively. The analysis of pesticides in European soils that receive direct scenario comparable to that observed in other continents such as the
applications shows that co-occurrence is common (Silva et al., 2019). North America and the Europe, showing the general consequences of the
However, this study shows that this pattern was repeated in agroeco­ CBA system, regardless of the country under consideration (Battaglin
logical soils with no pesticide application. et al., 2014; Farenhorst and Andronak, 2015; Silva et al., 2019).
Fig. 2 shows the pesticide detection frequencies, which were always
higher in CBA than in ABA, with all pesticides detected in the ABA 3.2. Pesticide in soils related to management in studied agricultural
systems

The frequency and concentrations obtained from the pesticides

detected, separated by production system and by sampling campaign are
shown in Table 3. As a general trend, an increase in the co-occurrence of
pesticides was observed in C2. In C1, at least 3 pesticides were found in
85% and 27% of the CBA and ABA samples, respectively; while in C2,
both values increased, reaching 100% and 35% of the samples in each
agroproductive system.
For CBA fields, GLP and AMPA were detected in every sample
analyzed in both C1 and C2. The frequency of detection of 2,4-D, atra­
zine, metsulfuron-methyl and epoxiconazole increased in the second
campaign for both systems, while tebuconazole decreased from C1 to C2
in the CBA samples (Table 3). As indicated in Table 2, most CBAfields
did not have crops at the time of C1 and were under chemical fallow
based on GLP, 2,4-D and metsulfuron-methyl. The detection of herbi­
cides and the highest degree of co-occurrence of residues in C2, might be
a consequence of the adding post-sowing spraying of these pesticides in
Fig. 2. Detection frequency of the pesticides analyzed by production model. agreement with the application schedule for the crops present in CBA
(CBA: chemical-based agriculture; ABA: agro-ecological-based agriculture). (Table 2). Atrazine is commonly associated with crops such as corn and

5
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

Table 3
Median concentration, minimum-maximum range (min-max), and detection frequency (%) of the studied pesticides, differentiated by campaign and production system
(ABA: agroecological-based agriculture; CBA: chemical-based agriculture). The limits of detection (LOD) and quantification (LOQ) of each pesticide are detailed. All
concentrations are expressed in μg kg-1 dry weight (dw).
C1 (July-Winter) C2 (November-Spring)

Pesticide LOD LOQ ABA CBA ABA CBA

Glyphosate 26.96 237.68 32.24 580.2

(GLP) 2.00 4.00 (8.52− 98.93) (65.42− 383.12) (LOD-47.77) (86.75− 1268.92)
100% 100% 82% 100%
15.46 758.73 51.93 1423.99
AMPA 2.00 4.00 (LOD-48.29) (455.9− 1787.58) (LOD-114.01) (707.43− 2919.17)
44% 100% 64% 100%
21.91 25.49
2,4-D 4.40 14.90 (LOD-31.12) (LOD-38.52) (LOD-LOQ) (LOD-LOQ)
11% 29% 23% 43%
0.38 0.45 0.08 0.07
Atrazine 0.01 0.04 (LOD-1.16) (one sample) (LOD-2.12) (LOD-0.28)
22% 14% 36% 71%
Desethyl-atrazine 0.07 0.37 (LOD-LOQ) (LOD-LOQ) <LOD <LOD
17% 71%
3.46
Metsulfuron Methyl 0.09 0.29 <LOD <LOD <LOD (LOD-5.02)
71%
0.53
Acetochlor 0.15 0.49 <LOD (one sample) <LOD <LOD
14%
2.96 9.96 1.2 11.65
Epoxiconazole 0.60 1.80 (LOD-4.73) (one sample) (LOD-18.41) (LOD-13.35)
11% 14% 14% 29%
1.91 1,27
Tebuconazole 0.47 1.50 <LOD (LOD-11.07) (LOD-LOQ) (LOD-1.97)
71% 14% 57%

<LOD concentration below detection limit.

sorghum (DP, 2015). Therefore, its detection, along with its desethy­ 3.3. Glyphosate and AMPA as most relevant pesticides in soils of
lated metabolite, was unexpected since the presence of such crops was agroproductive systems
not registered in the surrounding CBA fields and no application registers
were observed from 2006 to date in ABA (Table 1 y Table 2). The 3.3.1. Occurrence and concentrations of GLP and AMPA
presence of this compound could be a consequence of older applications Considering that GLP and AMPA were the most detected analytes,
in CBA fieldsand its high persistence and stability in the environment, and that they represented over 90% of the total mass load of the
with a half-life in soils of up to 4 years (de Albuquerque et al., 2020). quantified pesticides, a spatio-temporal analysis for both compounds
There is evidence of the presence of atrazine in soils from the same and a correlation analysis between them were carried out. When
studied region, and the relevance of its atmospheric transport and wet comparing the agricultural systems studied, it was observed that the
deposition (Alonso et al., 2018). The opposite was observed for median concentrations of both the herbicide and its environmental
metsulfuron-methyl, which is applied both for chemical fallow and for metabolite were higher (p < 0.001, n = 62 for both compounds) in the
post-seeding maintenance of the present crops in CBA fields (DP, 2015). CBA than in the ABA system (Fig. 3). Different studies reported GLP and
Furthermore, applications of the herbicide are registered in ABA farm­
during 2010 and 2011 (Table 1). However, it was only detected during
the second sampling campaign in the CBA fields. This result is consistent
with other studies in soils from the south of the Province of Buenos Aires,
that evidence its lower half-life (38–51 days; Bedmar et al., 2006), its
low adsorption (Zanini et al., 2009), and its consequent leaching po­
tential (Azcarate et al., 2015) and mobilization towards surface water
bodies (De Gerónimo et al., 2014).
In the case of fungicides, there are regular applications at the
growing stage for all the crops registered in CBA fields. The main active
ingredients used are epoxiconazole, for which higher median concen­
trations were observed, and tebuconazole, this being the most frequently
fungicide detected in CBA sites. Spraying of fungicides takes place be­
tween July and November for barley, and between October and
December for wheat, while for soy crops it is between January and April
(cf.Table 2). The input of pesticides in the CBA crops is then reflected in
the presence of these compounds in the ABA establishment (no appli­
cations of fungicides registered from 2006), where the maximum con­
centration of epoxiconazole was detected (cf.Table 3), evidencing the Fig. 3. Concentration of glyphosate (GLP) and AMPA in soils from each
influence of the chemical-based system practices on the agroecological establishment (ABA: agroecological-based agriculture; CBA: chemical-based
field. agriculture). Concentration (μg kg-1 dw) is plotted on the y-axis on a loga­
rithmic scale for each pesticide displayed on the x-axis. In the box plots, the
marker indicates the median, the box the 25% and 75% percentiles and the
whiskers the non-outlier range.

