Large-scale monoculture reduces honey yield: The case of soybean expansion in Argentina

https://doi.org/10.1016/j.agee.2020.107203Get rights and content

Highlights

  • Soybean expansion since 1996 has intensified industrial agriculture in Argentina.

  • This period of soybean expansion is associated with a 60 % decrease in honey yield.

  • Since 1996, annual variation in soybean area and honey yield are negatively related.

  • Industrial agriculture can threaten beekeeping and honey production.

Abstract

Large-scale changes introduced by industrial agriculture can affect other productive activities such as beekeeping, which heavily depends on floral resources and responsible management of agrochemicals. To assess the long-term effect of soybean expansion on honey production in Argentina, we evaluated the relationships between the area cultivated with soybean and honey yield and total production by managed honeybee (Apis mellifera) hives between 1961 and 2016. Results indicate that the expansion of the area cultivated with soybean since 1996, which involved the replacement of natural habitats with extensive cultivated fields, intensive use of transgenic seeds, agrochemicals and machinery, was associated with a reduction in honey yield of ∼60 %. Furthermore, since 1996 honey yield tended to be lower in years in which the area cultivated with soybean was larger. Causal modelling of this data confirmed a strong negative effect of increasing soybean cultivation on Argentinean honey production via decreasing yield during the period 1996–2016. Although the underlying mechanisms still need to be disentangled, this work provides support to the hypothesis that the beekeeping crisis in Argentina can relate to soybean expansion. More generally, it provides evidence that industrial agriculture has a negative impact on apiculture.

Introduction

Industrial agriculture based on the intensive cultivation of few crops, which prevails in several regions of the world (Foley, 2011), is associated with a profound transformation of production methods. In particular, the frequent use of genetically modified seeds, large inputs of agrochemicals and direct sowing, as well as of heavy machinery that covers large areas of land in less time have reduced cultivation and harvesting costs, leading to a reduction in labour costs (Kremen et al., 2012). Consequently, the spatial scale of production based on single crops has increased notably. These extensive monocultures have led to a simplification and homogenization of agricultural landscapes, posing challenges for biodiversity and the benefits derived from it (Bommarco et al., 2013; Aizen et al., 2019).

In particular, the dominance of monoculture in detriment of other land uses, together with the intensive use of non-selective herbicides, leads to a decrease in plant species diversity at both local and regional scales through different non-mutually exclusive mechanisms (Hillebrand et al., 2008). First, the expansion of arable land reduces areas dedicated to other productive activities such as raising livestock, which, at intermediate stocking rates, may actually increase plant species diversity in natural and semi-natural pastures (Smith et al., 2016). Second, monoculture expansion within agricultural lands is detrimental to crop diversity (Aizen et al., 2019). Third, demand for more cultivated land leads to the expansion of the agricultural frontier through the destruction of native forests and grasslands, with the consequent loss of large areas of natural and semi-natural vegetation (Gibbs et al., 2010). Finally, broad-spectrum herbicides eliminate not only target weeds but also vegetation growing on the edges of crop fields and roads, reducing plant diversity both inside and outside crop plots (Goulson et al., 2015; Potts et al., 2016). All these mechanisms lead to a loss of plant diversity with consequences for other trophic levels.

A reduction in plant diversity implies a loss in the amount, variety, and spatial and temporal availability of floral resources (nectar and pollen), which can strongly affect animals, such as bees (Apoidea), that feed exclusively upon these resources throughout their lives. Nectar provides mostly carbohydrates whereas pollen provides proteins, lipids, vitamins and minerals (De Groot, 1953; Haydak, 1970), with the nutritional value of floral resources varying between species and vegetation types (Keller et al., 2005; Odoux et al., 2012; Roulston and Cane, 2000). Nutritional requirements of bees may be inadequately satisfied when the habitat is dominated by extensive monocultures (Branchiccela et al., 2019; Brodschneider and Crailsheim, 2010; Naug, 2009). In fact, a pollinator-dependent monoculture represents an oversupply of a single floral resource only available for a few weeks, followed by a long period of food scarcity (Dolezal et al., 2019a; Requier et al., 2015). This strong reduction in the spatial diversity and temporal availability of floral resources may jeopardize bee survivorship as well as the pollination services provided by bees (Potts et al., 2010).

