Patterns and Relationships of Pesticide Use in Agricultural Crops of Latin America: Review and Analysis of Statistical Data

Abstract

The excessive use of pesticides in agriculture causes problems related to public health; biodiversity; the contamination of water bodies, soil and air; and general environmental degradation, including interactions with climate change effects. The aim of this work was to describe the patterns of pesticide use in 17 Latin American countries and their associations with the total harvested area and the harvested areas of the five main crops on the basis of statistics reported by the Food and Agriculture Organization from 1990 to 2021. Principal component analysis (PCA) revealed four different patterns among countries according to the magnitude of pesticide application: Brazil (G1) > Argentina (G2) > Colombia and Mexico (G3) > Central and South America (G4). Together, more than 1.2 million tons of active ingredients is applied annually, carrying harmful effects and risks. In the joint analysis of both datasets (applied pesticides and harvested area), different correlations were observed between the amount of pesticides applied and the harvested area; for example, in G1 and G2, positive and significant correlations were detected between the applied amounts of total pesticides, herbicides, insecticides and bactericides and the total area harvested by the main crop, but in G3, there was a negative correlation with the five main crops.

1. Introduction

The objectives of sustainable development include decoupling economic growth from environmental degradation; increasing the use efficiency of natural resources; and promoting sustainable and healthy lifestyles, consumption and production [1]. However, in the last 20 years, pesticide use has doubled, and pesticides now have a market value of USD 130.7 billion. In particular, the European Union (where four companies account for >70% of pesticide production) is the main exporter of pesticides to countries of the Global South; these pesticides have been associated with increases in health problems (385 million cases recorded), decreased biodiversity, the contamination of water bodies and soils and general environmental degradation [2]. In 2021, 3.5 million tons of pesticides was used worldwide, 35% of which (2.1 million tons) was applied in Latin America and the Caribbean; pesticide application amounts have rapidly increased over the past three decades [3,4]. The release of these volumes of pesticides into the environment unquestionably affects the health of agricultural workers/farmers and their families, as well as the ecosystems near or surrounding agricultural plots where pesticides are applied by contaminating water bodies and air and remaining in the soil.

In Latin America, enormous amounts of pesticides are being applied annually in cultivation plots, with variations in the frequency or intensity of application of each product or its active ingredients. Thus, the application of pesticides is a function of market dynamics (e.g., the promotion of new pesticides and the expiration and restriction of certain active ingredients or products); the regulatory laws of each country; technological innovations in production systems (e.g., the introduction of transgenic varieties, the transition to organic or agroecological agriculture and the use of drones to apply pesticides); and social, territorial and agroecological factors inherent to the production region and/or type of production system, such as traditional, subsistence, semi-intensive, intensive, or commercial agriculture [5,6,7,8].

In the ecological and geographic context of Latin America, changes in the amount and frequency of rainfall, temperature fluctuations, droughts, and changes in wind currents and factors related to climate change affect not only the efficiency of pesticides but also both the incidence and resistance of crop pests and diseases. Accordingly, the occurrence and application of pesticides inside and outside production areas change with increasing dose. Related influencing factors include the evaporation of pesticide-laden water, wind currents and the loss or infiltration of water from cultivation plots regularly amended with agrochemicals at levels that are harmful to health [9,10]. Consequently, climate change, in addition to altering the efficiency of pesticides, affects the application, distribution and human toxicity or health effects of chemical pollutants [11]. Moreover, these effects can lead to low pesticide efficiency or biotic resistance, meaning that more pesticides must be applied per cultivated area.

Increases and decreases in the area used for the cultivation of main crops affect the amount and frequency of pesticide use. For example, the area cultivated with soybeans in Brazil and Argentina, with an average production of more than 176 million tons (more than 50% of international exports), has increased since 2000 [12]. Recently, Uruguay and Paraguay have adopted similar dynamics of soybean cultivation [13,14]. In Ecuador, the cultivated area and pesticide use used for the cultivation of bananas for export are increasing [15]. Similarly, the cultivation of coffee for export in Latin America has transformed local and regional dynamics, with or without the use of pesticides [16]. Schreinemachers and Tipraqsa [17] determined a direct relationship between the intensification of agricultural land use and increases in the use of pesticides per hectare, but the increases in the use of insecticides, herbicides, or fungicides vary according to national income (low, medium, or high). For example, the pattern of increase was more evident in countries such as Brazil, Mexico, Uruguay, Cameroon, Malaysia and Thailand.