6
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

AMPA concentrations in agroproductive CBA-type soils from Argentina,

in concentrations of the same order of magnitude as those found in the
CBA fields of this work (Aparicio et al., 2013; Lupi et al., 2015; Primost
et al., 2017; Alonso et al., 2018; Soracco et al., 2018; Okada et al., 2018).
Until now, no concentrations of pesticides have been reported in real
scale agroecological fields of the country. However, in non-agricultural
fields (without application registers in over 10 years) immersed in
productive locations, Aparicio et al. (2013) reported maximum con­
centrations of 41.4 μg kg− 1 for GLP and 43.2 μg kg− 1 for AMPA,
approximately less than half of those found in the ABA field of the
present study. In addition, Lupi et al. (2015), who assessed an area
separated from the agricultural field with GLP applications by a live
windbreak, detected GLP and AMPA at concentrations of 2.0 and 5.6 μg
kg-1, respectively, even without having received direct application. In
the present study, the concentrations found in the agroecological field
were up to 20 times higher than those, with a maximum of 98.93 μg kg-1
dw of GLP, and 114.01 μg kg-1 dw of AMPA.
The most probable mechanisms by which pesticides may mobilize to
the agroecological field are subsurface runoff and atmospheric transport
(Chang et al., 2011; Lupi et al., 2019), which includes the primary drift
from the application spray (Jensen and Olesen, 2014), and secondary
processes from wind-blown soil particles. The latter gains relevance due
to the high affinity of GLP and AMPA for adsorption processes on agri­
cultural soils (Okada et al., 2016; Lupi et al., 2019), as they can be
transported by the particulate material generated by wind erosion or
during soil tillage (Gill et al., 2006; Bento et al., 2017).
Currently, the physical separation between the fields of study is just a
wire fence. Research on the subject indicates that windbreaks mitigate
spray drift from CBA fields, and have been proven effective in protecting
non target areas, organisms and crops (Ucar et al., 2001; Baker et al.,
2018). Moreover, planted tree belts can reduce the transport of pesticide
in dust particles (Zaady et al., 2018). This demonstrates the relevance of
Fig. 4. Comparison of glyphosate (GLP) and AMPA concentrations from the
management strategies such as the use of windbreaks to minimize
two agricultural systems in the two sampling campaigns. a)CBA: chemical-
pesticide drift from conventional fields. However secondary drift can based agriculture. b)ABA: agroecological-based agriculture. Concentration (μg
reach large distances. Previous studies have demonstrated the associa­ kg-1 dw) is plotted on the y-axis for each pesticide displayed on the x-axis. In
tion of these compounds to fine particles (Bento et al., 2017; Ramirez the box plots, the marker indicates the median, the box the 25% and 75%
Haberkon et al., 2020), followed by mobilization by atmosphere, from percentiles and the whiskers the non-outlier range.
the spray zone, and then return to the soil by means of wet or dry
deposition (Chang et al., 2011; Farenhorst and Andronak, 2015). From found by other authors (Okada et al., 2018; Silva et al., 2018), who
the Pampas region of Argentina, Alonso et al. (2018) detected GLP and proposed that the dissipation of AMPA in soils that received frequent
AMPA in 80% of the rainwater samples, even in places far away from applications is less than that of glyphosate in the same environmental
crops. conditions, due to its longer half-life (Battaglin et al., 2005; Bento et al.,
2016).
3.3.2. Spatio-temporal variation of glyphosate and AMPA concentrations The spatial variation of GLP and AMPA was studied considering CBA
No statistically significant differences were observed between sam­ as a "source" close to ABA and analyzing the soils samples at different
pling campaigns in the GLP concentrations within each agricultural distances from said source. The GLP concentrations (Fig. 5a) showed
system model (CBA: p = 0.0973 n = 14; ABA: p = 0.2109, n = 48) significant differences between the source and the different distances
(Fig. 4). These results are in accordance with that described by Primost within the ABA farm for both sampling campaigns, but no significant
et al. (2017) for CBA systems, where GLP was characterized as differences were observed between the samples obtained from the ABA
pseudo-persistent in soils as a consequence of spraying frequencies that system. The same trend was found for AMPA in C1 (Fig. 5b). On the
are between 5–7 times per year. It is important to note that this was also other hand, during C2 a gradient was observed, decreasing in concen­
observed in the agroecological system studied, where use of this herbi­ tration as it moved away from the source.
cide was not part of their agricultural practice since 2006 (Table 1). It is noteworthy that GLP was quantified at distances more than 300
Regarding the concentrations of AMPA, there were significant dif­ m from the CBA borders (S1, S2, and S3, Fig. 1), and at concentrations
ferences between the campaigns within each system (CBA: p = 0.0379, n that did not present significant differences with respect to those found in
= 14; ABA: p = 0.0194, n = 48), with the highest median concentration the area near the CBA fields (Fig. 5a), showing the ubiquity of the spread
in C2 for both systems (Fig. 4). This can be explained by the glyphosate of pesticides from CBA fields. GLP has been previously defined as
sprays that occurred in the CBA system between sampling campaigns, as pseudo-persistent pesticide in CBA fields in Argentina (Primost et al.,
evidenced by higher median and maximum concentrations of GLP in 2017).The results of GLP and AMPA concentrations and detection fre­
CBA-C2 (cf.Table 3), and an increase in the degradation rate, of up to 30 quencies in soils of the ABA farm has also show a pseudo-persistence
times greater degradation of GLP to AMPA at higher temperatures and condition. Given that agricultural formulations based on GLP repre­
humidity (seasonality) due to the microbial activity in the soil (Bento sent 62% of the market, this pesticide is the most sold in Argentina
et al., 2016), regardless of the further degradation of AMPA. (CASAFE, 2012) and that several environmental studies have frequently
Lastly, the median concentrations of AMPA were significantly higher detected presence of this herbicide and its main metabolite in large river
than those of GLP for both sampling campaigns in CBA fields(C1: p = basins (Ronco et al., 2016), shallow lakes (Castro Berman et al., 2018),
0.0006, n = 14; C2: p = 0.0111, n = 14; Fig. 4a). These trends were also

