High bee functional diversity buffers crop pollination services against Amazon deforestation

https://doi.org/10.1016/j.agee.2021.107777Get rights and content

Highlights

  • Assessed Amazon bee communities on açaí palm along tropical forest cover gradient.

  • Comprehensive trait dataset used to investigate bee response and effect traits.

  • Bee responses to forest loss were predicted by body size.

  • Pollination increased with bee functional diversity but resilient to deforestation.

  • High forest landscapes support both native bee diversity and ecosystem services.

Abstract

Predicting outcomes of land use change on biodiversity and ecosystem services remains a key priority for ecologists, but may be particularly challenging in diverse tropical ecosystems. Trait-based approaches are a key tool to meet this challenge. Such approaches seek functional mechanisms underpinning species’ responses to environmental disturbance and contributions to ecosystem services. Here, we use a functional trait approach to study effects of land use change on stingless bee communities and on pollination services to açaí palm (Euterpe oleracea, Arecaceae) in the eastern Brazilian Amazon. We compared traits of stingless bees visiting açaí inflorescences across a land use intensity gradient (low to high forest cover) to determine: (1) the role of traits in bee species’ responses to deforestation; (2) how deforestation affects functional composition of bee communities; and (3) whether bee traits better explain variation in açaí fruit production than species diversity metrics. We found that bee species’ responses to deforestation were non-random and predicted by body size, with small-sized bees more susceptible to forest loss, and changes in functional diversity of bee communities were important for pollination services. However, not all changes in functional diversity were associated with forest loss. Together, these results suggest that: (1) large tracts of minimally disturbed tropical rainforest are vital for the conservation of diverse stingless bee communities; (2) efficient pollination is contingent on bee species not only having divergent trait values (functional dispersion), but also traits’ relative abundance in communities (functional evenness); and (3) high functional diversity in stingless bee communities buffers açaí pollination services to loss of sensitive species. Thus, conservation strategies must focus on protecting wider biodiversity, not just ecosystem services, to guarantee conservation of native eusocial bee taxa. Doing so will safeguard crop pollination services, the pollination of native plant communities, and the long-term resilience of Amazon forest ecosystems.

Introduction

The conversion of natural habitats into agricultural land is a major driver of global biodiversity loss (Foley et al., 2005). As a consequence we lose wild species that provide essential ecosystem services (Dainese et al., 2019). Understanding the impacts of land use change on biodiversity and ecosystem functioning (BEF) is key to devising land management practices that support wider biodiversity and ecosystem services in croplands (Kleijn et al., 2015). Much of the evidence for positive BEF relationships comes from plant communities (Tilman et al., 2014). In contrast, evidence from arthropod-mediated ecosystem services, such as pollination and pest control, remains mixed (Dainese et al., 2019; Kleijn et al., 2015; Ricketts et al., 2016), mainly due to continued uncertainty over underlying mechanisms (Bartomeus et al., 2018).

To provide a more mechanistic understanding of BEF relationships, ecologists have developed ‘trait-based’ approaches (Dı́az and Cabido, 2001). These aim to identify morphological, physiological, and behavioral attributes of species (herein ‘traits’) that determine sensitivity to environmental change (‘response traits’), and contribute to specific ecological functions (‘effect traits’) (McGill et al., 2006). Trait-based approaches have been used to investigate impacts of land use and climate change on wild bee communities, and of bee diversity on pollination services (Giannini et al., 2020b, Williams et al., 2010, Woodcock et al., 2019). Despite recent advances, studies linking bee species’ environmental sensitivity and function (i.e., response-effect trait framework) remain scarce (but see Martins et al., 2015; Bartomeus et al., 2018). As such, trait-based approaches have so far failed to provide general predictions on how land use change alters bee pollination services (Bartomeus et al., 2018). We can improve this framework’s overall predictive power by testing it in diverse ecological contexts, especially where information on traits is limited, such as the tropics (Archer et al., 2014).

In tropical and subtropical regions, eusocial stingless bees (Hymenoptera: Apidae: Meliponini) are the dominant flower-visitor taxa in both natural and agricultural habitats (Bawa, 1990, Biesmeijer and Slaa, 2006), and vital crop pollinators (Heard, 1999). Most stingless bee taxa build their nests in trees and are generalist flower visitors (Roubik, 1989). Yet, among species, there exists a striking diversity of morphological, physiological, and behavioral adaptations to maximize survival and resource exploitation in diverse tropical habitats (Hrncir and Maia-Silva, 2013), even allowing some species to thrive in human-modified landscapes (Jaffé et al., 2016). Nonetheless, many species are poorly adapted to forest loss, leading to precipitous declines in stingless bee abundance and diversity in degraded landscapes (Brosi et al., 2008, Ricketts et al., 2008). Unlike most other tropical insects, information on traits that could influence species’ responses to land use change is widely available for stingless bees, and recent studies have found that body size (Brown and De Oliveira, 2014, Mayes et al., 2019, Smith and Mayfield, 2018) and dominance interactions (Lichtenberg et al., 2017) influence species’ local extinction risk. However, knowledge of the impacts of species loss on mechanisms driven by functional composition (e.g., niche complementarity) in stingless bee communities, and ecosystem functioning (e.g., crop pollination), remains limited.