The constant use of pesticides after the production cycle leads to both accumulation and rapid dissemination in agriculture, various ecosystems, the environment and water bodies and influences biodiversity. Moreover, this type of pesticide used directly and/or indirectly affects the health of the rural population through contact with or the handling of pesticides and the health of consumers in urban areas through the ingestion of pesticide-contaminated food or water. Health damage has been reported in Latin America; for example, in a community of farmers in Mexico, pesticide residues, including up to 17 active ingredients, were detected in the urine of children and adolescents [18]. Avila-Vazquez et al. [19] associated the application of pesticides (e.g., glyphosate) in farming communities in Argentina with two- to threefold increases in abortions and congenital abnormalities compared with communities that did not use pesticides. A similar pattern was observed in Brazil, with reports of low birth weight and mortality from congenital diseases at birth [20]. Thus, damage to human health and multiple effects on ecosystems have been continuously reported. In this context, the objective of this study was to describe the patterns of pesticide use in 17 Latin American countries and their associations with the areas harvested with the five main crops on the basis of statistics reported by the Food and Agriculture Organization (FAO) from 1990 to 2021 [4]. The ultimate goals were to provide information for assessing impacts on human health and ecosystems, to guide actions to monitor or document damage and to promote agroecological changes in food production.

2. Materials and Methods

2.1. Databases of Agricultural Pesticides Used in Mexico and Central and South America

The data analyzed in this work were retrieved from the FAOSTAT information databases comprising annual reports from 1990 to 2021 for 17 countries: Argentina, Brazil, Bolivia, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Peru, Paraguay, Uruguay and Venezuela [4]. These countries account for 66.9 to 87.3% of the total pesticides applied for agricultural purposes in Mexico and Central and South America, which we designate Latin America for geographic descriptive purposes. The analysis did not include the United States, Canada, or countries with insufficient annual reports for more than five years or with data estimated for 10 years or more that were reported by the institutional managers of the databases. The analysis considered the annual total amount of agricultural pesticides used and their breakdown, in terms of herbicides, insecticides, fungicides and bactericides, in tons of active ingredients. In addition to the pesticide data, the total area harvested annually and the areas corresponding to the top five crops in each country were considered, whereas the area sown was omitted to avoid overestimations of pesticide use since plots could have been lost for several reasons.

2.2. Statistical Analysis

The matrix of analyzed data included information on the amounts of pesticides used, the total harvested area and the areas corresponding to the five main crops in 17 Latin American countries over 32 years, from 1990 to 2021. The growth and pesticide use patterns by country were analyzed via principal component analysis (PCA) following the procedure proposed by Johnson and Wichern [21] and using the values of pesticide per area (total pesticides/total area harvested), total pesticides, herbicides, insecticides and fungicides plus bactericides applied per country and per year. Briefly, the procedure was as follows. (a) The analysis was based on the variances–covariances matrix (∑), and the general form is the following:

$$\sum =\left[\begin{array}{c}\begin{array}{ccc}{\sigma }_1^2& \rho {\sigma }_{12}^2\dots & \rho {\sigma }_{1p}^2\\ \rho {\sigma }_{22}^2& {\sigma }_2^2\dots & \rho {\sigma }_{2p}^2\\ ⋮& ⋮& ⋮\end{array}\\ \begin{array}{ccc}\rho {\sigma }_{p2}^2& \rho {\sigma }_{p2}^2\dots & {\sigma }_p^2\end{array}\end{array}\right]$$

(b) First, each variable was standardized to build a new matrix (Z_n×p) where n = 32 year × 17 countries, and p = variables (pesticides use), using the expression $Z={\displaystyle \frac{X-\overline{X}}{\sigma }}$. Then, (c) the analysis was performed using the ∑ matrix. Later, with a second database, another PCA was performed using the values of the harvested area by the five main crops per country, from 1990 to 2021, where a Z_nxp matrix was integrated using the standardized values of each variable, and the analysis was also performed using the ∑ matrix. In both cases, the main principal components that captured 95% or more of the explained total variance were retained.

Complementarily, the matrices corresponding to the non-standardized values of pesticide use and harvested area were combined, and Pearson’s correlation analysis was performed to evaluate the degree of correlation between the amounts of pesticides applied and the harvested area, as well as indicators of increasing or decreasing relationships between pesticide use and harvested area. Correlation analyses were grouped into four analysis groups and significance tests named G1, G2, G3 and G4, using values of 30, 30, 62 and 414 degrees of freedom (df = n − 2), respectively, and Student’s t tests (p ≤ 0.05, 0.01) to assess significance. All statistical analyses were performed using the SAS statistical package (version 9.0, SAS Institute Inc., Cary, NC, USA) [22].

3. Results

Pesticide application in 17 selected countries in Latin America has had different growth rates in different decades since 1990. For example, in 1990, 163 thousand tons was applied, followed by a growth of 147% until 2000. This increasing trend persisted: pesticide use increased by 110% in 2010 compared with that in the previous decade, and from 2010 to 2020, the increase was 37.5%, with its application increasing to a million tons. In all countries, the use of herbicides has continued to grow in comparison with the use of insecticides, fungicides, and bactericides, with a quantity of approximately 764 thousand tons (Figure 1). These figures reveal the high dissemination of pesticides to crops, the environment, water bodies and populated areas and suggest that several of the active ingredients are likely accumulating or dispersing.

