Air and rainwater contaminated with Glyphosate

Air and rainwater contaminated with Glyphosate

In 2012, we released scientific studies from the School of Public Health at the University of Minnesota and the United States Geological Survey of contamination of rainwater with glyphosate in the United States, which showed that up to 2.5 ugr of glyphosate could be collected for each liter of rain.

In Argentina, CONICET researchers have now collected up to 67.2 ugr per liter of rainwater, showing that the drifts of pesticides are uncontrollable.

Our previous Report for 2012 is accessible at the following link: http: //

In it you can also access scientific reports of contamination and persistence of glyphosate in surface and groundwater from Argentina, Spain and the US.

Now we spread the Spanish translation of the publication of the group of scientists of the National University of La Plata in:Science of the Total Environment 645 (2018) 89–96, entitled:Glyphosate and atrazine in rain and soils of agro-productive areas in the pampas region of Argentina so that it is accessible to all doctors in our Network, neighbors who breathe this polluted air throughout the country, journalists, farmers, agronomists and the general public, so that it is known as the current toxic agriculture model pollutes the entire environment generating increasing levels of exposure to pesticides that are applied uncontrollably.

Authors: Lucas L. Alonso, Pablo M. Demetrio, M. Agustina Etchegoyen, Damián J. Marino: Environmental Research Center (CIM), Faculty of Exact Sciences, National University of La Plata, La Plata, Buenos Aires, Argentina. National Council for Scientific and Technical Research (CONICET), Buenos Aires, Argentina

To access the original pdf click here: ARG rain glyph (302)

To access a pdf of the translation click here: Spanish rain (334)

Research Highlights:

GLP, AMPA and ATZ were found in 80% of the rainwater samples in the Argentine pampas.

Soils as a source of herbicides did not define a local atmospheric footprint.

Mean concentrations of GLP in rainwater were associated with the dynamics of precipitation.

The ATZ levels did not follow a specific pattern neither for the rainwater nor for the soil samples.


The presence in the atmosphere of glyphosate (GLP) and atrazine (ATZ), the pesticides that dominate the market in Argentina, was investigated through rain, as the main climatic phenomenon associated with wet deposition, both by analyzing the Source-receptor relationships with the soil as well as with the climatic influences that can condition this transport and by estimating the annual deposition on the surface of the Argentine pampas. Samples (n = 112) of rainwater were collected along each precipitation in urban areas of the Pampean zone with different degrees of land use in the production of extensive crops together with samples of subsurface soil (n = 58) of peri-urban sites.

Herbicides were analyzed by liquid chromatography with mass spectrometry and were detected in80% of rain samples at medium to maximum concentrations of 1.24-67.3 μg / L of glyphosate (GLP) and 0.22-26.9 μg / L of atrazine (ATZ), while aminomethylphosphonic acid (AMPA) was detected in 34% of the rain samples (0.75 -7.91 μg / L).On soils, GLP was recorded more frequently (41%; 102-323 μg · / kg) followed by ATZ (32%; 7-66 μg / kg) and AMPA (22%; 223-732 μg / kg).

Peak concentrations of GLP quantified in rainwater exceeded previous reported concentrations of levels reported for the United States and Canada. No associations were observed between the soil and the concentrations of rainwater in the same monitoring areas, despite the action of the soil as a source, as evidenced by the AMPA present in the rainwater. Median concentrations of LPG in rainfall were significantly associated with pluvial isobars, with an increasing gradient from east to west, with an inverse pattern to that of annual rainfall volumes (more rain less glyphosate, less mm of rain more glyphosate); while the ATZ levels in rainwater did not exhibit a characteristic spatial configuration. The estimated annual deposition of LPG, due to rain, indicated that more than one source of the herbicide can contribute to its presence in the atmosphere and indicates the relevance of rain to the surface concentrations of this pollutant.