Bee species vary in their contribution to pollination services based on differences in morphological (e.g., body size, hairiness; Larsen et al., 2005; Stavert et al., 2016), and physiological traits (e.g., thermal tolerance; Brittain et al., 2013), and behavior during flower visits (Martins et al., 2015). Yet, evidence on whether individual, or multiple traits best explain ecosystem functioning (Gagic et al., 2015) remains equivocal, with two hypotheses being prevalent in the literature. Firstly, if function is strongly linked to a particular range or level of a single trait (‘trait state’), then that trait’s abundance in the community will be the best predictor of ecosystem functioning (‘functional identity’ or ‘mass ratio’ hypothesis) (Garibaldi et al., 2015, Grime, 1998). Alternatively, if ecosystem function is dependent on the degree of complementarity among species’ traits (e.g., spatio-temporal partitioning of flower visits), then function may be predicted by trait diversity (‘functional complementarity’ hypothesis) (Dı́az and Cabido, 2001, Gagic et al., 2015). Under both hypotheses, if bee species’ local extinction risk covaries with pollination function, then ecosystem services may be at risk under land use change (Larsen et al., 2005, Nicholson et al., 2019). On the other hand, if these variables are decoupled, for instance if functional redundancy is high and species are mutually replaceable, or if pollination is driven by common species, loss of sensitive species will not influence ecosystem service provision (Kleijn et al., 2015).

Here, we investigate how functional traits influence stingless bee responses to deforestation and pollination services to açaí palm (Euterpe oleracea Mart., Arecaceae) in the eastern Brazilian Amazon, a global hotspot for stingless bee diversity (Pedro, 2014). Açaí fruit is vitally important for food security and rural livelihoods in the Amazon region (Brondízio, 2008, Borges et al., 2020a), and, due to rapid growth in domestic and international demand, one of Brazil’s most lucrative pollinator-dependent crops (Giannini et al., 2020a). It is produced in a wide range of contexts, from smallholder agroforestry systems in its native floodplain forest habitat to intensively managed plantations in uplands (Campbell et al., 2018). Pollinators, defined as species that visit both sexual morphs of palm inflorescences, include a diverse array of insects (bees, flies, wasps, beetles, and ants), that on average increase fruit yield by 80% relative to inflorescences where pollinators have been experimentally excluded (Campbell et al., 2018). Pollination services are positively related to pollinator species richness (Campbell et al., 2018). However, among pollinators, stingless bees are its most effective pollen vectors (Bezerra et al., 2020), and the only taxa whose visitation frequencies are dependent on surrounding forest cover (Campbell et al., 2018). Thus, pollination services may be contingent on a subset of environmentally sensitive stingless bees.

In this study, we address: (1) the role of functional traits in stingless bee species’ responses to deforestation; (2) how deforestation affects functional composition of stingless bee communities; and (3) whether stingless bee traits or functional composition explain more variation in açaí fruit production than overall pollinator species diversity. We expect that stingless bee species’ responses to deforestation are non-random and influenced by their functional traits, and not only lead to changes in species richness but also functional composition. For pollination services, we make three predictions. (i) If pollination services are enhanced by functional differences across a wide range of insect taxa (e.g., bees, flies, wasps, beetles), overall pollinator richness will remain the best predictor of açaí fruit production. (ii) If stingless bees are important pollinators, taxonomic or trait-based indices of stingless bee communities may replace or explain additional variation in pollination services on top of overall pollinator richness. (iii) Traits may interact with overall pollinator richness. This could occur if stingless bee trait diversity is a proxy of functional complementarity in wider flower-visitor communities, or behavioral interactions among stingless bees and other flower visitors have antagonistic or synergistic effects on pollination services (Carvalheiro et al., 2011).

Section snippets

Materials and methods

To investigate impacts of landscape structure (forest cover) and production system (upland or floodplain) on stingless bee communities and açaí pollination services, we focused on 18 sites used for intensive production of açaí palm fruit in the Amazon estuary region, close to Belém, Pará state, northern Brazil (Fig. A1, Supplementary Materials). This region is characterized by large tracts of wet tropical rainforest, separated by large rivers and land cleared for agriculture (e.g., pasture,

Stingless bee communities visiting açaí inflorescences

A total of 33 species (16 genera) of stingless bees were collected on E. oleracea inflorescences. The most common genera (species totals) included: Trigona (5 species), Trigonisca (5), Partamona (4), Plebeia (3), and Nannotrigona (3) (for full species list, see Table A1, Supplementary Materials). Stingless bee species displayed extensive variation in trait values, with body size (inter-tegular distance, ITD) varying between 0.7 and 2.6 mm (median = 1.3 mm, IQR = 0.5 mm), colony size between 390

Discussion

Evidence for covariance between biodiversity and ecosystem services is mixed, due to high variability in species’ responses to anthropogenic stressors and relative contributions to ecosystem services (Bartomeus et al., 2018; Kleijn et al., 2015), and differential spatio-temporal scales over which diversity effects are assessed (e.g., alpha vs. beta diversity, current vs. future contribution under environmental change) (Senapathi et al., 2015, Wilcox et al., 2017). Classifying organisms by

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

The authors would like to thank all those who helped with field work, including Jamille Veiga, Raphael Nunes, and all açaí producers who provided invaluable access to field sites, local knowledge, and accommodation to field teams. Authors would also like to acknowledge Dr Eduardo Freitas Moreira for providing example landcover maps for the graphical abstract, specialists who aided the identification of non-bee visitor taxa: Dr Orlando Tobias Silveira (Vespidae), Dr José Nazareno Araújo Santos

References (78)

  • N.M. Williams et al.