The honeybee (Apis mellifera), the most important species worldwide for honey production and managed pollination, depends on the access to diverse floral resources over time to survive and maintain healthy and productive colonies (Standifer, 1967; Herbert et al., 1977). Landscape homogenization, decreasing flower abundance and diversity, and different agrochemicals, including a plethora of pesticides, can affect honeybee colony performance by increasing the incidence of lethal and sub-lethal effects (Dolezal et al., 2016; Henry et al., 2012; Mullin et al., 2010; Scofield and Mattila, 2015; Smart et al., 2016; Yang et al., 2008). These sublethal effects can impact individual honeybee foraging efficiency (Dolezal et al., 2015; Toth et al., 2005), the capacity to deal with agrochemicals present in the environment (Di Pasquale et al., 2013, 2016; Goulson et al., 2015; Krupke et al., 2012; Tosi et al., 2017, 2018; Dolezal and Toth, 2018), and the ability to withstand pathogen infection (Di Pasquale et al., 2013, 2016; Dolezal et al., 2016, Dolezal et al., 2019b). The consequences of the additive and/or synergistic effects of these stressors on individuals can scale up to the health of entire beehives (Alaux et al., 2010; DeGrandi-Hoffman and Chen, 2015). In particular, hive development and productivity could be affected by pronounced cycles of bounty and dearth of floral resources associated with landscapes dominated by single crops (Dolezal et al., 2019b; Requier et al., 2017; Smart et al., 2016). Therefore, honey yield (i.e., honey production per hive) is expected to be negatively affected by the expansion of industrial agriculture.

In Argentina, as in other South American countries, soybean is currently the emblematic commodity crop of industrial agriculture (Dros, 2004; Manzanal, 2017; Pengue and Altieri, 2005). The area dedicated to the cultivation of this oilseed was fairly insignificant at the beginning of 1970 (FAOSTAT, 2018). Since then, the greater profitability of agriculture compared to livestock, coupled with improvements in agricultural techniques, has led to a drastic transformation of the countryside (Cadenazzi, 2009; Pengue, 2001). This transformation involved, among others, a more continuous and intensive use of land cover for monoculture crop production, greater use of agrochemical inputs (fertilizers, herbicides and pesticides), and the employment of larger and more impacting machinery.

The release of genetically modified soybean in 1996 in Argentina was the foundational event that triggered the exponential growth of this crop in South America during the last decades (Satorre, 2005), also facilitated by a global economic context favourable to the export of primary products and commodities (Páez, 2016). The initial release and success of different varieties of genetically modified soybean into the Argentinean agricultural economy opened the door for the intensification of industrial agriculture. More specifically, the technological package associated with the cultivation of genetically modified soybean included: (1) use of soybean seeds resistant to glyphosate-based herbicides; (2) massive application of these herbicides for the control of weeds, along with other agrochemicals; (3) direct sowing or zero tillage; (4) use of heavy machinery for sowing and harvesting; and, consequently, (5) an increase in the spatial scale of production and the average area of cultivated lots (Cadenazzi, 2009; Dominguez and Sabatino, 2006; Kahl, 2012; Pengue and Altieri, 2005). Through this process of land-use change, soybean became the dominant crop (i.e. the one that individually has the largest cultivated area) in Argentinean agriculture (Aizen et al., 2009).

Even though Argentina is one of the main producers and exporters of honey worldwide, national beekeeping trends are showing critical signs of decreasing honey production (Requier et al., 2018; Sanchez et al., 2018). Although some soybean cultivars produce nectar accessible to the honeybee (Erickson, 1975; Erickson and Garment, 1979; Villanueva-Gutiérrez et al., 2014), the increase in soybean production has been hypothesized as the single most important cause of this decline (Blengino, 2013). Beekeeping activity is best-developed in the central region of the country, including the Pampas plain, home to 68 % of the beekeepers, 80 % of the hives (Ferrari et al., 2011), and 93 % of the exported honey production nationwide (Rabaglio and Castignani, 2015). This region also accounted for about 90 % of the total national soybean production between 2010–2018 (Agroindustry Open Data, 2018). Given the large geographical overlap between agriculture and beekeeping in Argentina, a detailed analysis of the link between temporal changes in soybean cultivation and honey production in recent decades at the country level is urgently needed to address the hypothesis that industrial agriculture has a negative effect on apiculture.

In this study, we assessed the impact of the industrial agriculture on beekeeping and honey production, using the rapid expansion of soybean cultivation in Argentina as a study case. First, we evaluated the temporal relation between soybean cultivated area and honey yield over 55 years in the long and short term (i.e., long-term trend and detrended variability, respectively). Then, we tested a causal model that considers five variables (i.e., soybean area, honey export value, number of hives, honey yield, and total honey production) to assess the influence of soybean area on Argentinean honey production. We hypothesized that expansion of soybean cultivation has negatively affected overall honey production mainly through changes in honey yield. Therefore, after the implementation of genetically modified soybean in 1996, which led to the intensification of industrial agriculture in Argentina, we predict (1) a progressive decrease in honey yield over time, (2) a negative relation between honey yield and soybean area once the long-term trend has been removed (fine-scale temporal coupling), and (3) an indirect negative effect of soybean area on total honey production mediated by decreasing yield.