7
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

and bottom sediments sampled from watersheds in areas of intense

agricultural activity (Primost et al., 2017; Okada et al., 2018). Those
same authors concluded that, in solid matrices, biotic degradation is
favored over other removal processes, such as wind erosion, surface
runoff and leaching, and is strongly associated with the formation of
AMPA through the breakdown of GLP in situ. The degradation rate,
measured through %AMPA (Fig. 5c), showed no significant differences
between C1 and C2 in the CBA fields (p = 0.5754, n = 14). All the while,
GLP was applied in CBA fields between sampling campaigns, and,
therefore, degradation processes had already occurred at the time of
sampling for C2.
Likewise, in C2, the ABA soils presented a positive and significant
GLP-AMPA correlation, and, between sampling campaigns, there was an
increase in the median concentrations of AMPA in samples up to 10 m
from the border (Fig. 5b). Furthermore, %AMPA increased signi­
ficatively from C1 to C2 in samples at 2 m (p = 0.0179, n = 11) and 10 m
(p = 0.0431, n = 7) within the ABA system, to the point that no signif­
icant differences were detected with respect to the CBA field for C2
(Fig. 5c). These results show the environmental pressure of herbicide
application cycle in CBA crops on the neighboring ABA farm, as, after
application, the agroecological system responded as a conventional
system, by generating the metabolite AMPA in samples up to 10 m from
the field boundary, with medians of %AMPA greater than 50%, as
described for conventional agricultural soils in the region (Okada et al.,
2018). Based on the observations of the agronomist who manages the
agroecological field, and these experimental results, it can be suggested
that the distance of direct influence is between 10 and 40 m.

3.4. Effects in edaphic organisms by pesticide in soils

The negative effects of pesticides on key soil organisms have been

evidenced (Pelosi et al., 2016; Wołejko et al., 2020), particularly for
herbicides, which can affect the trophic networks of which these or­
ganisms are part (Van Bruggen et al., 2018; Sharma et al., 2018).
There are ecotoxicological studies that report biological effects
relative to concentrations of pesticides in soil (mass of active ingredient/
soil mass) (USEPA, 2020). Published effects on soil organisms for some
of the quantified pesticides in the present study (atrazine, 2,4-D, epox­
iconazole, and tebuconazole) were at concentrations at least 5 times
higher than those reported here. However, 64% and 43% of GLP and
AMPA concentrations were higher than the lowest concentration re­
ported to generate negative effects on different organisms.
Fig. 6 details the cumulative distribution for GLP and effect con­
centrations for soil organisms, considering lethality and different sub-
lethal effects. From the CBA fields GLP soil concentrations, it was
observed that 7% of the quantified samples were above the concentra­
tion which is lethal to 50% of a population (median lethal concentration,
Fig. 5. Spatial distribution of a) glyphosate (GLP) and b) AMPA concentrations
LC50 = 1130 μg kg− 1) for the springtail Folsomia candida (Santos et al.,
in soils, and c) %AMPA. Equal letters (uppercase C1, lowercase C2) indicate
2012), and the Lowest Observed Effect Level (LOEL = 790 μg kg− 1) for
that there are no significant differences. In the figures (a) and (b), the con­
centration (μg kg-1 dw) is plotted on the y-axis on a logarithmic scale, while in cell viability in mycorrhizal arbuscular fungi (Druille et al., 2013).
figure (c), the %AMPA is plotted on the y-axis. Distances are plotted in the x- Furthermore, at least 50% of the concentrations found in CBA fields
axis (CBA: source; ABA: 2 m, 10 m, 40 m, and >300 m from border fence). In were higher than the reproductive median effective concentration (EC50
the box plots, the marker indicates the median, the box the 25% and 75% = 330 μg kg− 1) in F. candida (Santos et al., 2010). Effects at a
percentiles and the whiskers the non-outlier range. biochemical level should also be considered, since more than 65% of the
concentrations of the herbicide were above the LOEL (216 μg kg− 1) for
in the atmosphere (Alonso et al., 2018; Ramirez Haberkon et al., 2020), the activity of the glutathione-S-transferase antioxidant enzyme in the
horticultural areas (Mac Loughlin et al., 2017) and a ABA farm, leads to oligochaeta Eisenia hortensis (Hackenberger et al., 2018).
them being proposed as environmental tracers of chemical-based agri­ For the environmental metabolite AMPA, the only soil effect con­
culture for Argentina. centration reported is the reproductive LOEL (1000 μg kg− 1) for the
Relationships between glyphosate and AMPA were evaluated species Eisenia fetida (Dominguez et al., 2016), which, as previously
through correlation tests and %AMPA. Positive and significant correla­ stated, 43% of AMPA concentrations were above said level.
tions were observed in all cases (CBA-C1: r = 0.82, p = 0.0442, n = 7; For the agroecological system studied, and to the best of our
CBA-C2: r = 0.96, p = 0.0182, n = 7; ABA-C2: r = 0.63, p = 0.0369, n = knowledge, no concentrations were observed that generate direct effects
12), except for ABA-C1 (r = 0.57, p = 0.1360, n = 8). This analysis has at the individual level for GLP and AMPA. However, concentrations in
already been performed with concentration data from soils of CBA fields the surrounding CBA fields are capable of harming key organisms
(worms, invertebrates, and fungi) in the ecosystem functions, necessary