The principal component analysis (PCA) of the behavior of countries over the studied years of pesticide application revealed distinct patterns (Figure 2). Eigenvectors for PC1 and PC2 were 0.001, 0.822, 0.540, 0.129 and 0.125 and 0.012, 0.298, −0.273, 0.365 and 0.570, corresponding to pesticide use per area, total, herbicides, insecticides and fungicides plus bactericides, respectively, resulting in 99.8% of the total variance explained with both components (more information in Supplementary Materials Table S1). Johnson and Wichern (p. 444) [21] stated that eigenvectors’ values indicate an equivalent and direct positive or negative relationship between variables used in the analysis performed and principal components. Therefore, PC1 was associated with total pesticides > herbicides > insecticides > fungicides and bactericides > pesticide use per area and PC2 with fungicides and bactericides > insecticides > total pesticides > herbicides (negatively) > pesticide use per area, both in descending order of correlation, using their eigenvector values as a reference. So, close to the PC1 and PC2 axes of Figure 2, just the main variables with more influence or association with each component were annotated. For example, the quantity and intensity of pesticide application in Brazil (G1) were unique, with values higher than 700 thousand tons in recent years (2020 and 2021), which was 50% greater than the amount applied a decade ago. A similar pattern was observed for Argentina (G2), with a growth in pesticide application of 100 tons over a decade; for example, in 2021, more than 241 thousand tons was applied. A third pattern was observed for Mexico and Colombia (G3), where more than 78 thousand tons was jointly applied, representing a slightly lower value than that in 2010, when more than 100 tons was applied annually. A fourth pattern was observed for countries in Central America (Guatemala, Costa Rica, El Salvador, Honduras and Nicaragua) and South America (Panama, Chile, Uruguay, Paraguay, Bolivia, Peru and Ecuador) (G4), where each country contributed significantly lower amounts than those in groups G1–G3, but the joint impact was on the order of 163 thousand tons per year; in particular, Bolivia, Ecuador, Paraguay and Uruguay had individual application amounts greater than 16 thousand tons (Figure 2).

With respect to the estimated average area harvested every five or six years, in the last decade, Brazil (G1) had harvest areas ranging from 72.4 to 80.9 million ha. These values were twice the values determined for Argentina (G2, 34.4 to 36.7 million), which, in turn, were approximately three times greater than the average harvested areas of Colombia and Mexico (G3); 13 countries in Central and South America (G4) accounted for a harvest area of approximately 2 million ha. The harvested area showed a proportional relationship with the total application amount of pesticides; for example, regarding application in the last decade, the identified groups followed the order G1 > G2 > G3 > G4, with values of 464 to 600 > 212 to 209 > 50.6 to 44.9 > 11.2 to 12.4 thousand tons of active ingredients. A similar trend was observed for the total application amount of herbicides: 361.3 > 197.2 > 18.5 > 6.2 thousand tons (

Table 1. Quinquennial and sexennial averages of harvested area and amounts of total pesticides, insecticides, herbicides, fungicides and bactericides applied from 1990 to 2021 in Latin American countries grouped by use patterns.

Quinquennial or Sexennial Averages	Brazil (G1)	Argentina (G2)	Colombia and Mexico (G3)	Countries of Central and South America 1 (G4)
Average harvested area (×100,000 ha)
1990–1995	510.8	190.9	102.6	12.7
1996–2000	488.0	239.4	102.8	13.9
2001–2005	581.9	268.5	100.7	15.4
2006–2010	644.2	301.5	99.3	17.8
2011–2015	724.9	344.8	100.5	20.0
2016–2021	809.1	367.3	102.2	20.1
Average total pesticides applied (×1000 tons)
1990–1995	71.9	28.8	21.8	3.6
1996–2000	120.4	70.4	37.0	5.7
2001–2005	186.2	116.9	59.1	7.0
2006–2010	311.5	183.1	57.1	9.4
2011–2015	464.1	212.1	50.6	11.2
2016–2021	600.8	209.4	44.9	12.4
Average herbicides applied (×1000 tons)
1990–1995	35.5	19.1	6.7	1.2
1996–2000	64.8	53.8	12.3	2.0
2001–2005	108.7	103.5	26.6	3.3
2006–2010	185.5	169.2	22.3	4.9
2011–2015	292.7	197.0	17.5	6.0
2016–2021	361.3	197.2	18.5	6.2
Average insecticides applied (×1000 tons)
1990–1995	21.3	4.2	4.5	0.9
1996–2000	27.4	9.1	6.9	1.3
2001–2005	36.0	7.6	10.8	1.4
2006–2010	61.4	8.2	8.7	1.8
2011–2015	92.1	8.1	9.0	1.7
2016–2021	99.2	4.5	8.0	1.6
Average fungicides and bactericides applied (×1000 tons)
1990–1995	11.7	5.2	10.4	1.1
1996–2000	18.4	6.3	17.8	1.4
2001–2005	21.6	3.4	21.6	1.7
2006–2010	38.6	2.5	26.1	2.3
2011–2015	68.8	3.9	23.9	2.8
2016–2021	123.3	4.0	18.4	3.7

¹ Including Chile, Costa Rica, Ecuador, Bolivia, Guatemala, El Salvador, Honduras, Nicaragua, Panama, Paraguay, Peru, Uruguay and Venezuela.