A solution to the need for greater increases in crop yields in extensive agriculture has been sought through the implementation of a technological package that involves the introduction of genetic modifications in species that are tolerant to pesticides (Leguizamón, 2014) within a context of pest management based mainly on the use of synthetic pesticides. Herbicides are the most widely used pesticides on the market, with particular emphasis on the use of glyphosate [N (phosphonomethyl) glycine: GLP] and atrazine (2 chloro 4 ethylamino 6 isopropylamino 1,3,5 triazine: ATZ), both of which regional level (Leguizamón, 2014) and globally (Benbrook, 2016). During the 2013-2014 agricultural season, 18.7 million ha were planted with herbicide-tolerant soybean and corn varieties in Argentina (MINAGRI, 2017), with 80% of the production corresponding to the pampas region, resulting in in a demand of 182.5 million liters (L) or kilograms of LPG formulations. Although no specific data were available for ATZ, this agent was reported as the third most used compound with 62 million kg or L of herbicides apart from GLP (CASAFE, 2013). Therefore, we would estimate that the use of that compound was probably about 10-15 million kg or L.

When these formulations are applied to fields, almost 20-30% of the spray dose does not reach the target area as a result of airborne transmission or primary drift. The magnitude of this effect depends on conditions that vary from the type of formulation and the climate during the operation to variables that are difficult to quantify, such as the experience of the applicator (Gil and Sinfort, 2005). Once these herbicides reach the surface layer, the persistence in the soil of GLP, its main degradation product aminomethylphosphonic acid (AMPA), and ATZ is reported to be for months or years (Simonsen et al., 2008; Vonberg et al. al., 2014). That the concentrations of these herbicides persist in the soil (Aparicio et al., 2013; Primost et al., 2017) points out the role of the soil matrix as a source for their eventual re-emission into the atmosphere.

Depending on the physicochemical properties of the active compounds, post-application emissions can occur, reaching losses of almost 90% of the product through volatilization that can last a few days or for weeks (Bedos et al., 2002) together with the action of erosion. wind power by dragging and finally lifting pesticide-laden soil particles from that matrix into the air column (Bidleman, 1988). The dynamics of the pesticide in the environment include continuous transfers between these two matrices. Although this movement normally occurs between only adjacent areas, studies have shown that pesticides can nevertheless travel long distances to be detected in extremely remote locations from widely removed locations from agricultural areas such as polar regions (Baek et al., 2011 ; Unsworth et al., 1999). ATZ and its metabolites have been detected predominantly in the vapor phase (Cooter et al., 2002) and in even 200-300 km from the closest cultivated fields (Thurman and Cromwell, 2000), while GLP and AMPA have been reported in the air near the areas of application (Chang et al., 2011; Morshed et al., 2011), which indicates a short-range transport within the atmosphere mainly in association with particulate matter (Bento et al., 2017; Chang et al., 2011). The probable atmospheric dynamics of these herbicides is that they are transported long distances and then returned to the surface by wet and dry deposition (Goel et al., 2005; Messing et al., 2013).

Wet deposition is considered the predominant route to remove the herbicide from the atmosphere, either by dissolution in rainwater for compounds in the vapor phase, or by washing of particles (Bidleman, 1988; Goel et al., 2005). In this sense, 97% of GLP can be eliminated with weekly rains greater than 30 mm, with maximum concentrations of 2.5 μg / L, as has been detected in rainwater in the United States (Chang et al., 2011). In a large worldwide study, atmospheric ATZ was detected in rainwater in France (Trautner et al., 1992), Poland (Grynkiewicz et al., 2003), the United States (Majewski et al., 2000; Vogel et al., 2008), Germany (Hüskes and Levsen, 1997) and Italy (Trevisan et al., 1993); with maximum values ​​of 40 μg / L registered in the United States (Nations and Hallberg, 1992).

Furthermore, Goolsby et al. (1997) estimated an annual contribution of 110,000 kg of ATZ to the Mississippi River basin from the atmosphere, a matrix that can be considered as an important source of this herbicide for surface water bodies.

Despite the widespread use of pesticides in Latin America, little information on the dynamics of herbicides within the atmosphere is available in this geographic region. And as GLP was recently categorized as "probably carcinogenic to humans" by the International Agency for Research on Cancer (Portier et al., 2016), and in view of the volumes of these agents applied to the fields and detected in the air; An analysis of the extent of the degree of herbicide transport and the possibility of depositing these compounds on the earth's surface is relevant and necessary. The objective of this work was, therefore, to study the presence of herbicides in the rain (as the main means of moisture deposition) and to evaluate the corresponding spatial and temporal variations and those relationships with the herbicide contents in the soil. and the climatic conditions in the Argentine pampas.