Section snippets

Materials and methods

We retrieved data from the United Nations Food and Agriculture Organization (FAO) database (FAOSTAT, 2018), which was analysed with different statistical techniques using the free software R (version 3.4.2) (R Core Team, 2015). From FAOSTAT (2018), we obtained annual values of land area cultivated with soybean (actually reported as harvested area), total number of beehives, and total honey production in Argentina between 1961 and 2016. We checked this FAO data against national statistics (see

Results

Soybean occupied <0.01 % of the total cultivated area of Argentina in 1961, representing ca. 1000 ha (Fig. 1a, Table S3). The segmented regression estimated two break points in the rate of expansion of soybean area, the first in 1975 (P < 0.001; CI = 1971–1978) and the second in 1995 (P < 0.001; CI = 1993–1997) (Fig. 2a). Since 1975, soybean cultivation began to expand at an average rate of 282,797 ha year−1, reaching an area of approximately 6,000,000 ha harvested in 1995. The area cultivated

Discussion

Industrial agriculture, which implies a radical landscape transformation and the intense use of agrochemical inputs, can affect other productive activities. This is the case of beekeeping that depends on the presence of diverse nutritional resources for the honeybee at landscape scale. Our results support the hypothesis that in Argentina the intensification of industrial agriculture, based on the release of transgenic soybean in 1996, has had a strong impact on honey production.

In particular,

Conclusions

Industrial agriculture, currently dominant in several agricultural regions of the world, has introduced numerous stressors that can directly and indirectly affect bees and other insects. We provide evidence that a decrease in honey yield can be linked to soybean monoculture intensification in Argentina since 1996. Future experimental and correlative studies should test the underlying mechanisms, interactions and synergies.

More generally, this study contributes to the growing scientific evidence

Author contributions

GSDG, MAA, AS & CLM conceived the idea and the analyses. GSDG collected data. GSDG & MAA performed analyses. GSDG wrote the initial manuscript. All co-authors contributed significantly to improving the manuscript. All authors gave final approval for publication.

Funding

This work is part of the “Safeguarding Pollination Services in a Changing World: theory into practice” (SURPASS2), an international collaboration funded by the Newton Fund Latin American Biodiversity Programme, awarded through the UKRI Natural Environment Research Council (NERC) NE/S011870/1 and the National Council of Scientific and Technological Research of Argentina (CONICET) RD 1984-19. Support from the National Fund for Scientific and Technological Research of Argentina (FONCYT) [PICT

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

We thank Amy Toth, Ron Miksha, Gherardo Bogo and Pablo Hünicken for constructive comments on an early version of this manuscript, and INTA ProApi, Beekeepers Association of the Comarca and the Secretary of Urban and Periurban Agriculture of Bariloche for their support.

References (98)

  • S. Tosi et al.

    A 3-year survey of Italian honeybee-collected pollen reveals widespread contamination by agricultural pesticides

    Sci. Total Environ.

    (2018)
  • M. Vallejos et al.

    Transformation dynamics of the natural cover in the dry chaco ecoregion: A plot level geo-database from 1976 to 2012

    J. Arid Environ.

    (2015)
  • D.E. Vázquez et al.

    Chronic exposure to glyphosate induces transcriptional changes in honey bee larva: a toxicogenomic study

    Environ. Poll.

    (2020)
  • Agroindustry Open Data

    Agroindustry Secretary, Ministry of Production and Labor, Argentina

    (2018)
  • M.A. Aizen et al.

    Expansión de la soja y diversidad de la agricultura argentina

    Ecol. Austral

    (2009)
  • M.A. Aizen et al.

    Global agricultural productivity is threatened by increasing pollinator dependence without a parallel increase in crop diversification

    Glob. Change Biol.

    (2019)
  • C. Alaux et al.

    Diet effects on honeybee immunocompetence

    Biol. Lett.

    (2010)
  • A. Basilio et al.

    Análisis polínico de mieles de la región chaqueña: comparación del origen floral entre las zonas: domo central y esteros, cañadas y selvas de ribera

    Revista de Investigaciones Agropecuarias

    (2002)
  • C. Blengino

    Sector apícola 2013. Informe de coyuntura 3

    (2013)
  • B. Branchiccela et al.

    Impact of nutritional stress on the honeybee colony health

    Sci. Rep.

    (2019)
  • R. Brodschneider et al.

    Nutrition and health in honey bees

    Apidologie

    (2010)
  • A.L. Cabrera et al.

    Biogeografía de América Latina. Washington DC: Programa Regional de Desarrollo Científico y Tecnológico

    (1973)
  • L.G. Cadenazzi

    De granero a aceitera del mundo. Las exportaciones agrarias argentinas entre 1930 y 1970

    Agroalimentaria

    (2009)
  • N.P. Chacoff et al.