). Thus, the magnitude of cultivated and harvested areas was related to the amount of herbicides, insecticides, fungicides and bactericides applied in each country.

The predominant crops in the 17 Latin American countries determined the types of pesticides used. For example, in Brazil (G1), soybeans, corn, sugarcane, beans and rice were the primary crops; furthermore, some transgenic varieties of soybeans and corn were used, which are resistant to the application of herbicides such as glyphosate (isopropylamine salt of glyphosate with N-(phosphonomethyl) glycine). These two crops showed an annual cultivation area of approximately 58 million hectares, corresponding to the application of 407 thousand tons of herbicides. Brazil was followed by Argentina (G2) in terms of herbicide application (228 thousand tons), with soybean, wheat, corn, sunflower, sorghum and cotton as the target crops (38 million hectares). In contrast to these countries, corn, coffee, beans, sorghum, sugarcane, wheat, rice and banana were predominant in Colombia and Mexico (G3), with cultivation areas on the order of 20 million hectares and herbicide application amounts of 28 thousand tons. The countries with low pesticide application amounts but a high diversity of main crops (corn, soybeans, wheat, rice, coffee, sorghum, sugarcane, banana, potato, etc.) were Chile, Costa Rica, Ecuador, Bolivia, Guatemala, El Salvador, Honduras Nicaragua, Panama, Paraguay, Peru, Uruguay and Venezuela, with approximately 26.7 million hectares of cultivated area and 82 million tons of applied herbicides. These data showed that the amount of herbicides applied relative to the total amount of pesticides applied (

Table 2. Quinquennial and sexennial averages of harvested area and amounts of total pesticides, insecticides, herbicides, fungicides and bactericides applied from 1990 to 2021 in Latin American countries grouped by use patterns.

Applied Pesticides and Harvested Area	Brazil (G1)	Argentina (G2)	Colombia and Mexico (G3)	Countries of Central and South America 1 (G4)
Total applied pesticides (ton)	719,507.4	241,520.0	82,902.1	162,958.5
Herbicides (ton)	407,462.7	228,429.2	28,053.0	82,477.0
Insecticides (ton)	122,182.4	6076.9	15,297.4	18,921.4
Fungicides and bactericides (ton)	168,169.4	3756.5	39,089.8	48,465.3
Total harvested area (ha)	86,549,711.0	38,099,766.0	20,557,313.0	26,694,095.0
First main crop (ha)	39,168,068.0	16,466,714.0	7,979,733.0	9,675,569.0
Second main crop (ha)	19,024,538.0	8,146,596.0	2,214,137.0	4,133,578.0
Third main crop (ha)	9,970,958.0	6,394,102.0	1,796,902.0	2,890,867.0
Fourth main crop (ha)	2,750,264.0	1,666,843.0	1,238,910.0	1,798,826.0
Fifth main crop (ha)	2,613,086.0	1,006,503.0	1,044,366.0	1,369,553.0

¹ Chile, Costa Rica, Ecuador, Bolivia, Guatemala, El Salvador, Honduras Nicaragua, Panama, Paraguay, Peru, Uruguay and Venezuela.

) was on the orders of 56.6, 94.6, 33.8 and 50.6% for G1, G2, G3 and G4 (total herbicides/total pesticides), respectively, indicating that herbicides were the greatest source of agricultural pesticide contamination of the environment, water table, surface waters, air and population centers.

The PCA, which was based on the harvested areas of the five main crops (Figure 3), revealed behavior patterns that were similar or homologous to the patterns associated with pesticide use (Figure 2). In this case, the eigenvectors for PC1 were 0.851, 0.437, 0.255, 0.111 and 0.085, and those for PC2 were −0.502, 0.725, 0.201, 0.326 and 0.275, corresponding to harvesting from the first to fifth main crops, respectively, accounting for 99.5% of the total variance explained by both components (more information in Supplementary Materials Table S2). In descending order of association between the original variables and principal components obtained, PC1 was associated with first > second > third > fourth > fifth main crop and PC2 with second > first (negatively) > fourth > five > third main crop, based on their eigenvector values [21]. So, as part of Figure 3, close to PC1 and PC2, the variables associated positively with each component were annotated. That is, Brazil had the largest area harvested with soybeans and corn, the two main crops, corresponding to the highest amount of pesticide use. Argentina presented a similar pattern for the same crops. Notably, Brazil and Argentina were large exporters of corn. In contrast, Colombia and Mexico differed from Brazil, Argentina and different countries in Central and South America, representing an intermediate phase between the countries with the largest and smallest cultivated and harvested areas. With respect to the diversity of cultivated species, the countries with the smallest surface areas had less pesticide use and vice versa. Thus, the order of crop diversity across countries was Central and South America > Colombia and Mexico > Argentina > Brazil, indicating that the latter country, with larger cultivated and harvested areas and monocultures, also tended to use greater amounts of herbicides (Figure 3).