8
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

Fig. 6. Accumulated empirical distributions of glyphosate (GLP) concentrations in soils from ABA (agroecological-based agriculture) and CBA (chemical-based
agriculture). The vertical lines indicate reported sub-lethal (LOEL: Lowest Observed Effect Level, and EC50: Median Effective Concentration) and lethal (LC50:
Median Lethal Concentration) concentrations of glyphosate for soil organisms.

for optimal functioning of agroecological systems (Gurr et al., 2016), not detection and concentration records across Argentina, GLP and AMPA
only because of the aforementioned direct effects at the individual level, are proposed as environmental tracers of conventional agroproductive
but also in terms of community/ecosystemic level consequences. activity. The information presented in this work indicates that exposure
For instance, the absence of spontaneous vegetation (seen as “weeds” to GLP must be taken into consideration in future studies focusing on
from the perspective of the CBA system), and therefore the absence of agricultural biodiversity, since in Agroecology, its conservation and
host plants, is considered an indirect effect of herbicides on beneficial management are central for biological control and nutrient recycling.
organisms, causing a reduction in the population (Sharma et al., 2018). The studied agroecological system is reached by pesticides, both from
In Argentina, it has been detected a decrease in the abundance of the neighboring conventional system, and as a consequence of being
beneficial predatory organisms due to the decrease of semi-natural en­ located in a region dominated by the pesticides-dependent production
vironments (spontaneous vegetation) in areas close to wheat crops system. In order to minimize this situation, management tools must be
(Marasas et al., 2010). adopted to reduce the use of pesticides, and thus protect ecosystemic
Finally, it is worth noting the concern about the lack of information equilibriums that are the foundation of agroecological production and
associated with the effect of pesticide mixtures, particularly due to the can be affected by the presence of these pollutants.
co-occurrence found in the ABA farm samples (Silva et al., 2019;
Wołejko et al., 2020). It is necessary to develop ecotoxicological criteria Declaration of Competing Interest
to understand the possible effects on biota in cases like this work, where
there is co-occurrence of up to 5 compounds in at least one site of both The authors declare no conflict of interest. This study was supported
establishments, even without considering the potential presence of other by national funds from the Universidad Nacional de La Plata and grants
pesticides as insecticides, of greater toxic effect (Pelosi et al., 2014). from the National Agency for Scientific and Technological Promotion
(ANPCyT).
4. Conclusions
Acknowledgements
The present study showed the occurrence of 9 herbicides and fun­
gicides out of the 22 analyzed pesticides, of which there is a paucity of We thank the reviewers for their careful reading of our manuscript
information for the region. This is the first report on pesticides dynamics and their many insightful comments and suggestions. This research was
in real conventional-agroecological scenarios. Moreover, there is suffi­ supported by grants from Universidad Nacional de La Plata (UNLP
cient evidence to affirm that pressure is being exerted to the agroeco­ X801), ANPCyT (PICT 214-0919) and CONICET (PIP 2012-2014-0090).
logical establishment "La Aurora" by the agricultural practices carried To Juan and Erna Khier, owners of “La Aurora”, for their invaluable
out in the surrounding fields, as pesticides were detected there, some of collaboration in the development of this work, and for their commitment
which have not been applied in more than 10 years. Detection fre­ to Agroecology and nature. Dr. Virginia Aparicio for granting access to
quencies, mass loads and co-occurrence of compounds in the ABA field the UPLC-MS/MS equipment from the INTA-Estación Experimental
were conditioned by the actions implemented in the CBA fields, such as Balcarce. The study is dedicated to the memory of Dra. Alicia E. Ronco
application cycles, stage and type of crop and, in at least one of the for her invaluable contribution to the field of environmental science in
sampling campaigns, the direct impact of the applications on the CBA Argentina.
system was identified up to 10 m within the ABA system. Similar find­
ings were observed in both establishments, such as the association be­ References
tween GLP and AMPA, and the pseudo‑persistence of GLP in soils. Soil
concentrations of GLP and AMPA in the CBA plots reached such values, Alonso, L.L., Demetrio, P.M., Etchegoyen, M.A., Marino, D., 2018. Glyphosate and
atrazine in rainfall and soils in agroproductive areas of the pampas region in
up to 1268.92 and 2919.17 μg kg− 1 dw for GLP and AMPA, respectively, Argentina. Sci. Total Environ. 645, 89–96. https://doi.org/10.1016/j.
in the CBA fields, capable of causing sub‑lethal and lethal effects to scitotenv.2018.07.134.
organisms like springtails, oligochaetes, and to the soil microbial fauna, Altieri, M.A., 2017. Breve reseña sobre los origenes
́ y evolución de la Agroecologiá en
América Latina. Agroecologa 10 (2), 7–8.
all of which have direct implications on the structure and function of the Altieri, M., 2018. In: Altieri, M. (Ed.), Agroecology: The Science of Sustainable
edaphic ecosystem. Given their higher detection frequency and envi­ Agriculture. CRC Press. https://doi.org/10.1201/9780429495465.
ronmental concentrations, the fact that they were both detected in
samples taken more than 300 m from the perimeter, along with high