  1. Materials and methods

2.1. Study area

The study area comprised four of the five Argentine provinces of the Pampas region (excluding La Pampa): Buenos Aires, Entre Ríos, Santa Fe and Córdoba; covering an approximate area of ​​60 million hectares. This region is the source of 90% of the soybeans and between 80 and 90% of the wheat, corn, sorghum, barley and sunflower produced in the country. The predominantly temperate and humid climate, with hot summers and no dry seasons, is responsible for these productions. The annual extent of precipitation is between 600 mm in the southwest and 1200 mm in the northeast, while the respective mean annual maximum and minimum temperatures are 18 and 6 ° C in the south and 26 and 14 ° C in the North. The annual precipitation gradient varies in direction according to the different areas, that is, in the north, the rainfall decreases from east to west, while in the south from north to south. The most frequent distribution of annual precipitation within this area implies a maximum in the summer that decreases from autumn to winter and spring (Magrin et al., 2007). Seven representative locations within the Pampas provinces were selected (Fig. 1), consisting of two from Buenos Aires (BA), three from Córdoba (CB), and one each from Santa Fe (SF) and Entre Ríos (ER ). Table 1 provides descriptions of each site. The cultivation cycles were taken into account because they define a high season (mainly spring) and low season (summer to autumn) of herbicide application that were considered for the analysis of temporal variations.

The circular graphs of diameters proportional to log10 of the mean concentration of GLP + AMPA (μg / kg), indicate the relative areas in hectares of soybeans planted in the different locations indicated in the figure with white representing GLP and gray AMPA. Main surveillance areas: BK, Brinkman + HE, Hersilia (Santa Fe); I + MA, Malvinas Argentinas + MJ, Marcos Juárez (Córdoba); UR, Urdinarrian (Entre Ríos); LP: La Plata + CS, Coronel Suárez (Buenos Aires). The box indicates in black the location of the entire monitoring area within Argentina.

2.2. Rainwater samples

Each rainfall was individually monitored at each location (Trevisan et al., 1993). The sampling period was according to the application campaigns from October 2012 to April 2014 (Table 1; Ghida Daza and Urquiza, 2014). The samples were collected by direct entry of raindrops into 1 L polypropylene containers (Sakai, 2002) containing 100 ng of [13C, 15N] glyphosate ([13C, 15N] GLP) and 100 ng of [5D] atrazine ([5D] ATZ), as quality control and quality assurance systems. After each rain, the particulate material in the samples was filtered off through nylon membranes with pores of 0.45 μm in size and the soluble fraction frozen at -20 ° C until further analysis.

2.3. Soil samples

The presence of the two herbicides in soils was studied in different regions in which no previous data or publications were available (BK, MA, IT, LP, CS and HE), with each cardinal point selected for the presence of an environment peri-urban (Fig. 1). The samples were collected slightly below surface level, collecting soil from a defined area of ​​40 cm × 40 cm and at a depth of 5 cm (Feng and Thompson, 1990). This procedure was repeated 5 times in each field, with a sample being removed from each of the four corners at a distance of 20 m towards the center and a final sample taken from the center (i.e., five points). These subsamples from each field were mixed and homogenized in situ and a representative fraction transferred refrigerated to the laboratory. There, the soil samples were manually homogenized, ground and filtered through a 2 mm pore size sieve for subsequent storage at -20 ° C until the time of analysis.

2.4. Chemicals and reagents

The solvents used in the chemical and chromatographic analyzes were high performance liquid chromatography (HPLC) grade, while all salts were analytical grade (JT Baker-Mallinckrodt Baker Inc., USA). Pure water to nanogram grade was obtained in the laboratory by means of a Sartorius Arium water purification system (Sartorius AG, Göttingen, The Netherlands). Standards for GLP (99%), AMPA (98.5%), [13C, 15N] GLP, [5D] ATZ and 9-fluorenylmethyloxycarbonyl chloride (FMOC-Cl, HPLC grade at N99%) were purchased from Sigma Aldrich (St. Louis, MO, USA).