The significance of the correlations in each group (G1–G4) were assessed with the same Student’s t test; for example, all correlation values within the G1 and G2 groups were valued with the same Student’s t test because they presented the same degrees of freedom (df = 30) but not for the G3 and G4 groups which had 62 and 414 df, respectively (

Table 3. The significance and magnitude of Pearson’s correlations between the amount of pesticides applied and the total area harvested and the areas of the first five main crops in seventeen Latin American countries.

Total Harvested Area and Areas of the Five Main Crops	Brazil (G1, n = 32)	Argentina (G2, n = 32)	Colombia and Mexico (G3, n = 64)	Countries of Central and South America 1 (G4, n = 416)
Total pesticides applied
Total harvested area	0.98 **	0.96 **	−0.36 **	0.51 **
First main crop	0.98 **	0.96 **	−0.36 **	0.49 **
Second main crop	0.95 **	0.22 ns	−0.41 **	0.54 **
Third main crop	0.94 **	0.57 **	−0.34 *	0.52 **
Fourth main crop	−0.89 **	−0.55 **	−0.32 *	0.40 **
Fifth main crop	−0.84 **	0.47 *	−0.27 *	0.41 **
Herbicides applied
Total harvested area	0.97 **	0.96 **	−0.63 **	0.56 **
First main crop	0.97 **	0.96 **	−0.63 **	0.64 **
Second main crop	0.95 **	0.23 ns	−0.64 **	0.61 **
Third main crop	0.95 **	0.58 **	−0.58 **	0.52 **
Fourth main crop	−0.89 **	−0.57 **	−0.57 **	0.35 **
Fifth main crop	−0.86 **	0.46 *	−0.56 **	0.36 **
Insecticides applied
Total harvested area	0.96 **	0.06 ns	−0.08 ns	0.43 **
First main crop	0.95 **	0.15 ns	−0.09 ns	0.37 **
Second main crop	0.92 **	−0.16 ns	−0.13 ns	0.45 **
Third main crop	0.96 **	−0.27 ns	−0.09 ns	0.49 **
Fourth main crop	−0.86 **	0.33 ns	−0.07 ns	0.34 **
Fifth main crop	−0.82 **	−0.02 ns	−0.05 ns	0.43 **
Fungicides and bactericides applied
Total harvested area	0.91 **	−0.45 *	0.05 ns	0.19 **
First main crop	0.93 **	−0.63 **	0.04 ns	0.10 ns
Second main crop	0.91 **	0.02 ns	−0.03 ns	0.17 **
Third main crop	0.84 **	0.03 ns	0.01 ns	0.24 **
Fourth main crop	−0.84 **	0.48 *	0.05 ns	0.22 **
Fifth main crop	−0.75 **	0.17 ns	0.14 ns	0.21 **

¹ Including Chile, Costa Rica, Ecuador, Bolivia, Guatemala, El Salvador, Honduras Nicaragua, Panama, Paraguay, Peru, Uruguay and Venezuela; ^ns not significant (p > 0.05); * significant at p ≤ 0.05; ** significant at p ≤ 0.01, where the significance of the correlations is based on Student’s t test with 30, 30, 62 and 414 degrees of freedom for the G1, G2, G3 and G4 groups, respectively.

). Therefore, Pearson’s correlation analysis (for each group: G1, G2, G3 and G4) between the amount of pesticides used and the total harvested area and areas of the five main crops revealed that the total amount of pesticides was positively and significantly correlated with the total harvested area and the areas of the first to third main crops in Brazil and the countries of Central and South America with less cultivated area. However, a negative and significant correlation was observed with the total area harvested and the areas of the five main crops in Colombia and Mexico, indicating that the cultivated area decreased or was maintained; however, if the use of pesticides increased, an inverse relationship was observed. These patterns were also observed for the total amount of herbicides, insecticides, fungicides and bactericides applied. In Brazil, significant and negative correlations were detected between the harvested areas of the fourth and fifth main crops and the application amounts of total pesticides, herbicides, insecticides, fungicides and bactericides; a similar pattern was observed between the harvested area of the fourth main crop and the application amounts of total pesticides and herbicides in Argentina (Table 3 shows all analyses for each group, G1, G2, G3 and G4).