9
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

Antweiler, R.C., 2015. Evaluation of statistical treatments of left-censored environmental Spectrometry. Methods of Analysis by the U.S. Geological Survey National Water
data using coincident uncensored data sets. II. Group comparisons. Environ. Sci. Quality Laboratory.
Technol. 49, 13439–13446. https://doi.org/10.1021/acs.est.5b02385. Giesy, J.P., Dobson, S., Solomon, K., 2000. Ecotoxicological risk assessment for roundup
Aparicio, V.C., De Geronimo, E., Marino, D., Primost, J., Carriquiriborde, P., Costa, J.L., herbicide. Rev. Environ. Contam. Toxicol. 167, 35–120.
2013. Environmental fate of glyphosate and aminomethylphosphonic acid in surface Gill, T.E., Zobeck, T.M., Stout, J.E., 2006. Technologies for laboratory generation of dust
waters and soil of agricultural basins. Chemosphere 93, 1866–1873. https://doi.org/ from geological materials. J. Hazard. Mater. 132, 1–13. https://doi.org/10.1016/j.
10.1016/j.chemosphere.2013.06.041. jhazmat.2005.11.083.
Azcarate, M.P., Montoya, J.C., Koskinen, W.C., 2015. Sorption, desorption and leaching Gurr, G.M., Lu, Z., Zheng, X., Xu, H., Zhu, P., Chen, G., Yao, X., Chen, G.J., Zhu, Z.,
potential of sulfonylurea herbicides in Argentinean soils. J. Environ. Sci. Health B 50 Catindig, J.L., Villareal, S., Chien, H.V., Cuong, L.Q., Channoo, C., Chen, G.,
(4), 229–237. https://doi.org/10.1080/03601234.2015.999583. wattana, N., Lan, L.P., Hai, L.H., Chaiwong, J., Nicol, H.I., Perovic, D.J., Wratten, S.
Battaglin, W.A., Kolpin, D.W., Scribner, E.A., Kuivila, K.M., Sandstrom, M.W., 2005. D., Heong, K.L., 2016. Multi-country evidence that crop diversification promotes
Glyphosate, other herbicides, and transformation products in midwestern streams. ecological intensification of agriculture. Nat. Plants 2, 16014. https://doi.org/
JAWRA 41, 323–332. https://doi.org/10.1111/j.1752-1688.2005.tb03738.x. 10.1038/nplants.2016.14.
Battaglin, W.A., Meyer, M.T., Kuivila, K.M., Dietze, J.E., 2014. Glyphosate and its Hackenberger, D.K., Stjepanovic, N., Loncaric, Z., Hackenberger, B.K., 2018. Acute and
degradation product AMPA occur frequently and widely in U.S. soils, surface water, subchronic effects of three herbicides on biomarkers and reproduction in earthworm
groundwater, and precipitation. JAWRA 50, 275–290. https://doi.org/10.1111/ Dendrobaena veneta. Chemosphere 208, 722–730. https://doi.org/10.1016/j.
jawr.12159. chemosphere.2018.06.047.
Bedmar, F., Perdigon, J.A., Monterubbianesi, M.G., 2006. Residual phytotoxicity and Iermanó, M.J., 2015. Sistemas Familiares Mixtos De Agricultura Y Ganadería Pastoril De
persistence of chlorimuron and metsulfuron in soils of Argentina. J. Environ. Biol.27 La Región Pampeana: Eficiencia En El Uso De La Energía Y Rol Funcional De La
175–179. Agrobiodiversidad. Tesis Doctoral Facultad De Ciencias Agrarias. Universidad
Benamú, M.A., Schneider, M.I., Sánchez, N.E., 2010. Effects of the herbicide glyphosate Nacional de La Plata, Buenos Aires, Argentina. https://doi.org/10.35537/10915/
on biological attributes of Alpaida veniliae (Araneae, Araneidae), in laboratory. 46343.
Chemosphere 78, 871–876. https://doi.org/10.1016/j.chemosphere.2009.11.027. INDEC, 2020. Instituto Nacional De Estadística Y Censos - Censo Nacional Agropecuario
Bento, C.P., Yang, X., Gort, G., Xue, S., van Dam, R., Zomer, P., Mol, H.G., Ritsema, C.J., 2018 : Resultados Preliminares : Agricultura : Enero De 2020. - 1a Ed. - Ciudad
Geissen, V., 2016. Persistence of glyphosate and aminomethylphosphonic acid in Autónoma De Buenos Aires : Instituto Nacional De Estadística Y Censos -. ISBN 978-
loess soil under different combinations of temperature, soil moisture and light/ 950-896-571-4. https://cna2018.indec.gob.ar/ (accessed 10 August 2020).
darkness. Sci. Total Environ. 572, 301–311. https://doi.org/10.1016/j. INTA, 2002. Carta De Suelos De La República Argentina. Instituto Nacional de
scitotenv.2016.07.215. Tecnologìa Agropecuaria.. http://anterior.inta.gov.ar/suelos/cartas/series/Laprida.
