Effects of natural habitat composition and configuration, environment and agricultural input on soybean and maize yields in Argentina

https://doi.org/10.1016/j.agee.2022.108133Get rights and content

Highlights

  • Agricultural input correlated negatively with the percentage of natural habitat.

  • Environmental and crop management variables explained spatial variation in yield.

  • Fields yields increased when they were recently transformed from natural habitat.

  • Effects of natural habitat on yield were not detected, probably due to high input.

Abstract

A fundamental challenge of land use management is to sustain the production of food, energy and fiber whilst preserving biodiversity and ecosystem functions. Some promising solutions to current resource-use conflicts are rooted in (agro) ecological intensification, which proposes that ecosystem functions provided by natural habitat can largely replace agrochemical inputs. Here, we evaluate how natural habitat is distributed in relation to agricultural input and the environmental potential for crop production, and whether natural habitat can explain the variations in yield not explained by management and environmental factors. In our analysis, we relied on environmental and management variables from 2858 soybean and 1548 individual maize fields provided by a farming organization in Argentina, and assessed landscape metrics of natural habitat composition (percentage of natural habitat) and configuration (edge density) for each one. We found that fields with higher fertilizer and seed input had lower percentages of natural habitat. Spatial variation in yield was well explained by environmental and management variables for both soybean and maize fields, and landscape metrics showed no relationship to the residuals of the models. However, fields recently transformed from natural habitat had higher yields than those with a long history of agricultural use. We conclude that compensatory management may mask the beneficial effects of natural habitat to some extent, especially in fields with intensive agrochemical use.

Introduction

Agricultural expansion and conventional land-use intensification have led to landscape homogenization and global biodiversity loss (Diaz et al., 2019, Martin et al., 2019, Seppelt et al., 2014). Decreases of biodiversity in agricultural landscapes have also resulted in a decline in ecosystem services supporting sustainable crop production (Tscharntke et al., 2005). Such decreases are not surprising, since conventional intensification has largely ignored the positive role of biodiversity in crop production systems (Seppelt et al., 2020).

Over the last few years, the establishment and conservation of natural habitat in agricultural landscapes have been promoted under the paradigm of ecological intensification (Garibaldi et al., 2019). The presence of natural areas in agroecosystems is expected to create win-win situations for biodiversity and agriculture through the ecosystem functions these areas provide for crops (Garibaldi et al., 2020). Some of the most important ecosystem functions in this context are generated by mobile species and their interactions with crops (Kremen et al., 2007). The importance of migration between natural and agricultural habitats has been shown for multiple crop types and ecosystem functions such as pollination (Ricketts et al., 2008) and biological pest control (Karp et al., 2018, Tscharntke et al., 2016). However, the effect of the composition and configuration of natural habitat on field crop yield is still little understood.

Farmers are responsible for managing most of the world´s populated land (Ramankutty et al., 2008). Improved knowledge of on-farm benefits of changes in the composition and configuration of natural habitat on a landscape scale could have worldwide environmental implications. For example, previous studies showed that both soybean and maize, two of the most widely grown crops, benefit from natural habitat as it promotes the natural enemies of pests (Gonzalez et al., 2020, Santana Sousa et al., 2012). Additionally, soybean yields are also increased through more effective pollination (Garibaldi et al., 2021, Monasterolo et al., 2015).

Our main objective was to assess the relationships between the environment, land management, landscape structure and yields of soybean and maize in Argentina. We targeted soybean and maize crops because, globally, 120.5 and 197.2 million ha were harvested of these crops, respectively, during 2019 (Anon, 2019). Argentina accounted for 14% of the global soybean and 4% of the global maize production area in that year (Anon, 2019). To understand the relationships between landscape, environmental and management variables, we first evaluated how the landscape metrics (percentage of natural habitat and edge density) were distributed with regard to agricultural input and environmental potential for crop production. We then assessed the main drivers of crop yield and evaluated whether landscape metrics could account for the spatial variation in yield that was not explained by management and environmental factors. This information would indicate whether the ecosystem services provided by natural habitat substantially impact yields. In total, we gathered data from 2858 soybean and 1548 maize fields.