4. Discussion

The use of agricultural pesticides is commonly associated with the prevention, management and control of crop pests. The pesticide market is highly lucrative (USD 43.2 billion), and the main exporting companies are based in Europe (3.54 million tons of active ingredients). Despite initiatives such as ‘Save Bees and Farmers’, which was signed by 1.2 million people living in Europe and proposed an 80% reduction in pesticide use by 2030 and total elimination by 2035, progress has been scarce [2]. From 1990 to 2021, the magnitude of pesticide use increased by 206% in Oceania, 191% in America, 175% in Africa, 67% in Asia and approximately 1% in Europe. These figures indicate that America is the continent with the second highest amount of pesticides dispersed in cultivated areas, with 26% growth in the last decade and with estimated application amounts of 3.01 kg/ha or 1.23 kg per capita of active ingredients [4].

Brazil, Argentina, Colombia, Mexico, Chile, Costa Rica, Ecuador, Bolivia, Guatemala, El Salvador, Honduras, Nicaragua, Panama, Paraguay, Peru, Uruguay and Venezuela are the focus of this work and are referred to as middle- to low-income Latin American countries; these countries account for 66.9 to 87.3% of the total pesticides (herbicides, insecticides, fungicides and bactericides) that are applied in America (1772 million tons in 2021). In 2021, 457,385 and 92,960 tons of active ingredients were applied in the United States and Canada, respectively, with the greatest application to corn, soybean and wheat crops [4]. From 1990 to 2021, the application of pesticides in Latin America experienced exponential growth from 0.163 to 1206 million tons (Figure 1). This increase was due in part to the continued expansion of the green revolution in the 1990s in Latin America, where pesticide use was implemented to solve problems such as weeds, pests and/or diseases in wheat and corn crops. Schreinemachers and Tipraqsa [17] noted that in middle-income countries (e.g., Brazil, Mexico, Argentina and Uruguay), the intensity of pesticide use is consistently increasing compared with that in low-income countries, which are characterized by recurring economic crises and consequently do not increase their use of pesticides.

In the last decade (2020s), the application of pesticides has continued to grow, with magnitudes greater than one million tons per year and with a relatively high proportion of herbicides (61.8%). Given the lack of evidence of growth in cultivated and harvested areas, it can be inferred that the intensity of pesticide application per cultivated hectare has increased. This scale of pesticide application involves several environmental pathways and the deterioration of natural resources. For example, Ryberg and Gilliom [23] sampled the main rivers of the United States and detected high concentrations of pesticides with the greatest agricultural use, such as cyanazine, alachlor, atrazine, deethylatrazine, metolachlor and carbofuran, as well as pesticides with low agricultural use or nonagricultural use, such as simazine, chlorpyrifos, malthion, diazinon and carbaryl. These findings implied that the tributaries of the rivers transported pesticides from cultivated areas via groundwater infiltration and indicated that different soil removal practices, agricultural crop management practices and changes in the types of pesticides and active ingredients influenced pesticide transport. Additionally, these results suggested that production increased but carried environmental effects and impacts on human health [24,25].

In terms of the magnitude of agricultural pesticide use among the 17 studied Latin American countries, different response patterns were observed (Figure 2, PCA). For example, from 1990 to 2021, Brazil (G1) was the country with the highest application amounts of total pesticides, herbicides, insecticides, fungicides and bactericides, followed by Argentina (G2), Colombia and Mexico (G3) and 13 countries in Central and South America (G4). These different patterns are due in part to the cultivated area; 80 to 88 million hectares is cultivated per year in Brazil, more than 36 million is cultivated in Argentina, approximately 50 million is cultivated in Colombia and Mexico and more than 10 million is cultivated in 13 countries of Central and South America (Table 1). Thus, Brazil and Argentina consume more than 600 and 209 thousand tons of pesticides, respectively, but this pesticide use also has strong impacts on human health, natural resources (e.g., contamination of aquifers) and the environment. The main documented types of damage to health as a consequence of prolonged or chronic exposure to pesticides in Brazil include damage to the central nervous system, cancer, deleterious or reproductive effects on agricultural workers, poisoning, malformation and endocrine changes [26].

In Argentina, exposure to pesticides in not only agricultural workers but also people in nearby population centers or areas near fields has led to permanent risks of certain diseases and impacts on aquifers where water is obtained for human consumption. For example, Filippi et al. [27] monitored urinary metabolites in a population sample from Córdoba, Argentina, a region of high agricultural activity and pesticide application, and found evidence of pirimiphos, parathion and chlorpyrifos. Mas et al. [28] sampled water collected from rain, shallow wells and dams in Santiago del Estero, Argentina, and reported high concentrations of glyphosate and aminomethylphosphonic acid above the standard levels established by the European Union. Verzeñassi et al. [29] reported that eight population centers in Argentina close to agricultural crops with high application rates of pesticides presented higher frequencies of cancer and death than did those in all of Argentina. The authors mentioned that frequent exposure to pesticides, active ingredients and pesticide formulations increases the risk of cancerous tumors even though they did not identify direct relationships with specific pesticides.

