Seed-dispersing vertebrates and the abiotic environment shape functional diversity of the pantropical Annonaceae

Abstract

Introduction

The diversity of functional traits (i.e. characteristics of organisms that influence their fitness, growth, and survival; Violle et al., 2007) in assemblages is important for ecosystem multifunctionality, because co-occurring species with contrasting trait values may exploit different resources, and thus increase resource utilization (Gross et al., 2017). In turn, multifunctionality may lead to diversity of ecosystem services, emphasizing the importance of understanding the origin and maintenance of functional diversity for conservation of ecosystems. Functional richness (i.e. trait space occupied by an assemblage, hereafter FRic; Mason et al., 2005; Villéger et al., 2008) has been increasingly used to understand how species interact with the environment and contribute to ecosystem functioning (e.g. Cadotte et al., 2011; Mason et al., 2013). Functional richness may differ between assemblages at different latitudes and biogeographical realms, as a result of differences in historical (e.g. paleoclimatic) and present-day (e.g. current climate) processes shaping assemblages and traits (Lamanna et al., 2014; Svenning et al., 2015). The role of biotic interactions in shaping FRic across spatiotemporal scales, however, remains unclear and requires quantifying trait diversity in both interaction partners.

One such interaction type is the mutualism between fleshy-fruited plants and frugivores (i.e. fruit-eating and seed-dispersing animals). This mutualism is prominent in tropical rainforests, where up to 90% of woody plants depend on frugivores for seed dispersal (Jordano, 2000). Both fruits and frugivores have evolved adaptive traits to facilitate interactions (Fleming & Kress, 2013). For example, frugivore gape width (and corresponding body size) constrains which fruit sizes can be swallowed, leading to the largest fruits generally being dispersed by the largest animals (Fleming & Kress, 2013; Galetti et al., 2013). This ‘trait matching’ (Dehling et al., 2014; Bender et al., 2018) may also explain the distribution of fruit sizes across broad-scale global assemblages (Lim et al., 2020; McFadden et al., 2022; Wölke et al., 2023). Similarly, frugivores have evolved traits to facilitate the detection and handling of fruits (Fleming & Kress, 2013), such as primate color vision in relation to palm fruit colors (Onstein et al., 2020). This may explain the evolution of ‘fruit dispersal syndromes’, that is sets of matching frugivory-related traits between plants and animals (Onstein et al., 2019, 2020; Valenta & Nevo, 2020). Frugivory-related trait distributions and trait matching are, however, also influenced by abiotic variables, such as climatic conditions, productivity, and total plant or animal species richness (SRic; McFadden et al., 2022). Nevertheless, it remains unclear how frugivores, in addition to the abiotic environment, have influenced FRic of plants across broad-scale assemblages and biogeographical regions (but see Albrecht et al., 2018).

Here, we use a quantitative trait-based approach to evaluate whether global variation in the frugivory-related FRic of the pantropical custard apple plant family (Annonaceae) (Fig. 1a–f) can be explained by the (potential) mutualistic interaction with frugivorous birds and mammals (Fig. 1g,h). Annonaceae is the most species-rich family within the Magnoliales (Chatrou et al., 2012), comprising c. 2500 predominantly tropical rainforest species (Couvreur et al., 2011; Chatrou et al., 2012; Erkens et al., 2022). Annonaceae have been well-studied from taxonomic (e.g. Johnson & Murray, 2018), spatial (e.g. Erkens et al., 2022), biogeographic (e.g. Couvreur et al., 2011), and functional trait perspectives (e.g. Onstein et al., 2019; Xue et al., 2020), providing the basis for our study. Annonaceae show striking diversity of fruits (see Fig. 1a–e for examples), with sizes ranging from c. 0.25 cm (e.g. in certain apocarpic species such as Greenwayodendron littorale Lissambou, Dauby & Couvreur) up to 50 cm (e.g. in syncarpic species such as Anonidium mannii (Oliv.) Engl. & Diels) and brightly colored moniliform fruits with monocarps as beads in a necklace (e.g. a number of species within Monanthotaxis Baill.; see van Setten et al., 1992). This variation makes Annonaceae fruits attractive to an equally wide diversity of seed-dispersing guilds of terrestrial vertebrates, especially birds, bats, primates, and other mammals (Supporting Information Table S1; Coates-Estrada & Estrada, 1988; Kessler, 1993; McConkey et al., 2018; Onstein et al., 2019).

Fig. 1

Open in figure viewerPowerPoint

Conceptual framework of the quantitative trait-based approach adopted in this study. (a–e) Examples of species and frugivory-related functional traits of our model group Annonaceae (Table 1, Supporting Information Table S2): (a) Anaxagorea phaeocarpa Mart.; (b) Annona hypoglauca Mart.; (c) Guatteria pichinchae Maas & Westra; (d) Letestudoxa bella Pellegr.; (e) Monanthotaxis sp. (f) Inference of frugivory-related functional trait spaces that reflect functional richness of co-occurring Annonaceae species in an assemblage. (g) Comparison between frugivory-related trait spaces of co-occurring Annonaceae, frugivorous birds, and frugivorous mammals. Trait spaces were calculated separately for each assemblage and subsequently applied in our structural equation models. (h) Trait matching of Annonaceae, frugivorous birds, and frugivorous mammals across sites: (h.1) integration of plant (matrix R) and animals (matrix Q) trait with co-occurrence (species interaction) data (matrix L) to calculate the ‘fourth corner’ – matching plant and frugivore traits; (h.2–h.3) examples of hypothesized matching plant and frugivore traits (Table S2); (h.2) fruit size with animal body size; and (h.3) plant habit with animal foraging strata. Photographs by: (a, c, d) T.L.P. Couvreur; (b) D. Sasaki; (e) L. Chatrou.

Due to reciprocal resource utilization and co-evolutionary selective pressures from interacting partners, we hypothesize (H1) that frugivory-related bird and mammal FRic (Fig. 1g) explains global variation in frugivory-related Annonaceae FRic (Fig. 1f) across broad-scale assemblages, even after accounting for the direct and indirect effects of SRic and the abiotic environment. Furthermore, we hypothesize (H2) that distinct biogeographical histories may have led to differences in frugivory-related effects between realms. For example, we expect an overall stronger effect of frugivorous birds on Annonaceae diversity in the Neotropics and/or Asia-Pacific realm than in the Afrotropics, due to the higher diversity and dominance of the frugivorous bird guild (compared with mammals, e.g. primates) in the respective realms (Kissling et al., 2009; Fleming & Kress, 2011). Finally, we hypothesize (H3) that individual functional traits of co-occurring Annonaceae and frugivorous bird and mammal species are ‘matching’, underlying the association between frugivore and Annonaceae FRic (Fig. 1h; Table S2). For example, we expect that large-fruited Annonaceae predominantly co-occur with large-bodied mammals, leading to a trait matching correlation between fruit sizes and animal body sizes.