Section snippets

Data collection

Data was collected using an extensive, standardized protocol co-developed with the Regional Consortiums for Agricultural Experimentation (CREA) through the DAT CREA project. Regional Consortiums for Agricultural Experimentation is a non-profit civil association integrated and directed by agricultural entrepreneurs (>1800 farms) who meet in groups to share experiences and knowledge (https://www.crea.org.ar/). In total, the assessed area covered more than 324,000 ha of agricultural land

Results

During the 2018–2019 growing season, we collected data from 2858 soybean and 1548 maize fields across almost all extensive crop regions of Argentina (Fig. 1). Average single field size was 70.83 ha for soybean and 63.90 ha for maize; maximum field sizes were 500 and 360 ha, respectively. Landscapes with soybean fields were characterized by an average of 36.9% natural habitat and an edge density of 33.6 m ha-1. Maize field landscapes had an average of 35.3% natural habitat and 31.3 m ha-1 edge

Discussion

In this study, we gathered data from hundreds of fields in Argentina to assess the relationships between the environment, land management, landscape structure and yields. Due to the complex correlation structure between environmental, management and landscape variables, we implemented a three-step approach. We found that fields with greater agricultural input negatively correlated with the percentage of natural habitat. Land with higher environmental potential for grain production (i.e., field

Conclusions

In this paper, we covered almost all major field-crop regions of Argentina, where soybean and maize production represents one of the country’s main sources of income. Agricultural input was negatively correlated with natural habitat. Environmental and management variables explained yield variability in both crops, and yield models considering these effects satisfactorily described the large spatial variation of yield in the study regions. Neither percentage of natural habitat nor edge density

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.

Acknowledgments

We thank Dr. Facundo Oddi for previous discussions, Dr. Juan Aguero for reading the manuscript in previous versions and CREA research and development unit for valuable feedback during the research process. Our thanks also go to two anonymous reviewers that made valuable comments on the manuscript, and to the Consejo Nacional de Investigaciones Científicas y Técnicas - Argentina, and the Asociación Argentina de Consorcios Regionales de Experimentación Agrícola for co-funding a scholarship to the

References (56)

  • L.N. Vitantonio-Mazzini et al.

    Management options for reducing maize yield gaps in contrasting sowing dates

    Field Crops Res.

    (2020)
  • J.N. Volante et al.

    Agricultural expansion in the Semiarid Chaco: Poorly selectivecontagious advance

    Land Use Pol.

    (2016)
  • L. Yang et al.

    Non-crop habitats promote the abundance of predatory ladybeetles in maize fields in the agricultural landscape of northern China

    Agric. Ecosyst. Environ.

    (2019)
  • W. Zhang et al.

    Ecosystem services and dis-services to agriculture

    Ecol. Econ.

    (2007)
  • N. Alexandridis et al.

    Archetype models upscale understanding of natural pest control response to land-use change

    Ecol. Appl.

    (2022)
  • FAO, 2019. 〈https://www.fao.org/faostat/en/#home〉. Accessed in November...
  • INTA 2019 Mapa Nacional de Cultivos campaña...
  • G. Baldi et al.

    Land-use and land cover dynamics in South American temperate grasslands

    Ecol. Soc.

    (2008)
  • D. Bates et al.

    Fitting Linear Mixed-Effects Models Using lme4

    J. Stat. Softw.

    (2015)
  • K.P. Burnham et al.

    Model Selection and Multimodel Inference: A Practical Information-theoretic Approach

    (2002)
  • K.P. Burnham et al.

    AIC model selection and multimodel inference in behavioral ecology: Some background, observations, and comparisons

    Behav. Ecol. Sociobiol.

    (2011)
  • M. Butinof et al.
  • N.P. Chacoff et al.

    Pollinator dependence of Argentinean agriculture: current status and temporal analysis

    Am. J. Plant Sci. Biotechnol.

    (2010)
  • M.S. Cid et al.

    Grass-fed beef production systems of Argentina's flooding pampas: understanding ecosystem heterogeneity to improve livestock production

    Outlook Agric.

    (2011)
  • S. Díaz et al.

    Pervasive Human-Driven Decline of Life on Earth Points to the Need for Transformative Change

    Science

    (2019)
  • B.L. Gambini et al.

    Exploring genotype, management, and environmental variables influencing grain yield of late-sown maize in central Argentina

    Agric. Syst.

    (2016)
  • L. Garibaldi et al.

    Working landscapes need more than 20% native hábitat

    Cons. Lett.

    (2020)
  • L. Garibaldi et al.

    SmalleR Agricultural Fields, More Edges, and Natural Habitats Reduce Herbicide-resistant Weeds

    (2022)
  • Cited by (0)

    View full text