2.5. Chemical analysis

2.5.1. Sample preparation for chemical analysis LPG and AMPA. From each sample, 5 g of soil was weighed into a 50 ml Falcon ™ propylene tube and enriched with 500 ng of [13C, 15N] GLP. GLP and AMPA levels were determined according to Aparicio et al. (2013). The analytes were extracted with 25 mL of a 0.1 M K2HPO4 solution and the resulting extracts were sonicated 3 times for 10 min with shaking between cycles followed by centrifugation for 10 minutes at 3500 g. A 2 ml aliquot of both rainwater samples and the soil extracts were first adjusted to pH = 9 with sodium tetraborate (40 mM) and then 2 ml of FMOC-Cl solution in acetonitrile were added (Sancho et al. , nineteen ninety six). The preparation of the standard solution for the calibration curves and the reagent blanks was carried out under operating conditions equivalent to those used in the tests. All the derivatized samples were finally extracted with 5 mL of dichloromethane, centrifuged and the aqueous supernatant filtered through a membrane with a pore size of 0.45 μm for the determination with HLPC mass spectrometry. ATZ. From each sample, 5 g of soil were added with 150 ng of [5D] ATZ (at a nominal concentration in the instrumental analysis of 10 ng · mL-1) and extracted using the so-called QuEChERS method - which means "Quick, easy, cheap, effective, resistant and safe ”: described in Bruzzoniti et al. (2014). For the extraction, 15 ml of acetic acid 1% (v / v) in acetonitrile and the mixture was stirred manually for 1 minute, sonicated for 10 minutes; then 7.00 g of anhydrous MgSO4 and 2.00 g of sodium acetate were added, followed by manual stirring for 1 minute. Then the sample was centrifuged for 10 minutes at 3500 g and 1 ml of the upper organic solution was mixed with 1 ml of water. The resulting solution was finally filtered through a 0.45 µm pore size membrane for later instrumental analysis.

2.5.2. Instrumental analysis

Analysis was performed with an Agilent 1100 binary pump HPLC system (Agilent Technologies Inc., Miami, FL, USA) in conjunction with a VL quadrupole mass spectrometer with electrospray-ionization source (Agilent Technologies Inc. , Miami, FL, USA For GLP a reverse C18 chromatographic column (X-SELECT ™ 75mm × 4.6mm 3mm pore size from Waters Corp., Milford, MA, USA) was used which was kept at 25 ± 1 ° C. A gradient of methanol: water (with the mobile phases previously conditioned with 5 mM ammonium acetate) in 0.5 mL · min-1 was used. As described in Meyer et al. (2009) , the monitoring of selected ions in negative ionization mode was applied for the detection of GLP-FMOC, [13C, 15N] GLP-FMOC, and AMPA-FMOC. Quantification of ATZ and [5D] ATZ was carried out in the isocratic mode with 0.1% (v / v) formic acid in acetonitrile / water (70/30) as mobile device phase and with the same column used for GLP analysis. ction, the electrospray ionization source was applied in positive mode. Nitrogen was used as an auxiliary gas throughout the run at 8 l / min at a source temperature of 330 ° C with ion adjustments corresponding to deprotonated and protonated compounds and two daughter ions for quantification and identification, respectively. Data acquisition and analysis were conducted using Agilent Chemstation Rev. 10A.02 software. (Ronco et al., 2016).

2.6. Quality controls and quality assurance

Quality controls during sampling and analysis of the main components involved the use of blank reagents, duplicate samples, and isotopically labeled GLP ([13C, 15N] GLP) and ATZ ([5D] ATZ) to evaluate retention time and recovery for the entire procedure on each sample.

For quality control and assurance in the laboratory analysis of GLP, AMPA and ATZ the linearity, reproducibility, detection and limits of quantification; the matrix effect; and the recovery was tested according to SANCO (2009).