Bento, C.P.M., Goossens, D., Rezaei, M., Riksen, M., Mol, H.G.J., Ritsema, C.J., htm
Geissen, V., 2017. Glyphosate and AMPA distribution in wind-eroded sediment INTA, 2014. Zonas Agroecológicas III Y IV Del Área De Influencia EEA Balcarce.
derived from loess soil. Environ. Pollut. 220, 1079–1089. https://doi.org/10.1016/j. Available in: https://inta.gob.ar/documentos/zonas-agroecologicas-iii-y-iv-del-area-
envpol.2016.11.033. de-influencia-eea-balcarce. (accessed may 2018).
Benton, T.G., Vickery, J.A., Wilson, J.D., 2003. Farmland biodiversity: is habitat ISAAA, 2016. Global Status of Commercialized Biotech/GM Crops: 2016. ISAAA Brief
heterogeneity the key? Trends Ecol. Evol. 18, 182–188. https://doi.org/10.1016/ No. 52.ISAAA, Ithaca, NY. ISBN: 978-1-892456-66-4.
S0169-5347(03)00011-9. ISAAA, 2017. Global Status of Commercialized Biotech/GM Crops in 2017: Biotech Crop
CASAFE, 2012. Cámara Argentina de sanidad agropecuaria y fertilizantes, mercado Adoption Surges As Economic Benefits Accumulate in 22 Years. ISAAA Brief No. 53.
argentino 2012 de productos fitosanitarios KLEFFMANNGROUP Argentina. Buenos ISAAA, Ithaca, NY. ISBN: 978-1-892456-67-2.
Aires. Iturburu, F.G., Calderon, G., Amé, M.V., Menone, M.L., 2019. Ecological Risk Assessment
Castro Berman, M., Marino, D.J., Quiroga, M.V., Zagarese, H., 2018. Occurrence and (ERA) of pesticides from freshwater ecosystems in the Pampas region of Argentina:
levels of glyphosate and AMPA in shallow lakes from the Pampean and Patagonian legacy and current use chemicals contribution. Sci. Total Environ. 691, 476–482.
regions of Argentina. Chemosphere 200, 513–522. https://doi.org/10.1016/j. https://doi.org/10.1016/j.scitotenv.2019.07.044.
chemosphere.2018.02.103. Jensen, P.K., Olesen, M.H., 2014. Spray MassBalance in PesticideApplication: A Review
Cerdá, E.O., Sarandón, S.J., Flores, C.C., 2014. El caso de “La Aurora”: un ejemplo de Crop Protection, 61, pp. 23–31. https://doi.org/10.1016/j.cropro.2014.03.006.
aplicación del enfoque agroecológico en sistemas extensivos del sudeste de la Leguizamón, A., 2014. Modifying Argentina: GM soy and socio-environmental change.
provincia de Buenos Aires, Benito juárez, Argentina. In: Sarandón, S.J., Flores, C.C. Geoforum 53, 149–160. https://doi.org/10.1016/j.geoforum.2013.04.001.
(Eds.), Agroecología: Bases teóricas para el diseño y manejo de Agroecosistemas Lupi, L., Miglioranza, K.S.B., Aparicio, V.C., Marino, D., Bedmar, F., Wunderlin, D.A.,
sustentables. EDULP, La Plata, pp. 437–463. E-Book: ISBN 978-950-34-1107-0. 2015. Occurrence of glyphosate and AMPA in an agricultural watershed from the
Chang, F., Simcik, M.F., Capel, P.D., 2011. Occurrence and fate of the herbicide southeastern region of Argentina. Sci. Total Environ. 536, 687–694. https://doi.org/
glyphosate and its degradate aminomethylphosphonic acid in the atmosphere. 10.1016/j.scitotenv.2015.07.090.
Environ. Toxlcology Chem. 30, 548–555. https://doi.org/10.1002/etc.431. Lupi, L., Bedmar, F., Puricelli, M., Marino, D.J.G., Aparicio, V.C., Wunderlin, D.,
Conover, W.J., 1999. Practical Nonparametric Statistics. John Wiley & Sons, Inc., New Miglioranza, K.S.B., 2019. Glyphosate runoff and its occurrence in rainwater and
York. subsurface soil in the nearby area of agricultural fields in Argentina. Chemosphere
de Albuquerque, F.P., de Oliveira, J.L., Moschini-Carlos, V., Fraceto, L.F., 2020. An 225, 906–914. https://doi.org/10.1016/j.chemosphere.2019.03.090.
overview of the potential impacts of atrazine in aquatic environments: perspectives Mac Loughlin, T.M., Peluso, L., Marino, D.J.G., 2017. Pesticide impact study in the peri-
for tailored solutions based on nanotechnology. Sci. Total Environ. 700, 134868 urban horticultural area of Gran La Plata, Argentina. Sci. Total Environ. 598,
https://doi.org/10.1016/j.scitotenv.2019.134868. 572–580. https://doi.org/10.1016/j.scitotenv.2017.04.116.
De Gerónimo, E., Aparicio, V.C., Bárbaro, S., Portocarrero, R., Jaime, S., Costa, J.L., Mac Loughlin, T.M., Peluso, M.L., Etchegoyen, M.A., Alonso, L.L., de Castro, M.C.,