In Colombia and Mexico, from 2019 to 2021, 78,392 to 118,851 tons of pesticides was applied annually, with similar proportions in both countries, and the use of herbicides (33.8 to 50.0%) was greater than that of insecticides, fungicides and/or bactericides, which are used in corn crops in Mexico and coffee in Colombia. The corresponding release of pesticides has increased the risks to producers and the environment. For example, 4,4′-DDDT, endosulfan II, endosulfan sulfate, endrin, parathion, chlorpyrifos, endrin aldehyde, heptachlor, heptachlor epoxide, endrin ketone and methoxychlor, among others, were detected in tributaries or surface waters in the coffee zone of Quindío, Colombia, which indicated that pesticide residues in water affected the use of this water for animal feed and other agricultural–horticultural purposes and threatened human health. Navarro et al. [30] measured pesticide residues, including imidacloprid, lambda-cyhalothrin and chlorpyrifos, in agricultural soils designated for onion cultivation in Boyaca, Colombia. Pesticide residues, which mainly include organochlorines, carbamates and organophosphorus pesticides, have also been detected in urban or treated water supply systems [31].

In Mexico, organophosphate pesticide exposure in infants has been shown to have neurotoxic and neurobehavioral effects. Yañez-Estrada et al. [32] evaluated the presence of the metabolite dialkyl phosphate (DAP) in urine as an indicator of exposure to organophosphate pesticides and reported that children aged 5 to 14 years with relatively high concentrations of DAP had low cognitive performance and low scores in evaluations of working memory. Such effects are more common in rural communities with recurring exposure to pesticides at home or at work on farm plots. Pérez-Herrera et al. [33] reported that a family primarily involved in agricultural activities is exposed to 12, 6 and 6 active ingredients of insecticides (methomil, methamdophos, chlorpyrifos, etc.), herbicides (glyphosate, 2,4-D, paraquat glufosinate, etc.) and fungicides (mancozeb, captan, benomyl, metalaxyl, etc.), respectively. Zuñiga-Venegas et al. [34] comprehensively reviewed the health effects of pesticide exposure in Latin America and the Caribbean and reported that the main effects are as follows: genotoxicity, neurobehavioral issues, placental effects, teratogenic effects, cancer, thyroid function imbalances, reproductive effects, birth effects, child growth abnormalities, respiratory effects, allergy, kidney function imbalances, liver injury and hematological parameter and lipid profile effects.

The PCA of the total harvested area and areas of the five main crops (Figure 2) revealed clusters of countries similar to those identified in the PCA of applied pesticide amounts (Figure 1). That is, the cultivated areas of Brazil (G1), Argentina (G2), Colombia and Mexico (G3) and countries in Central and South America (G4) are markedly different and are correlated with the magnitude of pesticide application to crops. For example, from 2016 to 2021, the main crops in Brazil and Argentina were transgenic and nontransgenic soybeans, which received the highest amounts of herbicides. In Colombia and Mexico, coffee and corn crops presented the highest application rates of fungicides, insecticides and herbicides. However, this trend was not observed in the 13 countries of Central and South America, where a greater diversity of species was grown without specific patterns (Table 2, Figure 2). Larsen et al. [35] studied the effects of the abandonment of or reduction in cultivated areas on the magnitude of pesticide application in California, United States, and determined an increase in environmental benefits. However, they reported that in the first few years (1–8), the use of pesticides increased in areas that remained under cultivation; according to the authors, this increase occurred because the continuous application of pesticides led to resistance in weeds and pest microorganisms, necessitating certain increases in application depending on the species remaining under continuous cultivation. The authors observed a direct relationship between the cultivated area and the amount of pesticides applied, but they also suggested conducting more in-depth studies and studies considering the long-term environmental effects on groundwater.

Correlation analysis was performed for the applied pesticides and the total harvested area and areas of the five main crops. In Brazil, positive and significant correlations were observed between the total area and areas of the three main crops and the application amounts of total pesticides, herbicides, insecticides and fungicides plus bactericides (0.84 ≤ r ≤ 0.98, p ≤ 0.01). However, an inverse and significant relationship was observed between the areas of the fourth and fifth main crops and the total amount of pesticides applied (−0.75 ≤ r ≤ −0.89, p ≤ 0.01). These trend differences indicated that as the total cultivated area and areas of the first three main crops (soybeans, corn and sugarcane) increased significantly, the application of pesticides rose. Among countries in Central and South America, positive and significant correlations were consistently detected between pesticide use and harvested area (0.17 ≤ r ≤ 0.64, p ≤ 0.01), except for the area of the first main crop and the application amounts of fungicides and bactericides. These correlation patterns indicated that from 1990 to 2021, the use of pesticides increased, in some cases (e.g., Uruguay and Paraguay) with exponential growth in relation to the growth of the cultivated area. In Argentina, positive and significant correlations were observed between the total area harvested and areas of the first and third main crops (soybeans and corn) and the application amounts of total pesticides and herbicides; this correlation has increased in recent decades with increases in glyphosate application and the areas cultivated with soybeans, maize tolerant to imazapic and sunflower tolerant to imazapyr [36].