2.7. Data analysis

A descriptive statistical analysis was carried out for both matrices in the entire region. The non-parametric Kruskal-Wallis test (Conover, 1999) and pairwise comparisons were used with GLP, AMPA and ATZ concentrations (in μg / L) in the different sites, after verifying that a normal distribution does not apply. Only single measurements above the limit of detection (LOD) were considered, and the concentrations between the LOD and the limit of quantification (LOQ) were replaced by a value of half the LOQ (Delistraty and Yokel, 2007). To study the association between the temporal pattern (that is, the high and low application campaigns) and the frequency of detection (NLOD versus bLOD), contingency tables (2 × 2) were used for each analyte and an exact test of independence. Fisher's was done. An analysis of the spatial variation of pesticides in rainfall was carried out to include the three different categories in the entire region and the 4 provinces studied along with an analysis involving the accumulated precipitation categories (high zone, HZ, to 1000 mm / year; Middle Zone, MZ, at 900-1000 mm / year; and Low Zone, LZ at less than 900 mm / year). The Spearman correlation coefficient (Conover, 1999) was used for the complete information on the LOD and to evaluate the correlation between analytes in both precipitation (n = 112) and soil matrices (n = 58). The correlation between the matrices was analyzed by grouping the median concentrations of the different sites (n = 7). The MJ site was not considered due to the lack of soil analysis and published data available. All tests were established at a significance level of 0.05 and statistical analyzes were performed using INFOSTAT ™ software. The meteorological information of each precipitation was obtained from the Argentine Ministry of Agroindustry (MINAGRI) and was later correlated with the concentrations of herbicides. Regional climate information, such as wind patterns and accumulated annual rainfall, was also used to analyze the dynamics of the compounds. All maps were built using QGIS v.2.2.0 software.

  1. Results and Discussion

3.1. Analytical parameters

The analytical method used was linear within the range studied (that is, 1-1000 μg / L) for all analytes with an r N 0.993 (critical value = 0.549, 95%, n = 10). The LOD and LOQ for rainwater and soils (in parentheses) with respect to GLP and AMPA 0.5 and 1 µg / L (2 and 5 µg / kg), respectively; while for ATZ the corresponding values ​​were 0.1 and 0.2 μg / L (0.2 and 0.5 μg / kg), respectively. An analysis of the overall recovery for liquid and solid samples, including the electrospray-ionizction-matrich fountain effect, performed on the isotopically labeled standards gave values ​​for rainwater of 93 ± 5% for [13C, 15N] GLP and 90 ± 7% for [5D] ATZ, quantified for all samples (n = 112). For soils, recoveries were 80 ± 10% for [13C, 15N] GLP and 92 ± 5% for [5D] ATZ. Those matrix effects, measured through ionic suppression based on the characteristics of the analyzed samples, were in agreement with Taylor (2005) for this type of analytical methodology. These results are in accordance with the requirements established by the SANCO regulation (2009) for the analysis of pesticide residues.

3.2. Herbicides in rainwater in the pampas region

3.2.1. Spatial patterns

Analysis of the rainfall data from the different regions (n ​​= 112) indicate detection frequencies (NLOD) of 80% for GLP and ATZ (that is, 81.3% and 80.4%, respectively). Previous studies involving different regions in the United States for this environmental matrix have reported similar detection ranges for both compounds, between 61 and 100% for GLP and 69 and 94% for ATZ (Vogel et al., 2008; Chang et al. , 2011; Coupe et al., 2000; Farenhorst and Andronak, 2015). These results reveal the ubiquity of these herbicides in the atmosphere (Majewski et al., 2014).

In the present study, 65% of GLP and 51% of ATZ were detected at concentrations above LOQ. A more detailed analysis revealed that Córdoba was the province with the highest frequency of AMPA detections (42%; Table 2), but that value was notably lower than those of other studies, where AMPA and GLP were detected at similar frequencies with both by above 70% (Battaglin et al., 2014). Fig. 2 represents the spatial distribution of the concentrations of GLP and AMPA in rainwater.

The regional mean and median GLP concentrations were 5.5 ± 11.3 μg / L and 1.29 μg · L, respectively. The maximum GLP concentration (at 67.3 μg / L) was measured at the IT site in the province of Córdoba, where the recorded value was significantly higher than those found in other places (Table 2), in addition to being higher than those reported for weekly samples by Farenhorst and Andronak (2015), at 16.9 μg / L, and by Quaghebeur et al. (2004), at 6.2 μg / L. That in Argentina, the usual doses of GLP are approximately 12 L / Ha / year (CASAFE, 2013) in contrast to the doses of 0.5-2.0 L / Ha / year in the aforementioned The country is also notable, with doses at least 5 times lower.