2014. Presence of pesticides in surface water from four sub-basins in Argentina. Percudani, M.C., Marino, D.J.G., 2018. Pesticide residues in fruits and vegetables of
Chemosphere 107, 423–431. https://doi.org/10.1016/j.chemosphere.2014.01.039. the argentine domestic market: occurrence and quality. Food Control 93, 129–138.
Dominguez, A., Brown, G.G., Sautter, K.D., De Oliveira, C.M.R., De Vasconcelos, E.C., https://doi.org/10.1016/j.foodcont.2018.05.041.
Niva, C.C., Bartz, M.L.C., Beda, J.C., 2016. Toxicity of AMPA to the earthworm MAGyP, 2019. Datos Abiertos Del Ministerio De Agricultura Ganadería Y Pesca. https
Eisenia andrei bouche, 1972 in tropical artificial soil. Sci. Rep. 6, 19731. https://doi. ://datos.magyp.gob.ar/series/api.
org/10.1038/srep19731. Marasas, M.E., Sarandón, S.J., Cicchino, A., 2010. Semi-Natural Habitats and Field
DP (Defensor del Pueblo), 2015. Relevamiento De La Utilización De Agroquímicos En La Margins in a Typical Agroecosystem of the Argentinean Pampas as a Reservoir of
Provincia De Buenos Aires. Mapa de Situación e Incidencia Sobre la Salud. Defensor Carabid Beetles’. J. Sustain. Agric. 34, 153–168. https://doi.org/10.1080/
del Pueblo, Buenos Aires, Argentina, pp. 96–99. 10440040903482563.
Druille, M., Omacini, M., Golluscio, R.A., Cabello, M.N., 2013. Arbuscular mycorrhizal Masiá, A., Vásquez, K., Campo, J., Picó, Y., 2015. Assessment of two extraction methods
Fungi are directly and indirectly affected by glyphosate ApplicationAppl. Agric. to determine pesticides in soils, sediments and sludges. Application to the Túria
Ecosyst. Environ. 72, 143–149. https://doi.org/10.1016/j.apsoil.2013.06.011. River Basin. J. Chromatogr. A 1378. https://doi.org/10.1016/j.
Etchegoyen, M.A., Ronco, A.E., Almada, P., Abelando, M., Marino, D.J.G., 2017. chroma.2014.11.079, 19–3.
Occurrence and fate of pesticides in the Argentine stretch of the Paraguay-Paraná Okada, E., Costa, J.L., Bedmar, F., 2016. Adsorption and mobility of glyphosate in
basin. Environ. Monit. Assess. 189, 63. https://doi.org/10.1007/s10661-017-5773- different soils under no-till and conventional tillage. Geoderma 263, 78–85. https://
1. doi.org/10.1016/j.geoderma.2015.09.009.
Evans, A.N., Llanos, J.E.M., Kunin, W.E., Sophie, E.F., Evison, S.E.F., 2018. Indirect Okada, E., Perez, D., Gerónimo, E.De, Aparicio, V., Massone, H., Costa, J.L., 2018. Non-
effects of agricultural pesticide use on parasite prevalence in wild pollinators. Agric. point source pollution of glyphosate and AMPA in a rural basin from the southeast
Ecosyst. Environ. 258, 40–48. https://doi.org/10.1016/j.agee.2018.02.002. Pampas, Argentina. Environ. Sci. Pollut. Res. 25, 15120–15132. https://doi.org/
FAO, 1990. Food and agricultureorganization of the United nations. Artículo 2◦ Del 10.1007/s11356-018-1734-7.
Código Internacional De Conducta Para La Distribución Y Utilización De Plaguicidas. Pelosi, C., Barot, S., Capowiez, Y. Hedde, M., Vandenbulckeet, F., 2014. Pesticides and
FAO, 2016. 52 Profiles on Agroecology: La Aurora. http://www.fao.org/3/a-be861s.pdf. earthworms. A review. Agron. Sustain. Dev. 34, 199–228. https://doi.org/10.1007/
Farenhorst, A., Andronak, L.A., 2015. Bulk deposition of pesticides in a canadian city: s13593-013-0151-z.
part 1. Glyphosate Agric. Pesticides. https://doi.org/10.1007/s11270-015-2343- Primost, J.E., Marino, D., Aparicio, V.C., Costa, J.L., Carriquiriborde, P., 2017.
2344. Glyphosate and AMPA, “pseudo-persistent” pollutants under real-world agricultural
Furlong, E.T., Anderson, B.D., Werner, S.L., Soliven, P.P., Coffey, L.J., Burkhardt, M.R., management practices in the Mesopotamic Pampas agroecosystem, Argentina.
2001. Determination of Pesticides in Water by Graphitized Carbon-Based Solid- Environ. Pollut. 229, 771–779. https://doi.org/10.1016/j.envpol.2017.06.006.
Phase Extraction and High-Performance Liquid Chromatography/Mass Ramirez Haberkon, N.B., Aparicio, V.C., Buschiazzo, D.E., De Gerónimo, E., Aimar, S.B.,
Costa, J.L., Mendez, M.J., 2020. Glyphosate and AMPA concentrations in the