    Ecological and life-history traits predict bee species responses to environmental disturbances

    Biol. Conserv.

    (2010)
  • J. Aguirre-Gutiérrez et al.

    Functional traits help to explain half-century long shifts in pollinator distributions

    Sci. Rep.

    (2016)
  • D.R. Anderson et al.

    Avoiding pitfalls when using information-theoretic methods

    J. Wildl. Manag.

    (2002)
  • C.R. Archer et al.

    Economic and ecological implications of geographic bias in pollinator ecology in the light of pollinator declines

    Oikos

    (2014)
  • J. Barlow et al.

    Anthropogenic disturbance in tropical forests can double biodiversity loss from deforestation

    Nature

    (2016)
  • I. Bartomeus et al.

    On the inconsistency of pollinator species traits for predicting either response to land-use change or functional contribution

    Oikos

    (2018)
  • Barton, K., 2015. MuMIn: Multi-Model Inference. R Package Version 1.9.13. Version 1, p. 18....
  • K.S. Bawa

    Plant-pollinator interactions in tropical rain forests

    Annu. Rev. Ecol. Syst.

    (1990)
  • F.E. Benjamin et al.

    Pollinator body size mediates the scale at which land use drives crop pollination services

    J. Appl. Ecol.

    (2014)
  • L.A. Bezerra et al.

    Pollen loads of flower visitors to Açaí Palm (Euterpe oleracea) and implications for management of pollination services

    Neotrop. Entomol.

    (2020)
  • J.C. Biesmeijer et al.

    Information flow and organization of stingless bee foraging

    Apidologie

    (2004)
  • J.C. Biesmeijer et al.

    The structure of eusocial bee assemblages in Brazil

    Apidologie

    (2006)
  • R.C. Borges et al.

    The value of crop production and pollination services in the Eastern Amazon

    Neotrop. Entomol.

    (2020)
  • Rafael Cabral Borges et al.

    A dataset of multi-functional ecological traits of Brazilian bees

    Sci. Data

    (2020)
  • Brazilian Native Vegetation Protection Law No 12.651 (Presidência da República do Brasil,...
  • C. Brittain et al.

    Biodiversity buffers pollination from changes in environmental conditions

    Glob. Chang. Biol.

    (2013)
  • E.S. Brondízio

    The Amazonian Caboclo and the Açaí Palm: Forest Farmers in the Global market

    (2008)
  • B.J. Brosi et al.

    Bee community shifts with landscape context in a tropical countryside

    Ecol. Appl.

    (2007)
  • B.J. Brosi et al.

    The effects of forest fragmentation on bee communities in tropical countryside

    J. Appl. Ecol.

    (2008)
  • J.C. Brown et al.

    The impact of agricultural colonization and deforestation on stingless bee (Apidae: Meliponini) composition and richness in Rondônia, Brazil

    Apidologie

    (2014)
  • A.J. Campbell et al.

    Anthropogenic disturbance of tropical forests threatens pollination services to açaí palm in the Amazon river delta

    J. Appl. Ecol.

    (2018)
  • L.G. Carvalheiro et al.

    Natural and within-farmland biodiversity enhances crop productivity

    Ecol. Lett.

    (2011)
  • G. Carvalho-Zilse et al.

    Threats to the stingless bees in the Brazilian Amazon: how to deal with scarce biological data and an increasing rate of destruction

  • M. Dainese et al.

    A global synthesis reveals biodiversity-mediated benefits for crop production

    Sci. Adv.

    (2019)
  • L. Elizalde et al.

    The ecosystem services provided by social insects: traits, management tools and knowledge gaps

    Biol. Rev.

    (2020)
  • P. Ferreira et al.

    Responses of bees to habitat loss in fragmented landscapes of Brazilian Atlantic Rainforest

    Landsc. Ecol.

    (2015)
  • J. Ferreira-Ferreira et al.

    Combining ALOS/PALSAR derived vegetation structure and inundation patterns to characterize major vegetation types in the Mamirauá Sustainable Development Reserve, Central Amazon floodplain, Brazil

    Wetl. Ecol. Manag.

    (2015)
  • J.A. Foley et al.

    Global consequences of land use

    Science

    (2005)
  • V. Gagic et al.

    Functional identity and diversity of animals predict ecosystem functioning better than species-based indices

    Proc. R. Soc. B Biol. Sci.

    (2015)
  • Cited by (0)

    1

    Equal contribution.

    View full text