The AMPA metabolite was detected in 33.9% of all samples, with 34.2% having concentrations higher than LOQ. The mean and median concentrations were 1.5 ± 1.8 μg / L and 0.75 μg / L, respectively. The maximum concentration observed in the present work was 7.91 μg / L, higher than the value reported by Chang et al. (2011) for the States States of 0.97 μg / L. These results of levels and frequencies of AMPA show wind erosion as the main source of these compounds for the atmosphere since the presence of the metabolite is restricted to microbiological degradation in the soil (Grunewald et al., 2001).

For ATZ, the mean and median concentrations in rainwater 0.93 ± 3.36 μg / L and 0.22 μg / L, respectively. The provinces can be organized in the following ascending order based on the median values ​​recorded: Buenos Aires = Santa Fe

The vapor pressure of 0.039 mPa (PPDB, 2017) for ATZ, compared to GLP and AMPA which are considered non-volatile compounds (EU, 2002; USEPA, 1993), indicates a high concentration in the vapor phase of the herbicide ATZ (Pankow, 1994), in whose phase it is swept by rain less efficiently than in its particle phase (Majewskiet al., 2014; Goolsby et al., 1997).

3.2.2. Temporal patterns

Although the volumes of herbicide applied differ depending on the annual cultivation cycles (especially in soybeans, see table 1), non-significant differences were observed between the mean concentrations of GLP, AMPA and ATZ in rainwater, measured in the phases of the campaigns that involve the application of "high" and "low" herbicide according to Marino and Ronco (2005). However, current agricultural practices involve the use of herbicides not only for instant weed control but also for induced fallow, which implies a continuous entry of herbicides throughout the annual cycle (DP, 2015). The resulting regularity of the spraying produces a continuous movement in the atmosphere through the primary drift, in addition to supplying the soil with the herbicide to a point where pseudo-persistence of GLP was observed in different soils of Argentina (Primost et al., 2017 ; Soracco et al., 2018). As discussed in the next section and alluded to in the Introduction, soils constitute another source of atmospheric herbicides through wind erosion: in fact, even during a high application season, transport to the atmosphere via wind erosion can contribute 20-40% of atmospheric herbicides; while in weeks without any application, this contribution reaches as high as 50-100% (Chang et al., 2011).

3.3. Soil as a source of herbicides in the atmosphere

The global frequencies of herbicide detection in soils obtained in this work were 41% for GLP, 22% for AMPA and 32% for ATZ. These results are of particular relevance since soils act as an emission surface, both for airborne parental compounds (Tao et al., 2008) loaded with particles and metabolites such as AMPA (Bento et al., 2017). The GLP and AMPA contained within the upper centimeters of the soil are susceptible to wind erosion and subsequent atmospheric transport. Silva et al. (2018) has estimated the elimination of GLP and AMPA through the action of wind erosion that will be around 1900 mg / ha / year for soils with concentrations of GLP and AMPA below 0.5 mg / kg and up to 3000 mg / ha / year for soils with higherisconcentrations. The levels of GLP and AMPA detected in soils of the provinces of Buenos Aires, Córdoba and Santa Fe (Fig. 1), were not significantly different, having an average concentration of 125 ± 87 μg / kg (together with a maximum of 323 μg 7 kg). These GLP concentrations are within the range previously observed for the southeastern region of Buenos Aires by Aparicio et al. (2013) at 35-1502 μg / kg, but they are one eighteenth of that of the soils of the province of Entre Ríos, as reported by Primost et al. (2017), at 2299 ± 476 μg / kg. Estas diferencias entre los los niveles cuantificados podrían estar relacionados con el sesgo inherente al diseño del muestreo dentro del marco de los objetivos propuestos. Tanto en el presente estudio y en Aparicio et al. (2013) los suelos muestreados estaban asociados con diferentes tipos de cultivos, mientras que Primost et al. (2017) estudió exclusivamente campos de soja, cuyas plantas en particular tienen un mayor requerimiento de GLP.