    Pollinator dependence of Argentinean agriculture: current status and temporal analysis

    Am. J. Plant Sci. Biotechnol.

    (2010)
  • A.P. De Groot

    Protein and amino acid requirements of the honeybee (Apis mellifera L.)

    Physiol. Comp. Oecol.

    (1953)
  • G. Di Pasquale et al.

    Influence of pollen nutrition on honeybee health: Do pollen quality and diversity matter?

    PLoS One

    (2013)
  • G. Di Pasquale et al.

    Variations in the availability of pollen resources affect honeybee health

    PLoS One

    (2016)
  • A.G. Dolezal et al.

    Native habitat mitigates feast–famine conditions faced by honeybee in an agricultural landscape

    Proc. Natl. Acad. Sci. U. S. A.

    (2019)
  • A.G. Dolezal et al.

    Interacting stressors matter: diet quality and virus infection in honeybee health

    R. Soc. Open Sci.

    (2019)
  • A.G. Dolezal et al.

    Pollen contaminated with field-relevant levels of cyhalothrin affects honeybee survival, nutritional physiology, and pollen consumption behaviour

    J. Econ. Ento.

    (2015)
  • A.G. Dolezal et al.

    Intensively cultivated landscape and varroa mite infestation are associated with reduced honeybee nutritional state

    PLoS One

    (2016)
  • D. Dominguez et al.

    Con la soja al cuello: Crónicas de un país hambriento productor de divisas

  • J.M. Dros

    Manejo del boom de la soya: Dos escenarios sobre la expansión de la producción de la soya en América del Sur. AID Environment

    (2004)
  • E.H. Erickson

    Effect of honeybee on yield of three soybean cultivars

    Crop Sci.

    (1975)
  • E.H. Erickson et al.

    Soya-bean flowers: nectary ultrastructure, nectar guides, and orientation on the flower by foraging honeybee

    J. Apic. Res.

    (1979)
  • European Commission

    Commission Implementing Regulation (EU) 2018/783, 2018/784 and 2018/785. Off. J. Eur. Union

    (2018)
  • FAOSTAT

    Food and Agriculture Organization Corporate Statistical Database

    (2018)
  • W.R. Fehr et al.

    Stages of Soybean Development, Special Report 80

    (1977)
  • C. Ferrari et al.

    La apicultura argentina y sus regiones. Una visión panorámica. Consejo Federal de Inversiones

    (2011)
  • FILAPI

    Declaración de la ciudad de Montevideo. Congreso Latinoamericano de Apicultura

    (2018)
  • J.A. Foley

    Can we feed the world & sustain the planet?

    Sci. Am.

    (2011)
  • L.A. Garibaldi et al.

    Seguridad alimentaria, medioambiente y nuestros hábitos de consumo

    Ecol. Austral

    (2018)
  • H.K. Gibbs et al.

    Tropical forests were the primary sources of new agricultural land in the 1980s and 1990s

    Proc. Natl. Acad. Sci. U. S. A.

    (2010)
  • D. Goulson et al.

    Bee declines driven by combined stress from parasites, pesticides, and lack of flowers

    Science

    (2015)
  • H.R. Grau et al.

    Agriculture expansion and deforestation in seasonally dry forests of north-west Argentina

    Environ. Conserv.

    (2005)
  • M. Haydak

    Honey bee nutrition

    Annu. Rev. Entomol.

    (1970)
  • M. Henry et al.

    A common pesticide decreases foraging success and survival in honeybee

    Science

    (2012)
  • E.W. Herbert et al.

    Optimum protein levels required by honeybee (Hymenoptera, Apidae) to initiate and maintain brood rearing

    Apidologie

    (1977)
  • L.T. Herbert et al.

    Effects of field-realistic doses of glyphosate on honeybee appetitive behaviour

    J. Exp. Biol.

    (2014)
  • Cited by (19)

    • Biological-based and remote sensing techniques to link vegetative and reproductive development and assess pollen emission in Mediterranean grasses

      2022, Ecological Informatics
      Citation Excerpt :

      Large surfaces covered by monospecific cereal crops reduce considerably the diversity and abundance of wild plant species in favour of the dominance of mainly self-fertile and cleistogamous cultivars (Thom et al., 2018). The effect of the agriculture expansion process on pollinators is evidenced by the reduction of honey yield related to soybean expansion in South America which has been hypothesised as the origin of an environmental and beekeeping crisis in Argentina by de Groot et al. (2021). The satellite remote sensing time-series during the period 2014–2021 confirmed the results of greater pollen emission of natural and semi-natural grasslands (Figs. 8–9).

    View all citing articles on Scopus
    View full text