10
C. Bernasconi et al. Agriculture, Ecosystems and Environment 313 (2021) 107341

respirable dust emitted experimentally by soil aggregates, shortly after herbicide Silva, V., Montanarella, L., Jones, A., Fernandez-Ugalde, O., Mol, H.G.J., Ritsema, C.J.,
application. Geoderma 369, 114334. https://doi.org/10.1016/j. et al., 2018. Distribution of glyphosate and aminomethylphosphonic acid (AMPA) in
geoderma.2020.114334. agricultural topsoils of the European Union. Sci. Total Environ. 621, 1352–1359.
Rauchecker, M., 2019. The territorial and sectoral dimensions of advocacy – The conflicts https://doi.org/10.1016/j.scitotenv.2017.10.093.
about pesticide use in Argentina. Polit. Geogr. 75, 102067 https://doi.org/10.1016/ Silva, V., Mol, H.G.J., Zomer, P., Tienstra, M., Ritsema, C.J., Geissen, V., 2019. Pesticide
j.polgeo.2019.102067. residues in European agricultural soils – a hidden reality unfolded. Sci. Total
Ronco, A.E., Marino, D.J.G., Abelando, M., Almada, P., Apartin, C.D., 2016. Water Environ. 653, 1532–1545. https://doi.org/10.1016/j.scitotenv.2018.10.441.
quality of the main tributaries of the Paraná Basin: glyphosate and AMPA in surface Soracco, C.G., Villarreal, R., Lozano, L.A., Vittori, S., Melani, E.M., Marino, D.J., 2018.
water and bottom sediments. Environ. Monit. Assess. 188, 458. https://doi.org/ Glyphosate dynamics in a soil under conventional and no-till systems during a
10.1007/s10661-016-5467-0. soybean growing season. Geoderma 323, 13–21. https://doi.org/10.1016/j.
SANTE 11945, 2015. Guidance Document on Analytical Quality Control and Method geoderma.2018.02.041.
Validation Procedures for Pesticides Residues Analysis in Food and Feed. European Trudeau, V.L., Thomson, P., Zhang, W.S., Reynaud, S., Navarro-Martin, L., Langlois, V.S.,
Commission Directorate - General for Health and Food Safety. 2020. Agrochemicals disrupt multiple endocrine axes in amphibians. Mol. Cell.
Santos, M.J.G., Ferreira, M.F.L., Cachada, A., Duarte, A.C., Sousa, J.P., 2012. Pesticide Endocrinol. 513–110861. https://doi.org/10.1016/j.mce.2020.110861.
application to agricultural fields: effects on the reproduction and avoidance US-EPA, 2020. US-EPA ECOTOX database, USA (accessed June 2020). https://cfpub.epa.
behaviour of Folsomia candida and Eisenia andrei. Ecotoxicology 21, 2113–2122. gov/ecotox/.
https://doi.org/10.1007/s10646-012-0963-7. Van Bruggen, A.H.C., He, M.M., Shin, K., Mai, V., Jeong, K.C., Finckh, M.R., Morris, J.G.
Sarandón, S.J., Flores, C.C., 2014. La insustentabilidad del modelo agrícola actual. In: Jr., 2018. Environmental and health effects of the herbicide glyphosate. Sci. Total
Sarandón, S.J., Flores, C.C. (Eds.), Agroecología: Bases teóricas para el diseño y Environ. 616–617, 255–268. https://doi.org/10.1016/j.scitotenv.2017.10.309.
manejo de Agroecosistemas sustentables. EDULP, La Plata, pp. 13–41. E-Book: ISBN Wołejko, E., Jabłońska-Trypuć, A., Wydro, U., Butarewicz, A., Łozowicka, B., 2020. Soil
978-950-34-1107-0. biological activity as an indicator of soil pollution with pesticides –A review. Appl.
Sarandon, S.J., Marasas, M.E., 2017. Brief history of agroecology in Argentina: origins, Soil Ecol. 147, 103356 https://doi.org/10.1016/j.apsoil.2019.09.006.
evolution, and future prospects. Agroecol. Sustain. Food Syst. 41, 238–255. https:// Woodburn, A.T., 2000. Glyphosate: production, pricing and use worldwide. Pest Manag.
doi.org/10.1080/21683565.2017.1287808. Sci. 56, 309–312.
SENASA, 2019. Servicio Nacional de sanidad y calidadagroalimentaria “Situación de la Zanini, G.P., Maneiro, C., Waiman, C., Galantini, J.A., Rosell, R.A., 2009. Adsorption of
producciónorgánica en la Argentina durante el año 2018”. Dirección Nacional de metsulfuron-methyl on soils under no-till system in semiarid Pampean Region,
Inocuidad y Calidad Agroalimentaria. Dirección de Calidad Agroalimentaria Argentina. Geoderma 149, 110–115. https://doi.org/10.1016/j.
Coordinación de Productos Ecológicos, Buenos Aires, Argentina. geoderma.2008.11.025.
Sharma, A., Jha, P., Reddy, G.V.P., 2018. Multidimensional relationships of herbicides
with insect-crop food webs. Sci. Total Environ. 643, 1522–1532. https://doi.org/
10.1016/j.scitotenv.2018.06.312.

11