La concentración total media de ATZ en el suelo fue 13 ± 17 μg 7 kg, junto con un máximo de 66 μg / kg en Córdoba. No se observaron diferencias entre la concentración de ATZ en los suelos de Córdoba y los suelos de Buenos Aires; ambas provincias concentran también la producción de maíz que tiene un mayor requerimiento de ATZ (MINAGRI, 2017). Sin embargo, no hubo datos disponibles de Entre Ríos y solo se detectó un único valor positivo entre los diferentes muestras de suelo de Santa Fe.

3.4. Relaciones entre las concentraciones de herbicidas en el suelo y el agua de lluvia

En agua de lluvia, se observó una correlación significativamente positiva (r = 0.66) entre los valores para GLP y AMPA. Además, esta correspondencia también se observó en el suelo (r = 0,88), como se ha informado para otros matrices ambientales por Primost et al. (2017). Dado que los suelos son las únicos fuente de AMPA para la lixiviación a la atmósfera (Majewski et al., 2014), la correlación entre la concentración de este metabolito y los niveles de GLP en la precipitación son un indicador del papel clave del suelo como una fuente de emisión además de generada en la deriva primaria (Chang et al., 2011). Sin embargo, correlaciones positivas significativas entre esos dos compuestos y ATZ en lluvia (r = 0.46 yr = 0.47 respectivamente) fueron observados, evidenciando el uso combinado de los dos herbicidas en los protocolos agrícolas (Ghida Daza y Urquiza, 2014). Una comparación entre las concentraciones de todos los herbicidas encontrados en el agua de lluvia versus los niveles correspondientes en los suelos locales de los diferentes sitios de estudio, en particular, no revelaron correlaciones significativas entre los compuestos medidos en esos dos entornos de matrices. En vista de este hallazgo, podríamos inferir que los suelos son una fuente de estos compuestos para la atmósfera; pero esa falta de correlación impide la definición de una huella digital atmosférica local, con erosión eólica y volatilización, además de tener un papel importante en el dinámica de esos herbicidas. Por otra parte, la mayor adsorción de GLP y AMPA a las partículas más finas de los suelos (<10 μm) aumenta el transporte aéreo fuera de sitio (Bento et al., 2017).

3.5. Herbicidas en lluvia como consecuencia de factores climáticos

Explorando las variables climáticas registradas para cada precipitación – volumen de precipitación, temperatura máxima y velocidad del viento-, no registramos correlaciones entre cada variable y las concentraciones de GLP, AMPA y ATZ. Además, no hay una asociación significativa entre las concentraciones de los tres herbicidas, según lo registrado por Waite et al. (2005). Debido a la presión de vapor insignificante de ambos compuestos, mencionado por Majewski et al. (2014), en consecuencia, esperábamos que la detección de GLP y AMPA estaría relacionada principalmente con la dinámica de recarga atmosférica de material particulado y por lo tanto a la frecuencia de las precipitaciones en lugar de las condiciones climáticas en el momento de muestreo. Sin embargo, al evaluar la precipitación anual acumulada por siobarras, una asociación entre las concentraciones medias de GLP y AMPA y volúmenes de precipitación anual se observó.

La Fig. 2 ilustra los sitios de monitoreo, subdivididos según los parámetros de precipitación, derivados de las isobarras de precipitación anual acumuladas (AAPI) en particular, las concentraciones medias de GLP en la lluvia cuando fueron agrupados por isobarras exhibieron diferencias significativas (p ≤ 0.05) entre las tres zonas. LZ (zonas de baja lluvia), correspondiente a la AAPI más baja, se caracterizó por concentraciones de GLP significativamente más altas que las registradas las otras zonas, mientras que concentraciones más bajas que las de LZ se detectaron en HZ (zonas de alta lluvia) a pesar de todo un AAPI más alto. Estos resultados están de acuerdo con Messing et al. (2013) y Hill et al. (2002), donde las concentraciones más altas de herbicidas se detectaron en sitios con bajas frecuencias de lluvia (considerando solo la deposición húmeda). AMPA mostró un similar comportamiento con la zona alta que tiene concentraciones medianas significativamente más bajas que los otros dos, mientras que no se encontraron diferencias significativas entre las concentraciones registradas en LZ y MZ.