From 2011 to 2021, the harvested area ranged from 3.8 to 4.4 million hectares and from 15.2 to 16.8 million hectares in Colombia and Mexico, respectively, whereas pesticide use decreased. For example, in Colombia, a total of 73,144 tons was applied in 2014, and this value fluctuated downward to 39,324 tons in 2021. In contrast, the applied amount in Mexico fluctuated from 53,219 to 41,681 tons without defined patterns, with a negative and significant correlation between the application amounts of total pesticides and herbicides and the total area harvested and areas of the five main crops (−0.27 ≤ r ≤ −0.64, p ≤ 0.01) (Table 3). In terms of herbicides, in Colombia, the application of herbicides increased from 2011 to 2021, with certain fluctuations ranging from 30,037 in 2013 to 25,157 tons in 2017 and from 40,352 tons in 2019 to 17,752 tons in 2021, indicating that, in certain years, increased application probably corresponded to illicit crop plantings. In Mexico, the application of herbicides remained constant from 2011 to 2021, with fluctuations from 11,809 to 8973 tons per year applied to corn, bean, sorghum, sugarcane, coffee and/or wheat crops. Notably, Mexico is the third largest agrochemical market in North America, after the United States and Canada, with total pesticide application amounts ranging from 40 to 50 thousand tons per year [37]. The inverse relationships between applied pesticides and cultivated areas in Colombia and Mexico suggest that the amount of pesticides applied per unit area is generally decreasing, probably due to the effects of high pesticide costs, the introduction of agroecological or organic practices and the nonapplication of herbicides to counteract illicit crops in the case of Colombia, among other factors.

The statistical analysis of the pesticide data and cultivated areas revealed interdependence, with implied damage and risks to human health, biodiversity, ecosystems and ground and surface waters; air pollution; accumulation in soils; and alterations in the microbial community and environment not only in populations and environments close to cultivated areas or exposed people but also in urban areas or more remote areas, especially under the influence of climate change [38]. There are no reported strategies in the study region for the continuous environmental monitoring of pesticides, and each country has different approaches according to the thematic concerns of research groups [28,34,39]. Additionally, alternative strategies for the use of pesticides have been tested, but such strategies are not part of agricultural production policies, and there is little or no regulatory legislation on products or residual limits [40,41,42]. In recent decades, agroecology and organic agriculture have emerged as strategies for transitioning to production modes without the use of pesticides, but they are associated with low and high fluctuations in productivity [43].

The information analyzed herein reveals the need in commercial agriculture for alternative options to pesticides to manage weeds and pathogenic microorganisms or pests and to reduce production costs; one option is the major diffusion or demonstration of experimental and practical experiences to manage these problems (i.e., microbial herbicides, phytopesticides and nano-biopesticides), which suggests the need to identify options for every crop and agroclimatic conditions [42]. A second intervention area is related to national polices on pesticides use, in which all Latin America countries confront the serious difficulties posed by the lax or permissible regulatory systems; for example, 72 pesticides are used in Latin America that are prohibited in the European Union (EU) [44]. Based on the magnitude of cultivated area in each country, the implementation of alternative strategies differs in terms of difficulty or complexity from smaller to larger areas; for example, all Central American countries could face fewer difficulties than Brazil, Argentina, Colombia or Mexico, which are compounded by major commercial interests. Every country has local or regional initiatives on organic, agroecological or low-input agriculture, but these initiatives have not yet been consolidated to build a national or Latin American movement that influences country policies.

5. Conclusions

In the 17 studied Latin American countries, more than one million tons of pesticides is applied annually to agricultural crops, with direct health effects for farm workers, their families and the residents of all population centers close to the cultivated areas. The PCA of pesticide application amounts (1990–2021) revealed four clusters of countries with distinct trends from higher to lower application amounts: Brazil (G1) > Argentina (G2) > Colombia and Mexico > Central and South American countries. These patterns in pesticide use were echoed in the multivariate PCA of the total harvested area and the areas of the five main cultivated crops in each country. In this work, pesticide use and harvested area were positively correlated, except in Colombia and Mexico, where the correlation was negative or not significant. The trends in correlation varied among country groups, with a marked influence on the magnitude of pesticide use and harvested area, which implied differences in complexity; for example, in Brazil, Argentina, Mexico and Colombia, the areas cultivated with export crops were larger than those of all Central American countries. The major increase in herbicide use merits greater attention from research institutions for practical novel strategies and from governmental institutions to improve the regulatory systems of pesticide use and also annual insecticide and fungicide applications. Given these trends, more in-depth studies should be performed to monitor the damage and risks of pesticides for each country in the long term, although some progress has been reported in countries such as Brazil, Argentina, Colombia and Mexico. For small-scale farmers, alternative options for the use of pesticides have been developed, but more practical experience is needed on the basis of experimental evidence.