En cuanto a ATZ, las concentraciones medias evidenciaron un patrón similar, aumentando hacia el suroeste; pero se observaron diferencias significativas solo entre los sitios de las zonas media y alta.

3.6. El papel de la lluvia en la deposición masiva de GLP en el ambiente superficial

En vista de la deposición de lluvia y los cálculos realizados por Coupe et al. (2000) y Vogel et al. (2008), estimamos la contribución de estos herbicidas a nivel superficial de los ambientes en la isobarras de MZ (ver Fig. 2) como el escenario más desfavorable. La precipitación anual para esa zona se estimó en 950 mm, correspondiente a valor medio de las isobarras limitadoras y, por lo tanto, se consideró uniforme sobre toda la superficie. Si entonces las concentraciones regionales medianas (a 1.24 μg / L) se tienen en cuenta, la masa anual de GLP depositada ascendería a unos 11.780 mg/ha/ año. Para evaluar estos resultados, tomando como propuestos por Silva et al. (2018), la aportación anual de la erosión eólica proporciona a la atmósfera unos 1940 mg /ha / año de suelos con concentraciones de GLP por debajo de 0.5 mg / kg. Entonces, desde esas consideraciones, la deposición anual estimada por lluvia indica la extensión de otras fuentes del herbicida para el aire.

Aunque se espera que la escorrentía sea la principal fuente de estos herbicidas para contaminar cuerpos de agua (Messing et al., 2011; Sasal et al., 2015), bajo ciertas condiciones específicas, pòr ejemplo, una tormenta torrencial: la deposición húmeda de pesticidas podría exceder su contribución por escorrentía (Donald et al., 2005). De acuerdo con los resultados encontrados en nuestro estudio, la lluvia definitivamente debe considerarse una fuente relevante de estos contaminantes para entornos de nivel superficie. Como se informó anteriormente por Majewski et al. (2000) y Nations y Hallberg (1992), estos resultados refuerzan la noción que los herbicidas son aeroptrasportados hacia comunidades urbanas y periurbanas, agregando así una posible vía de exposición para humanos y animales en la región de las pampas, como fue citado por Bento et al. (2017) y Battaglin et al. (2014) para otros países. En vista de tales implicaciones, proponemos una actualización de las directrices argentinas para controlar la calidad ambiental incorporando debidamente los criterios de herbicida para el aire ambiente.


Los resultados de este estudio de herbicidas en agua de lluvia, el primero en Argentina handemostrado la alta frecuencia de detección (80%) de GLP y ATZ junto con la ubicuidad de esos compuestos en la atmósfera asociada con las precipitaciones anuales. El máximo de las concentraciones de ambos herbicidas fueron más altas que las detectadas en otros países, posiblemente como consecuencia de las dosis agronómicas más altas utilizadas en Argentina. GLP, AMPA y ATZ se detectaron en suelos, con niveles mayores de concentración de GLP asociados con cultivos de soja que con otros cultivos. Por lo tanto, aunque esta matriz constituye una importante fuente, no se asoció con las concentraciones atmosféricas observadas en la escala local. Una variabilidad espacial de la concentración de plaguicidas se observó entre la precipitación acumulada por isobarras, esto fue más evidente para GLP y AMPA que para ATZ. Por lo tanto, la recarga atmosférica de material particulado determinó la concentración de ambos compuestos en la lluvia. Porque la deposición atmosférica de herbicidas a través de la lluvia en cuerpos de agua superficiales y suelos así como en los sitios urbanos de la región podría constituir una fuente de exposición de la población a estos contaminantes del aire, es necesario incluir esos compuestos en las directrices de calidad del aire y en los programas de monitoreo. Tras la necesaria consideración adicional de la más amplia gama de sustancias activas utilizadas en las prácticas agrícolas actuales en todo el país, también sugerimos futuras investigaciones que involucren la inclusión en los análisis de otros pesticidas que además se sabe que se dispersan en regiones fuera del área de aplicación.

Video: You Wont Believe How Roundup, or Glyphosate, Affects You! (October 2020).